PROCEEDINGS OF THE Biological Society of Washington VOLUME XXIV 1911 fX.^AS' WASHINGTON PRINTED FOR THE SOCIETY 1912 COMMITTEE ON PUBLICATIONS W. P. HAY, Chairman N. HOLLISTER J. W. GIDLEY H. L. & J. B. McQueen, Inc. 1108 E St. N. W. Washington, d. 0. OFFICERS AND COUNCIL OF THE BIOLOGICAL SOCIETY OF WASHINGTON For 1911 (ELECTED DECEMBER 10, 1910) \Y. P. HAY .1. X. ROSE OFFICERS President • DAVID WHITE Vice-Presidents E. L. GREENE E. W. NELSON Recording Secretary D. E. LANTZ Corresponding Secretary N. HOLLISTER Treasurer J. AY. GIDLEY COUNCIL WILLIAM H. DALE* THEODORE GILL* L. 0. HOWARD* FREDERICK V. COVILLE * F. A. LUCAS* C. HART MERRIAM* FRANK H. KNOWLTON* B. W. EVERMANN* GEORGE M. STERNBERG * L. STEJNEGER* T. S. PALMER* A. D. HOPKINS A. B. BAKER A. K. FISHER VERNON BAILEY PAUL BARTSCH Paul Bartsch STANDING COMMITTEES— 1911 Committee on Communications Vernon Bailey, Chairman Albert Manx N. HOLLISTER Committee on Publications W. P. Hay, Chairman J. W. (ilDLEY Ex-Presidents of the Society. (iii) EX -PRESIDENTS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Theodore X. Gill, 1SS1, 1882 ♦Charles A. White, 1883, 1884 *G. Bkowv Goode, L885, 1886 William II . Dall, 1887, 1888 Lester F. Ward, 1889, 1890 C. Hart Merriam, 1891, 1892 *C. V. Riley, 1893, 1894 Geo. M. Sternberg, 1895, 1896 L. (). Howard, 1897, 1898 Frederick Y. Coville, 1899; 1900 F. A. Lucas, 1901, 1902 B. W. Evermaxn, 1903, 1904 F. H. Know i. ton-, 1905, 1906 L. Stejneger, 1907, 1908 T. S. Palmer, 1909, 1910 * Deceased. (iv) TABLE OF CONTENTS Officers and Committees for 1911 iii Proceedings for 1911 vii A New Musk-deer from Korea, by N. Hollister 1-2 Descriptions of Two New Raccoons, by Gerrit S. Miller, Jr. . . 3-6 The Nomenclature of the Scyphomedusse, by T. D. A. Cocke'rell 7-12 The Generic Name of the Muskrat, by N. Hollister 13-14 New Lizards and a New Toad from the Dutch East Indies, with Notes on Other Species, by Thomas Barbour 15-22 A New Bell-bird from Auckland Island, by Outram Bangs . . . 23-24 Descriptions of Six New Mammals from the Malay Archipelago, by Gerrit S. Miller, Jr 25-28 Two New Birds from the Island of Molokai, by Outram Bangs . 29-30 Diagnoses of Some New Forms of Picidse, by Robert Ridgway . 31-36 General Notes 37-40 The Type Locality of Mellivora abyssinica, by N. Hollister, 37; Discovery of a Fossil Delphi noid Cetacean with Tuberculate Teeth, by F. W. True, 37; Note on the Mus commissarius of Mearns, by Gerrit S. Miller, Jr., 38; On the Correct Name of the Inca Tern, by J. H. Riley, 38; New Names for Two European Voles, by Gerrit S. Miller, Jr., 39; Note on the Scales of the Osteoglossid Fishes, by T. I). A. Cockerel!, 39. A New Fantail from the Chatham Islands, by Outram Bangs . 41-42 Description of a New Dryonastes from China, by J. II. Riley . . 43-44 Notes on Percopsis guttatus Agassiz and Salmo omiscomaycus Walbaum, by William Converse Kendall 45-52 Four New Chinese Mammals, by Gerrit S. Miller, Jr 53-56 Notes on Some Birds from the Island of Grenada, by Thomas Barbour 57-60 A New Pocket Mouse from Wyoming, by Merritt Cary .... 61-62 Some Observations Made on Little River, near Wichita, Kansas, with Reference to the Unionidae, by H. Walton Clark and George H. Gillette 63-68 Notes on the Coloration of Fishes, by Barton A. Bean and Alfred C. Weed 69-76' The Nomenclature of the Hydromedusae, by T. D. A. Cockerell 77-86 A New Crinoid Genus from the Indian Ocean, by Austin H. Clark 87-88 Two New Species of Epimys from Luzon, by X. Hollister . . . 89-90 Breeding Habits of the Viviparous Fishes Gambusia holbrookii and Heterandria formosa, by William P. Seal 91-96 General Notes 97-100 A New Name for Sorex macrurus Batchelder, by Charles Foster Batchelder, 97; Note on the Occurrence of the Whale Shark, Rhinodon typicus, in the Philippine Islands, by (v) vi The Biological Society of Washington. Hugh M. Smith, 97; Owen's Recent Encriniteldentified, by Austin H. Clark, 98; Note on Agarna carinata Schioedte and Meinert, by Harriet Richardson, 98 ; Sciurus poliopus Lyon Changed to Sciurus conipus, by Marcus Ward Lyon, Jr., 98; Capture of Sorex dispar in West Virginia, by Arthur II. Howell, 99. New Names for Two Subspecies <>f Peromi/scus maniculatus (Wagner), by Edgar A. Mearns 101-102 Notes on Genera of Paniceae, by Agnes Chase . . . . 103-160 A New Bat from the Caroline Islands, by Gerrit S. Miller, Jr. . 161-162 A New Blue Grosbeak from California, by Joseph Grinned . . 163^164 A New Mouse-deer from the Rhio Linga Archipelago, by Gerrit S. Miller, Jr 165-166 Descriptions of Four New Treeshrews, by Marcus Ward Lyon, Jr. 167-1 70 Recent Additions to the Fish Fauna of the District of Columbia, by Barton A. Bean and Alfred C. Weed 171-174 The Louisiana Puma, by N. Hollister 175-178 Some Observations on a Photogenic Micro-organism, Pseudo- monas lucifera Molisch, by F. Alex. McDermott 179-184 Description of a New Philippine Flying-squirrel, by X. Hollister 185-186 Descriptions of New American Birds, by Outram Bangs .... 187-190 The Generic Name of the African Buffalo, by N. Hollister . . . 191-194 A New Swift from Palestine, by Outram Bangs 195-196 Some New North American Ixodidse with Notes on Other Species, by F. C. Bishopp 197-208 Some Notes on Fish Scales, by T. D. A. Cockerell 209-214 A New Teal from the Andaman Islands, by J. II. Fleming . . 215-210 Description of a New Subspecies of the Painted Bunting from the Interior of Texas, by Edgar A. Mearns 217-218 A New Race of Chameleons from British East Africa, by Thomas Barbour 219-220 Three New Shrewsof the < renus Cryptotis, by < lerrit S. Miller, Jr. 221-224 General Notes 225-230 The Nomenclature of the Cheetahs, by X. Hollister, 225; On the Correct Name for the Red-winged Blackbird of the North- eastern United States, by Edgar A. Mearns, 226; Note on the Mexican Bats of the Genus Dasypterus, by Gerrit S. Miller, Jr., 227; The Volcano Rabbit of Mount Iztaccihuatl, by Gerrit S. Miller, Jr., 228. A New Roe-deer from China, by Gerrit S. Miller, Jr 231-232 Descriptions of Three Xew Birds fr©m Canada, by J. II. Riley . 233-236 Three New Mammals from Central and South America, by E. A. Goldman 237-240 Two New Shrews from Kashmir, by Gerrit S. Miller, Jr 241-242 The American Species of Fagonia, by Paul C. Standley .... 243-250 Description of a New Melospiza from California, by Harry C. Oberholser 251-252 A New Jumping-mouse from Xew Mexico, by ( ierrit S. Miller, Jr. 253-254 Vol. XXIV, pp. vii-xii January 23, 1912 PPxOCEEDINGS <>K in ]•: BIOLOGICAL SOCIETY OF WASHINGTON PROCEEDINGS. The Society meets in the Assembly Hall of the Cosmos Club on alternate Saturdays at S p. m. January 7, 1911-477th Meeting. President David White in the chair and 51 persons present. The president appointed the following committees for the year 1911: On Communications: Vernon Bailey, Albert Mann and Paid Bart sch. On Publications: W. P. Hay, N. Hollister and J. \V. Gidley. Barton \V. Evermann reported observations of the brown rat's catching flies, at Kokomo, Indiana. C. D. Marsh made comments on a recent report by Aars on biological investigations in Pake Tanganyika, Africa. The following communications were presented: 1 Notes on the Aspens ": I. Tidestrom. ' Some Nesting Habits of the Water Ousel and the Cuckoos " : N. Dearborn. "A Recently Imported Enemy of Alfalfa": F. M. Webster. January 21, 1911 — 478th Meeting. The president in the chair and 175 persons present. The following communication was presented: ' Collecting Big Game with Roosevelt in Africa": Edmund Heller. February 4, 1911— 479th Meeting. The president in the chair and (34 persons present. The following communications were presented: (vii) viii The Biological Society of Washington. "A New Translation of Aristotle's History of Animals": Theodore Gill. " The Insect Fauna of Telegraph Poles ": T. E. Snyder. " The Domestication of Wild Animals": D. E. Lantz. February 18, 1911— 480th Meeting. The president in the chair and 45 persons present. General Wilcox reported having heard a flock of migrating geese on February 14, and also noted that trailing arbutus may be transplanted successfully if taken up with a mass of the soil without loosening the roots. The following communications were presented: "The Mechanics of Cell Activity ": Carl L. Alsberg. " Some Cactus Insects, with special reference to Cochineal ": W. I). Hunter. March 4, 1911 — 481st Meeting. Vice-President W. P. Hay in the chair and 95 persons present. The following communications were presented: " The Desert Country of Northern Lower California " : Dr. W. J. Mixter. "Afoot across the British Columbia Rockies": George Mixter. E. A. Preble exhibited a series of skins of mammals and birds secured by George Mixter and himself in the trip referred to in the second communication. March 18, 1911— 482d Meeting. Vice-President W. P. Hay in the chair and 53 persons present. C. 1). Marsh reviewed briefly a recent paper on the Geo- graphic Distribution of Diatoms, by Hollinger, published in the Jena Zoologische Jahrbiich. The following communications were presented : " Raising Trailing Arbutus from the Seed " : F. V. Coville. Notes on Javan Natural History and the Salt Makers of Tjihara, Java ": William Palmer. Proceeding*. ix April 1, 1911 -483d Meeting President White in the chair and 93 persons present. F. Y. Coville exhibited a pot of flowering trailing arbutus grown from the seed. The following- communications were presented : "A Day in the Galapagos Islands ": W. E. Safford. "Tin- Keys, Corals and Coral Reefs of Florida ": T. Way- land Vaughan. 'e' April 15, 1911— 484th Meeting. The president in the chair and 57 persons present. 0. P. Hay exhibited a fossil skull of a large fresh water snap- ping turtle from Texas. ('has. Warded Stiles reported progress of the International Committee on Zoological Nomenclature. David White exhibited specimens of bituminous coal, show- ing fragments of carboniferous plants. The following communication was presented: ' The House Fly and Typhoid Fever" : L. 0. Howard. The paper was discussed by Dr. C.Gordon Hewitt, Dominion Entomologist, of Ottawa, Canada, and several visiting physicians of the city. May 13, 1911 — 485th Meeting. The president in the chair and 46 persons present. A. S. Hitchcock gave a brief review of a recent number of the North American Flora and of a paper by Nieuwland on the type of Ptliiiriini . The regular programme consisted of Biological Notes from the recent Albatross Expedition to Lower California. Chas. H. Townsend gave a general account of the trip. J. N. Rose dis- cussed the desert vegetation, and Paul Bartsch the fauna of the trip. October 21, 1911— 486th Meeting. President White in the chair and 54 persons present. The following communication was presented : 'The Ice Age and its Extinct Animals " : 0. P. Hay. x The Biological Society of Washington. November 4, 1911— 487th Meeting. The president in the chair and 74 persons present. W. P. Hay exhibited lantern slides of living Amphioxus and of the blue crab, showing the egg masses of the latter. The program consisted of talks on Recent Explorations in Panama. The results of the recent exploring expedition under the auspices of the National Museum were presented by S. E. Meek, E. A. Schwarz, E. A. Goldman and August Busck. Numerous illustrations were used. November 18, 1911— 488th Meeting. The president in the chair and 56 persons present. The following communications were presented: "A Study of Distribution based upon the family Pyramidel- lidae ": Paul Bartsch. 'Peculiar Migration of the Evening Grosbeak": W. W . Cooke. December 2, 1911— 489th Meeting. The [(resident in the chair and 49 persons present. General Wilcox noted having found numerous particles of white quartz in the stomachs of the blue grouse. Barton W. Evermann reported the safe arrival in the United States of ten young fur seals from the Pribilof Islands. L. 0. Howard reported that R. S. Woglum of the Bureau of Entomology had returned from Asia, bringing six cases of living parasites of the white fly, so destructive to citrus fruits in Florida. The following communications were presented : "The Hooting of the Blue Grouse " : E. A. Preble. " Notes on the Fishes of the District of Columbia " : A. C. Weed . ' On another supposed Fruit-bearing Fern-like plant from the American Permian " : David White. December 16, 1911 — 490th Meeting. THIRTY-SECOND ANNUAL MEETING. President White in the chair and 32 persons present. Reports of the secretaries, the treasurer, and chairmen of standing committees were received and placed on file. Proceedings. xi The election of officers for the year 1912 resulted as follows : President : E. W. Nelson. Vice-Presidents: W. P. Hay, J. N. Pose, A. D. Hopkins, Paul Bartsch. Recording Secretary : D. E.'Lantz. Corresponding Secretary : X. Hollister. Treasurer : J. \V. Gidley. Memhers of the Council : Vernon Bailey, A. K. Fisher, A. B. Baker, Hugh M. Smith, William Palmer. Appointment of standing committees was deferred. D. E. Lantz, Recording Secretary. Vol. XXIV, pp 1-2 January 28, 1911 PROCEEDINGS or THE BIOLOGICAL SOCIETY OF WASHINGTON A NEW MUSK-DEER FROM KOREA. BY N. HOLLISTER. TBy permission of the Secretary of the Smithsonian Institution.] ^ During the winter of 1902-3, Dr. William Lord Smith, of Boston, -hunted in southwestern peninsular Korea, and later, with other spoils of the trip, presented to the United States National Museum the skin of an undescribed musk-deer. Doctor Smith writes of his trip:* 'I started from Mok-po (lat. 35°) and went down the coast, partly by boat and partly on foot, to the end of the peninsula. The specimens in question were collected during November and December, 1902, and Jan- uary, 1903. I was never far from the western coast line and was really tiger shooting, bagging two tigers myself, my hunter getting one." Korea appears not to have been included in the range of the musk-deer as given in recent works, though the animal was recorded from there as far back as 1867 by Pere David. f Had Doctor Smith undertaken his trip in summer, he would, doubt- less, have missed the capture of this handsome new species. Moschus parvipes sp. nov. Type from mountains near Mok-po, South Tscholla Province, Korea. Cat. No. 143,184 United States National Museum. Skin only, collected winter of 1902-3, by Dr. William Lord Smith. General characters.— Smaller than Moschus moschiferus or any of the described related forms; legs slender and feet small, with much smaller main and lateral hoofs. Color strikingly ricli and dark, with sharp mark- ings; hair of winter coat much softer and shorter than in M. moschiferus, only about 35 mm. long on shoulders. * Letter to Austin H. Clark, January 27, 1910. tRull. nouv. arch. nnis. d'hist. nat. Paris, III, p. 29, ISC". 1— Proc. Biol. Soc. Wash.. Vol. XXIV. 1911. (1) 2 Hollister — A New Mask-Deer from Korea. Color of type. — Top of head, from nose to back of nape, dark blackish brown; cheeks, spot between eye and ear, and sides of muzzle, mixed with yellowish white; ears outside mixed black and white, inside pure white, tips rimmed with black; a white streak from eye to throat; throat mixed black and white. A snow-white streak, sharply denned, extends from ear, joins the white of throat on side of neck, and extends back- ward to shoulders. Back, from nape to rump, pale russet, much mixed with dark blackish brown, and with many indistinct creamy white spots, which become especially numerous and more distinct on posterior half of body and hips. Sides dark blackish brown; underparts mixed brown and white. Shoulders and forelegs outside, dark brown; inside of legs with a streak of mixed brown and white the entire length to lateral hoofs. Hips and hind legs dark brown with stripe of mixed black and white down front of leg to near hoofs. Tail dark blackish brown above; mixed with white beneath; an indistinct whitish area below tail. Measurements of dry skin. — Length from nose to root of tail, S75 mm. ; hind foot, to end of hoofs, 255; fore foot, to end of hoofs, 170; ear, 77; length of front surface of forward hoofs, 25; of hind hoofs, 27; length of hind lateral hoofs, 20.5. Remarks. — Unfortunately, the skull of this specimen was destroyed for food by Doctor Smith's native assistants before it could be saved, and the cranial characters of the species are therefore unknown. Vol. XXIV, pp. 3-6 January 28, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON DESCRIPTIONS OF TWO NEW RACCOONS. BY GERRIT S. MILLER, Jr. [By permission of the Secretary of the Smithsonian Institution.] The United States National Museum contains specimens of two species of Procyon which appear not to have been hitherto described. Procyon pumilus sp. now Type. — Young adult (rostral sutures open, those of braincase closed; teeth slightly but evidently worn), skin and skull, No. 171,983, U.S. National Museum. Collected at Ancon, Panama, by Dr. Allan H. Jennings. Diagnosis. — Resembling Procyon pygmseus Merriam of Cozumel Island, Yucatan, but size somewhat less diminutive, color less yellowish, teeth relatively less reduced, and nasal hones not distinctly widened posteriorly. Color. — Upper parts and sides a clear grizzled gray without evident yellowish tinge, the underfur drab, darkening ajipreciably at ends of hairs, the longer hairs with black tip and a subterminal annulation of huffy white; the whitish element in excess on neck, shoulders and sides, the black along middle of hack, especially across lumbar region, where it lends to produce a median line; crown like sides but more finely grizzled and with light element a clearer white ; face markings normal, but black band broader than in P. pygmseus and light area above eyes correspondingly reduced ; tail dull hurt' in strong contrast with gray body, the seven black rings well defined, about half as wide as the light interspaces ; underparts drab mixed with the whitish of the longer hairs, but without black except for the broad dark interramial band; fore legs like sides but strongly tinged with drab ; hind legs similar but becoming blackish on inner side above heel. Skull. — The skull differs from that of Procyon pygmseus in its slightly less reduced size, in the form of the audital bulla3 and apparently also in the outline of the nasal bones. Audital bulla? with meatal region distinctly more tubular than in either P. pygmivus or P. heruandezii, the difference apparently due to reduction in the area of inflated portion. Nasal bones 2— Proc. Biol. Soc. Wash., Vol. XXIV, 1911. (3) 4 Miller — Descriptions of Two New Raccoons. scarcely widened posteriorly and showing no tendency toward the broadly rounded form of the terminal region characteristic of P. pygmseus. General form of skull decidedly wider than in P. hernandezii, a peculiarity equally noticeable in the relatively greater zygomatic breadth and in the more inflated anterior portion of braincase. Teeth. — Though showing no special peculiarities of form the teeth are readily distinguishable from those of Procyon hernandezii by their much smaller size, and from those P. pygmseus by their less reduced condition, this character especially noticeable in pm4. Measurements. — External measurements of type ( from skin ) : head and body, 430 ; tail, 235 ; hind foot, 97 (93). Skull of type and of a second specimen of the same age, the measurements of the latter in parenthesis: condylobasal length, 101 (96.6); zygomatic breadth, 63.6(67.0); inter- orbital constriction, 20.4 ( L9.0); breadth of braincase, 46.6 (4.">.0); nasal (median), 25.2 (27.6); mandible, 7.i (72.4). For dental measurements see table. Measurements of the skulls of P. pygmseus, the type, a young male with sutures on upper part of braincase open (basal suture closed), and a female of same age as type of pumilus: condylobasal length, 02.8 (91.8); zygomatic breadth, 58.6 (HO. 2); interorbital constriction, 19.6 (19.8); breadth of braincase, 43.0 (4:1.2); nasal (median), 23.2 (22.2) ; mandible, 69.2 (00.8). Measurements of a skull of P. hernandezii from Colima, Mexico, same age as type of P. pumilus : condylobasal length, 115.4; zygomatic breadth, 70.2; interorbital constriction, 24.2; breadth of braincase, 50.4; nasal (median), Ml. 2 ; mandible, 84.4. Specimens examined. — The type, from Panama, and a skull collected by Lieutenant Couch and labeled Matamoras (No. 1388). Remarks. — In all essential features the Couch specimen agrees so closely with the skull of the type that there can be scarcely any question as to their identity, at least as compared with the large Procyon hernandezii and the small-toothed P. pygmseus. It may, perhaps, be reasonably doubted whether this individual was actually killed at Matamoras, but its presence in Couch's collection appears to be a clear indication that the species ranges far to the north of Panama. Procyon minor sp. nov. Type. — Young male, (permanent dentition in place but basal suture open), skin and skull No. fllsf U. S. National Museum. Collected at Ponite-a-Pitre, Guadeloupe, Lesser Antilles, by L. Guesde. Received from the l'Herminier museum. Diagnosis. — Size and general appearance as in the other small members of the genus (P. maynardi, P. pygmzeus, P. pumilus); skull with posterior extension of palate so widened that its least breadth is decidedly greater than distance from last molar to anterior border of mesopterygoid fossa. Color. — The skin is not in good condition, having apparently been subjected to the action of some kind of preservative fluid. The color Millet — Descriptions of Two Nev. Raccoons. 5 indicates no special peculiarities; back rather uniformly clouded with black, the light element of the grizzle bully white with an evident yellowish tinge <»n neck; face markings normal, the light area above eye well developed; whitisb terminal area on ear large ami conspicuous ; tail not noticeably more yellowish than back, the black rings narrow but well denned; underparts dark wood brown thinly overlaid by the whitish longer hairs; interramial dark area narrow and ill defined. Skull. — The skull is too young to furnish a satisfactory basis for comparison. It differs conspicuously, however, from that of other members of the genus in the unusual breadth of the posterior extension of the palate. The least width of this is 16.4 mm., the distance in median line from level of posterior edge of last molar to anterior border of mesopterygoid fossa (disregarding median spine or notch) is 12 mm. In a specimen of /'. hernandezii <4.(i; zygomatic breadth, 55.4; interorbital constriction, 17.8; breadth of braincase, 44.8; nasal (median), 27.2; mandible, 70.4. For dental measurements see table. Specimen examined. — The type. Remarks.— Although represented by a single rather unsatisfactory specimen this animal shows such marked characters that 1 have no hesitation in regarding it as a distinct species. 6 Millei — Descriptions of Tiro New Raccoons. © o o o -+ ^ CI -f © lC LC lO — IC iO U3 "C © •( UAVOI0 1 X X X X X X X X X JB[OUl .l.i\\t>[ puoa9s © o CI o -r -f © © X »>. X X © X oa 02 © ~- © "M -. 1 -t- o -1- -t X -t- iB[ora '( L1AVOJD) J9A\0[ l^-'Li V, 1 1 Lt X CI LC X -t- X o X o 'C X © LC X © X CI X -t- X X X © 1 — 1 o -• * ~ ^H f~* CI © © X; X o Tl -f -f -f -* to -i- -t- -t- — LC ( »A\O.I.) .IU[OUl -aad I.t.WOl . [Ul.UIJSUrl X X X X X X X X X © CI © -+ © — ~ X CI - - - I- i~ - 1 - X X (Sll U(l.l.l) I|).P,l)>[.'.M|.) -t- 1 1 CI CI » X © CI IT J9AV0[ o3jh[ ■'•'•i'IX J-j CI CI — -1- -t< 1Q CI l~ ' ' ' ' t ■ ■UOSI.HII IO .'MS -f ~ © ■■C CI -t- © o -11[.IX.) A\n.n[](i(ii j.mo-j ; — © -t- iC "* iC -t- iC CI CI CI X CI © © X CC •(UAVoao) IB[OUI .I.i'lilll JS.IIj X X X © © OS x x x x CI CI X X I - ~' X I -' X X © © l~ I- X X I-' © I - M ■ji x" ■(II UIII.ll .I«l< 1III -.U'l .I.tildlt JOU3)SOj X X J I c X X x -- © cc - X X 1 - 1- — X '(suayoio) qjririj>i9ai{o .i-i'Mii .iS.ri![ aaiqi i- © X X ~ 1 CI © C 1 c"i © M X -1- Cl c"i ■SIOSI0UJ JO 9A1S -f •+ -1- CI !D -r w -n[.)xe uoai[jooj .i< t>a X ■< >3 H ^ H cS i- cc • dS 73 !<5 s: C s: o O 99 S CJ ^ n *** 5 a ^ o O Si, s o © © e s © © © © © © © Vol. XXIV, pp. 7-12 January 28, 1911 PROCEEDINGS or in i ; BIOLOGICAL SOCIETY OF WASHINGTON THE NOMENCLATURE OF THE SCYPHOMEDUS/E. COMPILED BY T. D. A. COCKERELL. Having become greatly interested in Dr. A. G. Mayer's great work Medusae of the World (Carnegie Institution, 1910), and being at the same time concerned with matters of nomenclature, I was led to take up the nomenclatural questions involved in the classification of the Medusae, and in consequence to enter upon a correspondence with Dr. Mayer regarding a number of difficult points. Dr. Mayer's work is so well done that it affords material for the investigation of almost any question that may arise regard- ing the Medusae. From the standpoint of correct nomenclature, and in some cases proper grammatical form, fault may often be found with the names of the subfamilies, families and higher groups. The objectionable forms are of course in no case due to Dr. Mayer, but have been received from older authors who were more or less indifferent to the preferred usages of zoologists. There are certainly good reasons for adopting a fairly uniform system of nomenclature for the higher groups in zoology, and in particular the names of families and subfamilies should accord with the International Code. In the following list I have included the Scyphomedusa? of the world down to genera (excepting those which are of quite uncertain status), and in each case have given the carefully ascertained type species of the genus. The whole list has been very kindly gone over by Dr. Mayer (who is substantially its author, although he declines to stand in that position) and approved by him, with two exceptions presently to be men- tioned. In several cases, where I had suggested an innovation with a query, he ran his pencil through the query-mark; but o— Pkoc. Biol. Soc. Wash., Vol. XXIV, 1911. (7) 8 Cockerell — The Nomenclature of the Scyphomedusse. in one instance he deleted a change which 1 had (as I now recognize) mistakenly proposed. The two exceptions concerned the genera Phyllorhiza and Thysanostoma of L. Agassi/,, for which it seemed extremely desirable to take as types well-known and recognizable species, instead of the almost hypothetical forms on which these generic names were based. Unfortunately it is quite impossible under the rules to take as the type of a genus n species not originally included, or even (as in these cases) published at the time when the generic name was first proposed. In the case of Thysanostoma the imperfectly known type species may well represent the genus as defined by Mayer, so there is no disturb- ance of the nomenclature; but the original Phyllorhiza was probably Cephea, and there seems no way of avoiding a new name for Phyllorhiza in the sense of von Lendenfeld and Mayer. It is proposed later to treat the Hydromedusse in the same manner, though there are some exceptional difficulties in the way of this undertaking. Names within square brackets are those of Dr. Mayer's work. Order CARYBDEIDA [Carybdei.be]. Fa m . Car ybdei d.e . Carybdea Per. & Less. marsupialis ( L. ) . Tamoya F. Mull. haplonerria F. Mull. Tripedalia ( lonant. cystophora Conant. ( 'hyropsalmus L. Ag. quadrumanus (F. Mull.). Chirodropus Haeck . gorilla Haeck . Order STAUROMEDUSIDA [Stauromedusse]. Fam. Luceenakiid^e. Subf. Tesseranthin.h. Tesseranthe Haeck. connecteus Haeck. Subf. LUCERNARIIN;E. Depastrum Gosse. cyathiforme (M. Sars). Cocherell — T%e Nomenclature of the Scyphomedusse. 9 Stenoscyphus Kish . inabai (Kish. ). Thaumatoscyphus Kish. distinctus Kish. Lucernaria ( >. F. Mull. quadricornis 0. F. Mull. Kishinonyea Mayer. nagatensis (Oka). Haliclydus Clark. auricula ( Rath ke ) . Halimocyathus Clark. platypus Clark. Capria Antipa. sturdzii Antipa. Subf. Lipkein^e. Lipkea Vogt. ruspoliana Vogt. Order CORONATjE. Fam. Periphylltd^s. Pericolpa Haeck. quadrigata Haeck. Periphylla Steenstr. hyacinthina Steenstr. Peripkyllopsis Vanhoffen . braueri Vanhoffen. Fam . Paraphyllimd.k. Paraphyllina Maas. intermedia Maas. Fam. Ltnuchid^ [Ephyropsidse] . Palephyra Haeck. antiqua Haeck. Nausithoe Koelliker. punctata Koelliker. Linuche Esch. unguiculata (Schw.). Fam. Atolltd/E [Collaspidae]. Atoll a Haeck. wyvillei Haeck. 10 Cockerell — The Nomenclature of the Scyphrnnedusw . Fam. Atorellid.*:. Atorella Vanli. subglobosa Vanli. Order SEM^OSTOMATA [Serrueostomeffi]. Fam. Pelagiid^. Pelagia Per. & Less. noctiluca (Forsk.). Chrysaora Per. & Less. hysoscella (L.). Dactylometra L. Ag. lactea (Esch.). Kuragea Kish. depressa Kish. Sanderia Gotte. malayensis Gotte. Fam. Cyaneid^. Desmonema L. Ag. gaudichaudii (Lesson). < 'yanea Per. & Less. capUlata (L.). Drynonemd Haeck. dalmatina Haeck. Fam. Aurelliid^ [UlmaridaeJ. Subf. Discomedusin^ [Umbrosinae]. Discomedusa Claus. lobata Claus. Parumbrosa Kish. polylobata Kish. Undosa Haeck. undulata Haeck. Diplulmaris Maas. antarctica Maas. Subf. Sthenoniin^e. Sthenonia Esch. albida Esch. Phacellophora Brandt. camtschatica Brandt. Pon 1 1 in Vanli. rufescens Vanh. Cockerell — The Nomenclature of the Scyphomednsse. 11 Subf. Ari(i:r.i.iiNVE. Aurellia Per. & Less. atirita (L. ). Aurosa Haeck. furcata Haeck. Order RHIZOSTOMATA [Rhizostomse]. Fam. Cassiopeiums. Ton izma Haeck. dieuphila (Pers. & Less.). Cassiopea Pers. & Less. andromeda (Forsk. ). Fam. Cepheid^B. ( 'epKeu Per. & Less. cephea (Forsk. ). Cotylorhiza L. Ag. tuberculata (Macri). Polyrhiza L. Ag. vesiculosa (Ehrenb. ). Fam. Catostylih.k. Catostylus L. Ag. mosaicus (Quoy. & Claim.)