.- PROCEEDINGS OF THE Biological Society of Washington VOLUME IX 18^4-1895 WASHINGTON: PRINTED FOR THE SOCIETY, \ j d 1894-'95. COMMITTEE ON PUBLICATIONS THEODORE GILL, Chairman T. H. BEAN L. O. HOWARD F. H. KNOWLTON T. W. STANTON JUDD & DETWEILEK, Printers CONTENTS Page. Officers and committees for 1894 , iv Officers and committees for 1895 Proceedings Social Insects from Psychical and Evolutional points of view, by C. V. Riley 1-74 Fossil Oycadean Trunks of North America, with a Revision of the genus Cycadeoidea. Buckland, by Lester F. Ward 75-88 Note on some Appendages of the Trilobites, by Chas. D. Walcott. 89-97 Synaptomys cooperi in Eastern Massachusetts, with note on & stonei, by Outram Bangs 99-lt)4 A new Rabbit from Western Florida, by Gerrit S. Miller, Jr., and Outram Bangs 105-108 Preliminary descriptions of eleven new Kangaroo Rats of the genera Dipodomys and Perodipus, by C. Hart Merriam 109-116 Abstract of a study of the American Wood Rats, with descriptions of fourteen new species and subspecies of the genus Neotoma, by C. Hart Merriam 117-128 Description of a new Field Mouse (Arvicola terrxuwvx} from Codroy, Newfoundland, by Outram Bangs 129-132 Description of a new Muskrat from Codroy, Newfoundland, by Outram Bangs 133-138 LIST OF THE OFFICIOUS AXD COUNCIL OF THE BIOLOGICAL SOCIETY OF WASHINGTON IFOR 1894- (ELECTED DECEMBER 30, 1893) OFFICERS President C. V. RILEY Vice- Presidents FRANK BAKER RICHARD RATHBUN B. E. FERNOW C. D. WALCOTT Recording Secretary F. V. COVILLE Corresponding Secretary F. A. LUCAS Treasurer F. H. KNOWLTON COUNCIL TARLETON H. BEAN C. HART MERRIAM* WILLIAM H. DALL* T. S. PALMER THEODORE GILL* THEOBALD SMITH G. BROWN GOODE* FREDERICK AV. TRUE L. O. HOWARD LESTER F. WARD* STANDING COMMITTEES 1894 Committee on Communications B. E. FERNOW, Chairman T. H. BEAN CHAULKS SCIII'CHERT F A. LUCAS. EUWFN F. SMITH Coininitlet' on Publications THEODORE GILL, Chairman L. O. HOWARD T. W. STAXTON F. H. KNOWLTON T. H. BEAN Delegates to the Joint Commission of Scientific Societies of Washington RICHARD RATHBUX C. V. RILEY LESTER F. WARD * Ex-Presidents of the Society. LIST OF THE OFFICERS AND COUNCIL OP THE BIOLOGICAL SOCIETY OF WASHINGTON FOR 1895 (ELECTED DECEMBER 29, 1894) OFFICERS President GEO. M. STERNBERG Vice-Presidents RICHARD RATHBUN B. E. FERNOW C. D. WALCOTT L. O. HOWARD Recording Secretary M. B. WAITE Corresponding Secretary F. A. LUCAS Treasurer F. H. KNOWLTON COUNCIL WM. H. ASHMEAD C. HART MERRIAM* TARLETON H. BEAN C. V. RILEY* WILLIAM H. DALL* THEOBALD SMITH THEODORE GILL* CH. WARDELL STILES G. BROWN GOODE* FREDERICK W. TRUE LESTER F. WARD* STANDING COMMITTEES 1895 Committee on Communications B. E. FfJKNOw, Chairman F. A. LUCAS L. O. Ho WARD Committee on Publications THEODORE GILL, Chairman L. O. HOWARD T. W. STANTOX F. H. KXOWLTON T. H. BEAN Delegates to the Joint Commission GEO. M. STERNBERG RICHARD RATHBITX LESTER F. WARD * Ex-Presidents of the Society. (v LIST OF ILLUSTRATIONS PLATES Page I. Appendages of Trilobites 98 II. Skull and teeth of Arvicola terrsenovse 132 TEXT FIGURES Figure 1. Modifications of the hind legs of different Bees 13 2. Modifications of the hind legs of different Bees 15 3. Wax Discs of Social Bees 16 4. Architecture of Bees 17 5. Architecture of Bees 18 6. Development of Formica rufa 27 7. Honey Ants (Myrmecocistus mexicanus) 28 8. Sensory Organs in Insects 37 9. Sensory Organs in Insects 40 10. Some Antenme of Coleoptera 41 11. Antenna of male Phengodes with portion of ray 42 12. Some Antennae of Insects. . 44 VOL. IX, PP. i-xvi; 139-145 FEBRUARY 8, 1896 -PROCEEDINGS BIOLOGICAL SOCIETY OF WASHINGTON. PROCEEDINGS. January 13, 1894 220th Meeting. The President in the chair and twenty -three persons present. The following communications were presented : R. T. Hill : A New Fauna from the Cretaceous Formation of Texas.* Ch. Ward ell Stiles : The Teaching of Biology in Colleges. January 27, 1894 221st Meeting. The President in the chair and twenty-two persons present. The following communications were presented : J. N. Rose : A Botanical Trip to Northwestern Wyoming. B. T. Galloway : A Rust of Pine Leaves and the Effect of the Parasite on the Host. Theodore Gill : The Segregation of the Osteophysial Fishes as Fresh Water Forms.f February 10, 1894 222d Meeting. The President in the chair and twenty-three persons present. The following communications were presented : M. B. Waite : The Treatment of Pear Leaf-Blight.! * Abstract in Am. Journ. Sci., 3d ser., xlvii, 141, Feb., 1894. f The Early Segregation of Fresh Water Fishes. . Workers. 7. Soldiers. 10. Winged forms. 11. True royal pairs. 34 Riley Presidential Address. The fecundity of the true queen Termite is something remark able, and, based on Smeathman's observations on an African species (TrniH'* hcllir.Dxti*) the fact that an egg is produced every second, or some 80,000 a day in the height of the breeding- season, has been commonly quoted among writers on the subject. In this species the queen is sealed up in a cell which is as hard as a stone, in the central and most protected part of the termi tary, the cell being opened and enlarged from time to time by the workers, and being also perforated by holes which admit the workers to care for and feed her, while preventing the egress of the female and her attendant male escort. Among the more curious facts connected with these Termites, because of their exceptional nature, is the late development of the internal sexual organs in the reproductive forms and the existence of a single long-lived male a condition not parelleled among other insects, so far as I am aware. Further, as Dr. Hagen has pointed out, the queen represents a unique instance among insects of actual growth taking place in the imago state; for the in- tra-segmental ligaments not only expand, but grow with the in creasing gravity of the abdomen, the stigmata actually taking part in this growth, though the dorsal abdominal plates remain unaffected. In the Hive Bee multiplication of colonies takes place by divi sion, but the colonizing swarm carries in itself all the elements " 5 necessary for the foundation of a new colony. In the more typi cal Termites multiplication of colonies also takes place by division, but this is carried out by the neuters and the various adolescent stages, since there is usually but one true queen, which can not be moved. The new colony, therefore, can only obtain a true queen by introducing one of the royal pairs that wander about after they have swarmed and thrown off their wings. That great difficulty attends the establishment of such a royal pair of indi viduals in a colony is illustrated by the fact that they are rarely discovered among colonies of our commoner species of Termes proper.* *From the accounts of authors there is no , organ of smell in May Beetle (after Hauser) ; C, organ of smell in Vespa (after Hauser) ; D, sensory organs of TermesJIavipes, , tibial auditory organ, c, enlargement of same, 6, sensory pits of tarsus (after Stokes); K, organ of taste in maxillae of Vespa vulgaris (after Will) ; F, organ of taste in labium of same insect (after Will); G, organ of smell in Caloptenus (after Hauser); //, sensory pilose de pressions on tibia of Termes (after Stokes) ; /, terminal portion of antennae of Myrmica ruginodis, c, cork shaped organs, s, outer sac, /, tube, w, posterior chamber (after Lub- bock) ; K, longitudinal section through portion of flagellum of antennae of worker bee, showing sensory hairs and supposed olfactory organs (after Cheshire). All very greatly enlarged. HEARING. In regard to the sense of hearing the most casual experimentation will show (and general experience confirms it) that most insects, while keenly alive to the slightest movements Social Insects. 41 or vibrations, are for the most part deaf to sounds which affect us. That they have a sense of sound is equally certain, but its range is very different from ours. A sensitive flame arranged for Lubbock by the late Prof. Tyndall, gave no response from ants, and a sensitive microphone arranged for him by Prof. Bell gave record of no other sound than the patter of feet in walking. But the most sensitive tests we can experimentally apply may be, and doubtless are, too gross to adjust themselves to the finer sensibil ities of such minute, active and nervous creatures. There can be no question that insects not only produce sounds, but receive the impression of sounds entirely beyond our own range of per- FIG. 10. SOME ANTENNAE OF CCLECPTERA : a, lyiidius ; b, Corymbites ; c, Prionocy- phon ; d, Acneits ; e, Dendroides ; f, Diueutes ; g, I Y acrmosterna ; h, Bolbocerus ; t, Adraiies, (after L,eConte and Horn). All greatly enlarged. ception, or as Lubbock puts it, that "we can no more form an idea of than we should have been able to conceive red or green if the human race had been blind. The human ear is sensitive to vibrations reaching at the outside to 38,000 in a second. The sensation of red is produced when 470 millions of millions of vibrations enter the eye in a similar time ; but between these two numbers, vibrations produce on us only the sensation of heat. We have no especial organ of sense adapted to them." It is quite cer tain that ants do make sounds, and the sound-producing organs 011 Riley Presidential Address. some of the abdominal joints have been carefully described. The fact that so many insects have the power of producing sounds that are even audible to us, is the best evidence that they possess auditory organs. These are, however, never vocal, but are situ ated upon various parts of the body or upon different members thereof. SPECIAL SENSES AND SENSE-ORGANS. While from what has preceded it is somewhat difficult to compare the more obvious senses possessed by insects with our own. except, perhaps, the sense of touch, it is, I repeat, just as obvious to the careful student of insect life that they possess special senses which it is difficult for us to comprehend. The sense of direction, for in stance, is very marked in the social Hymenoptera which we have been considerering, and in this respect insects remind us of many of the lower vertebrates which have this sense much more strongly developed than we have. Indeed, they manifest more especially what has been referred to in man as a sixth sense, viz., a certain intuition which is essentially psychical, and which undoubtedly serves and acts to the advantage of the species as fully, perhaps, as any of the other senses. Lubbock demonstrated that an ant will recognize one of its own colony from among the individuals of another colony of the same species, and when we consider that the members of a colony number at times not thousands but hundreds of thousands, this remarkable power will be fully ap preciated. The neuter Termites are blind and can have no sense of light in their internal or subterranean burrowings; yet they will undermine build ings and pulverize various parts o f elaborate f u r n i- ture without once gnawing through to the s u r f a c e, and those species which use clay will fill up their burrowings to strengthen the supports of structures which might otherwise fall and injure the insects or betray their work. The bat in a lighted FIG. ir. Antenna of male Phengodes with portion of ray Greatly enlarged. (Original.) Social Insects. 43 room, though blinded as to sight, will fly in all directions with such swiftness and with such infallible certainty of avoiding con cussion or contact, that its feeling at a distance is practically in comprehensible to us. The manner in which anything threatening its welfare thrills and agitates one of these insect communities, and causes every individual to act at once for the common good, has been noted by all observers, and is a good illustration in point. It may be likened to the manner in which the same conditions influence communities of other animals, including man. There are emer gencies when intuitive feeling dispossesses reason, and every cap able person seems blindly urged to definite action for the protc-c- tion of the community, regardless of consequence. The war-cry of a nation is an example in point, and violations of otherwise just, but tedious, processes of law, are under certain circumstances deemed justifiable. I shall never forget the emotion that in fluenced the citizens of Chicago the day following their great fire in 1871. Reason, argument, judgment, were in abeyance. The quicker, intuitive processes prevailed, and to meet lawlessness and the tendency to incendiarism, every right-minded citizen was ready to do vigilant duty, regardless of personal interest, every incendiary being hung to the nearest lamp-post, without ado or delay. It was the universal and deep-seated instinct of self-pres ervation. TELEPATHY. But however difficult it may be to define this intuitive sense which, while apparently combining some of the other senses, has many attributes peculiar to itself, and however difficult it may be for us to analyze the remarkable sense of direc tion, there can be no doubt that many insects possess the power of communicating at a distance, of which we can form some con ception by what is known as telepathy in man. This power would seem to depend neitner upon scent nor upon hearing, in the ordinary understanding of these senses, but rather on certain subtle vibrations, as difficult for us to apprehend as is the exact nature of electricity. The fact that man can telegraphically transmit sound almost instantaneously around the globe, and that his very speech may be telephonically transmitted, as quickly as uttered, for thousands of miles, may suggest something of this subtle power, even though it furnish no explanation thereof. The power of sembling among certain moths, for instance, es- 44 Riley Presidential Address. pecially those of the family Bombycidse, is well known to ento mologists, and many remarkable instances are recorded. (Note 7.) I am tempted to put on record, for the first time, an individual experience which very well illustrates this power, as, on a num ber of occasions when I have narrated it, most persons not familiar with the general facts have deemed it remarkable. In 1863 I ob tained from the then Commissioner of Agriculture, Col. Capron, eggs of Samia cynthia, the Ailanthus silk worm of Japan, which had been recently introduced by him. I was living on East FIG. 12. SOME ANTENNA OF INSECTS: a.Telea polyphemus, male, x 3 ; b and c, tips of rays of same still more enlarged ; d, Chirouomus x 6 : statement that the bee "begins by performing these actions before it has itself had any individual experi ence of cell-making and without its parents ever having had any ancestral experience." While this statement represents accepted belief, it follows from what I have already said of the bee that it is essentially untrue. The worker could no more begin to secrete wax and build cells until it had ac quired a certain age than could mammals secret the lacteal fluid before a certain age ; and during its early life as an adult it had the experience of its older fellows to guide it, were such guidance necessary. The example chosen by Romanes was simply unfortunate. To understand the development of the cell-building instinct, we must consider the stages of its development as illustrated in the varying forms of cells yet existing, from the cruder cells of Bombus on, and remember that each step in the more perfect building has been accompanied by structural modifications, and that the instincts have been accumulated and perfected by heredity pari pawn with the struc tures ; further that the habit probably became so firmly fixed before the neuters had been differentiated, that it has been transmitted since that time through the queen, though she herself no longer possesses it; further that while instinctive performance is ordinarily inevitable, it yet varies in the amount of its fixity and accuracy and often leads astray or fails; and, finally, that it is often modified by individual experience or reason, or by communal interest or necessity these truths applying particu larly to the social insects, and in a variable degree to all animals. 50 Riley Presidential Address. to within recent years, have been so effectually discarded that they are even renounced by the more advanced theologians. I refer to the belief that organisms were specially created as they now exist, and that man was apart from, and not a part of, the rest of the animal world. It is my judgment that a third equally prevalent notion is essentially false, and will have to be abandoned before we can properly appreciate the psychology of animals. I refer to the notion that the lower animals do not reason, and are incapable of conscious reflection and thought. It would be easy to occupy your time for hours with accounts of their actions which can be explained only upon the views here set forth, and which are utterly at variance with the popular notions and prejudices. The insects to which I have referred to-night are admitted to be among the more intelligent of their class; but they are only illustrations of an intelligence which is found throughout the other orders, and which impresses us in proportion as we study it and come to realize and recognize it. We can never properly appreciate, nor properly bring ourselves into sympathy with these lower creatures, until we recognize that they are actuated by the same kind of intelligence as we ourselves. There are certain acts which all creatures necessarily perform, as an outgrowth of their organization. These are essentially the instinctive acts, and are, for the most part, inevitable and often unconscious. A great many of the acts of rational men are, in this view, instinctive, and from birth to maturity many of them are prompted solely by the consecutive development of different parts of the organization, and are much less the result of training and teaching than is generally believed. Most of the acts of insects are instinctive and explicable upon this same view, but no one can study them carefully and without bias and not feel that these instinctive and inevitable actions are associated with many others which result from the possession of intelligence of conscious reasoning and reflective powers. In this view of the case is the whole world truly kin, and is man brought more fully into sympathy with and appreciation of it. Is it not significant also, that, just as in man, among mammalia, the higher intellectual development and social organization is found correlated with the longest period of dependent infancy; that this helpless infancy has been, in fact, as Fiske has shown, Social Insects. 51 a prime influence in the origin, through family, clan, tribe and state, of organized civilization; so in the insect world we find the same correlation between the highest intelligence and depend ent infancy, and are justified in concluding that the latter is, in the social Hymenoptera as in man, in the same way the cause of the high organization, and division of labor so charateristic of them ! HEREDITY: NATURAL SELECTION. The application of the principle of natural selection to the pro duction of neuter insects, and especially to the production of neuter insects of diversified form, seems, at first sight, impossi ble. Indeed, we know that Darwin felt that this question of neuter insects was one of the most difficult to deal with in con nection with his grand generalization. Weismann, who believes in the all-sufficiency of natural selection, insists, and has within the past year, in his controversies with Herbert Spencer, empha sized his belief, that these neuter insects absolutely preclude the idea of the transmission of acquired characters, and endeavors to explain their occurance by his own peculiar theories as to modi fication taking place in the germ plasm. I shall certainly not attempt, in the limited time that I may yet hope to hold your attention, to discuss in detail the views held whether by Weis mann or his opponents ;* but I will venture to show that, while *The chief argument in favor of Weismann 's theory of heredity is that it is an earnest attempt to establish a basis in observed histologic and em- bryologic facts. The idea of the continuity or "immortality" (using the word in his own qualified use of it) of the germ plasm is a bold one which gives us at least a conceivable and material basis of reproduction, and is justified, though not absolutely, in the facts referred to and in the history of the Protozoa. One of the chief arguments against it is, in my judgment, that, inasmuch as it precludes the transmission of impressions on the soma, i. e., individually acquired characters, Weismanu has, in order to sustain the theory, been led to question and finally to deny the transmissibility of such ac quired characters. It is difficult to formulate the later modifications of the orig inal theory without using many Weismannisms, themselves requiring chapters of explanation ; but that variation is due to direct effects on the germ plasm by inequalities of nutrition, is, I believe, a correct statement of his latest views. The trouble with all theories of reproduction and heredity based solely on observed microscopic facts, is that the essence, the life principle, the potential factors, must always escape such methods. Wherefore any theory that will hold must cover the psychical as well as the physical facts the total of well established experience and this truth was doubtless 52 Riley Presidential Address. the social insects offer the most serious obstacles to the accept ance of the theory of natural selection as an all-sufficient theory to explain the phenomena, yet the facts are perfectly explicable upon the general principles that have governed the modification of organisms, among which that of natural selection plays an im portant, but limited part. In the economy of the Hive Bee we have seen that all the neuters are structurally alike, and that the different functions which they perform result from inherited tendencies or structural peculiarities developed at different ages. There are some records of abnormal workers, small drones, and slight variations in the amount of arrestation of development ; but on the whole the three classes of queen, worker and drone are remarkably well differentiated and fixed. We have seen that the differences in the two former classes result from conditions of food, treatment and environment of the young, and are under the control of the colony. Each fertile egg has the potentiality of developing a fertile queen, and as the neuters, under exceptional conditions, are able to lay eggs which invariably produce drones, the queen, through such drones, must occasionally inherit indirectly from the workers. At bottom, however, the differentiation between the workers and the queen is purely a matter of food and bring ing up, or education, as the French would more correctly call it. In other words, the ultimate result is decided for each generation in the treatment of the young or the larvse. The drone results from an unfertilized egg, and as the egg is only fertilized when the tip of the queen's abdomen is pressed into a worker cell, and not when thrust into a drone cell, the production of drones is also under control of the colony. I have already called attention to the fact that other species recognized by Darwin in framing his tentative theory of pangenesis. Weismann's efforts to derive a physical theory of reproduction and evo lution find a paralell in the efforts of those entomological histologists who, starting with the conception that the development of the individual was but an unfolding of structures already nascent in the embryo, expected to find and even claim to have found all the structures of the imago repre sented, en petit, hi the larva. In truth, however, there is a total re-adjustment of cells, and development de. novo of organs, with each important change or molt, and the vital force which impels this development, whether of the minutest bodily structure or the subtlest intellectual attribute, is the great mystery beyond explanation. Social Insects. 53 of bees show gradations in these two kinds of females, and that some species permit more than one queen or fertile female in the colony and would refer for further details, both as to present gradations and variations to the Notes, especially numbers 1, 2, 3, and 4. Natural selection, if it has played any part at all, must have done so chiefly in the manner ingeniously suggested by Darwin himself, namely, not as between individuals, but as be tween colonies. The tendency to produce arrested females or neuters doubtless became fixed in some ancestral form through social selection, and is kept up by this and colony selection. In the wasps we have a very different state of things, involv ing the parthenogenetic production of arrested females and the seasonal production of fully developed forms of both sexes. Here again, the evidence all goes to show that the differences de pend for each generation on the environment, food and method of nurture of the larva, the tendency having become fixed in vary ing degrees in the different species, and only so fixed by being transmitted through the queen or sexually perfect females. So far as natural selection has acted at all, it has acted on the poten tiality or inherited tendencies of these females. Very exact in formation is not yet at hand as to how far the neuters are vari able, whether as to condition of the reproductive organs or as to size. But judging merely by mounted specimens which I have examined in various species, it is probable that there is some variation in these respects, though the three classes are quite neatly differentiated, much as in the bees. When it comes to the ants, the problem is more compli cated ; but we may safely assume that the different forms have been brought about by the same influences. In a large colony of individuals, where size and character are not fixed by a definite cradle, but where the young larvae are free and are carried about, nursed and fed by the workers, there would naturally arise greater variations between individuals, and while the kind of nourishment, or the kind of nurture, or the age of the female at the time the ova are produced, or the season of the year, have doubtless all contributed to the variation, and may still independently contribute to it at the present time ; yet, what ever the causes of this variation, it has become fixed in certain definite lines that are more or less useful to the species. Whether or not the proportion of the different individuals is under the 54 Riley Presidential Address. control of the colony as a whole, by virtue of the treatment of the larva, it will always be difficult to prove, though there is every reason to believe that, as in the bees, there is, to some ex tent, such control, and that the relative proportions of the differ ent forms will depend upon circumstances. But the fact re mains that, in ants, as in bees and wasps, the neuters are but arrested females, and are capable of becoming, under exceptional circumstances, fertile, and that we see in the different species all gradations, not only as to the number of forms of the workers, but as to the number of fertile females that are allowed in the same colony to provide for the continuance of the species. We also find in the same species great variation and gradation in the characters of the different sets which form the community, es pecially between the different forms of workers, in contrast to what I have remarked as to bees and wasps. This has been re corded not only by writers like Darwin and Lubbock, but by all who have given close attention to the subject; while Ch. Lespes (Ann. des Sciences Nat. (4) 20, pp. 241-251) in his " Observations sur les Fourmis Neutres" has shown that all neuters have traces of the female reproductive organs; that these traces vary in the different species; and that where there are two forms of neuters these pass insensibly into each other through intermediate forms. The ants thus furnish us with varying degrees of social organiza tion when the different species are considered, while the different classes in the same species are not as definitely fixed as in the bees or the wasps. Now it were comparatively easy to account for these neuters among the social Hymenoptera and the different forms and attri butes which they present, by putting aside natural selection, as expounded by Darwin, and substituting therefor social selection acting not on generations in time, but on the individual at once by the manner of its bringing up ; and surely there would seem to be sufficient justification for this course when we find not only such great physiological and functional, but such profound structural modifications induced by larval environment and nur ture, as I have pointed out, especially between the queen and the worker bee.* This has, in fact, been the chief explanation which *Mr. Herbert Spencer, in one of his rejoiners to Prof. Weismann, (Con temporary Review, December, 1893) refers to a chapter on The Determination of Sex by Prof, Geddes and Mr. Thompson in their " Evolution of Sex," Social Insects. 55 I have offered for the facts, in discussions with friends and be fore the society, limiting the action of natural selection to colonies as a whole. Few persons who have not had large experience in rearing insects can appreciate the full influence of larval environ ment and food on the ultimate imago, or the power of larval accomodation to various conditions. All insects in the larva state possess this power, within varying limits, and it is nowhere more marked than in the Aculeate Hymenoptera. I have called attention to it on numerous occasions* when treating of parasitic species, and it is particularly noticeable in the fossorial Hymen optera and the Melo'idse. Size, especially, may easily be dimin- where they state that "such conditions as deficient or abnormal food." and others " causing preponderances of waste over repair tend to re sult in the production of males," while " abundant and rich nutrition " and other conditions which "favor constructive processes " result in the production of females." He then cites J. H. Fabre's statement that in the nests of Osmia tricornis the eggs at the bottom of the cell which are first laid and accompanied by much food, produce females, while those at the top, laid last and accompanied by one-half or one-third the quantity of food, produce males. (Souvenirs Entomologiques, Seme serie, page 328). He further refers to Hiiber's observations, that the queen bee only lays eggs of drones when declining nutrition or exhaustion has set in, and that when the workers in bees and wasps lay eggs, these produce drones. These statements are not entirely justified. I cannot speak positively of Fabre's observations, though I suspect something back of the larval food- supply which has fixed the sex and determined the treatment of the larva. But the queen bee produces drones at any age by the egg passing into the drone cell and not being impregnated in passing the spermotheca. She pro duces drones only when she is superannuated, because the spermatozoa have become exhausted. In wasps it is just the contrary, the unimpregnated egg producing ordinarily, not a drone or a male, but a female. I have already called attention to the ease with which erroneous conclusions are drawn in this matter of regulating sex by food of larvae, ex ovo (Am. Naturalist, Vol. VII, pp. 513-531, September, 1873) and the evidence would seem to how that the influence is confined to arrestation or modification of the sex without changing it. The subject is, however, most intricate, and further experimental facts are needed. Spencer's conclusion is, nevertheless, gen erally true, namely : ' ' that one set of differences in structure and instincts is determined by nutrition before the egg is laid, and a further set of differences in structure and instincts is determined by nutrition after the egg is laid." *See notes on Tiphia inornata, Sixth Report on the Insects of Missouri, p. 123, and upon Blister-beetles, First Report U. S. Entomological Commission, pp. 295-302, 56 Riley Presidential Address. ished one-half or more, or fully doubled, from the normal, by limiting or increasing the supply of food, as I have proved with Pelopaeus. But when we come to the facts in the economy of the Termites, this explanation does not hold good to the same degree. Here we find still greater diversity in form than even among ants, under circumstances where control of these forms by the colony itself must be much less, but nevertheless does occur. The young Termite is to a limited extent, and during early life only, provided with food by members of the colony, and from birth is essentially a free moving agent, less dependent on the adults. We have much yet to learn as to the actual facts, which would seem also to vary in different species. Thus in Eutermes Mr. Hubbard believes, but I think wrongfully, that the young feed on nodules, specially prepared, of comminuted and doubtless partly digested material, while Fritz Mu'ller belie v 7 es that they feed on a fungus mycelium which develops on such prepared substance. The truth with most species seems to be that they are fed on a semi-liquid fluid disgorged from the mouth, whether of -the workers or the undeveloped queens; while in some cases they are fed from a secretion from the anus. (See Note 6.) In these respects and in the early helplessness of the larvae, they closely approximate the social Hymenoptera. Similar variations to those found in social insects, whether sexual or seasonal, are extremely common among insects which are not social, as is well exemplified by the long category of phy- tophagic variation, secondary sexual characters, and of dimorph ism and heteromorphism among insects. These variations in non- social insects are often equally as marked and as curious, struc turally, as they are among social species. They are also, except, perhaps, the secondary sexual characters and the- variations which take on the form of mimicry, equally difficult to explain on any view of natural selection that is all-sufficient. On the whole, then, it may safely be said that the production of neuter insects is determined in each generation by the colony itself, in the man ner in which the larvae are fed and reared. In so far as this is true, it is outside the domain of natural selection, and speaks eloquently in favor of the various other causes of variation and modification which have been insisted upon by many of our lead ing American biologists, and which I have repeatedly urged in Social Insects. 57 my own writings.* The tendency to such production was doubt less developed in the ancestors of the present species, and we may even trace the steps by studying the gradations in existing species. The facts connected with the social insects which I have con sidered, present the strongest argument in favor of the heredity of acquired characters and tendencies. Competition has been between colonies rather than individuals, and those colonies which have acquired, through heredity, the habit of producing, through one or more fertile females, the different forms which have proved useful in the social economy, have, in the course of time, survived others in which such tendency was less pro nounced. Yet various steps in the process are yet manifest in the different species, and under these circumstances it seems to me foolish to insist that the fixed habit in one species has, per se, any especial advantage over the less fixed habit in others which still maintain themselves. I need hardly say to the members of this Society who are familiar with my views as to the causes of variation, that it does not follow in my mind that the different forms of Termites, for instance, that are found in the colonies of some species, are all essential, but that some of the forms may be advantageous, others only par tially so, and still others purely fortuitous. The tendency to vary an inherent property in all organisms has shown itself among the individuals of these different colonies. These variations have been guided by natural selection among col onies, and by what I have just referred to as social selection among individuals, along certain lines which are most useful. In other cases the variation has accumulated along lines of secondary utility ; while in yet others it has gone along lines which are purely fortuitous and still most variable and unfixed natural selection playing little or no part in these. In species with the less complete social organization, the existing variations will be greatest; while the structures and functions have become most fixed and show least tendency to vary in those species which have become most specialized and perfect in their social economy. It is very questionable, however, whether, in the struggle for ex istence, this greater specialization and fixity give the species any *See more particularly the address before Section F, at the Cleveland (1888) meeting of the A. A. A. S , and the paper before this Society "On the Interrelations of Plants and Insects," Vol. VII, pp. 81-104 (May, 1892). 58 Riley Presidential Address. advantage over another which is more elastic and variable. On the contrary there are many facts which go to show that extreme specialization is a disadvantage and the precursor of decrease and ultimate extinction. So that natural selection, in this light, if limited, as its exponents have limited it, to the production of characters absolutely essential or useful to the species, must play a yet more restricted part in organic variation than even I have allotted to it. Social selection, as here expounded, implies, it is true, a degree of intelligence which has unusually been denied these creatures ; but the phenomena are some of them inexpli cable upon any other theory, and I have, I hope, already shown how little reason we have for denying them such intelligence. In a certain way the production of these specialized individuals in a colony of insects may be likened to the production of specialized individuals in a human community. In new coun tries, like our own, the specialization has not become so marked, but in the older communities of the world, the life of the indivi dual, and especially the early training and environment, produce certain characteristics which permit us to stamp at once the typi cal sailor, soldier or butcher, the various artisans and the men of various professions. They undergo essential modifications in mind and body. Yet there is no question or very little of selection, whether natural or artificial. The tendency to vary in given directions becomes fixed through heredity, since the char acteristics of different nationalities in comparison with each other cannot be so well explained upon any other view. Certain types persist, and the same laws which will explain the recur rence and persistence in a promiscuous community of, say, the red-headed type, whether that of atavism or any other be ad duced, will undoubtedly apply to the persistency of types in the social insects. That no material or mosaic theory of heredity yet propounded is satisfactory, as accounting for the facts, does not affect the question, and that natural selection, as expounded by Weismann and the ultra-Darwinians, fails to explain the phenomena, is the very best evidence that too much is claimed for the theory. INVERTEBRATE vs. VERTEBRATE. I used to be fond of speculating as to the possibilities of the articulate type as exemplified in the ant, in comparison with the Social Insects. 59 vertebrate type as exemplified in man, had the former continued its development so as to approximate, say, the eagle in bodily size and man in brain development. That the Arthropod type could attain to such dimensions is evidenced in the Euryp- terus or water scorpion which prevailed in early geologic times, and attained a length of six feet; while a modern Japanese crab (Megachilus kaempferi) has a spread of ten or twelve feet, and is a formidable creature. For very much the same selfish reasons that begot most of our earlier notions as to man's origin and place, it has been assumed that he represents the perfection of the animal organization, the highest expression of an all- wise Creator. Following this same idea, our own world, it has been reasoned, is the only one peopled. Now it has never seemed to me that there was any justification for the assumption that existing forms of plants or animals must of necessity have assumed the physical or mental characteristics which belong to them, considering the myriad forms which have preceded us and gone, or the many which are yet with us, but fast going. Remembering, also, that the race is not always to the swift, nor the battle to the strong, there would seem to be no valid reason why, on some other sphere, under like, or even under unlike conditions, life may not have taken on other distinctive types or attained developments inconceivable to us ; or, for that matter, why it might not have been differently mani fested upon our own little earth. Place the directing enginery of the human brain in a body with a hard, external skeleton, which should at once be a defensive armor against exterior attack, a protection to all the vital organs, and yet allow free play to every possible movement; with a breathing system that is multiple, and therefore less liable to get out of order than where it is concentrated in one place ; with six or more legs ; extremities variously differentiated, so as to enable one pair of them to perform the functions of our hands, while other pairs possessed greater prehensile, tactile or other special ized powers; with powerful primary and with supplemental jaws; with all the senses and sense organs we possess and others added ; with simple and compound or telescopic eyes combined in the same individual; with a venomous, offensive and defensive weapon ; with a social organization in which working, fighting and reproductive elements are well differentiated and yet under 60 Riley Presidential Address. control ; with the power of aerial flight developed when wanted ; with a reproductive system that permits of great prolificacy and yet avoids all the dangers of placental birth ; with the power of temporarily suspending the active life functions when necessary ; and, finally, with the power of such renewal of both the softer and harder tissues of the body as ecdysis involves and you have in fancy a creature which would easily make the earth and all the fullness thereof its own. The great industry exhibited by social insects has been a fav orite topic wherewith to point a moral to the sluggard; but I ven ture to suggest that their economies, if they do not point other morals, are extremely suggestive to man. With all their other traits, so comparable to those characteristic of human society, they will hardly be charged with the possession or practice of any theology ; yet we may look in vain, among all the nations of the earth, unless, indeed, among the similarly unblessed aborigines of Borneo and some other lands, for greater self-sacrifice or cour age in defending the common weal ; for greater loyalty to the sovereign head of the community, not made by divine right, but practically chosen by the commoners; for greater attention or care in the education of the helpless young, or for more har monious or friendly action between the individuals that form the community. Without form or ceremony they have developed an altruism which with us is believed to exemplify the highest phase of civilization. Nor am I quite sure that they have not solved the social prob lem in a way that, so far as the good of the community as well as the individual is concerned, has marked advantages over the many varied attempts in the same direction by mankind in dif ferent parts of the world. If a large proportion of the units of both sexes which go to make up human society could be so brought up and trained that the sexual instincts remained per manently arrested and undeveloped, while along with this arresta- tion in this particular there went an increasing intellectual de velopment and energy, to be expended in profitable industry, what a large share of vice and misery in human society might be avoided, and what a large amount of increased happiness among the multitude might thus be secured, since in the end, intel lectual and bodily activities, freed as far as possible from all baser passions, bring us the highest happiness that we can realize ! Social Insects. 61 APPENDIX. NOTE 1. The principal Races of Apis mellifica. The common form of this species, known as the Brown, the Black or the German bee, is the best-known. It is found throughout northern Europe, and as far south as central Austria, central Switzerland, and southern France to the Italian frontier. It also occurs in Portugal and Spain, and extends into Siberia, and, during later centuries, has been introduced into North and South America, many of the Paciiic islands, and into Aus tralia. Its chief merits are that it has a moderate swarming propensity and is an excellent comb- builder and honey gatherer, and accommodates itself to tne greatest extremes of climate. Its disadvantages, as compared with some other varieties, are a disposition to rob, to attack persons who approach the hive and to be somewhat less industrious. The general color is a dull brown, lighter on the thorax, the queens nearly black. The Heath and Brabant bees, sub- varieties, occuring in the heath districts of northern Germany, are much given to swarming, a habit which has be come fixed by the stimulative feeding in spring practised by the bee-keepers there for at least two hundred years. The Italian or Ligurian bee, originally confined to Italy, Sicily, Sardinia, the southern Tyrol, and southern Switzerland, has now been introduced into most countries where the common black bee occurs. It is gentler in dispo sition, but not so good a comb-builder arid, with a more tender constitution, does not thrive in extreme northern climates. The color of the Italians is in general much brighter, and the first three seg ments of the abdomen are golden-yellow on their dorsal surfaces. Its qual ities and its color have become fairly well fixed by artificial selection which there is every reason to believe has been practised in Italy for some two thousand years. Both Virgil and Columella evidently refer to it, the former (Georgics IV, 98) speaking of two kinds of bees, the better of which he de scribes as having shining bodies, variegated like drops of gold. The tend ency to vary under domestication at the present time would indicate that the the race is a composite one, and Mr. Frank Benton informs me that by cross ing the Egyptian, the Palestine or the Syrian with the common brown Ger man race, workers are produced in a few generations that can scarcely be dis tinguished from Italians ; a fact which as regards the Egyptains, was ascer tained by the Berlin Acclimatization Society which, some 30 years ago, experimented with the honey bees native to Egypt, and which Mr. Benton has since confirmed by tests with the other two races (Palestine and Syrian). He finds also, that the Syrian tpye leads, when crossed with the common brown race, most commonly to the Italian type, a fact which is significant when we remember that the Phoenicians ancient inhabitants of Syria established colonies in southern Italy at a very early date. We can hardly realize to-day the importance that was attached to the production of honey and wax in Egypt and the surrounding countries in those days, until we remember the uses to which these articles were put in connection with the religious rites of the people, and especially the embalming of the dead, as w r ell as the relative importance of honey in those early days in the absence of the many other sweets which we possess. In the United States the Ital ian race, by selection since its introduction a third of a century ago,* has undergone more rapid modification than any of the other races, though *See a paper by the author on "What the Department 9f Agriculture has done for Apiculture." Proc. North American Bee Keepers' Association, 1893. 