- Lychnorhiza Haeck. luceriia Haeck. Crambione Maas. mastigophora Maas. Mastigias L. Ag. papua ( Lesson). Pseudorhiza v. Lendenf. avrosa v. Lendenf. Megamastigias n.u. (Phyllorhiza v. Lendenf. not L. Ag. ). punctata (v. Lendenf. ). Versura Haeck. palmata Haeck. Lobonema Mayer. smithii Mayer. Fam. Leptobrachiid^;. Thysanostoma L. Ag. brachyura (Less.). 12 Cockerell — The Nomenclature of the Scyphomedusse. Lor if era Haeck. lorifera (Ehrenb. ). Leptobrach ia Brandt . leptopus Ghana. & Eys. Fam. Rhizostomid^;. Rh izostoma Cuvier . pulmo (Macri). Rhopilema Haeck. rhopalophora Haeck. Eupilema Haeck . scapalare Haeck. Stomolophus L. Ag. meleagris L. Ag. Vol. XXIV, pp. 13-14 January 28, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON THE GENERIC NAME OF THE MUSKRAT. BY N. HOLLTSTER. For over sixty years the propriety of Fiber as the generic name of the muskrat has not been questioned. A careful review of the history and synonymy of the genus brings to light a surprising state of uncertainty as to the validity of this name. Linnaeus, after having previously confounded the American muskrat with his Castor moschatvs (= Desmana moschata) of Asia, in 1766 recognized it as a distinct species, naming it Castor zibethicus. Gmelin, 1788, transferred it to the Linnsean genus Mas, placing it in a starred section (cauda apice com- pressa) with the Mus roi/pus of -Molina. Kerr, 1792, gave this section a subgeneric name, Myocastor.* Link, 1795, erected the genus Ondatra,^ with the same two component species, Ondatra coypus and 0. zibethicus. Lacepede, 179'), listed a genus Ondatra, + with Ondatra zibethicus as the typical specific example. Presumably, he, like the others, considered the coypu congeneric with the muskrat. Cuvier, 1800, named a subgenus Fiber, § based on the ondatra of his "Tableau," 1798, and Illiger, 1811, formally adopted the combination Fiber zibethicus. || Various other generic names were later proposed for the animal, but these have no bearing on the case in question. Lesson seems to have been the last author to con- sider Ondatra seriously, after which the name dropped into the synonymy of Fiber. * Anim. Kingd., p. 225. 1792. iBeytr. Nat.. I, pt, II. p. 7fi. 1795. tTabl. Miimm., p. 9. 1799. fcLecons, d' Anat. Comp., I, tabl. I, 1800. II Prodr. s.n st. Mamm., i>. 88, 1811. 4— Proc. Biol. Soc. Wash.. Vol. XXIV, 1911. (13) 14 Hollistci — The Generic Name of the Muskrat. According to Canon XXVI of the A. 0. U. code (XXII of the revised edition) inasmuch as no genotypes were designated for Myocastor Kerr and Ondatra Link the latter at once became a synonym of Myocastor. Lacepede's Ondatra was of course preoccupied by Ondatra Link. According to the International code, however, it is necessary to fix the types of these various genera to settle their respective standings. The types of Castor and Mus are fixed by tautonymy, and Allen has, by the pro- cess of elimination, fixed the type of Myocastor.* It remains then only to fix the type of Ondatra Link. Every logical argument is in favor of naming Ondatra zibethicus as the type of this genus, and thus reinstate the name Ondatra for the muskrat. The name is distinctively applicable to the animal, and with a slightly broader interpretation of tautonymy than seems at present allowed by the code, zibethicus would automatically be the type, " Ondatra. Buffon hist. not. X. p. I. t. I " appear- ing in the synonymy of Gmelin's Mus zibethicus. It seems unsafe to choose it as such at this late day, however, with the uncertainty as to whether Palmer (Index Generum Mamma- Hum, Appendix, p. 951) has not already unintentionally fixed Mus coypus in that position. Palmer, under Ondatra Lacepede, 1799, states: " Not Ondatra Link, 1795, a synonym of Myo- castor Kerr, 1792 (type Mus coypus Molina), which is a genus of Octodontidse." That a difference of opinion regarding this question might always obtain was at once apparent when it was submitted to several nomenclatorial experts. In order to settle it definitely I, therefore, choose Ondatra coypus as the type of Ondatra Link, 1795. This selection will, at any rate, bring about the same results by both codes; satisfy the few American mammalogists wrho are, unfortunately, still loyal to the A. 0. U. code; and per- petuate the name Fiber, under which the great mass of literature relating to the animal has been published. * Myocastor coypus; Allen, Bull. Amer. Mus. Nat. Hist., VII, ]>. 183, 1895. Vol. XXIV, pp. 15-22 January 28, 1911. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON NEW LIZARDS AND A NEW TOAD FROM THE DUTCH EAST INDIES, WITH NOTES ON OTHER SPECIES. BY THOMAS BARBOUR. The identification of the herpetological material which I col- lected in the East Indies during a part of 1906-7 has just been completed. I find some pointsof interest regarding geographic races in several of the species, and then/ are other specimens which can not be referred to any form described up to the present time. A complete list is now almost ready but it seems advisable to publish the following preliminary descriptions. Leiolepisma pullum sp. now Type. — Xn. 7486, Museum of Comparative Zoology, from Humbolt's Bay, Dutch New Guinea, collected by T. Barbour, January, 1907. Closely related to L. fuscum ( I). & B. ) and L. beccarii I Ptrs. & Doria), but differing from both these species as the appended description shows: Habit lacertiform. Fingers four, toes five; distance between the end of the snout and the fore-limb contained once and one-fourth in the distance between axilla and -ruin ; snout obtuse. Lowereyelid with an undivided transparent disc, smaller than the ear-opening. Nostril pierced in a single nasal; supranasal absent; frontonasal much broader than long, broadly in contact with the rostral, and very narrowly in contact witli the frontal ; latter shield a little shorter than frontoparietal, in contact with the first and second supraoculars ; four supraoculars, seven superciliaries ; fronto- parietal single, followed by a small interparietal, parietals forming a median suture; a pair of nuchals, and a pair of temporals border the parietals; four labials anterior to the subocular. Ear-opening oval, larger than the transparent disc; auricular lobules present, and entering the aural aper- ture from all sides. 42 rows of scales round the body; dorsals and laterals all distinctly tricarinate; preanal scales very slightly enlarged. The hind limb reaches a little beyond the anterior side of the fore-limb. Subdigital lamellae under the fourth toe, 32. Tail almost exactly twice as long as head and body. Color dark mahogany brown above, uniform except for 5— Peoc. Biol. Soc. Wash., Vol. XXIV, 1911. (15) 16 Barboui — Ncir Lizards and New Toad from Dutch East Indies. two crossing, irregular, zig-zag lines of light spots, many of which have black centers. The crossing lines enclose areas which are roughly diamond shaped. The region where these lines occur is that between the fore and hind-limbs. On each side an irregular dark hand, more or less broken up into squarish blotches by white vertical lines. This lateral hand is not edged below with a lighter color, but fades away gradually into the yellow of the belly. There are two light longitudinal stripes on the sides of the tail, with brown above and yellow below. Although the great majority of the wide-ranging skinks do not seem to show any definite variations connected with the locality from which they come, nevertheless two of these lizards do separate into distinct geographic races. The first which I shall discuss is Dasia smaragdinum (Lesson). Lesson, Voy. Coquile Zool. II, 1830, p. 43, pi. ."., tig. 1. Boulenger, Cat. Liz. B. M., Ill, 1887, p. 250. Type locality. — Oualan, Caroline Islands. Lesson figured and named not only this species from Oualan, but also what he called Scincus viridipunctus (pi. 4, tig. 1). A comparison of these figures with large scries from New Guinea, Waigiu, Johi, and New Britain islands, from the Pelew, Marshal, and Caroline islands, and then from Cerain, Ternate, Obi, and Hahnahera, shows that Lesson's figure of smaragdinum was undoubtedly drawn from a Papuan individual. The examples are all the same and unvarying in color for three races based upon definite and unchanging color-phases which do not seem to be correlated with modifications in the squamation. All of the Papuan specimens without exception — I have before me more than 50 — are vivid green over the anterior part of the body; while the posterior may lie either green or may fade to a reddish buff color. This type of coloration, so far as the material in hand shows, never occurs in the Marshal, Caroline, or Pelew islands, nor from the Moluccas; and this skink may he called D. smaragdinum smaragdinum (Lesson), based upon pi. 3, fig. I, the characters being there excellently shown, though the locality of the specimen evidently is incorrect. It may be mentioned here that some of the other lizards recorded or described from the Carolines by Lesson need confirmation. All of the Polynesian specimens at hand, — ten from Ebon, Marshal Islands ; two from Pelew, and two from Ruk in the Caro- line Islands, — are all alike and unvarying in color. This is the form represented by Lesson in pi. 4, fig. 1. It may be known as: D. smarag- dinum viridipunctum (Lesson). The specimens from the Moluccas which I collected are as follows: two from Wahaai, one from Piru, Ceram ; six from Ternate ; six from Obi ; two from Galela, Hahnahera Island, all fall together into another series, not a single specimen varying in the least towards the two races which we have mentioned. These individuals may be taken to represent: Barbow — New Lizards and New Toad from Dutch East Indies. 17 D. smaragdinum moluccarum subsp. nov. Type. — Adult from VVahaai, Ceram ; Museum of Comparative Zoology; No. 74X1, T. Barbour, collector. The coloration of this may be described as that of a light gray ground color, more or less irregularly spotted and blotched with dark brown and black, giving what is commonly called a salt-and-pepper appearance. The writer has seen an enormous number of these lizards alive in very many localities, and his notes, made in the field, show that no variation from these conditions came under observation. I have published the only known record for D. smaragdinum occurring in Formosa (Proc. N. E. Zool. Club, IV, Nov., 1909, p. 05). This was a young example and hence can not be definitely placed in a particular color phase as all the young seem to have a more or less similar color pattern. Colored drawings based upon notes made in the field have been made to illustrate these three races. Cryptoblepharus boutonii (Desjardin). Desjardin Ann. Sci. Nat. (1), XXII, 1831, p. 298. Boulenger, Cat. Liz. B. M., Ill, 1SS7, p. 346. Type locality. — Mauritius. This very widely distributed species is undoubtedly frequently carried about l>y human agency, and therefore it appears colonized in certain localities where no definite local races have arisen. However, local races seem to occur which often may be recognized by a definite color pattern, where, as in I), smaragdinum, structural differences do not seem to have become definitive as yet. Thus we have from the Bonin Islands C. b. nigropunctatus (Hallowell); from Papuasia, C. b. peronii (Cocteau); while Carman has described and provided notes on others (Bull. M. C. Z. LII, 1, p. 12, June, L908). He has proposed the varietal name paschal is for those upon Easter Island, based upon differences which " may at least he said to indicate the process of forming new species by means of heredi- tary tendencies in variation." Lizards from other islands have been discussed as though they represented full species. Thus, Carman con- siders C. nigropunctatus as of equal value to C. boutonii, from which it has undoubtedly been derived; and he speaks of G. poecilopleurus as a full species " likely to have sprung from C. boutonii." These forms are all better considered as geographic races of a single species; and only trinomial names express this condition. From the East Indian area we know, besides peronii, the subspecies furcata, of which Max Weber col- lected and described five specimens which did not vary in color. My examples from New Guinea, Jobi, and Waigiu all fall under peronii; while on the other hand examples from Bali and Lombok differ widely in coloration from others hitherto described, and may, 1 believe, with justice be considered to represent two other undescribed local races, judg- ing by what we know of the distribution of the species as a whole. Thus from Bali we have IS Barbom — New Lizards and New Toad from Dutch East Indie*. Cryptoblepharus boutonii balinensis subsp. nov. Type. — Museum of Comparative Zoology, No. 7480; collected at Bule- leng, Bali Island, by T. Barbour. Scales in L'4 rows. Olive above, with four darker areas on the back, separated by three light olive lines, and bounded below by the olive sides. The median light line forks near the region above the axil, and the two resultant lines continue down the back, bounding a median dark area which is wide and continued to the root of the tail, which latter is olive above and below. Cryptoblepharus boutonii cursor subsp. now Type. — Museum of Comparative Zoology, No. 747i>, from Ampenan, Lombok Island, collected by T. Barbour. The coloration of this form is as follows: Middle of the back light olive bounded on each side by a dark line, each of which meets the other at the root of the tail, and fades away. Outside of these two dark lines are first a silvery white line, beginning just above the eye, and running back to the base of the tail; ami below this again there are wide black lines running down each side from back of the eye to the tail; they continue down the sides of the tail only a short distance, when they break up into series of spots, and become indistinct and disappear. The lower surface is silvery white. The limbs are checked with black and white square spots. The tail is olive above, lighter below, fading to reddish towards the tip. Colored figures are being prepared of these specimens. After the foregoing remarks on Cryptoblepharus had gone into type I received a paper from Dr. Jean Roux on the Reptiles and Amphibians of the Aru and Ke Islands (Abb. Senck. Xaturf. Ges., Vol. :'.:'», L910, pp. L'l 1-1*47, (ils. 13-14). On page 240-241 he discusses at considerable length what he calls Ablephararus boutoni var. keiensis n. var. This is figured; and shows that in the Ke Islands, as in the islands of the Lesser Sunda chain, we have a perfectly recognizable race which may be known as Cryptoblepharus boutonii keiensis (Roux); while on page 218 he shows us that the specimens from the Aru Islands may be referred to the sub- species peronii ( ( !od . ). Among the lizards captured by the naturalists on board the I*. S. F. C. Steamer Albatross were a number of sail-tailed lizards. These Dr. Stej- neger kindly lent me for study. Among them, most fortunately, was a single specimen from Amboina, the type locality of Schlosser's Lacerta amboinensis.* Now, a comparison of these specimens with others which I have from the Moluccas shows that we have been confounding several * Lophura lias Keen recently used for the genus by all authors. This generic name is, however, preoccupied, so that ii becomes necessarj to use Hydrosaurus (Kaup, Isis, 1:828, p. 1H7). In Boulenger's list of synonymous genera, Tstiurus (Cuvier, K. A., 2nd Ed., II, p. -11) appears before Kaup's name. Ii was not, however, proposed until is;;;i Barbowi — New Lizards and. New Toad from Dutch East Indies. 19 perfectly distinct forms under one name. A fully adult specimen from ('cram agrees perfectly with the topotype taken by the Albatross. These two specimens, however, are different from two other adults which 1 shol neat' W'eeda, I lalnialiera ; and from the others, which came from various localities in the southern Philippines. Judging from what Max Weber says (Zool. Erg. Reise Ned.-Ind., I, 1890, p. 167), we may conclude that the examples from Celebes represent another distinct form. We have then in the genus as it stands at present, Hydrosaurus amboinensis (Schlosser), from Amboina and ('cram. H. weberi sp. now, from Halmahera. II. microlophus (Bleeker), from Celebes. H. pustulosus (Eschscholtz) from the Philippines. For the sake of comparison with the description of //. weberi, sp. now, I add here a similar diagnosis of the specimen from Ceram. Unfortunately no examples from Celebes are available for study, though we may expect a full description of //. pustulosus (Esch.) when Dr. Stejneger publishes his " Herpetology of the Philippines." A few notes on the main diagnostic characters of this species are also added, thank- to Dr. Stejneger's courtesy, and with his permission. Hydrosaurus amboinensis (Schlosser). Described from an adultmale, M. ('. Z. No. 7oii4, taken at Pirn, Ceram, February :;, li>o7, by T. Barbour. Head small; snout rather elongate, strongly compressed, with a very inconspicuous longitudinal crest of enlarged scales in the male. Vertical diameter of tympanum contained exactly twice in the distance from the anterior border of the eye to the posterior limit of the nostril. Upper head scales minute, very Strongly keeled, larger in the frontal region than between the eyes; pillar scales small, granular and of very unequal size; a series of enlarged scales on each side, parallel with the lower labials, beginning at the mental and extending backward, decreasing gradually in size until 1 hey disappear almost opposite the posterior border of the eye. Dorsal and nuchal crests continuous, composed of compressed. straight, lanceolate spines on the nuchal region, and of backward-curved, almost falciform, spines on the sacral region; the dorsal crest beginning at the shoulder region and extending backward about half way to the sacrum, is composed of small, inconspicuous scales much less in size than those of the nuchal or sacral areas; dorsal scales small, imbricate, keeled, the keels directed upwards and backwards, intermixed with a very few scattered, enlarged, roundish, short-keeled scales which vary somewhat in size; ventral scales larger than dorsals, subquadrangular, smooth, arranged in regular transverse series ; an area of conspicuously enlarged scales on each side of the chest just in front of the insertion of the fore- limbs. Limbs rather long; the adpressed hind-limb reaches almost to the tip of the snout. Anterior face of fore-limb with greatly enlarged scales in three complete rows. Femoral pores nine on one side, ten on the 20 Barbow — New Lizards and New Toad from Dutch East Indies. other. Tail covered with small quadrangular keeled scales above and on the sides, with larger, more heavily keeled ones inferiorly. Caudal crest high on the male, with denticulated border; the crested portion of the tail being contained three and one-third times in the length of the tail ; which is two and two-thirds times as long ;ls head and body. Olive above, spotted and vermiculated with black. Fold in front of shoulder black. Hydrosaurus weberi sp. nov. Type. — Museum of Comparative Zoology No. 7505; collected atWeeda, Ilalmahera, February, 1907, by T. Barbour. Paratype.— M. C. Z. No. 7506, taken at same time and place. Type an adult male, paratype an adult female. Head deep and massive. Snout rather short, with a prominent, up- raised area covered with greatly enlarged keeled scales instead of a longi- tudinal crest between the nostrils; tympanum small, its vertical diameter being contained almost three times in the distance from the anterior border of the eye to the posterior limit of the nostril, which is round instead of oval as in H. amboinensis. Upper head scales extremely small, strongly keeled, not conspicuously enlarged in the frontal area as com- pared to the region between the eyes; gular scales very small, granular; row of enlarged shields on each side parallel with the lower labials, com- mencing from the mental, and extending to a point below the anterior third of the eye. Dorsal and nuchal crests interrupted in the shoulder region; nuchal crest composed of short, thick, compressed spines; dorsal and sacral crests with uniform, elongate, recurved spines, these being most developed in the mid-dorsal region. Dorsal scales small, imbricate, keeled, the keels directed upwards and backwards, along each side seven groups, each composed of two or three very large, roundish, plate-like scales, shortly keeled; these are not so regularly distributed in the female (paratype). Limbs rather long, the adpressed hind-limb reaching the tympanum. Five or six series of enlarged and strongly keeled scales on the anterior face of the fore-limb; these rows are surrounded by very many partially complete series of scales almost as greatly enlarged, making a strong armor over the whole anterior face of the limb as against the three rows seen on H. amboinensis. Femoral pores 11' on one side, i'.\ on the other. Tail as in //. amboinensis, the crested portion being con- tained three and two-thirds times in the entire length of the tail; which is not quite twice as long as head and body. Color uniform dark brown, lower surface more yellowish. This species may thus be distinguished at once from the type of the genus by its massive, bull-dog head, covered above with almost uniform minute scales; its small tympanum; and its entirely different nuchal ami dorsal crest, its shorter tail and by the greater number of lateral enlarged scales. This species is named in honor of Professor Max Weber, of Amsterdam, to whom we owe so much for his studies of East Indian zoogeography. H. pitstulosus, from the Philippines, has the crest between the nostrils Barboui — New Lizardsand New Toad from Dutch East Indies. 2\ much more developed; a continuous dorsal and nuchal crest of very elongate, falciform spines; a large tympanum, the vertical diameter of which is contained only one and one-third times in the distance from the anterior border of the eye to the nostril; and a head of fundamentally different shape, larger than in //. amboinensis, and its elongate form reminds one of this species. The head is, however, longer in specimens of the same size, and the snout is declivitous, which gives it a more acuminate appearance There are many groups of a considerable number each of enlarged plate-like scales on the side. In this character II. weberi is intermediate between II. amboinensis and H. pustulosus. The type of the new toad which forms the last notice of this papier was taken under an upturned stone on the road between Ampenan and Mataram, on the island of Lombok. It may he known as Bufo cavator sp. now Type: Museum of Comparative Zoology, No. :M70, from Ampenan, Lombok Island. Closely related to Bufo biporcatus Tschudi. Crown with bony ridges, rather feebly developed, viz., a supraorbital and a parietal, forming together a straight, or nearly straight, line, and a short orbital tympanic; snout short, with prominent can thus rostralis ; interorbital space slightly broader than uppereyelid; tympanum distinct, circular, about | diameter of eye; first linger as long as second, toes fully webbed except the fourth toe, which extends far beyond the web. Subarticular tubercles simple; two moderate metatarsal tubercles; tarsus without a fold. The hind limb being carried forward along the body, the tar.so-metatar.sal articula- tion reaches to the nostril. Upper surface with scattered warts which are not spiny; paratoids very prominent, small, almost circular. Brownish above, marbled with light yellowish ; fore and hind limbs cross-barred with, yellowish; beneath yellowish, throat dark brown. This species may be distinguished from Bufo biporcatus at once by its much smaller tympanum, which is almost circular instead of vertical- oval. In //. biporcatus the upper boundary of the tympanum is formed by the orbito-tympanic ridge, which is not the case with this species. The shape of the tympanum is quite different in the two species; and in this new one the cephalic crests are not prolonged as far posteriorly as they are in biporcatus. Vol. XXIV, pp. 23-24 February 24, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A NEW BELL-BIRD FROM AUCKLAND ISLAND. BY OUTRAM BANGS. Among the birds the Museum of Comparative Zoology received from Mr. H. H. Travers was one skin of an adult male Bell- Bird from Auckland Island, which upon comparison proves to be so different from Anthornis m.elanura (Sparrm.) of New Zealand or A. melanocephala Gray of the Chatham Islands that it, as well as its two congeners, must be given specific rather than subspecific rank. Upon inquiry I found that the IT. S. National Museum had one adult male from Auckland Island from the same source as ours, and this together with their entire series of A. melanura and A. melanocephala was kindly sent me by Dr. Chas. VV. Richmond, acting Curator of Birds. The two specimens exactly match and arc not approached by any individual variation among skins from New Zealand or the Chathams. The new bird may be known as— Anthornis incoronata sp. now Type from Auckland Island, No. 40, COS, Museum of Comparative Zoology, adult d\ Collected by H. H. Travers. Characters. — Similar in general to A. melanura (Sparrm.) of New Zealand and A. melanocephala ; tail, 90.5; tarsus, 28; ex- posed culmen, 15. No. 175,193, U. S. Nat. Mus., topotype, adult d\ wing, 94; tail, 92; tarsus, 29; exposed culmen, 14.5. Averages of adult males of A. melanura are, wing, 88; tail, 84; tarsus, 25.5; exposed culmen, 14, while one fully adult male of A. melanocephala affords the following: wing, 100; tail, 96; tarsus, 34; exposed culmen, 15.5. Vol. XXIV, pp. 25-28 February 24, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON DESCRIPTIONS OF SIX NEW MAMMALS FROM THE MALAY ARCHIPELAGO. BY GERRIT S. MILLER, JR. [By permission of the Secretary of the Smithsonian Institution.] The large collection of Malayan mammals in the United States National Museum includes representatives of the follow- ing six forms, none of which appears to have been previously described. Pipistrellus curtatus sp. nov. Type.— Adult female (in alcohol) No. 141,019, U. S. National Museum. Collected on Engano Island, Sumatra, December 1, P)04, by Dr. W. L. Abbott. Original number, 3785. Diagnosis. — Resembling the Javan Pipistrellus imbricatus, but ear larger, muzzle shorter and crown area of anterior lower premolar much reduced. External characters. —Size and general form essentially as in P. imbri- catus, but tail slightly longer and foot less robust. Membranes with no special peculiarities, the wing inserted at base of outer toe. Calcar ter- minating indistinctly, its posterior border with a slight keel that shows no tendency to develop into a distinct lobe like that of P. imbricatus. Ear extending a little beyond nostril when laid forward, its general outline as in P. imbricatus but antitragus better developed. Head shorter and broader than in the Javan species, its general outline when viewed from above nearly circular, with slightly projecting, moderately tumid muzzle. Color after six years immersion in alcohol apparently not so dark as in P. imbricatus. Skull and teeth. — The skull is at once distinguishable from that of Pipis- trellus imbricatus by the decided reduction of rostrum as compared with braincase, a difference equally appreciable in dorsal or palatal view. Palate so much shortened that length in median line is only about equal to width between last molars. Audital bullae as in P. imbricatus. Basi- sphenoid pits rather well defined. Teeth as in Pipistrellus imbricatus except that premolars are decidedly more reduced; anterior upper premolar with crown area barely equal to half that of outer incisor; crown area of anterior lower premolar slightly 7— I'koc. Biol. Soc. Wash., Vol.. XXIV, 1911. (20) 26 Miller — Six New Mammals from the Mala]/ Archipelago. more than"half that of succeeding tooth; length of crowns of the two lower premolars together decidedly less than that of first molar. Measurements. — Type: head and body, 40; tail, :!4; tibia, 14.8; foot, 7.2; forearm, 33.6; thumb, 6.2; third finger, 60; fifth finger, 45; ear from meatus, 13.6. Skull of type: condylobasal length, LI. 6; zygomatic breadth, 9.0; interorbital constriction, 3.8; breadth of braincase, 7.0; depth of braincase, 5.0; mandible, 8.6; maxillary toothrow exclusive of incisors, 4.2; mandibular toothrow exclusive of incisors, 4.4. Specimens examined. — Two, both from Engano. Epimys maerens sp. now Type. — Adult female (teeth so much worn that enamel pattern is 1 >egi li- ning to be obscured ) No. 141,193, V. S. National Museum. Collected at mouth of Mojeia River, Nias Island, Sumatra, March 11, 1905, by J>r. W. L. Abbott. Original number, 40<>2. Diagnosis. — A member of the Epimys rattus group; color as in the dark E. shnalurensis of Simalur Island, Sumatra, but size decidedly less, the hind foot not attaining a length of 40 mm., the maximum condylobasal length of skull among twelve adults, 42 mm. Measurements. — Type: head and body, L78; tail, 168; hind foot, :!4; (32). Average and extremes of the six largest specimens: head and body, 17S (170-185) ; tail, 159(155-168); hind foot, 34.9 (33.6-36); hind toot. without claws, 33.1 (32-34). Skull of type and of largest specimen in the series: condylobasal length, 39.8 and 42.0; zygomatic breadth, 20.2 and 19.8; interorbital constriction, 6.8and 6.6; breadth of braincase, 10.0 and 1(').0; depth of braincase at middle, 11.8 and 12.0; nasal, 14.2 and 15.2; depth of rostrum behind incisors, 8.2 and 8.0; mandible, 25.0 and 26.0; maxillary toothrow, 6.6 and 7.0; mandibular toothrow, 6.2 and 7.0. Specimens examined. — Nineteen, all from Nias Island. Remarks. — The Nias representative of the Epimys rattus group shares the dark color of the other Barussan local forms. The upper parts have more of the burly cast than is usual in E. simalurensis and the belly is a clearer smoky gray, but the difference is very slight. In size there is a marked contrast witli E. simalurensis and E. lugens, in both of which the length of the hind foot usually exceeds 40 nun., and the condylobasal length is usually more than 45 mm. In both size and color the scries is remarkably uniform. Two Nias specimens, however, represent a light col- ored species essentially like Epimys neglectus (Nos. 121,8(12 and 141,184); this animal has probably been artificially introduced. Epimys barussanus sp. now Type. — Adult male (teeth moderately worn), skin andskull No. 141,208, U. S. National Museum. Collected at mouth of Mojeia River, Nias Island, Sumatra, March 10, 1005, by Dr. W. L. Abbott. Original number, 4010. Diagnosis. — A member of the Epimys cremoriventer group resembling the previously known species in general appearance, but size decidedly greater, and skull with anterior portion of rostrum widened. Millci — Six New Mammals from the Malay Archipelago. 