62 Riley Presidential Address. greater efforts, proportionately, have been made with these another fact which would indicate that the Italian type is less fixed than some of the oriental races. The Oarniolan race is confined to Carniola, Austria, and the adjoining provinces, and is a local type developed by some centuries of peculiar treat ment with little intermixture of outside blood. This race is somewhat larger than the others, exceedingly robust, the distinctive color-mark being light gray varying to steel blue, the abdominal segments being all edged with pubescence of this color and the thorax thickly set with the same. The race is characterized by great prolificacy, which can be traced to the constant stimulative feeding early in the season, and by a very mild disposition, a result which would seem to be due to the frequent manipulation of the hives, migratory bee-keeping having been practised for centuries in Carniola. The Cecropian, Attic, or Hymettus bees of Greece, on the other hand, though similar to the Carnioian race in markings, are exceedingly irritable, as a result, doubtless, of their being very little manipulated or interferred with. The Tunisian bees are found in Tripoli, Tunis, and Algeria, where they are extensively cultivated by the natives. The type is uniformly dark in color. The queens are very prolific and when preparing to swarm 200 to 300 queen- cells are often constructed, instead of only 8 to 10 as is usual with the ordi nary race. The workers are small, very active, irritable and vindictive. Be cause of this and the fact that they do not winter well, in consequence of prolonging the brood season, their introduction has been very limited. The Egyptians, or the bees found all over northeastern Africa, and which for several thousand years have been extensively cultivated in Egypt, pos sess very marked characteristics as regards color, form and habits, and have been regarded by many as worthy of specific rank, having been described by Latreille as Apis jasciata. The workers are small-bodied, slender, cov ered with a dense, light gray pubescence, and the abdominal segments are edged on their dorsal surfaces with a lemon-yellow color, giving with the gray pubescense a banded effect. They do not withstand our winters and are easily angered by manipulation, not being amenable to smoke like European bees. The queens are prolific and when the colonies are made queenless great numbers of workers commence depositing eggs at once. The Palestines and /Syrians possess many of the qualities and characteris tics of Egyptians; yet the queens, workers and drones are readily distin guishable from those of the latter, being less yellow and larger bodied, es pecially the Syrians. They are marked varieties, more fixed than the Italian, and evidently forming, with other eastern Mediterranean bees, an Oriental group having allied characteristics and of which the Egyptian is the extreme type. The Caucasian and Smymian races vary more than the other Oriental races. In specimens from Smyrna the light yellow coloration of the abdominal segments noted farther south is found to be replaced by a darker yellow and the light gray pubescence by a less dense and darker gray, often brownish, pubescence. Queens, workers and drones are larger bodied and variations in temper and habit may also be noted. The Cyprian race, having been isolated for a long period, is, as might be expected, a very fixed one the most thoroughly so of any race of bees yet brought to this country, and transmits its peculiar markings and character istics through many generations of crosses with any other known type. In general it resembles the race found on the adjacent mainland, whence it was probably brought by the early Phoenicians who colonized Cyprus. Very characteristic markings of this variety are the bright yellow lunule which the postscutellum shows and the bright yellow of the ventral surface of the ab domen clear to the tip. The conditions under which this race has been established have resulted in the survival of a hardy, active race, capable of procuring a living and storing a surplus where others could barely subsist. The literature refers almost entirely to the older countries of Europe and Social Insects. 63 the East. Some modification has doubtless taken place in the tropical parts of America but the subject has not yet been sufficiently studied in those countries. NOTE 2. The Species of Apis with their Varieties. (1) Apis mellifica, L. as indicated in Note h, is found in all the coun tries of Europe, and extends over the whole of Asia Minor into the Syrian Desert and south into Arabia. It occupies all the islands of the Med iterranean and has spread through all the northern countries of Africa southward into the Desert of Sahara. South Africa has one or two varieties belonging to the species, while the representatives of the genus found in Senegal and the Congo country doubtless belong to this species, as do those of Madagascar. It has been permanently introduced into Australia, Tasma nia, New Zealand and many of the islands of the Pacific ocean. Whether the honey bees reported from northern India belong to this species or not, has not been definitely ascertained. It is also more than probable that the honey bee of China, described under the name of Apis sinensis, is but a variety of this species. In North and South America it is evidently intro duced, and has spread into some of the adjacent islands. There is a differ ence of opinion as to whether the honey bee native to Egypt, which Latreille describes as Apis fasciata, should have specific rank or be regarded as a variety of melliftca. While Frederick Smith , who was one of our best author ities, was inclined to attribute to it specific value, the fact that it interbreeds with mellifica, producing fertile offspring, would rather confirm the opposite view. Respecting the honey bees of Tasmania, Senegal, the Congo and Madagascar, our information is insufficient to permit us to say whether they are specifically distinct or not, and the same may be said of the Hazara, Bhootan, and Bushar bees of northern India and other more or less distinct types found in Japan. (2) Apis indica Fabr. The extent of territory occupied by this small East Indian bee is not definitely known, although it has been definitely reported from northern and southern India, Ceylon, Farther India and Java. Apis nigrocincta ; A. socialis, Latr.; A. delesserti Guer.; A. perrottetii Guer. and A. peronii Latr. are probably only varieties of A. indica. (3) Apisflorea Fabr. This, the smallest bee of India, is found generally in southern India and Ceylon, and there are indications, that it is common to other portions of the East Indies. Apis lobata described by F. Smith in his first catalogue, is dropped from the second edition. (4) Apis dorsata Fabr. nigripennis Latr. =bicolor Klug. =testacea. It is somewhat questionable whether the names here given as synonymous are such, or names of true varieties of dorsata. A. dorsata, known as the Giant East Indian Bee, is found in British India, Ceylon, Farther India and the Dutch P^ast Indies. (5) Apiszonata Guerin. Found in the Philippine Islands and Celebes. Mr. F. Smith enumerated this as worthy of specific rank, when he revised his catalogue in 1876. He referred to its greater size and difference in form of the metatarsus compared with that of A. dorsata. But Gerstaecker as serted in 1865 that this difference in structure of the metatarsus does not exist is " purely imaginary ". Mr. Frank Ben ton, to whom I am under obligations for valuable infor mation on this subject, has kindly prepared for me the following table as indicating his own ideas of the grouping of the species of Apis, and the known varieties of these. 4) .22 OH 84 Riley Presidential Address. The Species of Apis with tiieir Varieties. f Race. Common Brown, Black, or German. Hab.: Central, north ern and northwestern Europe ; introduced into N. and S. America, Australia, New Zealand and Pacific Islands. Sub-var. Heath. Hab.: Heath districts of North Ger many. Sub-var. Brabant or Small Holland. Hab.: Brabant (Hol land and Belgium). Race. Carniolan. Hab.: Carniola, Carinthia (Aus.). A distinct var. Sub-var. Hungarian. Hab.: Northwestern Hungary. Race. Dalmatian. Hab.: Dalmatia (Austria). Race. Herzegovinian. Hab.: Herzegovina (Austria). Race. Cecropian, Attic or Hymettus. Hab.: Greece and the adja cent islands. Var. ligustica Spin., Ligurian or Italian. Hab.: Italy and adja cent islands. S. Switzerland, and S.Tyrol; introduced HH -{ into other parts of Europe, N. and S. America, Aus tralia and New Zealand. Var. -\rufescens Hab.: Tasmania (acc'd to M. Girard). Var. \mgritarum St. Farg. Hab.: Congo (Africa). Var. \*adnnsoni Latr. Hab.: Senegal (Africa). Var. scutellata St. Farg. Hab. : South Africa. Var. cfiffra St. Farg. Hab.: South Africa. Race. Tunisian. Hab.: Tunis, Algeria. Sub-var. Minorcan. Hab.: Balearic Islands (Spain). Var. t*ttrac07or Latr. Hab.: Madagascar ; intr. into islands of Bour bon and Mauritius. Race. Smyrnian. Hab.: Asia Minor. Race. Caucasian. Hab. : Caucasus. Race. Cyprian. Hab.: Island of Cyprus. A very distinct var. Race. Syrian. Hab.: Syria, northward from Mr. Carmel. Race. Palestine. Hab. : Palestine. Var. -fasciuta Latr. Hab.: Egypt. Race. Hazara. Hab.: Hazara District, Punjab (India). Var. *sinensis. Chinese bee. Hab.: China. =cerana Fabr. Race. Busbar. Hab.: Busbar District, Punjab (India). Rac,-japanese J I'. 1 3. " Small brown bee. ' 'Hab. : Hikigoie (Satsuma) Race. Boohtan. Hab.: Boohtan (India). It is very probable that further investigation of this group will bring four of its varieties under A. mellifica, and the last one under A. indica. *indica. Fabr., Small East Indian bee. Hab.: British and Dutch East Indies. =8odali8 Latr. Hab.: Bengal. =delesserti Gue"r. Hab. : Pondicherry. =perrottetii Gue>. Hab.: India. =peronii Lat. Hab.; India. Var. (?) *nigrocincta. Hab.: "Celebes, Borneo, etc." (acc'd to F. Smith. ^Regarded by Frederick Smith as good species. fNot positively known that they will interbreed with Apis mellifica. All others named under A. mellifica will do so. Social Insects. 65 _ , *florea Fabr. Hab.: India, Ceylon, Borneo. < 1 =lobata Smith. Hab.: India. f*dorsata Fabr. Hab. : British India, Ceylon, Farther India, Dutch East Indies. =bicolor King. Var. nigripennis Latr. Hab. : Bengal, t Var. testacea Smith. Hab.: Timor. *zonata Guer. Hab.: Philippine Islands, Celebes. This may prove but a variety of A. dorsata. NOTE 3. Polliniferous Organs in Bees. The modification of structure and hairy vestiture (see Fig. 2) to facil itate the collection and transportation of pollen is, perhaps, exhibited in its most perfect development in the Hive Bee. That these peculiarities have been evolved from those possessed by less specialized species of social bees, represented by existing Meliponae and Bombi, and still more remotely from those of solitary bees, will not be questioned by those who study the steps in the process as exemplified in modern species. The pollen of liowers is variously collected by different bees, and different parts of the body are specially developed for this purpose. But in the Hive Bee the specialized polliniferous apparatus is limited to the posterior legs, and in these to the tibia and the basal joint of the tarsus, so that the development of these pails only need be traced. In the case of the tibia the first thing to be noted is the entire absence of the tibial spurs, which are present in all Hymenoptera except the genus Apis, and its near allies Melipona and Trigona. The tibia and first tarsal joint are greatly broadened and more or less concave exteriorly, and the latter is extraordinarily enlarged, so that it is nearly equal m size to the tibia. The outer surface of this modified tarsal joint'is not remarkable and has no specific function, but the inner surface is divided into transverse rows of stiff spines or combs, reddish in color, the rows slightly overlapping and elevated at a slight angle from the surface of the joint. The function of this series of combs is to collect the pollen grains which become entan- led in the feathery hairs of the thorax of the insect, and an examination will almost invariably discover more or less of the grains of pollen in these combs. During the collecting of honey and pollen, the bee is constantly passing the face of this tarsal joint over its abdomen, removing the pollen grains from time to time, and emptying the load of pollen into the pollen- basket proper or corbiculum, on the" outer face of the tibia. This, as noted, is concave, with a smooth, almost hairless exterior surface, provided at the sides with several rows of Jong curved hairs, which arch over either side, forming a veritable basket in which the pollen may be securely packed. As soon as the collecting combs of the tarsus are filled, the bee draws them across the strong, curved hairs of the corbicula, the right tarsus emptying into the left corbiculum and vice versa, until both are filled. These baskets or masses of pollen are emptied by means of the single strong tibial spine on each of the middle pair of legs, the spine being thrust beneath the load of pollen and used as a pry to loosen and remove it. A very remarkable peculiarity of the posterior legs, but having no con nection with the polliniferous apparatus, is seen at the union of tibia and first tarsal joint. These are articulated at the extreme anterior angles in such a manner that the broadened apex of one and the base of the other, work together as a sort of nippers or pincers. The tibia is armed on the inner margin with a strong, uniform row of short spines extending two- thirds of the way across. This apparatus is employed by the bees in re moving the wax scales trom the abdomen. 66 Riley Presidential Address. Examination of these parts in other species of Apis fails to indicate any particular modification or deviation in structure from mellifica. In Ajm indica no differences whatever can be discovered ; in A. dorsata the leg is somewhat more hairy and a few hairs occur on the outer surface of the tibia. In A. florea the smallest species known, the spines on the apex of the tibia are somewhat shorter and stouter and the hairs forming the cor- biculum are somewhat less regular in length and arrangement. This statement of the structure of these parts in the species of Apis will enable us to compare intelligently the similar parts in those genera most nearly allied to them, tracing the variation through these to the more widely divergent forms. The genera Melipona and Trigona include bees which are closest to Apis in general structure and habits, and agree also in the absence of the tibial spines of the posterior legs. We rind, as might be inferred, a very close correspondence in the polliniferous apparatus, which, in all essential details, is practically the same as in Apis. The pollen-col lecting combs on the inner surface of the first tarsal joint are absent, or rather their place is supplied by a uniform clothing' of short stiff spines which are not arranged transversely in rows, as in Apis, but serve the same purpose. This joint also differs in shape from that in Apis, by being sud denly narrowed or excavated toward the base so that the nippers noted in the former genus for the removal of the wax are practically wanting, although the row of stiff spines at the apex of the tibia is still present, but somewhat reduced. A very peculiar tuft of strong, curvet! spines occurs, in the two genera mentioned, at the anterior outer angle of the tibia. This has no counterpart in any other bees and its function is problematical. In the case of Bombus, the lowest of the social bees, there is at once a greater divergence from Apis and at the same time a resemblence to it in certain features of the hind legs and pollimferous apparatus. The tibial spines are very strongly and prominently developed, allying this genus to the solitary bees and other Hymenoptera, but the general structure of the tjbia and first tarsal joint is practically identical with that of Apis, and the tarsal joint in this particular does not present the divergence which was noted in the case of the genera Melipona and Trigona, but has the broadly truncated basal margin forming the lower blade of the nippers, even more strongly developed than in Apis. The pollen-collecting spines on the inner face of the tarsal joint are uniformly distributed over the surface, practically as in the two genera last mentioned (Melipona and Trigona). The border ing hairs of the corbiculum are somewhat stronger and more abundant, but in all essential details the structure is identical with the same in Apis. The solitary bees of the genus Anthophora, which is somewhat nearer Apis than any other, present distinct traces of the specialized polliniferous apparatus of this last. The enlargement of the tibia and of the first tarsal joint is quite marked, and the corbiculum is imperfectly indicated by the longer growth of hairs on the edge of the tibia, the face of the latter being also covered with shorter hairs. The brush or pollen comb on the inner surface of the tarsal joint is practically the same as in Bombus. The small row of spines at the apex of the tibia are entirely wanting, and the nippers at the junction of the tibia and metatarsus are not particularly noticeable; in fact this structure is not seen in any except the social bees which alone pro duce and use wax in their economy. The genus Melissodes presents a dis tinctly wider divergence from Apis, in that the hairy vestiture on the outer surface of the tibia and metatarsus is equally long and dense over the entire surface, showing little if any approach to the corbiculum, which, as we have seen in Anthophora, begins with the shortening of the hairs on the outer face of the tibia, In other particulars the bees of this genus are similar to Anthophora, and in both genera the pollen collected is carried interspersed among the hairs of the tibia and tarsus, being doubtlass emptied or combed into them from the brush of the inner surface of the first tarsal joint, and probably removed again by the same brush in storing it in their larval cells. Going still lower in the scale of bees, we find in Perdita a yet wider Social Insects. 67 divergence from Apis in the absence of any particular dilation of the tibia and metatarsus, the posterior legs being similar to the anterior members, simple in structure, and armed with long, scattered, feathered hairs, which are generally distributed over all their surface and which entangle more or less of the pollen grains. The brush of the inner surface of the metatarsus is still present, and in fact occurs in all Apidre and Andrenid;e. The genus Nomada is still less specialized, in that the legs are simple, not dilated and also practically hairless ; or rather the hairs are short and simple and have no pollen-collecting capacity. In this genus the brush of the metatarsus can hardly have any other use than to keep the body of the insect clean, as these bees are pseudo-parasitic or inquilinous and do not collect or store pollen. It is a mere modification of the normal or original structure and doubtless a degeneration due to the semi-parasitic habit. From the above review of the modification of the posterior legs as pol- liniferous organs in various genera of the family Apidye, it will be seen that there are first developed on the leg, hairs which are feathery and which will entangle the grains of pollen. The next step in the development is an in crease in the abundance of this hairy vestiture, and a further advance occurs in the widening of the tibia and first tarsal joint, to give a greater surface for the pollen-collecting, plumose hairs. This reaches its maxium in the genus Melissodes in which the external hairs of both the tibia and the metatarsus are very long and dense and the feathering very decided. The next step toward the condition found in Apis is exhibited in Anthophora, and consists in the partial disappearance and shortening of the hairs on the outer face of the tibia and metatarsus, by which means an im perfect corbiculum is formed, foreshadowing the more complex structure of the social bees, in which it becomes quite well developed in Bombus and perfectly so in Trigona, Melipona, and Apis. In Anthophora a further modification is noted in that the hairs of the legs are practically simple and unfeatherecl as in the higher social bees. In the other family of bees, the Andrenidpe, we have a similar condition of things, the variation in the pollen-collecting character of the posterior legs ranging from Agapostemon to Prosopis, and showing the same grada tions noted in the Apidse from Melissodes to Nomada. The reader interested in studying how the mouth-parts and the legs have been modified in the bees by their honey and pollen gathering habits, can not do better than consult Hermann Miillers' w T orks* on the subject. There is almost an unbroken chain of these characters, from the highly de veloped bees to such as are hardly distinguishable from the fossorial wasps. NOTE 4. Wax- producing organs. In all the wax-producing bees the specialized discs (see Fig. 3) on which the wax is deposited when secreted by the true glands beneath, occur on the basal half of the second to the fifth ventral segments of the abdomen, the overlapping half of each segment covering and protecting the disc of the succeeding segment. With the Hive Bee these discs are com pound and two in number on each segment. They are broad, ovate, pale yellow in color, smooth, delicate and transparent, and are surrounded by a narrow thickening of the chitine of the sclerite and separated by an un modified rnedio- ventral septum. This specialized structure occurs only in the workers. The queen, however, has a sub-obsolete, undivided area on the same five abdominal segments, and which in structure bears a striking resemblance to the similar area in the workers of the lower forms of social bees. The wax discs of Melipona and Trigona are practically identical, and are narrow, extending entirely across the base of the segment, not being broken, as in Apis, with a dividing septum, and also extending laterally *The Fertilisation of Flowers, by Prof. Hermann Muller. Translated and edited by D'Arcy W. Thompson, B. A,, L,ondon, 1883. 68 Riley Presidential Address. nearly to the apex of the sclerite as in the case of the fertile female in Apis. In Bombus the structure is almost identically the same as in Melipona. * NOTE 5. Ant Economy. Considering the large number of species of ants, a book would be re quired to treat of them in detail, and volumes have been written. In this note I shall only treat of a few of the better known, to supplement the mere summary in the body of the address. The most interesting of our North American species which I have had an opportunity of studying are the mound-building species of the East, the leaf-cutting species of Florida and Texas, and the honey ants of Colorado. With the aid of Mr. Th. Pergande, who has been assiduous in his studies of the family, and is per haps our best-informed myrmecologist, I have brought together a number of notes on the habits of our North American species of Carpenter Ants and others ; but they are excluded as the least important in connection with the text, and with a view of duly limiting the pages. MOUND-BUILDING ANTS. In this category may be classed by far the larger number of our better-known ants. The term is, however, particularly applicable to the species of the genus Formica. These ants are very much more active and industrious and typical of the family, than are the carpen ter ants. Our own species inhabit, by preference, pine woods. They are pugnacious and valiant, and whenever their mound is disturbed, however slightly, will speedily cover the whole surface in one surging mass, spread ing over the mound and attacking in their fury any living creature within reach. They are in fact so fierce and fearless that even man does well to avoid their mounds ; for the bite is quite severe, and when multi plied indefinitely is unbearable. The Fallow ant (Formica exsectoides Forel), one of our best known species and a close ally of F. exsecta of Europe, builds large mounds of earth, more or less mixed with other materials, especially small sticks and dried leaves of pine. These will measure all the way from two to eight feet in diameter at the base, and may be from one to three feet high. They are more or less regular and conical, full of galleries, with larger or smaller chambers which communicate with a general system of subterranean cells or cavities, which are used as store-rooms, nurseries for the young, parlors for the queens, and other purposes. The purpose of the superstructure hi most mound- build ing ants appears to be for aeration, for the more rapid development of the larvae, and, apparantly, to facilitate social intercouse between the individuals when not engaged in actual work. Except for the extrane ous matter which gives it firmness, all the material of the mound is brought up from beneath the surface, and the inhabitants are incessantly at work, night and day, in constructing, altering and repairing. Very large colonies are often connected by secondary hills. I once had a good oppor tunity of studying these mounds around Ithaca, N. Y., and Dr. H. C. Mc- Cook has published a most interesting and detailed account of his observa tions upon this ant in the Trans. American Entomological Society for 1877, Vol. VI, page 253, and also in The American Naturaliat for July, 1878, Vol. XII, pp. 431-445. It is particularly common in the Alleganies. There are three forms of workers, viz, major, minor and dwarf. His interesting ob servations will well repay reading. It is in these mound-building ants that we find the true economy of the division of labor. While large numbers are ceaselessly building and min ing, so as to keep the formicary in good condition, repairing or increasing its size, so as to accommodate the growing numbers, others are busily engaged in scouring the surrounding country for food, both for themselves, for the multitude of those who stay at home, and for the young. In these expedi tions they never hesitate to attack any other insect that may be in their way, no matter how much larger than themselves, and what they lack in power individually they make up in numbers. Still others again are run- Social Insects. 69 ning over the trees and shrubs and other plants, searching for plant-lice, from which they gather the sweet rejectamenta, gorging themselves fre quently to such an extent that ithey return home with difficulty. This honey is used chiefly for feeding the larvae. HONEY ANTS. There is really but one Honey Ant, strictly speaking, viz, Myrmecocystus melliger Llave (M. mexicanus Westm.), in North America, and this ranges from Mexico to Colorado. Other species occur in other parts of the world, with somewhat similar habits, and one is especially mentioned by Lubbock from Australia (Camponotusinflatus Lubb.) which has undergone precisely the same modifications, though belonging to a distinct genus, a most interesting fact, since it shows that the modification has arisen inde pendently. The honey collected and stored by these ants has little value commercially, first, because of its rather poor quality ; secondly, because of its small quantity barely more than half a pint to each colony obtainable ; and, thirdly, because of the difficulty of colonizing or in any way commer cially manipulating the ants. The insect must be crushed to obtain the honey. Yet it is sought for by the Mexican Indians, and used to a con siderable extent. The formicaries are little truncated cones from two to three inches high, and usually less than a foot in diameter. They have a tubular channel, a few inches in diameter, leading from the central opening to the interior, to a depth of six inches or more below the general surface. Here are often found one or more dome-like vaults or honey -chambers, about an inch deep by about three inches in width. Hanging from the roughened roof of these chambers may, at any time, be found numbers of the honey-bearers, with immensely swollen abdomens and looking, when congregated, like a series of small grapes or large currants, with the same translucency which these possess. These individuals have little capacity for movement, and indeed move but little. They are but living receptacles of the sweets which are gathered by the real workers, and the food-supply of the rest of the colony is only drawn from these stationary honey reserves, or animated honey pots, as Lubbock calls them, when necessity requires. The modifications are confined to the abdominal portion of the digestive organs. The honey is gathered from a little Cynipid oak-gall which I have described as Cynips quercus-niellaria and which abounds on a small scrubby oak (Qaercus undulata) frequent in those regions. The ants always work at night, making their way in long strings to the nearest gall-bearing tree, the branches of w T hich they carefully search for the young and succulent galls which secrete a small globule of a clear saccharine liquid. The gathered liquid is then, upon the return to the formicary, emptied into the mouths of those individuals which serve as honey stores. LEAF-CUTTING ANTS. These are represented almost solely by the genus Atta, which abounds in tropical and sub-tropical countries, where the species are dreaded by planters because of their great destructiveness to culti vated plants and trees. These ants have been denominated agricultural ants, and recent observations have confirmed the explanation originally urged by Belt, that the leaves are cut into pieces and gathered into small heaps, as a nidus for the cultivation of a fungus (Kozites) the mycelium form of some mushroom, so that they may be said to have anticipated man in this kind of culture. The only two species belonging to the genus so far observed in this country, are Altafervens Say, and Atta tardigrada Buckley. The former is our commonest species, occurring in Texas. Its formicaries are often twenty feet in diameter and several feet high, with numerous smaller moundlets scattered over the surface. They have a crater-like de pression in the top, with a central opening running down into the formicary, sometimes to a very great depth. Each formicary contains immense num bers of individuals, and during the day appears to be empty and deserted. After dark, however, the entrances are opened, first by smaller workers who remove the particles of sand and earth, then by individuals of larger form who aid in removing the refuse. When the way has been sufficiently cleared, the inmates pour forth, both workers and soldiers, and march to 70 Riley Presidential Address. some plant or other near by. They are generally seen in double column, one column ascending the plant and cutting off the leaves, and the other returning loaded to the nest. Great intelligence is shown by this ant in its foraging expeditions. The cut leaves, either whole or in circular pieces, are usually thrown on the ground by those who ascend the tree, while others below receive and bear the fodder home. Each piece of leaf is grasped by the jaws, and, with a quick motion of the head, thrown back over the head and thorax in such manner that it lodges edgewise in a deep fur row and between two spines which characterize the head, so as to cover the insect more or less and offer little or 110 obstacle to its progress. Very long underground tunnels are sometimes excavated from the main formicary to some shrub or tree so as to facilitate access thereto. The stories told by southern planters of the ravages of this insect seem almost incredible, but I have myself witnessed the utter denudation of a large tree in a single night, in which case all the forces of the formicary seemed to be concentrated on a single object. Atia tardigmda is found east of the Mississippi River, occurring through out the gulf States from Florida to Texas. In Florida what is evidently this species builds rather large cells from two to four inches in diameter in fine white sand, the walls very firm and smooth. In some instances the walls are said to be lined with a kind of curtain composed of particles of different colored sands brought up from a lower stratum and interwoven with fine white threads, by which is doubtless meant shreds of the refuse vegetation collected a kind of spongy mass, manufactured from the vege tation and somewhat resembling the comb made by certain bees. This spongy mass contains small irregular pockets, apparently designed for the reception of the young, and in this we have the nearest tendency in ants to the building of cells which is so common in some of the other social Hymen- optera. This species prefers the fine needle-like leaves of tender pine seed lings, and a row, marching in single file, each carrying a piece of one of these needles, suggests a file of soldiers armed with rifies. Atta mexicana Sin. abounds in the temperate regions of Mexico, its formi caries being twenty or more feet in diameter, and a funnel is said to extend through its center to facilitate drainage, which would seem to be necessary in a country subject to very heavy rains. The damage done by this species, especially to coffee plantations, is said to be very great. Atta ccpkalotes L. is dreaded in .Brazil because of its destructiveness to vegetation and of its tendency to enter houses and carry off the mandioca rneal. Its formicaries often reach a diameter of more than 100 feet. NKST-BUILUING ANTS. Though we have in the United States no species which constructs nests similar to those of wasps, yet such are known to occur in other parts of the w r prld, especially in tropical and sub-tropical countries. The genera Polyrhacis, Doliehoderus and Oernastogaster imitate wasps in the construction of their nests. Some of the Brazilian species of Cremastogaster construct more or less globular, black nests, about the size of a human head, fastened between the branches of trees, large numbers of which may often be noticed among the mangrove bushes bordering the shores of the ocean, and frequently so low down as to be but a few inches above high tide. Similar nests are common in the West Indies, and look very much like young nests of Eutermes. The nest of Cremastogaster arboreus Sm., found at Port Natal, Africa, is very large, measuring about fifteen inches in length, by nine inches in diameter. It is always built around a branch, resembles in texture and ap pearance the nest of our common paper wasp, Vespa macnlata, and contains thousands of the insects. (See Smith, Cat., Hym. Ins. Brit. Mus. Pt. VI. pi. XIV. We see the beginnings of the nest- building habit in some of our North American species, especially in Cremastogaster lineolata Say, which builds coverings over colonies of Aphides, the coverings composed of minute par ticles of vegetable and earthy matter firmly glued together ; or else makes Social Insects. 71 a more or less conspicuous loose nest by massing together the exuviae of the Aphides and portions of dead leaves, generally around some twig or branch. (See Practical Entomologist, Vol. II, No. 3, Dec. 1866, p. 41.) In this case the object is doubtless to prevent the robbing of the coveted sweets by other nectar loving species ; while the more elaborate nests of the tropics are for self protection and social economy, the nearest approach to these in N. A. being made by a Florida ant (Cremastogaster Ixviuscula Mayr) which makes large brown chambered nests in long grass, recalling somewhat in color and character those of Entermes. NOTE 6. Termite Economy. TRUE ROYAL PAIRS. There are many recondite phenomena con nected with the life-history of the Termites that yet remain unexplained. But all the species annually produce large numbers of male and female adults, i. e., winged individuals which are capable, normally, of reproducing. These are recognizable after the first moult by the larger thoracic segments, which bear the first indication of wing-pads. Daring flight or swarming, and the subsequent walks on the ground, no real union of the sexes has so far been observed. In fact the reproductive organs are at this period not fully developed, and it is not until a pair have succeeded in establishing themselves amid a certain number of Avorkers that the sexual organs be come functional. The wings are thrown off and at this stage these indi viduals are knoAvn as true royal pairs, the wing stumps showing in con tradistinction to the wing-pads of the larva and pupa, while their darker color otherwise distinguishes them. They are long-lived, coition taking place repeatedly. The male increases but little in size, but the abdomen of the female increases enormously with increasing fecundity. SUPPLEMENTARY KINGS AND QUEENS. The absence of a true royal pair by no means impairs the vitality and prosperity of a Termite colony; for a certain number of individuals are met Avith which, in the absence of the true queen may become sexually mature, the female laying fertile eggs, from which, in due course of time, all the forms composing the colony are developed. The true nature of these secondary or supplemantary males and females Avas first fully recognized by Fritz Miiller, and their develop ment is explained as follows : At first indistinguishable from the larvoe of individuals which produce winged specimens, they are, in the nymph or pupa state, thicker and clumsier. The internal sexual organs are more strongly developed, and they have short Aving-pacls placed sideAA'ays instead of long and broad Aving- pads as in the nymphs Avhich produce the true kings and queens. In short, they undergo one moult less, and, as a consequence, do not acquire A\ T ings or swarm. They acquire sexual maturity later in the season than the winged individuals, from AA'hich they are ahA'ays distinguished in maturity by the possession of wing-pads instead of the Aving stumps. They are also lighter in color, the males having smaller eyes, and the females a broader thorax, Avhereas in the true royal individuals there is no difference in this respect. They are not as long-lived, either, as the royal pair, the males dying within a few months and the females probably not sur\ T iving more than a year. It will be seen from the above stated facts that if through the death of a queen, or in the absence of a queen, a colony has not been able to secure another royal pair from the swarming individuals "nymph-like males and females, safely kept m the nest " step in as substitutes and save the colony from becoming extinct. Furthermore it has been observed that if, in very small and fragmentary colonies, the supplementary males and females should be absent, the colony may yet be perpetuated by the substitution of larva-like males and females, which have been called complementary kings and queens. A remarkable observation made by Fritz Muller deserves mention here. He found in a Eutermes colony, in the passages of what appeared at first to be a true royal cell, not less than 31 supplementary females and among 72 Riley Preside iitial Address. them a single true king. i. e., with distinct wing-stumps. "Instead of a royal palace " he says, "in which the king lived in chaste matrimony with his equal consort, I had a harem before my eyes in which a sultan satisfied himself with numerous coquettes." This observation would seem to indi cate that, in the economy of the Termite colony, a true king and queen may not only be replaced by" supplementary kings and queens, but that this substitution may take place for both sexes at the same time, or for each sex separately. I would observe, in this connection, that during the swarming season many species of true ants forcibly detain some of the winged males and fe males and prevent their leaving the formicary by biting off their wings, and that the pairs thus forcibly detained supply the colony with eggs. A simi lar condition may prevail among the Termites, and if so, would throw light on some of the facts which have been observed. INFLUENCE OF FOOD AND TREATMENT. The effect of food and treatment has less, perhaps, to do with the differentiation of individuals among ter mites than among the bees, wasps or ants. All Termite larvrc are supposed to partake of the same kind of food, as to the nature of which there is conflict of opinion, due doubtless to the vary ing habits in the different species. From my own observations on Termes and Eutermes, I am inclined to believe that, as in the Social Hynienoptera, the food and treatment of the young larva, during the first stage more par ticularly, have much to do in determining the development or suppression of the sexual organs, and, as a consequence, in determining the character of the full grown individual. The eg^s are, first of all, brought together in special parts of the termitary, and it is quite probable that the workers ex ercise some judgment and" discrimination in the grouping, as has been proved to be the case with Hymenoptera, with a view to future larval treat ment. Judging from the delicacy of their mouth-parts and of the general integument, the young are at first more or less dependent upon either the forethought or the direct action of the adults, and I cannot resist the con clusion that the infancy of the termites is dependent, as it is in the Social Hymenoptera, if not to the same extent; for they have soon perished where I have hatched them away from adults, and have developed where the adults had access to them. But further exact observations, which, in the nature of the case, it is difficult to make, are needed before definite con clusions can be drawn. Fritz Miiller believes that the young feed on a fungus which develops on the walls of the cells, a peculiar white fungus being not uncommon in such situations, though I have more often found nothing of the sort where the young were abundant. Mr. Hubbard found many small hard bodies among the eggs of Euierm.es rippertii which were recognized as the sclerotium of a fungus by Prof. F. G. Farlow, and other observers have referred to the presence of fungi in Ter mite nests. Mr. Hubbard also records the feeding of the young upon hard and tough rounded masses found in the nests of the above-named species. They could not do so, however, without the assistance of the nasuti or work ers to soften these nodules, for their mouthparts are too feeble, while the nodules are of very irregular occurrence and in some nests not present at all. Where the young are crowding, the material of the nest is moister than elsewhere and their chief food must be a liquid regurgitated from the mouth, by the workers or by the partly developed sexed individuals, just sis in the social Hymenoptera, and either taken directly or from the moistened substance of the cavities. Indeed, though Mr. P. H. Dudley in some interesting observations on Eutermes on the Isthmus of Panama (Journal N. Y. Micros. Soc. V. p. 62, April, 1889) describes the nasuti as being able to fire an " offensive glutinous shot, which puts an antagonist twice his size luws de combed," I have never been able to confirm this statement. The nasuti have seemed to me defenseless and I suspect that the liquid so readily secreted from the tip of the nose is chiefly designed for nourishment. That comminuted, decayed wood, as well as the faeces are Social Insects. 73 also used for food has been shown by Grass! and others, while the tendency to feed freely upon one another is matter of common record, and indeed all the dead and dying are devoured.* In Calotermes the excrement consists of dry and hard sub-ovoid particles which accumulate in the burrows, so that the faeces are not used here whether as food or to line the burrows. Consequently the young must de pend entirely on liquid from the mouths of the females. The food is, how ever, from what has gone before, sufficiently varied in those species which exhibit the greatest number of colony forms, to justify the belief, here set forth, that it has much to do in the development of those forms. It is, however, definitely known that differentiation of the sexes takes place at an early period, and can be recognized by anatomical and exter nal characters in the larva, immediately after the first moult. Freshly hatched larvae appear to be sexually undifferentiated, although it is probable, as suggested by Newman in 1853 and Hagen in 1855, that this is simply be cause the differences are too minute to be observed. Sex is doubtless de termined in the egg, but the different forms of either sex are, in all proba bility, due to food and treatment in the first larval stage, and to an innate tendency confirmed by heredity. The mode of treatment of the mother, in insects generally, may influence the sex of the offspring; but there is no evi dence to show that the sex can be altered when the egg has once passed. Fecundity varies in individuals of any community, and a certain number are always sterile. In the social insects this condition is simply controlled to the advantage of the species, and the tendency, associated with various other modifications, has been emphasized by heredity. Prof. B. Grassi (Bull. Mensuel Acad. Gioenia, 1889; Entom. Nachrichten, 1889) has of fered a rather curious explanation of the origin of the sex in Termites. He finds in the ccecum of the young larvae, as well as in the fully developed workers and soldiers, an abundance of protozoon parasites. With each moult these parasites disappear, but immediately commence to reappear, and the ccecum is inflated in a sac which presses on the sexual organs so that the development of the latter is prevented, the protozoons not appear ing, after the first moult, in those individuals which are to become truly sexual, or at least in only the smallest quantities. He bases this view upon the examination of many hundreds of individuals, but the probabilities are tiiat the presence of the protozoons has no essential part in the result, as he offers no explanation as to why they are absent or less numerous in the one case than in the other. COMPOSITION OF THK TERMES COLONY. Remembering that in Termes the adolescent stages actively participate in the work and composition of the colony, and accepting the nomenclature most recently used by the latest and best observers, the forms already indicated in the diagram on p. 33 may be enumerated as occurring in the species of the genus Termes, as ex emplified by the commoner European and American species: Prof. Grassi has enumerated some three additional forms, but this con fusing complexity of forms really occurs only among those which are re productive and they never all occur at one and the same time, while some of them only occur under certain peculiar conditions. The youngest larvae, i. e., the indistinguishable freshly hatched larvae of all forms (No. 1) are very small, in no species attaining 2 mm. in length. They are delicate, feebly chitinized creatures, blind, the thoracic segments not specialized, and with short 9-to 10-jointed antennae. After the first moult the differentiation into neuters and sexed individuals becomes appre ciable, not only in the beginnings of the development of the sexual organs, but in the increase in the number of an tennal joints. The larvae and sub- *By placing a small quantity of arsenic or calomel mixed with sugar in their burrows or nests, the termites will greedily devour the mixture, and by means of the poisoned individuals being fed on as fast as they perish, the whole colony will in time be de stroyed. 74 Riley Presidential Address. sequent stage's of the neuters remain eyeless and the thoracic segments are very little altered, since they develop no wings. But after the second moult a further differentiation takes place between the larvae of the ordinary workers and soldiers, those of the former being recognized by the email head, smaller mandibles, large maxillae and lahiuin, while those of the lat ter have a much larger head, very prominent mandibles, variously modified according to species, and much smaller maxillae and labial parts. In the perfect workers and soldiers these differences are still more strongly marked, and both forms may at once be distinguished from other larvae by the darker color and the shining and harder integuments. A peculiar form of neuter^ occurring in Euterrnes, the so-called nasuti, remained a puzzle for a long time. In this form the head is pear-shaped and prolonged anteriorly into a tube or nose which possesses a channel leading backward into the head. The nasuti have the power of secreting a viscid liquid from the tip of this nose. The mandibles are not prolonged and are unfitted for biting, while the lower mouth -parts are but little better developed than in the common soldiers. Dr. Hagen in the Appendix to his famous monograph of theTermes, recognized this form as a soldier form, characteristic of the genus Eutermes, which replnces the large-headed and mandibulate soldiers of the other genera. Mr. Hubbard, however, records having found in one colony of Eutermes nppertii in Jamaica a few of these nasuti among the soldiers (Boston Soc. Nat. Hist., 1877, pp. 270-2). It is believed, and I think justly, by Fritz M tiller that when found in colonies of other Termites having mandibulate soldiers, these nasuti are mere inquilines or intruders, and the opposite view is justifiable, that when the mandibu late soldier is found among the nasuti, it also Ls an intruder.* Acknowledgment. Figures 1, 2, 3, 8, 9, 10 and 11, are made from illustrations belonging to the Department of Agriculture, and are used by the kind permission of Chas. R. Dabney, Jr., Assistant Secretary of Agriculture. *Since this address was written, I have had an opportunity of studying Eutermes in the West Indies, E. morio, at St. Thomas, St. Kitts, Monserrat, Dominica, Mar tinique, St. Lucia and Barbados, and both it and E. rippettii in Jamaica. The nasuti are here the smallest individuals in the colony and also somewhat the darkest. They have no power of biting, and no organ of offense, as the liquid exuded from the tip of the nose has no pungent property. They may, therefore, be handled with perfect impunity Of some forty nests examined none have furnished a mandibulate soldier. The nasuti, though having no weapon of offense (so far at least as man is concerned) are nevertheless active guards, and undoubtedly take the place of the soldiers in Termes proper. They crowd around the queen, when the colony is disturbed, and rush to the outside and about the borders of any breakage or hole made in the nest or the tunnels thereto. They thiow up the head and play the antennee and palpi in a comically threatening way, considering their inoffensiveness, and they watch around the borders on the inside of such breakage while the workers run up rapidly now and again to deposit the soft excrement which is to mend the gap, and of which the tunnels and nests are for the most part formed. Kggs and young larvze are frequently borne on the nose and on the feelers of these nasuti; but I have not yet satisfied myself that they are thus purposely carried, and are not accidentally stuck by the exuding liquid, the latter view comporting best with most of the cases. But that these nasuti perform some function in the economy of the colony other than that of soldiery defence, is rendered almost certain by their relatively large numbers com pared with the real soldiers in Termes, for they are generally as numerous as the man dibulate workers and sometimes as numerous as all the other individuals together. While the liquid from the nose may be used in cementing the walls of the tunnels, I am inclined to believe that it is of more importance in furnishing the first pabulum of the young. Eutermes rippertii differs little from E. morio in habit except that the hard, paler nodules generally found in its older nests do not occur in those of the latter. But the most interesting experience, which is born out by the observations of Mr Dudley on the species in Panama, is that I have found as many as nine queens in one nest and often three or four. In fact there is every variation, even in independent nests which appar ently have no accessory mother-nest, from those without queen to those with one up to nine (or more according to Dudley), while in one nest I found scores of true royal pairs in which the queens had undergone no material enlargement. I have also found either no male or sometimes two and once three males associated with a single queen. Ordinarily, however, there is but a pair, i. e., one queen and her escort. VOL. IX, pp. 75-88 APRIL 9, 1894 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON FOSSIL CYCADEAN TRUNKS OF NORTH AMERICA, WITH A REVISION OF THE GENUS CYCADEOIDEA BUCKLAND. BY LESTER F. WARD. The recent discovery of a large number of fossil cycadean trunks in the Cretaceous rim of the Black Hills, has furnished a new stimulus to the study of these forms in America. Such objects have been found at no less than six distinct North Ameri can localities. The oldest and best known forms are those first mentioned by Tyson from the Lower Cretaceous of Maryland. Dr. Emmons found one such in the Trias of North Carolina, and Sir Wm. Dawson another in the Trias of Prince Edward Island. All the rest, with one exception, are from Cretaceous strata, the age probably not widely differing from that of the Maryland specimens. These are from the Trinity division of the Comanche group in Southern Kansas, and from two localities among the foot-hills outside of the Red Beds of the Black Hills region in South Dakota. The exception to this is the Cycadeoidea mira- Mlis (Lx.) Solms (Zamiostrobus mirabilis Lx.), found on the surface of the ground by Dr. F. V. Hayden, near Golden, Colo rado, within the Laramie, or Post-Laramie terrane. This locality is at the foot of the Front Range, and it would have been very easy for an erratic block to be borne down the mountain side and lodged in the valley where this was found. As is well known, older formations are encountered on ascending the eastern slope 76 Ward Fossil Cycadean Trunks of of the Rocky Mountains, and Cretaceous and Jurassic strata un doubtedly crop out immediately above this locality. Early in the spring of 1893, the National Museum obtained possession of a collection of six fine cycadean trunks from parties residing at Hot Springs, South Dakota, who had collected them at that vicinity.* One of these specimens measures thirty- one inches in height and twenty-four in greatest diameter, and weighs nine hundred pounds ; the others are comparatively smaller, the smallest of all not exceeding a foot in height. Most of them are considerably flattened, but one or two are nearly cir cular in cross section. One of them exhibits a number of lateral branches, and in most cases the apex is depressed, forming the "crows nests" so characteristic of the specimens from the Isle of Portland, Dorsetshire, England. In the Geology of the Black Hills, prepared by Profes sors Newton '-and Jenney, from their survey of 1875, and published at Washington in 1880, none of the Cretaceous strata below the Dakota group of Meek and Hayden, are recognized; and while I presumed from the general his tory of this class of vegetation that these remains came out of the Triassic Red Beds, or the overlying Jurassic, I was still so greatly interested to ascertain their true source that early in September last I made an expedition to the region, and in coop eration with Professor Jenney discovered the locality and made further collections, including one very much branching and- very large trunk and many interesting fragments. All the re mains of this class that have been thus far found in the southern part of the Black Hills, occur in the area mapped as Dakota group by Professor Newton, and, although no cycadean vege tation had yet been found amidst the extensive collections from the Dakota group of Kansas, Nebraska, and other more eastern localities, we were at first disposed to accept this as proof of their occurrence at that horizon in this region. But the great im probability of this assumption led us to make a careful examina tion of the series that had been thus classed by Professor New ton. The result was that we came to the conclusion that the Dakota group of Newton is much more extensive than the No. 1 of Meek and Hayden, and while the upper portion of it cer- *See Science, Vol. XXI, New York, June 30, 1893, p. 355. North America. 77 tainly belongs to the true Dakota, the lower portion very proba bly extends to near the base of the Cretaceous. The evidence upon which these conclusions rest will soon be published, and it need only be added that the cycadean trunks belong to this lower portion though not very near the base and may not differ greatly in age from those found in Southern Kansas and Maryland. At another part of the Black Hills region, within the foot hills on the Eastern side, some six or eight miles north of Rapid City, and between that place and Piedmont, and also probably in the Cretaceous area, two other specimens have been found and are now at the State School of Mines at Rapid City. The where abouts of these specimens was not known at the time that I vis ited that section, but since my return Professor Jenney, ap pointed at about that time Dean of the Faculty of the State School of Mines, has discovered them there and has furnished me the data for tne brief description given below, together with rough drawings, and measurements, From him I learn that in 1877, Mr. J, M. Leedy, then of Rapid City, now residing in Florida, found these specimens at the place stated, that they re mained at his ranch for some time, were then placed on exhibi tion at a fair held at Library Hall, and not being supposed to have any value, were subsequently thrown out into a vacant lot, where they remained until removed to the School of Mines. These forms are much more cylindrical than those found in the southern section, and seem without doubt to constitute a new species. I have therefore named this species Cycadeoidea Jen- ncyana, in commemoration of Professor Jenney's great services to the people of that section as well as to science in general. I have not seen Professor Cragin's specimens from Southern Kansas, and he unfortunately did not figure them, but he stated in his description that they very closely resemble the Maryland specimens, of which he had obtained a photograph and had learned some particulars as to size. While he thought these two forms were specifically identical, it is probably best to let them remain as distinct species for the present. All of our American forms appear to belong to the genus Cycadeoidea of Buckland. None of the taller, more slender, palm-like, or branching trunks, belonging to the Old World genera Bucklandia and Cylindropodium, have yet been discovered this side of the Atlantic. The genera Fittonia, Yatesia, and 78 Ward Fossil Cycadean Trunks of North America. Platylepis, in which the leaf -bases are persistent, seem also to be absent. I have therefore made a careful revision of the genus Cycadeoidea condensing into it the Bolbopodium and Clathropo- dium of Saporta, and also referring to it all the species of Ben- nettites of Carruthers. The greater part of all this had already been done by the recent researches of Count Solms-Laubach, and it only remained to pick up a few of the outlying forms that did not come within the purview of his studies. If his results are accepted at all there is no logical stopping-place short of the embodiment of all these forms under the genus Cycadeoidea. It is of course possible that future exhaustive study, especially from the standpoint of internal structure, may result in the subdivi sion of this genus into several. But at present the tendency is toward consolidation, and a great uniformity is found in both the external and internal characteristics of the extinct Cycadaceae. In a much more extended paper, which is now in preparation ? I hope to bring out the special characteristics of our American forms and to compare them with those of the Old World. Sec tions are now being made of some of the specimens from the Black Hills, and it is proposed to illustrate the internal structure of these specimens as fully as possible. Prof. F. H. Knowlton has consented to superintend the work of section cutting and to prepare the part of this paper relating to internal structure. Thus far we are in possession only of the Black Hills material and the single specimen of C. miraMlis described by Lesquereux in his Tertiary Flora. This specimen was loaned several years ago to count Solms who made sections of it and prepared several slides, duplicates of which he sent back with the specimen. I also have a somewhat careful description of what he found, not only in letters received from him, but also in his memoir on the fossil cycads of Italy. Should other material come into our hands it will also be treated from the same standpoint. I have endeavored in all cases to conform strictly to the law of priority now so rigidly enforced in all departments of natural history. I have been careful to give dates, so that the reasons for the deviations from the more current designations may be clear. If I have made any mistakes in this respect I shall be very thank ful to receive corrections before the final paper is completed, this being one of the objects of this preliminary one. Species. 79 REVISION OF THE GENUS CYCADEOIDEA BUCKLAND. Genus Cycadeoidea Buckland. 1827. Cycadeoidea Buckland, Proc. Geol. Soc, Lond., Vol. 1, No. 8, pp. 80-81 (Session of June 6, 1827). 1828. Cycadeoidea Buckland, Trans. Geol. Soc. Lond., 2 Ser., Vol. II, pp. 375-401, pi. xlvi-xlix. This genus seems to be the ultimate destiny of all cycadean trunks of dwarf bulb-like or conical form, deciduous leaf stalks and rhombic leaf scars. Count Solms-Laubach has already re ferred many of the species of Bennettites and Clathropodium to it, and the Marquis Saporta admits that one species of Bolbopo- dinm belongs to the same genus as C. pygmcea. The fact alone that fruit has been found in one species (C. Oibsoni) seems in sufficient ground for retaining the genus Bennettifces. The only other name that has any claim to retention for this group is Mantellia of Brongniart, but his publication of it at the same date with Buckland's Cycadeoidea was a nomen nudum, and had moreover been used for an animal fossil. It is therefore gener ally given up. Brongniart himself conceded this, but wrote Cycadoidea on grounds of euphony. Even this cannot be al lowed by the now more and more strictly enforced rules of no menclature, and Cycadeoidea must stand as originally written by Buckland. Cycadeoidea megalophylla Buckland. 1827. Cycadeoidea megalophylla Buckland, Proc. Geol. Soc. Lond., Vol. I, No. 8, p. 80. 1828. Trans. Geol. Soc. Lond., 2d Ser., Vol., II, pp. 397-401 , pi. xlvii, figs. 1-4; pi. xlviii, figs. 1-3. 1828. Mantellia nidiformis Brongn., Prodrome, pp. H6, 199. 1837. Mantellia megalophylla (Buckl.) Bronn, Lethaea Geognostica, p. 227, pi. xv, fig. 2. 1837. Cycadites megalophyllus Buckland, Geology and Mineralogy, etc., Vol. I, p. 497; Vol. II, p. 98; pi. Ix, figs. 1, 2. 1838. Zamites megalophyllus (Buckl.) Presl, in Sternberg's Versuch, etc., Vol. II, Hefte 7 and 8, p. 196. 1842. Eacephlartos Bucklandii Miquel, Monogr. Cycad., p. 60. 1849. Echinostipes nidiformis (Brongn.) Pomel, Materiaux, etc., p. 1H. 1874. Clathropodium megalophyllum (Buckl.) Saporta, PL Jurass., Vol II, p. 285, Ipl. Ixxvi, fig. 1. Purbeck beds, s _Isle of Portland, Dorsetshire, England. 80 Ward Fossil Cycadean Trunks of North America. Cycadeoidea microphylla Buckland. 1827. Cycadeoidea microphylla Buckland, Proc. Geol. Soc. Lond., Vol. I, p. 81. 1828. Trans. Geol. Soc. Lond., 2d Ser., Vol. II, pp. 398-401, pi. xlix, figs. 1, 2. 1834. Strobilites Bucklandii L. & H., Foss. Fl. Gt. Brit., Vol. II, p. 1H3, pi. cxxix. 1837. Mantellia microphylla (Backl.) Bronn, Lethaea Geognostica, p. 227. 1837. Cycadites microphyllus Buckland, Geology and Mineralogy, etc., Vol. I, pp. 497, 498 ; Vol. II, pp. 98, 99, 100, pi. Ixi, figs. 1-3 ; pi. Ixii, fi<^s. 2, 3. 1838. Zamites microphyllus (Buckl.) Presl, in Sternberg's Versuch, etc., Vol. II, Hefte 7 and 8, p. 396. 1849. Echinostipes microphyllus (Buckl.) Pomel, Materiaux, etc., p. 16. 1874. Clathropodium microphyllum (Buckl.) Sap., PL Jurass., Vol. II, p. 284. Purbeck beds, Isle of Portland, Dorsetshire, England. Morris gives as locality for SfirobilUes Bucklandii, not stated by Lindley and Hutton, the Upper Greensand of Wiltshire, and Presl says that the species is also found in the Lower Lias of Lyme Regis. Cycadeoidea pygmaea L. & H. 1835. Cycadeoidea pygmrca L. & H., Foss. Fl. Gt. Brit., Vol. II, p. 175, pi. cxliii. 1841. Zamites pygmseus (L. & H.) Morris, Ann. and Mag. Nat. Hist., Vol. VII, p. 116. 1849. Echinostipes pygmreus (L. & H.) Pomel, Materiaux, etc., p. 17. 1870. Mantellia pygmrea (L. & H.) Carruthers, Trans. Linn. Soc. Lend., Vol. XXVI, p. 703. Lower Lias of Lyme Regis, England. Pomel thought he recognized the species in his material from France, but this may have been C. Pictaviensis. Cycadeoidea Saxbyana (R. Brown) Morris. 1851. Cycadites Saxbyanus R. Brown, Proc. Linn. Soc. Lond., Vol. II, p., 180. 1854. Cycadeoidea Saxbyana (R. Brown) Morris, Cat. Brit. Foss., 2d ed., p. 7. 1867. Bennettites Saxbyi Carruthers, Brit. Assoc. Rep., 37th Meeting, Pt. II, p. 80. 1870. Bennettites Saxbyanus (R. Brown) Carruthers, Trans. Linn. Soc. Lond., Vol' XXVI, pp. 681, 698, 706, pi. Ivii, figs. 1-8. Wealden of Brook Point, Isle of Wight, England. Cycadeoidea Gibsoni Carruthers sp. 1867. Bennettites Gibsoni Carr., Brit. Assoc. Rep., 37th meeting, Pt. II, p. 80. 1870. Bennettites Gibsonianus Carr., Trans. Linn. Soc, Lond., Vol. XXVI, pp. 681, 700, pi. Iviii, figs. 1-5 ; pi. lix, figs. 1-9 ; pi. lx. figs. 1-12. Lower Greensand of Luccomb Chine, Isle of Wight, England. Species. 81 Cycadeoidea Portlandica (Carr.) Solms. 1870. Bennettites Portlandicus Carr., Trans. Linn. Soc. Lond., Vol. XXVI, pp. 681, 700, 707, pi. Ixi, figs. 1-5. 1892. Cycndeoidea Portlandica (Carr.) Solms, Mem. Accad. Sci. 1st. Bologna, Ser. V, Tom. II, p. 187. Lower Purbeck beds, Isle of Portland, England. Cycadeoidea maxima (Carr.) Solms. 1870. Bennettites maximus Carruthers, Trans. Linn. Soc. Lond., Vol. XXVI, pp. 6S1, 699. 1892. Cycadeoidea maxima (Carr.) Solms, Mem. Accad. Sci. 1st. Bologna, Ser. V, Tom. II, p. 187. Wealden of Shanklin, Isle of Wight, England. Cycadeoidea Carruthersi. 1870. Mantellia intermedia Carruthers, Trans. Linn. Soc. Lond., Vol. XXVI, pp. 681, 702, 708, pi. Ixiii, figs. 4, 5. 1874. Cycadeoidea intermedia (Carr. ) Schimp. (non Ranzani) , Paleontologie Vegetale, Vol. Ill, p. 556. Lower Purbeck beds, Isle of Portland, England. The name C. intermedia being preoccupied by Ranzani in 1836 (see below) it was necessary to change it. Cycadeoidea Peachii (Carr.) Solms. 1*67. Bennettites Peachii Carruthers, Brit. Assoc. Rep., 37th meeting, Pt. II,. p. 80. 1870. Bennettites Peachianus Carruthers, Trans. Linn. Soc. Lond., Vol. XXVI, pp. 681, 700, 707, pi. Ixii, figs. 1, 2. 1892. Cycadeoidea Peachii (Carr.) Solms, Mem. Accad. Sci. 1st. Bologna, Ser. V, Tom. II, p. 187. Coral Rag of Helmsdale, Sutherlandshire, Scotland. Cycadeoidea inclusa (Carr.) Schimper. 1870. Mantellia inclusa Carruthers, Trans. Linn. Soc. Lond., Vol. XXVI, pp. 681, 703, 708, pi. Ixiii, figs. 2, 3. 1874. Cycadeoidea inclusa (Carr.) Schimper, Paleontologie Vegetale, Vol. Ill, p. 556. Lower Cretaceous of Potton, Cambridgeshire, England. Cycadeoidea Bucklandi Corda sp. 1845. Zamites Bucklandi Corda, Beitr. z. Flora der Vorwelt, pp. 38, 120, pi. xvii, figs. 1-10. Locality and formation unknown. Corda says that the specimen prob ably came from England. It resembles C. Saxbyana. Cycadeoidea Morieri Renault sp. 1887. Clathropodium Morieri Renault, Bull. Soc. Linn. Normand.,4eSer,, Vol. I, pp. 143-151, pi. iv, v. Purbeck beds, Isle of Portland, England, 82 Ward Fossil Cycadean Trunks of North America. Cycadeoidea forata (Sap.) Solms. 1875. Clathropodium foratum Saporta, PI. Jurass., Vol. II, p. 297, pi. cxxiv, figs. 1, 2. 1892. Cycadeoidea forata (Sap.) Solms, Mem. Accad. Sci. 1st. Bologna, Ser. V, Tom. II, p. 190. Gault of Cauville near Havre, France. Saporta's original supposition that this form came from the Oolite of Mans (Sarthe) was subsequently found to be erroneous. Cycadeoidea Pictaviensis (Longuemar) Saporta, ms. 1870. Cycadeoidea Pictaviensis (Longuemar) Saporta, ms., inSchimper: Paleontologie Vegetale, Vol. II, p. 188; Atlas, pi. Ixxi, fig. 12. 1870. Araucaria Pictaviensis Longuemar, Et. geol. et agron. sur le depart. de la Vienne, Vol. I, p. 491. 1874. Bolbopodiurn Pictaviense (Longuemar) Saporta, PL Jurass., Vol. II, p. 258, pi. cxviii, fig. 2. Upper Oxford of Montanaise near Poitier (Vienne), France. Cycadeoidea Sarlatensis Saporta sp. 1849. Cycadeoidea sp. Brongniart, Tableau, p. 59. 1875. Clathropodium Sarlatense Saporta, PL Jurass., Vol., II, p. 293, pi. cxxiii, figs. 1, 2. Upper Jurassic of Sarlat (Dordogne), France. Cycadeoidea Trigeri Brongniart. 1B49. Cycadeoidea Trigeri Brongniart, Tableau, p. 59. 1849. Cyca dites Trigeri Brongn. ms., cf. Saporta, PL Jurass., Vol. II, p. 288. 1849. Echinostipes sp. Pomel, Materiaux, p. 17. 1874. Clathropodium Trigeri (Brongn.) Saporta, PL Jurass., Vol. II, p. 288, pi. cxxii, figs. 1-3. Upper Jurassic of Mans (Sarthe), France. Cycadeoidea micromera Saporta sp. 1874. Bolbopodium microinerurn Saporta, PL Jurass., Vol. II, p. 262, pi. cxviii, fig. 1. Corallian of Tonnerre (Yonne), France. Cycadeoidea Mamertina Crie sp. 1879. Bolbopodium Mamertinum Crie, Les Anciens Climate et les Flores Fossiles de POuest de la France, pp. 15, 18. Bathonian of Mamers (Sarthe), France. Cycadeoidea Montiana Capellini & Solms. 1755. Lapideorum balanorum insignis congeries Monti, Bonon. Sci. et Art. Inst. at. Acad. Comment., Tom. Ill, p. 323, tav. fol. 1892. Cycadeoidea Montiana Capellini & Solms, Mem. Accad. ScL 1st, Bologna, Ser. V, Tom. II, pp. 169, 181, 214, pi. iii, fig. 1. Rio della Cavaliera, Bolognese, Italy. Cretaceous ? Species. 83 Cycadeoidea intermedia Ranzani. 1836. Cycadeoidea intermedia Ranzani, Resoconto Accad. 1st. di Bologna, 23a Sess., 26 maggio 1836. 1839. Nov. Corn. Acad. Sci. Inst. Bonon., Tom III (Bull. Sci. Med., Vol. I), p. 385, tab., figs. 2, 3, 5. Fiume Reno, Bolognese, Italy. Cretaceous? Cycadeoidea Scarabellii (Mgh.) Cap. & Solms. 1854. Mantellia? Scarabellii Meneghim, Ann. dell Universita Toscana, Tom. Ill, p. 74, nota 14. 1892. Cycadeoidea Scarabellii (Mgh.) Cap. & Solms, Mem. Real. Accad. Sci. 1st. Bologna, Ser. V, Tom, II, pp. 170, 171, 176, 181, 207, 214, pi. iii, fig. 3. Fiume Santerno, Iniolese, Italy. Cretaceous? Meneghini maintained that this species belonged to the Miocene in which it was found, but Capellini does not doubt that, like most of the other Cycadean trunks of Italy, it was redeposited from the argillaceous shales of the underlying Cretaceous. Cycadeoidea Pirazzoliana Massalongo, ms. 1858. Cycadeoidea Pirazzoliana Massalongo, ms. 1892. Mem. Real. Accad. Sci. 1st. Bologna, Ser. V, Tom. II, pp. 171, 176, 181, 208, 212, pi. ii, fig. 1. Torrente Correcchio, Imolese, Italy. Cretaceous ? Cycadeoidea Veronensis Massalongo. 1858. Cycadeoidea Veronensis Massalongo, Atti d. R. 1st. Veneto, Ser. 3a, Tom. Ill, Venezia, p. 816. 1859. Syllabus PI. Foss. Agri Veneti, pp. 20, 132. 1892. Mem. Real. Accad. Sci. 1st. Bologna, Ser. V, Tom. II, pp. 173, ]81, 206. In the garden Feruzzi-Malagnini, wall of the Padri in Verona, artificially so placed. Original source unknown. The specimen was discovered in this position by Massalongo and Scarbelli in 1858 mingled with stalactites and other objects. Capellini states that it was not mentioned in print until 1859 in the Syllabus on page 20, and seems not to have been aware that Massalongo embodied it under the same name in his " Elenco dei modelli di piante fossili donati al R. Istituto Veneto," published in 1858 in the Atti, Ser. 3, Tom. Ill, on page 816. He also includes it, along with C. Bianconiana, in the "Elenchus Specierum Vegetalium et Animalium Fossilium," etc., placed at the end of the Syllabus (see p. 132). Cycadeoidea Bianconiana Massalongo. 1859. Cycadeoidea Bianconiana Massalongo, Syllabus PL Foss. Agri. Veneti, p. 132. 1892. Mem. Real. Accad. Sci., 1st. Bologna, Ser, V, Tom. II, pp. 172, 181, 205, pi. ii, fig. 2. Torrente Samoggia, Bolognese, Italy. Cretaceous ? Capellini seems not to have observed Massalongo's record of this plant in his Syllabus, p. 132. He there says : "Ex form, ignota agri Bononiensis. Caudex." 84 Ward Fossil Cycadean Trunks of North America. Cycadeoidea Cocchiana (Caruel) Solms. 1870. Rauineria Cocchiana Caruel, R. Com. Geol. Ital. Bol., Vol. I, pp. 183, 186; figs, on p. 186. 1892. Cycadeoidea Cocchiana (Caruel) Solms, Mem. Real. Accad. Sci. 1st. Bologna, Ser. V, Tom. II, pp. 174, 181, 206, 215, pi. v, figs. 2, 5. Torrente Marnia in Valdarno, Italy. Cretaceous? The specimens were found erratic in the Pliocene. Cycadeoidea Maraniana (Scarab.) Solms. 1875. Bennettites Maranianus Scarabelli, ms. 1892. Cycadeoidea Maraniana (Scarab.) Solms, Mem. Real. Accad. Sci. 1st. Bologna, Ser. V, Tom. II, pp. 176, 179, 181, 201, 212, 214, pi. ii, fig. 3 ; pi. iii, fig. 4. Castel S. Pietro and Torrente Correcchio, Imolese, Italy. Creataceous? Cycadeoidea Capelliniana Solms. 1879. Cycadacea specie Ferreti, Atti Soc. Ital. Sci. Nat., Vol. XXI. p. 832. 1892. Cycadeoidea Capelliniana Solms, Mem. Real. Accad. Sci. 1st. Bologna, Ser. V, Tom. II, pp. 174, 181, 207, 212, 214, 215, pi. i, figs. 3, 4; pi. v, figs. 1, 3, 6. Fiume Idice, Bolognese ; Torrente Tresinaro presso Scandiano ; Paullo nel Regiano; Vallestra, Regiano, Italy. Cretaceous? Cycadeoidea Masseiana Cap. & Solms. 1890. Raumeria Masseiana Capellini, Mem. Real. Accad. Sci. 1st. Bologna, Ser. IV, Tom. X, pp. 446, 450, pi. ii. 1892. Cycadeoidea Masseiana Cap. & Solms, Mem. Real. Accad. Sci. 1st. Bologna, Ser. V, Tom. II, pp. 165, 168, 175, 178, 181, 205, 212, pi. i, fig. 1. Cretaceous (Cenomanian?) clay shales of the Idice Valley, near the Villa di Ozzano in Emilia, Italy. Cycadeoidea Etrusca Cap. & Solms. 1892. Cycadeoidea Etrusca Cap. & Solms, Mem. Real. Accad. Sci. 1st. Bologna, Ser. V, Tom. II, pp. 177, 181, 204, 212, 214, 215, pi. i, fig. 2 ; pi. iv, fig. 1 ; pi. v, figs. 7, 8. Etruscan Necropolis of Marzabotto, Bolognese, Italy. Cretaceous? Original source unknown. Specimen found placed on a tomb as an orna ment or symbolic rite by the ancient inhabitants. It is the largest of the Italian specimens. Cycadeoidea Ferretiana Cap. & Solms. 1892. Cycadeoidea Ferretiana Cap. & Solms, Mem. Real. Accad. Sci. 1st. Bologna, Ser. V, Tom. II, pp. 178, 181, 209. Monte Babbio, Regiano, Italy. Cretaceous ? Cycadeoidea Imolensis Cap. & Solms sp. 1892. Cycadea Imolensis Cap. & Solms, Mem. Real Accad. Sci. 1st. Bologna, Ser. V, Tom. II, pp. 176/181, r ^200. Fiume Santemo? Ijnolese, Italy, Cretaceous? Species. 85 I have not hesitated to place this species in the genus Cycadeoidea be cause Count Solms gives as his only reason for not doing so that the speci men was too imperfect to be certain that it belonged there. He therefore created a new genus (Cycadea) for its reception. Such a course is certain to lead to great confusion. New genera should only be created where the material is so abundant and complete that it can be adequately character ized. This new genus is not even described, and as he admits, could not be from his specimen. It therefore can have no existence. On the other hand the large number of specimens found in Italy, all referable to Cycadeoidea make it altogether probable that this also belongs there. The only other course would be to hold it entirely in reserve. This he has not done but has given it a specific name. Cycadeoidea sp. indet. Cap. & Solms. 1892. Cycadeoidea sp. indet. Cap. & Solms, Mem. Real. Accad. Sci. 1st. Bologna, Ser. V, Tom. II, pp. 176, 181. Fiume Santerno ? Imolese, Italy. Cretaceous ? Cycadeoidea Reichenbachiana (Gopp.) Cap. & Solms. 1755. Vegetabilische Versteinerung Walch, Knorr's Petrefacten Samm- lung, Text, p. 150; Atlas, pi. iiia, fig. 6. 1844. Eaumeria lieichenbachiana Goppert, in Wimmer : Flora von Schlesien, Ed. TI, Vol. II, p. 217. 1853. Jubilaums-Denkschr. d. Schles. Ges. f. vat. Cult., 1853, pp. 262, 265, pi. viii, figs. 4-7 ; pi. ix, fig. 1. 1892. Cycadeoidea Reichenbachiana (Gopp.) Cap. & Solms, Mem. Real. Accad. Sci. 1st. Bologna, Ser. V, Tom. II, pp. 186, 187, 188. Lednice near Wieliczka, Galicia. This is the large and now celebrated specimen in the Dresden Museum. Its geologic age is still unknown, but is almost certainly not Permian as conjectured by Geinitz. Cycadeoidea Schulziana (Gopp.) Cap. & Solms. 1844. Raumeria Schulziana Goppert, in Wimmer : Flora von Schlesien, Ed. II, Vol. II, p. 217. 1853. JubMaums-Denkschr. d. Schles. Ges. f. vat. Cult., 1853, pp. 259; 264, pi. vii, figs. 1-5 ; pi. viii, figs. 1-3. 1892. Cycadeoidea Schulziana (Gopp.) Cap. & Solms, Mem. Real. Accad. Sci. 1st. Bologna, Ser. V, Tom. II, pp. 186, 187. Klodnitz Canal near Gleiwitz, Silesia ; formation unknown. Cycadeoidea Schachti (Coem.) Cap. & Solms. 1867. Cycadites Sehachti Coemans, Mem. Cour. des Savants Etrangers de 1'Acad. Roy. de Belgique, Vol. XXXIII, No. 3, p. 7, pi. iii, figs. 1, 2, 5. 1870. Clathraria Schachti (Coem.) Schimper, Paleontologie Vegetale, Vol. II, p. 212. 1^70. Bennettites Schachti (Coem.) Carruthers, Trans. Linn. Soc. Lond., Vol. XXVI, p. fi99. 1892. Cycadeoidea Schachti (Coem.) Cap. & Solms, Mem. Real. Accad. Sci. 1st. Bologna, Ser. V, Tom. II, p. 187. Gault of La Louviere, Hainaut, Belgium. 86 Ward Fossil Oycadean Trunk* of North America. Cycadeoidea Marylandica (Font.) Cap. & Solms. 18GO. Cycas sp. Tyson, First Report State Agric. Chem. Maryland, p. 42. 1870. Bennettites sp. Carruthers, Trans. Linn. Soc. Lond., Vol. XXVI, p. 708. 1879. Cycadeoidea sp. Fontaine, Am. Journ. Sci. 3dSer., Vol. XVII, p. 157. 1889. Tysonia Marylandica Fontaine, Flora of the Potomac Formation, p. 193, pi. clxxiv-clxxx. 1892. Cycadeoidea Marylandica (Font.) Cap. & Solms, Mem. Real. Accad. Sci. 1st, Bologna, Ser. V, Tom. II, pp. 179, 180, 18>. Potomac formation (Lower Cretaceous) at various points in Maryland, chiefly along the Baltimore and Ohio Railroad between Washington and Baltimore and in the vicinity of the latter city. Cycadeoidea Emmonsi Font. sp. 18-)7. Trunk of a cycad Emmons, American Geology, Vol. VI, pp. 123, J24; fig. 92a. 1883. Zamiostrobus Emmonsi Fontaine, Older Mesozoic Flora, p. 117, pi. lii, fig. 5. Upper Trias of North Carolina, exact locality not known. Judging from the excellent figure of Dr. Emmons, of which that of Professor Fontaine is not a true reproduction, I consider it much more probable that this was a "trunk of a cycad" than that it was a strobile. Cycadeoidea mirabilis Lx. sp. 1876. Zamiostrobus mirabilis Lx., Bull. U. S. Geol. and Geogr. Surv. Terr., Vol. 1, 2d Ser., No. 5, p. 383 (issued January 8, 1876); Hayden's Ann. Rep. U. S. Geol. and Geogr. Surv. Terr, for 1874, p. 309. 1878. Tertiary Flora, p. 70, pi. Ixiii, figs. 1, la. 1884. Nelumbium sp. James. Science, Vol. Ill, p. 434. 1884. Clathropodium 'mimbilie (Lx.) Ward, Science, Vol. Ill, pp. 532 533. 1890. Bennettites mirabilis (Lx.) Solms, in litt. (Sept. 10). 1892. Cycadeoidea Zamiostrobus Solms, Mem. Real. Accad. Sci. 1st. Bologna, Ser. V, Tom. II, pp. 210, 211. Found lying on the surface of the ground near Golden, Colorado, in the Lararnie terraue, but probably belonging to a more ancient formation from which it had been transported. Cycadeoidea munita Cragin. 1889. Cycadeoidea munita Cragin, Bull. Washburn College Lab. Nat. Hist., Topeka, Kansas, Vol. II, No. 10, pp. 65, 66. Cheyenne Sandstone, Trinity Division of the Comanche Series (Lower Cretaceous), at Cheyenne Rock, Belvidere, Southern Kansas. Cycadeoidea Dacotensis McBride sp. Bennettites Dacotensis McBride, American Geologist, Vol. XII, p. 249, pi. xi, figs. 1, 2; Bull. Lab. Nat. Hist. State Univ. of Iowa, Vol. II, No, 4, p, 391, pi. xii, figs. 1, 2. Species. 87 Lower Cretaceous strata, valley of Minnekahta Creek near Minnekahta Station of the Burlington and Missouri Railroad, Fall River County, South Dakota (Black Hills). Cycadeoidea Jenneyana n. sp. Trunks cylindrical-conical, 15 to 17 inches in diameter and 2 to 3 feet high with concave depression ("crow's nest") at the summit; cross section of leaf stalks very irregular, rhombic or trapezoidal, two of the angles often very acute or prolonged indicating wings, the other angles obtuse. Divide between Box Elder Creek and Elk Creek, six or eight miles north of Rapid City, South Dakota (Black Hills). Formation not yet determined but probabJy same as last. The above description and data as to location are taken from letters re ceived from Prof. W. P. Jenney, Dean of the Faculty of the State School of Mines at Rapid City where the specimens now are. There are two specimens, one of which shows the summit but lacks the basal portion and is 21 inches high and 15 inches in diameter at the lower end. The other shows the base but not the summit, is 17 inches in diameter and quite cylindrical, but truncated at the height of 16 inches. This form clearly indicates that the species at least is distinct from the last and it is possible that when better material is discovered it may require to be referred to some of the less dwarfted genera, such as Bucklandia or Cylmdropodium. The distinction is further emphasized by the difference in the shape of the leaf bases or perforations left by their disappearance. I have named the species for Professor Jenney to whose assistance I am so greatly indebted in determin ing the geological position of the fossil plant beds in the southern portion of the Cretaceous rim of the Black Hills, a region which scientifically he has made his own. Cycadeoidea Abequidensis Dawson. 1871. Cycadeoidea Abequidensis Dawson, Geol. Struct. Prince Edward Island, p. 45, pi. iii, fig. 29. Trias of Gallas Point, Prince Edward Island. Sir Wm. Dawson referred this deposit doubtfully to the Lower Trias, but some regard it as the equiva lent of the Newark System. VOL. IX, pp. 89-97, PL. I MARCH 30, 1894 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON NOTE ON SOME APPENDAGES OF THE TRILOBITES, BY CHAS. D. WALCOTT. The results of Mr. W. S. Valiant's long search for the ap pendages of trilobites have recently been made known by Mr. W. B. Matthew, who described the material sold to the Columbia College of New York by Mr. Valiant.* Mr. Valiant informs me that he discovered traces of what he considered to be anten nae, and that for several years he continued collecting until he found a locality where the specimens were well preserved and show, not only the antennae, but legs and what he supposed to be the swimming appendages. Not having confidence that he could properly describe the specimens he sold part of his ma terial, and in this way it came to be first described by Mr. Mat thew, a student at Columbia College. His step-brother, Mr. Mitchell, continued to collect; and in August, 1893, through the courtesy of Mr. Valiant, I visited the locality with Mr. Mitchell and obtained a few specimens for the National Government. The most important part of the discovery, announced by Mr. Matthew's paper, is that the trilobita have true antennae. The discovery of the legs and plumose appendages is also of great interest, as it adds to our information respecting the appendages of the trilobite some of the details of another genus. A collection was made for the Yale College Museum by Dr. Read March 24, 1894. *Am. Jour. Sci., Vol. 46, 1893, p. 121. 90 Walcott Appendages of the Trilobites. C. E. Beecher, and in some notes on the thoracic legs of Tri- arthrus* he describes and illustrates a dorsal view of the legs of the second and third free thoracic segments. These show that the endopodite of the leg is essentially the same as in Calymene and Asaphus, and that the exopodite is unlike that of Calymene or Ceraurus. Through the courtesy of Prof. J. F. Kemp of Columbia Col lege, I have examined the material studied by Mr. Matthew; and Prof. A. H. Chester, of Rutgers College, kindly loaned me for study five specimens that he purchased from Mr. Valiant. From these and the specimens in the National Museum a few notes have been taken that permit of some comparisons with the extremities found in Ceraurus, Calymene and Asaphus.f The limbs of Triarthrus differ in the details of the joints of the in ner branch of the limb (endopodite) and still more in the char acter of the exopodite. Cephalic limbs. The antennae are uniramose, and, judging from the position in which they are found, were attached to the body near the postero-lateral angle of the hypostoma (Fig. 1, e, Plate 1). In one specimen a cephalic limb somewhat detached from its true position shows a large basal joint and six slender joints (Fig. I,/), The basal joint does not show conclusive evidence of the presence of a masticatory ridge. On another specimen, however, the form of the basal joint strongly suggests that it subserves the purpose of mastication. This is illustrated at g in Fig. 1. A slender jointed appendage like that attached to the basal joint of g occurs between it and the antennae and is probably a portion of another one of the cephalic limbs. No other cephalic appendages have been observed in the material at hand. Since the publication of my articles on .The TrilobiteJ I found in a section of the head of Calymene senaria a slender jointed limb that appears to have been an antennule. It is unlike any limb found beneath the head and thorax, and, if not an anten nule, it may represent a fifth pair of cephalic limbs. This is *Am. Jour. Sci., Vol. 46, 1893, pp. 467-470. fThe Trilobite ; New and Old Evidence Relating to its Organization. Bull. Mus. Comp. Zool., Vol. 8, 1881, p. 6. JBull. Mus. Comp. Zool., Vol. 8, 1881, p. 191-224. Science, Vol. 3, 1883, p, 279. Thoracic Limbs. 91 also suggested by a section of the limbs within the head of Caly- mene, illustrated on Plate 1, Fig. 9, Bull. Mus, Comp. Zool., Vol. 8, 1881. In this, a fifth limb is indicated close to the hy- postoma. The trilobite was enrolled so as to include the anten- nule entirely within the border of the head, A sketch, taken from a photograph of the thin section by transmitted light, is shown by Fig. 8, PI. 1. The hypostoma of Oeraurus* shows a rounded indentation of the antero-lateral sides, where an antennule probably passed by it. This character is strongly marked in Sao hirsuta, Proetus bohemicus, Amphion fischeri, etc., as illustrated by Barande. The character and position of the remaining cephalic limbs of Triarthrus are not shown in any specimens that I have examined, but, from the relations of Calymene, Ceraurus and Triarthrus, especially the two latter, it is probable that their arrangement is essentially the same. Thoracic limbs. Many specimens show the thoracic limbs extending out from beneath the carapace of Triarthrus. It was not until by a fortunate dissection that I obtained the material illustrating the limbs in position beneath the thorax. The an terior limbs are formed of a protopodite and a somewhat com plex exopodite. The protopodite consists of a short basal and a long joint, (Fig. 2, d, e,) to which the endopodite and exopodite are attached. This appears to be direct in the posterior limbs of the thorax (Fig. 3, a), but as yet the point of attachment of the basal joint of the exopodite has not been seen in the anterior limbs. The endopodite of the anterior portion of the thoracic limbs varies in the number of joints and in their relative length (Fig. 1, a, a). Two show four long proximal and three shorter dis tal joints. Other limbs show two smaller distal, and three or four proximal, while in several there is a more or less uniform gradation from the protopodite to the distal joint. In Fig. 1, some of these variations are indicated. In Fig. 2, eleven limbs are shown, as seen from the under side. The basal (coxal) joint is seen at b, d, e, and nine show the long second joint of the pro topodite. At e and f a new phase is indicated by the enlarge ment of the proximal joints. This is marked in a, b, c, d, and in Fig. 3, the details are more fully shown. These joints occur *Loc. cit., PL iv., Fig. 5. 92 Walcott Appendages of the Trilobites. on the seven posterior thoracic limbs of Fig. 2 ; and in the spec imen from which Fig. 3 was drawn the limb opposite the tenth segment from the pygidium shows a slightly triangular second (meropodite) and third (carpodite) joint. In Fig. 2, the limb a is opposite the second free segment of the thorax anterior to the pygidnm. The limbs a and &, Fig. 3, clearly show that the four proximal joints are broad and subtriangler in outline. A glance at the abdominal swimming legs of the Phyllocarida (Parane- balia), Schizopoda and Cumacea, suggests that the functions of these legs were both natatory and ambulatory. The expododite illustrated by Beecher shows the dorsal sur face (Fig. 6). A number, presenting the ventral surface, are shown on the right side of Fig. 2. They occur on the same specimen as the endopodities, on the left side, but have been pushed out of place. The most perfect is represented by m. The proximal portion is formed of a rather large basal joint and a number of short joints, 7 or 8. The distal end is formed of an inner and outer segmented portion. The inner side is divided into numerous segments by oblique divisions that give the im pression of a closely coiled spiral. The outer side is a cylindri cal, jointed, stem-like rim that is attached to the inner side, a narrow, distinctly impressed line separating the two, except at the somewhat flattened tip where they merge into each other. On the outer or upper surface of the outer side numerous crenu- lations occur that extend into long setae, n, Fig. 2; 6, ft, Fig. 1. Dr. Beecher considers the expedite as a swimming organ ; but from the manifest branchial character of the exopodite and at tached epipodite in Calymene (Fig. 7), it seems probable that this exopodite of Triarthrus served largely as a gill, and that the animal used the broad proximal joints of the posterior limbs of the thorax as its principal propulsion in swimming. The exopodite of Triarthrus looks like a consolidated exopodite and epipodite, very much as though these two organs as they occur in Calymene were merged into one. Several specimens illustrate appendages beneath the pygidium. Some have the broad proximal joints, d, Fig. 1, while others show the outer rim of the exopodite c, Fig. 1. The material I have seen indicates very little difference between the appendages of the posterior half of the thorax and the pygidium, except Classification. 93 that those of the latter are less developed in size and details. Mr. Matthew sospected the presence of a flap, formed by the anchylosing of the appendages beneath the pygidium. From the appearance of a similar structure, where the limbs are mat ted together along the side of the thorax, this tentative view is received with doubt. More perfect material may show distinc tions not recognizable at present. If future investigations prove, as it now seems probable, that the modified swimming joints of the endopodite are attached to ten or more of the thoracic segments, the anterior eight seg ments can be grouped together as the typical thorax, and the re maining segments of the body as the abdomen. Mr. Matthew suggests that the homology between Triarthrus and Limulus may not be as close as between Limulus, Calymene and Ceraurus. This is true from what we now know of Triar thrus, but, if a sixth pair of cephalic limbs should be discovered in Triarthrus the resemblance would be strengthened. Triarthrus does not differ from Ceraurus and Calymene more than would be anticipated in such unlike genera. Triarthrus is essentially a "Primordial" type that has continued until upper Ordovi- cian time. It represents a large group of Cambrian trilobites, while Calymene and Asaphus represent the more highly de veloped Ordovician and Silurian forms. Dr. Lang held the view that if a fifth pair of cephalic limbs were found, comparable to the anterior antennas "Trilobites might then be regarded as original Entomostraca, to be derived from the same racial form as the Phyllopoda." He says further, "Xiphosura, Hemiaspidse, and Gigantostraca are themselves again perhaps racially connected with the Trilobites. In any case, however, in the present state of science, it seems probable that all these groups are only connected at their roots with the Crustacea.*" From the paleontological record I am essentially in accord with this view, but I am not yet prepared to abandon the posi tion taken in 1881, that all these groups should be arranged under one class and not as an appendage to the Crustacea, as pro posed by Dr. Lang. Text Book of Comparative Anatomy, Eng. Ed., 1891, p. 415. *Loc. cit., p. 421. 94 Walcott Appendages of the TriloMtes. I would go still further and form a class of the Trilobita, and one of the Merostomata. Two general facts lead me to think that the modern crustacean is descendant from the Phyllopod branch and the Trilobita from a distinct branch.* 1st. The Trilobita branch ex hausted its initial vital energy in Paleozoic time and disap peared. 2nd. The Phyllopod branch developed slowly until after the Trilobita passed its maximum and then began its great dif ferentiation that approaches culmination in recent times. When the trilobite and phyllopod diverged from their com mon ancestral crustacean the trilobite began at once to differen tiate and to use its initial vital energy in developing new species, genera and families. Probably two thousand species and one hundred or more genera are known from the Paleozoic strata. With this great differentiation the initial vital energy was im paired and the Trilobita died out at the close of Paleozoic time. The Phyllopod branch continued with little variation until after the trilobite passed its maximum, and then began to differ entiate until to-day its descendents form the class Crustacea, that corresponds to the class Trilobita in Paleozoic time. Springing from a common crustacean base the two groups have many features in common, and in carrying out of details of structure in the limbs and gills many striking resemblances occur. It does not impress me that trilobites were true Entomostracans or Malacostracans ; they have certain character istics in common, but these are not necessarily the result of lineal descent one from the other but are the result of descent from a common ancestral crustacean type of pre-Oambrian time that lived in the pelagic fauna in which all the earlier types of life were probably developedf and from which, as time passed on, additions must have been made to the paleontologic record of geologic time. The Phyllopods, Ostracods and Trilobita are clearly differentiated in the lower Cambrian fauna. Bernard is *Tkis view is only confirmatory of the result of the profound study of the Apodida: by Bernard (The Apodidse Nature Series, 1892). tSee Brooks' beautiful memoir on Sal pa, with its suggestive theory of the origin of the bottom faunas of the ocean and the early geologic faunas. The Genus Salpa, Memoirs from the Biological Laboratory of the Johns Hopkins University, II, 1893, pp. 140-177. Mode of Occurrence. 95 confidant that the Trilobites may take a firm place at the root of the Crustacean system, with the existing Apus as their nearest ally.* There is yet much to be learned from the study of Triarthrus. A great amount of material can be readily collected at the local ity near Borne, X. Y. It is also of interest to note that the lo cality at Trenton Falls, N. Y., from which the specimens of Calymene and Ceraurus were obtained, is only seventeen miles from the Eome locality ; that both occur within the Ordovician ; and that the stratigraphic position of the bed at Rome is be tween six and seven hundred feet above that at Trenton Falls.f *Nature, Vol. 48, 1893, p. 582. fThe appendages of Triarthrus are replaced by iron pyrites and are usu ally \vell preserved. The specimens of Calymene and Ceraurus from the Trenton limestone of Trenton Falls, N. Y., were replaced by calcite and in them there were preserved even more delicate parts than I have yet ob served in Triarthrus. Thin sections were made of the latter and photo graphs obtained by transmitted light, that were used in illustrating the paper in the Bulletin of the Museum of Comparative Zoology, Vol. 8, 1881. 96 Walcott Appendages of the Trilobites. Description of Plate. Fig. 1. Triarthrus becki (X3). Outline of carapace, with appendages rep resented as they occur on several specimens, their relative position being retained. a, a, a, a. Endopodites of limbs showing variation in joints. b, b. Plumose portion of exopodite. c, c. The outer or supporting portion of the setae or fimbriae of b, b. d, Limbs extending from beneath the pygidium, showing large proximal joints. Those of the left side are imperfectly preserved. e, Antenna extending back nearly to the postero-lateral margin of the hypostoma. /. One of the cephalic limbs. The basal joint may be broken away on the inner side. g. Cephalic limb. Fig. 2 (X7). Limbs attached to the under surface of an individual preserv ing 13 thoracic segments and the pygidium. The limbs (a to k) on the left side are mainly in place. A fracture cuts out one limb between g and h. a to g. Limbs preserving traces of the enlarged proximal joints. b, d. Limbs preserving the two joints of the protopodite and two of the large proximal joints. I, m, o. Exopodites, showing under or side views. n. Enlargement of fimbrise of m. r, s. Distal joints of endopodites of right side. y. Portion of an exopodite showing its inner support. Fig. 3. Limbs occurring on the under side of an individual of 14 thoracic segments, a, b, c, d. Limbs with flattened, enlarged proximal joints and slender distal joints. a. Limb preserving large joint of protopodite, four enlarged prox imal joints and three slender distal joints. At x the point of attachment of an exopodite is shown, and in the speci men it looks as though / had been broken away from x. Fig. 4. Restoration of the thoracic limbs of the fifth segment anterior to the pygidium. en. endopodite. p. protopodite. a. four proximal swimming joints, b. three distal joints. ex. exopodite, attached to same joint of the protopodite as the endopodite. Description of Plate. 