27 Color. — The color does not differ essentially from that of Epimys cre- moriventer, though the oehraceous of the hack is duller (less yellow) and the intermingling of blackish hairs is somewhat more conspicuous. Tail more noticeably blackish than in E. cremoriventer, its annulations coarser and more evident. Skull and teeth. — Aside from its conspicuously greater size the skull of Epimys barussanus differs from that of E. cremoriventer in a distinct broadening and deepening of anterior portion of rostrum at once appre- ciable on comparison of specimens. Contrast between width of nasals anteriorly and posteriorly more evident than in the smaller animal. Teeth with no peculiarities of form. Measurements. — Type: head and body, 1.73; tail L>4; hind foot, 34 (32). Skull of type: condylobasal length, 36.6 (33.2);* zygomatic breadth, 18.2 (16.4); interorbital constriction, 6.6 (6.0)'; breadth of hraincase, 15.4 (14.S); depth of hraincase at middle, 11.0 (10.0); nasal, 14.2 (13.2); depth of rostrum behind incisors, S.O (6.8); mandible, 21.8 (20.0); max- illary toothrow, 7.2 (0.0); mandibular toothrow, 7.0 (0.0). Specimen* examined. — Four, all from the same locality as the type. Epimys mengurus sp. nov. Type. — Adult male ( teeth moderately worn ), skin and skull No. 125,021 , r. S. National Museum. Collected at Bukit Menguru, Billiton Island, Sumatra, August 14, 1904, by Dr. W. L. Abbott. Original number, 3581. Diagnosis. — Like Epimys cremoriventer but with relatively longer tail, smaller teeth and more slender skull ; nasals showing a tendency to widen anteriorly as in E. barussanus. Measurements. — Type: head and body, 134; tail, 182; hind foot, 26 (25). Skull of type: condylobasal length, 30.6; zygomatic breadth, 15.0; interorbital constriction, 5.6; breadth of hraincase, 13.4; depth of brain- ease at middle. 9.8; nasal, 12.2; depth of rostrum behind incisors, 7.0; mandible, 17.2; maxillary toothrow, 5.8 ; mandibular toothrow, 5.6. Specimens examined. — The type, from Billiton; also two immature indi- viduals from Banka (Nos. 124,887 and 124,889) probably referable to the same form. Epimys batus sp. nov. '/'///"'• — Adult female (teeth moderately worn), skin and skull No. 121,792, U. S. National Museum. Collected on Pulo Pinie, Batu Islands, Sumatra. March 4, 1903, by Dr. W. L. Abbott. Original number, 2368. Diagnosis. — A member of the Epimys asper group, with the large size and heavy teeth of E. batamanus Lyon, and E. mandus Lyon, but color very dark and rich, the upperparts essentially as in E. mandus but more strongly oehraceous, the underparts clear yellowish ochraceous-buff with only a slight trace of gray in region of chin. Measurements. — Type: head and body, 133; tail, 107; hind foot, 32 (30.4). Two adult females: head and body, 135 and 14(i; tail, 105 and •Measurements in parenthesis are those of mi old male Ep im ys cremoriventer with much worn teeth (No. 104,154, The Dindings). 28 Miller — Six New Mammals from the Malay Archipelago. 105; hind foot, 30(28.6)and :51 (29). Skull of type: condylobasal length, 31.4; zygomatic breadth, 16.0; in terorbital constriction, 6.0; breadth of braincase, 14.2; depth of braincase at middle, 10.0; nasal, 11.2; depth of rostrum behind incisors, li.li; mandible, 10.0; maxillary toothrow, 5.8; mandibular toothrow, 5.8. Specimens examined. — Three, all from Pulo Pinie. Ratufa bicolor major subsp. nov. Type.— Adult female (skin and skull) No. 155,666, V. 8. National Mu- seum. Collected at Tjibodas, Mt. Gede, Java, alt. 4,500 ft., August 15, 1909, by Win. Palmer (Bryant Expedition). Original number, 505. Diagnosis. — Similar to Ratufa bicolor bicolor of the lowlands of eastern Java but larger. Measurements. — Type: head and body, .">(>5; tail, 435; hind foot, 87 (80). Skull of type (teeth moderately worn ) : condylobasal length, 70.4 (66.0);* zygomatic breadth, 45. .4 ( 13.2); mandible, 47.0 (45.2); maxillary toothrow (alveoli), 13.6 (13.8); mandibular toothrow (alveoli), 14.0 (14.2). Specimens examined. — Three, all from the same locality as the type. * Measurements in parenthesis are those of an old female (teeth much worn) from Tjinmnggoe, Java, the largest in a series of twelve skulls of the lowland form. Vol. XXIV, pp. 29-30 February 24, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON TWO NEW BIRDS FROM THE ISLAND OF MOLOKAI. BY OUTRAM BANGS. While identifying and arranging the extensive series of Hawaiian birds in the Museum of Comparative Zoology, I noticed that both the Iiwi and On of Molokai showed characters by which they could be picked out easily, despite the fact that ornithologists dealing with the birds of these islands have not considered them separable. I therefore propose the following names for these two subspecies: Vestiaria coccinea suavis subsp. nov. Type from the island of Molokai, Hawaiian Islands; Museum of Com- parative Zoology No. 15,059, Bangs Coll., adult d\ Collected February 5, 1895, by M. J. Flood. Characters. — Similar to true V. coccinea (Forster) of Hawaii; but larger, bill slightly stronger, and red of general plumage orange-vermilion instead of scarlet-vermilion. Measurements. — Adult d\ type: wing, 85; tail, 60; culmen, 28; tarsus, 25. Adult 9, topotype, No. 15,058: wing, 80; tail, 55; culmen, 25.5; tarsus. 22.* Remarks.— A young male of the new form, also collected by Flood, that was changing from a spotted nestling to the red dress of the adult, when compared with similar skins from Hawaii, shows exactly the same differ- ence in the shade of the vermilion portions of its plumage as do adults. While this particular difference in the shade of vermilion is very striking in the symmetrical, smooth skins of even and regular make, which I have just compared, I must confess that it probably would not be in rough skins such as some European ornithologists appear still content with. At all events I find no mention of it in literature, although Wilson did notice certain dissimilarities in size, the shape of the bill, and in the color of the nestlings, in comparing birds from these two islands. * In a long series of true V. coccinea from Hawaii, the wing in adult males averages 79.6 (78-81), in adult females, 73.75 (72-75). 8— Proc. Biol. Soc. Wash., Vol. XXIV, 1911. (29) 30 Bangs — Two ATew Birds from the Island of Molokai. The island whence the type of V. coccinea, brought back by Captain Cook, really -came can probably never be known; it seems safe, however, to assume that it was Hawaii, as Captain Cook stopped longer there than at any of the other islands. Psittirostra psittacea oppidana subsp. nov. Type from the island of Molokai, Hawaiian Islands; Museum of Com- parative Zoology No. 15,047. Bangs Collection, adult d\ Collected Feb- ruary 8, 1895, byM. J. Flood. Characters. — Similar to true P. psittacea (Gmel.) of Hawaii, and quite as large; larger, therefore, than P. psittacea deppei Rothsch. (now supposed to be extinct) of Oahu. In color somewhat intermediate between the two. The fully adult male compared with fully adult males of true P. psittacea has the upper parts paler, more yellowish, — rather nearer oil-green than olive-green, instead of the reverse; the chest much paler, about gray No. i) of Ridgway, and without darker median streaks to the feathers, instead of gray No. (>, or rarely No. 7, the feathers with darker median streaks; middle of belly with more extended grayish white patch, and sides and flanks dull olive-yellowish instead of oil-green. The adult female is more yellowish, less olivaceous green above and the sides and Hanks are dull olive-yellowish, instead of oil-green. Measurements. — Adult cT , type: wing, 07.5; tail, 54; cnlmen, 16.5; tarsus, '2:\. 5. Adult 9,topotype, Xo. 15,018: wing, 93; tail, 57; culmen, 15.5; tarsus, L'L'.5. Remarks. — Both Wilson and Rothschild seem to have been aware that the Molokai On was not quite the same as that of Hawaii, Wilson saying: " Examples from Molokai do not present any definite points of difference, though perhaps they are somewhat duller beneath"; while Rothschild specially mentions a youngish male in his collection from that island which he considered to be about intermediate between P. jisittacea and P. deppei. The three skins — two adult males and one adult female — in our collec- tion from Molokai are all very different from any in a large series, includ- ing many adults and young of various ages from Hawaii. Vol. XXIV, pp. 31-36 February 24, 1911 PROCEEDINGS of Tin: BIOLOGICAL SOCIETY OF WASHINGTON DIAGNOSES OF SOME NEW FORMS OF PICIDjEj BY ROBERT RIDGWAY. [By permission of the Secretary of the Smithsonian Institution.] Colaptes auratus borealis subsp. nov. Type from Nulato, lower Yukon R., Alaska. Xo. 49,922, U. S. Nat. Mus. Adult male. June 23, 1867. W. H. Dall. Similar to C. a. auratus and C. a. luteus in coloration, but decidedly larger than the latter, much larger than the former. Adult male (type): Wing, 170; tail, 113; exposed culmen, 36.5; tarsus, 30; outer anterior toe, i':;. The case of Colaptes auratus is precisely parallel to those of Dryobates villosus and D. pubescens, all three species gradually increasing in size from the extreme southern to the extreme northern parts of their range, without material change in coloration. Evidently the three cases require identical treatment; and since three forms seem to best express the conditions in the two species of Dryobates (east of the Rocky Mts.) we can not have less than three forms of Colaptes without being conspicuously inconsistent. While ( 'olaptes auratus luteus Bangs includes both the medium-sized speci- mens from the more northern portions of the United States and the very large ones from the far North, the type is an example of the resident bird of .Massachusetts, though an exceptionally large one. Under the circum- stances, it seems best to restrict the name luteus to the mid-region form, and give a new name to tbe large northern bird. By so doing, we have, as in the cases of Dryobates villosus and D. pubescens, a small Lower Austral or Austrori parian form, a medium-sized form of the Upper Austral and Tran- sition life-zones, and a large form in the Canadian and Hudsonian zones. Tbe case of Phlaiotomus pileatus, in its variations east of the Rocky .Mountains, is nearly the same, but different in this respect: The species does not extend so far northward, and the lines separating (somewhat arbitrarily, as in the other species) the ranges of the two more southern forms are shifted farther southward; the extreme southern form, instead of inhabiting the Lower Austral zone as a whole, being restricted to middle and southern Florida. 9— Proc. Biol. Soc. Wash., Vol. XXIV, 1911. (31) 32 Ridgway — Diagnoses of Some New Forms of Picidse. Colaptes chrysoides mearnsi subsp. nov. Type from Quitovaquito, Arizona. No. 132,871, U.S. Nat. Mas. Adult male. February 2, 1894. Dr. Edgar A. Mearns, U. S. A. Similar to C. c. chrysoides but decidedly larger and paler, with pileum more strongly cinnamomeous, black bars on back, etc., narrower, spots on outer web of primaries more conspicuous (primary coverts also sometimes spotted), gray of throat, etc., lighter, and spots on under parts usually smaller. Adult male (type): Wing, 146; tail, !)6; exposed cuhnen, 35.5; tarsus, 27; outer anterior toe, 2:!. There are apparently three easily characterized geographic forms of this species, as follows : Colaptes chrysoides chrysoides ( Malherbe). Southern Lower California. Colaptes chrysoides brunnescens Anthony. Middle Lower California. Colaptes chrysoides mearnsi Ridgway. Arizona, extreme southeastern California and northern Lower California, and southward to southern Sonora. Centurus chrysogenys flavinuchus subsp. nov. Type from Acapulco, Guerrero, southwestern .Mexico. No. 154,935, U.S. Nat. Mus. (Biological Survey Coll.). Adult male. January 13, 1895, Nelson and Goldman. Similar to C. c. chrysogenys (Vigors)* but adult male with nape bright orange-yellow (instead of orange-red), strongly contrasted with red of crown; adult female with nape yellow, instead of orange or orange-red. Adult male (type): Wing, 124; tail, 75; exposed culmen, 26; tarsus, 21.5; outer anterior toe, 20. Centurus uropygialis brewsteri subsp. nov. Type from Santiago, southern Lower California. No. 151,827, U. S. Nat. Mus. Adult male. November 22, 1887, M. Abbott Frazar. Similar to C. u. uropygialis, of Arizona, etc., but smaller, with rela- tively ( often absolutely ) larger bill; bars on back, etc., averaging decidedly narrower, black bars on lower rump and upper tail-coverts narrower and more numerous, and white bars on lateral rectrices, as well as black ones on inner web of middle rectrices, narrower. Adult male (type) : Wing, 129; tail, 79.5; exposed culmen, 30; tarsus, 23; outer anterior toe, 18. Chloronerpes rubiginosus trinitatis subsp. nov. Type from Princestown, Trinidad. No. 59,416, Am. Mus. Nat. Hist. Adidt male. March .">, 1893, Frank M. Chapman. Smaller and more richly colored than C. r. rubiginosus, from the arid coast district of Venezuela, the back, etc., brighter, more tawny, olive, chest more brownish dusky with narrower bars of yellowish and more or less strongly suffused with dull orange or tawny. Adult male (type): Wing, 104 mm.; tail, 61.5; culmen, 2:5.5; tarsus, 20.5. * The name Centurus elegans (Picas elegans Swainson, 1S27) can not be used for this species, being preoccupied by Picas elegans Midler, 1776. Centurus chrysogenys chrysogenys is confined to the State of Sinaloa and Territory Of Tepic. Ridgway — Diagnoses of Some New Forms of Picidas. 33 Chloronerpes rubiginosus tobagensis subsp. now Type from Tobago. No. 74,968, U. S. Nat. Mus. Adult male. May. F. A. Ober. Similar in coloration to C. r. trinitatii hut decidedly larger. Adult male (type): Wing, 115; tail, 66; culmen, 23.5; tarsus, 20. (Average measurements of adult males of the two forms are: — C. /■■ trinitatis (5 specimens): Wing, 105.4; tail, 60.7; culmen, 22.9; tarsus, 20. C. r. tobagensis (7 specimens): Winy, 111; tail, 65.5; culmen, 24.1; tarsus, 17.6. ) Chloronerpes rubiginosus meridensis subsp. now Type from near Merida, Venezuela. No. 100,734, Am. Mus. Nat. Hist. Adult male. S. Briceno Gabaldon. In coloration nearest C. r. uropygialis from Costa Rica and Panama, but posterior under parts more strongly barred, and bill more slender. Adult male (type): Wing, 121 mm.; tail, 7:\; culmen, 22.5; tarsus, 21. (Another adult male from Punta, near Merida, measures as follows: Wing, 123.5; tail, 74; culmen, 23; tarsus, 20.5.) Veniliornis kirkii darienensis subsp. nov. Type from El Real, Darien, eastern Panama. No. 150,795, II. S. Nat. Mus. Adult male. September 3, 1889, Heyde and Lux. Similar to V. k. continentalis Hellmayr, of Venezuela, but wing-coverts without yellowish spots or streaks, and under parts much more narrowly barred, the pale bars pale brownish buff instead of buffy white. Similar also to V. k. cecil ii (Malherbe), of central Colombia, but back, etc., brighter, more tawny, and under parts more broadly barred with pale brownish buff (instead of dull whitish), and size slightly less. Adult male (type): Wing, . 1908. § Rev. et Mag. de Zool., 2nd ser.. IV. i>. 151). March, 1852. || Atti Sesta Riun, Sci. Ital., Torino. 1844, p. 350. 1845. Vol. XXIV, pp. 41-42 February 24, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A NEW FANTAIL FROM THE CHATHAM ISLANDS. BY OUTRAM BANGS. Several years ago the Museum of Comparative Zoology received from Mr. H. H. Travers a collection made in Snares, Auckland, Chatham and Antipodes islands. It contained one skin of a Rhipidura from the Chathams (probably Mangare, though the label says merely Chatham Islands) that differs in several respects from New Zealand specimens of Rhipidura flab ill if era (GmeL). As this island form appears to be unnamed it may be known as — Rhipidura flabillifera penitus subsp. nov. Type from the Chatham Islands, No. 39,984, Museum of Comparative Zoology, adult Q. Collected by II. II. Travers.* Characters. — Similar in general to true R. flabillifera (Gmel.) of New Zealand, but tail longer and larger — each rectrice wider; two central rectrices grayish or hoary black instead of brownish black and with a white stripe 3 mm. wide fringing the inner web for the apical half of the feather and joining the white tip which is more extensive. Measurements. — Type adult 9 : wing, 76; tail, 99; culmen, 8.5; tarsus, 19.5. In a series of six skins of true R. flabillifera from New Zealand of which only two — a male and a female— had the sex determined by the collector, the wing varies in length from 70 to 75, and the tail from 1)1 to !»4. * Unfortunately the skins we hart of Travers had no data other than locality and sex. 11— Proc. Biol. Soc. Wash., Vol. XXIV, 1911. (41) Vol. XXIV, pp. 43-44 February 24, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON DESCRIPTION OF A NEW DRYONASTES FROM CHINA. BY J. H. RILEV. [By permission of the Secretary of the Smithsonian Institution.] Iii a small lot of birds collected by Arthur de C. Sowerby in the provinces of Shansi and Shensi, China, there is a single specimen resembling Dryonastes perspicillatus but paler through- out and in my opinion represents a recognizable geographical race. It may be known from the following description: Dryonastes perspicillatus shensiensis subsp. now Type, No. 212,0(17, Collection of the U. S. National Museum. Adult male. Fifteen miles south of Si-an-fu, Shensi, China, 1,500 feet altitude, February 28, 1909. Collected by Arthur de C. Sowerby. Subsperific character*. — Similar to Dryonastes p. perspicillatus G-melin of south and south-east China, but much paler both above and below and with the feathers of the throat and neck with a mere indication of a dark shaft-streak. Description. — Forehead, supra-orbital, lores, and auriculars, black, forming a mask; crown, cervix, anil occiput smoke gray, deepening into brownish on the upper back, all the feathers with very narrow edgings and indistinct shaft-streaks of hair brown; back and rump broccoli brown; upper tail-coverts Isabella color; throat, jugulum, and sides of neck, smoke gray, the feathers of the throat and jugulum with hardly perceptihle dusky shaft-streaks; breast, abdomen, and wing-lining, pink- ish buff; thighs buff; crissum ochraceous ; carpo-metacarpal bordered with clove brown; wings externally color of the back; the primaries and secondaries blackish on the inner well and then edged with huffy towards the base of the feathers; tail wood brown, all the feathers broadly tipped with black, except the middle pair, the black increasing towards the outer pair where it occupies about half the feather and extends some distance up the outer wel). Wing, 134; tail, loo; culmen, 25.5; tarsus, 41.5; middle toe, 25.5 mm. 12— PROC. BIOL. Sue. Wash., Vol. XXIV. 1911. (43) 44 Riley — Description of a New Dryona&es from China. Remarks. — Besides the paler colors, Dryonastes perspicillatus shensiensis has the feathers of the crown and cervix less strongly edged with hair brown ; the middle tail feathers are wood brown while in D. p. perspicil- latus they are mars brown. Shensi seems to be out of the given range of D. p. perspicillatus and in a different faunal area. Among the birds from the type locality of the above described race there was a single specimen of Nannus troglodytes idius (Richmond), which extends the range of this bird considerably to the west. Vol. XXIV, pp. 45-52 February 24, 1911 PROCEEDINGS OF THE rc>v BIOLOGICAL SOCIETY OF WASHINGTON NOTES ON PERCOPSIS GUTTATUS AGASSIZ AND SALMO OMISCOMA YCUS WALBAUM . BY WILLIAM CONVERSE KENDALL. | Published by permission of George M. Bowers, U. S. Commissioner of Fish and Fisheries.] At the meeting of the Boston Society of Natural History of November, 18 18, Professor Louis Agassiz gave an account of two fishes obtained by him at Lake Superior, which he regarded as the types of new genera. Regarding one of these Agassiz said (Proc. Bost. Soc. Nat. Hist. IV, 1851, pp. 80-81): "The first of these two new species is a small fish, five or six inches long, in general shape resembling a Leuciscus. It has the adi- pose fin of the Salmonidse, but not the jaws of that family ; these strongly resemble those of the Percoids. In its scales which are serrated on their margins, it also resembles the Percoids. Its characters are sufficiently peculiar to justify the establishment of a new family from this single species. Fossil species with similar characters are found in the Cretaceous formation. This is the second of the old ' fashioned ' fishes, so to speak, corres- ponding in their structure to fossil species, which has been observed in this country. The other, the Lepisosteus, is the only living representative of a large family of fossil species. ' The existence of these two species has undoubtedly refer- ence to the fact, that America is the oldest extensive continent which has been upheaved above the level of the sea." He called the genus " Percopsis on account of its resemblance to the Percoids." In his Lake Superior, 1850, Agassiz gives a fuller description of the genus and states that his " new genus, Percopsis, is just intermediate between Ctenoids and Cycloids: and that is what an ichthyologist, at present, would scarcely think possible — a 13— Proc. Biol. Soc. Wash., Vol. XXIV, 1911. (45) 46 Kendall — Notes on Percopsis guttatus and Salmo omiscomaycus. true intermediate type between Pereoicls and Salmonidse, and should be considered as the type of a distinct family, under the name of Percopsides." Following which is a complete description of the species which he names Percopm guttatus, remarking that the species was found in great abundance at the " Sault St. Mary, at Michipictin and at Fort William." It seems that the Rev. Zadock Thompson, author of " History of Vermont, Natural, Civil and Statistical," had discovered the same fish and named it, without publication however, sometime before Agassiz announced his discovery. Concerning which Mr. Thompson has the following in the second edition of his History, 1853, Appendix, p. 33. "The first knowledge I had of tins fish was in the summer of 1841, when I found a specimen of it, 5 inches long, which was dead, and had been drifted up by the waves on the lake shore in Burlington. On examining it I found it to possess the adipose and abdominal fins of the trout, but in its teeth, gill - covers and particularly in its hard, serrated scales, to bear con- siderable resemblance to the perch family. After searching all the books within my reach without finding it described I con- cluded that it might be new, both in genus and species, and accordingly, in allusion to the above-mentioned properties, I described it in my journal under the provisional generic name of Salmoperca. A notice of this fish was omitted in my History of Vermont, published in 1842, because I had then only one specimen, and upon that one, with my little experience, I did not think it prudent to found a new genus and species. When Professor Agassiz was at Burlington in 1847 I submitted the above-mentioned specimen to his inspection, having at that time obtained no others. At first sight he thought it might be a young fish of the salmon family, but upon further examination he said it was not a salmon, nor any other fish with which he was acquainted. ' ' During the summer of 1847 I found three other specimens of this fish, dead, on the lake shore. One of these I took with me to Boston in September to the meeting of the Association of American Geologists and Naturalists, and put it into the hands of my friend, D. H. Storer, M. D., with a request that he would ascertain what it was and let me know. " In May, 1849, I obtained from Winooski River a number Kendall — Noteson Percopsis gvlttatus and Salmo omiscormycus. 4Tt of living specimens, which I kept alive for some time; and, observing the great translucency of the living fish when held up toward the light I gave it the specific name of pellucida, having previously called it, in my journal, eoceta, from its wing-like pectoral fins. " About this time I noticed, in the proceedings of the Boston Society of Natural History, that Professor Agassiz had laid before the Society an account of a new genus of fishes dis- covered by him in Lake Superior, which he proposed to call Percopsis. Suspecting, from the brief description given of it, that it was identical with my Salmoperca, I wrote to Dr. Storer and inquired of him if the specimens from Lake Superior pre- sented to the Society by Professor Agassiz were like the one I put into his hands in 1847. He wrote me that he could not say — that the specimen went out of his hands soon after he received it and he had not seen it since. " In Professor Agassiz's Lake Superior, page 248, I find an account of his genus Percopsis and his species P. guttatas, and I have no doubt that it is identical with my Salmoperca pellucida. Still, I have thought it best to let it remain, in this Appendix under the name I had given." Since Agassiz first named this species the name has withstood the vicissitudes that taxonomy has so frequently meted to many other systematic names of fishes as well as other animals. It is perhaps unfortunate that the nomenclatural rules of priority of publication are so hard and fast that Thompson's name for this fish could not have been retained, in justice to Mr. Thompson. It is surely unfortunate that those same rules require a name however barbarous and unmeaning to be accepted if it has proper qualifications of form and priority of publication . Accordingly it comes about that another specific termination antedating both Agassiz and Thompson, for the genus Percopsis, must be substituted for that of guttata*. In 1784 was published the Introduction to Arctic Zoology by Thomas Pennant. In the list of Fishes of Hudson's Bay appears on page CXCII : ' The Omisco Mai/cus is a new species of trout taken in May in Albany River, not exceeding four inches and a half long. It has five branchiostegous rays: first dorsal fin has eleven rays, ventral eight, anal seven, pectoral thirteen: tail forked: in the jaws are minute teeth: back, as 48 Kendall — Notes on Percopsis guttatus a nd Salmo omiscomaycuS . low as the lateral line, is of a pale color, marked with two lon- gitudinal rows of black stelliform spots: below the lateral line the color silvery: the belly white." Again on page 65, Walbaum's Petri Artedi Genera Pisciuni, Additamentum, 1792, is " 18. Salmo, Omisco Maycus, radiis 5 membranaebrancbiostegae; Cauda bifurca. Pennant. Arct. Zool. mi rod. p. 192. D. 11. P. 13. V. 8. A. 7. C. 'Corpus fere spitbameum. Dentes in maxillis minuti. Color in dorso usque ad lineam lateralem pallidus, duplici serie macularum nigrorum, stellatarum notatus; infra lineam later- alem argenteus; sub abdomine albus. Habitat in sinu Hud- sonis." According to the rules of nomenclature ichthyologists accept all of Walbaum's names as binomial and notwithstanding its form this name falls in the same category as the others. But it serins to have been disregarded or overlooked by the systematic writers prior to Jordan and Evermann (Fishes of North and •Middle America, part 1, page 487, 1896), who place it in the synonymy of Salmo salar, probably assuming that it was a young salmon. The description of Salmo Omisco Maycus and an examination of specimens of Percopsis guttatus conclusively show that they are specifically identical, and must be designated as Percopsis omiscornaycus (Walbaum), however regretable this cir- cumstance may be. Comparison of essential characters in Pennant's (Walbaum's) description of the Omisco Maycus with the same characters in Agassiz's description of Percopsis guttatus: 1. Maximum length . . 2. No. of branchiostegals :!. No. of first dorsal rays 4. No. of anal ray.s . . . 5. No. of ventral rays 6. Shape of caudal fin 7. No. of pectoral rays . S. Teeth 9. Color Pennant ( Walbaum) \% inches 5 11 7 8 " forked" 13 " minute " 2 rows of black spots on side of hack. Agassiz 6 inches (J 2, Hi 1, 7 s "furcated" 12 "excessively tine" Sometimes in 2 rows Kendall — Notes on Percopsis guttatus and Salmo omiscomaycus. M> Notes applying to the above comparison. 1. Jordan and Evermann give the maximum length of Per- copsis [villain* as ('» inches. The species will not average over 4 or 4^ inches in specimens at hand. 2. The branchiostegals are not uniformly 6. In several specimens examined by the writer there are only 5 on a side. 3. There are only eleven developed dorsal rays, the spines being small, inconspicuous and adherent to the first ray. 4. Anal uniformly with 7 developed rays with inconspicuous rudimentary spine adherent to the first ray. 5. Ventral rays are uniformly 8. 6. Caudal fin always strongly forked. 7. Pectoral varies in number of rays. 8. The teeth of even a 4i inch trout or salmon would hardly be called " minute." 9. Seldom more than two rows of spots on the side of the back and one row along the dorsal median line. The characters conspicuously distinguishing the Omisco May- cus from any Salmonoid are: No Salmonoid has so few branchiostegal rays. Coregonus quadrilateralis according to Richardson, sometimes with 8 rays, comes nearest to it, but the number of anal rays in the white- fish far exceed the number in Percopsis. No Salmonoid other than some white fish has so few ventral rays, and as previously mentioned the anal of the whitefish exceeds the Percopsis in number of rays. Besides no whitefish * has teeth on its jaws. The teeth of the salmon or trout are comparatively large. Salmo salar is not recorded from Hudson Bay. Lowe men- tions it as occurring in Ungava Bay, the most westward point of its range. Cridivomer namaycush and perhaps one or more species of Salvelinus occur in the region but the characters above mentioned preclude these forms. In his Families of Fishes, 1872 (Smithsonian Miscl. Collec- tions), Gill included the Family Percopsidae in the order Isos- pondyli, to which, from the structure of the species, it would seem that it was more closely related than to the Acanthopteri, owing to the fact that the so called spines are very weak and similar to the usual simple or rudimentary rays of the Cyprin- 50 Kendall — Notes on Percopsis guttatus and. Salmo omiscomaycus. idae and the majority of the characters are of that order. But- later another fish of the family was discovered by Eigenmann, in the Umatilla River, Oregon, still more decidedly percoid in its structure, justifying the provisional disposition of the family in the Acanthopteri by Jordan and Evermann. This fish was described by Eigenmann and Eigenmann as a new genus and new species under the name of Columbia trans- montana, in Science, Oct. 21, 1892, p. 233. In apparent recognition of the generic name bestowed upon the fish by Thompson, Jordan and Evermann have established the suborder Salmopercae for these fishes, of which they have the following to say: We place provisionally as a suborder of the Acanthopteri, a singular group of archaic fishes, relics of some earlier fauna, and apparently derived directly from the extinct transitional forms through which Haplomi and Acanthopteri have descended from allies of the Isospondyli. The group shows the remarkable combination of true fin spines, ctenoid scales, and a percoid mouth, with the adipose fin, abdominal ventrals, and naked head of the Isospondyli. The relations of the Percopsidse with such archaic spiny rayed fishes as Aphredoderus and Elassoma are certainly not remote and the close resemblance of the head of Percopsis to that of Gymnocephaliis (Acerina) may be more than accidental." Accordingly the classification of this little group is as fol- lows: Order. Acanthopteri. Suborder. Salmopercae. Family. Pereopsidae. ,-, f Percopsis, omiscomaycus. (jrenera. i r, -, , • ' [ Columbia, transmontana. Key to the Genera. [After Jordan and Evermann.] a. Dorsal fin with 2 feeble, slender spines or simple rays; anal with 1 slender spine ; scales most strongly ctenoid on cau- dal peduncle; posterior margin of preopercle entire or with feeble crenulations; lateral line developed, the tubes small; form slender; the body translucent. Percopsis. aa. Dorsal and anal each with two very strong spines; ven- tral spine evident; scales most strongly ctenoid on anterior part of body ; posterior margin of preopercle with a few short but Kendall — Noteson Percopsis guttatus and Sal hid omiscomaycus. 51 strong spines ; lateral line imperfect, the tubes more or less obsolete; form robust, the substance more or less opaque. Colli illbii I . Percopsis omiscomaycus. (Walbaum). Omisco Maycus, Pennant Arctic. Zool. In trod., p. CXCII, 1784; (ibid, Second Ed. p. CCXCVIII, 1792). Salmo Ormisco Maycus, Walbaum, Artedi Gen. Pise., p. 65, 1792; (after Pennant). Percopsis guttatus, Agassiz, Lake Superior, p. 286, pi. I, fig. 1 and 2, 1850. Salmoperca pellucida, Thompson, Hist. Vermont, second ed., appendix, p. 33, with figure, 1853. Percopsis hainmondi, Gill, Proc. Ac. Sci., Phila., 1864, p. 151. "Head 3f ; depth about 4£ ; D. II, 9 ; A. I, 7; V. I, 8; scales 50, head slender and conical ; mouth small, subinferior, maxillary not nearly reaching front of orbit. Caudal peduncle long and slender. Pale olivaceous, a silvery stripe along the lateral line, becoming obsolete forward; upper parts with obscure round dusky spots made of dark points, peritoneum silvery. Length 6 inches. Spawns in spring. Delaware River (Abbott) to Ohio River (Sloan; Gilbert); Kansas and north- ward ; very abundant in the Great Lakes ; in all streams trib- utary to Hudson Bay, Red River of the North, and found by Eigenmann in the Saskatchewan as far as Medicine Hat ; rare in streams south of Lake Erie, although occasionally taken throughout the upper Mississippi Valley." [Jordan and Ever- mann.] Specimens from Hudson Bay region in the general locality of Albany River are reported from Moose Factory, at the mouth of Moose River, emptying into James Bay or southern extrem- ity of Hudson Bay, which is the next considerable river shown on map, south of Albany River. Specimens also from Nelson River a long distance north of Albany River and flowing directly into Hudson Bay. [Bean in Proc. U. S. Nat. Mus., 1881, pp. 127 and 128.] Vol. XXIV, pp. 53-56 February 24, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON FOUR NEW CHINESE MAMMALS. BY GERRIT S. MILLER, JR. [By permission of the Secretary of the Smithsonian Institution. Among some Chinese mammals collected by Mr. Arthur deC. Sowerby and presented to the Tinted States National Museum are representatives of four that have not hitherto been described. They were taken in the provinces of Kansu and Shansi, north- west China. Eptesicus serotinus pallens subsp. nov. Type. — Adult male (skin and skull ) No. [55, L56,U. 8. National Museum. Collected at Ch'eng-yuamhsien, 70 miles west of Ch'ing-yang-fu, Kansu, China, August 4, 1909, by Arthur de ('. Sowerby. Original number 246. Diagnosis. — In general like the European Eptesicus serotinus serotinus hut color somewhat less dark, particularly below, the contrast between that of dorsal and ventral surfaces evident enough to produce a slight line of demarcation along side of neck ; forearm shorter than in true serotinus ; skull tending to be rather short and broad. Measurements. — Type: head and body, 70; tail, 50; tibia, 22; foot, 13. 8; forearm, 40; thumb, 10.(5; third finger, Oil; fifth finger, 65; ear, 10. Skull : condylobasal length, 10.0; zygomatic breadth, 14.0; interorbital constriction, 4.0; breadth of braincase, 9.6; mandible, 15; maxillary toothrow exclusive of incisors, 7.2; mandibular tooth row exclusive of incisors, 8.2. Specimens examined. — Four, from the following localities: Ch'eng-yuan- hsien, 70 miles west of Ch'ing-yang-fu, Kansu (altitude 4,000 feet), 1 ; 18 miles east of Ku-yuan-chow, Kansu (altitude 5,300 feet), 1; 80 miles southwest of Yen-an-fu, Shensi (altitude 3,500 feet), 2. Microtus pul Ins sp. nov. Type. — Adult male (skin and skull) No. 155,047, U.S. National Museum. Collected at Chiao Ch'eng Shan, 00 miles west of Tai Yuan Fu, Shansi, China (altitude 7,000 feet), October 11, L908, by Arthur de C. Sowerby. Original number 32. 14— Proc. Biol. Soc. Wash., Vol. XXIX, 1911. (53) 54 Miller — Four New Chinese Mammals. Diagnosis. — Size and general characters as in Microtus (Ph;romys) Johannes Thomas, but color dark, the general effect approaching mars- brown instead of dull ochraceous-buft*. Color. — Upperparts a uniform fine grizzle of cream-buff" and black with a distinct brownish wash on crown and along middle of back, the general effect suggesting Kidgway's mars-brown and not in the least resembling the pallid ochraceous-buff of Microtus Johannes. Sides not so dark as back, the grizzling less evident. ' Underparts light cream-buff much clouded by the blackish-slate under color. Feet dull huffy white with an evident brownish wash. Tail whitish below, brownish above, not sharply bicolor. Skull and teeth. — The skull and teeth do not differ appreciably from those of Microtus Johannes. They are conspicuously smaller than in M. mandarin u*. Measurements. — Type: head and body, 104; tail, IS; hind foot without claws, 15.5; ear, 9. Skull: condylobasal length, 24.8(27.4)* zygomatic breadth, 16.2 (17.8); interorbital constriction, 3.6 (4.0); occipital breadth, 13.0 ( 13.0); occipital depth, 7.0 ( 7.2) ; nasal, 6.2 (7.0) diastema, 7.8 (9.2); mandible, 16.4(18.2); maxillary toothrow, 5. 8 H>.4); man- dibular toothrow, 5.8 (0.4). Specimens examined. — Four, all from the type locality. Allactaga mongolica longior subsp. now Type. — Adult female (skin and skull) No. 155,183, TJ. S. National Museum. Collected 15 miles northeastof Ching-ning-ehow, Kansu, China. Altitude 0,200 feet, July 2(5, 1909, by Arthur de C. Sowerby. Original number 204. Diagnosis. — Similar to true Allactaga mongolica (Pallas), but ear and hind foot longer, and audital bullae larger. Measurements. — Type: head and body, 150; tail, 230; hind foot with- out claws, 70. Skull: condylobasal length, 38.4; zygomatic breadth, 20.0; interorbital constriction, 11.4; breadth of braincase, 19.8; depth of braincase at middle, 13.6; nasal (median), 15.4; diastema, 12.6; man- dible, 24.8; maxillary toothrow, alveoli, 8.0; mandibular toothrow, alveoli, 7.8. Specimens examined. — Twelve, all from the type locality. Ochotona annectens sp. nov. Type. — Adult male (skin and skull) No. 155,164, U.S. National Museum . Collected 15 miles northeast of Ching-ning-chow, Kansu, China (altitude 6,200 feet, July 27, 1909, by Arthur de C. Sowerby. Original number 225. Diagnosis. — Similar to Ochotona dauurica (Pallas) from the Mongolian Plateau, but dorsal outline of skull less convex and audital bullae slightly larger. Differs from 0. bedfordi Thomas, of Shensi, and Shansi, China, in smaller size (particularly of skull) more convex upper cranial outline, and much smaller audital bullae. * Measurements in parenthesis are those of an adult male Microtus mandarinus from the vicinity of Tai-yuan-fu, Shansi. Millet — Four New Chinese Mammals. 55 Color. — As in the related species the general color is a light huff darkened by a plentiful sprinkling <>f black hair-tips, the huff everywhere predominating, and becoming clearer and more yellowish along sides, particularly in region of shoulder. A faintly indicated pale area between cars. Underparts and feet dull white with a slight huffy tinge; throat crossed by a broad hand of dull huff, this area sometimes extending backward along median line. Ear with the usual dark marking on outer side. Measurements. — Type: head and body, 181; hind foot, 29(26); ear, 20. Skull: condylobasal length, 40.0 (42.4);* zygomatic breadth, 20.4 (21.2); interorhital constriction, 4.2 (4.2); postorbital constriction, 13.2 (13.S); mastoid breadth including bullae, 20.8 (22.0); depth of braincase at middle, 11.0(11.4); nasal, 15.0 (15.0); diastema, 9.8(10.8). mandible, 27. '2 (30.2); maxillary toothrow, alveoli, 7.4 (8.6); mandibu- lar toothrow, alveoli, 7.4 (S.4). Specimens examined. — Seven; six from the type locality and one from Tsai-chiah-tsuei, 116 miles east of Lanchow, Kansu (altitude 6,300 feet). * Measurements in parenthesis are those of an adult male Ochotona bedfordi from Wu-tsai, Shansi, China (No. 172,601). Vol. XXIV, pp. 57-60 February 24, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON NOTES ON SOME BIRDS FROM THE ISLAND OF GRENADA. BY T. BARBOUR. During part of August and September of the past summer, Dr. Glover M. Allen and Mr. C. T. Brues visited Grenada for the purpose of studying the fauna. Their collections are now all in the Museum of Comparative Zoology. The ornithological collection from Grenada already in the Museum was so complete that no attempt was made to preserve series of the common birds. Special efforts to secure some of the more interesting forms were rewarded by the examples upon which these notes are based. Hybris* nigrescens noctividus subsp. nov. Type, from St. George's, Grenada, No. 53,586, Museum of Comparative Zoology, adult male. Collected by Glover M. Allen, September 15, 1910. Characters. — Size tbat of Aluco nigrescens (Lawrence) of Dominica, but differing as follows: tarsus somewhat longer; under parts darker in color, and much more heavily spotted with dusky; upper parts darker, and more profusely marked with larger white spots than in the owl from Dominica. Measurements. — Wing, 230 mm. ; tail, 100 mm. ; tarsus, 110 mm. ; cul- men, 30 mm. Remarks. — Thanks to the courtesy of Dr. Charles W. Richmond we have been enabled to compare the type of Lawrence's A. nigrescens, a male from Dominica (U. S. X. M. No. 77,843), as well as another from the same locality (U. S. N. M. No. 80,017), with an adult female also from this island (Bang's Coll. M. C. Z. No. 13,614). These three birds are all pale ochraceous below, somewhat mixed with whitish; while the dusky spots are confined almost wholly to the belly. * Hybris is used instead of Tyto, which Matthews suggests (of. Matthews, Nor. Zool., XVII, 1910, p. 500), because Tyto, the name for an owl, is preoccupied by Tyta Billberg, the name for an owlet moth. The two names have the same derivation from the Greek word tvtu}, meaning an owl. 15— Peoc. Biol. Soc. Wash., Vol. XXIV, 1911. (.57) 58 Barbour — Notes on some Birds from the Island of Grenada. Our Grenada material consists of the type already mentioned, and seven others, of which three are loaned hy the United States National Museum; the Museum of Comparative Zoology having specimens from Bequia in the Grenadine Islands, as well as St. Vincent. All the hirds from these southern islands of the Lesser Ant il lean chain are similar; and have the underparts tawny, somewhat darker than Ridgway's color on plate 5. Planesticus nigrirostris personus subsp. nov. Type, from the Grand Etang, 2,000 feet altitude, Grenada. No. 53,59S, Museum of Comparative Zoology, adult male. Collected by Glover M. Allen, September 6, 1910. Characters. — Similar to P. nigrirostris (Lawrence) of St. Vincent, but differing in that the upper parts are much more darkly olivaceous, less reddish olive; the lower parts more grayish, less rufescent ; and the under- wing covers rather paler. Remarks. — Mr. Ridgway, in Birds of North and Middle America, has already called attention to these differences. The characters being con- stant, as a comparison shows, these birds may be considered separate subspecies, as he suggested possible. Among the birds credited to Grenada was one described from a single young specimen in very bad preservation, the only one ever known, and called Blacicus flaviventris Lawrence. Wells, known for so many years as a collector of Grenadian birds, tried in vain to find other specimens of this extremely rare Tyrant bird. Dr. Allen was especially cautioned to watch for it, and by rare good fortune secured an adult male in the deep forest near the Grand Etang, a lake in the central part of the island at an altitude of about 2,000 feet. This speci- men was the only one which he saw, in spite of a long and careful search It was obvious at once that the bird was very different from any other Blacicus; and Mr. Ridgway, when he saw it, decided definitely that it was really an Empidonax; although there was no doubt as to the identity of this specimen with the remains of the type, now in the United States National 'Museum. As the specific name flaviventris is preoccupied, it becomes necessary to give this bird a new name. A note regarding its relationships, and a description follow. Empidonax johnstonei nom. nov. Described from an adult male, No. 53,591, Museum of Comparative Zoology; from the Grand Etang, Granada, B. W. I., 5 Sept., 1910, col- lected by Glover M. Allen. This Tyrant bird apparently has been derived from Empidonax law- renci Allen, of Trinidad, but it has become differentiated by isolation to a considerable extent. I have been enabled to make direct comparison with a fine male of E. lawrencei (American Museum of Natural History No. 50,067) thanks to the courtesy of Mr. Dewitt A. Miller. Characters. — Similar to K. lawrencei but smaller; with a narrower bill; Barboui — Notes on some Birds from the Island of Grenada. .^9 color pattern very similar but the upper parts a very dark greenish olive, the cap and cheeks dusky; instead of dull olive green with the cap hut little darker. Wing bars narrower and darker, being tawny ochraceous instead of ochraceous buff. Throat more purely white, underparts other- wise the same. Measurements. — M. C. Z. No. 53,591, d? : Wing, 60; tail, 54; tarsus, L4.5; exposed culmen, 12. A. M. N. H. No. 59,0(i7, c? : Princestown, Trinidad, March 1(1, L893, F. M. Chapman, collector. Wing, 64; tail, 58; tarsus, 13.5; exposed culmen, 13. The new specific name of this rare bird, which is especially interesting as being the only Antillean species of its genus, is given in honor of His Honor Robert S. Johnstone, Chief Justice of Grenada and dependencies, who has often aided the members of expeditions from this Museum by his kind hospitality and in many other ways. Vol. XXIV, pp. 61-62 March 22, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A NEW POCKET MOUSE FROM WYOMING. BY MERRITT GARY. Specimens of Perognathus in the Biological Survey collection from Sweetwater Valley and the northern edge of the Red Desert, Wyoming, represent a well marked pale race, hitherto undescribed, of P. fasciatus. , This pocket mouse may be known from the following characterization : Perognathus fasciatus Iitus subsp. nov. Type from Sun, Sweetwater Valley, Wyoming. No. 160,600, Q ad., U. S. National Museum, Biological Survey Collection. Collected Sept. 18, 1909, by M. Cary. Original No. 1778. Range. — Lower Sweetwater Valley and adjacent parts of Red Desert, Wyoming. General Characters. — A little smaller than P. fasciatus, about the size and proportions of P. flavescens, but with slightly larger hind foot than either; color extremely pale; pelage very soft. Color (September specimens). — Upperparts and sides palest shade of cream buff, sparsely lined with black-tipped hairs, the strong olivaceous tints of fasciatus entirely absent; underparts pure white; orbital ring, auricular spots and lateral line pale cream buff; tail indistinctly bicolor, slightly dusky above, white below; feet white. Skull. — As in fasciatus. Measurements. — Type: Total length, 128; tail vertebrae, 59; hind foot, is. Que topotype: total length, 127; tail vertebrae, 57; hind foot, 18. Remarks. — This pocket mouse is related to P. fasciatus of the north- ern Great Plains, with which it agrees closely in all respects except color. The only specimens known — two from Sun and one from near the old Lost Soldier Stage Station 40 miles northwest of Rawlins — were col- lected on sandy sage and .[triplex flats. A September specimen from the North Platte Valley at Casper is considerably paler than typical fasciatus from the Yellowstone Valley, Montana, and shows an approach to litus. 16— Proc. Biol. Soc. Wash., Vol. XXIV, 1911. (F THE BIOLOGICAL SOCIETY OF WASHINGTON SOxME OBSERVATIONS MADE ON LITTLE RIVER, NEAR WICHITA, KANSAS, WITH REFERENCE TO THE UNION I ICE.* BY II. WALTON CLARK AND GEORGE II. GILLETTE. [Published with the permission of the Hon. < }& irgc M. Bowers, U.S. Commissioner of Fisheries.] Little Arkansas River, more commonly known locally as Little River, is one of the largest northern tributaries of the Arkansas River in Kansas. It rises in the great wheat fields of central Kansas and takes a winding southeasterly course, emptying its waters into the Arkansas River at Wichita. The country through which it flows is wholly an agricultural region. broad fields of corn occupying the bottom land lying next to the stream on each side, while on the higher land farther away lie extensive wheat fields. Little River, in marked contrast with the usual type of western stream, keeps up a steady flow throughout the year, and is not materially affected by the dry season of July, August and September. This would seem to indicate that the stream is fed by numerous springs or underground How. The upper course was not examined, but in the lower course the well-water of the bottom lands is reached at a depth of from 15 to 20 feet, which is about the depth of the hanks of the stream. The water is unusually uniform in temperature, being cooler in summer than other streams of the vicinity, and abnormally warm in winter, so that it is rarely ever frozen over very thickly even in the coldest weather, and in some portions it has never been known to freeze entirely over. These circum- stances also tend to indicate that the river owes its steady How *The field investigations on which this paper is based were made by the junior author in August. 1910. 17— Pkoc. Biol. Soc. Wash., Vol. XXIV.1911. (63) 64 Clark and Gillette — Unionidas of Little Arkansas River. in a considerable measure to many springs, or that its bed lies about on the level of the underground flow. The water is moderately clear, excepting after heavy rains, when it has a dark gray appearance. This is in marked con- trast to the waters of the Arkansas River proper, the waters of which are light ashy in color at normal stage. The banks of Little River are lined with trees and underbrush. Among the trees are found Cottonwood, willow, birch, walnut and elm, the trees and underbrush extending back from a few to several rods. The banks on one side are rugged and steep, while on the other they are low and in some places boggy, — the high and low side depending on the curvature of the stream ; on the convex side of the curve will be found the high bank, and on the opposite or concave side is the low and boggy portion . Observations were made from near the mouth of the river to about 10 miles upstream, near Valley Center. About two miles above Wichita a dam, known as Sullivan's dam, has been built, for the purpose of dividing the flow of the river, causing a por- tion of the flow to run through the eastern part of the city. It is said that this dam was built for two purposes : ( 1 ) to protect the city from overflow and (2) to furnish means of flushing a small stagnant creek, Chisholm Creek, which flows through the eastern part of the city, and into which the refuse of manufac- turing plants, packing houses, etc., is dumped. This dam is built principally of cement, is 4 to 6 feet high and from 50 to 75 feet long. As it has no fish way, the only means by which fish below the dam could get up above would be by passing down stream into the Big Arkansas and then up Chisholm Creek, and finally pass through this artificial ditch or cut-off to the waters above the dam. At the head of the ditch near the dam is a floodgate, also built of cement. When the water in the river is running high this gate is kept closed, so as not to let too much water into Chisholm Creek; during low water, however, the gate is always kept open. Above the dam the river is considerably deeper than below ; it has a much slower current, and spreads out to a breadth of from 75 to 100 feet. Most of the collecting was done above the dam. This was very difficult because the bottom was full of broken logs, rubbish, etc. The bottom consists mostly of muck and coarse gravel, with very little shifting sand, the Clark and Gillette — Unionida of Little Arkansas River. 65 gravel being in nearly every case in the center of the stream, and the black muck along the shores and in all the eddies. About 100 shells were collected. While the shells were not especially abundant, excellent examples of some of the good commercial species are present; among them the heelsplitter (Symphynota com/plancda) , buckhorn or pistolgrip (Tritogonia tubercuiata) , j'ellow sandshell (Lampsilis anodontoides) and mapleleaf (Quadrula lachrymosa). Many of the specimens, especially of the lighter-shelled species such as the floater (Anodonta grandis) and heelsplitter, were obtained in the eddies and protected places in the stream where there was very little current. The heavier-shelled species, such as the buckhorn, yellow sandshell and mapleleaf were found both in the eddies and protected places, and in some instances in the swiftly running stream, but no shells of any kind were found where there was any shifting sand. A favorite place for the heavy-shelled varieties was the lower edges of the sand or gravel bars, while on the tops of these bars where the water was swifter, none was found. We were unable to determine to what extent they were found in similar situations in the deepest water, as we had no tongs or grappling hooks. So far as we could learn no mussels were being taken from the river for commercial purposes, although we heard of a party who had obtained quite a valuable collection of pearls from mussels taken in the river. In many of the specimens taken, small poorly developed pearls were found, but not of sufficient size and quality to be of much value. Observations were also made on the Big Arkansas above Wichita on August 26. The river at this point is wide and shal- low, the water at this time covering only about one-fourth of the bed. The main channel of the stream did not follow the center of the bed, but crossed diagonally from one side to the other. The current was swift and carried with it a large amount of shifting white sand which gave the water a yellowish creamy appearance. About 3 miles were covered during these observa- tions. Owing to so much shifting fine sand and the changing of the main channel of the river during high water, it is impos- sible for mollusks requiring the support of a permanent bottom to exist in this river at the place of observation. In this three- mile course only one valve of a mussel shell was found and this 66 Clark and Gillette — Unionidse of Little Arkansas River. was picked up in a rubbish pile. It was bleached and very brittle, and gave evidence of having been carried a long distance. It had mi doubt been washed into the river from one of its tributaries. List of Species. 1. Quadrula pustulosa (Lea). WARTYBACK. There arc 15 examples of this species, L2 from near Wichita and ."> from Valley Center. These are all rather small or medium-sized shells and quite uniform in character, being markedly compressed and unusually smooth, most of them being entirely free from elevations and the most pustulous one containing a few very low, hardly perceptible nodules near the ventral border. The whole collection presented a considerable differ- ence from Q. pustulosa as usually seen, and it was only after some con- sideration and comparison that they were identified as this species. They represent the form originally described as a distinct species under the name Unio sehoolcraftii. This flat, smooth form is of occasional occurrence in collections, and examples are now and then found mixed in with beds of the more common inflated pustulous form. This is our first experience in finding it the predominant form. The existence of this and intergrading forms, along with other aberrant and peculiar types, has long made (J. pustulosa a puzzling species. The history of the studj of this species would be merely an account of the various attempts to assemble a motley but well-connected series of forms, and authorities differ somewhat as to the number of forms to include. Baker* speaks of baving before him at one time 19differen1 varieties of Q. pustulosa from a number of States, and Calif discusses its variability and synonymy at consid- erable length, and says: " From the Little Arkansas, at Wichita, Kansas, come numbers of magnificent examples of sehoolcraftii, some entirely covered with pustules, others absolutely devoid of even a semblance of one ; indeed the w liter's collection contains some fifty exam pies from that stream, exhibiting every phase of nodulation from absolutely smooth specimens to those showing o;reat numbers of small pustules. The characters of the cardinal teeth alone would have sufficed, in the hands of species mongers, to make a dozen extremely characteristic species." In view of this statement it is rather remarkable that all our specimens should he so uniform. These smooth flattened shells are even superior to the inflated pustulous forms for commercial purposes. All our specimens, however, are of rather small size. 2. Quadrula lachrymosa (Lea). MAPLELEAF. Of this species there are 22 from near Wichita in the collection and 10 from near Valley Center. They are all of large size and very uniform in general appearance. * Mollusca <>! the Chicago Area; Bull. Ill, part 1. Nat. Hist. Survey, Chicago Acad. fThe Unionidse of Arkansas; Trans. Acad. Sci. St. Louis, Vol. VII, Xo. 1, p. 43. Clark and GilletU — Ifnionida of Little Arkansas Hirer. 