97 Fig. 5. Restoration of the thoracic limbs of the fourth thoracic segment posterior to the head. en. endopodite. ex. exopodite. Fig. 6. Diagramatic restoration of the second thoracic limb. (After Beecher.) Fig. 7. Restoration of thoracic limb of Calymene senaria. en. endopodite. ex. exopodite. ep. epipodite. (Bull. Mus. Cornp. Zool. Vol. 8, 1881.) Fig. 8. Cephalic limb of Calymene X 3 ; supposed antennule. Fig. 9. Cephalic limb figured by Dr. Henry Woodward. (Quart. Jour. GeoL Soc. London. Vol. 26, 1870, p. 487. a. side of hy- postoma. Fig. 10. Slender jointed legs associated in same beds with Calymene at Cincinnati, Ohio. PROC. BIOL. Soc. WASH., VOL. IX, 1894. PLATE I. (1) (9) (10) VOL. IX, pp. 99-104 APRIL 14, 1894 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON SYNAPTOMYS COOPERII BAIRD IN EASTERN MAS- SACHUSSETTS; WITH NOTES ON SYNAPTOMYS STONEI RHOADS, ESPECIALLY AS TO THE VALIDITY OF THIS SPECIES. BY OUTRAM BANGS, BOSTON, MASS. Ever since I began to trap small mammals in the modern im proved manner, I have been on the lookout for this species and so was not surprised to find, on June 9th, 1893, a fine adult female in one of my traps. The trap was set in an old cranberry bog that had been allowed to run out, and had grown up to clumps of Viburnum and Vaccinium bushes, and under these, grasses and sphaguum and carices had crowded out the cran berry vines to a considerable extent. It was in the middle of the Plymouth woods, about seven miles from the town of Wareham, Plymouth County, Mass. The ground was traversed in every direction by 'the run-ways of Arvicola riparius and in one of these run-ways I caught the Synaptomys. She was nursing young at the time, although repeated trapping in the same bog yielded nothing but innumerable Arvicolas, a Zapus hudsonius or two, and a few Evotomys gapperi. I now had a slight notion of the sort of place to look for Synaptomys in, and tried all such localities I could find without success until September 21, 1893, when in an almost precisely similar bog about six miles distant from the first place, in the township of Wareham, I caught an adult female, also nursing, and in an Arvicola run -way; and on September 24, an adult 160 fiang8^=*Synfiptomy$ Cooperii male in another trap in the same bog, in an Arvicola run-way. Following is a list of small mammals caught in this last bog, which, as I trapped it pretty clean, may be of interest as showing the species inhabiting such a place, and their relative abundance : Twenty [20] traps set. Sept. 19, 1893. 6 Arvicola riparius. 1 Zapus hudsonius. Sept. 20. 5 Arvicola riparius. 1 Evotomys gapperi. Sept. 21. 3 Arvicola riparius. 1 Evotomys gapperi. 1 Sorex person at us. 1 Synaptomys cooperii. Sixty-five [65] traps set. Sept. 22. 17 Arvicola riparius. 1 Evotomys gapperi. Sept. 23. 10 Arvicola riparius. Sept. 24. 6 Arvicola riparius. 1 Evotomys gapperi. 1 Synaptomys cooperii, Sept. 25. 3 Arvicola riparius. 1 Evotomys gapperi. Sept. 26. 1 Arvicola riparius. 1 Evotomys gapperi. Sept. 27. 1 Evotomys gapperi. Sept. 28. Nothing; took up traps. Totals. Arvicola riparius, 54. Evotomys gapperi, 7. Sorex personatus, 1. Zapus hudsonius, 1. Synaptomys cooperii, 2. This bog contained about an acre-and-a-half, and was bordered on one side by thick swampy woods and on the other three by open fields of grass, and had a small brook running through it, Synaptomys cooperii is, I think, rare, or at any rate very local in this section, as I have trapped persistently for two years in every sort of locality the county affords, and have only taken these three examples. As the country about Wareham, Mass., is not unlike that of And Synaptomys Stonei. 101 South Central New Jersey, I was anxious to see if my specimens were not referable to S. stond Rhoads rather than to S. cooperii For this purpose Dr. 0. Hart Merriam kindly lent me a fine series of fourteen skins and many skulls of S. cooperii, partly from his own private collection, and partly from the collection of the Department of Agriculture at Washington. I also, through the kindness of Mr. S. N". Rhoads, of the Academy of Natural Sciences, Philadelphia, had a chance to examine his type of Synaptomys stonei and a topotype in the collection of Mr. Whitmer Stone, for whom the species was named. In the light of this fine material, the specific character claimed for S. stonei faded away to mere individual variation, and S. stonei will have to stand as a synonym of S. cooperii, pure and simple. The list of specimens I had to work with is as follows : No. Sex. Date. J -ocality. Collector. Measurement of hind foot. *215 9 ad. June 9, 1893 Mass., Wareham. O. Bangs 191T 216 9a. PQ 1 'e eg & Ot> 3 O tt ^" . ^ ^w |b8 |i 5 ld 3 | .3^ ! !" poj d.a Collec Measu 108 M'lllfr-Iiuins A nctr Rabbitom \\~cxfrrii Flnr'nln. T* c3 8 3 *e A, pal W VOL. IX, pp. 109-116 JUNE 21, 1894 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON PRELIMINARY DESCRIPTIONS OF ELEVEN NEW KAN GAROO RATS OF THE GENERA DIPODOMYS AND PERODIPUS. BY DR. C. HART MERRIAM. The following brief descriptions are here published in advance of a monographic revision of the group which will appear shortly. Of the eleven new forms here defined, Dipodomys elator from northern Texas ; D. drnatus from the state of Zacatecas, Mexico ; D. m. nitratus from Owens Lake, California; Perodipus streatori from the west slope of the Sierra Nevada, and P. panamintinus from the Panamints Mts. in California require no comparison with previously described species. The others are less sharply differentiated. Dipodomys elator sp. nov. Type from HENRIETTA, CLAY Co., TEXAS. No. 64,802 $ ad. U. S. Nat. Museum, Department of Agriculture Collection. Collected April 13, 1894, by J. Alden Lormg (Original number 1,804). Measurements (taken in flesh). Type: Total length 292; tail vertebrae 173 ; hind foot 47. Ear from anterior base 14 (in dry skin). Average measurements of 2 specimens from type locality : Total length 290 ; tail vertebra 170 ; hind foot 45 5. General characters. Similar to Dipodomys spectabilis but con siderably smaller, with much smaller ears ; tail more slender and paler, with shorter white pencil ; hind feet relatively longer ; 110 Merriam New Kangaroo Rats. facial crescents heavier ; nose blacker. Cranial characters uni que. Color. Upper parts clay-color, lined with dark-tipped hairs on head and back, becoming pale ochraceous buff on flanks; thigh patches large ; facial crescents broad and indistinctly continuous to end of nose which is broadly blackish ; inner side of legs dusky ; dorsal and ventral tail stripes barely meeting in front of white pencil, the white lateral stripes being almost con tinuous to the white tip ; ventral dark stripe pale ; dorsal dark stripe pale for proximal I, becoming blackish on crested part; white pencil rather short, measuring about 23 mm. beyond the tips of the black hairs in the two specimens at hand. Cranial characters. Skull small for the size of the animal; rather highly arched on top as in D. phillipsi ; supraoccipital between mas- toid bullce broader than in any other known species; interparietal nearly as broad as long; ascending branches of premaxilke broad and slightly expanded posteriorly ; nasals somewhat narrowed posteriorly; top of skull broad but not broad enough to hide zygomatic arches, which are far apart ; sides of frontals sloping strongly inward from point slightly anterior to plane of fronto-parietal suture ; nasals decidely longer than frontal breadth immediately behind lachrymals. Mandible small for size of skull; angle large and pointed. Upper premolar an incompletely double prism, its crown with a well developed aiitero-internal lobe. Dipodomys ornatus sp. nov. Type from BERRIOZABAL, ZACATECA*, MEXICO. No. 57,990 9 ad. U. S. Nat. Mus. Department of Agricultural Collection. Collected December 29, 1893 by E. A. Goldman. (Original number 5,613.) Measurements (taken in flesh). Type: Total length 274 ; tail vertebrae 167 ; hind foot 39. Ear from anterior base 15 (in dry skin). General characters. Similar to Dipodomys phillipsi in size and pattern of markings but brighter and more golden in color; dark markings more extensive and blacker, ears somewhat larger ; hind foot shorter ; tail crested penicillate, its tip white. Color. Upper parts bright golden clay-color, darkest on head and median back, brightest on sides ; thigh patches large; facial crescents large, broad, and very black, meeting broadly over bridge of nose which Is solid black for ^ the distance from nostrils to eye except a small white spot over extreme end of nose ; narrow ring round eye, inner sides of hind legs, and dorsal and ventral tail stripes black ; tip of tail pure white. The white side stripes of the tail disappear near the junction of the distal and middle thirds, the black stripes uniting in a broad belt anterior to the white pencil. The face is mainly white between the eye and facial crescent, though the white is somewhat obscured, particularly near the eye, by dark-tipped hairs. Cranial character's. Skull similar to that of J). pkillijuri in general size and form, but proportions different. Mandible larger and heavier with much longer and broader angular processes. The basal length and mastoip Mcrriam New Kangaroo Rats. Ill breadth are essentially the same in the two species, but the naso-occipital length in ornatus is much less and the zygornatic breadth very much greater- While the zygomatic breadth is actually greater in ornatus, the breadth across the top of the skull is decidedly less : hence when viewed from above, the zygomatic arches stand out beyond the sides of the cranium, while in phillipsi they are hidden beneath the edges of the frontals and parietals. In ornatus the top of the cranium is much flatter than in phillipsi ; the supraoccipital is narrower between the mastoid bullne ; the nasals are not narrowed behind, and the ascending branches of the premax Okie are shorter and more slender and have no trace of the posterior expansion commonly present in phillipsi. The upper premolar is a single prism and its crown has no trace of the antero-internal lobe of phil Dipodomys perotensis sp. nov. Type from PEROTE, VERA CRUZ, MEXICO. No. 54,285 9 ad. U. S. Nat. Mus. Department of Agriculture Collection. Collected May 21, 1893 by E. W. Nelson (Original number 4840). Measurements (taken in flesh). Type: Total length 265 ; tail vertebrae 162 ; hind foot 40. Ear from anterior base 14 (in dry skin). Average measurements of 8 specimens from type locality : Total length 271 ; tail vertebra 168 ; hind foot 40.4. General characters. Similar in size and general appearance to Dipodomys phillipsi and ornatus and intermediate between them in coloration ; white terminal pencil short, and in one specimen absent. Cranial characters substantial. Color. Upper parts brownish clay color, intimately mixed with and darkened by blackish-tipped hairs on head and back ; strongly suffused with ochraceous buff 011 sides and flanks; facial crescents large and black, meet ing across the nose ; inner side of leg and sole blackish ; lateral white stripes of tail disappearing near junction of distal and middle thirds ; white termi nal pencil small and in one specimen absent (possibly the result of injury in early life). Cranial characters. Skull similar to that of D. ornatus, but even narrower on top [consequently very different from pliillipsi]; zygoma visible from above ; top of skull more strongly arched anteropostenorly than any other known species ; breadth of supraoccipital between inflated mastoids greater than m phillipsi or ornatus. Angle of mandible larger than in phillipsi but smaller than in ornatus. Dipodomys merriami nevadensis subsp. nov. lype from PYRAMID LAKE, NEVADA. No. 54,552 9 ad. U. S. Nat. Mus., Department of Agriculture Collection. Collected June 26, 1893, by Vernon Bailey (Original number 3,990). Measurements (taken in flesh). Type: Total length 240 ; tail vertebrte 140 ; hind foot 39. Ear from anterior base 13 (in dry skin). Merriam New Kangaroo Rats. Average measurements of five adults from type locality : Total length 243 ; tail vertebra? 143.5 ; hind foot 39.9. General characters. Similar to D. merriami but with shorter tail and longer hind foot ; coloration paler and more buffy. Color. Upper parts pinkish buff, darkened on head and back by inter mixture of dark- tipped hairs ; facial crescents distinct but hardly meeting across nose, though bridge of nose is somewhat darkened ; face in front of eyes pure white except where interrupted at base of whiskers by facial cres cents; underparts and thigh stripes pure white; dorsal and ventral tail stripes dusky, meeting at end of tail ; inner side of legs to heel dusky. Dipodomys merriami nitratus subsp. nov. Type from KEELER, EAST SIDE OF OWENS LAKB, CALIFORNIA (No. f^ tf ad. U. S. Nat. Mus. Department of Agriculture Collection). Collected December 29, 1890 by E. W. Nelson (Original number 160). Measurements (taken hi flesh). Type: Total length 237; tail vertebra? 140; hairs 26; hind foot 39. Ear from anterior base 13 (dry skin). Basilar length of skull 22 mm. Average measurements of 23 specimens from type locality : Total length 239 ; tail vertebra? 141 ; hind foot 37.8. General characters. Smaller than D. merriami, with relatively larger hind feet and wholly different coloration ; dusky markings obsolete. Color. Upper parts uniform intense ochraceous or tawny-buff" not mixed with black-tipped hairs ; facial crescents obsolete ; no dusky or blackish markings anywhere ; no superciliary stripe, but a distinct white spot over eye ; upper and lower tail stripes concolor with back ; white side stripes continuous. Dipodomys merriami nitratoides subsp. nov. Type from TIPTON, SAN JOAQUIN VALLKY, CALIFORNIA. No. 54,674 tf ad. U. S. Nat. Mus. Department of Agriculture Collection. Collected June 25, 18V)3, by Clark P. Streator (Original number 2,978). Measurements (taken in flesh). Type: Total length 246, tail vertebra? 148 ; hind foot 36. Ear from anterior base 12 (in dry skin). Average measurements of 13 specimens from type locality : Total length 237 ; tail vertebra? 144; hind foot 35. General characters. Similar to D. m. nitratus in size and color, but with strongly marked facial crescents meeting over bridge of nose; ears smaller. Color. Upper parts everywhere uniform fulvous ; facial crescents dusky and meeting over bridge of nose ; dorsal tail stripe darker than back ; crested part of tail same color as fcack ; ventral tail stripe dull fulvous, con- Merriam New Kangaroo Rats. 113 tinuous to en 1 of tail ; inner aspect of hind legs to heel dull fulvous ; under parts and thigh stripe white ; spot ever eye obscured by dark tipped hairs. Dipodomys merriami exilis subsp. nov. Type from FKKSNO, SAN JOAQUIN VALLEY, CALIFORNIA. No. ^f/ 3 (^ yg. ad. U. S Nat. Mus. Department of Agricultural Collection. Col lected September 23, 1891, by Vernon Bailey (Original number 3,277). Measurements (taken in flesh). Type: Total length 241; tail vertebrae 143 ; hairs 21 ; hind foot 33. Ear from anterior base 12 (in dry skin). Bisilar length of skull 21 mm. Average measurements of 20 specimens from type locality : Total length 227; tail vertebrae 135.5 ; hind foot 34. General characters. Similar to Dipodomys merriami but smaller and darker, with tipper surface of nose and posterior aspect of ankles black. Color. Upper parts nearly uniform clay color, darkened with sepia from abundant admixture of black-tipped hairs, and darkest on the head ; sides and flanks tinged with ochraceous-buff; black crescents at base of whiskers sharply defined and meeting in median line so that the bridge of the nose is black: superciliary stripe whitish, not interrupted as in D. merriami; ears dark ; posterior aspect of ankles and lower leg black ; upper and lower tail stripes sooty blackish, meeting along terminal third, thus interrupting the white side stripes ; under parts silky white. Cranial characters. Skull similar to that of D. merriami but much smaller ; nasal bones shorter. Dipodomys merriami atronasus subsp. nov. Type from HACIENDA LA PARADA, SAN Luis POTOSI, MKXH o. No. 50,270 tf ad. U. S. Nat. Mus. Department of Agriculture Collection. Collected August 20, 1892, by E. W. Nelson. (Original number 3,2.'9). Measurements (taken in flesh). Type: Total length 267 ; tail vertebrae 162 ; hind foot 40- Average measurements of 4 specimens from type locality : Total length 250; tail vertebras 152 ; hind foot 38.5. General characters. Similar to D. merriami but darker ; pelage coarser, particularly on head. Color. Upper parts dark clay-color, everywhere mixed with dark-tipped hairs and suffused with ochraceous buff, which is strongest on the sides ; nose from black tip to between eyes grizzled with coarse yellowish, dark- tipped hairs ; facial crescents large, black, meeting over end of nose ; inner side of thighs and dark tail stripes blackish ; white lateral tail stripes mixed with dark hairs and disappearing in middle third of tail. Perodipus streatori sp. nov. Type from CARBONDALE, MARIPOSA Co., CALIFORNIA (at west foot of Sierra Nevada). No. 64,310 9 ad. U. S. Nat. Mus. Department of 114 Merriam AV/r Kangaroo Rats. Agriculture Collection. Collected April 3, 1894, by Clark P. Streator. (Original number 3,673). Measurements (taken in flesh). Type: Total length 29 .i; tail vertebrae 179 ; hind foot 43. Average of 2t> specimens from type locality : Total length 295 ; tail vertebra 180 ; hind foot 43. General characters Similar to P, agilis but larger ; ears smaller; tip of tail normally white. Color. Upper parts Isabella brown, darker along the middle of the back and on sides of neck ; sides and flanks suffused with ochraceous buff ; a distinct white spot over eye and at base of ear ; top of nose, crescent through base of whiskers, and narrow ring around eye blackish; a band of white overlaid by dark-tipped hairs runs from base of whiskers to ear, including the eye: innerside of thigh and sole of foot blackish; dorsal and ventral tail stripes dusky, meeting in a broad subapical dark ring beyond which the end of the tail is normally pare white as in many species of Dipodonnjs ; under parts, thigh stripes, and ring at base of tail pure white. Two very young specimens have the white tip of the tail sharply defined but short ; some of the old specimens lack the white tip, in others the white side-stripes are nearly continuous to the tip. Cranial characters. Skull similar to that of P. agilis but larger and heavier ; parietals longer an tero- posteriorly (inner border decidedly longer than anterior) ; fronto-parietal suture strongly sinuous, convex forward at median line ; supraoccipital broader between mastoid bull;e on top of skull. Dental characters. Molariform teeth larger and heavier ; crown of last upper molar longer antero-posteriorly and usually more subquadrate ; osteodentine islands dark. Perodipus panamintinus sp. nov. Type from PANAMINT MTS., CALIFORNIA (onhead of Willow Creek). No. 44 ; tail vertebrae 148 ; hind foot 40. Ear from anterior base 13 (in dry skin). Average measurements of 15 specimens from type locality : Total length 244.5 ; tail vertebra 140.5 ; hind foot 39. General characters. Similar to P. ordi but less ochraceous in color and with markings more pronounced; ears blackish instead of flesh color ; supraorbital white spots more conspicuous. Color. Upper parts clay-color, finely mixed with dark-tipped hairs ; thigh patches large, more tinged with ochraceous than rest of upper parts, this color reaching down on inner side of leg in a narrow stripe to heel [no dusky on inner side of leg] ; facial crescents blackish and nearly continuous with a blackish spot on end of nose ; supraorbital white spots unusually conspicuous ; interior of ear conch and reflexed upper border blackish ; back side of ear whitish ; dorsal and ventral tail stripes dusky, the ventral failing or nearly failing on terminal fourth, where the white sides stripes commonly meet or nearly meet for a short distance ; tip of tail dusky all round from upper stripe. Cranial characters. The skull of columbianus differs from that of typical ordi in having the basioceipital much broader, the postero-external angle of the maxillary wing of the zygoma more strongly produced backward and downward and the angle of the mandible larger. VOL IX, pp. 117-128 JULY 2, 1894 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ABSTRACT OF A STUDY OF THE AMERICAN WOOD RATS, WITH DESCRIPTIONS OF FOURTEEN NEW SPECIES AND SUBSPECIES OF THE GENUS NEOTOMA. BY DR. C. HART MERRIAM. The following brief abstract of a study of the North Ameri can Wood Rats and Desert Rats, with descriptions of a dozen new forms from Mexico and the western United States, based on the rich collections of the United States Department of Agricul ture, is here published in advance of a more formal paper on the group. The genus Neotoma is here restricted to the species in which the crown of the last lower molar is made up of two trans verse loops; the species having the crown of this tooth shaped like the letter S are transferred to the genus Ptyssophorus of Ameghiuo, previously known from a single fossil species from South America. As thus restricted, the genus Neotoma is divided into two subgenera, Neotoma proper and Teonoma, which are complementary in their geographic distribution, Neotoma proper being Sonoran or Austral, while Teonoma is Boreal. It is con venient to subdivide the former into four minor groups, none of which is worthy of the distinction of subgeneric rank. These groups may be designated, from a typical species in each, as fol lows : (1) the leucodon group ; (2) the mexicana group ; (3) the desertorum group, and (4) the arizonae group. Subgenus NEOTOMA Ord, 1825. Type, Neotoma flondana Ord, from Florida. Tail commonly round, scant-haired and tapering, but in one species mod erately bushy ; hind feet small or moderate. 118 Merriam American Wood Rats. Rostrum of moderate length, never more than one-third the length of cranium ; sagittal area usually rounded, the broadest part always con siderably anterior to plane of interparietal, whence the sides curve gradually backward to interparietal shield ; spheno- palatine vacuties always open. (1) Neotoma leucodon group. Neotoma leucodon, latifrons, micro- pus, baileyi, floridana and pennsylvanica form a fairly well cir cumscribed group, differing from the other subdivisions of the genus in having the f rentals abruptly spreading and flattened immediately behind the intcrorbital constriction, the orbital mar gins upturned and pinched in, almost forming a bead ; the nasal bones short and cuneate, tapering evenly to a dull point behind ; the postpalatal notch moderately or broadly excavated (moder ately in leu-cndon, very broadly in floridana) ; the upper molar series very much broader anteriorly than posteriorly (m - 1 nearly i broader than m -) ; m 1 comprising three transverse loops, the anterior of which is but slightly indented by the antero-internal sulcus never divided by the deepening of this sulcus as in the mexicana series; color of teeth white or nearly white (except in floridana, which is an aberrant member of the group*). N. pennsylvanica has certain primitive characters not shared by the others, and is more nearly intermediate between the subgenera Neotoma and Tcononia than any known living species. The group inhabits the Lower and Upper Sonoran Zones, from Perote in Vera Cruz and Berriozabal in Zacatecas, northward to southern South Dakota. (2) Neotoma mexicana, group. Neotoma mexicana'f, pinetorum, orizabse tenuicauda, fidviventer, fallax and fuscipes form a group of closely allied species agreeing in certain important cranial char acters whereby they differ from all the other subdivisions of the genus. N. fu.xci-pc*l and fallax are somewhat aberrant members of the series. Neotoma torquata Ward probably belongs here also . The group seems to occupy a midway position in the genus, lack ing the more specialized characters that distinguish the others. *In most species of Neotoma the osteodentine is dark and the reentrant angles are filled with a blackish substance. fOtie subspecies of mexicana is here recognized : N. mexicana bullata from the Santa Catalitia Mts. in Arizona. JThe subspecies of fuscipes here recognized are macrotis Thomas from the southern coast region of California ; streatori from the western slope of the Sierra Nevada and adjacent parts of the upper Sacramento Valley, and dispar from the east foot of the Sierra along the western edge of the Mohave Desert region. N. monochroura Rhoads and N. splendens True seem to be typical fuscipes, and N. macro/is simplex an inter- grade. Merriam American Wood Rats. 119 The upper molar series is of more nearly equal breadth through out, the anterior molar not being so broad relatively as in the other groups. The postpahiUl notch is usually narrow, though it is broadened anteriorly in N. fuscipes macrotis of southern Cali fornia. The frontals increase in width but slightly from before backward, never expanding abruptly behind the interorbital con striction as in the leucodon series. The anterior lobe of m - is completely divided by a deep sulcus on the inner side into two loops, except in fuscipes, in which the sulcus is relatively shal lower and more anterior in position, the division being less com plete than in other species. So far a known the group is re stricted to the Upper Sonoran and Transition Zones, where it ranges from southern Mexico (States of Jalisco, Michoacan, Mexico, Puebla, and Vera Cruz) northward in the interior to Colorado and northern Arizona, and along the Pacific Coast to Oregon. (3) Neotoma desertorum group. Neotoma desertorum and inter media* constitute the third group into which it is convenient to divide the restricted genus. The group is not very sharply de fined, some forms of intermedia coming very close to aberrant forms of the leucodon series. The frontals increase in breadth gradually from before backwards, much as in the pinetorum, group not suddenly behind the constriction as in the leucodon. series. There is no supraorbital bead in typical desertorum, but intermedia shows a decided tendency to the formation of such a bead. The postpalatal notch is narrower than in any other division of the genus. In dental characters the group resembles the leucodon series, the molars being decidedly broader anteriorly than poster iorly, and m L being made up of three transverse loops, the an terior of which is but faintly indented by the antero-internal sulcus. The members of the group inhabit the Sonoran deserts of northern Mexico and the southern United States, ranging from Chihuahua and Sonora northward to northern Utah, northern Nevada, and middle California. * Neotoma intermedia Rhoads inhabits the valleys of the coast reigon of California, south of Monterey Bay. A somewhat paler form, usually more or less suffused with pale ochraceus buffy, inhabits San Gorgonio Pass and the western edge of the Colorado Desert. It was provisionally named gilva by Rhoads, and has just been renamed venusta by True (in a publication received since the present paper went to press), but seems hardly entitled to the distinction of a separate name. N. californica Price seems to be a typical intermedia. Two subspecies, albigula Hartley from south and west Arizona, and melanurq nob, from Sonora, are here recognized. 120 Merriam American Wood Rats. (4) Neotomci nrlznutc group. Neotoma arizonss and N. lrj>i cb cb O^> C^> O5 OO 00 00 QO CX) OO &5 ?5 3 :oO"*iLOOt^GOCft COfMCOOOCOCOCOCO i 132 bC I i i ^ ^ 8 sp $ a I o 43 o> I Ea Crania LhtiKjx AV/r I^/i/tl Mntisr from Newfoundland. 10 o 2 rH Of o ^ o os rH Of 5 % S (MOO 00 00 - GO lOrflO (M M rH 00 (M C^l 00 co^-OTt^ (M (M -H 00 M SO OO 00 M < ^ Th LO TJH >o co t^ co 06 *' o ~- 06 CM C CO GO CO O >P 00 00 CO 9C9 . e -3 | . S ^ S H " ' ; * H li a 1 1 1 ' ' i fi 1 1 1 ' ALPHABETICAL INDEX Names of new species and subspecies are printed in heavy type A. Agapostemon, 67. Amphion fischeri, 91. Andremidse, 67. Annual meeting, xi, xvi. Anoplotermes, 33. Anthophora, 66, 67. Ant economy, 24, 68-71. fallow, 6. sauba, 29. wars, 24-25. Ants, 23-36, 47, 53, 54, 68-71. Apidse, 16, 67. Aphides, 26. Apis, 14. species and varieties, 63-65. Apus, 95. Araucaria pictaviensis, 82. Army worm, 14. Arvicola chrotorrhinus, 129, 130. riparius, 99, 100, 104, 129, 130. Arvicola terraenovae, 129-132. xanthognathus, 129, 130. Asaphus, 93. Atta cephalotes, 70. fervens, 69. mexicana. 70. tardigrada, 69, 70. B. Bailey, L. H., The Plant Individual in the Light of Evolution, xi. Bailey, Vernon, Bones from a Cave in Arizona, viii. Baker, Frank, Nomenclature of Nerve Cells, xvi. , Some Peculiarities of Lumbar Vertebrae, xiii. Bangs, Outram, New Field Mouse from Newfoundland, 129-132. , New Musk Rat from Newfound land, 133-138. , Synaptomys cooperii in Eastern Mass., with notes on Synaptomys stonei Rhoads, 99-104. Bangs, Outram, and Miller, G. S., Jr., A New Rabbit from Western Florida, 105-108. Beal, F. E. L., Food Habits of Wood peckers, xii. Bees, 4-18, 47-48, 61-68. division of labor among, 9. polleniferous organs in, 65-67. social organization of, 9. Bennettites, 79, 86. dacotensis, 86. gibsonianus, 80. gibsoni, 80. maranianus, 84. maximus, 81. mirabilis, 86. peachianus, 81. peachii, 8L. portlandicus, 81. saxbyanus, 80. saxbyi, 80. schachti, 85. Benton, F., Species of Bees, 64-65. Blattidre, 40. Blister beetles, 55. Bolbopodium, 79. mamertinum, 82. micromerum, 82. pictaviense, 82. Bombus, 18, 49, 6(5. Bombycidse, 44. Bottle bees, 18. (139) 140 Alphabetical Index. Burial grounds of ants, 26. Bumble bee, 18. C. Calotermes, 33, 35. Calymene, 92, 93, 95. senaria, 90. Camponotus inflatus, 69. Carleton, M.-A., Artificial Infection with Uredospores, x. Cell, Discussion of living, ix. Ceraurus, 9i, 93, 95. Chartergus nidulans, 21. Chironomus, 45. Clathraria schachti, 85. Clathropodium foratum, 82. megalophyllum, 79. microphyllum, 80. mirabile, 86. morieri, 81. sarlatense, 82. trigeri, 82. Claviger, 30. Clisiocampa sp., 3. Coville, F. V., Botanical Explorations of Thomas Coulter in Mexico and Cali fornia, xv. , Remarks on the New Botanical Check List, xiii. Cremastochilus, 30. Creinastogaster, 29, 70. arboreus, 70. lineolata, 70. Cycadean Trunks of North America, 75-87. Cycadeoidea, 79-87. abequidensis, 87. bianconiana, 83. bucklandi, 81. capelliniana, 84. carruthersi, 81. cocchiana, 84. dacotensis, 86. emmonsi, 86. etrusca, 84. ferretiana, 84. Cycadeoidea forata, 82. gibsoni, 79, 80. imolensis, 84. inclusa, 81. intermedia, 81, 83. jenneyana, 77, 87. inamertina, 82. maraniana, 84. marylaiidica, 86. masseiana, 84. maxima, 81. megalophylla, 79. micromera, 82. microphylla, 80. mirabilis, 75, 78, 86. montana, 82. morieri, 81. munita, 86. peachii, 81. pictaviensis, 82. pirazzoliana, 83. portlandica, 81. pygmsea, 79, 80. reichenbachiana, 85. sarlatensis, 82. saxbyana, 80. scarabellii, 83. schachti, 85. schulziana, 85. trigeri, 82. veronensis, 83. zamiostrobus, 86. Cycadites megalophyllus, 79. microphyllus, 80. saxbyanus, 80. schachti, 85. trigeri, 82. Cycas, 86. Cynips quercus-mellaria, 69. D. Ball, W. H., Exhibition of Remains of Mammoth, xv. Dipodomys, 109-113. Dipodomys elator, 109-110. merriami, 112, 113. Alphab etica I In dex. 141 Dipodomys merriami atronasus, 113. exilis, 113. nevadensis, 111-112. nitratoides, 112-113. nitratus, 112. ornatus, 109, 110-111. perotensis, 111. phillipsi, 110, 111. spectabilis, 109. Dolichoderus, 70. Dorilidae, 24. Dragon flies, 38. E. Echinostipes, 82. microphyllus, 80. nidiformis, 79. pygmseus, 80. Economy of Hive, 9. Election of Officers, xi, xvi. Encephlartos bucklandii, 79. Eucheira socialis, 4. Eutermes, 33, 35, 74. morio, 74. rippertii, 72, 74. Evermann, B. W., Fishes of Missouri River Basin, xv. , Red Fish of the Idaho Lakes, xi. Evotomys gapperi, 99, 100. P. Fall web-worm, 3. Fiber obscurus, 133-138. Fiber zibethicus, 133, 134, 135, 137. Field mouse, new, from Newfoundland, 129-132. Food habits of ants, 26. Formica, 29. exsecta, 25, 68. exsectoides, 68. fusca, 26, 29. pratensis, 25. subpolita, 26. Formicary, Division of individuals in, 29. Formicidse, 24. Galloway, B., T., Effect of Spray ing with Fungicides on Growth of Nursery Stock, ix. , A Hexenbesen of Rubus, viii. , Physiological Significance of Trans piration of Plants, x. , Rust of Pine Leaves and Effects of Parasite on Host, vii. , Size and Weight of Seed in Rela tion to Size and Weight of Plant, ix. , Winter Coloration of Evergreen Leaves, viii. Gigantostraca, 93. Gill, Theodore, The Belone and Sarginas of Aristotle, xv. , Relation of Ancient and Modern Ceratodontidse, xiv. , On the Torpedoes, xiii. , Pithecanthropus, xii. , Remarkable New Bassalian Family of Crabs, x. , Segregation of Osteophysial Fishes as Fresh Water Forms, vii. Goode, G. Brown, Horizontal and Ver tical Distribution of Deep Sea Fishes, xiv. , Location and Record of Natural Phenomena by a Method of Reference to Geographical Coordinates, xiv. Greene, Edward L., Some Fundamen tals of Nomenclature, xvi. Hemiaspidre, 93. Hepialus, 39. behrensi, 40. 'hectus, 40. Heredity in insects, 51-58. Hill, Robert T., A New Fauna from the Cretaceous Formations of Texas, vii. Histerid^, 30. Hive, Economy of, 9. Holm, Theodor, Anatomy of a Leaf Gall of Pinus virginiana, xii. 142 A Ip ha betica I In dex. Holm, Theodor, Contributions to Flora of District of Columbia, xvi. , (Edema of Violet Leaves, xiii. Hornet, baldfaced, 20. Howard, L. O., An Enemy of the Hell- gramite Fly, xv. , New Cotton Enemy from Mexico, xi. , Notes on Spider Bites, viii. Hyphantria cunea, 3. Hyponorneutidte, 4. I. Insect societies, organized, 3. Insects, color sense in, 38-39. heredity in, 51-58. intelligence in, 46-51. senses in, 30-46. social, 1-74. generalizations on, 36. Invertebrates vs. vertebrates, 58-60. J. James, Joseph F., Remarks on Daimo- nelix and allied Fossils, xiii. Judd, S. D., Food of the Catbird, Brown Thrasher, and Wrens, xv. K. Kine-keeping, etc., by ants, 26-27. Knowlton, F. H., Amount of AVater Transpired by Plants, xi. L. Las ins, 25, 29. Lepus paludicola, 105-108. palnstris, 105-108. Libellulidie, 38. Limulus, 93. Lucas, F. A., Abnormal Feet of Mam mals, xii. , Extinct Gigantic Birds of Pata gonia, xv. M. Mantellia, 79. inclusa, 81. intermedia, 81. niegalophylla, 79. microphylla, 80. nidiformis, 79. pygmtea, 80. scarabellii, 83. Mearns, Edgar A., The Hares (genus Lepus) of the Mexican Border, xiv. Megachilus kaempferi, 59. Melipona, 65, 66, 67. Meloidse, 55. Melissodes, 66, 67. Merriam, C. Hart, A Remarkable new Rabbit from Mexico, viii. , American Wood Rats (Neotoma), with Descriptions of 14 new Species, 117-128. , Dental Armature of Pocket Gophers (Geomys), ix. , Mammals of Pribilof Islands, xv. , North American Shrews, xiv. , Descriptions of 11 new Kangaroo Rats (Dipodomys and Perodipus), 109-116. , Short-tailed Shrews of America, xiv. Miller, G. S., Jr., and Bangs Outram, A new Rabbit from Western Florida, 105-108. Monomorium pharaonis, 24. Musk Rat, new, from Newfoundland, 133-138. Myrmicidae, 24. M yrmecocystus melliger, 69. mexicanus, 69. Myrmicophilye, 29-31. N. Nelumbium, 86. Neotoma, 117-128. albigula, 126, 127. arizonse, 120. A Iphabetical Index. 143 Neotoma baileyi, 123. cinerea drummondi, 127, 128. desertorum, 119, 125-126. sola, 126. fallax, 123-124. floridana, 117, 118. fulviventer, 121-122. fuscipes dispar, 124-125. streatori, 124. intermedia angusticeps, 127. melanura, 126-127. latifrons, 121. leucodon, 118, 120, 121. mexicana, 118. mexicana bullata, 122, 123. orizabae, 122. orolestes, 128. tenuicauda, 121. Nomada, 67. O. Oecodoma cephalotes, 29. (Ecophylla, 24. Organs, important special, of bees, 11. sense, 42. wax-producing, 12. P. Palmer, Wm., Albinistic Birds' Feet, xv. , Nesting Sites of Blue Gray Gnat Catcher, x. , Rare Birds in District of Columbia, viii. Parorgjria, 39. Perdita, 66. Perodipus, 109, 113-115. agilis, 114. ordi, 115. Perodipus ordi columbianus, 115. panamintinus, 109, 114. streatori, 109, 113-114. Plant lice, 26. Polistes, 20, 23. gallica, 21, 22. Pollard, Charles L., Genus Cassia in America, x. Polleniferous organs in bees, 16, 65 67. Polyergus rufescens, 25. Polyrhacis, 70. Poneridte, 24. Powell, J. W., Classification of Subject- matter of Biology, xiii. Prosopis, 67. Proetus bohemicus, 91. R. Rabbit, New, from Western Florida, 105-108. Raumeria cocchiana, 84. masseiana, 84. reichenbachiana, 85. schulziana, 85. Riley, C. V., Social Insects from Psy chical and Evolutional Points of View, 1-74. ; Some Interesting Results of In juries to Trees, xi. , Transmission of Acquired Char acters, viii. Romanes on instinct in neuter insects, 49. Rose, J. N., A Botanical Trip to North western Wyoming, vii. S. Salpa, 94. Samia cynthia, 44. Sao hirsuta, 91. Sciara, 4. Simpson, Charles T., Geographical Dis tribution of Fresh-water Mussels, xiv. , Geographical Distribution of Land Shells in Jamaica, x. , Respective Values of Shell and Soft Parts in Naiad Classification, xiii. , Validity of genus Margaritana, xi. Slave-making among ants, 25-26. Smith, Erwin F., Biology of Bacillus trachseiphilus, xiv. 144 A Iphabetical Index. Smith, Edwin F., Last Phase of the Root Tubercle Question, xi. , Length of Vessels in Higher Plants, ix. , The Other Side of the Nomenclat ure Question, xiv. Smith, Theobald, Infectious Entero- Hepatitis of Fowls due to Protozoa, xiii. , Significance of Variation among Species of Pathogenic Bacteria, viii. Snake worms, 4. Stejneger, Leonhard, Exhibition of Spade-Foot Toad (Spea), viii. Sternberg, George M. , Explanation of Acquired Immunity, xiii. , Explanation of Immunity from Infectious Diseases, xii. , Practical Results of Bacteriological Researches, xvi. Stiles, Ch. Wardell, Adult Cestodes of Herbivorous Animals, ix. , Distoma westermanni in the Lungs of a Cat, viii. , Double-pored Cestode with Occa sional Single Pores, xii. . , Experimental Trichinosis in a New Host, x. , Presence of Adult Cestodes in Hogs, xiii. , The Rudolph Leuckhart Memorial, xv. , The Third International Zoolog ical Congress, xv. , Teaching of Biology in Colleges, vii. Stylopidre, 30. Swarming of bees, 10. Synaptomys cooperii, 99-104. stonei, 99-104. T. Telepathy in insects, 43-46. Tent caterpillars, 3. Teonoma, 127. Termes, 74. Ternies bellicosus, 34. colony, forms in, 33. composition of colony, 73-74. flavipes, 32, 35. lucifugus, 31, 33, 35. Termite economy, 71-74. Termites, 31-36/42, 47, 56, 57. influence of food and treat ment on, 72-73. supplementary kings and queens of, 71-72. true royal pairs of, 71-72. Tetramorium, 25, 29. Thompson, Ernest E., Means of Inter communication among Wolves, xiv. Townsend, C. H., Ornithology of Cocos Island in Relation to that of the Gala pagos Archipelago, viii. Triarthus, 92, 93, 95. becki, 96. Trigona, 17, 65, 66, 67. Trilobites, appendages of, 89-97. Tysonia marylandica, 86. V. Vespa, 22. maculata, 20, 70. Vespidre, 19, 23. W. Waite, M. B., Hexenbesens of Wash ington and Vicinity, viii. , Flora of Washington and Vicinity, xii. , Structure and Method of Opening of Anthers of the Pomeae, viii. , Treatment of the Pear Leaf-Blight, vii. Walcott, Charles D., Appendages of Trilobites, ix, 89-97. , Occurrence of Fossil Medusrc in the Middle Cambrian Terrane, ix. Ward, Lester F., Fossil Cycads from Potomac Formation of Maryland, ix. , Fossil Cycadean Trunks with Re vision of genus Cycadeoidea, 75-87. Alphabetical Index. 145 Ward, Lester F., Mesozoic Flora of Por tugal compared with that of United States, xii. , Kemarks on the genus Caulinites Brongn., xv. Wasps, 19, 47, 48, 53, 54. Life history of, 21. Paper, 20. Social, 19-23. Wax discs, 16. glands, 12. pincers, 13. producing organs in bees, 12, 67-68. Webber, H. J., Dissemination of the Yucca, x. Weismann's theory of heredity, 51-52. Woods, A. F., Calorific Effect of Light upon Plants, ix. Woods, A. F., Some Effects of -Spraying Mixtures on growth of Plants, x. X. Xiphosura, 93. Y. Yellow jackets, 20. Z. Zamites bucklandi, 81. megalophyllus, 79. microphyllus, 80. pygmseus, 80. Zamiostrobus emmonsi, 86. mirabilis, 75, 86. Zapus hudsonius, 99, 100. VOL IX, pp. 133-138 SEPTEMBER 15, 1894 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON DESCRIPTION OF A NEW MUSK RAT FROM CODROY, NEWFOUNDLAND. BY OUTRAM BANGS. A fine series of musk rafcs lately received from Codroy, New foundland, shows such differences from Fiber zibethicus (Linn.) of Eastern North America that I think it must be regarded as representing a distinct species. This Fiber may be defined as follows : Fiber obscurus sp. nov. Diagnosis. Smaller than F. zibethicus (Linn.), with the under parts and sides less ferruginous and the upper parts much darker. The lips and the hairs about the under side of the nose are much lighter (pure white). The fringe of long hair around the toes of the hind feet is decidedly lighter than the feet, while in F. zibethicus this fringe is generally of about the same shade as the feet. The under fur is darker than in zibethicus. Skull smaller and smoother, not rising into such pronounced bong ridges with age, with the rostrum relatively larger and the audital bullae rela tively smaller, while the interorbital constriction is actually broader. 134 Bangs A Neiv Musk Rat from Newfoundland. Description. Type. No. 1155 Collection of E. A. and O. Bangs, Boston, Mass. From Codroy, Newfoundland, May 14, 1894, Ernest Doane, collector. Total length, 476 mm.; tail, 200 mm., hind foot, 70 mm.; ear from notch, 22 mm. (Measurements taken in flesh by the collector). Upper parts: General color Front's brown,* shaded a little with Vandyke brown,* darkening on the top of the head to almost black. The long shin ing hairs are black all over the back, but on the sides they shade off a little to a very dark reddish brown. Under parts and sides of the head of a shade between Prout's brown,* and broccoli brown,* paling off on the under side of the neck and legs to almost fawn color.* Lips and hair under the nose white. Under fur slate grey everywhere. Feet blackish slate,* with the fringe of long hair around the toes of the hind feet Isabella color.* Cranial characters: Skull small and smooth, and broad between the orbits, with large rostrum and small audital bullae. Specimens taken later in the season than the type are rather more reddish (ferruginous) in general color, but compared with specimens of F. zibethicus of corresponding dates from Massa chusetts, Nova Scotia, etc., the difference is quite as great as be tween the type and examples of F. zibethicus of the same date as the type from those localities. I have carefully compared musk rats from our Eastern Coast from Connecticut to Cape Breton and can find no difference in specimens of the same age and season and no tendency towards a small dark northern race,f and therefore incline to the belief that F. obscurus is an insular form peculiar to Newfoundland. I have however never seen any musk rats from Labrador, and it is possible that there the animal may be more like the one found in Newfoundland. *Ridgway's Nomenclature of colors. fThe largest musk rat I have examined is from Shenacadie (Cape Breton) Nova Scotia. It is a very old adult male. No. 2007, Collection of E). A. and O. Bangs, and measures, total length, 632 mm.; tail, 291 mm.; hind foot, 87 mm.; ear from notch, 26.5 mm. Bangs A New Musk Rat from Newfoundland. 135 snatupads ua; 91^ jo sjuauiaanseam 9SB.I9AV BAON Of BI;OOS BAON o XBJltBH co co iot^ocoio T i i I-O5COCO C * ^ CO t^- O* GO iO CO CM CO ^CMCOGO^COGO^ ^GO^t ^kOCO CM CMCOrHrHrHrHCO TJH CO lo ^ GO CM GO COCMTfGOCM Tt< GO t-^ CM OO rH ^f iO CO CM CM CO 1 C7^ O^ O^ O^ OS CS GOCOCOOOOOGOGOCOGOCO ; aJ CS a (M (M rH n CO >fcCbt,^r2 >>>j>>^>i 5 4*5 3 3'3'dd'd l-5-^-i^t^i>-5Ocor^i>. t^ GO t! GO QO GO OO OO QO GO QO OO GO ^ " 03 c3 * 33 , J S I ; 1 i s m - n a ^ 1 I * I 5 B I O = iO^:OX>O 1 CQ ll a* 'cS o ^ 5 S, 'o Name. Locality. a 2 | S "5 s '-2 11 sl "So Cfc "So 03 i bO c3 o ^ O "U i 'o c t-, a, a e? i 03 p. C^ 50 .2 U '*' o -w J-J 0) "^ c _o |^2 ** . Q^ o -^ c a Q} s s if! 3 "cS "cS * 11 1 O2 * CQ o O O M s O P. lonqicauda Alberta, Sask., and N. Dak cfad. 6 47.4 46.8 29.6 25.8 14.0 3.7 29.0 Alberta, Mont., and N. Dak 2 ad 6 42.8 42.4 25.7 22.9 12.0 3.9 25 4 P. 1. spadix Fort Sn'elling and Elk river, qp n\4i cf ad. 5 47.6 25.8 14.4 3.3 30.4 Minn. 9 ad tt tt 3 "o 1 | T3 a 5 No. of specimens in average. Putorius lonqicaudci.... Wingard, Sask cf old ad. 466.0 173.0 50.0 1 Osier, Sask (f ad 440.5 158 2 (i u Wingard Sa^k 9 old ad 417 162 48 1 (t South Edmonton, Alberta 9 yg. ad. 389.0 145 45 1 Putorius longicctuda spadix Putorius brasiliensis frenatus... . Fort Snelling, Minn, (type loc.) Brownsville, Texas cf ad 9 old ad. cf ad Q ad 461.0 375.0 496.4 416.0 170.5 123.0 220.0 172.0 54.4 425 47.1 36.6 5 1 7 3 Tarpon Springs, Fla . .. 9 ad.. 374.0 127 44 5 1 Putorius noveboracensis Liberty Hill, Conn cf ad 407.9 138.4 47 5 10 9 ad 3190 106 5 33 7 6 Putorius richardsoni.. Fort Franklin, Great Bear lake cf ad * 1020 1 Fort Simpson Mackenzie river cf ad f 107 5 43 1 Codroy Newfoundland cf ad 327 3 94 7 46 1 7 Hucksport, Maine Ossipee N H cf ad cf ad 2SK9 277.8 82 1 79 6 39.2 36 6 5 5 it Liberty Hill, Conn. cf ad.. 2fil.O 600 34.0 1 U 1C (( Codroy Newfoundland 9 ad . 2567 77 3 33 3 3 u tt <( VI t Forest Ontario 2 ad 258 5 74 31 5 1 Putorius rixosus (type) Osier Sask 9 ad. 181 31 I Putorius ermineus Yukon, mouth of Porcupine river, Alaska. England 9 ad cfad 9 ad 177.5 393.0 3140 17.5 123.0 88 470 400 1 1 Scoiland. cf ad .. . 257.5 52.0 31 2 9 ad 237 49.0 26 1 * Head and body, 280. f Head and body, 261. EXPLANATION OF PLATES. PLATES I AND II. (Explanation of figures the same in both plates. ) Fig. 1. Putorius longicauda (Bonap.). 1. C? ad. Wingard, Saskatchewan (No. 73183, U. S. Nat. Mus., Dept. Agric. coll.). la. 9 ad. Wingard, Saskatchewan (No. 75483, TJ. S. Nat. Mus., Dept. Agric. coll.). 2. Putorius noveboracensis Emmons. 2. J* ad. Adirondacks, New York (No. 3843, Merriam coll.). 2a. 9 ad. Adirondacks, New York (No. 5598, Merriam coll.). 3. P"torius richardsoni (Bonap.). 3. $ ad. St. Michaels, Alaska (No. 36243, U. S. Nat. Mus.). 3a. 9 ad. St. Michaels, Alaska (No. 36246, U. S. Nat. Mus.). 4. Patorius richardsoni cicognani (Bonap.). 4. cf . Bucksport, Maine (No. 4247, coll. E. A. and O. Bangs). 4a. 9 . Mt. Forest, Ontario (No. 789, coll. E. A. and O. Bangs). 5. Patorius peninsulss Rhoads. 9 old. Tarpon Springs, Florida (No. 2379, coll. S. N. Rhoads). 6. Patorius rixosus nob. 9 ad. (type). Osier, Saskatchewan (No. 642, coll. E. A. and O. Bangs). PLATE III. Fig. 1. Patorius longicauda (Bonap.). 1. cT ad. Wingard, Sask. (No. 73183, U. S. Nat. Mus., Dept. Agric. coll.). la. 9 ad. Wingard, Sask. (No. 75483, U. S. Nat. Mus., Dept. Agric. coll.). 2. Putorius richardsoni cicognani (Bonap.). 2. H ,Q -2 t O 1 1 r^ o 'a, I 1 '_S "H 1 S "S Q) 1 | i o> CO PQ 1 I 3 E CD JS 6 H O a> C. americanus. 4999 (f old ad. Maine, Bucksport 296.0 259.5 136.0 109.0 82.0 96.0 26.0 79.5 240.5 82.0 5000 9 old ad. 259.0 231.0 121.5 90.0 72.0 90.0 24.0 66.5 221.0 73.5 c. osceola. 2391 (f old ad. Florida, Blitches Ferry, 254.5 229.5 113.0 91.0 70.0 87.5 17.5 76.0 216.0 78.0 Citrus county. 2392 (f old ad. Florida, Citronelle, Cit 263.0 241.0 1145 94.0 67.0 93.0 18.0 77.0 229.5 84.0 2394 9 ad rus county. Florida, Citronelle, Cit 220.0 202.0 91.0 70.5 51.5 83.0 14.0 730 194.5 80.0 rus county. 2393 $ad Florida, Citronelle, Cit 211.0 197.5 94.5 690 59.9 80.0 15.5 72.5 189.0 82.0 rus county. VOL. X, PP. 29-43 FEBRUARY 26, 1896 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON. FOURTH LIST OF ADDITIONS TO THE FLORA OF WASHINGTON, D. C. BY THEO. HOLM.* The present paper is supplemental to Prof. Lester F. Ward's 1 Guide to the Flora of Washington and Vicinity.' f It con stitutes the fourth addition, and I have followed the same prin ciples of nomenclature and arrangement as in the above-cited work of Professor Ward. I have also, in accordance with my previous addition, J not only enumerated such species as are new to the flora, and which are marked with an 'asterisk, but have also included a number of the rarer species recorded from new localities. In spite of the fact that comparatively few of the Washington botanists are interested in the local flora, we have, nevertheless, succeeded in accumulating a very considerable amount of ma terial for the study of the geographical distribution of plants within the limits of the flora. It is very interesting to see that several of the plants that have been recorded as new in the later additions are only new so far as concerns Professor Ward's flora. Several of them were already recorded by Brereton in his Pro- dromus, but not included in the list given by Professor W^ard. * Presented at a meeting of the Biological Society of Washington, No vember 30, 1895. t Bulletin No. 22, U. S. National Museum, 1881. J Holm, Theo. : Proc. Biol. Soc., vol. vii, 1892, pp. 105-132. % Brereton, John A.: Florae Columbians Prodromus. Washington, 1830. 5 BIOI,. Soc. WASH., VOL. X, 1896 (29) 30 Holm Additions to the Flora of Washington. These species belong almost exclusively to the fourth category, under which Professor Ward (1. c., p. 14) has designated a num ber of plants which he thought had either been exterminated or had accidentally disappeared since the publication of the Pro- dromus. Some of these plants are now known to be quite abun dant in certain localities ; for instance, Apocynum androsaemi- folium, Rhexia mariana, Polygala verticillata, Polygonum tenue and Cyperus flavescens. This fact seems to show that the flora of the District is yet but imperfectly known. There are, indeed, many and most in teresting localities that have not yet been explored, and I will merely recall the fact that the southern and southeastern part of the District is still almost unknown. There is that interest ing locality where Mr. G. W. Oliver obtained so many rare plants, which was recorded in the third addition as " the vicinity of Silver Hill." This locality comprises the large forests, with creeks and sphagnum swamps, which lie between Silver Hill post-office and Surattsville, Prince George's county, Md. A visit to these forests shows us only too clearly that there is certainly a good deal of work to be done before we should venture to declare that we know our local flora. We meet here with species which are either not recorded in Professor Ward's flora or only enumerated as rare and of which several have been observed to occur in the greatest abundance, such as Ilex glabra, Rhyncospora cephalantha, Agrostis elata, Uniola gracilis, Aristida purpurascens, Xyris flexuosa, etc. The same is the case with several other places which have only been lately explored. Brookland, for instance, is the home of twenty species of Panicum, several of which have never been noticed before, although they are very common; there grow, also, Apocynum androsaemifolium, Sporobolus vaginseflorus and Agrostis elata quite abundantly. But, besides these truly indigenous species, I have also noted some others, which have been accidentally introduced e. g., Tribidus maximus * and Leptochloa mucronatarf both of which ap peared suddenly in the Agricultural grounds. Leptochloa was very abundant, with ripe seeds, and it may, therefore, also spread to other parts of the city and become a well-established citizen. Among the species enumerated in the following list are several * Found by G. H. Hicks, f Found by the author. Additions to the Flora of Washington. 31 which Professor AVard had considered so common that no spe cial locality was recorded. Extended researches have, however? induced me to arrange some of these among the rarer plants, and I thought, also, that in this way a better idea of their dis tribution might be obtained. 3. Clematis Virginiana L. Along the Canal road near the Eighth lock. G. H. Hicks. 10. Anemone nemorosa L. Banks, Four-Mile Run. R. R. Gurley. 13. Ranunculus pusillus Poir. A very large specimen was collected in a stagnant pool by Piney Branch, at the north end of Seventeenth street. G. H. Hicks. 26. Aconitum uncinatum L. Near Chevy Chase. G. H. Hicks. 40. Papaver dubium L. Bank of Potomac above Rosslyn, Va., near the Seventh lock along the Canal road. G. H. Hicks. 45. Fumaria officinalis L. Along Bates road near Bunker Hill. The author. 47. Nasturtium sylvestre R. Br. Potomac flats near the Seventh lock. G. H. Hicks. 51. Nasturtium Armoracia Fries. Along; a ditch near Terra Cotta, escaped. The author. 55. Arabis dentata T. & G. Along the Canal road near the Seventh lock. G. H. Hicks. 61. Cardamine hirsuta L. (C. intermedia Horn.) Rock Creek. G. H. Hicks. The Zoological Park; in full bloom the 10th of May, 1895. The author. * 62a. Cardamine parviflora L. Shaded woods in Brookland ; in moist, rich soil near Terra Cotta. The author. * 62b. Cardamine silvatica Link. Terra Cotta swamp ; Rock Creek. The author. *62c. Cardamine Pennsylvanica Muhl. Meadow near Rosslyn. T. H. Kearney, Jr. Rock Creek near the Zoological Park ; in creeks on the Virginia shore of the Potomac, one mile above Aqueduct Bridge. In flower the second week of May. The author. It seems, according to the Synoptical Flora of North America,* that these species of Cardamine are not well understood in this country. The difficulty in distinguishing them is evidently due to insufficient European material for comparison. The name Cardamine hirsuta L. is very mislead- * Vol. I, part I. Continued and edited by B. L. Robinson, 1895, page 158. 32 Holm Additions to the Flora of Washington. ing, and I have therefore added as a synonym C. intermedia Horn. This name is used by Professor Lange,* since Linnaeus undoubtedly included Link's sylvatica under his hirsuta, until Hornemann separated them, naming the last of these ' intermedia.' f They are very different from each other, and we add the following characters as supplemental to those already given in the Synoptical Flora (1. c.) : Cardamine hirsuta L. (C. intermedia Horn). Basal leaves very numerous and persisting for a long time, large and forming a dense rosette; they are most often smooth; the terminal leaflet is larger than the lateral ones, orbicular or reniform. Flowering stems, several, simple or with a few branches, ascending from a decum bent base. The uppermost pods surmount the flowering part of the raceme. This species is said to be "abundant about Washington, D. C.," which is evidently a mistake. It might have been confounded with C. sylvatica Link. Cardamine parviflora L. The basal leaves are few in number and early fading ; the plant is strictly annual ; otherwise the description in the Synoptical Flora (1. c. ) corresponds very well to this species. Cardamine silvatica Link. This species is not mentioned in the Synoptical Flora. The basal leaves are few, early fading; those of the stem numerous, with large and broad divisions, the margin dentate. The pods are borne on more or less hori zontal pedicels, and the uppermost ones are hardly surpassing the flower ing part of the raceme. The plant is most often biennial in the vicinity of Washington, but occurs also as a perennial in Europe. 63. Dentaria heterophylla Nutt. The Zoological Park. The author. *63a. Dentaria cardiophylla Robinson. Described in the Synoptical Flora (1. c., p. 155). Collected in Rock Creek by George R. Vasey. 72. Camelina sativa (L.) Crantz. Near Rosslyn, Va. T. H. Kearney, Jr. *77a. Lepidium Draba. Lincoln Park. A. G. Maesius. 80. Helianthemum Canadense Michx. Sand hills near Terra Cotta. The author. 82. Viola lanceolata L. Bladensburg. M. B. Waite. 90. Viola striata Ait. Near Glen Echo, along the Canal road. G. H. Hicks. On the Virginia shore of the Potomac, two miles above Aqueduct Bridge. The author. *Lange, Joh. Haandbog i den Danske Flora. Kjobenhavn, 1586-'88, page 629. t Hornemann, I. W. Oekonomisk Plantelsere, 1821-'39, page 714. Additions to the Flora of Washington. 33 94. Viola tricolor L., var. arvensis Ging, is V. tenella Muhl. 96. Polygala incarnata L. Near Chevy Chase ; Takoma. G. H. Hicks. In pine woods in several places around Brookland. The author. 99a. Polygala Curtissii Gr., var. pycnostachya Gr. Near Carlins, Va.; Takoma. G. H. Hicks. Terra Cotta swamp; near Surattsville, Prince George County. G. W. Oliver and the author. 100. Polygala ambigua Nutt. In a dry field along the Walker road near Surattsville. The author. 101. Polygala verticillata L. Bunker Hill; Brookland. The author. Near Chevy Chase. G H. Hicks. * 104ffl. Saponaria Vaccaria L. Along the railroad track near University Station ; in bloom third week of May, 1 894. The author. 106. Silene nivea D. C. Bank of Potomac, two miles above Aqueduct Bridge, Va. G. H. Hicks. 108. Silene Armeria L. Among wheat near Garrett Park. The author. 122. Anychla dichotoma Michx. Open places in the pine woods around Brookland. The author. 145. Linum striatum Walt. Very abundant in Terra Cotta swamp, near the railway track. The author. 149. Geranium columbinum L. Hillside near Aqueduct Bridge, on the Virginia shore. The author. 158a. Ilex glabra (L.) Gr. Abundant in woods along the Walker road between Camp Spring P. 0. and Surattsville. G. W. Oliver, W. T. Swingle, and M. B. Waite. 161. Ilex leevigata Gr. With the preceding. 174. Acer saccharinum Wang. This species has spread from the city parks to the woods between Eckington and the Catholic University. The author. 194a. Trifolium medium L. Reported by Mr. G. B. Sudworth and published in the Third Addition to the Flora of the District ; not this species, but merely a form of T. pratense L. 216. Desmodium ciliare D. C. Common in open pine woods, Brookland. The author. 217. Desmodium Marilandicum Boott. Very scarce in dry fields, Brookland. The author. 230. Clitoria Mariana L. Near Carlins, Va. , near Fort Reynolds, Va. G. H. Hicks. 34 Holm Additions to the Flora of Washington. 234. Phaseolus perennis Walt. Near Carlins, Va., and near Fort Reynolds. G. H. Hicks. 250. Spiraea Aruncus L. Very common near Garrett Park. The author. 256. Rubus hispidus L. Exceedingly abundant in the swamps near Surattsville. W. T. Swingle, M. B. Waite and the author. 259. Rubus cuneifolius Pursh. Dry hillside near Silver Hill P. O. The author. 270. Poterium Canadense B. & H. Terra Cotta swamp. The author. 272. Rosa setigera Michx. Along Rock Creek below Pierce' s Mill. G. H. Hicks. 278. Pirus coronaria L. South Brookland. Robert Ridgway. 295. Ribes floridum THer. Near the Baptist Church in Brookland, 1895. Now destroyed. The author. 300. Drosera rotundifolia L. . Common in swamps between Camp Spring P. O. and Surattsville. G. W. Oliver and the author. 304. Proserpinaca palustris L. Not common. One mile above Glen Echo. T. H. Kearney, Jr. 304a. Callitriche Austin! Eng. Several places in the woods around Brookland. The author. It is evi dently not rare, but overlooked. It grows in deep shade on black, rich soil. 305. Callitriche verna L. Pool near Piney Branch, at the north end of Seventeenth street. G. H. Hicks. 316. Oenothera pumila L. Near Surattsville. W. T. Swingle and M. B. Waite. 317. Oeiiothera sinuata L. A single specimen was collected in a dry field, South Brookland. The author. 327. Hydrocotyle ranunculoides L. Under Bennings bridge. The author. Above Rosslyn. G. H. Hicks. 339. Chaerophyllum procumbens Crantz. Not common, and has only been observed along the Potomac shore, on both sides of the river, but especially on the Virginia side. It has been collected on the Maryland side along the Canal, near the Seventh lock, by G. H. Hicks. 345. Pastinaca sativa L. Clover field in South Brookland ; along the electric car track near Eck- ington. The author. Additions to the Flora of Washington. 35 *348. Caucalis Anthriscus Huds. A single specimen near Bathing Beach, back of the Monument grounds ; July 5, 1894. G. H. Hicks. 351. Aralia nudicaulis L. High Island. T. H. Kearney, Jr. Blagden's Run. G. H. Hicks. 384. Fedia olitoria Vahl. Island of the Potomac, near the Seventh lock. G. H. Hicks. South Brookland. The author. The specimens which Mr. Hicks has collected were rather robust for the typical species. The corolla was, however, light blue. Therefore I have not hesitated in recording them under this species. 3906. Eupatorium semiserratum D. C. Great Falls. The author. 391. Eupatorium hyssopifolium L. Woods near Glen Echo. G. H. Hicks. Brookland. The author. 391a. Eupatorium altissimum L. Island in Potomac, near the Seventh lock. G. H. Hicks. 400. Eupatorium aromaticum L. Rock Creek. G. H. Hicks. 402 Mikania scandens L. Swamp between Eckington and the Catholic University. Robert Ridg- way. 426. Sericocarpus solidagineus Nees. Rather rare. Dry field, Surattsville. The author. 430. Aster concolor L. Surattsville. The author. 494a. Bidens connata Muhl., var. comosa Gray. Bank of Potomac, opposite mouth of Cabin John's Run, Md. G. H. Hicks. Ditch near Metropolis View ; along a creek near the Catholic Uni versity. The author. 497a. Galinsoga parviflora Cav. Near Bureau of Engraving and Printing. G. H. Hicks. Massachusetts avenue, between Fourteenth and Fifteenth streets. The author. 504a. Senecio vulgaris L. Potomac flats near Bureau of Engraving and Printing. G. H. Hicks. 536. Tragopogon porrifolius L. Near Eckington, escaped. The author. 539. Lobelia puberula Michx. Rock Creek. G. H. Hicks. Swamp near Silver Hill. The author. 542. Specularia perfoliata A. DC. A form without corolla is common in shaded woods around Brookland. The author. 543. Campanula Americana L. Island of the Potomac, opposite the Eighth lock. G. H. Hicks. 36 Holm Additions to the Flora of Washington. 543a. Campanula aparinoides L. Terra Cotta swamp ; quite abundant among Lilium snperlmm. The author. 551. Gaultheria procumbens L. Burnt Mills ; near Surattsville. M. B. Waite. 564. Pyrola secunda L. Garrett Park. P. H. Dorsett. 565. Pyrola chlorantha Swtz. Garrett Park. P. H. Dorsett. 566. Pyrola elliptica Nutt. Garrett Park. The author. 567. Pyrola rotundifolia L. Near Chevy Chase. G. H. Hicks. 569. Monotropa Hypopitys L. The typical form is not rare in the District and seems to correspond exactly to the European plant. But a very peculiar form has also been collected, which differs from the description given of M. Hypopitys so much that it seems to represent a distinct species. It is, for instance, of a red, almost blood-red, color and densely pubescent all over. I have secured some material of the European plant, preserved in alcohol, and it is my intention to give a detailed account of the characters, morpho logical and anatomical, so as to decide whether we shall consider this plant to represent a species or only a variety. It has been collected in Rock Creek by G. H. Hicks and near Arundel (Anne Arundel County, Md.), by M. B. Waite. 570. Dodecatheon Meadia L. Near Cabin John's Bridge. A. J. Pieters. 585a. Apocynum androssemifolium L. Very common in low grounds in South Brookland. The author. 589. Asclepias rubra L. Sphagnum swamp near Surattsville. J. Krause and the author. 594. Asclepias obtusifolia Michx. Not common. Brookland. The author. 595. Asclepias variegata L. The Zoological Park ; Blagden's mill-race. G. H. Hicks. Brookland. The author. 597. Asclepias verticillata L. In pine woods at Brookland. The author. 605. Gentiana ochroleuca Froel. Blagden's Run ; Glen Echo. G. H. Hicks. Brookland. The author. 608. Phlox paniculata L. Near Glen Echo. G. H. Hicks. 615a. Phacelia Covillei Wats. Island near the Seventh lock. G. H. Hicks. Very abundant on High Island. M. B. Waite. Additions to the Flora of Washington. 37 617. Phacelia parviflora Pursh. About two miles above Aqueduct Bridge, on the Virginia shore of the Potomac. The author. Along the Canal road near the Seventh lock. G. H. Hicks. 623. Myosotis laxa Lehm. In a pool near the north end of Seventeenth street and Piney Branch. G. H. Hicks. 634. Ipomaea lacunosa L. The Canal road near the Seventh lock. G. H. Hicks. 635. Convolvulus spithamaeus L. Very abundant in open thickets in North and South Brookland. The author. Ivy City. T. H. Kearney, Jr. 637. Convolvulus arvensis L. Along the electric car track between Eckington and the Catholic Univer sity. The author. 643. Physalis pubescens L. Bank of the Potomac back of the Twelfth lock. G. H. Hicks. 643a. Physalis Philadelphica Lam. Bank of Potomac opposite mouth of Cabin John's Run. G. H. Hicks. 657. Pentstemon laevigatus Soland. Common in South Brookland. The author. 660. Herpestis nigrescens Bth. Great Falls (Virginia side). M. B. Waite. 666. Veronica Americana Schwein. Potomac Boat Club landing, Virginia. G. H. Hicks. 669. Veronica serpyllifolia L. Woodley Park. The author. 671a. Veronica hederifolia L. The Virginia shore of the Potomac, two miles above Aqueduct Bridge. The author. 680. Melampyrum Americaiium Michx. Takoma Park. G. H. Hicks. 681. Orobanche minor L. Magnolia Run. G. H. Hicks. 682. Aphyllon uniflorum Gr. North Brookland. The author. 686. Utricularia gibba L. Opposite the Eleventh lock, Maryland. G. H. Hicks. 724. Monarda punctata L. Not common. Rock Creek near Brightwood. G. H. Hicks. North Brookland. The author. 725. Lophanthus nepetoides Bth. Bank of the Potomac near the Twelfth lock, Maryland. G. H. Hicks. 732a. Scutellaria parvula Michxi On dry, sandy soil, North and South Brookland. The author. 6 BIOL. See. WASH., VOL. X, 1896 38 Holm Additions to the Flora of Washington. 745. Plantago Patagonica Jacq. var. aristata Gr. East Brookland ; around Allentown, Prince George County, Md. The author. Near Fifteenth and W streets. G. H. Hicks. Near Chevy Chase. M. B. Waite. 747. Amarantus paniculatus L. Near Pierce' s mill. G. H. Hicks. Around Silver Hill, Prince George County. The author. 749. Amarantus albus L. Near Bureau of Engraving and Printing. G. H. Hicks. *769a. Polygonum Muhlenbergii Watson. Potomac flats, opposite the Eighth lock, Maryland. G. H. Hicks. 773a. Polygonum tenue Michx. Abundant in dry fields, Brookland ; sandy soil near Terra Cotta. J. Krause and the author. 780. Rumex Britannicus L. Potomac flats near Rosslyn, Va. G. H. Hicks. 781. Rumex verticillatus L. With the preceding. G. H. Hicks. 800. Euphorbia commutata Engelm. Glen Echo. T. H. Kearney, Jr. and G. H. Hicks. 843. Quercus ilicifolia Wang. Abundant near Laytonsville and Goshen, Montgomery County, Md. George B. Sudworth. 901. Habenaria tridentata Hook. Near Chevy Chase. G. H. Hicks. North Brookland. The author. 904. Habenaria lacera R. Br. Terra Cotta swamp. The author. Swamp near Surattsville. J. Krause and the author. 911. Pogonia ophioglossoides Nutt. Sphagnum swamp near Surattsville. W. T. Swingle and M. B. Waite. 912. Pogonia verticillata Nutt. North Brookland ; near Terra Cotta. The author. 922. Cypripedium pubescens Willd. Blagden's Mill, Rock Creek. R. R. Gurley. Burnt Mills. M. B. Waite. South Brookland. The author. 925. Aletris farinosa L. Very common around Surattsville. The author. Near Arundel, Md. M. B. Waite. 955. Trillium sessile L. Island near the Seventh lock. G. H. Hicks. 957. Veratrum viride Ait. Magnolia Run. L. H. Dewey. Garrett Park. The author. *962a,. Muscari racemosa Nutt. Island near the Seventh lock. G.' H. Hicks. Additions to the Flora of Washington. 39 982. Commelina hirtella Vahl (C. erecta L) Bennings Bridge. The author. *982. Commelina communis L. Magnolia Run, Rock Creek Park. G. H. Hicks. 983. Commelina Virginica L. Flats of the Potomac under Chain Bridge. The author. 985. Xyris flexuosa Muhl. Along the railroad track near Carlins, Va. G. H. Hicks. Sphagnum swamps in the woods near Surattsville. The author. 988. Cyperus aristatus Rottb. (C. inflexus Muhl.). Muddy bottom of Potomac, opposite mouth of Cabin John Run, Md. G. H. Hicks. 998. Cyperus retrofractus Torr. South Brookland. The author. *999a. Kyllinga pumila Michx. Holmead swamp. M. B. Waite. * lOOOa. Hemicarpha subsquarrosa Nees. Potomac flats opposite the Ninth lock, Md. G. H. Hicks. *1004rt. Bleocharis intermedia Schult. Opposite the Eighth lock, Md. G. H. Hicks. 1007. Scirpus planifolius Muhl. On dry, grassy slopes in the Zoological Park. The author. 1010. Scirpus debilis Pursh. ' Potomac flats opposite the Ninth lock, Md. G. H. Hicks. Sphagnum swamp near Silver Hill. The author. 1017. Eriophorum Virginicum L. Sphagnum swamp near Surattsville. The author. 1019, Fimbristylis capillaris Gr. Along the railroad track near Carlins, Va. G. H. Hicks. North Brookland. A. J. Pieters. 1020. Rhyiicospora alba Vahl. Sphagnum swamp near Surattsville. The author. 1021a. Rhyncospora cephalantha Gr. Common in swamps around Surattsville and Silver Hill. The author. 1024. Scleria pauciflora Muhl. Grassy knoll near Aqueduct Bridge on the Virginia shore of the Poto mac. The author. 1025. Carex polytrichoides Muhl.* Magnolia Run, Rock Creek Park ; Sphagnum swamp near Suratts ville. The author. 1026. Carex Willdenovii Schk. Bunker Hill ; Corco ran's woods. The author. * Professor Charles F. Wheeler has kindly revised the entire collection of the genus Carex recorded here. 40 Holm Additions to the Flora of Washington. *1030a. Carex conjuncta Boott. High Island. The author. * 10306. Carex alopecoidea Tuckerm. Low meadow near the Insane Asylum. The author. 1031. Carex stipata Muhl. Rock Creek woods. G. H. Hicks. Near Surattsville. The author. *1035a. Carex Muhlenbergii Schk., var. enervis Boott. High Island and along the Canal road ; not rare. G. H. Hicks and the author. Bunker Hill. The author. 1040. Carex lagopodioides Schk. The name should be changed to C. tribuloides Wahlbg. *1040a. Carex tribuloides Wahlbg., var. reducta Bailey. Common in shady places around Brookland ; Insane Asylum. The author. 1047. Carex torta Boott. Cabin John Run. G. H. Hicks. 1051. Carex Shortiana Dew. High Island, near the river shore. The author. 1055. Carex glaucodea Port. Bunker Hill. Open woods in North Brookland. The author. Along the towpath, C. and O. Canal, near Georgetown. T. H. Kearney, Jr. 1056. Carex pallescens L. Grassy knoll near Aqueduct Bridge, on the Virginia shore of the Potomac. The author. 1063. Carex platyphylla Carey. Shaded places on the rocks near Aqueduct Bridge, on the Virginia shore of the Potomac. The author. 1065. Carex retrocurva Dew. On the rocks near Aqueduct Bridge, on the Virginia shore of the Potomac. The author. 1067. Carex laxiflora Lam. The type seems to be rare ; near the Insane Asylum. Lester F. Ward. Brookland. The author. * 107 2. Carex laxiflora Lam., var. divaricata Bailey. Near Washington, D. C. George Vasey. Rock Creek. G. H. Hicks. 10726. Carex laxiflora Lam., var. patulifolia Carey. Common in the woods around Brookland. The author. 1074. Carex oligocarpa Schk. Rare. On shaded rocks on the Virginia side of the Potomac near Aqueduct Bridge. The author. 1075. Carex umbellata Schk. Bunker Hill ; sandy hills around Terra Cotta. The author. 1076. Carex Emmoiisii Dew. In Professor Ward's catalogue is C. varia Muhl. Additions to the Flora of Washington. 41 *1078. Carex communis Bailey. This species has undoubtedly been observed before in the District, but confounded with C. Pennsylvanica Lam. It is closely related to this last, from which it differs, however, by its cespitose growth without stolons. The scales of the staminate and fertile inflorescences are usually purplish, but lighter than those of C. Pennsylvamca. It is very common on the rocks of the Potomac shore, Virginia. The author. *1078/>. Carex communis Bailey, var. Wheeleri Bailey. With the preceding. The author. 1080. Carex pubescens Muhl. Cabin John Run ; the Zoological Park. The author. 1081. Carex prasina Vahl. (C. miliacea Muhl.). Woodley Park ; the Potomac shore, three miles above Aqueduct Bridge, Va. The author. 1090. Carex lupulina Muhl. Near Rosslyn, Va. ; Potomac flats. G. H. Hicks. The Zoological Park. William Hunter. 1091. Carex folliculata L. Takoma. T. H. Kearney, Jr. Magnolia Run, Rock Creek Park ; abundant in the woods along Walker road, between Camp Spring P. O. and Surattsville. The author. 1092. Carex squarrosa L. Near Chevy Chase circle. G. H. Hicks. Garrett Park. The author. 1093. Carex stenolepis Torr. Common in South Brookland. The author. 1101. Sporobolus asper Beauv. High Island. L. H. Dewey. llOla. Sporobolus vaginaeflorus Torr. Common in low grounds, Brookland ; the lawns in the Smithsonian Park, and around the Catholic University. The author. *1104a. Agrostis elata Trin. Common in pine woods, Brookland ; also in the deciduous forests along Walker road, near Silver Hill, Surattsville, etc. The author. 1107. Muhlenbergia sobolifera Trin. Woods in North Brookland , but very scarce. The author. 1114. Calamagrostis Nuttalliana Steud. Takoma. F. L. Scribner, T. H. Kearney, Jr., and L. H. Dewey. 1119. Aristida purpurascens Poir. Along the roads ; very common near Allentown and Surattsville. 1125a. Eatonia obtusata Gr. Dry fields, North Brookland. The author. *1126a. Eatonia Dudleyi Vasey. Brookland. Terra Cotta. The author. Rock Creek. L. H. Dewey. This species is recorded in Professor Ward's Flora of the District as a slender wood form of E. Pennsylvanica Gr. 42 Holm Additions to the Flora of Washington. * 11266. A hybrid between Eatonia Pennsylvanica Gr. and Trise- tum palustre L. Has been described by Dr. George Vasey,* who found several speci mens of this peculiar form in a low meadow on the banks of Hunting Creek, near where it empties into the Potomac River, a mile below Alex andria, Va. 1136. Poa sylvestris Gr. High Island; on the Virginia shore of the Potomac near Aqueduct Bridge. The author. 1137. Poa flexuosa Muhl. The Zoological Park. The author. 1143. Eragrostis Purshii Schrad. Abundant along the railroad track between University Station and Terra Cotta. The author. 1153o. Bromus tectorum L. Near the Navy Yard. The author. 1158. Uniola gracilis Michx. Rock Creek. G. H. Hicks. Several places in the woods between Silver Hill and Surattsville. The author. 1172. Phalaris Canariensis L. Waste grounds south of the Bureau of Engraving. G. H. Hicks. 1172a. Phalaris arundinacea L. Ditch near Bates Road. The author. 1179. Panicum proliferum Lam. Very common along the streets in Brookland. The author. *1180a. Panicum capillare L., var. minima Engelm. Common on dry, sandy soil in Brookland ; near Silver Hill. The author. * 1183a. Panicum commutatum Schultes. Extension of Kenesaw Avenue, Rock Creek. G. H. Hicks. Takoma ; Four Mile Run. T. H. Kearney, Jr. Common in Brookland ; near Surattsville ; Terra Cotta. The author. 1185. Panicum microcarpon Muhl. Takoma. T. H. Kearney, Jr. Brookland. The author. 1186. Panicum viscidum Ell. Brookland ; Terra Cotta swamp near the railroad track. The author. 1187. Panicum scoparium Lam. Great Falls. F. L. Scribner. 1187. Panicum sphaerocarpon Ell. Bunker Hill ; North Brookland. The author. Takoma. T. H. Kear ney, Jr. * 1188a. Panicum ramulosum Michx. Sphagnum swamp near Takoma. F. L. Scribner and T. H. Kearney, Jr. * Botanical Gazette, vol. ix, 1884, p. 165. Additions to the Flora of Washington. 43 Panicum nitidum Lam. North Brookland ; Bunker Hill. The author. * 1188c. Panicum lanuginosum Ell. Brookland ; Garrett Park. The author. * 1183d Panicum pubescens Lam. Brookland. The author. 1197. Tripsacum dactyloides L. Near Silver Hill. The author. 1198 Erianthus alopecuroides Ell. Takoma. F. L. Scribner. Field between Jackson City and Arlington, Va. G. H. Hicks. 1219. Woodwardia angustifolia Smith. Very abundant in the woods near Surattsville. The author. 1229. Aspidium cristatum Swtz. Numerous fruit-bearing specimens were observed near Chevy Chase. G. H. Hicks. 1236. Cystopteris fragilis Bernh. Flats under Chain Bridge. The author. 1240. Lygodium palmatum Swtz. Near Arundel, Md. M. B. Waite. 1245. Botrychium ternatum Swtz., var. dissecta Milde. Near Chevy Chase circle. G. H. Hicks. Along Bates road near Terra Cotta swamp. The author. 1247. Ophioglossum vulgatum L. In the woods near Garrett Park. M. B. Waite. 1252. Selaginella rupestris Spring. Rediscovered near Great Falls by M. B. Waite. 1253. Selaginella apus Spring. Swamp near Silver Hill. The author. VOL. X, pp. 45-52 MARCH 9, 1896 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON. ON A SMALL COLLECTION OF MAMMALS FROM LAKE EDWARD, QUEBEC. BY OUTRAM BANGS. Early in September, 1895, my brother, E. A. Bangs, and I made a short collecting trip to Quebec. Our original plan was to spend all our time at Roberval, on Lake St. John, the most northern locality reached by railroad in eastern North America. But Lake St. John proved a great disappointment. The town of Roberval lies in a dreary valley, that seems wholly destitute of mammalian life. The forest has been cleared away and the barren fields and desolate scrub are wholly unfit to supply the needs of even the smaller mammals. Had we been fitted for camping out we could undoubtedly have found a rich field up one of the many rivers that pour their waters from every direc tion into this great basin; but we were not. After wasting two days in a vain endeavor to find any place within walking dis tance of Roberval suited to our work, we turned our backs on Lake St. John and went down the railroad about sixty-five miles to Lake Edward. The town of Lake Edward is on the northern end of the lake of the same name, and lies in the heart of a rich Hudsonian forest. The lake is about twenty-three miles long and termi nates in the Jeannotte river. A great part of the shores of both lake and river are still clothed in primeval forests, but the busy saw-mill at Lake Edward, with its daily consumption of five hundred logs, is fast eating up this old growth and leaving be hind only white birch and small second-growth spruce and fir. 7 BIOL. Soc. WASH., VOL. X, 1896 (45) 46 Bangs Mammals from Lake Edward, Quebec. This forest contains very few species of trees, of which the white birch is the commonest, with spruce and fir in about equal num bers next, and now and then a solitary white pine. The moun tain ash and the spiked maple are very common, but hardly attain to the dignity of trees. In many places where the forest has been burnt a dense growth of raspberry bushes and dwarf cherry immediately springs up, and it is many years before the trees again take possession of the land. The monotony of the forest is here and there broken by little alder swamps along the many brooks, or by open sphagnum barrens with their clumps of Ledum latifolium and Kalmia glauca. In this northern latitude the fallen trees lie on the ground for a long time without decay ing, and the accumulation of centuries covered by a luxuriant growth of moss makes walking through the forest a matter of the greatest difficulty. There are no roads anywhere, all the logging being done by water, but the abundance of lakes con nected by rivers or brooks makes the country very accessible by canoe. Trapping in the northern forest in the tangled mass of fallen trees and granite boulders covered by a deep growth of moss is a very different thing from trapping in open country. In the open southern woods, with but little rubbish on the ground, one takes as much in traps that have been set a week or ten days in one spot as one does the first day, and when the supply is used up, it is then little use to move the trap, as all the small mam mals from near about have already found it. It is not so in the northern forest, where distance means much more and the small mammals are very local and do not travel far. The first day or two will exhaust the supply in one spot, but a move of only a few yards will again yield specimens in about the same number. We were disappointed in not getting Phenacomys, but it is possible that the animal does occur here locally. There were a few mammals we knew to occur in the imme diate vicinity of Lake Edward that we were unable to get, and perhaps it is as well to mention these. Flying squirrels and chipmunks were said by the Indians and French Canadians to occur, but we saw none. Moose and caribou were both quite plentiful. I found a fresh caribou track one morning where the animal had come out of the forest and walked along the railroad for about a mile. The red fox was abundant, and we found many signs. The section man on the railroad told me foxes were sometimes killed Mammals from Lake Edward, Quebec. 47 by the train, and that he had picked them up on several occasions when going over the road on his hand-car in the morning. The trappers get otter every winter, and the black bear is fairly com mon. The wolverine is still sometimes met with and occasion ally this expert trap robber proves a great nuisance to the trapper in the winter by rinding his line of deadfalls, following it up, de molishing every one, and eating the bait and any animal that may have been caught. Sciurus hudsonicus Erxl. Red Squirrel. 5 specimens. Red squirrels were extremely abundant and a great nuisance, as they persisted in getting into our mouse traps, and as the traps were usually not strong enough to kill them outright they carried away a great many. A few that were caught around the neck in the Schuyler mouse traps w r ere killed. We also caught a great number in steel traps baited with salt pork or meat. Castor canadensis Kuhl. American Beaver. 3 specimens. Beaver are still quite common in all this region, but are relentlessly pursued by the Indians and are decreasing very fast. The nearest beaver to Lake Edward were on the .Teannotte river. We were too busy to go after them ourselves and so hired two Indians and sent them down the Jeannotte. In five days they returned with a whole family of beaver an old male and female and three young. Unfortunately they had utterly ruined the old female and one of the young by shooting them in the heads with their rifles. The old male was a very fine, large beaver and according to the Indians was five years old. The specimen measured : total length, 1,130; tail, 410; hind foot, 176. The same two Indians, in the winter of 1894-1895, killed sixty beaver and told me they expected to get about forty this winter. In addition to the Indians, there are many other trappers working this country every season with great thoroughness, and the beaver stand but a poor chance. Synaptomys fatuus sp. nov. Northern Lemming Mouse. 9 specimens. Type No. 3857, coll. of E. A. and 0. Bangs ; female adult, from Lake Edward, Quebec, September 28, 1895. Total length, 125; tail, 16; hind foot, 19. E. A. and 0. Bangs, collectors. General characters. Slightly smaller and darker than S. cooperi, with smaller and lighter skull and much narrower and shorter incisors. Coat very long and full. Color. Upper parts sepia brown, thickly interspersed with black- tipped hairs ; under parts slate gray, with in places a slight brownish tinge ; feet drab ; tail nearly unicolor, slightly paler below, darker at the tip, and sparsely haired. Skull. The skull, as compared with that of S- cooperi, is rather smaller and narrower, with less spread to the zygomata and more slender rostrum. Teeth. The molar teeth are substantially the same as in S. cooperi, but the incisors are very much narrower and shorter. 48 Bangs Mammals from Lake Edward, Quebec. Measurements of nine Specimens of S. fatuus. No. Sex and age. Date. Total length. Tail. Hind foot. 3857 9 ad. Sept 25 1895 125 16 19 3855 cT yg. ad. . . Sept. 27, 1895 114 11 (bobtail) 18 3854 c? ad Sept. 24, 1895 123 20 18.5 3858 9 yg. ad. . . Sept. 27, 1895 114 15 17.5 3859 9 vg. ad. . . Sept. 24, 1895 113 19 19 3856 9 yg. ad. . . Sept. 27, 1895 114 15 17 3861 cfyg--'- Sept. 19, 1895 110 17.5 18 3860 cT ys Sept. 17, 1895 111 16 18 3862 cf yg Sept. 25, 1895 93 15 17 This strange little animal was common about Lake Edward and inhab ited every variety of country the sphagnum bogs, the deep spruce forest, and the banks of little streams. It lived everywhere in the deep moss. It was hard to trap and seemed not to care for any kind of bait, but blun dered into the traps that happened to be in its way. We caught thirteen examples of S. fatuus, four of which were so badly eaten by shrews or mice as to be worthless. Microtus fontigenus * sp. nov. Forest Meadow Mouse. 8 specimens. Type No. 3837, coll. of E. A. and O. Bangs, female adult from Lake Edward, Quebec, September 28, 1895. Total length, 151; tail. 41.25; hind foot, 21. E. A. and O. Bangs, collectors. General characters. Size small ; colors dark, with no rufous shades ; rostrum very slender; audital bullse very large and round. Color. Upper parts dark sepia brown, with a slight admixture of black- tipped hairs ; under parts olive gray to smoke gray ; tail sparsely haired and bicolor, black above, gray beneath. Skull. The skull is small, with very slender rostrum, and differs from that of any Microtus I am familiar with in having very large and round audital bullie, about as in the genus Evotomys. The basioccipital is nar row and does not have a distinct median keel. Teeth. The pattern of enamel folding of the molar teeth is substantially as in M. pennsylvanicus. Size. No. 3837, female adult (type) : total length, 151 ; tail, 41.5; hind foot, 21. No. 3840, male adult: total length, 150; tail vertebrae, 45; hind foot, 21. This Microtus was not common. We found it usually along the banks of the little spring brooks in the deep forest and in small sphagnum bogs, where it lived under old logs or in holes in the moss, after the manner of an Evotomys. Nowhere did it make runways like those of M. pennsylvanicus, * Fontigena = born beside springs or fountain heads ; a poetical term applied to the Muses, and therefore appearing in literature only in the feminine. Mammals from Lake Edward, Quebec. 49 and it appeared to be confined to the forest. I hunted in vain the marshy spots and alder swamps and the cleared fields, places M. pennsylvanicus would have delighted in, but found no trace of any Microtus there, and trapping in such localities yielded nothing but shrews. We caught only eight examples of M. fontigenus. Microtus chrotorrhinus (Miller). Rufous-nosed Meadow Mouse. 9 specimens. This beautiful little inhabitant of the deep spruce forest was not com mon. I consider it one of the rarest of our small mammals. It is easy to catch, and a day or two of trapping in any place is usually sufficient to capture all that are there. M. chrotorrhinus is apparently wholly diurnal. On account of the depredations of shrews I visited our traps regularly twice a day once at daylight in the morning and again just before dark. I never found a chrotorrhinus on any morning visit. Although these specimens were taken nearly three hundred miles north of the type locality (Mt Washington, New Hampshire), they are in every way typical and show no approach to M. xanthognathus. Fiber zibethicus (L.) Muskrat. 9 specimens. Exceedingly abundant on all the marshy shores of the lakes and rivers. We set a line of sixteen traps one afternoon and on visiting them next morning found fourteen muskrats. One trap I set on a floating log that lay across a little brook where it emptied into Lake Edward and caught a muskrat in it every night during our stay. Evotomys gapperi (Vig.). Red-backed Mouse. 36 specimens. The commonest small mammal at Lake Edward. The red-backed mouse of this region is the small, dark-colored form of the spruce belt, true gapperi. Evotomys fuscodorsalis Allen. Dusky-backed mouse. 4 specimens. Apparently this little known Evotomys was rare, four examples being all we caught. These were taken in two localities about three miles apart and two in each place. In both places they were caught among loose boulders on side hills covered by moss and overgrown by spruce, fir, and white birch. Peromyscus canadensis abietorum subsp. nov. Hudsonian White- footed Mouse. 4 specimens. Type No. 2205, coll. of E. A. and 0. Bangs, female adult, from James river, Nova Scotia. Coll. by C. H. Goldthwaite, August 8, 1894. Total length, 200; tail, 103; hind foot, 20 (measured in flesh by collector). General characters. Similar to Peromyscus canadensis (Miller), from which it differs in being a uniform dark gray above in both young and adult, never showing the russet and yellowish shades of old examples of P. canadensis. 50 Bangs Mammals from Lake Edivard, Quebec. Color. Old adult: upper parts dark smoke gray, slightly darker along the middle of the back, causing an indistinct median band ; under parts white, the hairs plumbeous at their base ; feet and hands white ; tail bi- colored, black above, white below, hairy, and longer than the head and body ; pencil long. The size, proportions, and skull are the same as in true canadensis. This white-footed mouse is the Northern representative of P. canadensis, which it resembles very closely in everything but color. When a large series of each is laid out side by side the difference in color is very strik ing, the uniform gray of the adults of abietorum being in marked contrast to the russet and yellow shades of the adults of canadensis. P. abietorum has a wide range in the spruce and fir forests of the north. It was not common at Lake Edward, and, as all we caught were immature, I have taken for the type a fine old adult from James river, Nova Scotia, from whence I have a good series, collected by Mr. C. H. Goldthwaite in the summer of 1894. Zapus insignis Miller. Woodland Jumping Mouse. 1 specimen. Either Zapus insignis was very rare at Lake Edward or they had already hibernated, the weather being quite cold, with a heavy frost nearly every night during our stay. This species is very easy to catch and we set many traps in its favorite haunts along the little brooks in the forest. The only one caught was exceedingly fat. Lepus americanus Erxl. American Hare. 4 specimens. Very abundant. We caught a number in steel traps baited with salt pork. These traps were set after the Indian fashion, a semicircle of slabs cut from the spruces being set up and the top covered over with spruce boughs. The bait was put inside and the trap in the opening. One morn ing I shot a hare asleep on top of a board fence three feet high, beside the railroad in the settlement. How he could have jumped onto this fence and balanced himself there is a mystery. Vespertilio subulatus Say. Bat. 1 specimen. Two bats of this species flew into the house on different evenings. Blarina brevicauda (Say). Short- tailed Shrew. 5 specimens. Common everywhere. Sorex (Microsorex) hoyi Baird. Hoy's Shrew. 1 specimen. Apparently rare. Sorex (Neosorex) albibarbis (Cope). Water Shrew. 1 specimen. Apparently rare. Sorex personatus Geoff. St. Hilaire. Common Shrew. 18 specimens. Extremely abundant and inhabiting every variety of country. Mammals from Lake Edward, Quebec. 51 Condylura cristata (L.) Star-nosed Mole. 1 specimen. No work of this mole was seen anywhere. The one taken was caught in a cyclone trap set under an old log. Probably the animal lives below the deep layer of moss with which everything is covered, and therefore gives no sign of its presence. Mephitis mephitica (Shaw). Hudsonian Skunk. 5 specimens. Skunks were common about the settlement. We trapped four and took another skull from an animal that had been killed some months pre viously. These skunks are highly interesting, being extreme examples of the Northern short-tailed form to which I have restricted Shaw's name mephitica.* They measure as follows : No. Sex and age. Total length. Tail. Hind foot. 3801 rT old ad. 585 193 75 3803 tf ad 617 202 79 3804 3" ad... 592 202 76 3802 9 old ad. 565 159 75 The skulls of all lack the median palatal spine usually seen in the skulls of Southern skunks. Putorius (Lutreola) vison (Schreber). Little Black Mink. mens. speci- Mink were abundant in spite of the fact that great numbers are trapped every winter. All we took are very small and dark-colored and are ex treme examples of the beautiful northern form, true vison. Putorius (Gale) richardsoni cicognani (Bp.) Small Brown Weasel. 3 specimens. We caught four of these little weasels, but one was partly eaten and ruined by some animal. All were caught in traps set for marten and baited with salt pork. Mustela americana Turton. Marten or Sable. 1 specimen. We set many traps for this elusive pirate of the forest, but succeeded in catching only one, a very dark-colored old female. It is of interest that the trappers here never get the fisher (M. pennanli) and say that it does not occur at all in this whole region. *Proc. Bost. Soc. Nat. Hist., vol. XXVI. Author's edition, July 31, 1895, p. 5. 52 Bangs Mammals from Lake Edward, Quebec. jo jo jo j[Bq ajSuis- jo uo sai.tas jB|ouuuaddn jo CC CO O --O ffl id o ; ic >^ tunaSrjai uaurB.ioj jo di[ q q oo e = 8Aoqu uinitmio jo aqSiaq ^sajsajg t- J 00 00 -* ;O 00 | CO 07 CO CO : t>J 00 CO -i !M CO X qjSaaj q ox/ 8 &S: = jo ^uojj oj uois O CM ' a 3 si a CS 'f - 6 W es >>, a ee &?^1 VOL. X, pp. 53-54 MARCH 14, 1896 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON DESCRIPTION OF A NEW SPECIES OF PLOVER FROM THE EAST COAST OF MADAGASCAR. BY CHARLES W. RICHMOND. The apparently new species of plover here described is repre sented in the United States National Museum series by five specimens. Three of these were in a collection of birds lately received from Dr. W. L. Abbott ; the other two were obtained by exchange some years ago from the Paris Museum. ^Igialitis thoracica sp. uov. Type No. 151,174, U. S. National Museum, 9 adult, Loholoka, east coast of Madagascar, June 3, 1895. Dr. W. L. Abbott, collector. Crown, back, scapulars, tertials, and wing-coverts hair brown, the feathers edged with pale or deep buff, those of the greater wing-coverts edged and tipped with white ; primaries, secondaries, rump, median upper tail-coverts, and middle rectrices dark clove brown ; shafts of primaries (including the third) with white on terminal half; primary coverts brownish black, tipped with white; lateral upper tail-coverts white; inner primaries narrowly bordered on inner web and tipped with white ; base of outer webs white ; secondaries tipped with white, which become broader toward the innermost. Forehead, lores, cheeks, throat, axillars, under wing-coverts, sides of body, and flanks white ; a line from upper mandible to lower anterior border of eye, continued posteriorly through and including ear-coverts black, connecting with a narrower black band extending across lower border of nape, and with a broad black pectoral band, the latter more extensive on sides of chest; an interocular crescent-shaped black band borders the white forehead and separates it from a white line over eyes, ear-coverts, and passing across nape as a con spicuous ruchal band (leaving the black crown patch entirely surrounded by a white band and the latter isolated from other white markings) ; a white band below the black pectoral band passes abruptly into cinnamon buff on the abdomen and under tail-coverts, that of the abdomen extends up on sides of body to the black band across breast, intercepting the white. Three outer tail feathers white, with more or less dusky markings, 8 BIOL. Soc. WASH., VOL. X, 1896 (53) 54 Richmond A New Species of Plover from Madagascar. especially on the two inner ones ; next inner pair (4th) dusky, with white tips; 5th pair hair brown, becoming black snbterminally, with a deep buff tip. Bill, legs, and feet black in dried skin. Wing, 4.00; tail, 1.72; tarsus, 1.20; culmen (exposed), .69 inches. In another female (No. 151,169) the wing measures 4.20 inches; the other measurements of the five specimens are very much the same. This species seems to be most nearly related to ^Egialitis varia (Vieillot) of Africa, and also found in Madagascar, but differs from it mainly in the presence of the black pectoral band and the absence of a wholly black shaft in the third primary ; the white line posterior to the black crescent between eyes is more pronounced and the lesser wing-coverts and primar}' coverts are not decidedly blackish. There is also a slight difference in size, particularly noticeable in the bills. The two specimens received from the Paris Museum are sexed as males, and are precisely similar to those collected by Dr. Abbott. They were collected by M. Lantz, in 1882, on the southeast coast of Madagascar. In addition to this information the labels bear the names 'Charatlrius tenett-us,' and, in a later handwriting, ' pecuarius ' [~=varia]. From an examination of the specimens in the National Museum and a careful comparison of descriptions, it appears that no described plumages of either JEgialitis tenetta or J varia possess black pectoral bands. I was rather loth to consider the species unnamed after examining the two specimens from the Paris Museum, as the bird must be well known to the French authors, particularly Milne-Edwards and Grandidier, whose great work on Madagascar birds I have had no opportunity to consult. Thinking there might be some reference to the black pectoral band in the account of jE. varia in this work, I wrote to Mr. Witiner Stone, of the Philadelphia Academy, who has access to it, and he has very kindly fur nished me with the folio wing extract* under Charadrius pecuarius Temm. (as they prefer to write it) : ' ' Ce Pluvier africain se trouve aussi a Madagascar, sur ies cotes de Test comme sur celles de 1'ouest. II est en dessus d'un brun roussatre clair avec une couronne blanche autour de latete qu'un diademe noiratre separe du front, qui est egalement blanc ainsi que Ies joues ; la gorge, la poitrine, que traverse, chez Ies adultes, une large bande noire, et Ies sous-caudales, sont blanches ; le ventre est roussatre. Cette bande noire qui traverse la poitrine chez Ies adultes n'a pas encore ete signalee chez les individus Africains." It is very remarkable that the black pectoral band should be present in adults from Madagascar and absent in those from Africa, where the species is said to be common in many places and breeds and from whence it was originally described. The two species, varia and thoracica, are apparently found together on the east coast of Madagascar, where Dr. Abbott collected a specimen of each at Loholoka on June 3. It was probably this association of the species that led the authors of the above-mentioned work to consider them adult and young of one species. * Hist. Phys. Nat. et Polit. de Madagascar, XII, Ois. tome I, pp. 511-512. j"i L;<->-TVU -^x W 18/896 VOL. *, PP. 55-64 ^<*^RCH 19, 1896 PROCE: OF THE BIOLOGICAL SOCIETY OF WASHINGTON REVISION OF THE LEMMINGS OF THE GENUS SYNAP- TOMYS, WITH DESCRIPTIONS OF NEW SPECIES. BY DR. C. HART MERRIAM. The genus Synaptomya has an interesting history. It was de scribed by Professor Baird about forty years ago from a speci men received from William Cooper, of Hoboken, New Jersey, for whom the species was named Synaptomys cooperi. 