67 Compared with the species as generally known, these shells exhibit the same departures from the general type as those exhibited by the Q. pus- tulosa just mentioned; that is, they are unusually compressed, and note- worthy for the fewness and smallness of pustules which are frequently altogether absent on the posterior ridge. The sulcus is also unusually shallow. The shells, however, average larger than in ordinary collections. The nacre is white and clear, and free from stains. Q. lachrymosa is a very fair commercial species, and the specimens from the Little Arkansas are considerably superior to the ordinary type on account of the flatness and smoothness of the shell. :;. Symphynota complanata (Karnes). HEELSPLITTER. Fifteen examples of this species were obtained, ll' from near Wichita, and :; from Valley ( !enter. The heelsplitter dwells in a variety of situations such as ponds, sloughs and the more quiet portion of rivers. It varies considerably in minor details such as thickness of shell, development of wing, etc., but is gener- ally pretty uniform in general shape. Bayou-dwelling examples are usually roughened, thin-shelled and badly stained, so that they have no commercial value. In some places (the upper Mississippi) the shells are peculiarly truncate posteriorly, as if they had abruptly stopped growing in that part of the shell. We have had no opportunity to study the bodies of S. complanata from Little River. < Jenerally speaking the species is, like the Anodontas, quite markedly subject to the attack of parasites, leeches being occasional, Atax frequent and Aspidogaster concliicola abundant. River-dwelling mussels are usually not so badly affected as those living in sloughs, and from these perfectly formed and excellent shells it would appear that parasites are not especially abundant here. Symphynota complanata is usually too thin, and sometimes too badly Stained to be of any value in the manufacture of buttons so that it is not generally regarded as a commercial species. On account of its flatness and broad expanse of shell, when it develops a sufficient thickness of shell it is an excellent button species. No other native freshwater shell except the immense thick (Jandrula linos and the beautiful but valueless Anodonta suborbiculata equals it in surface, and where it can be w^'d more buttons can be cut from a ton of shells of this species than from any Other freshwater mussel. The specimens from the Little Arkansas are exceptionally fine. We have seen none to approach them in excellence except a few examples much like them collected in Perche Creek, Missouri, by Doctors Lefevre and Curtis, of the University of Missouri. Some of the shells, both from Perche Creek and Wichita, are unusually elongate, and all are remarkably smooth and heavy. The nacre is of a clear soft satiny texture, pearly white in all but '2 examples, in which it is a very faint warm yellowish pink, too faint to be more than a mere suggestion of color and very 68 Clark and (J ill cite — Unionidse of Little Arkansas River. attractive to the e}Te. Although they lack the iridescence of the best niggerhead material the soft satiny luster is scarcely less attractive, and these shells would make excellent button material. •Inst within the nacre of one of the shells are numerous tine whitish, long, crooked tracks, each ending in a small elevation which shows dark be- neath. The dark point proves, on examination, to be a larval Atax which has been coated over with nacre. The peculiar tendency of young Atax to bore entirely through the mantle, crawling along on the nacre and finally becoming covered over with a nacreous deposit, was first noticed during the past summer (1910) by the senior author in »S'. com- planata from the sloughs near Fairport, Iowa. So far, we have observed it in no other species of mussel. It seems improbable that the larvte could ever work their way out again after being thus entombed. They appear to have no other effect upon the nacre than to leave the whitish tracks ending in the low raised point already described, and both of these marks are probably soon covered up and concealed by a new growth of shell. 4. Anodonta grandis Say. FLOATER. Fifteen examples, all from near Wichita. These are all mature shells of fairly large size, of the unit us type. None of the shells shows the salmon- colored and diseased nacre usually frequent where parasitic trematodes are common. There are a few pearl-like blisters in some, but as a rule they are free from abnormalities. No. 41 is rather thick and No. 42 has pink nacre. The species has no commercial value. 5. Tritogonia tuberculata (Barnes). lUVKHORN : PISTOLGRIP. Three rather large shells, all from near Wichita. All are moderately elongate and have white nacre. They are of good commercial quality. <;. Lampsilis anodontoides (Lea). YELLOW SAND8HELL. One medium-sized example from near Wichita. This shell is con- siderably deformed and twisted, and the greater part of the epidermis is stained and unusually imbricate. Where it reaches good development, this is the most valuable of the freshwater shells. Vol. XXIV, pp. 69-76 March 22, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON NOTES ON THE COLORATION OF FISHES! BY r.AKTOX A. BEAN AND ALFRED C. WEED. [Published by permission of the Secretarj of the Smithsonian Institution.] While it is well known that the color of fishes of the same species varies more or less it may not be amiss to call especial attention to some of the more marked changes that may be noticed. The color changes in living fishes may be due to various emotions as fright, curiosity, anger, or sexual excite- ment, or they may he due to environment, as the color of the bottom, the color of the water, distance of the fish from the surface or from bottom, distance from objects attached to the bottom or floating at the surface, presence in the immediate vicinity of grass, sticks, etc. Many people who are familiar with the fact that the color and color pattern may change in the same species have an idea that this takes place slowly. We hope to show that this does not necessarily follow and will give some illustrations of color changes which we have noted in various species in aquariums, principally that of the Bureau of Fisheries at Washington, D. ('. The Large Mouth Black Bass (Micrepterus salmoides) when viewed as an ordinarily preserved specimen usually shows more or less traces of a black lateral band. In the large specimens this may be very faint but can usually be discerned. Similar specimens of the Small Mouth Black Bass (Micropterus dolomieu) show no traces of this black band but are plain olivaceous in color. There are several living specimens of each species in the aquarium at the Bureau of Fisheries where we have spent some time watching the color changes. One large specimen which showed very strongly the typical black lateral band of the Large 18— Proc. Biol. Soc. Wash., Vol. XXIV, 1911. (69) 70 Bean and Weed — Coloration of Fishes. Mouth Black Bass changed in less than two seconds, so that the band was entirely invisible and the fish appeared to be a Small Mouth Black Bass. In preserved specimens of both species the color, except for the black lateral band, is plain; that is, there are no mottlings or marmorations, but in the aquarium, where the fish are so used to seeing visitors that they pay no attention to their presence, the color may be seen to change from plain olivaceous without the black lateral band to plain olivaceous with such a band, or to a mottling of light and dark green with or without the black band. The Large Mouth Black Bass when showing the mottled color usually shows the black band ; speci- mens of the Small Mouth Black Bass usually do not show the black band and the mottlings have a greater tendency to appear as dark cross bars. It is difficult to determine the exciting causes of these changes as fish hiding in crevices of the rocks will show both colorations and fishes swimming show both color patterns. It appears, however, that fish swimming freely or hiding in rather dark holes have a greater tendency to exhibit the plain coloration, while those resting on the bottom or swim- ming close to the rocks at the side of the aquarium have a greater tendency to show the marbled or mottled color pattern . Preserved specimens of Tautog (Tautoga onitis) are usually nearly plain black but with more or less distinct vertical light and dark bars of about equal width. In the aquarium they show remarkable variations of these two color patterns. When swimming freely the color is usually plain black and when rest- ing on the bottom or at the sides of the aquarium the bright pattern is visible. However, these colors are not constant and swimming fish may be seen showing the bars while those resting on the bottom show the dark color. The color changes are frequently very rapid. We have seen a fish swimming in a vertical circle about two feet in diameter showing the bright coloration only near the bottom of the aquarium and the black color in the upper half of the circle at each turn. In the New York Aquarium at Battery Park notices are posted in front of certain tanks containing some of the tropical angel fish, calling attention of visitors to rapid changes in color and color pattern of this fish. These changes are truly marvelous. Waves of blue starting at the head pass to the tail and disap- pear. While one is watching the fish the color will become all Bean and Weed — Coloration of Fishes. 71 blue or all brownish or half blue and half brown, etc. Indeed the changes are too numerous to mention. The Pigfish (Orthopristis rhn/sopterus) shows very remarkable changes in color and color pattern. There are three perfectly distinct color patterns which appear to be about equally com- mon. One is a perfectly plain light or dark gray without spots or bars of any kind. Another has this same gray ground with several heavy black cross bars, which sometimes are more or less broken up into dark mottlings. A third shows instead of the cross bars a black line from the end of the snout to the base of the dorsal fin and along each side of this fin. Below this is a narrow light stripe with a heavy black band below it. A second dark band extends from the eye straight to the caudal peduncle. The lower edge of this band is just above the upper lip. Between this and the dark band above it is a light stripe about the same width as either of these dark bands. The upper dark band crosses the front of the head about midway between the lower band and the front of the dorsal fin. There are also a few small black lines across the front of the head which make the markings appear much like a bridle. One of these lines passes from the upper edge of the opercle to the upper edge of orbit and then across front of head. Sometimes the two sets of bars fade out except at points where they cross, giving rise to a series of square black blotches on the side of the fish. Any of these color patterns may appear instantane- ously, following any other. Sometimes the change may be likened to the fading of one picture and the appearance of another in a stereopticon . Specimens of the Pinfish (Lagodon rhomboides) in the aqua- rium show mostly a color pattern made up simply of longitudi- nal blue stripes of greater or less intensity on a grayish ground. However, at times black cross bars appear. The appearance and disappearance of these bars may be practically instantane- ous or it may be quite slow. The ground color between the blue lines often becomes a bright shell pink. Many of our common fresh-water sunfishes show remarkable changes in their color in life. The Common Sunfish or Tobacco Box " (Lepomis gibbosus) does not show these changes so well as some of the other species. The ' Green Sunfish " (Lepomis cyanellus), a widely distributed fish in the 72 Bean and Weed — Coloration of Fishes. Mississippi Valley, which has been introduced into the Potomac River, shows these color changes more than any other sunfish with which we are acquainted, in fact the changes of color and color patterns are almost as numerous and as rapid as in the angelfish. In addition to the normal color changes, an increase or reduction in the intensity of the color, or a change from dark to light color, due to emotions or environment, there are many other changes for which we can only guess at the cause. We shall probably never know why one fish when frightened will turn pale while another in the same aquarium will become very strongly barred. The commonest color, visible in preserved specimens, in dead fish and in fish just taken from the water, is a plain dark olivaceous with more or less dark red in the vertical fins and with two blue lines across the cheek. In the aquarium this may be seen to change to a rich golden brown with or without vertical cross bars or to a plain light color almost silvery. The vertical cross bars are sometimes as pronounced as in the Yellow Perch ( Perca fiavescens) , at other times the whole side of the fish is covered with minute shining specks without any apparent regular arrangement. Any of these color patterns may appear practically instantaneously. The change from vertical bars to the plain coloration with light specks may occur apparently in a flash. Where a few speci- mens of approximately equal size are confined in a small aquarium these color changes may be most readily noted. The intensification in color of the largest specimen when a smaller one happens to stray into forbidden territory and the sudden paling of the latter as he retreats may lie seen very frequently, especially when the fish are being fed. The Long Eared Sunfish (Lepomis auritus), known also as Red Breasted Bream, is commonly a more or less yellowish fish with orange spots and mottlings over the entire body, these sometimes showing as more or less indistinct cross bars. There are many specimens of this species in the aquarium at the Bureau of Fisheries where we have watched their color changes. Most of the specimens showed a dark olive color with the verti- cal fins and the breast dark red. Sometimes faint cross bars show. This coloration is so similar to that of Lepomis cyanellus that only the position of some blue lines on the operculum and the shape of the " ear ' ' indicated the specific difference. From Bean and Weed — Coloration of Fishes. /3 this olive coloration fishes would quickly change to a golden brown with or without dark cross bars or with bars formed of small spots slightly darker than the ground color. One fish which we watched for some time and which, hy its actions, appeared to be a male more or less approaching the breeding condition was a bright yellow color mottled over the whole side with dark orange spots each about the size of a scale. The breast was a clear bright orange. The pectoral fins were a clear lemon yellow. There was hardly a trace of cross bars although at times a few could be seen quite indistinctly. This fish was observed several times and seemed to show no color changes except, on one occasion, a slight tendency of the yellow ground color to appear grayish. The Blue Sunfish (Lepomis pallidas) which is known also as Bluegill, Copper Nosed Bream, Blue Bream, Black Sunfish, Roach (Sodus Bay, New York), etc., shows various color pat- terns. Sometimes the fish will be a plain iron color with no markings at all except the black spot on the opercular Hap. At other times it is a plain yellowish green with no markings or it may show either of these colors with faint or dark cross bars, or the light ground color may he almost entirely obscured by heavy dark bars with dark mottlings between them. In Eftneacanthus gloriosus, a common sunfish of the vicinity of Washington, color changes may be noted which are very sim- ilar to those of Lepomis cyannellus. The changes from vertical dark and light bars to the coloration of bright specks is practi- cally the same as in the Green Sunfish except that in Enneacanthus the heavy black stripe through the eye never, so far as we know, becomes entirely invisible. The War Mouth (Chsenobryttus gulosus) shows little change in color except that the extent of the dark patches varies consid- erably so that the same fish may be grayish mottled with black or nearly all black with a few grayish streaks. Fish caught on hook and line near Washington often show much red or dark orange in the ground color. Some of our common fish, which are frequently seen in aqua- riums, as the Gar Pike (Lepidosteus osseus) and Strawberry Bass or Calico Bass (Pomoxis sparoides and Pomoxis annularis) , appar- ently show no changes except an increase or reduction in the intensity of the coloration. 74 Bean and Weed — Coloration of Fishes. Preserved specimens of the Pickerel (Esox reticulatus) show very slight differences in color pattern, which is usually more or less of a network of dark lines on a lighter ground, but great differences in its visibility. These differences are probably largely due to the condition of the fish when taken from the water and difference in manipulation in preserving the speci- men. Fishes from clear water usually have the color pattern more pronounced, while those from muddy water, particularly water of a distinctly yellow color, frequently show no traces of the color pattern when first taken and if killed in the preserva- tive immediately will show none. However, if these fish are placed for a few minutes in clear water the color pattern becomes visible, and it may appear in cold water even after the fish has died, showing that the color cells may remain active for a time under favorable conditions. Fishes of this species swimming near the surface in the aquarium show a very strongly marked color pattern. In some of them the reticulations extend entirely over the back, while in others the back is more or less plain dark olivaceous. When these fish are resting on a bottom of clear yellow sand the vertical bars of the reticula- tions tend to disappear and portions of the horizontal bars also become fainter so that the markings resemble wave marks on the bottom. The color of the back becomes plain greenish yel- low, except for the presence of scattered black specks resembling, in color and arrangement, black sand grains in the neighbor- hood. The fish resting near the surface resembles a floating stick covered with spots of light refracted by the small waves. In this species the color changes are not in the nature of a change in color pattern, but simply variations in its intensity. In the Yellow Perch (Perca flavescens) the color pattern does not change, but the intensity of the black cross bars is quite variable. Sometimes they are almost invisible and the fish ap- pears to be a clear yellow color, while at other times they are so broad and intense that the fish appears almost black. The ground color changes with the bands from a light yellow to a dark yellowish gray. The Sea Robin (Prionotus evolans), the Puffer (Spheroides mar tdatus) , and the Filefish (Alutera scripta) show no change in the color pattern, but considerable changes from golden to grayish in the ground color. Specimens of the Sea Robin Bean and Weed, — Coloration of Fishes. 75 which had buried themselves in the sand were light gray with black markings, while those that were swimming were various shades of yellow and brown marked with darker brown. Early in the summer of 1908 a small catfish (Amciurus sp.) was taken in a dip net in the Chesapeake and Ohio Canal, at Chain Bridge, near Georgetown, D. C. When first taken it was entirely black. Tins color persisted for several days and was then replaced by a most unexpected pattern. The body was black, but at the base of dorsal, adipose dorsal, anal and caudal fins was a narrow white band so that each fin appeared cut off from the body, which seemed to be a little piece of black stick with some little pieces of rotted leaves (the black part of the fins) near it. The caudal fin was in reality truncate, but there was a large white patch in it which made it appear that the upper and lower rays were greatly produced. This color in turn, after a few weeks, was replaced by the black color of the adult. Although many notes on the coloration of fishes have been published they are scattered through the literature of every branch of ichthyology. We offer the following partial list of the more recent papers: 1897 — A. E. Verrill — Nocturnal Protective Coloration of Fishes. Am. Nat., Vol. XXXI, Feb., 1897, 99-103. Am. Journ. Sci., Vol. Ill, 1897, 132-136. Abstract of paper read before Am. Morph. Soc, Dec. 30, 1896. 1905— D. S. Jordan— A Guide to the Study of Fishes. New York, 1905. Vol. I, pp. 226 to 236. 1907— Raveret-Wattel— " Le Pecheur" 18 anee No. 421, Sept. 30, 1907, p. 7(37. " La coloration des Poissons " gives an account of the slow color changes in fishes under the influence of great changes in color of environment. 1908 — Francis Ward — Markings and colors in fish. How they protect their wearers. Scientific American, Supplement No. 1714, p. 297. New York, 1908. A few notes on protective relation of color to surroundings in Stone Loach, Pike and Gudgeon. Experiments on Pike with head in dark and body in light and vice versa. This paper was apparently published or offered for publication in " Country Life," but we have not found it in that magazine. 76 Bean and Weed — Coloration of Fishes. 1909— C. Tate Regan— Proc. Zool. Soc. London, Feb. 16. 1909, p. 130. Gives notes on four species and criticises Dr. Evermann's note on species of Bodianus based on color in Fishes of Porto Rico, Bull. U. S. F. C, 1900 (1902). 1909— Jacob Reighard — An experimental study of warning coloration in coral reef fishes. Carnegie Inst. Pub. 103, pp. 257-325 ( 1909). Experiments in feeding variously colored baits to wild risbes. 1909—0. H. Townsend— 13th Ann. Rep. N. Y. Zool. Soc, Jan. 1909. Observations on instantaneous changes in color among tropical lis lies. 1910— 0. H. Townsend— "The Century Magazine," September, 1910. Chameleons of the Sea. Notes instantaneous color changes in many tropical risbes and slower ones in some northern ones. Practically a reprint of 13th Ann. Rept. N. Y. Zool. Soc, Jan., 1909, pp. 1-28. 1910 — F. B. Sumner — Adaptive color changes among hsbes. Bull. Zool. Soc. X. Y., No. 42, p. 699, 1910. ( v>lor changes of Pleuronecti- d.r. Five photographs of the same hsh on different bottoms. 1910— R. C. Osburn— Bull. Zool. Soc. N. Y., No. 42, p. 704, 1910. Cich- lid fish at the Aquarium. These fresh-water fishes have the power of instantaneous color changes. Vol. XXIV, pp. 77-86 May 15, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON THE NOMENCLATURE OF THE HYDROMEDUS.dE. COMPILED BY T. D. A. COCKERELL. The list given below is compiled from Dr. A. G. Mayer's 'Medusae of the World," in the same manner as that of the Scyphomedusse previously published (Proc. Biol. Soc. Wash., XXI V, p. 7).* Like the other list, it has been very kindly scrutinized by Dr, Mayer, who has made a number of sugges- tions. Owing to the difficulties inherent in the classification of the Hydromedusse, the list is to be considered provisional, but as it is based on Dr. Mayer's monograph, while at the same time correcting a number of nomenclatural errors, it should be serviceable. The type-species is cited after each generic name. Order ANTHOMEDUS^E. Fam. SARSiiDiE [Codonidae]. Subf. Sarsiin^. Tribe Pachycordylini. Pachycordyle Weismann. napolitana Weismann. Tribe Amalthseini. Amalthxa Schmidt. uvifera Schmidt. Tribe Pennariini . Pennaria Oken. pennaria (L.). * In the list of Scyphomedusae are a few errors, unfortunately overlooked by me. but detected by Dr. Mayer, as follows: p. 8, read Tesserantha eonnectens; p. 9, read Kishinouyea; p. 10, read Drymonema; p. 12, read Eupilema scapulare. Also for Pe>.& Less, read P6r. & Les. 19— Proc. Biol. Soc. Wash., Vol. XXIV, 1911. (77) 78 Cockerell — Nomenclature of the Hydromedusas. Tribe Trichorhizini. Trichorhizci R u ssell . brunnea Russell. Tribe Steenstmpiini. Steendrvpia Forbes. rubra Forbes. Hybocodon L. Ag. prolifer L. Ag. Microcampana Fewkes. conica Fewkes. Tribe Dicodoniini. Dicodon iurn Haeck . cornutum Haeck. Tribe Sarsiini. Sarsia Lesson. tubvlosa Lesson. Moerisia Ron] en g. lyonsi Bouleng. Hyi Irichth ys Fewkes. mirus Fewkes. Eucodonium Hartlaub. brownei Hartlaub. Ectopleura L. Ag. dumortieri (Van Rened.). Linvillea Mayer. [Corynitis McCrady, preocc.] agassizii ( McCrady ) . Dipurena McCrady. [Slabberia Forbes, preocc . ] strawgidata, McCrady. Subf. Margelopsin^e. Margelopsis Hartlaub. haeckelii Hartlaub. Pelagohydra Dendy. mirabilis Dendy. Fam. Cladonemid^;. Subf. Pteronemin^. Zanclea Gegenbaur. codata Gegenbaur. Cockerel! — Nomenclature of the ttydromedusse. 79 Zancleopsis Hartlaub. dichotoma ( Mayer). Pteronema Baeck. darwinii Haeck. Eleutheria Quatref. dichotoma Quatref. Mnextnt Krohn. parasites Krohn. Subf. Cladonkmix.e. Ctenaria Haeck. ctenophora Haeck. Cladonema Duj. radiatum Duj. Dendronema Haeck. stylodendron Haeck. Urashimea Kishinouye. globosa Kishinouye. Fam. Cyt.-eid.e. Subf. Pandein.e. Tribe Protiarini. Protiara Haeck. tetranema (Per. & Les.). Heterotiara Maas. anonyma Maas. Tribe Stomotocini . Stomotoca L. Ag. dinema (Per. & Les.). Dissonema Haeck. saphenella Haeck . Tribe Panel eini. Pandea Lesson. conica (Quoy & Gaim.). Conis Brandt. mitrata Brandt. Clavula Wright. neglecta (Less.); syn. gossel Wright. Subf. Cyt^in^. Tribe Cytaeini. Cytseis Eschsch. tetrastyla Eschsch. 80 Cockerell — Nomenclature of the Uydromedusse . Podocoryne Sars. carnea Sars. Turritopsis McCrady. nutricula McCrady. Modeeria Forbes.* formosa Forbes. Stylartis Allm. fuciola (Sars). Tribe Limnoreini. Thamnostylus Haeck . dinema Haeck. Thamnitis Haeck. tetrella Haeck. Lymnorea Per. & Les. triedra Per. & Les. Tribe Bougainvilliini. Bougainvillia Less. bougainvillii (Less. ) ; sy n . macloviana Less. Nemopsis L. Ag. bachei L. Ag. Rathkea Brandt. blumenbachii (Rathke). Chiarella Maas. centripetalis Maas. Subf. PROBOSCIDACTYLIN^e. Tribe Bytbotiarini. Bythotiara Giinth. murrayi Giinth. Calycopsis Fewkes. typa Fewkes. Niobia Mayer. dendrotentaada Mayer. Tribe Proboscidactylini. Proboscidactyla Brandt. flavicirrata Brandt. Willsia Forbes. sellata Forbes. * For the Oceania of Mayer's monograph I had proposed to use Callitiara Haeck, on account of the uncertainty which seemed to surround Modeeria. Dr. Mayer, however, advises the use of Modeeria. Oockerell — Nomenclature of the Hydromedusse. . 81 Order LEPTOMEDUSjE. Fam. Cannotid/E. Subf. Melicertin^e. Tetranema Haeck. eucopium Haeck. Laodicea Less. cruciata (Forsk.). Melicer'um Oken. campantda Oken . Melicertissa Haeck. clavigera Haeck. Orchistoma Haeck. pileus (Less.); syn. steenstrupii Haeck . Timoides H. B. Bigelow. agassizii H. B. Bigelow. Subf. POLYORCHIDIN^. Staurodiscus Haeck. tetrastaurus Haeck. Ptychogena A. Ag. lactea A. Ag. Poly orchis A. Ag. penicillata (Eschsch.). Scrippsia Torrey. pacific, a Torrey. Spirocodon Haeck. saltatrix Haeck. Subf. Cannotin^;. Cannota Haeck. dodecantha Haeck. Cuvieria Per. carisochroma Per. Dichotomia Brooks. cannoides Brooks. Dipleurosoma Axel Boeck. typicum Axel Boeck. Toxorchis Haeck. arcuatvs Haeck. Netocertoides Mayer. brachiatum Mayer. 82 Cocker ell — Nomenclature of the Hydromedusse. Fam. Eucopiid^;. Subf. Eucopiin^;. Eucopella v. Lendenf. bilabiata (Coughtrey). Agastra Hartl. mira Hartl. Euc opium Haeck. pictum (Kef erst. & Ehlers.). Obelia Per. & Les. marina (Slabber).* Tiaropsis L. Ag. multicirrata (Sars). Acosmetira n. n. (Cosmetira Hartlaub, not Forbes). pilosella (Forbes). Subf. Clytiin^. Clytra Lamx. volubilis (Ellis & Soland.). Phialidium Leuckart. t hemisphaericum (Per. & Les.) Phialopsis Torrey. diegensis Torrey. Phialucium Maas. virens (Bigelow). Blaclxfordia Mayer. manhattensis Mayer. Pseudoclytia Mayer. pentata Mayer. Gastroblasta Keller. timida Keller. Eucheilota McCrady. ventricularis McCrady. Mitrocoma Haeck. annse Haeck. *Some of Slabber's names have been rejected, not being binomial, but the present one is truly so, Dr. Mayer informs me. + For Phialidium I had substituted the much earlier name Thaumantias Eschsch., with the same type-species, but as the case is not clear, I retain Phialidium in the list. I thought that Oceania Per. <& Les., with the same type, might be available, following Agassiz. Dr. Mayer tells me that Leuckart in 1856 applied Oceania to what we now call Clavnla, the Oceania coccinea Leuck. being Clavula pileala (Forsk.). It will be necessary to reexamine the older authors to settle this case, but it seems improbable that Phiali- dium can stand. Cockerell — Nomenclature of the Hydromedusse. 83 Staurophora Brandt. mertensii Brandt. Subf. Eutimin^S. Saphenia Eschsch. bitentaculata (Quoy. & Gaim.). Eutima McCrady. mira McCrady. Eutonina Hartlaub. socialis Hartlaub. Subf. Eirenin^;. Phortis McCrady. gibbosa McCrady. Irenopsis Goette. hexanemalis Goette. Eirene Eschsch. viridula (Per. & Les. ). Tima Eschsch. flavilabris Eschsch. Fam. MqvoREiBM. Octocanna Haeck. polynema Haeck. Octogonade Zoja. mediterranea Zoja. Stomobrachium Brandt. lentlculare Brandt. Halopsis A. Ag. ocellata A. Ag. JEquorea Per. & Les. ssquorea (Forsk.). Rhacostoma L. Ag. globular e (Modeer); syn. atlanticum L. Ag. Zygocanna Haeck. pleuronota (Per. & Les.). Zygocannula Haeck. undulosa (Per. & Les.) 8f Cocker ell — Nomenclature of the Uydromeduscs . Order TRACHYMEDUSiE. Fam. Olindiid^;. Gonionemus A. Ag. vertens A. Ag. Cubaia Mayer. aphrodite Mayer. Vallentinia Browne. falklandica Browne. Olindias F. Mull. sambaquiensis F. Mull. Olindioides Goto. formosa Goto. Fam. Petasid^;. Petasus Haeck. atavus Haeck. Aglauropsis F. Mull. agassizii F. Mull. Craspedacusta Lank . sowerbil Lank. Microhydra Potts. ryderi Potts. Mseotias OstroumofT. inexspertata OstroumofT . Gossea L. Ag.