1 The locality at which it was collected is unknown. For many years the species continued to elude the notice of naturalists, and it was not until 1874 that additional information was published con cerning it. In this year Coues recorded specimens from Indiana, Illinois, Minnesota, and Kansas. He also mentioned specimens from Oregon [= Washington] and Alaska; but these, as will be shown later, do not belong to the present species.' 2 In 1881 Dr. F. W. Langdon recorded its occurrence u in num bers " at Brookville, Indiana, and described the locality at which it had been found by E. R. Quick.* In 1885 Edgar R. Quick and Amos W. Butler described its habits as observed at Brookville, Indiana. 5 In December, 1892, I published a notice of the occurrence of the species on Roan Mountain, North Carolina, and of the dis covery of its remains in ' pellets 1 of the long-eared owl found in Virginia, near Washington, D. C., by Dr. A. K. Fisher, and of others taken from the stomachs of hawks and owls killed at Sandy Spring, Maryland, and Alfred Center, New York. 6 At the close of this paper I suggested that mammal collectors would " do well to keep a sharp lookout for this species in the cooler parts of Pennsylvania and New Jersey." In January, 1893, S. N. Rhoads recorded the species from May 1 The numeral references in the present paper refer to titles in the bib liography at the end of the article. 9-Bioi,. Soc. WASH., Von. X, 1896 (55) 56 Merriam The Lemmings of the Genus Synaptomys. Landing, New Jersey, but unfortunately gave it a new name. Synaptomys stonei.*" 1 In the same year (1893) J. B. Steere re corded it from Ann Arbor, Michigan. 8 In April, 1894, Outram Bangs recorded specimens from Ware- ham and Plymouth, Massachusetts, and showed that S. stonei is the same as S. cooperi Baird. 9 In December, 1894, J. A. Allen recorded the northward exten sion of Synaptomys to Andover and Gulquac Lake, New Bruns wick. 12 Early in January, 1895, S. N. Rhoads published a record of the capture of a specimen of S. cooperi on Big Bushkill creek, Monroe county, Pennsylvania. 13 This completes, so far as I am aware, the published records of the type species. Although remains of the species had been found both in ' pel lets ' and stomachs of hawks and owls from the vicinity of Wash ington, D. C., and although the species had been persistently trapped for by a number of experienced mammal collectors, still no specimen ' in the flesh ' was actually obtained until February of the present year (1896), when Vernon Bailey captured several in a sphagnum bog at Hyattsville, Maryland, only seven miles from Washington. Mr. Bailey has also secured a number at Elk River, Minnesota, and I have a specimen from Knoxville, Iowa. During recent explorations in the great Dismal Swamp in southern Virginia. Dr. A. K. Fisher secured specimens of a new Synaptomys, which is here described under the name S. helaletes. Specimens collected at Neosho, Kansas, many years ago by the late Captain B. F. Goss, and labeled S. gossii by Baird, are here described as a subspecies -under that name. A few months ago Napoleon A. Comeau, of Godbout, on the north shore of the St. Lawrence, near the Gulf, sent me a speci men of Synaptomys which differs materially from S. cooperi. This animal has just been described by Outram Bangs under the name S. fatuus, from specimens collected by him at Lake Edward, Quebec. 14 Dr. Allen's New Brunswick specimens, which he has kindly loaned me for examination, also belong to this northern form. It is not improbable that all of the four forms here recog nized will be found to intergrade. In 1894 F. W. True described a new lemming mouse collected by Lucien M. Turner at Fort Chimo, Ungava, and named it Mic- * In the same paper Mr. Rhoads stated that the species " had previously been detected by the U. S. Department of Agriculture in the rejects of a barn owl living in the tower of the Smithsonian Institution" (Am. Nat., Jan., 1893, 53). This statement was unauthorized and incorrect. The Lemmings of the Genus Synaptomys. 57 tomys innuitus The characters that separate it from Synaptomys proper seem of subgeneric rather than generic weight, and in the present paper Mictomys is treated as a subgenus of Synaptomys. In 1874,' 2 and again in 1877, 3 Coues referred to Synaptomys cooperi, a specimen from Skagit valley, Washington, collected in 1859 by C. B. Kennerly, and one from Nulato, Alaska, collected in 1867 by William H. Dall. These specimens are still in the U. S. National Museum, and through the courtesy of Mr. True I have been enabled to compare them with his type of Mictomys innuitus, which they closely resemble. Both belong to the sub- genus Mictomys, but differ sufficiently from innuitus and from each other to warrant separation. They are here described under the names truei and datli. In September, 1895, Clark P. Streator collected, at Wrangel, Alaska, still another member of the same group, which is here named wrangeli. Summary. The material now available shows that the genus Synaptomys, instead of being monotypic, as until recently sup posed, comprises 2 well marked subgeneric groups Synaptomys proper and Mictomys ; that Synaptomys proper inhabits eastern Canada and the northeastern United States from Minnesota to New Brunswick and New England, and contains 4 fairly well denned forms ; that Mictomys has a transcontinental distribution from Labrador to Alaska, and contains at least 4 species. Synaptomys, like the other genera of lemmings, is a distinctly boreal group. Of the two subgenera, Mictomys is decidedly the more boreal, being strictly confined, in the east at least, to the Hudsonian zone. The subgenus Synaptomys pushes southward to the northern edge of the Austroriparian zone, but after it leaves the Boreal zone it occurs only, so far as known, in cool swamps. Genus SYNAPTOMYS Baird. Subgenus St/naptomys Baird, 1857. Inferior molars with well denned closed enamel loops on outer side ; upper incisors very broad and heavy, with enamel face deep orange throughout ; posterior end of palate without median azygos ridge or projection. Subgenus Mictomys True, 1894. Inferior molars with 110 closed enamel loops on outer side ; upper incisors relatively narrow and weak, with en amel face pale yellow and part on outer side of sulcus nearly white ; posterior end of palate with a strongly marked median azygOS FIG. i. Enamel pattern of lower molars, ridge and projection. T_. Synaptomys 2. Mictomys, 58 Merriam The Lemmings of the Genus Synaptomys. Subgenus SYNAPTOMYS Baird. Synaptomys cooperi Baird. Synaptomys cooperi Baird, Mammals N. Am., pp. 556-558, 1857. Coues, Proc. Acad. Nat. Sci., Phila., p. 194, 1874; Monog. N. Am. Rodentia, pp. 235-236, 1877. Quick and Butler, Am. Naturalist, XIX, 113-115, Feb., 1885. Merriam, Proc. Biol. Soc. Wash., VII, 175-177, Dec., 1892. Bangs, Proc. Biol. Soc. Wash., IX, 99-104, April, 1894. Synaptomys stonei Ehoads, Am. Naturalist, XXVII, pp. 53-54, Jan. 11, 1893. Type locality unknown ; probably northern New Jersey or southern New York. Geographic distribution. Boreal and parts of Transition zones from Min nesota eastward to eastern Massachusetts and south to Iowa, Indiana, and Maryland, and in the mountains to North Carolina and Tennessee. South of the Boreal zone it appears to be confined to cold sphagnum swamps, which give it a boreal atmosphere. General characters. Similar in size and general appearance to Microtus pennsylranicus, but tail very much shorter. Contrasted with Synaptomys helaletes the feet are smaller and the rostrum, mandible, and upper incisors are much narrower and less massive. Color. Upper parts grizzled gray and yellowish brown abundantly mixed with black-tipped hairs ; under parts soiled whitish, the plumbeous under fur showing through ; tail bicolor ; brownish above, whitish below. In the adult the color of the back varies from pale yellowish brown to almost rusty, always ' grizzled ' by a bountiful admixture of black-tipped hairs. In the young the color is at first very dark, almost blackish slate ; it then becomes grayish brown and approaches sepia before taking on the yellowish brown of the adult. Cranial and dental characters. Contrasted with ^' ne ^ etes ^ rom Dismal Swamp, the skull and teeth of S. cooperi are smaller and weaker, the zygomata more bowed outward, the rostrum and mandible very much narrower, the nasals nar rower posteriorly, and the brain case shorter. FIG. 2. Enamel pattern of Measurements. Average of 4 specimens from upper and lower molars ^nn Arbor, Michigan: total length, 118; tail in Synaptomys cooperi. vertebrffi| 17>5 . hind footj 18> Average of 2 from Roan Mountain, North Carolina: Total length, 121 ; tail vertebrae, 20 ; hind foot, 19.5. Synaptomys fatuus Bangs. Synaptomys fatuus Bangs, Proc. Biol. Soc. Wash., X, 47-48, March 7, 1896. Type locality. Lake Edward, Quebec. Geographic distribution. Hudsonian zone from Lake Edward, Quebec (and probably much farther west), to Victoria county, New Brunswick, and Godbout, Quebec. Limits of range unknown. General characters. Similar to S. cooperi, but slightly smaller; skull de. cidedly smaller, with much narrower upper incisors. The Lemmings of the Genus Synaptomys. 59 Color. Upper parts grizzled yellowish brown, abundantly mixed with black-tipped hairs ; under parts varying from slate gray to whitish, washed with buff on the belly ; tail nearly concolor, only slightly paler below than above. Cranial and dental characters. Skull similar to that of S. cooperi, but smaller and weaker ; rostrum narrower ; basisphenoid broader posteriorly. Upper incisors very much narrower than in cooperi. Measurements. Average of 2 adults from type locality (measured in flesh by O. Bangs): total length, 124; tail vertebrae, 18; hind foot, 18.7. Measurements of an alcoholic specimen ( 9 ) from Godbout, Quebec : total length, 106; tail vertebrae, 19; hind foot, 18. Synaptomys helaletes sp. nov. Type from Dismal Swamp, Virginia, No. 75172, 9 adult, U. S. National Museum, Department of Agriculture collection. Collected October 14, 1895, by Dr. A. K. Fisher. Original number 1818. General characters. Similar to S. cooperi, but with larger head and feet, longer tail, much broader rostrum and mandible, and larger and more massive skull and teeth. Color. Upper parts grizzled gray and yellowish brown, abundantly mixed with black-tipped hairs; under parts plumbeous, washed with white; tail bicolor, brown ish above, whitish below; toes usually partly white. Cranial and dental char acters. Contrasted with S. cooperi, the skull and teeth are larger, heavier, and more massive ; the zygomata less strongly bowed outward ; the na sals broader posteriorly, and the brain case longer. The rostrum, upper in cisors, and under jaw are remarkable for breadth and massiveness. Measurements. Type specimen: total length, 125; tail vertebra, 22; hind foot, 20. Average of four adults from type locality : total length, 118.5; tail vertebrae, 21 ; hind foot, 20.2. General remarks. Synaptomys helaletes, while of essentially the same size as S. cooperi, has very much larger fore and hind feet and a longer tail. The difference in the breadth and massiveness of the rostrum, mandible, and upper incisors is so great that skulls of the two require no compari son. Still, specimens recently collected by Vernon Bailey in a sphagnum swamp near Washington, D. C., are somewhat intermediate and indicate that intergradation may exist. FIG. 3. Skull of Synaptomys helaletes ?, WTO?, an ear. The leaves of Cerastium alpinum very closely resemble those of a Myosotis, so that it can * Johannes Loeselius : Flora Prussica, Regensburg, 1703. 106 Holm The Earliest Record of Arctic Plants. easily be seen how the mistake occurred. "Auriculae muris qffinis herba Spitzbergensis " is the name given by Ray to this plant, but his diagnosis, "Supremo cauliculo Flos innascitur albus," is the only feature which is characteristic of this Cerastium. Martens has, indeed, pointed out the characteristics in a much clearer way. " Kraut als Singrun '' (Table G, Fig. 6) represents Salixpolaris Wahlbg. If it were not that the illustration is so good, it would hardly have been possible to identify this plant. " Singriin " is now the name for Vinca. The stem is described as knotted and woody and the leaves as occurring in pairs. The flowers were not seen, and Martens is therefore not certain that the plant belongs to Pyrola minima. It is called " Vinca permncse, similix herba Spitzbergensis " by Ray. The leaves of this willow are very small and coriaceous, brilliant green. They occur in about two alternately on each branch, and to a certain extent resemble those of some species of Pyrola. " Erdbeer-Kraut " (Table H, Fig. 6) is Potentilla fragiformis Willd. The description is very good, and the statement that the leaves only had three leaflets shows that we have this species before us and not P. maculata Pourr., the leaves of which are quinate. The same statement is also given by Ray, u foliis tri- partitis divisis ... ," who has called, it "Fragarise affinis Spitzbergensis.'''' Two Algae are enumerated under the name " Klippen-Kraii- tern," of which the figure /; in Plate F represents Fucus vesicu- losus. The vesicles are described very accurately, and Martens states that he did not observe whether these contained any seeds. His sailors informed him, however, that the small sea snails (Pteropoda), upon which the whales feed, originate from the seeds of this Alga. Martens does not seem to have shared this opinion, however, and says that he is inclined to believe that these snails have, like others, originated from eggs! The large Alga (Fig. c in Plate I) is undoubtedly a species of Laminaria. Several other plants were observed, but were not collected. Only two of these have been described, but these have not been figured. One of these, u der weisse Ma/m" is evidently Dry us octopetala L. " Mahn " is undoubtedly a misprint for " Mohn," the common poppy (Papaver dubium or Rhceas). Since the only poppy that grows on Spitzbergen, P. nudicaule L., has yellow The Earliest Record of Arctic Plants. 107 flowers, it is not likely that Martens meant this plant, but rather the common white Dry as, which is not so very unlike a poppy. The other plant is u der rothe Sauerampffer," which probably was Oxyria dlgyna Campd., now called " Sauerling " by the Germans. If the list of plants collected by Martens be compared with the most recent publication on the flora of Spitsbergen,* it will be seen that all the species named in the list have actually been rediscovered by later expeditions. As to the locality where they were collected, it appears that they were found in the neighbor hood of Smeerenberg, on the northwestern shore of Spitzbergen, designated by Martens as a Harlinger Kocherey." * Nathorst, A. G. , Nya Bidrag till Kannedomen om Spetsbergens Karl- viixter, Stockholm, 1883, Kgl. Sv. Vet. Akad. Hdlgr., vol. 20, No. 6, 88 pp. PROC. BIOL. SOC. WASH., X, 1896 VOL. X, PP. 109-112, PL. VII JULY 22, 1896 PROCEEDINGS OF TIIK BIOLOGICAL SOCIETY OF WASHINGTON THE CENTRAL AMERICAN THYROPTERA. BY GERRIT S. MILLER, JR. Three specimens of Thyroptera, collected by G. E. Mitchell on the Escondido River at a point about fifty miles from Bluefields, Nicaragua, and now in the collection of the United States De partment of Agriculture, are clearly referable to the species de scribed by Lichtenstein and Peters in 1855 as Hyonycteris dis cifera* This bat was recognized as a distinct species by Tomes in a paper published in the Proceedings of the Zoological Society of London for 1856 (p. 179), but Dobson, in 1878,f placed the name Hyonycteris disci/era, together with Hyonycteris albiventer Tomes J and Thyroptera bicolor Cantraine among the synonyms of the Brazilian Thyroptera tricolor Spix. While no specimens of the three nominal and probably valid South American species || are available for comparison with the Nicaraguan bat, there can be no doubt that the latter differs widely from any of these. It may be redescribed as follows : Thyroptera discifera (Lichtenstein and Peters). Hyonycteris discifera Lichtenstein and Peters, Monatsber. K. Preuss. Akad. Wiss., Berlin (1854), p. 335, 1855. Tomes, Proc. Zool. Soc. London, 1850, p. 179. * Monatsber. K. Preuss. Akademie Wiss., Berlin (1854), p. 335, 1855. t Catalogue of the Chiroptera in the British Museum., p. 245, 1878. JProc. Zool. Soc. London, 1856, p. 179. i Bull. Acad. Roy. Sci. Bruxelles, VII, p. 489, 1845. || The type localities of these are : - Thyroptera tricolor, Brazil ; T. bicolor, Surinam ; T. albiventer, Napo River, near Quito, Ecuador. 18 RIOL. Soc. WASH., Vor,. X, 1896 (109) 110 Miller The Central American Thyroptera. Thyroptera tricolor Dobson, Catalogue of the Chiroptera in the British Museum, p. 345, 1878 (in synonymy only ; the description refers strictly to South American specimens). Type locality. Puerto Caballos, Honduras. Geographic distribution. Central America from Puerto Caballos, Hon duras, south to Bluefields, Nicaragua. General characters. Size small ; length about 45 mm.; tail, 26 ; forearm, 31. Calcar slender, distinct, slightly longer than free border of interfemoral membrane, terminating in an ill-defined lobllle ; the posterior edge with a well- formed keel supported by one strong cartilaginous process. Terminal 2 mm. of tail free. Free border of uropatagium with a few scattered hairs. Ears short, funnel-shaped, acutely pointed, when laid forward reaching barely to tip of nose. Wings from middle of claws. Third and FIG >.i.-Teeth of Thyroptera dis- fourth toeg d ]y approximated and a/era ; a, upper ; 6, lower (X 5)- firmly bound together. 99 11 ^ *? Teeth. Dental formula as usual in the genus: i- t c- -, pm 5,, b 11 o 3 q_q m ' = 38. The teeth (Fig. 1) are small and weak for the size of the oo skull. Upper incisors bifid,* in pairs, the outer tooth half as large as the inner and separated from the canine by a space about as wide as the crown of the larger incisor. Premolars all in the tooth row, not separated by spaces from each other or from the adjoining molar and canine, first slightly smaller than second, third slightly more than half as large as first molar. Crown of first molar broadest, crown of second longest. Lower incisors trifid, the crown of the outer as broad as that of the first and half of the second. First lower pre- inolar smaller than second, but larger than canine. Middle lower molar largest. Ears. The ears are short, acutely pointed, funnel-shaped, and directed forward. The tips , , , . ,. , f , FIG. 2. Head of Thyroptera do not reach tip of nose when the ears are laid disdfera (x 3) forward. The anterior border is strongly con vex from base to small concavity just below very narrowly rounded off tip. Posterior border concave immediately below tip, then convex to basal notch. The basal notch is strongly developed and isolates a very large lobe which joins side of head below line of lips (Fig. 2). Tragus short and broad, the tip thickened and bent abruptly forward ; a large thickened basal lobe directed forward and outward, and a minute process directed backward just above posterior base. * Dobson states that in Thyroptera tricolor the outer incisor is unicuspi- date. The Central American Thyroptera. Ill FIG. 3. Foot and uropatagium of Thyroptera disci/era (X 2). Membranes. The membranes are thin and semitransparent, broad and ample. Wings attached at middle of claws, sparsely hairy from sides of body to line connecting elbow and knee. The free edge has also a narrow hairy border. Antebrachial membrane hairy near base and along humerus and fleshy part of forearm, which in turn are covered with hair. Uropatagium sparsely haired throughout on dorsal surface, otherwise naked, except at extreme base and along veins on ventral surface. Feet. The feet are small, weak, and so turned outward as to be nearly in line with calcar (Fig. 3). Toes with two phalanges, of which the second is very small and serves merely to support the long claw. All the fingers are bound together by membrane to about the middle of the claws, while the third and fourth digits are firmly united, so that the two claws, although really separate, form what is apparently one strong nail, shorter and more abruptly curved than the others (Fig. 4). Calcar strong, distinct, longer than free border of uropatagium, termi nating in a small lobule and bear ing a well-formed keel, supported by one strong cartilaginous pro cess. Sucking disk circular, the margin next the phalanges distinct, that toward the keel not sharply marked off from sole. Fur and color. In distribution, the fur is peculiar in its extension on the wings and dorsal surface of the interfemoral membrane. Color dull yellowish brown throughout, scarcely paler ventrally, the hairs without darker bases. Ears and membranes dusky brownish. Measurements. No. 51538, 9 a< 3-> Escondido River, Nicaragua; total length, 66 mm. ; head and body, 37.6 ; tail, 26 ; free tip of tail, 2 ; femur, 13 ; tibia, 13.4 ; foot, 4 ; forearm, 31 ; third finger metacarp., 29.8 ; first ph., 14; second ph., 7.8; fourth finger metacarp., 28.6; first ph. ,10; second ph., 4.6; fifth finger metacarp., 26; first ph., 8; second ph., 5.6; ear, 11.6; width of ear, 12; tragus, 4 ; diameter of disk on thumb, 3 ; diam eter of disk on foot, 2. No. 51539, $ ad., same locality and date; total length, 65; head and body, 38 ; tail, 26; free tip of tail, 1.8; femur, 14 ; tibia, 14.8 ; foot, 4.8 ; forearm, 31.6; third finger metacarp., 29; first ph., 13.4; second ph., 8.4; fourth finger metacarp., 29; first ph., 9; second ph., 5.4; fifth finger metacarp., 25.6; first ph. ,7.4; second ph., 6; ear, 12; width of ear, 12; tragus, 3.6; diameter of disk on thumb, 3.4; diameter of disk on foot, 2. FIG. 4. Right foot of Thyroptera disci/era greatly enlarged to show syndactylism of third and fourth digits. 112 Miller The Central American Thyroptera. General remarks. Of the three South American species of Thy- roptera, two (T. bicolor and T. albiventer} are described as sharply bicolor. brownish above and white beneath, while the third (71 tricolor) is said by Dobson to be reddish brown on the back and pale yellowish white on the abdomen, and also to have dental characters not found in the Nicaraguan animal. In Thyroptcra tricolor and T. bicolor the free part of the tail equals one- fourth or one-third of its whole length. In T. disci/era, on the other hand, only the terminal joint and part of the penultimate joint project beyond the edge of the interfemoral membrane. T. albi- venter is said to have the terminal joint only of the tail free, but the type specimen of this species was so mutilated that no de pendence can be placed on this character. In size the four forms apparently agree very closely at least it is impossible to find any important differences in the measurements given in the original descriptions. The characters of Thyroptera disci/era and of the South Ameri can species as described may be thus contrasted : Both upper incisors bifid disci/era Only the inner upper incisor bifid tricolor Sharply bicolor, or color of back distinctly dif ferent from that of belly albiventer, bicolor, tricolor Essentially unicolor disci/era One-fourth to one- third of tail free from inter- femoral femoral membrane tricolor, bicolor Only tip of tail free albiventer (?) disci/era The syndactylism of the third and fourth digits of the foot may prove to be peculiar to Thyroptera disci/era. It is mentioned by Lichtenstein and Peters in the original description of the species, but none of the authors who have described South American specimens make any allusion to such a condition, al though in most cases they have mentioned the form of the feet and claws with considerable detail. Another character of Thyroptera disci/era not mentioned in descriptions of the South American species, but probably com mon to all, is the large and conspicuous clitoris (see pi. VII). In the adult female this measures 1.6 mm. in length and is about half as long as the penis of a nearly full grown male. The vulva opens longitudinally with the anterior commissure encroaching on the basal third of the clitoris. VOL. X, PP. 113-114 JULY 22, 1896 PROCEEDINGS OF THK BIOLOGICAL SOCIETY OF WASHINGTON NOTE ON THE MILK DENTITION OP DESMODUS. BY GERRIT S. MILLER, JR. Some immature specimens of Desmodus rufus, taken by Mr. E. W. Nelson, at Etzatlan, Ja lisco, Mexico, in June, 1892, and now in the collection of the United States Department of Agriculture, retain the greater part of the milk dentition, though it is probable that none are young enough to present a complete set of deciduous mo lars. The extraordinary special ization of the teeth of this bat correlated with the animal's strictly sanguivorous habits make any facts relating to the *- early development of the teeth of special interest. In the adult (Fig. 1, e, and 2, c) the dental formula is i 7>i9> 1 1 22 c ^i pm gqj> = 20. The milk i- dentition, so far as it can be de- 2-2 termined, is as follows : di -^^> dc j3^' dm 2^9 = 18. The largest of the deciduous ^c* teeth are the Upper incisors Fro. i.-Maxillary teeth of Desmodus (Fig. 1 , di 1 and di 2). These CUt rufu ^ showi milk dentition and gradual change in form of permanent teeth from the gums some time before the ve r y young (a) to adult (e) (x 5). 19 BIOT,. Soc. WASH., Vor,. X, 1896 (113) 114 Miller Note on the Milk Dentition of Desnwdus. permanent incisors (Fig. 1, i), and even after the appearance of the tips of the latter remain for a considerable period tiie most conspicuous teeth in either jaw. Their strongly recurved tips are probably of great service to the young when clinging to the nipple of the female during flight. At first the anterior decidu ous incisor lies on the outer side of the permanent incisor, while the posterior deciduous incisor occupies the space between the permanent incisor and canine (Fig. 1, a). As the permanent incisor increases in size, it gradually extends backward until both milk incisors appear closely appressed to its outer face. In this condition (Fig. 1, c) the teeth re main until the animal acquires a large size, while the second milk incisor often persists in nearly full- grown individuals (Fig. 1, d). The deciduous canine (Fig. 1, a. dc) lies imbedded in the gum on the outer side of the prominence caused by the growing permanent canine. It is shed at an early age, and never becomes in any way functional. A deciduous molar (Fig. 1, b, dm} was found occupying a position near FIG 2.-Mandibuiar teeth of Desmo- ^ posterior outer border of the aus riifus, showing part or milk den tition (a) and gradual chang* in form first permanent upper prcmolar in to adult (ocx 5). two specimens. This tooth lacks the recurved tip and must be wholly functionless. In the lower jaw (Fig. 2) the deciduous teeth are smaller and less conspicuous than those in the upper jaw. They are also shed at a much earlier period. On opening the mouth of one specimen (No. 52130) I found two deciduous molars lying loose on the gum over the permanent premolars. The exact position of these milk teeth could not be determined, and I failed to detect any trace of them in other individuals. The deciduous mandibular canine closely resembles its counterpart in the upper jaw, both in size, position, and history. Its tip is, however, less strongly hooked (Fig. 2, dc). The deciduous lower incisors (Fig. 2, di) are very small and loosely attached to the gum through which they scarcely pierce before they are shed. In form they are totally different from the corresponding upper teeth. Their tips are somewhat widened and faintly notched, thus suggesting the form presented by the permanent lower incisors of many bats. VOL. X, pp. 115-118 NOVEMBER 3, 1896 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A NEW FIR FROM ARIZONA, ABIES ARIZONICA. BY C. HART MERRIAM. In the summer of 1889, when making a biological survey of the San Francisco Mountain region in Arizona, I discovered a new fir with whitish corky bark. The scales of the cones were shed before the end of September, and no perfect cones were brought back. In my report on the ' Forest Trees of the Sail Francisco Mountain Region ' I described this ' white cork-bark fir ' under the head of Abies subalpina Engelm. [= Abies lasiocarpa Hooker], with the following explanation : " I believe this tree to be distinct from true A. subalpina, but in the absence of material for direct comparison I am unwilling to separate it." My original account of the species is as follows : " This beautiful fir, unique in the color and character of its bark, is one of the most con spicuous trees on San Francisco Mountain between the altitudes of 2,725 and 2,900 meters (8,950 to 9,500 feet). On the north side of a large butte, just south of Walker Lake crater, it de scends to 2,600 meters (8,500 feet). The bark is a fine elastic cork of uniform texture, and free from hard particles. It aver ages about 6 millimeters in thickness and is very durable, fre quently remaining intact while the wood rots away. Large pieces of it, still retaining their elasticity, may be stripped from dead trees and from logs upon the ground. It is sculptured by irregu- , larly interrupted longitudinal depressions or grooves, and is i ornamented by fine, parallel, wavy lines. Its color varies from creamy white to gray, and the surface has a velvety texture. The leaves are short, and the scales of the large cones are deciduous while still on the tree. In fact, it was almost impossible to secure a perfect cone as early as the latter part of September. * * * 20-BiOL. Soc. WASH., Voi,. X, 1896 (115) 116 Merriam A New Fir from Arizona. " On Kendrick Peak it grows from the south rim of the crater (altitude about 2,800 meters, or 9,200 feet) to the summit (a little above 3,050 meters, or 10,000 feet).' 1 * Early in July of the present year (1896) I again visited San Francisco Mountain and, in company with Dr. B. E. Fernow, had the satisfaction of ob taining upper and lower branches, fresh cones, and bark of the new tree, which may be defined as follows : Abies arizonica sp. nov. Type from west slope of San Francisco Mountain, Arizona. Altitude, about 3,000 meters (approxi mately 10,000 feet). Col lected July 2, 1896. No. 270,604. U. S. National Herbarium. Range. Hudsonian Zone of San Francisco and Kendrick Mountains, Arizona; not reaching timber line. Characters. Size of tree, medium or rather small, averaging about 15 meters in height and rarely 300 millimeters in diameter at base; bark a highly elastic fine-grained cork, whitish or grayish in color, usually creamy white, with irregularly sinuous grayish ridges (Fig. 24) ; leaves of cone- bearing branches thick, subtriangular in transverse section, and sharp-pointed at apex (about 20 millimeters in length) ; leaves of lower branches much longer, natter, FIG. 24. Bark of Abies arizonica (natural size). * North American Fauna, No. 3, pp. 120-121, September, 1890. A New Fir from Arizona. 117 blunt, and notched at apex (about 25-30 millimeters in length) ; cones dark purple, slender, medium, or rather small, those of type specimen (not full grown) measuring about 50 x 20 millimeters ; scales much broader than long, strongly convex laterally (Fig. 25, c), purple on both sides ; bract (without awn) reaching to or past middle of scale ; body of bract much broader than long. Remarks. The only tree with which the white cork-bark fir needs comparison is the subalpine fir (Abies lasiocarpa Hooker = A. subalpina Engelm.), from which it differs in leaves, bark, and cones. In Abies lasiocarpa the leaves of the lower branches aver age much shorter than in A. arizonica ; the bark is hard instead of elastic-corky, and is variable in color, usually dark grayish brown blotched with whitish ; the cones are larger, and the scales and bracts differ widely in shape and proportions. In A. lasiocarpa (Fig. 25, a and 6) the scales are longer than broad, the body of the bract is less than one-third the length of the body of the scale, and the seed wings are about twice as long as broad ; in A. arizonica (Fig. 25, c) the scales are much broader than long, the body of the bract is more than half the length of the scale, and the seed wings are about as broad as long. The form of the scale and relative size of the bract probably change somewhat with age, but in the ac companying figures the immature scale of Abies arizonica (Fig. 25, c) is contrasted with a still younger scale of A. lasiocarpa (6), as well as with the mature scale of the latter (a). The young cone of A. lasiocarpa, from which the scale figured (6) is taken, is decidedly smaller than the cone of A. arizonica, from which figure c is taken, while the adult cone of A. lasiocarpa is more than twice as large. Both of the specimens figured of A. lasiocarpa came from Mount Hood, at the north end of the Cascade Range in Oregon, which is prob ably near the type locality of the species. I am indebted to Mr. F. V. Coville for the opportunity of figuring the young cone FIG. 25. Scales of cones (natural size). a. Abies lasiocarpa, mature. b. Abies lasiocarpa^ young. c. Abies arizonica, young. a, b, c. Upper side, showing seed wings, a', b', c'. Under side, showing bracts. 118 Merriam A New Fir from Arizona. of lasiocarpa. Abies arizonica is a much smaller tree than A. lasiocarpa. Both are highly boreal species, belonging to the Hudsonian Zone, though A. arizonica fails to reach the upper or timber-line belt of this zone. Abies lasiocarpa ranges from southern Alaska and British Columbia southward, over the Rocky Mountains into Utah and Colorado, and over the Cascade Range to southern Oregon. Abies arizonica, on the other hand, is restricted, so far as known, to San Francisco Mountain and neighboring peaks on the summit of the plateau in northern Arizona. VOL. X, PP. 119-125 NOVEMBER 5, 1896 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON THE COTTON MOUSE, PEROMYSCUS GOSSYPINUS. BY OUTRAM BANGS. The present revision of the subspecies of Peromyscus gossypinus is based on the study of several hundred specimens in the col lection of E. A. and O. Bangs and the type and five topotypes of Peromyscus gossypinus mississippiensis kindly lent me by Mr. Samuel N. Rhoads. LeConte, in 1853, bestowed the name Hesperomys gossypinus upon the large dark-colored, white-footed mouse of Georgia. Two years later the same author named what he supposed to be another species from the same general region, calling it Hesper omys cognatus. This last name has troubled subsequent mam- malogists not a little, until Mr. Rhoads, in his ' Mammals of Tennessee,'* in 1896, relegated it to its proper place, and it became a synonym of P. gossypinus, based on the young in the pelage assumed after the plumbeous first coat has disappeared. There is, however, a name earlier than LeConte's H. gossypinus that must be considered. It is Hesperomys polionotus of Wag ner, described in 1843. f This animal is said to have come *Proc. Acad. Nat. Sci., Phila., 1896, p. 189. t Archiv fur Naturgeschichte von Wiegmann, 1843, 2ter. Bd., pp. 51-52. "Aus eben diesem Staate [Georgia] herriihrend liess mir Prof. Schinz zwei Miiuse zur Ansicht zukommen, unter denen die eine mit M. Lecontii ubereinstimmt. Der andern, die mir unbeschrieben scheint, habe ich den Namen Mus polionotus beigelegt : M. supra flavido-plumbeus subtua pedibusque albidus ; auriculis mediocribus, dent. prim, integris, cauda pilosa abbreviata. Korper 2" 4 /x/ , Schwanz 1" 2 //x , Ohren 4 X// , Hinter- fuss 7 /7/ . Wie schon erwiihnt, werden beide nicht zu Mus gehoren, doch ist mir ihr Gebiss unbekannt." 21-Bioi,. Soc. WASH., Vol.. X, 1890 (119) 120 Bangs The Cotton Mouse, Peromyscus Gossijpinus. from Georgia. The measurements and description of the colors of Wagner's specimen show it to have been a very young indi vidual, and now impossible to identify. Wagner gives no defi nite locality in the State of Georgia, and as P. aurcolus is found generally distributed throughout that State and as P. leucopus undoubtedly occurs in the mountains, it would be unwise to assume that the specimen in question was certainly the young of P. gossypinus, and thus allow Wagner's name to stand for that species. Two names have been given lately to subspecies of gossypiims by Mr. Samuel N. Rhoads. One of these, the so-called Sitomys megacephalus, from Woodville, Alabama,* becomes a synonym of P. gossypinus. I have not seen the type, which is in alcohol, but there are no characters attributed to it that can in any way separate it from true gossypinus of Georgia, an animal Mr. Rhoads was wholly unfamiliar with, he making his comparisons with the Florida form, which is subspecifically distinct. The cranial characters claimed for megacephalus are individual and in nowise diagnostic. The other is the Peromyscus gossypinus mississlppien- sis of the bottom lands of the Mississippi in Tennessee, and is a well-marked race. I now describe two more races, one from the peninsula of Florida, the other from the bayou region of Louis iana, thus dividing P. gossypinus into four subspecies. Peromyscus gossypinus has been given by authors in recent years as a subspecies of P. leucopus, not because any intermedi ates were forthcoming, but on general principles, until Rhoads, in his ' Mammals of Tennessee,' in 1896, gave it full specific rank. Mr. Rhoads, in the summer of 1895, found gossypinus and leucopus in the Mississippi bottoms in Tennessee, where, he says, it was possible to catch both species in the same trap, and yet the two kept perfectly distinct. This undoubtedly will prove to be the case wherever the ranges of P. leucopus and P. gossypinus overlap. Most of the closely related forms of white-footed mice look very different from each other when one is trapping and hand ling them in the flesh. This ' aspect difference,' as Professor Shaler aptly calls it, is subtle and hard to define, and may dis appear almost entirely when the animals are made into the con ventional museum skins or preserved in spirits, thus leaving the characters on which species and subspecies are based very *Proc. Aead. Nat. Sci., Phila., 1894, p. 254. The Cotton Mouse, Peromyscus Gossypinus. 121 slight in comparison with what they were in life. This is strik ingly true of P. gossypinus, and I well remember, when I first trapped this beautiful mouse, being astonished to see a creature so wholly different from P. leucopus, of which I had previously supposed it merely a subspecies. Since the cranial characters presented by the members of the genus Peromyscus are so slight that it is often difficult to tell apart the skulls of very different species, they are naturally of little help in distinguishing closely related forms. Peromyscus gossypinus has a wide range in the lower Austral Zone, extending north along the Atlantic coast to North Carolina, up the Mississippi Valley to Tennessee, and west along the Gulf coast to Louisiana ; but it is not found on the higher land between the most northern, eastern, and western points of its range. * Peromyscus gossypinus inhabits a variety of situations, but my experience with the typical form in Georgia has been that it is rare. About St. Marys, Georgia, they lived in the hammocks and margins and around the edges of some of the cleared fields, but were not numerous anywhere. I could not find them in the pine woods at all. but their absence there may be due to the annual firing of these woods to make pasture. The Florida form is very abundant in many parts of peninsular Florida. At Oak Lodge, on the east peninsula opposite Micco, I trapped them by the hundred. Their favorite abodes there were the edges of the salt savannah, the piles of brush and rubbish around the cleared fields, and along the edge of the beach in the saw palmetto thickets. In these dense thickets and among the plants and grasses of the upper beach Peromyscus gossypinus palmarius and the exquisite little Peromyscus niveiventris occurred together in great numbers, feeding largly on the seeds of the sea oats, Uniola paniculata. Peromyscus gossypinus meets or overlaps the ranges of at least four and probably five other white-footed mice. All along its northern limits it must come in contact with Peromyscus leucopus, and judging from Mr. Rhoads' experience in Tennessee the two species overlap, but keep distinct. P. gossypinus can always be told from P. leucopus by its much larger size, stouter build, bigger hind foot, shorter tail, browner and less fulvous coloration of the upper parts, and the gray (not white) under parts. Major LeConte states in his description of P. gossypinus that it has *Reelfoot Lake, Tennessee, and Bertie County, North Carolina. 122 Bangs The Cotton Mouse, Peromyscus Gossypinus. longer front legs than leucopus, and consequently a different gait, progressing in an even run, while leucopus goes by little leaps. I regret to say that while I had the opportunity I never studied the movements of P. gossypinus in life. Major LeConte undoubt edly did, and I see no reason to doubt his statement. The skull of P. gossypinus averages larger than that of P. leucopus when in dividuals of the same age are compared, but apart from this difference in size the two are indistinguishable. Peromyscus aureolus overlaps the greater part of the range of P. gossypinus, but reaches farther north and probably not so far south, the southernmost examples, so far as I know, coming from Enterprise, Florida. It can always be told from gossypinus by the bright ochraceous of the upper parts, the under parts being also extensively washed with this color, and its smaller size about that of P. leucopus. In Florida two white-footed mice, very different from each other and equally different from gossypinus, occur in many places associated with gossypinus. The commoner of these is the most beautiful of all white-footed mice, the little, ghost-like Peromys cus niveiventris. This species is about half the size of gossypinus, with pale gray and fawn-colored upper parts and snowy white under parts. The other is Peromyscus floridanus, an animal very unlike P. gossypinus and belonging to a different group of the genus. It is a large mouse, with big, nearly naked ears, short tail, and very large hind foot. The sides are a bright ochraceous buff and the under parts white. The fur is very soft and silk} 7 ". In the west Peromyscus gossypinus may meet the range of P. mearngi of the lower Rio Grande and coast of Texas. P. mearnsi is about the size and proportions of P. leucopus, and is dark gray (purplish gray in fresh pelage) above, without a marked darker dorsal band, and white below. Peromyscus gossypinus (LeConte). 1831. Hyp\udxus] gossipinus LeConte, M'Mnrtrie's Cuvier's Animal Kingdom, I, 1831, app., p. 434 (nomeii nudum). 1853. Hesperomys gossypinus LeConte, Proc. A cad. Nat. Sci. Phila., 1853, p. 411. 1855. Hesperomys cognatus LeConte, Proc. Acad. Nat. Sci. Phila., 1855, p. 442. 1874. Hesperomys ( Ve^xt-itnas) leucopus gossypimis Coues, Proc. Acad. Nat. Sci. Phila., 1874, p. 179. 1877. Hesperomys leucopus gossypinus Coues, Monog. N. American Mu- ridte, p. 76. The Cotton Mouse, Peromyscus Gossypinus. 123 1894. Sitomys megacephalus Rhoads, Proc. Acad. Nat. Sci. Phila., 1894, p. 254. 1896. Peromyscus gossypinus Rhoads, Proc. Acad. Nat. Sci. Phila., 1896, p. 189. Type locality. The LeConte plantation a few miles above Riceboro, Liberty County, Georgia. Geographic distribution. From northern Florida north along the coast at least to Bertie County, N. C. ; west through the non-mountainous parts of Georgia to Alabama and perhaps Mississippi. Subspecific characters. A large heavily built mouse ; hind foot large ; tail shorter than head and body, bicolored ; ears dusky, nearly naked, of moderate size ; general color of upper parts dark brown, with broad darker dorsal band ; under parts gray ; feet and hands grayish white. Color. Adult: Upper parts dark brown, varying from Prouts brown to sepia, darkening along middle of back into a broad dorsal band, which ranges from clove brown to black ; a black orbital ring. Under parts smoke gray, the hairs plumbeous at base ; feet grayish white ; ears dusky ; tail bicolored, dusky above, grayish white below. Nursing young : Black ish slate above, slate gray below ; tail and feet as in adult. Young in second pelage: General color of upper parts duller, more hair brown, often with a sooty cast ; otherwise like adult, dorsal stripe well marked.* Size. Average measurements of twelve adult specimens from St. Marys, Ga. : total length, 177.66; tail vertebra, 70.25; hind foot, 22.35. Max imum size (of largest old adult in above average) : total length, 197 ; tail vertebra, 82.5 ; hind foot, 22. Specimens examined, 37, from the following localities : Georgia: St. Marys, 35. North Carolina : Bertie County, 2. Peromyscus gossypinus mississippiensis Rhoads. 1896. Peromyscus gossypinus mississippiensis Rhoads, Proc. Acad. Nat. Sci. Phila., 1896, p. 189. Type locality. Samburg, Reelfoot Lake, Tennessee. Geographic distribution. The Mississippi bottoms in Tennessee ; limits of range unknown. Subspecific characters. Size about that of typical gossypinus; tail a little longer ; hind foot larger ; colors paler and more yellowish ; dorsal band less well defined, without black orbital ring. Color. Adult: Upper parts varying from cinnamon brown to russet, darkening on middle of back into an ill-defined dorsal band about mummy brown ; no dark orbital ring ; under parts grayish white, the hairs plumbeous at base ; ears dusky ; tail bicolored, dusky above, white below ; feet grayish white. * The young in this pelage are much smaller than the adults, but as they frequently breed they have the appearance of full-grown animals, and gave rise to LeConte' s species Hesperomys cognalus. 124 Bangs The Cotton Mouse, Peromyscus Gossypinus. Size. Average measurements of six adult specimens from type locality : total length, 183; tail vertebrae, 79.5; hind foot, 24.45. Maximum size (of largest old adult in above average) : total length, 196 ; tail vertebrae, 84 ; hind foot, 25. Specimens examined, 6, all from the type locality. Peromyscus gossypinus palmarius subsp. nov. Type from Oak Lodge, on east peninsula opposite Micco, Brevtird County, Florida. No. 3224, 9 old adult, collection of E. A. and O. Bangs. Collected by 0. Bangs February 23, 1895. Total length, 183 ; tail verte brae, 74; hind foot, 21. Geographic distribution. Peninsular Florida, north at least to Brevard County on the east and Citrus County on the west. Subspec'ific characters. About the size of typical P. gossypinus ; hind foot shorter; colors much paler and more yellowish ; no decided darker dorsal band ; a black orbital ring. Color. Adult: Upper parts varying, according to freshness of pelage, from bright russet to wood brown, usually a few darker hairs scattered along middle of back, but not enough to form a dorsal band ; a black orbital ring ; under parts grayish white, the hairs plumbeous at base ; ears dusky ; tail bicolored, dusky above, white below ; feet grayish white. Size. Average measurements of twenty adult specimens from type locality: total length, 181; tail vertebrae, 71.88; hind foot, 21.55. Maxi mum size (of largest old adult in above average) : total length, 206 ; tail vertebrae, 83 ; hind foot, 22. Remarks. Peromyscus gossypinus palmarius often shows a pectoral spot of yellowish brown, sometimes of large size. It is often difficult to tell the young in the second pelage of palmarius from the young of typical gossypinus, but as a rule they are lighter in color, more grayish, less sooty, and have the dorsal stripe much less well defined. Specimens examined, 166, from the following localities in Florida: Oak Lodge, east peninsula opposite Micco, Brevard County, 111 ; Micco, 3 ; Flamingo, 19 ; Miami, 2; Jupiter Inlet, 3 ; Crystal River, 4 ; Citronelle, 3 ; Blitches Ferry, Citrus County, 21. Peromyscus gossypinus nigriculus subsp. nov. Type from Burbridge, Plaquemines Parish, Louisiana. No. 2731, 9 adult, collection of E. A. and 0. Bangs. Collected by F. L. Small January 30, 1895. Total length, 174; tail vertebras, 79; hind foot, 24. Geographic distribution. Bayou region of the coast of Louisiana. Subspeclfic characters. Size smallest of the gossypinus series ; hind foot about as in typical gossypinus ; tail proportionally longer ; colors verv dark ; a broad dorsal band nearly black ; ears and upper surface of tail black ; a black orbital ring. Color. Adult: upper parts varying from vandyke brown to sepia, often with a sooty cast ; darkening along middle of back into a broad dorsal The Cotton Mouse, Peromyscus Gossypinus. 