* corynetes (Gosse). Fam. LimnocnididjE. Limnocnida Gunth. tangavjicse Gunth. Fam. Ptychogastriid^. Ptychogastria Allm. polar is Allm. Fam. Trachynemid^. Sllbf. TRACHYNEMINiE. Trachynema Gegenb. ciliatum Gegenb. Sminthea Gegenb. eurygaster Gegenb. * Gossea Bate, in Crustacea, is of the same year (1S62). Dr. Mayer has looked this up, and decides that Agassiz has priority. Cockeretl — Nomenclature of the Hydromedusse. 85 Homoeonema Maas. typicum Maas. Tetrorchis H. B. Bigelow. erythrogaster H. B. Bigelow. I'n nttichogon Maas. hseckelii Maas. Holier eas Fewkes. minimum Fewkes. Halitrephes H. B. Bigelow. maasi H. B. Bigelow. Botrynema Browne. brucei Browne. Crossota VanhofT. hninnea Vanhoff. Subf. Aglaurin^. Aglaura Per. & Les. hemistoma Per. & Les. Aglantha Haeck. digitalis (Per. & Les.). Amphogona Browne. apsteini ( Yanhoff. ). Stauraglaura Haeck. tetragonima Haeek. Persa McCrady. incolorata McCrady . Fam. Geryoniid^E. Liriope Less. exigua (Quoy. & Gaim.); syn. cerasiformis Less. Geryonia Per. & Les. proboscidalis (Forsk.). Order NARCOMEDUSiE. Fam. Solmarid^. SolmariH Haeck . leucostyla (Will.) Pegantha Haeck. pantheon Haeck. 86 Cockerel!— Nomenclature of the Hydromedusge . Fam. ^Eginid^. Mgina Eschsch. citrea Eschsch. Solmundella Haeck. bitentacvlata (Quoy. & Gaim.). Hydroctena Dawycloff. salenskii Dawydoff. Cunoctantha Haeck . octona ria ( M cC rad y ) . Mgimira Haeck. myosura Haeck. /Eginopsis Brandt. laurentii Brandt. ( 'mi i»a Eschsch. globosa Eschsch. Sid missus Haeck. albescens (Gegenb.). Cunissa Haeck. polyporpa Haeck. JEginodiscus Haeck . actinodiscus Haeck. Vol. XXIV, pp. 87-88 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON. A NEW CRINOID GENUS FROM THE INDIAN OCEAN. BY AUSTIN II. CLARK. While studying the magnificent collection of recent crinoicls at the British Museum, I detected among the undetermined material representatives of several new species, one of the most interesting of which is the form described below. \ urged Pro- fessor F. J. Bell, who is in charge of the echinoderm collections, to describe these new species; but, with bis characteristic gen- erosity, he suggested that, as I bad become so deeply engrossed in the study of the recent crinoids, it would be more fitting that I should do it. lam therefore deeply indebted to Professor Bell for the privilege of being able to make known the following interesting new genus. Cyclometra gen. nov. Genotype. — Cyclometra flavesc.ens sp. nov. Systematic position. — Family Antedonidae, subfamily Heliometrinse. Characters. — This new genus is closely related to Heliometra, which it replaces in the East Indian region; it is distinguished by having P2 markedly shorter than Pi instead of approximately of the same length or slightly longer, as in Heliometra, and by the outer segments of Pi, which are somewhat elongated instead of very short. Cyclometra flavescens sp. nov. Description. — Centrodorsal rounded conical. I lirri xxvin, 41-4:!. 30 mm. long; first segment short, second nearly as long as broad, the following gradually increasing in length and becoming twice as long as broad on the fourth; the next five are similar; the fol- lowing gradually decrease in length, after about six more becoming about as long as broad and remaining of those proportions until the end of the cirrus. The earlier segments are slightly constricted centrally, with the 20— Peoc. Biol. -<«'. Wash., Vol. XXIV, 1911. (87) 88 Clark — A New Crinoid from the Indian Ocean. ends all around finely spinous; the short outer segments are carinate, this carination appearing convex in lateral view. The ten arms are 130 mm. long; the i Br series and brachials are essen- tially as in Antedon bifida, but the distal ends of the brachials are more prominent and are finely spinous. Syzygies occur between the third and fourth brachials, again between the ninth and tenth and fourteenth and fifteenth, and distally at intervals of three or four oblique muscular articulations. Pi is very long, 17 mm. to 11* mm. in length, with from forty-two to forty-five segments, of which the first seven are broader than long, and the distal slightly elongated; P2 is apparently similar, but shorter; P3 1 tears a genital gland. The distal pinnules are exceedingly slender, 17 mm. long. The color in spirits is yellow. Locality. — South of Ras Sharwein, Arabia; or, northwest of Sokotra ( 14° 20' N. lat., 52° 30' E. Long. ) ; 1200 fathoms. Remarks. — The type of this species, which was collected by the cable repair ship Electra, is in the British Museum. The following species should also be referred to this new genus: Cyclometra clio (A. H. Clark). Antedon clio 1907. A. H. Clark, Proc. U. S. Nat. Mus., Vol. 33, p. 79. Habitat. — Southwestern Japan. Depth. — 107 fathoms. Vol. XXIV, pp. 89-90 May 15, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON TWO NEW SPECIES OF EPIMYS FROM LUZON. BY N. HOLLISTER. [Published by permission of the Secretary of the Smithsonian Institution.] The large series of rats of the Epimys concolor group, collected in Luzon by Dr. Edgar A. Mearns for the United States National Museum, is readily divisible into two lots, representing what appear to be distinct species. A small, bright colored form is decidedly nearest to Epimys vulcani (Mearns), described from Mindanao; and a larger, darker form appears to be a local representative of the widely ranging Malayan species Epimys ephippium (Jentink). No rats of this group have been previ- ously named from Luzon, though a number of forms are known from the more southern of the Philippine Islands. Epimys calcis sp. nov. Type from Baguio (Limestone Hills, near Lime Kiln, at 5,000 ft. ), Ben- guet Province, Lnzon, P. I. United States National Museum No. 145,771, skin and skull, young adult cT, molars little worn. Collected May 8, 1907, by Dr. Edgar A. Mearns. Original No. 6412. General characters. — A small member of the concolor-ephip/rium group; pelage soft at all seasons. Nearest to Epimys vulcani, from which it diners in its generally paler coloration, with the sides and underparts especially lighter; and in a few slight average cranial characters. Color. — General tone of upperparts dark russet, finely mixed with black ; shading to a dark buff on sides and a pale cream on belly. The underparts of some specimens almost primrose yellow. Cheeks like sides; feet dirty white; tail brownish, slightly paler beneath. May and Decem- ber skins are virtually alike. Compared with E. vulcani, which it most resembles, it is lighter on the sides, the light color of underparts reach- ing nearer to back with outsides of legs colored like belly instead of like upperparts as in vulcani. Skull like that of E. vulcani but averaging slightly smaller, with more slender rostrum ; incisive foramina smaller, shorter and narrower. 21— Proc. Biol. Soc. Wash., Vol. XXIV, 1911. (89) 90 Holllster — Two Neiv Species of Epimys from Luzon. Measurements of type. — Total length, 253 mm. ; tail vertebrae, 130; hind foot, 26. Skull: Condylobasal length, 20.1; zygomatic breadth, 15; palatal length, 16.4; interorbital constriction, 4.fi; greatest breadth of braincase, 13.4; length of nasals, 11.0. Remarks. — This rat is quite different from Epimys negrinus of Negros Island. It is much lighter colored and has a more slender skull with conspicuously smaller teeth. Curiously enough it seems much more closely related to Epimys vulcani of Mindanao. The series from the type locality contains nineteen specimens, collected in May and December. Epimys querceti sp. nov. Type from Haights-in-the Oaks, Benguet Province, Luzon, P. I. Alti- tude 7,000 feet. United States National Museum Xo. 145,833, skin and skull, adult d\ molars considerably worn. Collected August 1, 1907, by Dr. Edgar A. Mearns. Original No. 6491. General characters. — Nearest in appearance to true Epimys ephippium but averaging decidedly larger, with longer tail. Color. — Back and sides mixed with black and cinnamon, giving a general tone of dark umber; very slightly paler on sides; outside of legs like back; feet dirty white, almost gray. Tail dark brown, usually uni- color, though a few specimens show a paler color near base below. Skull essentially like that of E. ephippium but averaging slightly heavier. Measurements of type.— Total length, 293 ; tail vertebra2, 160 ; hind foot, 28. Skull: Condylobasal length, 30.7; greatest breadth of braincase, 13.7; palatal length, 17.2; interorbital constriction, 5.2; length of nasals, 11.6; length of upper molar series, alveoli, 5.1. Average flesh measure- ments often adult specimens from the type locality and often adults of Epimys ephippium from Borneo and Sumatra, the latter in parentheses : Total length, 270.3 (258); tail vertebra?, 143.8 (130.2); bind foot, 27.4 (25.8). Remarks. — This form is readily distinguishable from Epimys calcis by its larger size and darker coloration. It diffei'S from E. negrinus much as does E. calcis, having smaller teeth and a somewhat more slender skull. It also has much coarser pelage. It is more closely related to true E. ephippium than are the other described Philippine forms of the group, but its large size and long tail are specific characters of sufficient value. Specimens examined, fifty-one, all from near the type locality. Vol. XXIV, pp. 91-96 May 15, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON BREEDING HABITS OF THE VIVIPAROUS FISHES GAMBUSIA HOLBROOKII AND HETERANDRIA FORMOSA. BY WILLIAM P. SEAL. The two species under observation belong to a well-defined group of small fishes the members of which are variously known as ' ' Top Minnows, " " Mud Minnows, " " Pike Minnows, " " Kil- lifishes," etc. The generic name Gambusia was suggested to the late Professor Poey by the common name in use in Cuba, Gambusino, meaning small and of no importance. The mem- bers of fhis genus are larger than those of Heterandria and have a more northerly range; the females, however, though usually twice as large as the males, rarely reach a length of two and one half inches. Until the discovery by Dr. Smith that the little fish named by him Mistichthys reached the adult stage at less than an inch in length Heterandria was about the smallest known vertebrate. According to Garman the range of Gam- busia holbrookii is from Virginia to Alabama and is represented in the tributaries of the Mississippi and westward to Mexico by its close ally G. patruelis. Heterandria fomiosa is abundant in swamps and ditches coastwise from the Carolinas to Florida. It- rarely exceeds an inch in length. Gambusia was introduced into New Jersey in 1905 and from observations made by me of both it and Heterandria since 1906 the habits of the two are, so far as discernible, exactly similar. They are the most northerly representatives of the viviparous genera of the family Poeciliidae. They are easily kept in small glass jars where they breed readily, eating anything that a gold- fish will eat, either vegetable or animal, and as insect destroyers have no superiors. 22— Proc. Biol. Soc. Wash., Vol. XXIV, 1911. (91) 92 Seal — Breeding Habits of Viviparous Fishes. The only recorded observation of the breeding habits of Gam- busia is to be found in the proceedings of the XL S. National Museum for 1885, credited by the late Prof. John A. Ryder to A. A. Duly an employee of the Museum. This account appears also as a footnote to Bulletin No. 47, U. S. Nat. Mus., Fishes of North America. Having observed the act of coition and of the extrusion of the young of both of these species a great many times every year, beginning with 1906, I am satisfied that Mr. Duly's account of it is, in many respects at least, incorrect. First, he says, " In coitus the male's head is turned in the direction of the tail of the female, the prolonged anal fin seeming to be thrust into the external opening of the ovarian duct or genital pore of the female, which lies just in advance of the anal fin." I have never witnessed anything like that, nor is it an analogical method. On the contrary the male follows incessantly and warily after the female, on the left side and to the rear, the female frequently turning and making savage dives at him causing him to turn and flee, but to return immediately and follow, watching for a moment when her attention will be dis- tracted when he will make a sudden dash, sometimes .succeeding in inserting the intromittent organ into the genital pore, but oftener, apparently, missing because of a quick turn of the female from which he flees in apparent terror. The contact is so sudden and brief that it required many observations to verify it. In these movements the male organ is thrust forward and to the right toward the female. In small jars the males are frequently killed, especially when the female is full sized, or if there are two or three females to one male. It has been noted that when catching them there appears to be a great preponder- ance of females, but that might be due to the fact that as the males are much smaller than the females they could more easily escape through the meshes of a net. In the aquarium the males are continually engaged in a pur- suit of the females while the females are apparently adverse to sexual dalliance and at all times unwilling participators and quick to resent the advances of the males. I have never wit- nessed anything to indicate a reciprocity of desire in coitus it being always a chance touch and go on the part of the males. There is never more than the one male following a female. If others approach, the male turns and drives them off. Seal — Breeding Habits of Viviparous Fishes. 93 Of course, normal action in fishes confined in aquaria is de- pendent on a satisfactory environment — warmth, proper food, etc. If conditions arc unfavorable and they are not comfortable there will he no sexual demonstration. Prof. Ryder says further: ' The young, when born, are stated by Mr. Duly to be about f of an inch in length and to be ex- pelled in a single mass, consisting of 8 to 11 young fishes at a single effort. This mass as soon as it escapes is seen to be composed of the infant Gambusia, which at once separate and move away. No membrane seemed to be expelled together with the mass of young, so that it is probable that in this species, as in Anableps and the Embiotocidse the foetuses rupture the follicles in which they developed a short time before birth." One has only to open a gravid female to see that Mr. Duly's account of the birth of Gambusia would be a physical im- possibility. The ova of a full sized Gambusia are, when fully developed, about an eighth of an inch in diameter, transparent and non-adhesive. Each one is held, apparently, by a thread of membrane to a central nucleus the character of which could only be determined by microscopic examination. The young fish can be seen fully formed, their eyes moving as they turn around in the egg. They are expelled one at a time and the ejection of each fish is so rapid that they appear as though shot out with some force. This, however, might be due to the bursting of the follicle and the uncoiling of the fish as it is released from restraint. When they first appear they are still in a somewhat curved form but they quickly straighten out and swim into hiding. The follicles are undoubtedly ruptured at the moment of extrusion, whether inside or out, 1 never suc- ceeded in observing, but it appears the more probable that it is inside. The young fish are already pigmented and altogether unlike the generality of embryo fishes. Oviparous species, when hatched, having an umbilical sac and being for some time unable to feed and quite helpless. They are also transparent and very delicate, while the young of the viviparous fishes are apparently, except in size and the development of the male sexual organ, as fully developed as the adult. I would say that tV to i inch is nearer the size of the fish when born than f inch, although the size, no doubt, varies with the size of the female. The intervals between the extrusions vary from several 94 Seal — Breeding Habits of Viviparous Fishes. minutes to as many hours. The number given off at one time varies also. After the extrusion of such young as have been developed, there is a period of rest of several days during which another lot is developed, a limited number of eggs only being fertilized at each sexual contact. I can not do better than to copy a series of notes taken from my observations. 1906. July 14 to 16. Heterandria female extruded 16 young. Some appeared on 14th, some on 15th, and some on 16th. Last previous delivery from same female was seven young. Three prior deliveries at intervals not noted, but not greater than the last mentioned one, resulted in the delivery of five, seven, and seven, respectively. It is impossible to say whether those num- bers represent accurately the respective deliveries, as some were born at night and at other times when they were not under observation, and some may have been eaten. July 21st (five days later). Found 9 young and on 23d found 2 more. Saw the female eat one of these. This may only occur when other animal food is wholly absent, or in the case of such young as are imperfect — deficient in vitality or deformed in some way. On the other hand there seems to be a universal tendency on the part of the females of all nest build- ing fishes to devour their eggs and young. This may also be characteristic of females of viviparous species. The writer has recorded the action of male sticklebacks in frightening their young by darting at them as though to devour them, causing them to rush to cover among the plants. This is undoubtedly a continuance of the protective habit from the eggs to the young until they are capable of self protection. August 1st (8 days interval), found 4 young. August 2d, found 12; August 4th and 5th, 2 more, making 18 in all. August lOth-llth (5 days interval), found 4 more. August 18th (7 days interval), found 10. These fish grow and develop very rapidly if properly fed. I had paired two that had been hatched early in May. On July 27th I found one young in the jar but at first supposed it must have been in the jar hiding among the plants when the breeders were introduced, as I could not conceive of two genera- tions of fish in one season, it being so contrary to the habits of fishes in general. But on August 18th I found 6 young about which there could be no doubt. Seal — Breeding Habits of Viviparous Fishes. 95 August 26-27, found 5 more. September 3d to tOth found several more. Also found several young with a pair hatched early in July. It is thus positively demonstrated, not only that there can be two or more generations of these species horn in a summer, but also a second generation in less than three months, so that although the fecundity is not so great as with oviparous species there is more rapid reproduction as well as protection both through the egg stage and by reason of the greater development and activity of the young when first hatched, the danger period with the others. This will account for the great numbers of these species where they abound, notwithstanding they are the prey of all the larger species. I have repeated these observations several times with both species with the same general results — any difference being in the varying numbers produced, which, however, were a close approximation . Notwithstanding that these two species live together appar- ently in perfect harmony there is sexually an apparently rigid racial antipathy. I have succeeded a number of times in hybridizing a male Gambusia with a female Heterandria, but not the reverse, probably because of the very small size of the males of the latter species. The offspring of these crosses had the appearance of Gambusia, none of the conspicuous markings of Heterandria appearing on any of them. They were fertile and reproduced, but the young also bad the appearance of Gambusia. This fact would have caused me to suspect a mistake somewhere in the course of the experiment if I had beard or read of it, but I know that there was no possibility of error because they were isolated in small jars with no change or interference. If it were not for such an antipathy, which appears to be confined to the females, the species would soon be intermixed. The attitude of the female Heterandria toward the male Gam- busia was even more hostile and savage than that of females of the same species. She would not only repulse him savagely but at times would follow him biting him viciously until he succeeded in hiding from her. Notwithstanding these vicious rebuffs he would soon be back, following persistently at the left and rear, rushing and retreating all through the hours of day- light at least. 96 Seal — Breeding Habits of Viviparous Fishes. The female Heterandria is at least double the bulk of a male Gambusia, though not of much greater length, and much more than a match for him. In the attacks of the females of either species they seem to endeavor to bite the long slender organ of the male, which is no doubt the most vulnerable point. This pair were together for two summers and the production of young throughout was approximately the same as that of the pair of Heterandria previously mentioned. Notwithstanding the viciousness displayed by the females with apparently no sign of yielding, there may be a cessation of hostilities during the darkness of night as they appear much more quiet in a dim light. There may be times also when there is a yielding from sexual impulse, but I have never been able to observe it. This is probable and would be analogous to the conduct of higher animals. The male generative organ, as may be seen by the accom- panying figures, is extremely long in proportion, being almost a third as long as the fish. It can be thrown completely in a circle close against the body of the fish as though working on a pivot. Proc. Biol. Soc. Wash.. Vol. XXIV. 1911. I'l \TE I. Fig.l. Fig. 3. Fig. 1. Gambusia holbrookii, male. Fig. 2. Gambusia holbrookii. female. Fig. 3. Heterandria formosa, female. (Much enlarged.) Vol. XXIV, pp. 97-100 May 15, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON GENERAL NOTES. A NEW NAME FOR SOREX MACRURUS BATCHELDER. In 1896 I described in these Proceedings (Vol. X, pp. 133-134, Decem- ber 8, 1896) a new species of shrew from the Adirondack Mountains, New York, giving it the name of Sorer macrurus. Through the kindness of Mr. Gerrit S. Miller, Jr., I recently have learned that this use of the name is invalidated, owing to the fact that it was applied in 1822 by J. G. C. Lehmann (Observationes Zoologicae praesertim in Faunam Hamburgensem. Pugillus Primus. ) to the common European water shrew, Neomys fodiens (Schreber). Lehmann's type locality was Sach- senwald, near Friedrichsruh, Sehleswig-Holstein, Germany. It therefore becomes necessary to give another name to the Adirondack shrew, and I propose to call it Sorex dispar. — Charles Foster Batchelder. >* &' NOTE ON THE OCCURRENCE OF THE WHALE SHARK, RHI- NODON TYPICUS, IN THE PHILIPPINE ISLANDS. In the issue of the Philippine Free Press for September 10, 1910, there is published a photograph, with brief description, of a marine monster from Negros Occidental. The photograph was submitted in competition and received the prize for the current week. The newspaper states that two other photographs of the same subject were sent in by other competitors, but the one selected was the best. Throughout the article the animal is referred to as a whale, but the photograph clearly indicates a whale shark, Rhinodon lypicus. The creature was 6 meters long, and was caught in a fish trap near Bacolod, Island of Negros, on September 4, 1910. This species, so far as known, has not heretofore been reported from the Philippine Archipelago, although its distribution is world-wide.* Outside of the Indian Seas it occurs only as a rare straggler. There have been several records from the Dutch East Indies, so that the capture of a specimen in the Philippines is not unexpected although none the less interesting. — Hugh M. Smith. * See Bean's " The History of the Whale Shark," Smithsonian Miscellaneous Collec- tions, vol. xlviii, 1905. 23— Proc. Biol. Soc. Wash., Vol. XXIV, 1911. (97) - 98 General Notes. OWEN'S RECENT ENCRINITE IDENTIFIED. In 1862 Sir Richard Owen wrote (Ann. and Mag. Nat. Hist. (3), vol. 9,* p. 486) "I learn from a correspondent at Melbourne, Mr. J. S. Poore, that during his visit to King George's Sound, Western Australia, he there dredged up from 8 fathoms a living Encrinite. The stem, which was attached to a stone, was about 6 inches long; the arms about 1^4 inch, of a beautiful rose-colour or pink, fading to white." Dr. P. H. Carpenter, commenting upon this, says (Challenger Reports, vol. 11, Zoology, p. 428) " This may perhaps have been a Pentacrinoid larva, but if so, it was of most unusual size." It seems most probable, however, that this was really a small Umbellularian ; one of Kolliker's figures of the young of Umbellularia carpenteri, taken by the Challenger in the seas southwest of Australia, shows an animal sufficiently like a crinoid to deceive even a fairly skilled zoologist, and of the size described by Owen. The color as given is certainly suggestive of an Umbellularian, and, moreover, does not occur in any of the small stalked crinoids, nor in any pentacrinoid larvae. — Austin H. Clark. NOTE ON AGARNA CARINATA SCHICEDTE AND MEINERT. Recently in referring to Haller's paper "Ueber einige neue Cymothoi- nen,"* my attention was called to a new species described therein as Livoneca cumulus from Guadeloupe. The species described later by Schicedte and Meinertt as Agarna carinata from the West Indies is un- doubtedly identical with this form. The correct name for the species is therefore Agarna cumulus (Haller). — Harriet Richardson. SC1URUS POLIOPUS LYON CHANGED TO SCIURUS CONIPUS. Sciurus poliopus Lyon (Proc. U. S. Nat. Mus., vol.40, p. 88, April 25, 1911) applied to a gray-footed plantain-squirrel from southeastern Borneo is preoccupied by Sciurus variegatus poliopus Fitzinger (Sitz. k. k. Akad. wiss. Wien, math.-naturw. CI. vol. 45, Abth. 1, p. 478, March, 1867), applied to a Mexican squirrel. The former may be called Sciurus conipus. — Marcus Ward Lyon, Jr. CAPTURE OF SOREX DISPAR IN WEST VIRGINIA. This rare shrew, originally described as Sorex macrurus and since re- named dispar,i has previously been known only from 10 specimens taken in the Adirondack and Catskill Mountains of New York. The capture of a specimen in West Virginia is, therefore, quite unexpected and of great interest. The specimen referred to I trapped on July 24, 1909, on the cool north slope of Winding Gulf, about four miles southwest of * Archiv-fur Natur-Geschichte, 1880, 46th year, Vol. I, pp. 383-386, 391, figs. 13-15. + Naturhistorisk Tidsskrift (3), Vol. XIV, 1883-84, pp. 328-334, pi, xni, figs, 1-6, J See page 97, this volume. General Notes. 99 Pemberton, Raleigh County, at an altitude of about 2,200 feet. Thesteep slope of the canon is heavily forested with hemlock and a variety of deciduous trees, with a dense undergrowth of rhododendron, striped maple (Acer pennsylvanicum) , purple-flowering raspberry (Rubacer odo- ratum) and other shrubs. The soil is wet and spongy, with a carpet of mosses and ferns growing among rocks and bowlders. On this cool, wet, Transition Zone hillside I found shrews very abundant, taking in one nighl '.* Blarina brevicauda, 3 Surer fumeus and 1 Sorex dispar. The latter appears to lie in every way identical with specimens taken by Dr. Mearns in the Catskill Mts., New York. It is an adult male and meas- ured in the flesh: total length, 131 mm.; tail vertebrae, 62mm.; hind foot, 14mm. — Arthur If. Howell. Vol. XXIV, pp. 101-102 May 15, 1911 PROCEEDINGS OF Til E BIOLOGICAL SOCIETY OF WASHINGTON NEW NAMES FOR TWO SUBSPECIES OF PEROMYSCUS MA NICULA TUS ( W AG N ER ) . BY EDGAR A. MEARNS. Mr. Gerrit S. Miller, Jr., has drawn my attention to the fact that the names applied by me to two forms of the genus Peromyscus, defined in 1890, were preoccupied, and are there- fore not available. The name Hesperomys leucopus arcticus Mearns ( 1890 )* must be consid- ered as preoccupied by " [Hesperomys] arcticus, Sauss." of Coues 1877, t in accordance with "opinion 4," rendered by the international commission on zoological nomenclature,* which affirms that "manuscript names acquire standing in nomenclature when printed in connection with the provisions of Art. 'S>,§ and the question as to their validity is not influenced by the fact whether such names are accepted or rejected by the author respon- sible for their publication." In this case it is evident that the author did not intend to set up arcticus as the name of a valid form, as, indeed, he expressly stated that he considered it to show not the slightest departure from the ordinary form of leucopus. The name arcticus can not be regarded as a noma/ nudum, as Coues gave a complete history, with no less than eight measurements of specimen No. :iu24. Moreover, as the animal from Labrador to which Coues attached the name arcticus, in 1877, is a synonym of Hesperomys maniculaius Wagner, 1845, the name arcticus disappears from nomenclature || and can not be used for the dif- ferent form from Fort Simpson, Mackenzie, Canada, which I described, in 1890, under the name Hesperomys leucopus arcticus ; for the latter I now propose the name borealis. •Bull. Am. Mus. Nat. Hist., N. V. II, 1890. p. 285. tMonogr. N. Am. Rodentia, 1877, p. 67 (in text) and p. 61 (where the name is mis- printed articus), table xi, line l (.Labrador). t Opinions rendered by the International Commission on Zoological Nomenclature. Opinions 1 to 25, publication No. 1938 of the Smithsonian Institution, July, 1910. § International Code of Zoological Nomenclature, 1905, p. 23. II" [Hesperomys] bairdii Sauss.", which Coues mentions in the same connection, also from Labrador, has the same status as arcticus; it being a young specimen of Wagner's maniculatus (Not Mus bairdii Hoy and Kennicott, 1856) 21— Proc. Biol. Soc. Wash., Vol. XXIV, 1911. (101) 102 Mearns — Names for Subspecies of Peromyscus maniculatus. Baird's Hesperomys sonoriensis nebrascensis* is a nomen nudum, though it is evident from a comparison of the text and synonymy on pages 4(32 and 474 of his Mammals of North America that he at first intended to describe this race as a form of sonoriensis, but that, later, he changed his opinion respecting its validity or possibly omitted it through oversight ; but Coues repeated the namef, and referred to specimens from Deer Creek, Nebraska, giving measurements and a critical remark on coloration. The name will therefore have to date from Coues. This Nebraskan form was redescribed by Osgood as Peromyscus luteus.i The Hesperomys leucopus nebrascensis Mearns §, from Calf Creek, Custer County, Montana, is subspecifically distinct from Hesperomys sono- riensisv&r. nebrascensis Coues. from Deer Creek, Nebraska, and as it must receive a new name, I propose osgoodi as a subspecific appellation. The synonymy of the three forms involved stands as follows: Peromyscus maniculatus borealis, new name. ARCTIC DEER MOUSE. Hesperomys leucopus arcticus Mearns, Bull. Am. Mus. Nat. Hist. N. Y., II, 1890, p. 285 (not of Coues). Fort Simpson, Hudson's Bay Ter- ritory, September 7th, R. Kennicott. Peromysriix maniculatus arcticus Osgood, N. Am. Fauna, No. 10, 1000, p. 33. Peromyscus maniculatus nebrascensis (Coues). NEBRASKA DEER MOUSE. ILsperomys sonoriensis var. nebrascensis Baird, Mamni. N. Am., 1857, p. 402 (noun, i a win in). Hesperomys sonoricnsisxav. nebrascensis ( !oues, Monogr. N. Am. Rodentia, 1877, p. 70, synonymy under Hesperomys leucopus sonoriensis, p. 28 (critical remark on coloration ), p. 80 ( mentions specimens from Deer Creek, Nebraska). Peromyscus nebrascensis of authors (not of Mearns). Peromyscus 1 ulcus Osgood, Proc. Biol. Soc. Washington, XVIII, 1005, p. 78 (Kennedy, Nebraska i. Peromyscus maniculatus luteus Osgood, N. Am. Fauna, No. 28, 1909, p. 77. Peromyscus maniculatus osgoodi, new name. BLACK-EARED DEER MOUSE. Hesperomys leucopus nebrascensis Mearns, Bull. Am. Mus. Nat. Hist., N. Y. 1800, p. 285 (not Hesperomys sonoriensis var. nebrascensis Coues, 1877). Calf Creek, Custer County, Montana. Peromyscus texanus nebrascensis Allen, Bull. Am. Mus. Nat. Mist., N. Y., VIII, 180(5, p. 251 (part). Peromyscus maniculatus nebrascensis Osgood, N. Am. Fauna, No. 28, L909, p. 75 (based on Mearns). * Maimn. N. Am.. ls.",7 . p. 462. I Monogr. X. Am. Rodentia. 1*77, pp. 28, 7;i and so. I Proc. Biol. Soc. Washington. 1905, p. 7s (Kennedy, Nebraska). § Bull. Am. Mus. Nat. Hist., N. Y., 1890, p. 285. Vol. XXIV, pp. 103-160. June 16, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON bo NOTES ON GENERA OF PANICEAE. IV.* BY AGNES CHASE. The work upon the genera of Paniceae as originally outlined contemplated a consideration bjr groups. It was the intention of the writer to prepare as a final paper a synopsis of the entire work, in a sequence representing, so far as possible in a lineal series, the natural relationship of the genera investigated. This synopsis with special reference to American genera together with notes on the North American genera are offered in a pre- liminary way at this time at the request of Mr. G. V. Nash, of the New York Botanical Garden, who wishes to cite the refer- ences in his forthcoming treatment of the tribe Paniceae in the North American Flora. PANICEjE. Spikelets ideally 2-flowered, the terminal floret perfect, the lower stami- nate or neuter (perfect in Isachne and Dissochondrus), its glume (the sterile lemma) unlike the flowering glume (the fertile lemma) in form and texture and simulating a third empty glume, a membranaceous or hyaline sterile palea (indurated at maturity in Ixophorus and in a few species of Panicum) present or wanting, the spikelet or at least the fruit (the fertile floret) dorsally compressed (laterally in Lithachne); fertile lemma and palea alike in texture, indurated, or at least tinner than the glumes and sterile lemma, firmly clasped together (open at the summit in Leptocoryphium and Hymenaehne) inclosing the free grain (in the cultivated Pennisetum americanum the enlarged grain forcing open the lemma and palea), awnless (sometimes mucronate, acuminate-pointed or, in Coridochloa, AUoterupsis, and species of Eriochloa, evenshort-awned), the nerves obsolete (present in Reimarochloa and Acritochaete) . * Notes on Genera of Paniceae I appeared in Proc. Biol. Soc. Wash. 19 :183-192. 1906; same II, op. cit. 21 : 1-10. 1908: same III, op. cit. 21 : 175-188. 1908. 25— Proc. Biol. Soc. Wash., Vol. XXIV, 1911. (103) ••■".- 104 Chaje — Notes on Genera of Panicex. IV. In the following synopsis and diagnoses this ideal spikelet is always in mind, the same name being used for corresponding parts, even when the first glume is wanting as in most species of Paspalum or when both glumes are wanting as in Reimarochloa and in a few species of Paspalum, the scale which appears to be a glume in the latter case, being the sterile lemma. The type species of each of the genera here recognized, except Thrasya, Odontelytrum and Dissochondrus, and of most of the generic names referred to synonymy, is represented in the National Herbarium by specimens or in a few cases by spikelets contributed from type specimens. Synoptical Key. Spikelets all alike (the lower of each pair aborted in Echinolaena). Spikelets neither sunken in the cavities of a spongy rachis, nor surrounded by an indurated leaf-sheath nor involucre-form bracts. Spikelets not subtended nor surrounded by bristles (sterile branchlets ) , ( axis of branchlet extending beyond the base of the uppermost spikelet as a point or bristle in Panicum, subgenus Paurochaetium) . Fruit cartilaginous-indurated, not rigid, papillose, usually dark colored, the lemma with more or less prominent, white hyaline margins not inrolled. ( Fruit not rigid in Hymenachne, Reimarochloa and in a few species of other genera, but texture and margin not as above.) Spikelets awnless. Fruit open at hyaline summit 1. Leptocoryphium. Fruit not open at summit. Lemma boat-shaped, hyaline margins narrow 2. Anthaeuantia. Lemma convex with broad hyaline margin. Fruit lanceolate-acuminate ; second glume and sterile lemma unusually long-silky 3. Valota. Fruit elliptic, pubescence short or none. Inflorescence of slender racemes, digitately or sub- digitately arranged 4. Syntherisma. Inflorescence a capillary panicle .... 5. Leptoloma. Spikelets with slender awns. Second glume reduced to a minute scale; sterile lemma with a straight awn 6. Chloridion. Second glume and sterile lemma with flexuous or tangled awns 7. Acritochaete. Fruit indurated, rigid (or if thin, not hyaline-margined). Alternate spikelets facing in opposite ways, that is the first glume alternately extrorse and introrse; inflorescence a solitary, unilateral raceme, the margins of the con- cave rachis partially embracing the subsessile spike- lets 8. Thrasya. Chast — Notes on Genera of Panlcese. IV. 105 Alternate spikelets not facing in opposite ways (or if s<> the lower of the pair aborted). Spikelets (or the primary one of a pair) placed with the back of the fruit turned away from therachis, usually solitary. First glume as long as the spikelet or nearly so. Spikelets in pairs, the secondary usually aborted, sometimes wanting, strongly tuherculate-hispid or uncinate at maturity, laterally compressed, in unilateral racemes 9. Echinolaena. Spikelets solitary, not tuberculate nor uncinate. Inflorescence a single spike-like raceme; spikelets swollen on the side toward the axis and fitting into alternate hollows 10. Mesosetum. Inflorescence a narrow panicle; spikelets swollen on the side away from the rachis or at least more convex on that side 11. Leucophri/s. First glume obsolete or not over }£ the length of the spikelet. First glume and rachilla joint forming a swollen ring- like callus below the spikelet, the glume reduced to an adnate sheath of the rachilla joint; fruit mucronate or shortly awn-pointed . . 12. Eriochloa. First glume present or wanting but no ring-like callus below the spikelet. First glume present ; racemes racemose along the main axis 13. Brachiaria. First glume wanting; racemes digitate or sub- digitate 14. Axonopus. Spikelets placed with the back of the fruit turned toward the rachis of the spike-like racemes, or pedicellate in panicles. Fruit long-acuminate, scarcely indurated, both glumes wanting; spikelets sessile, solitary in spike-like racemes, these re flexed or verticil late at maturity 15. Reimarochloa. Fruit not long-acuminate, indurated (if but slightly indurated, both glumes present and inflorescence paniculate). First glume typically wanting; spikelets plano-con- vex, subsessile in spike-like racemes . 16. Paspalum. First glume present; spikelets usually in panicles. Neither glumes nor lemmas awned . Spikelet with a single fertile floret. Second glume and sterile lemma not indu- rated, like the first glume in texture. Glumes not equal nor entirely enclosing the rest of the spikelet. 106 Chase — Notes on Genera of Panicese. IV. Fruit chartaceous-indurated the palea enclosed (rarely the tip free). Second glume not inflated-saccate. Culms not woody nor bamboo like. Fertile lemma neither with lateral appendages nor excavations at base, the inrolled margins clasping the palea; inflores- cence typically paniculate. 17. Panicum. Fertile lemma either with lateral appendages or excavations at base, the margins usually not inrolled; first glume large; blades usually contracted into a petiole-like base . .18. Ichnanthus. Culms usually woody, bamboo-like; spikelets globose, large, the glumes and sterile lemma papery ; fruit bony-indurated, a downy tuft at the apex 19. Lnsiacis. Second glume inflated-saccate, this and the sterile lemma much exceeding the stipitate fruit .... 20. Sacciolepis. Fruit membranaceous, the palea not in- closed above; spikelets lanceolate, subsessile in usually spike-like pan- icles 21. Hymenachne. Glumes equal, enclosing the rest of the dor- sally compressed spikelet; margins of the fertile lemma flat . . . 22. HomoJepis. Second glume and sterile lemma leathery- indurated; spikelets stipitate . . 2.°). Scutachne. Spikelets typically with two fertile florets. Florets alike in form ami texture; panicle compound 24. Isachne. Florets unlike, the lower lanceolate, exceed- ing the upper; panicle racemose. 25. Heteranthoecia. Glumes or lemmas or both awned, or if short- pointed only, the summit of fertile palea not enclosed. Inflorescence of unilateral racemes along a com- mon axis, never digitate; glumes 2-lobed (rarely entire) awned from between the lobes; fruit indurated, the palea enclosed at the summit 26. Oplismenus. Chase — Notes on Genera of Panicea. W. 107 Inflorescence paniculate or of slender subsimple racemes, these digitate. Spikelets never silky-pubescent, nor ciliate- margined, often scabrous or hispid. Spikelets ovate, not having a callus-like Imse 2,7. Echinochloa. ■ Spikelets lanceolate, with a long callus-like base below the long-awned glumes. 28. Chaetium. Spikelets silky pubescent or with a conspic- uously ciliate margin ; fruit subindurated. Second glume and sterile lemma 2-lobed, a slender awn from between the lobes; fruit awnless; first glume remote; in- florescence paniculate . . .29. Tricholaena. Second glume and sterile lemma not lobed ; fruit awned; first glume not remote; inflorescence of digitately arranged, subsimple racemes. Sterile lemma and glumes papery, not at all indurated; fruit stipitate. 30. Coridochloa. Sterile lemma subindurated similar to the fertile lemma 31. Alloteropsis. Spikelets subtended or surrounded by 1 to many bristles or spines (sterile branchlets), these distinct or more or less connate at base, forming a pseudo-involucre. Spikelets deciduous, bristles persistent. Spikelet with lower floret as well as the upper perfect and with indurated lemma and palea 34. Dissochondrus. Spikelet with upper floret only perfect. Sterile palea at maturity becoming cartilaginous and winged, much exceeding the spikelet in width; spikelets secund along the branches of a simple pan- icle, each subtended by a single viscid bristle. 35. Ixophorus. Sterile palea not enlarged at maturity. Second glume and sterile lemma very broad, many- nerved, the glume saccate, auriculate, the lemma lyre-shaped, indurated on the margins ; the spike- let subtended by a single flexuous bristle . 32. Setariopsis. Second glume and sterile lemma not many-nerved, saccate, auriculate nor lyre-shaped; subtending bristles 1 to many 33. Chaetochloa. Bristles falling with the spikelets at maturity (in the culti- vated Pennisetum americanum the globose grain falls from the usually persistent spikelet, with or without the lemma and palea attached). 108 Chase — Notes on Genera of Pamcece. PV . Articulation below the spikelet-bearing branchlets. A single bristle produced beyond the solitary spikelets. First glume present ; second glume and sterile lemma acuminate; fruit not acuminate . . 36. Chamaeraphis. First glume usually obsolete, second glume minute, sterile lemma and fruit long-acuminate . 37. Paratkeria. Bristles numerous below each spikelet or cluster of spikelets. Bristles not united at base, usually slender, often plu- mose 38. Pennisetum. Bristles more or less united at base into a bur-like pseudo-involucre 39. Cenchrus. Articulation at the junction of the primary branches with the main axis 40. Plagioselum. Spikelets either sunken in the cavities of a corky axis or sur- rounded by a sheath or by involucre-form bracts. Spikelets sunken in the cavities of the flattened corky axis, this disarticulating at maturity with the spikelets attached. 41. Stenotaphrum. Spikelets not sunken in a corky axis. Spikelets in small spikes, these surrounded by rigid sheaths. 42. Xerochloa. Spikelets solitary, subtended by two glume-like bracts, these placed cross- wise to the spikelet* 43. Odontelytrum. Spikelets not all alike. Plants monoecious or polygamous, that is with the different kinds of spikelets on one plant. Spikelets all perfect, but those of the aerial panicle not perfect- ing grains; the fruitful spikelets cleistogamous, borne on subterranean branches 44. Amphicarpon. Spikelets not all perfect. Spikelets hermaphrodite and sterile. Spikes several, crowded on a leaf-like axis . . 45. Phyllorachis. Spike solitary 46. Thuarea. Spikelets unisexual. Inflorescence paniculate; fruit bony-indurated. Panicles terminal on culms or leafy branches, pistil- late spikelets above, staminate spikelets below, in same panicle 47. Olyra. Panicles all axillary or axillary and terminal ; the terminal when present wholly staminate. Fruit laterally compressed, conspicuously gibbous on upper dorsum 48. Lithachne. Fruit dorsally compressed, lanceolate . 49. Raddia. Inflorescence of spike-like racemes. Raceme solitary ; spikelets geminate, a pistillate and a staminate forming a pair ... . . 50. Diandrolyra. * Taken from Hackel's description. We have not yet seen the plant. Chase — Notes on Genera of Panicecz. TV. 109 Racemes 2, digitate, one pistillate, the other starninate, the spikelets solitary 51. Mniochloa. Plants dioecious 52. Spinifex. 1. Genus LEPTOCORYPHIUM Nees.* Lrptocoryphium Nees, Agrost. Bras. 83. 1829. Two species are included, L. lanatum Xees, based on Paspalum lanatum H. B. K., and L. molle Xees, the first named being here taken as the type of the genus. 2. Genus ANTHAENANTIA Beauv. Anthaenantia Beauv. Ess. Agrost. 48. pi. 10. f. 7. 1812. Based on a single species, Phalaris villosa Michx. This name is spelled Anthenantia^ by Robert Brown (in Flinders, Voy. App. 2: 3582. 1814) and Athenantia t by Kunth (Mem. Mus. Paris 2:71. 1815). Aulaxanthus Ell. Bot. S. C. & Ga. 1:102. 1816. Two species, A. '■Hiatus Ell. and A. rufus Ell., the first named of which is here taken as the type, are included. The type of A. ciliatus, in Elliott's herbarium, is Anthaenantia villosa (Michx. ) Benth. Elliott refers " Phalaris villosa f Michx." to A. ciliatus. With the later fascicles of Elliott's work an illustration of A. rufus (pi. 6. f. 1), was given, but since this was not published until 1821 the first species published under the genus is taken as the type. Aulaxia Nutt. Gen. PI. 1:47. 1818. This is based on Aulaxanthus Ell., Elliott's description, slightly altered, being used and his species cited. Steudel (Nom. Bot. ed. 2. 1: 171. 1840) spells the name Aulaxis. 3. Genus VALOTA Adans. Valota Adans. Fam. PI. 2:495. 1763. This genus is established by a reference to "Sloan, t. 14. f. 2." This figure identifies the genus with Andropogon insulare L. (Syst. Nat. ed. 10. 2:1304. 1759). The type of the latter in the Linnaean Herbarium is from Jamaica, sent by Browne. After his diagnosis Linnaeus cites "Sloan, jam. t. 14. f. 2." Steudel (Nom. Bot. ed. 2. 2:744. 1841) spells the name Vallota. Acicarpa Raddi, Agrost. Bras. 31. pi. 1. f. 4. 1823. This is based on a single species, A. sacchariflora Raddi, the figure and description of which identify it with Valota insularis. Raddi also cites "Sloan. H. jam. I. p. 43. t. 14. fig. 2)." Trichachne Nees, Agrost. Bras. 85. 1829. Six species are included in the genus, T. insularis Nees which is taken as the type, based on Andro- pogon insulare L., T. sacchariflora (Raddi) Nees, which we now know to •The group containing the first five genera here given was earlier discussed and the fruit of the type species figured (Proc. Biol. Soc. Wash. 19:183-192. 1906). Only a summary of that discussion is here given, with such additional knowledge as has been gained. + These misspellings or changes of spelling are here given only because these names have found their way into synonymy, hence must be accounted for. 110 Chase — Notes on Genera of Panicece. TV. be synonymous with the first, and four new ones, T. recalva, T. tenuis, T. velutina and T. ferruginea. Since the notes on Valota were published* Prof. A. S. Hitchcock has examined the types of T. tenuis and T. recalva in Nees' herbarium at Munich, and duplicate types of T. ferruginea and T. velutina in the Trinius Herbarium. Trichachne recalvais seen to be allied to V. Pittieri ( Hack. ) Chase, while T. ferruginea and T. velutina approach Syntherisma through S. adusta. Trichachne tenuis Nees, upon which Kunth based Panicum tenerrimum (Rev. Gram. 1:39. 1.S20) is most nearly allied to V. insidaris, but has much smaller spikelets with less copious and shorter hairs. None of these species are the one referred to t as repre- sented in American herbaria by Nealley's Texas collections and passing under the name of Panicum tenerrimum Kunth. This very distinct species was collected in the summer of 1910 by Prof. A. S. Hitchcock, and is described below : Valota Hitchcockii sp. no v. Plants perennial, in dense clumps; culms erect, very slender, 30 to 50 cm. high, sparingly branching from the lower nodes, glabrous, usually very leafy, the lower nodes ascending-villous, the upper glabrous; sheaths about as long as the internodes or overlapping, the lower silky-villous, the middle and upper with a few, scattered, delicate hairs or glabrous or silky-ciliate only; ligule membranaceous, scarcely 0.5 mm. long, con- tinued as a brown scarious margin down the sheath ; blades ascending or erect, flat, 2 to 4.5 cm. long, 2 to 2.5 mm. wide, scarcely narrowed to the base, glabrous on the lower surface, minutely puberulent or glabrous on the upper, usually with a few long hairs near the base, and with a thin, white, cartilaginous margin; panicles consisting of few to several ascend- ing racemes rather distant along a slender, glabrous axis, the rachises slender, 3-angled, the spikelets mostly in pairs, one short-pediceled, the other on a pedicel about as long as the lower spikelet, thus forming slender racemes, the spikelets scarcely imbricated; spikelets whitish or purplish, 2.5 to nearly 3 mm. long, 0.7 mm. wide, or with the hairs expanded about 1.2 mm. wide; first glume less than A the length of the spikelet, obtuse, glabrous; second glume and sterile lemma equal, strongly 7- to 9-nerved, the internerves and margins densely silky hairy, the hairs at first appressed, at maturity spreading and matted, the spike- lets often matted together by the tangled hairs; fruit 2.1 to 2.2 mm. long, 0.6 mm. wide, abruptly short-acuminate, brown, the broad, white, hya- line margins of the lemma nearly covering the palea except at the base. Type collected June 24, 1910, on dry prairie soil, San Antonio, Texas, by A. S. Hitchcock (no. 5329). Valota Hitchcockii is most nearly related to P. saccharata (Buckl.) Chase, from which it is distinguished by its lower stature, short blades, short racemes of smaller spikelets, the sterile lemma densely hairy on all the internerves, while in V. saccharata the middle four internerves are *Proc. Biol. Soc. Wash., 19:186. 1906. + Op. cit. p. 188. Chase — Notes on Genera of Panicetz. IV. Ill glabrous, the hairs of the lateral intcrnerves and margins and of the second glume being also much longer than in V. Hitchcockii ; the fruit of the latter is much smaller, less acuminate and the lemma more broadly margined than in V. saccharala. This species is known only from Texas. It was also collected by G. C. Xealley, at Sanderson, Pecos County, in September, 1892, and at Del Rio, October, 1892, both collections being distributed as no. 109. Until the South American species referable to Valota are better known the transfer of Nees' species of Trichachne to this genus is deferred. 4. Genus SYNTHERISMA Walt. SanguineUa Gleichen, Mikrosk. Untersuch. 4. pi. 8. 1764. This includes a single species with a pbrase name wbich is not directly asso- ciate with a previously published binomial, hence the genus is not technically published. The plate is a crude colored representation of Syntherisma sanguinalis. Digitaria Haller, Stirp. Helv. 2:244. 1768, not Adanson, 1763. Haller includes two species, the first of which is Linnaeus' Panicum sanguinale , this name, however, being omitted and Linnaeus' diagnosis used instead as a phrase name. Syntherisma Walt. Fl. Carol. 7(i. 1788. Three species are included, the first of which, S. praecox Walt., is taken as the type. Though there is no specimen of this now in Walter's herbarium,* there is little doubt that it the same as Panicum sanguinale L. as stated by Michaux f and by Elliott, t Gramerium Desv. Opusc. 61. pi. 7. f. 1. 1881. This includes a single species, G. convolutwm Desv. We have not seen Desvaux's specimen but the description and figure apply well to the South American allies of Syntherisma adusta (Nees) Chase, and may possibly refer to a small specimen of that species itself. Sanguinaria Bubani, Fl. Pyren. 4 : 256. 1901. Four species are included, the first of which, here taken as the type, is S. nevenarae Bub. From the synonyms and illustrations cited this is seen to be S. sanguinalis. 5. Genus LEPTOLOMA Chase. Leptoloma Chase, Proc. Biol. Soc. Wash. 19 : 191. 1906. Based on Panicum cognatum Schult., L. cognatum (Schult. ) Chase. 6. Gems CHLORIDION Stapf. § Chloridion Stapf in Hook. Icon. PI. 27 : 2 pi. 2640. 1900. A single species, C. cameroni Stapf, from " British Central Africa : * * * Namasi, • For an account of the grasses in Walter's herbarium see Hitchcock, Sixteenth Ann. Rept. Mo. Bot. Gard. 44. 1905. + F1. Bor. Amer. 1 : 45. 1803. 1 Bot. S. C. & Ga. 1 : 131. 1816. &It is intended to discuss in detail this and other extra -American genera in a later paper, when the spikeletsof the type species will be figured. 112 Chase — Notes on Genera of Panicea. TV, Cameron, 15 (coll. of 1899)," is, included in the genus. The type has not been examined but the description and plate identify the species with the following. Stereochlaena Hack. Proc. Rhodesia Sci. Assoc. 7 : 2 65. 1908. One species, S. Jeffreyssii Hack., " Bulawayo, Maio, 1907, leg. Jeffreys, no. 46, 83." Professor Hackel kindly sent spikelets of the type for deposit in the National Herbarium, and stated in a letter that Stereochlaena " is identical with Chloridion Stapf, Hook. Ic. 2640 (1900)." In this genus the fruit is cartilaginous-indurated, brown, papillose, the margins of the lemma white and hyaline, as in the other genera of this group. The inflorescence consists of few to several slender racemes, digitate at the summit of the culm ; the short-pediceled, lanceolate, spikelets are in pairs, the first glume is wanting, the second a minute scale; the sterile lemma is attenuate into a slender scabrous awn, 4 or 5 times longer than the body of the spikelet, and encloses a minute cleft palea. Stapf says " Chloridion might be described as a Digitaria in which the lower glume is entirely suppressed and the upper reduced to a scale, whilst the lower (barren) valve runs out into a fine bristle-like awn." 7. Genus ACRITOCHAETE Pilger. Acritochaete Pilger in Engler, Bot. Jahrb. 32 : 53. 1902. The genus is based on one species, A . Volkensii Pilger ( op. cit. 54 ) ' ' Ost-Afrika, am Kili- mandscharo, im Griitelwald oberhalb Marangu verbreitet, 2100 m. s. m. ( Voi.kens n. 1278. ) " A portion of a raceme from the type specimen was kindly sent to the National Herbarium by Dr. Pilger. This genus like the preceding is very unlike any known American grass. In Acritochaete the scarcely indurated fruit is more like that of Reimaro- chloa or Hymenachne than like that of the preceding genera in texture, but the broad, white, thin-membranaceous margins of the lemma are like those found in this group. Chloridion is evidently related to Syntherisma, and Acritochaete is more nearly related to Chloridion than to any other known genus. The inflorescence consists of a few slender erect, distant racemes; the solitary, short-pediceled, lanceolate spikelets in two rows on one side of the slender raceme ; the first glume is reduced to an obscure scale, the second glume and sterile lemma are attenuate into long, slender, more or less twisted and irregularly flexuous awns several times longer than the body of the spikelet, the awns of the several spikelets more or less entangled. 8. Genus THRASYA H. B. K. Thrasya H. g. K. Nov. Gen. & Sp. 1 : 120. pi. 39. 1816. This is based on a single species, T. paspaloides (op. cit. p. 121). " Crescit in calidis, subinundatis insulae Orinocensis Panumana, inter vicos Atures et San Borja." The type has not been examined nor has any specimen of this species. The generic characters here given are based upon a study of specimens of Panicum thrasyoides Trin. (Thrasya hirsuta Nees) and P. Chase — Notes on Genera of Panicece. IV. 113 petrosum Trin. and the other two species here listed. Judging from Kunth's description and plate 39, and allowing for evident error in each, Thrasya paspaloides is closely related to P. thrasyoides Trin. Kunth places his Thrasya as the last genus of " Sectio I. Paniceae" (ahout the equivalent of our subfamily Panicoideae) immediately following M anisuris granulans (Hachiocldoa). He diagnoses Thrasya as being 2-flowered, having two glumes, "superior profunde bipartita," the halves aristate below the apex, " inferior [gluma] integra mutica." The male floret is said to have but a single palea [both valves of a floret being termed paleae] . From dissections of spikelets of P. thrasyoides and from plate 39 the follow- ing conclusions are reached: Knnth overlooked the small, hyaline first glume (which in P. thrasyoides is more or less buried in the cleft of the sterile lemma and might easily escape notice); his entire, awnless, lower glume is the second glume; his deeply divided upper glume, the sterile lemma, the aristae of the halves being not awns (prolongations of fibro- vascular bundles) hut stiff, cpiill-like hairs; the single palea of the male floret is the sterile palea. It is strange that Kunth, failing to note the first glume, did not, nevertheless, count it as obsolete, since failing to do so, and counting the second glume as the first, he describes a spikelet in which the scales are not distichous, for his male floret is immediately above the (supposed) second glume instead of on the opposite side above the first. The margins of the sterile palea (in the allied P. thrasyoides) are so narrow and so readily torn from the hyaline middle portion that the fact that they turned toward the supposed second glume might escape observation. Kunth describes this " flos masculus " as " subtrinervia " — being a palea it has no midnerve, where Kunth evidently looked for one. But even if this [supposed] incongruous structure of the spikelet escaped him (if it did escape him) Kunth considered his Thrasya a most curious grass. He states that though it resembles Paspalum platycaule in habit, the structure of the flowers is so singular as to he widely different from that of all known genera. Nees (Agrost. Bras. 93. 