125 band of nearly black ; a black orbital ring ; under parts grayish white ; the hairs plumbeous at base; ears black; tail bicolored, black above, grayish white below ; feet and hands grayish white. Size. Average measurements of three adult specimens from the type locality: total length, 168.33; tail vertebra?, 76.66; hind foot, 23.66. Average measurements of twenty adult specimens from Gibson, Terre Bonne Parish, Louisiana: total length, 169.85; tail vertebrae, 77.85 ; hind foot, 22. Maximum size (of largest old adult in above average) : total length, 184 ; tail vertebra, 86 ; hind foot, 22. Remarks. The young of Peromyscus gossypinus nigriculus are very dark colored, both in the nursing and the second pelage, and can usually be separated, both by their dark color and their smaller size, from the young of corresponding age of gossypinus or of palmarius. This form appears to be confined to the heavy swamps of the bayou region, and probably does not occur farther from the coast than the limits of these swamps. Although Mr. Small trapped persistently in several localities in the prairie and pine regions of central Louisiana, he failed to get a single specimen of any Peromyscus in such places and concluded that none occur north of the bayous. Specimens examined, 89, from the following localities in Louisiana : Bur- bridge, Plaquemines Parish, 5 ; Gibson, Terre Bonne Parish, 56; Pow- hatan Plantation (near Gibson), 28. VOL. X, PP. 127-130 NOVEMBER 14, 1896 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON JUNCUS CONFUSUS, A NEW RUSH FROM THE ROCKY MOUNTAIN REGION. BY FREDERICK V. COVILLE. In a collection of Juncaceae from Idaho, recently received for identification from Mr. A. A. Heller, were two specimens of an undescribed Juncus, which had long been confounded, even by Engelmann himself, with Juncus tenuis congestus Engelm. A de scription of the species, which was already well represented in the National Herbarium by specimens from other collectors, is given herewith. Juncus confusus Coville, sp. nov. Plant perennial, densely tufted, 15 to 60 cm. high, erect; stem 0.5 to 1.5 mm. thick at base, narrower above, striate, nearly terete; leaves all basal, the sheaths with well developed auricles, the blades erect, one- third to one-half or more the height of the stem, flat, usually involute in drying, narrow, 0.5 to 1 mm. in breadth ; inflorescence congested into a turbinate cluster 2 cm. or less in height, much exceeded by its lowest bract ; perianth 3 to 4 mm. long, its parts equal, ovate-lanceolate, acute, with green or at maturity stramineous midrib and a brown stripe on either side; stamens 6, about one-half as long as the perianth, the anthers shorter than their filaments; capsule oblong, equaling the perianth, re- tuse, completely 3-celled ; seed light brown, obovoid or oblong, .45 to .6 mm. in length, with oblique white apiculations connected by a usually evident white raphe, finely reticulated in about 16 longitudinal rows, the areolse smooth and 2 to 4 times broader than long. Type specimen in the U. S. National Herbarium, collected September 6, 1890, in an irrigated meadow, North Park, Colorado, by C. S. Crandall. 22 Bioi. Soc. WASH., VOL. X, 1896 (127) 128 Coville Junctus Confusus, a New Rush. Other specimens beside the type have been examined as fol lows: Colorado : Grand Lake, George Vasey, 1868, No. 576. Wyoming: Sherman, altitude 8,000 feet, G. W. Letterman, July 28, 1884. Big Horn Mountains, B. C. Buffum, August 6, 1892. Clarks Fork Valley, J. N. Hose, September 3, 1893, No. 530. Steamboat Point, Yellowstone Lake, Robert Adams, August 19, 1871. Montana: Spanish Creek, P. A. Rydberg, July 11, 1896, No. 3058. In a meadow, Spanish Basin, altitude 1,800 meters, P. A. Ryd berg, July 17, 1896, No. 3116. In a wet meadow, Blackhawk, P. A. Rydberg, August 5, 1896, No. 3282. Idaho : In the vicinity of Forest and about Lake Waha, Nez Perces County, Mr. and Mrs. A. A. Heller, June 25, 1896, No. 3319, and July 16, 1896, No. 3446. Washington : Near Spangle, Spokane County, W. N. Suksdorf, June 30, 1884, No. 1042. JUUCMS coiifasus is one of seven closely related species, all of which with the exception of/, tennis occur only in America and with the additional exception of J. dichotomus only in North America. Juncus tennis was formerly a very rare plant in Eu rope, but is now becoming widely disseminated there and in nearly all parts of the world, apparently by introduction from America. The following synopsis will be useful in distinguish ing the species of the group : SYNOPSIS OF JUNCUS TEN u is AND ITS ALLIES. Leaf blade flat, but sometimes involute in drying. Anthers much longer than their filaments 7. georgianus Coville. A densely tufted plant, with long leaves, reaching the unusually large inflorescence ; brown-striped perianth 4 to 6 mm. long ; and narrowly oblong-lanceolate completely 3-celled capsule. This species is known only from Georgia, where it occurs on Stone Mountain and adjacent knobs of similar geological structure. For full description see Bull. Torr. Club, 22 : 44. 1895. Anthers not exceeding their filaments. Perianth 2.5 to 4 mm. long, usually with some reddish or brownish coloration, equaling the completely 3-celled capsule ; apex of the capsule distinctly triquetrous, truncate or retuse. Junctus Confusus, a New Rush. 129 Inflorescence somewhat congested, much exceeded by its lowest involucral bract J. confusus Coville. A plant of the Rocky Mountain region from Colorado northward to Montana, Idaho, and Washington. Inflorescence not congested, the flowers Fecund on the some what incurved branches, seldom exceeded by the lowest involucral bract /. secundus Beauv. A species of common occurrence in the coastal plain from New Jersey to North Carolina and occasional in Illinois and Missouri. Perianth 3.5 to 5.5 mm. long, green or stramineous, without brown stripes along either side of the midrib (except in the variety) ; capsule obovate, broadly rounded, though sometimes retuse, incompletely 3-celled J. tennis Willd. Occurring almost throughout North America, especially as a weed along roadsides and paths, and now migrating to all parts of the world. Along the Pacific coast from middle California to Vancouver Island occurs a robust variety with congested inflorescence much exceeded, as is usually the case also in the type form of the species, by the lowest in volucral bract; the perianth 4 to 5.5 mm. in length, about one-half longer than the capsule ; its parts with a reddish brown stripe along either side of the midrib. This plant is here named Junctus tenuis occidentalis (J. tennis con- gestus Engelm. Trans. St. Louis Acad. 2 : 450. 1866. Not /. congestus Thuill. 1799). Leaf blade terete, channeled along the upper side. Seed not caudate. Perianth 3.5 to 5 mm. long, not exceeded by the capsule. /. dichotomus Ell. A species common to North and South America, occurring abundantly in the United States along the coast from Texas to New Jersey, and more rarely as far northward as Maine. The plant is often confounded with /. tenuis when not crit ically examined, but in addition to its leaf character it may be distinguished also by its darker green color and its fewer-ribbed (about 14 instead of 20 to 24) seeds. Perianth 2.5 to 3 mm. long, conspicuously exceeded by the cap sule /. greenei Oakes & Tuckerm. Occurring near the coast from New Jersey northward to New Brunswick ; in Michigan, Wisconsin, and Minnesota; and in the Canadian province of Ontario. The inflorescence is usually short, much exceeded by the lowest involucral bract, and the exposed portions of the completely 3-celled ovoid-lanceolate capsule are commonly brownish. The seeds are commonly but erroneously described as caudate. 130 Coville Junctus Confusus, a New Rush. Seed distinctly caudate J. vaseyi Engelm. Occurring from Michigan, Illinois, and Iowa northward to the plains of middle Canada; at Orono, Maine; in the Black Hills of South Dakota ; and, on the authority of a label in the Canby Herbarium, in Middle Park, Colorado. /. vaseyi differs from the last, in addition to its important seed and perianth characters, in its inflorescence usually exceeding the lowest involucral bract, and the green, or, at maturity, stramineous capsule being little or not at all contracted toward the apex. VOL. X, PP. 131-132 NOVEMBER 14, 1896 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON RISES ERYTHROCARPUM, A NEW CURRANT FROM THE VICINITY OF CRATER LAKE, OREGON. BY FREDERICK V. COVILLE. Crater Lake is a remarkable body of the purest water, nearly circular in form, about ten kilometers (6 miles) in diameter and 600 meters (2,000 feet) in depth, without a visible outlet, occupy ing the bowl of an extinct volcano in the southern part of the Cascade Mountains of Oregon, situated about latitude 43 and longitude 122. The surface of the water has an altitude of 1,902 meters (6,239 feet) and the surrounding cliffs rise 300 to 450 meters (1,000 to 1,500 feet) higher, some of the neighboring peaks reaching 2,400 and 2,700 meters (8,000 and 9,000 feet). The mountain slopes are densely forested, except where the trees have been burned off by sheep herders, and no settlements occur nearer than the plains below. It was the writer's good fortune to visit the place in August of the present year, at the time of the excursion of the Mazamas to that point. The Mazamas are an organization of mountain climbers, which originated in Port land, Oregon, and are doing a great deal to popularize the natural sciences, to make known the wonderful scenery of the Northwest coast, and especially to create and maintain a public sentiment toward the preservation of the magnificent forests of that region. Nothing seems to have been published on the botany of this part of the Cascades, and indeed no botanist appears heretofore to have made a collection of the plants of the Crater Lake region. The collection made by the writer and Mr. John B. Leiberg from August 13th to 20th of the present year is therefore of unusual 23-Bior,. Soc. WASH., Vor,. X, 1896 (131) 132 Coville Ribes Erythrocarpum, a New Currant. interest. Only a partial examination of the specimens has been made thus far, and a full report must be deferred, but an inter esting species, apparently undescribed, is here presented to the public. Ribes erythrocarpum Coville & Leiberg, sp. nov. Shrub trailing upon the ground, devoid of prickles, the stems rooting and giving rise to ascending branches commonly 10 to 20 cm. in height, the herbage and inflorescence clothed with short glandular hairs ; leaves angulate-orbicular in outline, rugose, commonly 2 to 3.5 cm. in diameter, on petioles nearly as long, 3 to 5-lobed, the sinuses extending one-half or two-thirds the way to the base, the lobes coarsely crenate and the crena- tures unevenly but finely dentate-serrate ; racemes erect, commonly 10 to 20-flowered, the bracts herbaceous, lanceolate to obovate, commonly 2 to 4 mm. long, persistent ; flowers erect, contiguous, when expanded 6 to 8 mm. in diameter, on pedicels equaling the bracts ; ovary beset with short glandular hairs; calyx not produced into a tube, the spreading lobes oblong, obtuse or broadly acute, yellow, minutely dotted with red, there fore appearing salmon-colored, sparingly and minutely pubescent without, glabrous within; petals broadly spatulate, glabrous, one-third to one-half the length of the calyx lobes and similar in color; filaments glabrous: style glabrous, 2-parted; fruiting racemes erect or sometimes declined by the weight of the berries ; fruit on erect pedicels, scarlet, subpyriform to spherical, commonly 8 to 10 mm. in length, provided with short glandular hairs, the flesh white or translucent, insipid. Type specimen in the United States National Herbarium, collected August 12, 1896, at an altitude of about 1,675 meters, in the canyon of Pole Bridge Creek, about 10 kilometers south of Crater Lake, Cascade Moun tains, Oregon, by Frederick V. Coville and John B. Leiberg. Th6 plant appears from the structure of its flowers to be most nearly related to the Ribes laxiflorum of Pursh and the Ribes hoivellii of Greene (R. acerifolium Howell), from both of which it is at once distinguishable by its creeping habit and its glandular pubescence, in the latter of these characters and in its general appearance closely resembling Pursh 's Ribes viscosissimum. Its herbage, however, possesses the rank odor of Ribes prostratum and R. hudsonianum, quite distinct from the citronella-like smell of viscosissimum. That species, too, has blue fruit and an elon gated calyx tube. Ribes erythrocarpum grows in abundance about Crater Lake, in the forests of Tsuga pattoniana, to an alti tude of at least 2,400 meters. VOL. X, PP. 133-134 DECEMBER 8, 1896 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON AN UNDESCRIBED SHREW OF THE GENUS SOREX. BY CHARLES F. BATCHELDER. On September 9, 1895, at Beetle's, Essex county, New York, I obtained a Shrew unlike any species known to me. It was caught in one of several ' cyclone ' traps, baited with rolled oat meal, that were set among some large, angular rocks at the head of a wooded talus of loose rock. Just above, shading the spot and keeping it moist and cool, rise the low cliffs from whose frag ments the talus has been formed. Nearly a year later, on August 1, 1896, 1 caught a second speci men of this Shrew on Mount Marcy, the highest of the Adiron dack Mountains. It, too, was caught with oatmeal in a ' cyclone ' trap. It was taken in a crevice between some rocks, on the bare, open summit of the mountain, about 5300 feet above sea-level. The locality where the first one was captured is about eight miles distant, in an air line, and lies at an elevation of only 1300 feet above the sea. I have compared this Shrew with other species of the genus Sorex (the material for comparison I owe in some cases to the unfailing kindness of Dr. C. Hart Merriam), and find it so dif ferent from them all that I am led to describe it as follows : Sorex macrurus sp. nov. Type from Beede's [sometimes called Keene Heights], in the township of Keene, Essex county, New York ; taken September 9, 1895. The type is a young adult male, No. 1384, collection of C. F. Batchelder. General characters. Size large ; tail long ; body stout. 24-BiOL. Soc. WASH., VOT,. X, 1896 (133) 134 Batchelder An Undescribed Shrew. FIG. 26. Skull of Sorex macrurus$ Colors (of type, noted in the flesh). Upper parts between ' slate-color ' * and ' blackish slate' ; 1 under parts dark ' smoke gray' 2 or brownish 'mouse-gray'; 3 tail, above, browner than back ; edge of lips and under side of tail, brownish flesh color; upper side of both hind and fore feet between ' fawn-color ' * and ' ecru drab.' 4 The specimen from Mt. Marcy ($, ad. Aug. 1, 1896, No. 1386, coll. C. F. B.) differs in color from the type only in having a slightly more plumbeous tint, a difference due, apparently, merely to its pelage having been exposed to several Type (x 2). " weeks less wear. Cranial and dental characters. Skull long and slender; brain-case low, narrow', and little inflated ; rostrum long, narrow, and low ; palate rather narrow. Posterior border of infraorbital foramen ly ing over a point consider ably behind the interspace between the first and sec ond molars. Unicuspidate teeth slender; the first and second about equal in size; the third and fourth small er, and subequal if any thing, the third slightly shorter than the fourth. Molariform teeth deeply excavated posteriorly. Measurements (of type, taken in the flesh). Total FIG. 28. Same tooth row, seen from below. length, 130 mm. ; tail vertebrae, 60 mm. ; hind foot, 15 mm. ; fore foot, 8 mm. ; height of ear, 10 mm. The Mount Marcy specimen measured : total length, 139 mm. ; tail vertebrae, 61 mm. ; hind foot, 15 mm. ; ear, 10 mm. The extreme tip of its tail appears to have been lost by some accident. This Shrew differs so widely from all others with which I am ac quainted that comparisons with any other species are quite unnecessary. In color and size it bears a slight superficial resemblance to Sorex fumeus and to S. trowbridgii, but it is at once distinguishable from them by its long tail, even without reference to its cranial and dental characters, in which it is totally unlike these species. In the general shape of the skull there is a suggestion of Sorex personality, but in this respect macrurus is even more remote from such species as trowbridgii or fumeus than is personatus itself. 1 Ridgway: A nomenclature of colors for naturalists, etc., 1886, plate II, Figs. 4-3. 2 Ibid., Fig. 12. 3 Ibid., Fig. 11. "Ibid., pi. Ill, Figs. 22-21. FIG. 27. I^eft side of upper jaw showing teeth. Type (X 6). VOL. X, pp. 135-138 DECEMBER 28, 1896 PROCEEDINGS OF THK BIOLOGICAL SOCIETY OF WASHINGTON SOME NEW MAMMALS FROM INDIAN TERRITORY AND MISSOURI. BY OUTRAM BANGS. In the summer of 1896 Mr. Thaddeus Surber undertook a col lecting trip to Indian Territory in the interests of the Bangs collection. After spending a short time in Missouri he went to Stilwell, in the Cherokee Nation, at the northwest part of the Boston Mountains. The country was suffering from an un precedented drought and all mammals were extremely hard to find. Mr. Surber was also handicapped by the unfriendliness of the Indians, who absolutely refused to help him in any way. He had collected but a few days when he was taken ill with an extremely malignant form of malaria, which compelled him to abandon the work. The Boston Mountains about Stilwell rise to a height of 2,500 feet (estimated), and are closed in by ranges of low lying hills, some 250 or 300 feet higher than the intervening narrow valleys of rich land. Beyond the hills west of Stilwell stretches a barren prairie that is said to have been formerly forest-covered. On the sides of the mountains are found black walnut, white oak, red oak, black jack, etc., but no pines. The mountains all top off in cliffs from five to fifty feet high, composed of sandstone or bastard limestone, in which there are many caves. The material collected at Stilwell, while small in number of specimens, is of great interest. Besides the new forms here de scribed, Mr. Surber got only three species of mammals the raccoon, Procyon lotor ; the southern gray squirrel, Sciurus caro- linensis, and the plains wood rat, Neotoma baileyi. 25 BIOL. Soc. WASH., Voi,. X, 1896 (135) 136 Bangs New Mammals from Indian Territory. My thanks are due to General Nelson A. Miles, who with great kindness secured for me the necessary permit allowing Mr. Surber to collect in Indian Territory. I am also indebted to Dr. J. A. Allen for presenting me with specimens of Lepus sylvaticus bach- mani, Peromyscus attwateri, and Scalops texanus for comparison with the Indian Territory forms. Lepus sylvaticus alacer subsp. nov. Type from Stilwell, Indian Ter., No. 5480, 9 young adult, collection of E. A. and O. Bangs. Collected by Thaddeus Surber August 14, 1896. Original No. 65. Two specimens from Stilwell, Indian Ter. ; 2 from Stotesbury, Vernon Co., Mo. General characters. About the size of Lepus sylvaticus bachmani, but differ ing from that form in being much darker and richer in color and in hav ing much smaller audital bullse. Color. Type in summer pelage: upper parts rich reddish brown (about hazel), many of the hairs with black tips ; nuchal patch and upper surface of legs and arms cinnamon rufous; sides and rump paler, shading towards wood brown ; band on under side of neck wood brown ; rest of under parts, including chin and throat, white. A specimen from Stotes bury, Mo., in winter pelage (No. 1677, February 27, 1894) : upper parts cinnamon rufous on back, wood brown on sides, very thickly mixed with black-tipped hairs, giving a dark and rich effect ; ears wood brown broadly edged with black ; no black mark between ears. Cranial characters. Skull small, about the size of that of L. sylvaticus bachmani, differing from other members of the sylvaticus series in having extremely small audital bullse. Size of type skull : basilar length, 536 ; occipitonasal length, 67.2; zygomatic breadth, 34; greatest length of single half of mandible, 51.6. Size. Type : total length, 370 ; tail vertebrae, 50 ; hind foot, 95 ; ear, 73. Average measurements of two adult specimens from Stotesbury, Vernon Co., Mo.: total length, 398; tail vertebrae, 30.5; hind foot, 79.35; ear, 82.6. General remarks. When I was at work on the cotton-tails of eastern North America in 1894 I had the two specimens referred to above, col lected at Stotesbury, Mo., in the winter of 1894, by Mr. Surber, and on account of their small size, peculiar coloration, and small audital bullse was unable to refer them to any known subspecies. They clearly belong to the same form as the Indian Territory specimens, which appears to be unnamed. The two examples taken at Stilwell were both shot in the low. rich valleys, and Mr. Surber did not find the animal on the moun tains. This form probably has an extensive range throughout the region where the wooded eastern country meets the great plains. Lepus syJvuticus alacer prohably merges into L. sylvaticus bachmani of Texas, but its smaller audital bullte and dark color at once distinguish it from the gray ba New Mammals from Indian Territory. 137 Feromyscus bellus sp. nov. Type from Stilwell, I. T. No. 5483, 9 adult, collection of E. A. and O. Bangs. Collected by Thaddeus Surber August 15, 1896. Original No. 67. Two specimens from Stilwell, I. T. General characters. Compared with P. attwateri (apparently its nearest geographical ally) P. bellus differs in being much darker and browner; in having a larger hind foot, a pectoral band of fawn color, and a fawn- colored nose patch (white in attwateri). It belongs to the group of so- called brush mice. Color. Upper parts broccoli-brown much mixed with black along back, becoming fawn color on lower sides ; patch at root of whiskers fawn color ; ears large, nearly naked, dusky; tail large and long, bicolored, black above, white below, well haired and with a decided pencil ; feet and hands white ; under parts white, the hairs plumbeous at base ; a band of fawn color extending across under side of neck in front of arms. Cranial characters. Skull of the same general appearance as that of P. attwateri, but larger and with deeper, broader brain case. Size of the type skull: basilar length, 24.2; occipitonasal length, 28; zygomatic breadth, 14; greatest length of single half of mandible, 15. Size. Type : total length, 190; tail vertebrae, 90; hind foot, 24; ear, 17. Average measurements of two adult specimens from Stilwell, I. T.: total length, 192.5; tail vertebrae, 93.5; hind foot, 24; ear, 16. General remarks. The two examples of this brush mouse were taken on one of the rocky hillsides at Stilwell. P. bellus differs from P. attwateri very materially, but P. attwateri seems very close, perhaps too close, to P. rowleyi, as I must confess I can hardly distinguish skins of the two species. Tamias striatus venustus subsp. nov. Type from Stilwell, I. T. No. 5478, J old adult, collection of E. A. and O. Bangs. Collected by Thaddeus Surber August 13, 1896. Original No. 63. Two specimens from Stilwell, I. T.; 1 from Noel, Mo. General ctiaracters. Size and proportions about as in T. striatus griseus ; colors very bright, especially on rump ; all the black dorsal and lateral bands much shortened; hair, especially on rump, hispid, but this char acter may be seasonal. Color. Rump and upper surface of legs deep, rich, lustrous chestnut rufous, this color extending up back and sides, narrowly bordering the black bands ; sides yellowish gray ; back (between the black bands) and upper neck and shoulders dark gray ; ears and face much suffused with chestnut rufous ; facial markings not conspicuous ; hairs of upper surface of tail yellowish at base, then black and slightly tipped with white; under parts yellowish white, somewhat washed on belly and under side of legs with cinnamon rufous ; under side of tail cinnamon rufous. Cranial characters. The skull is large, about as in T. striatus griseus. Size of type skull: basilar length, 38.6; occipitonasal length, 43. 6; zygo matic breadth, 24.4 ; greatest length of single half of mandible, 26.2. 138 Bangs New Mammals from Indian Terr it on/. Size. The type: total length, 260; tail vertebra, 100; hind foot, 37. Size of No. 5605, $ adult from Noel, Mo.: total length, 255 ; tail vertebrae, 105 ; hind foot, 36.5. General remarks. The two specimens of this fine chipmunk that Mr. Surber got at Stilwell were shot at the edge of an old field well up on a hillside. The specimen from Noel, Mo., was taken in a similar place. Tarnias striatus venastus is by far the handsomest of the striatus series and is easily distinguished from any of the other subspecies. Its large size and big hind foot place it nearest to griseus, but its bright, rich coloration will at once separate it from that form. With the pale yellow lysteri of the northeast it needs no comparison, and from the small, dull, dark- colored true stricitiis of the southeast it can always be told by its larger size, bigger hind foot, longer tail, and much brighter coloration. Scalops texanus aereus subsp. nov. Type from Stilwell, I. T. No. 5475, ? old adult, collection of E. A. and O. Bangs. Collected by Thaddeus Surber, August 13, 1896. One specimen from Stilwell, I. T. General characters. Size larger than typical S. texanus ; hind foot larger ; colors darker, without orange markings about nose and chest ; skull slightly different. Color. Rich coppery chestnut all over, without golden or orange suf fusions ; slightly duller below than above, and grayer on chin and throat. Cranial characters. The skull of S. texanus sereus as compared with that of true texanus is larger and of a slightly different shape. The skull of texanus h&B a short rostrum and is much bulged between the orbits. The skull of semis has a longer rostrum and does not present the bulged ap pearance between the orbits. Size of type skull: basilar length, 28.4; occipitonasal length, 33.4; zygomatic breadth, 15.2; greatest length of single half of mandible, 21.8. ^2f m The type: total length, 154; tail vertebrae, 24; hind foot, 19. General remarks. Mr. Surber caught the type specimen of Scalops texanus ; , n /'.s' while it was engaged in tunneling on a black-jack ridge at Stilwell. Dr. J. A. Allen* gives the following measurements for Sea lops texanus from Rockport, Texas: Average of twelve adult males, total length, 141 ; tail vertebrae, 25 ; hind foot, 17.8 ; and of eight adult females, total length, 137 ; tail vertebrae, 23 ; hind foot, 16.5. The largest male measured : total length, 147; tail vertebrae, 27; hind foot, 19; and the largest female: total length, 146; tail vertebrae, 25.5; hind foot, 18. Although Dr. Allen gives no cranial characters for the species, the two skulls of texanus that I have examined can be easily told from either the skulls of typical Scalops aqnaticus or 6'. aynuficus argentatu*, apart from the smaller size, by the much shorter rostrum and bulging interorbital region. The skull of ;r< "* is much more like that of aquatint*. Mr. Surber took a fine series of Scalops aqnaticus argentatus at Stotes- bury, Yernon County, Mo., which brings the range of that subspecies very near the range of /S. texanus lereus. JEreus, however, does not ap proach argentatus in any way, its affinities lying wholly with texanus. Bull. Am. Mus. Nut. Hist., vol. vi, 1894, p. 186. VOL. X, PP. 139-144 DECEMBER 28, 1896 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON THE SKUNKS OF THE GENUS MEPHITIS OF EASTERN NORTH AMERICA. BY OUTRAM BANGS. In 1895 * I described a new skunk from Florida as a sub species of the northern Mephitis mephitica (Shaw), and at the same time reviewed, in rather an informal way, the eastern mem bers of the genus Mephitis. Since then I have learned more about the distribution of the eastern skunks and have seen many additional specimens, so that some of my former views have changed. I now consider the Florida elongata entitled to specific rank, and still another form from the Mississippi Valley entitled to recognition. The latter form, which I shall call Mephitis me phitica scrutator, is common in the pine and prairie region of central Louisiana, and extends northward up the Mississippi Valley and eastward through the Alleghany Mountains, grad ually shading into true mephitica. Specimens from the central region from Virginia to Maine are typical of neither form. M. mephitica typica occurs only in high Canadian and Hudsonian regions. M. elongata is abundant, though locally distributed, over the greater part of peninsular Florida and extends up the Atlantic coast at least to southern South Carolina or northern Georgia, where it gradually ceases, and no skunk is found through out eastern North Carolina,! thus leaving elongata and mephitica * Notes on North American Mammals, Proc. Boston Soc. Nat. Hist., vol. xxvi, author's edition, July 31, 1895. f Messrs. H. H. and 0. S. Brimley, in fourteen years of constant col lecting about Raleigh, N. C., have never seen a skunk there, and have only known one to be reported as having been killed. I have made many inquiries of farmers throughout eastern North Carolina and have always got the same answer, that there are no skunks there. Of course, elongata or mephitica might be expected to occur occasionally as stragglers. 26-BioT,. Soc, WASH., Vor,, X, 1896 (139) 140 Bangs Skunks of Eastern North America. separated by a wide area in the east. Just what forms the western limit of the range of M. elongata I am unable to say, but probably it is the heavy swamps of the lower Mississippi. MEPHITIS MEPHITICA (Shaw). 1792. Viverra mephitica Shaw, Museum Lsverianum, 1792, p. 172. 1857. Mephitis mephitica Baird, Mamm. N. Am., 1857, p. 195. Geographic distribution. Whole of eastern North America from Ontario, Quebec, and Nova Scotia to Louisiana, except Florida and the coast belt from thence to Virginia. Its range may overlap that of M. elongata in the lower Mississippi Valley, and undoubtedly does overlap the range of M. hudsonica* in the upper Mississippi Valley. Description. Size small to medium ; tail short to medium, tapering to a pencil ; color pattern variable, but usually black all over except frontal stripe, nuchal patch, two strips extending from nuchal patch to and down sides of tail, tip of tail and some scattered hairs in black part of tail, all of which are white. Fully adult males vary in size according to locality : total length, 595- 682; tail vertebra, 171-241 ; hind foot. 64-83. M. mephitica is distinguished from M. hudsonica by smaller size, shorter tail, tapering to a pencil, and smaller and less elongated skull. It is sepa rated from M. elongata by heavier build, much shorter tail, and propor tionally shorter and broader hind foot. This species may be divided into two well-marked subspecies, Mephitis mephitica mephitica and M. mephitica scrutator. Mephitis mephitica mephitica (Shaw). 1792. Viverra mephitica Shaw, Museum Leverianum, 1792, p. 172. 1895. Mephili* mephitica Bangs, Proc. Boston Soc. Nat. Hist., Vol. XXVI. Author's edition, July 31, 1895. (Name restricted to the Hud- ponian form.) Type locality. North America. Geographic, distribution. Boreal eastern North America ; Nova Scotia, Quebec, and Ontario south to about the northern limits of the United States. Western limit of range unknown. General characters. Size large ; tail short, tapering off to a pencil ; feet very broad and long; heel usually covered with hair, rest of sole naked ; * Mephitis hudsonica (Richardson) extends eastward to Minnesota and probably to western Ontario. It is a very big skunk, fully adult males measuring : total length, 690 ; tail vertebrae, 255 ; hind foot, 83, and larger. The skull is large and long and the dentition heavy. The palate ends in an even curve, without median spine. The color pattern varies but little. The tail is long, very heavily haired, and has a blunt brush-like end, around which the long hairs of the sides of the tail fall, Skunks of Eastern North America. 141 markings very constant, varying only in a trifling difference of length and width of the two lateral white stripes. Color. A narrow frontal stripe, nuchal patch, and two lateral bands extending from nuchal patch to and downsides of tail, white; tip of tail often white ; many white or half white hairs mixed in tail ; rest of head, body, tail, arms, and legs, black. Varies in a slight degree only. Occa sionally the white stripes reach only to the middle of sides of back ; the stripes vary in width but little. Cranial characters. Skull large and massive, the palate ending in an even curve, without median spine. Size of an old adult male skull (No. 3805, Bangs collection from Lake Edward, Quebec) : basilar length, 71.6; occipitonasal length, 74.6 ; zygornatic breadth, 52 ; mastoid breadth, 43.8 ; greatest length of single half of mandible, 52.8. Size of an old adult female skull (No. 3802, Bangs coll. from Lake P'dward, Quebec) : basilar length, 65.2 ; occipitonasal length, 67.2 ; zygornatic breadth, 47.6 ; mastoid breadth, 40.8; greatest length of single half of mandible, 50. Size. Old adult h it leu ttfjtica, always with a large median spine at end of palate. Size of an old adult ^ skull (No. 3052, Bangs coll., topotype) : basilar length, 66.4 ; occipitonasal length, 71 ; zygomatic breadth, 49.2; mastoid breadth, 40; greatest length of single half of mandible, 50.8. An old adult 9 kull (No. 2484, Bangs coll., from P.] itches Ferry, Citrus Co., Skunks of Eastern North America. 143 Fla.) : basilar length, 59.6 ; occipitonasal length, 62.2 ; zygomatic breadth, 45.4 ; mastoid breadth, 35.8 ; greatest length of single half of mandible, 46.4. Size. Old adult $ (topotype, No. 3052, Bangs coll.) : total length, 719 ; tail vertebrae, 321 ; hind foot, 76. An old adult $ (No. 2483, Bangs coll., from Blitches Ferry, Citrus Co., Fla.) : total length, 673 ; tail vertebrae, 330 ; hind foot, 70. General remarks. Mephitis elongata is very different from M. mephitica, and its characters are constant throughout its range. Since I can find no indication of intergradation and the ranges of the two forms are sepa rated, at least in the east, by a strip of neutral ground, where no skunk occurs, M. elongata seems entitled to rank as a full species. (Measurements on next page.} 144 Bangs Skunks of Eastern North America. Individual Measurements of a Series of Eastern Skunks (genus Mephitis). 1 P I 1* 1 j Locality. c S | Measured l>v 0) "5 "g ~ Q) iill "o '3 >n fc 00 < EH r- 1 S Mephitis mephitica mephitica (Shaw). 3801* 3803 3804 3802 2022 2249 3942f 3941 3945 Quebec, Lake Edward cf April 16, 1886; Allen, Bull. Am. Mus. Nat. Hist., Vol. VI 1894, p. 171, foot-note. Type locality. Elk River, Minnesota. Geographic distribution The edge of the forest belt in Minnesota (a region having quite a distinctive mammalian fauna). Limits of range unknown. General characters. Size large, equaling that of leucotis. Color of upper parts rather darker than in leucotis and encroaching all round on under parts, leaving only a small central streak of white on belly. Soles of feet densely furred in winter between the pads, naked in summer. Ears well tufted in winter. Pelage in winter very long and full. Color. In winter upper parts dark iron gray, mixed with yellowish and rusty ; the hairs banded, yellowish rusty and black, and somewhat tipped with white; a small irregular central streak of white on belly; rest of under parts like back ; chest and under side of neck sometimes uniform yellowish brown ; tail dark, the black band of hairs longer and the white tips shorter than in leucotis ; ear tufts well developed in winter, yellowish white ; in summer the general color is darker and more yellowish, owing to the wearing down of the hair. The Squirrels of Eastern North America. 157 Cranial characters. Skull about the size of that of leucotis, showing no characters by which it can be separated from any of the carolinensis series. Size of an average adult skull : basilar length, 54.6; occipitonasal length, 63.2; zygoraatic breadth, 34.6; greatest height of cranium above palate, 19; greatest length of single half of mandible, 35.4. Size. Average measurements of nine adult specimens from Elk River, Minn. : total length, 496.3; tail vertebrae, 220.4; hind foot, 67.2. General remarks. Little is known of this fine squirrel. My knowledge of it comes wholly from Dr. Merriam's description and from four speci mens from the type locality, Elk River, Minn., kindly lent me by him, and ten topotypes in the Bangs collection. One point of some interest is that the northern gray squirrel (Sciurus carolinensis leucotis) occasionally occurs at Elk River in considerable num bers with hypophseus, but has not been known to breed there, appearing only in migrations. This fact suggests the possibility of hypophseus prov ing to be a distinct species when more is known of it. Specimens examined. Total number, 14, from Elk River, Minn. Sciurus carolinensis fuliginosus (Bachman). Bayou Gray Squirrel. 1838. Sciurus fuliginosus Bach. , Proc. Zool. Soc. London, 1838, p. 96 ; Aud. and Bach., Quad. N. Am., Ill, 1853, p. 240, pi. CXLIX. 1895. Sciurus carolinensis fuliginosus Bangs, Proc. Bost. Soc. Nat. Hist., XXVI, p. 543, 1895 ; Rhoads, Proc. Acad. Nat. Sciences, Phila., 1896, p. 196. Type locality. Near New Orleans, -La. Geographic distribution. The bayou region of the coast of Louisiana. General characters. Size larger than true carolinensis; colors rich and dark ; under parts never pure white and often clear ferruginous ; tail long and bushy, the hairs but slightly tipped with white ; feet large, soles naked ; ears with often a woolly tuft at base in winter. Color. Upper parts deep yellowish ferruginous, varied with black ; the hairs banded, many of them having the black band extending to the tip ; tail dark, the hairs yellowish ferruginous at base, then black and tipped with white, the black subapical band very broad and the white tips short ; under parts varying from clear buffy ferruginous, the chin only gray, to smoky gray ; line of demarkation between colors of upper and under parts always low down and irregular ; ear tufts well developed, ferruginous in the examples with ferruginous under parts, grayish white in the exam ples with gray under parts. Cranial characters. Skull a little larger than that of true carolinensis, otherwise similar. Size of an average adult skull: basilar length, 50.6; occipitonasal length, 60; zygomatic breadth, 33.4; greatest height of cra nium, above palate, 19.8 ; greatest length of single half of mandible, 36. Size. Average measurements of ten adult specimens from Gibson, La. : total length, 467 ; tail vertebra, 219.5; hind foot, 67. General remarks. Sciurus carolinensis fuliginosus is confined in its ex treme form to the heavy swamps of the bayou region of the coast of Louisiana. Farther north in the ' prairie ' regions of the same State it 158 Bangs The Squirrels of Eastern North America. begins gradually to approach carolinensis typicus, a large series from Acadia Parish, La., showing this tendency. Mr. Rhoads speaks of the living gray squirrels he saw in the park at Memphis, Tenn.,*and refers them to fuliginosus, on account of their large size and dark coloring. They are probably about like the examples from central Louisiana, which retain the large size and dark color above, but have pure white under parts, and can safely be called intermediates between carolinensis and fuliyinosus, though perhaps nearer fuliginosus. Specimens examined. Total number, 37, from the following localities: Louisiana: Gibson, Terre Bonne Parish, 13; Cartville, Acadia Parish, 3; Point Aux Loups Springs, Acadia Parish, 21. (Those from Acadia Parish not extreme.) Sciurus carolinensis extimus subap. nov. Everglade Gray Squirrel. Type from Miami, Dade Co., Florida. No. 4519, 9 young adult, collec tion of E. A. and O. Bangs. Collected March 12, 1895, by L. Brownell. Total length, 432; tail vertebrae, 194; hind foot, 54. Geographic distribution. Subtropical fauna of south Florida, northward about half way up the peninsula. Habitat. Everglades and oak and cabbage palmetto hammocks. Not found in the ' piney woods.' General characters. Size smallest of the carolinensis series ; tail and hind foot short; color much lighter, more gray than in carolinensis typicus; soles naked ; ears with sometimes a slight woolly tuft at base. Color. Upper parts yellowish gray, the hairs banded black and dull yellow, a few tipped with white (much the same color as the upper parts of leucotis in summer pelage) ; tail light colored, the hairs yellowish at base, then black and tipped with white ; under parts white ; ear tufts white. Cranial characters. Skull smaller than that of true carolinensis, other wise similar. Size of an average adult skull, the type : basilar length, 47 ; occipitonasal length, 55; zygomatic breadth, 31.2; greatest height of cranium above palate, 18 ; greatest length of single half of mandible, :>'!. 2. Size. Average measurements of seven adult specimens from Miami, Fla. : total length, 438.4 ; tail vertebrae, 190.9 ; hind foot, 47. General remarks. Sciurus carolinensis extimus represents in its small size, short tail, and small hind foot the extreme of differentiation of the caro linensis series, but differs widely from true carolinensis, its nearest geo~ graphical cousin, in its much grayer color. This yellowish gray color is probably highly protective, the animal spending most of its life among trees covered with the gray Spanish moss, Tillandsia usneoides, which its color almost exactly matches in tone. S. carolinensis extimus is only typical in the peculiar subtropical fauna of the everglades and southern part of the Florida peninsula. Specimens from Citrus Co., Fla., are larger and darker in color and are rather nearer to true carolinensis than to extimas. The gray squirrel of northern Florida is true carolinensis. *Proc. Acad. Nat. Sciences Phila., 1896, p. 196. The Squirrels of Eastern North America. 159 Specimens examined. Total number, 8, from the following localities: Florida: Miami, Dade Co., 7; Oak Lodge (east peninsula, opposite Micco), Brevard Co.,1; also 21 from Citrus Co. (Citronelle, 2 ; Blitches Ferry, 19), which are intermediates between carolinensis and exlimus. Sciurus hudsonicus hudsonicus (Erxleben). Red Squirrel ; Chickaree. 1777. Sciurus vulgaris e hudsonicus Erxleben, Mammalia, 1777, p. 416. 1778. Sciurus hudsonius Pallas, Nov. Spec. Glir.. 1778, p. 376. 1820. Sciurus rubrolineatus Desmarest, Mamm., I, 1820, p. 333 (Encyclo pedic Methodique). 1827. Tamia hudsonia Lesson, Man. Mamm., 1827, p. 231. 1843. Tamias rubrolineatus Schinz, Syn. Mamm., II, 1843, p. 48. 1877. Sciurus hudsonius var. hudsonius Allen, Monog. N. Am. Sciuridse, 1877, p. 672. Type locality. Hudson Strait. 'Ad fretum Hudsonis.' Geographic distribution. Boreal North America, from Labrador to Alaska, south to Maine and the northern peninsula of Michigan, and along the tops of the higher Alleghanies to Roan Mountain, North Carolina. Habitat. Spruce and fir forests. Feeds largely on the seeds of conifers. The northern red squirrel is excessively abundant in all favorable situ ations. In many places one can often count twenty or thirty individuals within sight or hearing at one time. Always noisy and jerky in its motions, the red squirrel is usually tame and unsuspicious. It feeds and lives both on the ground and in the trees, and is a very agile climber. General characters. Size smallest of the eastern squirrels; tail short, flat, and narrow ; a decided difference in color and markings between winter and summer pelage ; * dorsal stripe in winter chestnut rufous ; sides olivaceous gray ; white of underparts vermiculated with black ; in summer pelage hardly distinguishable from the next subspecies by color alone ; soles densely furred in winter and somewhat so in summer ; ear tufts in winter, long, protruding well beyond the ear; pelage in winter very full and soft. Color. Winter pelage : Upper parts with a broad dorsal band extending from between the ears down upper surface of tail, bright chestnut rufous ; sides, upper surface of legs and arms, and cheeks olivaceous gray, the hairs banded with black ; upper surface of feet and^hands often more yel. lowish ; under parts grayish white, thickly vermiculated with blackish, the hairs plumbeous at base. An indistinct blackish line usually shows on sides between colors of upper and under parts. On the upper surface of the tail the hairs are clear chestnut rufous, and only a few have black rings ; on the sides and lower surface they are dull yellowish at base and tips and black in the middle. Summer pelage : Upper parts with no dorsal stripe ; a peculiar ferruginous gray with an olivaceous cast, the hairs banded with black, becoming clear ferruginous on upper surface of hands and feet, and sometimes legs and arms also ; under parts white, often *In this connection, see Allen on 'Seasonal Variation in Color in Sciurus hudsonius,' Bull. Am. Mus. Nat. Hist., Vol. Ill, p. 41, 1890. 160 Bangs The Squirrels of Eastern North America. suffused with rusty yellow ; a broad black stripe along side separating colors of upper and under parts. Cranial characters. Skull light, developing very slight lateral ridges with age ; rostrum short and blunt ; nasals ending atfronto-premaxillary suture; postorbital process of frontal light and long; zygoma standing out squarely from root, and more flaring than in either the fox or gray squirrels; audital bullse large. Penultimate upper premolar either ab sent or present in the adult, though more often absent, and when present very minute. Size of an average adult skull: basilar length, 38.4; occipitonasai length, 45; zygomatic breadth, 26.2; greatest height of cranium above palate, 15.6; greatest length of single half of mandible, 26.4. Size. Average measurements of four adult specimens from Hamilton Inlet, Labrador: total length, 309; tail vertebra?, 120.5; hind foot, 47.75 (all four are very old adults and the averages therefore large). Average measurements of ten adult specimens from Digby, Nova Scotia: total length, 296.5; tail vertebrae, 118.2; hind foot, 45.2 General remarks. Sciurus hudsonicus has but one bad synonym the Sciurus rubrolineatus of Desmarest. Desmarest based his name wholly on the 'Ecureuil rouge (species nova)' of the French edition of Warden's De scription of the United States, published in 1820. Warden described under this name a red squirrel in winter pelage, assigning it no habitat. It is the only red squirrel Warden gives, and it is impossible to say which race it belonged to. In addition to Sciurus rubrolineatus Desmarest gives Sciurus hudsonius, his description of the latter being taken from a summer specimen. It is evident that the great difference between summer and winter specimens alone led Desmarest into the belief that there are two species of red squirrels. Sciurus hudsonicus typicus belongs to the spruce and fir belt and only extends south as far as these trees. Wherever the Transition and Cana dian faunas meet, as in central New York, New Hampshire, and Minne sota, intermediates between hudsonicus typicus and hudsonicus loquax occur. Only a very short distance south, however, into truly Transition country loquax is found in the typical form. Specimens examined. Total number, 89, from the following localities: Labrador : Hamilton Inlet, 4. Nova Scotia: Digby, 16; Granville, 4; James River, 1 ; Schenacidae (Cape Breton), 6. New Brunswick : Catnpobello Island, 9. Quebec : Lake Edw r ard, 5. Ontario : North Bay, 8 ; Nepigon, 1 ; Peninsula Harbor, 10. Saskatchewan:Batoche, 5. Maine : Greenville, 5; Upton, 5. West Virginia : White Sulphur Springs, 1. North Carolina : Roan Mountain, 4. Intermediates : New Hampshire : Franconia, 2 ; Antrim, 1. New York: Peterboro, 1. The Squirrels of Eastern North America. 161 Sciurus hudsoiiicus loquax subsp. nov. Southern Chickaree. 1815. Sciurus carolinensis Ord, Guthrie's Geog., 2d Am. ed., II, 1818, p. 292. (Name preoccupied by Gmelin for the southern gray squirrel.) Type from Liberty Hill, Conn., No. 4270, tf adult, collection of E. A. and O. Bangs. Collected by Outram Bangs December 24, 1895. Total length, 323 ; tail vertebra, 141 ; hind foot, 47. Geographic distribution. Transition and Carolinian zones, from southern Maine and southern Minnesota to Virginia, west to the edge of the plains. Not found in the tops of the higher Alleghanies where hndsonicus typicus takes its place. Habitat. Mixed woods, groves, and in fact almost everywhere ; per haps most numerous where there are large tracts of Pinus rigida, the seeds of which it is very fond of. Very abundant over the whole of its range except the southern part, where it becomes rare and local. General characters. Size somewhat larger than hudsonicus typicus ; tail longer ; color of dorsal stripe in winter pelage usually brighter red ; under parts pure grayish white, not vermiculated; soles and palms furred in winter, naked in summer. Color. Winter pelage : upper parts with a broad dorsal band extending from between ears down upper surface of tail, varying from bright fer ruginous to orange rufous ; sides and upper surface of arms and legs yellow or rusty gray, with sometimes an olivaceous cast, the hairs banded with black ; under parts clear grayish white, without verrniculations, the hairs plumbeous at base ; usually a black line shows indistinctly along sides between colors of upper and under parts ; hairs of upper surface of tail clear ferruginous ; those of lower surface and sides dull yellow at base and tip and black in middle. Summer pelage : Impossible to tell with cer tainty from summer pelage of hudsonicus typicus, but usually more fer ruginous gray and less olivaceous gray. Cranial characters. Skull averaging larger than that of hudsonicus typi cus; otherwise similar. Size of an average adult skull (the type) : basilar length, 40; occipitonasal length, 46.4; zygomatic breadth, 27; greatest height of cranium above palate, 16.4 ; greatest length of single half of mandible, 28. Size. Average measurements of eight adult specimens from Liberty Hill, Conn. : total length, 318.3; tail vertebra, 133.5 ; hind foot, 47.42. General remarks. Professor Baird, in his Mammals of North America, first pointed out the fact that northern examples of Sciurus hudsonicus had the under parts vermiculated with black and the southern examples did not. Dr. Allen, in his Monograph of the North American Sciuridse, dwelt at some length on the differences between the two races, but did not sep arate them by name. In winter pelage Sciurus hudsonicus typicus and Sciurus hudsonicus loquax can be told apart at a glance, but in their sum mer coats they are not so easily distinguished ; as a rule, however, loquax is more rusty and less olivaceous, and the difference in size between indi viduals of the same age is well marked, hudsonicus typicus being always the smaller of the two. 29-Biot. Soc. WASH., Vor,. X, 1896 1()2 Bangs The Squirrels of Eastern North Americo. Specimens examined. Total number, 56, from the following localities : Connecticut: Liberty Hill, 11. Massachusetts: Wareham, 24; Way land, 1; Brookline, 1. Indiana: Denver, 14. Minnesota: Steel Co., 1. Wisconsin : Waupaca, 2. North Carolina : Magnetic City, foot of Roan Mountain, 2 (not quite typical). Genus SCIUROPTERUS F. Cuvier. Tail flat, laterally expanded, densely haired with fine hairs; an expan. sion of the skin of the sides extends from wrist to ankle, and when spread acts, with the flat tail, like a parachute, enabling the animal to make long, slanting descents through the air ; pelage very fine and dense ; skull light ; audital bullte large ; end of the pterygoid process resting against audital bulla ; rostrum short; occipital region slightly drooping and turned under; interorbital constriction deep and zygoma drooping to make room for the large eye ; penultimate upper premolar always present ; nocturnal animals. Sciuropterus sabrinus (Shaw). Severn River Flying Squirrel. 1801. Sciums sabrinus Shaw, Gen. Zool., I, 1801, p. 157. 1778. " Sciuro volante majore" Pallas, Nova Spec. Glires, 1778, p. 354 (not a scientific name). 1788. Sciurus hudsonius Gmelin, Syst. Nat., I, 1788, p. 153 (preoccupied). 1815. Sciurus labradorius Ord, Guthrie's Geog. , 2d Am. ed., 1815, p. 292 (nomen nudum). 1828. Pteromifs sabrinus Richardson, Zool. Journ., Ill, 1828, p. 510; And. and Bach., Quad. North Am., Ill, 1853, p. 202. 1829. Pteromys hudsonius Fischer, Syn. Mamm., 1829, p. 365 ; Baird, Mamm. North Am., 1857, p. 288. 1874. Sciuropterus volucella var. hudsonius Allen, Proc. Boston Soc.. Nat. Hist., XVI, 1874, p. 289 ; Monog. N. Am. Sciuridse, 1877, p. 655. Ti/pe locality. Severn River, James Bay, Canada. Geographic distribution. Boreal North America, south in the east to northern New York and southern New Hampshire. Habitat. Mixed woods and forest; nocturnal, spending the day in hollows and nests in the trees. General characters. Size, largest of the eastern flying squirrels; tail broad, the hairs long ; hind foot large ; a decided difference in color be tween winter and summer pelage; under side of tail washed with sooty ; hairs of under parts, plumbeous at base, showing through, and giving a decidedly gray appearance to under parts; soles furred both in winter and summer, only the pads naked. Color. Winter pelage: upper parts very glossy, wood brown to cinna mon, often somewhat shaded with yellow, darkening on tail towards tip to sooty; hairs of back and sides dark plumbeous below r , the merest tip being colored, the plumbeous color therefore showing through whenever the fur is the least disturbed ; upper surface of feet and hands sooty gray ; The Squirrels of Eastern North America. 163 cheeks gray ; a black orbital ring ; ears sparsely haired, dusky ; under parts dirty white, the hairs plumbeous at base ; under side of tail yel lowish white washed with drab and sooty. Summer pelage : whole upper parts uniform sooty drab. Cranial characters. Skull large; audital bullse small and flat (for the genus) ; the bone dense ; nasals slightly turned up at end pug-nosed ; all the teeth, including penultimate upper premolar, large. Size of an average adult skull: basilar length, 32.4; occipitonasal length, 38.4; zygomatic breadth, 22.8; greatest height of cranium above palate, 12.4; greatest length of single half of mandible, 23.4. Size. Average measurements of seven adult specimens from Greenville, Maine: total length, 278.6; tail vertebra, 130.4; hind foot, 37.6. General remarks. Dr. Allen, in 1874, relegated this fine species to sub- specific rank, calling it a variety of Sciuropterus volans (alias volucella], and followed the same arrangement in 1877 in his Monograph of the Sciuridse, where he makes the statement "Grades insensibly into var. volucella." How or where Dr. Allen found intergrades I am at a loss to know. In reality Sciuropterus sabrinus and S. volans are two distinct species and never intergrade. Wherever their geographic ranges meet they occur together, often in the same wood, each species keeping distinct and retaining its characters as well as where far removed from contact with the other. S. sabrinus meets and overlaps the range of S. volans for a short distance, wherever the Canadian and Transition faunas meet. Dr. C. Hart Merriam found both species breeding in the Adirondack region of New York, and in his interesting accounts of the habits of these squirrels clearly shows the two to be specifically distinct, although he retained the varietal names of Allen. In the same wood lot at Peterboro, N. Y., Mr. Gerrit S. Mil ler, Jr., took on November 22, 1894, a fine example of S. volans, and on December 28, 1895, a pair of S. sabrinus. He has kindly lent me the specimens, which are now before me. I have both species from Han cock, N. H., but there volans is apparently the more common. I have yet to see a specimen that is in any way intermediate between S. sabrinus and S. volans, and if one did turn up it would be safe to consider it a natural hybrid and not an intergrade. Specimens examined. Total number, 24, from the following localities : Nova Scotia: Annapolis, 3; Digby, 1. Ontario: Nepigon, 1. Maine : Greenville, 8 ; Bucksport, 2. New Hampshire : Hancock, 1. New York : Peterboro, 2. Arctic America : Eed River, 2; Fort Resolution, 1 ; Big Island, 1 ; Moose Factory, 2. Sciuropterus silus sp. nov. Alleghany Mountain Flying Squirrel. Type from top of Katis Mountain, White Sulphur Springs, W. Va., at an altitude of 3,200 feet. No. 4931, tf adult, collection of E. A. and O. Bangs. Collected by Thaddeus Surber September 2, 1895. Total length, 214 ; tail vertebrae, 92 ; hind foot, 28. 164 Bangs The & aU my efforts to get ad _ ditional specimens having so far failed. Lieut. Wirt Robinson spent part of the summer of 1896 at White Sulphur Springs, W. Ya,, and kindly offered to get me flying squirrels. He succeeded in taking but one, a perfectly typical example of S. volans. It was taken at an altitude of about 1,200 feet lower than the type of *!?. silus. It is probable that the ranges of the two species overlap. Sduropterus silus bears no very close relationship to S. volans, although it is even smaller than that species. Its affinities lie with S. sabrinus, of which it is probably the Alleghany Mountain representative. It is, how ever, so very much smaller than that species and differs from it so much in other respects that I have accorded it specific rank. Specimen examined. The type. Sduropterus volans volans (Linn.). Southern Flying Squirrel. 1758. Mus volans Linn., Syst. Nat., ed. 10, I, 1758, p. 63. 1758. Sciurus volans Linn., Syst. Nat., ed. 10, I, 1758, p. 64 (in part). 1788. Sciurus volucella Pallas, Nov. Spec. Glires, 1788, p. 351. 1818. Pterornys volucella Desmarest, Nouv. Diet, d'Hist. Nat., XXVI F, 1818, p. 406; And. and Bach., Quad. N. Am., I, 1849, p. 216, pi. XXVIII; Baird, Mamm. N. Am., 1857, p. 286. 1828. Sduropterus volucella Geoffrey, Diet. Class. d'Hist. Nat., XIV, 1828, p. 132. 1874. Sduropterus volucella var. volncella Allen, Proc. Boston Soc. Nat. Hist, XVI, 1874, p. 189 ; Monog. N. Am. Sciuridse, 1877, p. 655. 1890. Sduropterus volans Jordan, Man. Vertebrates, 1890, p. 324, foot note. Tlie Squirrels of Eastern North America. 165 Type locality. North America (Virginia?).* Geographic distribution. Transition and Carolinian zones of the east ; from northern New York and southern New Hampshire south to Georgia ; west to the plains. Habitat. Forests and groves ; everywhere abundant, nocturnal, spend ing the day in hollow trees or nests made of bark, leaves, and moss. General characters. Size medium, considerably smaller than S. sabrinus ; hind foot smaller; tail narrower and of a rather different shape, tapering off more toward the end ; no decided difference in color between winter and summer pelage ; hairs of under parts white to the base ; soles furred in winter, only the pads naked ; wholly naked in summer ; palms naked throughout the year. Color. Winter pelage : upper parts drab, often shaded irregularly with russet, slightly darker on upper surface of tail, the hairs plumbeous below, only the tips being colored ; upper surface of hands grayish white ; upper surface of feet drab, the toes and inner edge grayish white ; cheeks gray ish white; a black orbital ring; ears nearly naked, the skin dusky; under parts pure white, usually washed on lower surface of tail and some times of legs and flying membrane with pinkish buff, the hairs white basally. Summer pelage not differing materially from the winter, the upper parts usually more russet, having the appearance of being due to fading and wearing rather than to a change in color ; the white of under parts often soiled, and the color of under surface of tail more intense. Cranial characters. Skull smaller and lighter than that of S. sabrinus, audital bullae larger, not flattened, the bone light and papery ; nasals not so much turned up at ends ; teeth, including penultimate upper premolar, lighter throughout. Size of an average adult skull : basilar length, 28.8 ; occipitonasal length, 34.2; zygomatic breadth, 21 ; greatest height of cra nium above palate, 12.2; greatest length of single half of mandible, 20.6. Size. Average measurements of seven adult specimens from Liberty Hill, Conn. : total length, 234.5 ; tail vertebne, 99.6; hind foot, 31.4. General remarks. Sciuropterus volans retains its characters, with only a slight range of individual variation, throughout the whole of the Transi tion and Carolinian zones, but in the lower Austral Zone begins to ap proach the slightly different form of peninsular Florida. A series from St. Marys, Ga., is intermediate both in cranial characters and color be tween S. volans typicus and S. volans querceti. In the north S. volans over laps the range of S. sabrinus for a short distance, the two meeting wher ever the Transition and Canadian faunas run into each other. Specimens examined. Total number, 28, from the following localities: New Hampshire: Hancock, 3. New York : Peterboro, 1. *No definite type locality can be assigned the southern flying squirrel. Linnaeus based his Must volans on Ray, Edwards and Seba, and himself gives Virginia and Mexico as its habitat. Ray tells us " In Nova Hispania atque etiam Virginia reperitur. " Seba does not specify where his specimen came from, though he calls it Sciurus volans virginianus. Edwards says : " They are brought to us from several parts of North America and have been of late discovered in Poland." 166 Bangs The Squirrels of Ext/picus, from which it differs in having the upper parts more uniform russet, and the under parts, especially the under surface of tail, strongly washed with the same color; soles naked ; audital bullre wheel-shaped, very large and deep. Color. Upper parts russet, shading to yellowish drab in places ; cheeks grayish white ; upper surface of feet and hands sooty gray ; toes rather lighter; under parts white, a good deal shaded with pinkish russet. Cranial characters. Skull similar to that of & volans, except that the audital bullre are much larger, more inflated, and broadly wheel-shaped (see figs. 31 and 32). Size of an average adult skull (the type) : basilar length, 30.2 ; occipitonasal length, 34.6 ; zygomatic breadth, 21 ; greatest height of cranium above palate, 12.8 ; greatest length of single half of man dible, 21.4. Size. Average measurements of three adult FIG. 31. Audital bulla of . _ e ~., , , , Sciuropterus volans. specimens from Citronelle, Fla.: total length, 237.66; tail vertebrae, 102.66; hind foot, 31.33. General remarks. Sciuropterus volants querceti passes into true volans in southern Georgia, a series of specimens from St. Marys, Ga., being in termediate between the two. A specimen from Powhatan Plantation, near Gibson, La., without a skull, seems referable to this form, and may thus , extend its range tc the coast of Louisiana. FIG. 32. -Audital bulla of Sciuropterus v. querceti. Specimens examined. Total number, 3, all from Citronelle, Citrus Co., Fla. EXPLANATION OF PLATES. All the figures are life size and were drawn by Dr. J. C. McConnell. PLATE VIII. Fig. 1. Sciurus nigerL. old adult, Citronelle, Fla. (No. 1978, Bangs coll.) 2. Sciurus ludovicianus Custis ^ old adult, Point aux Loups Springs, La. (No. 2929, Bangs coll.) 3. Sciurus ludovicianus vicinus Bangs 9 old adult; the type, White Sulphur Springs, W. Va. (No. 5215, Bangs coll.) 4. Sciurus carolinensis hypophseus Merriam ^ old adult, Elk River, Minn. (No. 3942, Merriam coll.) PLATE IX. Fig. 1. Sciurus carolinensis leucolis Gapper 9 old adult, Liberty Hill, Conn. (No. 1043, Bangs coll.) 2. Sciurus carolinensis fuliginosus (Bach.) ^ old adult, Gibson, La. (No. 2833, Bangs coll.) 3. Sciurus carolinensis Gmelin 9 old adult, St. Marys, Ga. (No. 5141, Bangs coll.) 4. Sciurus carolinensis extimus Bangs 9 adult; the type, Miami, Fla. (No. 4519, Bangs coll. ) PLATE X. Fig. 1. Sciurus hudsonicus loquax Bangs $ old adult ; the type, Liberty Hill, Conn. (No. 4270, Bangs coll.) 2. Sciurus hudsonicus (Erxleben) cT old adult, Hamilton Inlet, Lab rador. (No. 3956, Bangs coll.) 3. Sciuropterus sabrinus (Shaw) cT old adult, Greenville, Me. (No. 4962, Bangs coll.) 4. Sciuropterusvolansquerceti Bangs 9 old adult; the type, Citronelle , Fla. (No. 2451, Bangs coll.) 5. Sciuropterus volans (L.) ^ old adult, Liberty Hill, Conn. (No. 4269, Bangs coll. ) 6. Sciuropterus silus Bangs $ adult ; the type, top of Katis Moun tain, White Sulphur Springs, VV. Va. (No. 4931, Bangs coll.) (167) PROC. BIOL. SOC. WASH., X, 1896 PL. VIII 1 SCIURUS NIGER 2 SCIURUS LUDOVICIANUS 3 SCIURUS LUDOVICIANUS VICINUS 4 SCIURUS CAROLINENSIS HYPOPH.EUS PRCC. BIOL. SOC. WASH., X, 1896 PL. IX 1 SCIURUS CAROL1NENSIS LEUCOTIS 3 SCIURUS CAROLINENSIS 2 SCIURUS CAROLINENSIS FULIGINOSUS 4 SCIURUS CAROLINENSIS EXTIMUS PROC. BIOL. SOC. WASH., X, 1896 PL. X 1 SCIURUS HUDSONICUS LOQUAX 2 SCIURUS HUDSONICUS 3 SCIUROPTERUS SABRINUS 4 SCIUROPTEKUS VOLANS QUERCETI 5 SCIUROPTERUS VOLANS 6 SCIUROPTERUS SILUS VOL. X, PP. 169-174 DECEMBER 29, 1896 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ROMEROLAGUS NELSONI, A NEW GENUS AND SPECIES OF RABBIT FROM MT. POPOCATEPETL, MEXICO. BY DR. C. HART MERRIAM. Among the many new and interesting mammals collected by Mr. E. W. Nelson in Mexico during the past five years, one of the most remarkable is a small, short-eared, tailless rabbit dis covered high up on Mt. Popocatepetl, at and above an alti tude of 3,000 meters (approximately 10,000 feet). This singular animal has exceedingly short hind legs, and instead of moving by a series of leaps like ordinary rabbits, runs along on all fours, and lives in runways in the grass like the meadow mice. Mr. Nelson has prepared, at my request, the following account of his experience with this extraordinary animal. He says : "On my first visit to Mt. Popocatepetl in the spring of 1893, I learned that these little rabbits were found there, and on my return to the city of Mexico I prepared for an expedition to secure them. On January 5, 1894, my assistant, Mr. E. A. Gold man, and I made our camp on the side of a canon at an altitude of about 3,350 meters (11,000 feet) on the northwest slope of the mountain. We were accompanied by three Indian hunters and our packer. Among the firs and alders at this altitude the north erly slopes of the hills and canons are covered with a luxuriant growth of saccaton grass in huge bunches, from three to six feet across, and often reaching a height of 6 or 8 feet, which covers the ground so that the only open spaces are small spots scattered irregularly here and there. A search under the overhanging masses of long grass blades showed a perfect network of large arvicola-like runways tunneling through the bases of the tus- 30 BIOL. Soc. WASH., VOL. X, 189G (169) 170 Merriam A New Rabbit from Mexico. socks, and passing from one to another under the shelter of the outcurving masses of leaves. It was evident that the rabbits were very numerous here, and we all proceeded to hunt the vicinity carefully for them. The first day I saw three, but was unable to get a shot at any. One came running through the grass along one of the hidden trails and, seeing me, stopped in a little opening only seven or eight feet away. It was too near to shoot, and so escaped after looking at me with inquiring eyes for a few moments. The next evening I shot one by taking a stand on a large log, whence I could see several small openings in the grass, and saw one as it stopped a moment at the en trance of a runway. By persistent hunting for three days my Indians secured three more. " On our first night wire snares were set without success, so the next night we put out a lot of steel traps in the runways. This latter method was very successful, and three fine specimens were taken in a small area a few yards across. So far as ob served, these animals are strictly limited to the heavy growth of saccaton grass, between about 3,050 and 3,650 meters (10,000 and 12,000 feet), a few ranging a little above and below these limits in favorable places along canon slopes. I found them equally numerous in the heavy grass on canon slopes and hill sides and in the dense growth of grass about the sides of the small park-like openings in the forest. They make their forms within the matted bases of the huge grass tussocks by tunnel ing passageways along the surface of the ground through the mass of old grass leaves and stems and then hollowing out snug retreats within the weather-proof shelters thus obtained. Their concealed runways were intermingled with those of the common meadow mice of the mountains, and the striking resemblance in coloration and habits between the two animals was remarkable. Like the arvicolas, the rabbits are mainly nocturnal, but are occasionally found moving about by day. They become more active just at dusk, and on frosty mornings sometimes come out at sunrise into the small openings among the grass to bask in the warmth. My Indian hunters claimed that they often found them out sunning themselves in this way on summer afternoons after cold, heavy showers. " This species has practically no external tail, though in some specimens there is a small fleshy papilla two or three millimeters in length ; in others even this is absent. In this respect the animal resembles the pikas (Lagomys}." A New Rabbit from Mexico. 171 Animals differing so widely in habits and manner of progres sion as the present species and the ordinary rabbits would be expected to differ in their skeletons. Fortunately, Mr. Nelson preserved a perfect skeleton of the new rabbit, which on com parison with those of the several subgenera of Lepus shows differ ences of considerable morphologic weight. The clavicle is complete and articulates directly with the sternum (fig. 33) a thing that never happens in the genus Lepus. Huxley describes the clavicle of the rabbit as " incom plete at both ends," and Flower states that it "is very short and is suspended by long ligaments between the scapula and the sternum." The manubrium or presternum is broadly expanded between and anterior to the articulation of the first pair ^*)J of ribs (fig. 33), a condition unknown in the genus Lepus, in which it is always long and narrow (fig. 34). Flower calls attention to the corre lation existing between the form of the presternum and the degree of development of the clavicle, stating that " the presternum is compressed and produced forwards in FlG . 33 _ sternum of those rodents in which the Romeroiagus nei- clavicle is absent or rudimen- soni (nat> size) ' tary," as the hares, and " is generally broad in the forms which have the clavicle well developed, as the rats, beavers, &c." This interesting correlation is well exemplified in the Popocatepetl rabbit, which, having a complete clavicle, has also a broad manu brium. The segments of the mesosternum (between the pre sternum and xiphoid) are only three in number (fig 33), while in all the subgenera of Lepus the number is four (fig. 34). The ribs are correspondingly reduced, only six pairs instead of seven articulating with the sternum. The tubercles of the ribs are not produced into spiniform processes, as in Lepus, and disappear in the sixth pair. In Lepus they extend to the eighth pair. The scapula is rather narrow, with a long metacromial process, as in Lepus. There are four sacral vertebrae, as in Lepus (the first and anterior part of the second articulating with the ilia), and nine FIG. 34 Sternum of Lepus timidus (much- reduced). 172 Merriam A New Rabbit from Mexico. caudal vertebrae, the last three of which are upturned and rudi mentary. The fifth cervical vertebra is peculiar. Its transverse process projects directly outward instead of backward, and its inferior lamella has only a trace of the posterior extension usual in rab bits. The metapophyses begin on the tenth dorsal vertebra and are present in all the succeeding vertebrae to the last lumbar, in clusive. The anapophyses are much as in Lepus proper, being present, though small, on the ninth to twelfth dorsals, inclusive, and on all the lumbar vertebra? except the sixth and seventh. The transverse processes of the lumbar vertebrae are peculiar, each developing a broad posterior flange, which extends the full length of the side of the vertebra. Hypopophyses are present on the first, second, and third lumbar vertebrae, as in Lepus, though relatively short. The bones of the legs and feet show a number of more or less important differences, some of which may be mentioned here. The depression on the inner side of the trochlear facet of the humerus is small and flat instead of deeply sulcate ; the fibular malleolus is less strongly developed ; the navicular bone differs materially in form and its inferior crest is conspicuously shorter than in Lepus, and does not reach forward beneath the bases of the metatarsals. The skull, singularly enough, does not show the departure from Lepus that one would expect from a study of the other bones. It agrees in the main with skulls of the American cot tontails (subgenus Sylvilagus), but differs in the postorbital pro cesses, which are small, divergent posteriorly, and altogether wanting anteriorly, and in the jugal, which is greatly elongated posteriorly. The interparietal is distinct, and in old age becomes ankylosed with the supraoccipital. The thoroughly leporine character of the skull shows that the animal can hardly be re garded as ancestral to Lepus, as might have been inferred from its short ears, short hind legs, and various skeletal characters, but that it is a specialized offshoot from the genus Lepus itself. The taxonomic value of the characters which serve to dis tinguish the Popocatepetl rabbit from the true rabbits, and more particularly the peculiarities of its sternum and clavicle, require the erection of an independent genus for its reception. Hereto fore the genus Lepus has enjoyed the distinction of coincidence in characters with the family to which it belongs. Now the A New Rabbit from Mexico. 173 family circle of the Leporidse must be extended to include the new member.* From the foregoing it will be evident that the new animal is of unusual interest to naturalists. The curious combination of its anatomical characters, the peculiarity of its mode of locomo tion, the oddity of its habits, and the isolation of its home high up on lofty Popocatepetl give it an interest quite apart from that which attaches to most new discoveries. For this reason it affords me special pleasure to bestow upon the new and remarkable genus, of which it is the type, the name Romerolagus, in honor of the venerable Senor Don Matias Romero, Envoy Extraordinary and Minister Plenipotentiary from Mexico, as a slight token of appreciation for the active interest he has taken in the explorations of the United States Biological Survey in Mexico, and in recognition of the many courtesies he has ex tended to our field naturalists during the past five years. ROMEROLAGUS gen. nov. Type. Romerolagus nelsoni sp. nov., from Mt. Popocatepetl, Mexico. Diagnosis. Size small; ears exceedingly short, shorter than in any known species of Lepus ; hind legs and feet short ; skull much as in Lepus (submenus Si/lvilagus), except that the postorbital processes are small, di vergent, and wanting anteriorly, and the jugals much elongated poste riorly ; clavicle complete and articulating with both sternum and scapula ; presternum broadly expanded anteriorly, much broader than long in front of first pair of ribs [narrow and slender in Lepus} ; mesosternum of 3 seg ments [4 in Lepus} ; 6 pairs of ribs articulating with sternum [7 pairs in Lepus~} ; transverse process of 5th cervical vertebra directed straight out ward (instead of backward), its inferior lamella lacking the usual posterior extension ; transverse processes of all lumbar vertebrae broadly expanded, their bases covering entire length of vertebrae ; hypopophyses present on first 3 lumbar vertebrae, but small ; inferior crest of navicular bone short and not produced under base of metatarsal. Romerolagus nelsoni sp. nov. Popocatepetl Rabbit. Type from Mt. Popocatepetl, Mexico (altitude 3,350 meters or 11,000 feet). No. 57949, rT ad., U. 8. Nat. Mus., Dept. Agric. coll. Collected Jan. 6, 1894, by E. W. Nelson and E. A. Goldman. Original number 5639. Geographic distribution. Boreal Zone of Mt. Popocatepetl, between the altitudes of 3,050 and 3,660 meters (10,000-12,000 feet). * It is singular that of the four characters given by Flower and Lydek- ker in the first sentence of their diagnosis of the family Leporidse ("im perfect clavicles, elongated hind limbs, short recurved tail, and long ears " ) , not one applies to the Popocatepetl rabbit. 174 Mrrriam A New Rabbit from Mexico. General characters. Size small ; ears and hind feet very short ; no ex ternal tail ; coloration dark. Color. Upper parts, sides, and pectoral collar grizzled grayish brown, with a yellowish suffusion, and strongly mixed with black-tipped hairs (the yellowish due to a broad subapical zone of this color on each hair) ; belly and chin smoky grayish washed with buffy ; upper surfaces of feet buffy yellowish, much lighter than rest of upper parts ; ears without markings. Cranial characters. Skull similar in a general way to that of Lepns syl- rttticus, but much smaller ; supraorbital processes small, slender, divergent, and not approaching frontals posteriorly ; jugals much elongated and incurved posteriorly, not defined anteriorly (supraorbital notch absent) ; braincase less decurved and more depressed posteriorly than in Lepus si/I- vaticus and its allies; zygomata standing far out from sides of cranium ; palatal bridge relatively broad ; audital bullse moderately inflated. Measurements. Type specimen: total length, 311 ; tail vertebrae, 0; hind foot, 53 ; ear from notch in dry skin, 36. Average of 6 adults from type locality : total length, 295; hind foot, 52. The type is the largest of the seven specimens. Remarks. Mr. Nelson's account of the habits of this rabbit, as observed by him on Mt. Popocatepetl, has been given at the beginning of the pres ent article. Mr. Nelson saw runways which he believes were those of the same species, at an altitude of 3,050 to 3,350 meters (10,000 te 11,000 feet) on the southeast side of Mt. Iztaccihuatl. INDEX Names of new species and subspecies are printed in heavy type Page Abies arizonica 115-118 lasiocarpa 115, 117 subalpina 115, 117 Acer saccharinum 33 Aconitum uncinatum 31 ^Egialitis tenella 54 .Kijilii* thoracica 53~54 varia 54 Agrostis elata 30, 41 Aletris farinosa 38 Aloefolia 104 Amaraiitus albus 38 paniculatus 38 Anemone nemorosa 31 Anhalonium lewinii viii Annual meeting xii Anychia dichotoma 33 Aphylloti uniflorum 37 Apocynum androssemifolium 30, 36 Arabis dentata 31 Aralia nudicaulis 35 Arctic plants, Earliest record of. 103 Arctomys monax 95 Aristida purpurascens 30, 41 Arvicola (Synaptomys) gossii 60 Asclepias albicans viii obtusifolia 36 rubra 36 variegata 36 verticillata 36 Aspidium cristatum 43 Aster concolor 35 Atalapha borealis 101 cinerea ... 101 Bailey, Vernon : I/ist of mammals of Dis trict of Columbia 93-101 Tamarack swamps as boreal islands viii Two mammals new to vicinity of Washington viii Bangs, Outram : Florida deer 25-28 Squirrels of eastern North Amer ica 145-167 Weasels of eastern North America. 1-24 Skunks of genus Mephitis of east ern North America 139-144 Mammals from I^ake Edward, Que bec 45-52 31 BIOL. Soc. WASH., VOL. X, 1896 Page Bailey, Vernon : New mammals from In dian Territory and Missouri 135-138 The cotton mouse (Peromyscus gos- sypinus) 119-125 Batchelder, Charles F. : New shrew of ge nus Sorex 133-134 Beal, F. E. I/. : Food of blue jay viii Food of cow bird ix Bears, American 65-83 Bidens connata var. comosa 35 Bison bison 94 Blarina brevicauda 50, 100 parva 100 Botrychium ternatum var. dissecta 43 Bromus tectorum 42 Calamagrostis nuttalliana 41 Callitriche austini 34 verna 34 Camelina sativa 32 Campanula americana 35 aparinoides 36 Canis nubilus 94 Cardamine hirsuta 31, 32 parviflora 31, 32 pennsylvanica 31 silvatica 31, 32 Carex alopecoidea 40 communis 41 communis var. wheeleri 41 conjuncta 40 emmonsii 40 folliculata 41 glaucodea 40 lagopodioides 40 laxiflora 40 laxiflora var. divaricata 40 laxiflora var. patulifolia 40 lupulina 41 muhlenbergii var. enervis 40 oligocarpa 40 pallesceiis 40 platyphylla 40 polytrichoides 39 prasina 41 pubescens 41 retrocurva 40 shortiana 40 stenolepis 41 (175) 176 The Biological Society of Washington. Page Carex stipata 40 squarrosa 41 torta 40 tribuloides var. reducta 40 umbellata 40 willdenovii 39 Cariacus americauus 25, 27, 28 Cariacus osceola 26-28 Castor canadensis 47, 94 Caucalis anthriscus 35 Cerastium alpinum 105 Cervus canadensis 94 dama americana 25 virginianus 25 Chaerophyllum procumbeiis 34 Charadrius pecuarius 54 tenellus 54 Chesnut, Victor K. : Pfaff's recent investi gations on Rhus poisoning ix Clematis virginiana 31 Clitoria mariaua 33 Cochlearia britannica 104 fenestrata 104 spitzbergensis 104 Commelina communis 39 hirtella 39 virginica 39 Condylura cristata 51, 100 Convolvulus arvensis 37 spithamaeus 3y Cotton mouse (Peromyscus gossypinus).. 119-125 Coville, Frederick V. : Botanical explora tions near Mexican boundary ix Different editions of some Govern ment expedition reports ix Exhibition of hair ball of Trifolium incarnatum x Exhibition of Indian bow made of Taxus brevifolia xii Inflorescence of the Juncacese xii Juncus confusus, a new rush from Rocky Mountain region xi, 127-130 Exhibition of Protococcus nivalis and Nymphaea polysepala xi Notice of Brittou and Brown's illus trated flora of the northern United States and Canada xi Ribes erytlirocarpum, a new currant from vicinity of Crater Lake, Oregon 131-132 Cyperus aristatus 39 flavescens 30 retrofractus 39 Cypripedium pubescens 38 Cystopteris fragilis 43 Page Deer, Florida 25-28 Dentaria cardiophylla 32 heterophylla 32 Desmodium ciliare 33 marilandicum 33 Desmodus, Milk dentition of 113-114 rufus 113-114 Dewey, I,. H. : Sisymbrium altissimuni as a tumble weed x Didelphis virginianus 94 District of Columbia, List of mammals of. 93-101 Dixon, Henry H. : Recent researches on ascent of sap in trees ix Dodecatheon meadia 36 Dorcelaphus 25 Drosera rotundifolia 34 Dryas octopetala 106 Eatonia dudleyi 41 obtusata 41 pennsylvanica 42 Election of officers xii Eleocharis intermedia 39 Eragrostis purshii 42 Erianthus alopecuroides 43 Eriophorum virginicum 39 Euarctos 65, 78 Eupatorium altissimum 35 aromaticum 35 hyssopifolium 35 semiserratum 35 Euphorbia commutata 38 Evermann, Barton W. : Animals from an artesian well at San Marcos, Texas ix Fishes and Fisheries of Indian River, Florida x Story of two salmon ix Evotomys gapperi 49 fuscodorsalis 49 Fedia olitoria 35 Felis concolor 94 domesticus 98 Fe.rnow, B. E. : A pine coppice ix Spiny plants from Arizona xi Fiber zibethicus 49, 96 Fimbristylis capillaris 39 Fir, New, from Arizona xi, 115-118 Fish, Pierre A. : Action of electricity upon nerve cells viii Fucus vesiculosus 106 Fumaria officiiialis . 31 Dall, W. H. : Exhibition of skins of the glacier bear (Ursus emmonsi) ix Danis 65 Gale 2, 6, 51 Galinsoga parviflora 35 Galloway, B. T. : Action of copper in poi soning fungi ix Index. 177 Page Galloway, B. T. : Recent advances in our knowledge of the plant cell x Gaultheria procumbens 36 Gentiana ochroleuca 36 Geranium columbinum 33 Gill, Theo. : Category of family or order in biology xi Characteristics of families Salmon- idce and Thymallidae x Greene, Edward I,.: Distribution of Rham- nus and Ceanothus in America viii Salient features of flora of islands off Southern and Lower California x Habenaria lacera 38 tridentata 38 Helianthemumcanadense 32 Hemicarpha subsquarrosa 39 Herpestis nigrescens 37 Hesperomys cognatus 122 gossypinus 122 Hieracium pilosella 105 Hicks, G. H. : Mildews (Erysiphese) of Michigan xi Holm, Theo. : Alpine flora of Pikes Peak and Grays Peak in Colorado xii Earliest record of Arctic plants.. 103-107 Exhibition of rare books and discus sion of generic name Macounastrum. ... xii Fourth list of additions to flora of Washington, D. C 29-43 Howard, L. O. : Exhibition of photograph of triplets x Hymenopterous parasites of shade- tree insects xi The shade-tree question from an Kalmia glauca Kyllinga pumila Page ... 46 39 insect standpoint ix Hydrocotyle rammculoides 34 Hyonycteris discifera 109 Hypudseus gossypinus 122 Ilex glabra laevigata Ipomsea lacunosa 30,33 33 37 Judd, Sylvester D. : Peculiar eye of an am- phipod crustacean JllllfllS 4-.01lf~ll.SIIS 127-130 dichotomus 128, 129 georgianus 128 greenei 129 secundus 129 tenuis 128, 129 tenuis congestus 127, 129 Juiicus tenuis occideiUalig 129 vaseyi 130 Laminaria 106 Lasionycteris noctivagans 101 Lechea 86 Ledum latifolium 46 Lemmings (Synaptomys), Revision of..... 55-64 Lepidium draba 32 Leptochloa mucronata 30 Lepus americanus 50 sylvaticus 98 ~Lpus sylvaticus alacer 136 sylvaticus bachmani 136 Linum striatum 33 Lobelia puberula 35 Lophanthus nepetoides 37 Lutra hudsonica 99 Lutreola vison 51, 99 Lygodium palmatum 43 Lynx rufus 98 M Macounastrum xii Mammals of District of Columbia 93-101 Mammals from Lake Edward, Quebec 45-52 Mammals, New, from Indian Territory and Missouri 135-138 Mearns, Edgar A. : Mammals of the islands off Southern and Lower California x Melampyrum americanum 37 Mephitis elongata 139, 142-143, 144 hudsonica 140 mephitica 51, 99, 140-141, 144 Mephitis mepliitica scrutator 139, 141-142, 144 Merriam, C. Hart : American weasels viii Big bears of North America ix New fir from Arizona xi, 115-118 Notes on fauna of Oregon xi Preliminary synopsis of American bears 65-83 Revision of Lemmings of genus Synaptomys 55-64 Romerologiis iielsoni, a new genus and species of rabbit from Mex ico 169-174 Sorex veraepacis from Guatemala added to collection of shrews in the U. S. National Museum xi Supplementary notes on tropical American shrews xi Microsorex 50 Microtus chrotorrhinus 49 Microtus foiitigenus 48-49, 52 pennsylvanicus 49, 97 xauthognathus 49 178 The Biological Society of Washington. Page Mictorays 57,60-63 Mictomys dalli 62 innuitus 61 truei 62-63 wraugeli 63 Mikania scandens 35 Miller, Gerrit S., Jr. ; Central American Thyroptera 109-112 Milk dentition of Desmodus 113-114 Subgenera of Voles (Microtinae) viii Monarda punctata 37 Monotropa hypopitys 36 Moore, V. A. : Flagella of motile bacteria with special reference to their value in differentiating species vii Muhlenbergia sobolifera 41 Musdecumanus 94, 96 musculus 95 rattus 94 volans 164 Muscari racemosa 3 s Mustela americana 13, 5 1 boccamela 20 cicognani 18 erminea 7 freiiata 9 fusca J 8 longicauda 7 pennanti 5 1 pusilla 19 richardsoni 13 vulgaris 18, 20 Myosotis laxa 37 N Nasturtium armoracia.. sylvestre Neogale Neosorex Neotoma baileyi pennsylvanica 6 50 135 96 CEiiothera pumila 34 sinuata 34 Ophioglossum vulgatum 43 Orobanche minor 37 Oxyria digyna 107 Palmer, T. S. : Rabbit drives in the West... vi Panicum capillare var. minima 42 comniutatum 42 lanuginosum 43 microcarpon 42 nitidum 43 proliferum 42 pubescens 43 ramulosum 42 Page Panicum scoparium ..................................... 42 sphaerocarpoii ..................................... 42 viscidum .............................................. 42 Papaver dubium ...................................... 31, 106 midicaule .............................. .............. 106 Pastinaca sativa ........................................... 34 Pentstemon Isevigatus ................................. 37 attwateri ....................................... 136, 137 Peromyscus aureolus ............................. 120, 122 Peromyscus be 11 us ................................. 137 canadensis ........................................... 49 Peromyscus cauadeusis ametorum.. 49-50 floridanus ........................................... 122 gossypinus .................................... 119-125 gossypinus mississippiensis ......... 123-124 Peromyscus gossypinus iiigricu- lus ................................................... 124-125 nus palmariiis .................. 124 96, 120, 122 122 121 137 leucopus mearnsi niveiventris rowleyi Phacelia covillei .......................................... 36 parviflora ............................................. 37 Phalaris arundinacea .................................. 42 canariensis .......................................... 42 Phaseolus perennis ...................................... 34 Phenacomys ................................................. 46 Phlox pauiculata .......................................... 36 Physalis philadelphica ................................ 37 pubescens ............................................ 37 Pieters, Adrian J. : Influence of fruit-bear ing on mechanical tissue of twigs ......... viii Pirus coronaria ............................................. 34 Plantago patagonica var. aristata ............... 38 Plover, New, from Madagascar ............... 53~54 Poa flexuosa ................................................. 4 2 sylvestris ............................................. 42 Pogonia ophioglossoides .............................. 38 verticillata ........................................... 38 Pollard, Charles L,ouis : Further remarks on Britton and Brown's illustrated flora ....................................................... xii - Iresine paniculata, an addition to the flora of Washington ........................ xi - Observations on flora of District of Columbia ................................................ viii - Purple-flowered, stemless violets of Atlantic coast ...................................... 85-92 Polygala ....................................................... 86 ambigua .............................................. 33 curtisii var. pychnostachya ................ 33 incarnata ............................................. 33 verticillata ....................................... 30, 33 Polygonum muhlenbergii ........................... 38 bistorta ................................................ 105 tenue ................................................ 30, 3 s viviparum ........................................... 105 Potentilla fragiformis .................................. 106 maculata .... , ...................................... 106 Index. 179 Page Poterium canadense 34 Procyon lotor 99< J 35 Proserpinaca palustris 34 Pteromys hudsonius 162 sabrinus 162 volucella ^64 Putorius agilis 13 brasiliensis frenatus 6, 9-10 erminea 10, 13, 16, 19 frenatus 7 fuscus 1 9 longicauda 6, 7-8 Putorius longicauda spaclix 6, 8, 9 iioveboracensis 6, 13-16, 99 peninsulse 6, 10-13 pusillus 21 richardsoni 6, 13, 16-18 richardsoiii cicognani 6, 18-21, 51 Putorius rixosus 6, 21-22 vison 5 1 vulgaris 18, 19 vulgaris var. americana 19 Pyrola chlorautha 36 elliptica 36 minima 106 rotundifolia 36 secuiida 36 Quercus ilicifolia Rabbit, new genus from Mt. Popocatepetl 169-174 Ranunculus hyperboreus 104 pusillus. 31 pygmseus 104 sulphureus 104 Rhexia mariana 30 Rhyncospora alba 39 cephalantha 30, 39 Ribes acerifolium 132 Kibes erytlirocarpum 131-132 floridum 34 hudsonianum 132 laxiflorum 132 prostratum 132 viscosissimum 132 Richmond, Charles W. : Description of a new plover from the east coast of Mad agascar 53-54 Romerologtis iielsoui, a new rabbit. 169-174 Rosa setigera 34 Rubus cuneifolius 34 hispidus 34 Rumex britannicus 38 verticillatus 38 Rush, New, Juncus confusus 127-130 Salix polaris Sanicula Sapoiiaria vaccaria Saxifraga nivalis oppositifolia rivularis stellaris comosa Scalops aquaticus aquaticus argentatus.. texanus Scalops texauus nercus. Scirpus debilis Page 106 86 33 104 104 104 104 , 100, 138 138 128, 136 138 39 planifolius 39 Sciuropterus sabrinus 146, 162-163 Sciuropterus silus 146, 163-164 volans 94. 146, 164-166 Scinropterus volaus querceti 146, 166 volans virginiaiius 165 volucella 164 volucella hudsonius 162 Sciurus auduboni 149 capistratus 147 carolineusis 95- 135. H6, 153-154, 161 Sciurus carolineusis extimus.. 146, 158-1=59 carolinensis fuliginosus 146, 157-158 carolinensis hypophseus 146, 156-157 carolinensis leucotis 146, 155-156 cinereus 95, 147, 150, 155 hudsonicus 47, 94-95, 146, 159-160 Sciurus hudsouicus loquax 146, 161-162 hiemalis.. 155 labradorius 162 ludovicianus 146, 149, 150 Sciurus ludoviciaiius viciuus.. 146,150-153 macroura 149 magnicaudatus 149 migratorius 155 niger 146, 147-148, 155 occidentalis 149 permsylvanicus 155 rubicaudatus 149 rubrolineatus 159 rufiventer 149 sabrinus 162 sayi 149 subauratus 149 volans 164 volucella 164 vulpinus 147, 150, 155 Scleria pauciflora 39 Scutellaria parvula 37 Sedum 104 Selaginella apus 43 rupestris 43 Sempervivum tectorum 104 Senecio vulgaris 35 Seriocarpus solidagineus 35 Shrew, new, of the genus Sorex J 33-i34 180 The Biological Society of Washington. Page Silene armeria 33 nivea 33 Simpson, Charles T. : Extra-limital Missis sippi Unios vii Sisyrinchium 86 Sitomys megacephalus 120, 123 Skunks (Mephitis) of eastern North Amer ica I39-J44 Smith, Erwin F. : Action of sunlight on Bacillus tracheiphilus x Bacterial disease of potatoes, toma toes, and egg-plants xi Exhibition of L,euconostoc from a sugar vat in Louisiana xi Sorex albibarbis 5 fumeus, 134 hoyi 5 Sorex macrurus i33~ I 34 Sorex personatus 5. 99. T 34 trowbridgii 134 Specularia perfoliata 35 Spirsea aruncus 34 Sporoboius asper 4 1 vaginseflorus 30, 41 Squirrels of eastern North America 145-167 Stanton, T. W. : The genus Remondia x Stejneger, I,eonhard : Use of formalin in the field ix Sternberg, George M.: Malarial parasite and other pathogenic Protozoa xii Synaptomys cooperi 47, 52, 58, 59, 97 Syiiaptomys clalli 62 falling 47-48, 52, 58-59 helaletes 59 lielaletes gossii 60 innuitus 61-62 stonei 58 Synaptomys trnei 62-63 \vra iijjfH i 63 Tamia hudsonia Tamias rubrolineatus striatus Tamias striatus venuatus striatus griseus striatus lysteri. 159 159 95 137-138 137 138 Thalarctos 65, 69 raaritimus 66, 69 Thyroptera, Central American 109-112 albiventer 112 bicolor II2 discifera 109-112 tricolor no, 112 Topping, D. Leroy : Rediscovery of Ficaria ficaria in District of Columbia x Tragopogon porrifolius 35 Tribulus maximus 3 Trifolium medium 33 Page Trillium sessile 38 Tripsacum dactyloides 43 Trisetum palustre 42 Tsuga pattoniana 132 Uniola gracilis 30, 42 Urocyon cinereoargentatus 98 Ursus americanus 66, 78, 79, 94 arctos 78, 80 beringiana 66,69, 7 1 cadaverinus 80 Ursus clalli 69, 71-73 emmonsi 66, 82 Ursus floridanus 65, 66, 81-82 horriseus 66, 69, 75-77 horribilis 65, 69, 74-75 Ursus horribilis alasceiisis 74 horribilis califoriiicus 76 luteolus, 66, 79, 80 Ursus middeiidorffi 65,69-70 piscator 66 richardsoni 66, 69. 77-78 Ursus sitkeusis bs, 69, 73, 74 Utricularia gibba 37 Veratium viride 38 Veronica americana 37 hederifolia 37 serpyllifolia 37 Vesperimus I22 Vespertilio lucifugus 101 subulatus 5, 101 Vesperugo fuscus 101 georgianus 100-101 Viola affinis 92 asarifolia 9 2 ciliata 9. 9 2 congener 9 2 cucullata 88, 92 cucullata cordata 88, 89, 92 cucullata palmata 92 dentata 87, 90, 91, 92 emarginata 9. 9 2 heterophylla 9 2 lanceolata 3 2 obliqua 87, 88, 91, 92 ovata 9, 9 T 9 2 ovata hicksii 9 2 palmata 87,88,91, 92 papilionacea 9 2 pedata 9^ pedata bicolor 9* pedata iuornata 9 1 pedatifida 9 1 primulsefolia 9 2 primulifolia 9 2 sagittata 87, 89, 90, 91 Index. 181 Viola septemloba 87, sororia striata tenella tricolor var. arvensis triloba villosa 88, 89, villosa var. cordifolia Violets, Purple stemless Viverra mephitica Vulpes pennsylvanicus 98 w Waite, M. B. : I^ife history of pear-blight microbe vii Walcott, Charles D. : Preliminary notes on Middle Cambrian medusae ix Washington, Fourth list of additions to flora of. 29-43 Page 91,92 ,.. 92 .. 32 33 33 .. 92 91, 92 ... 92 85-92 ... 140 Page Weasels of eastern North America 1-24 White, David : New forms of Palaeozoic Algae from Central Appalachian Re gion viii Structure and relations of Butho- graptus, Plumulina, and Ptilophyton from the North American Palaeozoic viii Woods, Albert F. : Action of overdose of hydrocyanic acid gas on tomato plants., x Spotting of maple leaves xi Woodwardia angustifolia 43 Xyris flexuosa. Z Zapus hudsonius 97 insignis 50