1829) in his diagnosis of Thrasya notes a minute, scale-like inferior glume " (a cl. Kunth neglecta.)" The further important character that Kunth neglected, that is the alternately reversed position of the spikelets on the axis, Xees takes note of in his specific description of Thrasya hirsuta (based on Panicum thrasyoides Trin. ), though Trinius failed to do so. Description. — Inflorescence a single, terminal, more or less arcuate, spike-like raceme ; spikelets apparently suhsessile and solitary, in a single row on one side of a channeled, more or less winged rachis (the wings partially embracing the row of spikelets), hut actually in pairs (the spikelets of each pair back to back) the pedicel of the primary spikelet adnate to the midnerve of the rachis* (the spikelets spreading from the * To account for the alternation in position of spikelets. this was the interpretation arrived at by the writer from an examination of P. thrasyoides and P. petrosum, but I should not have ventured to give it at this time had I not found that it is the conclusion of Prof, Hackel (Oesterr. Bot. Zeitschr. 51 : 368. 1901) in the case of Panicum campylo- stachyum, in which species the paired arrangement is more evident than in P. thrasyoides and P. petrosum. 114 Chase — Notes on Genera of Panirea'. IV. rachis at an angle of about 45 degrees), alternate spikelets placed with the back and alternate spikelets placed with the face toward the rachis (that is, the first glume alternately introrse and extrorse), a short callus below the spikelet ; first glume minute, often hyaline and nerveless ; second glume membranaceous, shorter than the spikelet; sterile lemma subin- durated, thinner down the middle and early splitting to the base (the margins of the split rolling inward ) or deeply sulcate only, the sterile palea nearly as long as its lemma, the nerves and margins firm, the broad internerve very thin, a staminate flower present or wanting ; fruit oblong-elliptic, subacute, the lemma and palea cartilaginous-indurated, papillose, the summit of the lemma often clothed with stiff hairs, the thin margins flat, more or less pubescent. Slender, branching perennials, with narrow leaves, the genus confined to the tropics of the mainland of the western hemisphere. The strictly racemose inflorescence, the alternation in the position of the spikelets, the subindurated, split or deeply sulcate sterile lemma, and the cartilaginous ( not chartaceous ) indurated lemma and palea, the thin margins of the lemma flat, taken in combination, are here used to distin- guish Thrasya from Panicum and Paspalum. Thrasya approaches Pas- palum through Thrasya cultrata and Panicum campylostachyum ( in which the fruit is not hairy at the summit) on the one side and Paspalum ■monostachyum (H. B. K.) Vasey (in which the slightly indurated sterile lemma is somewhat sulcate and readily splits in dissection, and the firm margins of the papillose fertile lemma are not inrolled), on the other side. In the latter species, however, the paired spikelets are in two rows, and not in the alternately reversed position of those of Thrasya. If we con- ceive of a genus as bounded by an orbiculate line in places farther from, in places nearer to, the center (the species conceived to be the type of the genus) we have Paspalum monostachyum close to the line in one circle and Panicum, campylostachyum and Thrasya cultrata close to the line in the adjoining one, the theoretical common ancestor of both long extinct. In such cases of extra-generic affinity it seems to be the nearest approach to a natural arrangement if we place the anomalous species in the genus to the members of which it is on the whole most nearly allied. And Paspalum monostachyum finds its nearest allies in Panicum unispicatum Scribn. & Merr. and a few other species of the section Harpostachys or Dimorphostachys of Paspalum. The following species belong to this genus: Thrasya paspaloides H. B. K. Nov. Gen. & Sp. 1 : 121. pi. 39. 1816. (See above. ) Panicum thrasya Trin. Mem. Acad. St. Petersb. VI. Sci. Nat. 3 : 2228. 1834. Based on T. paspaloides H. B. K. Only the Humboldt and Bon- pland specimen is mentioned. It is possible that this species has not been collected a second time. Thrasya thrasyoides (Trin.). Panicum thrasyoides Trin. Gram. Pan. 126. 1826. "Brasil. (Langs- dorff)." The type specimen, in the Trinius Herbarium, is labeled Chase — Notes on Genera of Panicetz. IV. 115 Panicum thrasyoides in. ul Fig. 1 . Th rasya thrasyoides. (Two views of spikelet and fruit x 10 diam.) UDique in carnpis siccis, Brasiliae erescens, leg. incuse Dec. 1824. el. Langadorff." Thrasya hirsuta Nees, Agrost. Bras. !)4. 1829. Based on P. thrasyoides Trin. Nees distinguishes between this and T. paspaloides, a specimen of the original collection of which he exam- ined in the Willdenow Herbarium. We take it on his authority that the two are distinct. So far as we know they have not been united, but they are closely allied and it seemed possi- ble that Kunth's erroneous description might have led to failure to recognize his species. Thrasya petrosa (Trin.). Panicum petrosum Trin. Gram. Icon. 3: pi. 280. 1836. Trinius states that the illustration is "ad specimen Brasilianum." The type specimen, in the Trinius Herbarium, labeled "Panicum petrosum m. Cuyaba febr. 1827," was probably collected by Langsdorff. Tylothrasya petrosa Doell, in Mart. Fl. Bras. 2: z295. pi. 37. 1877. Based on Panicum petrosum Trin. Doell bases his genus Tylothrasya on this species, the basal callus being the character by which he differentiates it from Thrasya H. B. K. Thrasya thrasyoides, however, has this callus though it is less pronounced than in T. petrosa. Bentham (Linn. Soc. Journ. Bot. 1!» : 42. 1881) misspells the specific name "petraeum." Thrasya cultrata (Trin.) Nees, Agrost. Bras. 94. 1821). Panicum cultratum Trin. Gram. Pan. 126. 1826. " Brasil (Langs- dorff)." The type specimen, in the Trinius Herbarium, is labeled "Panicum cultratum m. In graminosis, m. da Congonhas, Brasil. leg. mense Jan. 1825. cl. de Langsdorff. Thrasya cultrata N. ab Esenb." Above this the name " cultratum " had been written and crossed out, apparently by Trinius himself, and " monostachyum Hb. Kth." added. In his later work, " Panicearum Genera" (Mem. Acad. St. Petersb. VI. Sci. Nat, 3 : 2 228. 1834) Trinius gives P. cultratum as a synonym of P. monostachyum. Thrasya campylostachya (Hack.). Panicum ca mpylostachyum Hack. Oesterr. Bot. Zeitschr. 51 : 367. 1901. " Costarica ; in savannis ad Canas Gordas leg. Pittier, nr. 11012 et 11018." The type (no. 11012) is in Hackel's herbarium, a duplicate is in the National Herbarium. This and the preceding species differ from the first three here listed in the lack of the quill-like hairs on the halves of the split sterile lemma, and also on the summit of the fruit. 116 Chase — Notes on Genera of Pankece. TV. There are several South American species, Panicum Schumannii Pilger among them, at present insufficiently known to us, which belong in this genus. 9. Genus ECHINOLAENA Desv. Echinolaena Desv. Journ. de Bot. Paris 1 : 75. 1813. This genus is based on a single species, E. hirta Desv. (1. c.) " Habitat in America equinoxiali." The type specimen, bearing in Desvaux's handwriting the name and date as published, is in the herbarium of the Museum d'His- toire Naturelle at Paris. Echinolaena was recognized as a genus by Kunth (H. B. K. Nov. Gen. & Sp. 1 : 118. 1816, Rev. Gram, 1 : 54. 1829, and Enum. PI. 1 : 171. 1833), and by Trinius in his earlier work (Gram. Pan. 75. 1826), though later (Mem. Acad. St. Petersb. VI. Sci. Nat. 3 : 2230. 1834) he includes it under his section Harpostachys of Panicum. Nees (Agrost. Bras. 127. 1829) makes it a section of Panicum. Steudel follows Trinius' later disposition of the genus, as does Doell (Mart. Fl. Bras. 2:'2179. 1877). Bentham (Linn. Soc. Journ. Bot. 19 : 50. 1881) maintains Echinolaena as a genus because of its having "the rigid single spike of some Chlorideae," and in the Genera Plantarum (Benth. & Hook. Gen. PI. 3 : 1107. 1883), also, he gives it generic rank. Hackel (Engler & Prantl, Pflanzenf. 2 : 235. 1887) reduces it to a section of Panicum. Description. — Inflorescence of one dense and spike-like, or several rather loose racemes ; spikelets in pairs, face to face, the primary short-pediceled and perfect the secondary subsessile, abortive (sometimes developed, rarely wanting ) , the back of the primary spikelet turned toward the axis ( that is the back of the fruit turned from the axis), in two rows along one side of a flat rachis and more or less divergent from it; fertile spikelets laterally subcompressed, echinate, at least at maturity; glumes broad, firm, acuminate-pointed, one or both echinate at maturity, the first straight and as long as the spikelet or longer, the second boat-shaped or becoming ventricose; sterile lemma broad, acuminate, enclosing a nearly equal sterile palea and usually a staminate flower; fruit indurated, less so than usual in Panicum, plano-convex, elliptical, the margins of the lemma flat or inrolled at the summit only, a minute membranaceous wing or spongy thickening on either side at base. Freely branching, more or less decumbent or creeping herbs, the type species confined to the South American tropics, another of widespread tropical distribution and a third of Madagascar. The statement made by Bentham (Benth. & Hook. Gen. PI. 3 : 1107. 1883): " rhachi spicse rigida ultra spiculas producta," and which appears from the Conspectus (page 1078) to express his chief reason for main- taining Echinolaena as a valid genus, is found to be erroneous. The uppermost spikelet in E. scabra (E. hirta), the one species Bentham admitted to the genus, (as well as in E. polystachya) is strictly terminal, the rachis not at all produced beyond its insertion. This uppermost spikelet is solitary and usually erect, and the first glume is long-acuminate Chas< — Notes on Genera of Panicecz. IV. 117 (longer than in the other spikelets) and lias much the appearance of a hirsute prolongation of the rachis. The characters to which we here give chief weight as generic are the strictly racemose inflorescence, the paired spikelets t'accti i face (exactly the opposite arrangement to that in Thrasya), the lower or secondary spikelet of the pair ahortive or usually so, the fertile spikelet with its hack to the axis (as in Brachiaria), the long first glume, and the fruit less indurated than in Panicvm. Raceme solitary, terminal on the culm or branches; second glume boat-shaped only . . . . . v E. inflexa. Racemes several, erect, distant along a common axis; second glume becoming ventricose and uncinate-spiny at maturity . . E.polystachya. Echinolaena inflexa (Poir. ). Cenchrus inflexus Poir. Encyc. Suppl. 6 : 50. 1804. "M. Richard l'a recueillie a Cayenne. ( V. s. in. herb. Jussieu & Lam. )" The type speci- men in the herbarium of the Museum at Paris was examined by Prof. Hitchcock and found to be the same species as the type of Echinolaena hirta Desv. in the same herbarium. Trinius (Gen. Pan. 75. 182(5) also re- ferred C. inflexus Poir. to E. scabra H. B. K. Poiret observes that he finds no involucre, that the species may be- long to another genus, and that it deserves further study. Cenchrus marginalia Rudge, PI. Guian. 19. pi. 25. 1805. No locality nor specimen is cited. The type has not been examined but the plate and de:cription identify the species. Echinolaena hirta Desv. Journ. de Bot. Paris 1 : 75. 1813. (See above.) Echinolaena scabra H. B. K. Nov. Gen. & Sp. 1 : 118. pi. 38. 1816. "Cres- cit in ripa umbrosa Atabapense juxta vicum San Balthasar et rupem Kema- rumo. ( Prov. de la Nueva Guayana. )" The type specimen has not been examined but the plate and description identify the species. Panicum echinolaena Nees, Agrost, Bras. 128. 1829. Nees divides the entire species into three varieties, a, p and 7 and cites " Echinolaena scabra ,H. et K." as a synonym without indicating to which variety he considers it synonymous. Since he also cites specimens we can not con- sider P. echinolaena a change of name only. " Var. a et /3 in Brasilia meridional!. (Sellow.) Vidi in Herb. Reg. Berol. )— Var. 7 in campis siccis ad Tanbate nee non Sahara, provinciarum S. Pauli et Minarum * * * (Martins)." The specimens referred to as a and /3 have not been examined. The Martius specimen, the type of var. 7, in the Munich Fig. Echinolaena inflexa. ( Spikelet and fruit x 6% diam.) 118 Chase — Notes on Genera of Panicece. IV. Herbarium, is the villous form represented in the National Herbarium by Novaes 1248, Campinas, Brazil. Echinolaena polystachya H. B. K. Nov. Gen. & Sp. 1 : 119. 1816. " Crescit in ripa fluminis Magdalense inter Tenerife et Zambrano." A part of the type, labeled " Echinolaena polystachya H. B. K. in ripa fluminis Magdalena, ex herb. Hnmb. & Bonpl.," was examined in the Berlin Herbarium. As indicated by Nees (Agrost. Bras. 129. 1829) and Trinius (Mem. Acad. St. Petersb. VI. Sci. Nat, 3 :2240. 1884) this is the same as Panicum uncinatum Raddi. Panicum uncinatum Raddi, Agrost. Bras. 41. 1823. "In sylvaticis prope Catumby, non procul ab Urbe Rio de Janeiro." An authentic specimen from Raddi (probably a duplicate type) was examined in the herbarium of the British Museum. Panicum heteranthum Link, Hort. Berol. 1 : 212. 1827. " Hab. in Brasilia." An authentic specimen from the Berlin Botanical Garden was examined in the Trinius Herbarium. The description also identifies the species. Panicum glandulosum Nees, Agrost. Bras. 129. 1829. " Habitat in sylvis ad Xipoto et ad Rio Jequetinhonha flumina (Martius)" is the first specimen cited, "in confinibus regni Paraguayani legit Sellovius (Vidi in Herb. Reg. Berol.)" an additional one. The Sellow specimen was examined at the Berlin Herbarium. Nees gives Echinolaena polystachya H. B. K. and Panicum uncinatum Raddi as synonyms. Trinius (Gram. Pan. 174. 1826) gives " Panicum glandulosum N. ab Es. in litt." (as well as P. uncinatum and Echinolaena polystachya) without description as a synonym of Panicum nemorosum Swartz var. /3. Echinolaena Trinii Zoll. & Mor. Syst. Verz. Zoll. 102. 1846. Based on "Panicum uncinatum Trin. v. 3. non Brown." The reference to Trinius is probably to the " Panicearum Genera " (Mem. Acad. St. Petersb. VI. Sci. Nat. 3 : 2 240. 1834). The name published by Brown is P. uncin- ulatum not P. uncinatum. Panicum echinatum Willd. ; Doell in Mart, Fl. Bras. 2 : 2 193. 1877. This herbarium name is given as a synonym of P. uncinatum Raddi. The specimen, in the Willdenow Herbarium, bears the data " Magdalena, Humboldt," and is doubtless a duplicate of the type of E. polystachya. Immature specimens of this species bear a superficial resemblance to Ichnanthus nemorosus (Sw.) Doell, and allied species. Trinius, as seen above, at first included it as an unnamed variety of Panicum nemorosum, but later (Gram. Icon. 2: pi. 216. 1829) gives Panicum uncinatum as a valid species, explaining that though it resembles P. nemorosum it differs in having a subcoriaceous lower floret, and, in the upper racemes, spikelets in which the prickles of the " inferior glume " [the plate shows the second glume is intended] are so dense and long as to produce a resemblance to spikelets of Lappago [Nazia] . Because of the racemose inflorescence Trinius here and in the Panicearum Genera (Mem. Acad. St. Petersb. Chase— Notes on Genera of Panicete. IV. 119 VI. Sci. Nat. 3 : 2 240. L834) places P. uncinatum in the section Brachiaria. Kunth (Rev. Grain. 1:54. 1829) doubtfully refers to Echinolaena his own E. polys tac hya, together with Panicum nemorosum&w., P. naviculare Nees (both of which belong in Ichnanthus) and P. brachystachyum Nees and P. procurrens Nees. In his synoptical heading the character " glum a superior echinato-glochidata " would exclude all but E. polystaehya. In the Enumeratio (Enum. PI. 1 : 173. 1833) Kunth again doubtfully includes these species and adds Panicum Inliaceum Bert., probably referable to Ichnanthus. While the uncinate and ventricose second glume of the mature spikelet of this species makes it look very different from spikelets of E. inflexa, the paired spikelets, the primary fertile, the secondary abortive or rudi- mentary, placed face to face, the back of the primary one turned toward the rachis, and the essentially like structure of the spikelet to that of E. inflexa show the two species to be congeneric. The second glume becomes indurated and spiny only as the spikelet nears maturity; the immature spikelets resemble those of E. inflexa. The sterile spikelet is often reduced to a mere rudiment, is sometimes wanting, and is rarely developed like the primary spikelet. In the fifty specimens of this species in the National Herbarium none are found without some of these sterile spikelets in the racemes. From the description, Echinolaena madagascarenxis Baker, a species which we have not seen, appears to belong unmistakably to this genus. 10. Genus MESOSETUM Steud. Mesosetum Steud. Syn. PI. Glum. 1 : 118. 1854. This is based on a single species, M. cayennense Steud. (1. c.) " Leprieur legit in Cayenne." The type specimen, labeled "Mesosetum cayennense Steud. Cayenne. Leprieux, 1825," in the Steudel Herbarium at Paris, is found to be the same species as Panicum rottboellioides H. B. K. Steudel earlier (Flora 33 : 228. 1850) mentions the name in a paper on the progress of his work on the " Synopsis plantarum." The generic description as given by Steudel is faulty and misleading. He evidently became confused as to the morphology of the parts of the spikelet. But reading his description with a dissected spikelet of P. rottboellioides under the microscope one can see that this is what Steudel is trying to describe. It seems likely from his description of the " flosculus hermaphroditus " that he mistook the sterile lemma with the fertile floret enfolded for the hermaphrodite floret (that is mistaking the fertile lemma for the "second valve" of the floret). Steudel's statement that M. cayennense approaches Panicum mesocomum Nees is further misleading. For this reason the name Meso- situm has been referred to various sections of Panicum but never, so far as we can find, to the group containing Panicum rottboellioides, until so used by Hitchcock (Contr. Nat. Herb. 12 : 211. 1909). Bentham (Linn. Soc. Journ. Bot. 19 : 42. 1881) says that Panicum leucophaeum H. B. K. (which is Panicum insulare (L. ) Meyer, Valota insularis (L. ) Chase) 120 Chase — Notes on Genera of Panice&. TV. "appears to have been the type of the proposed genera Acicarpa, Eaddi, Eriachne, Philippi, and Holosetum and Mesosetum Steud." (It is in fact the type of only the first-named of these. ) In the Genera Plantarum (Benth. & Hook. Gen. PI. 3 : 1101. 1883) he says that Mesosetum is per- haps (forte ) referable to the section Trichachne of Panicurn. At the same time Bentham (Linn. Soc. Journ. Bot. lit : 42. 1881) establishes as a section of Panicurn his Diplaria comprising " P. rottboellioides, H. B. K., P. exaratum and P. fcrrugineum Trim, P. pappophorum, Nees, and a few others." In the Genera Plantarum (1. c. ) this section is described and the same species mentioned as belonging to it. Dalla Torre and Harms (Gen. Siphonog. 14. IflOO) also include Mesosetum, together with Alloteropsis Presl, Coridochloa Nees, Eriachne Phil., as well as Acicarpa Raddi and Trichachne Nees which properly belong there, under Pdnicum, section Trichachne. Steudel himself failed to see the identity of his Mesosetum cayennense with P. rottboellioides or its affinity to the other species of this group which he includes under the section Harpostachys (Syn. PI. Glum. 1 : 55. 1854). It was only Prof . Hitchcock's examination of the type specimens of Mesosetum cayennense and P. rottboellioides, both now in the Herbarium of the Museum d' Histoire at Paris, that revealed the specific identity of the two. It is unfortunate that SteudePs generic name, with its inadequate description, must stand for this well-marked genus. But if the names of all incorrectly described genera were rejected the nomenclature of the Gramineae would undergo many changes, for this family seems particularly to have suffered from the misunderstanding by authors, of the morphology of the parts of its inflorescence. Anthaen- antia, Ichnanthus, Alloteropsis and Pentarrhaphis are examples of valid genera incorrectly described by the authors who bestowed the names we use for them. Description. — Inflorescence a single, terminal, erect, spike-like raceme, the spikelets subsessile, solitary, alternate in two rows on one side of a three-angled, rarely winged, tortuous or zigzag rachis, the spikelets with the back of the fruit turned from the median line of the rachis (the first glume towards it), ventricose on the side toward the rachis, and fitting into its concavities, the back of the spikelet flat or nearly so; glumes 3- to 5-nerved (when 5-nerved the lateral pairs of nerves approxi- mate) the lateral nerves often uniting with the midnerve below the acute or acuminate summit, one or both usually clothed on the margin with stiff hairs; sterile lemma like the glumes in texture and the distribution of the pubescence, usually appearing L'-keeled from the strong lateral nerves and firm lateral internerves and thin or hyaline middle internerves and weak midnerve; sterile palea wanting except in the section Bifaria ; fruit pointed, ventricose on the face (palea side) usually straight on the back, the lemma and palea less indurated than in Panicurn, the flat margins of the lemma not hyaline. Slender perennials with narrow leaves, the uppermost reduced to a bladeless or nearly bladeless sheath. This genus, like Brachiaria, differs from Panicurn in the strictly race- mose inflorescence and reversed position of the spikelets, and further in the form of the spikelets, swollen or ventricose on the face (or first glume Chase — Notes on Genera of Panicecz. IV. 121 side) instead of on the back as in Panicum, Paspalum and, in more or less pronounced form , in all other genera of the Paniceae. The fruit differs from that of Panicum in having less indurated lemma and palea, the margins of the lemma Hat. Doell (in Mart. Fl. Bras. 2 : 2 17:5. 1.S77) states that in this group of plants the caryopsis has a linear hilum, but the specimens in the National Herbarium are not mature enough to per- mit of verification of this statement. The genus is confined to the tropics of the western hemisphere, Brazil being the center of distribution, two species occurring in the West Indies. Besides Mesosetum proper the genus contains a well-marked section. First ghune awnless; lower floret neuter . . . . Mesosetum proper. First glume awned ; lower floret staminate . . . Section Bifaria. Mesosetum proper. This contains the following species : Mesosetum rottboellioides (H. B. K. ) Hitchc. Contr. Nat. Herb. 12 : 211. 1900. Based on Panicumrottbocllioides H. B. K. Panicum rottboelli- oides H. B. K. Nov. Gen. & Sp. 1 : 96. pi. 32. 1816. "Crescit in humidis ripa? Orino- censis inter Maypures et montem Sipapo." Mesosetum cayen- nense Steud. Syn. PI. Glum. 1 : 118. 1854. (See above. ) Fig. 3. Mesosetum rottboellioides. (Spikelet, first glume, second glume, sterile lemma and fruit x 10 diam.) Mesosetum Wrightii Hitchc. Contr. Wright 3859 no. 559961 in the U. S. Nat. Herb. 12:211. 1909. National Herbarium." This species is known only from Cuba. Mesosetum exaratum (Trim). Panicum exaratum Trim Gram. Pan. 100. 1826. "Brasil. (Langs- dorff. )" The type specimen, in the Trinius Herbarium, is labeled "Panicum exaratum in. In pratis paludosis S. da Lapa. Brasil. leg. cl. de Langsdorff." Panicum pappophorum Nees, Agrost. Bras. 104. 1829. ' Habitat * * * provinciae Piauhiensis." The type specimen, in the Munich Herbarium, bearing the name and locality as published, was collected by Martius. 122 Chase — Notes on Genera of Panicece. IV. Mesosetum loliiforme (Hochst.) Chase, Bot. Gaz. 51 : 302. 1911. Based on Panicum loliiforme Hochst. Panicum loliiforme Hochst. in Steud. Syn. PI. Glum. 1 : 56. 1854. " Herbr. Dr. Hostmann nr. 1071." The type specimen is in the her- barium at Leipzig. This is the not uncommon Cuban species which has been usually referred to P. rottboellioides in herbaria. It differs from M. rotlboellioides and M. Wrightii in having glumes each bilaterally unsymmetrically developed, the second shorter than the first instead of being the longest scale of the spikelet as in M. rotlboellioides ; M. loliiforme also differs in being sparingly stoloniferous. Nees' P. pappofthorum var. /3, bis specimen of which, also collected by Martins in Brazil, is now in the Munich Herbarium, is either M. lolii- forme or a very closely allied species. Mesosetum chlorostachyum (Doell). Panicum chlorostachyum Doell in Mart. Fl. Bras. 22 : 173. pi. 28. A. 1877. "Habitat in regionibus numinis Rio Negro (Spruce n. 885, 1310)." Spruce's no. 885 is in the Munich Herbarium. Mesosetum ferrugineum (Trin.) Panicum ferrugineum Trin. Gram. Pan. 1826. " Brasil (Langs- borff)." The type specimen, in the Trinius Herbarium, is labeled "Panicum ferrugineum m. In campis glareosis pr. S. Luzia, Brazil. 1. mense Oct. 1824. cl. Lajwsdorff." Panicum eriochryseoides Nees, in Trin. Gram. Pan. 160. 1826. This name is given as a synonym of P. ferrugineum. Nees ( Agrost. Bras. 103. 1829) publishes this as a new species citing a specimen collected in Brazil by Sellow, "Vidi in Herb. Peg. Berol." What is evidently a duplicate of this is in the Trinius Herbarium. Nees gives "Panicum ferrugineum Tr. in litt." as a synonym of P. eriochryseoides. Mesosetum sclerochloa (Trin. ) Hitchc. Contr. Nat. Herb. 12 : 212. 1909. Based on Panicum sclerochloa Trin. Panicum sclerochloa Trin. Gram. Icon. 3 : pi. 283. 1836. The illustra- tion is drawn "ad specimen Brasilianum." The type, in the Trinius Herbarium, is labeled "Panicum sclerochloa m. pr. Cuyaba febr. 1827." This peculiar species shows an approach to the species of the section Bifaria in the thickened midnerve of the first glume, keeled toward the apex and protruding as a short, laterally compressed point from between the lobes of the slightly cleft summit of the glume. The midnerve of the second glume and that of the sterile lemma are similarly keeled. Besides the species listed above, Panicum lolium Nees, the type speci- men of which has not been examined, and a few other insufficiently known South American species belong in this genus. Chase — Notes on Genera of Panioese. TV. 128 Section Bifaria (Hack.). Hackel (Oesterr. Bot, Zeitschr. 17 : 7.">. 1897) establishes Panicum, sec- tion Bifaria with three new Brazilian species, P. bifarium, P. caudicula- i a in and /'. elytrochaetum. Professor Hackel kindly contributed to the National Herbarium a spike from the type specimen of P. bifarium; the other two species have not been examined. The spikelets of P. bifarium have the structure and reversed position characteristic of Mesosetum but differ strikingly in the character of the first glumes. In these the mid- nerve, which (as in most species of Mesosetum proper) is joined by the lateral nerves, is keeled toward the summit of the notched glume and extends into a laterally compressed, sinuous, stout awn, varying from a short point to as long again as the spikelet. The species differs further from those of Mesosetum proper in having a staminate flower in the lower floret. Prof. Hackel points out the relationship of his section Bifaria to Bentham's section Diplaria (P. rottboellioides, P. loliiforme, etc.), and because of this relationship establishes for his three species a section of Panicum instead of a new genus. But unless the boundaries of the genus Panv-um are extended to the limits to which Trinius (Gram. Pan. 1826) stretched them, including Paspalum, Oplismenus, Setaria and other genera recognized as distinct by even the most conservative students of Gram- ineae, the species here referred to Mesosetum with their constant combi- nation of characters and their sharp delimitation from Panicum (there being no intergrading species as between Panicum and Paspalum and between Panicum and Chaetochloa) can not well find place therein. The only species of the section Bifariawe have seen is here transferred: Mesosetum bifarium (Hack. ). Panicum bifarium Hack. Oesterr. Bot. Zeitschr. 47:76. 1S97. " Serra da Baliza ad Cachoeiras da Vargem Grande, 5-1. 1895, Glaziou nr. 22455." Bifaria bifaria Kuntze, Gen. PI. 3:2359. 1898. Based on Panicum bifarium Hack. Kuntze, apparently without having seen the species, raises Hackel' s section, on account of the lobed and awned first glumes, to generic rank in order to be consistent, he having maintained Oplis- menus and Chaetium as distinct genera, he says, because of this charac- teristic. 11. Genus LEUCOPHRYS Rendle. Leucophrys Rendle, Cat, Afr. PI. Welwitsch 2 : : 193. 1899. This genus is based on a single species, L. mesocoma (Nees) Rendle (op. cit. 194) based on Panicum mesoc'omum Nees, an African species. In this genus the spikelets are placed with the back of the fruit turned from the axis, but, not being strictly racemose, this is not readily observed. The rachilla is produced into a short stipe below the first glume which is separated from the second by the slightly prolonged second joint of the rachilla. The nearly glabrous first glume about as long as the spikelet and the densely silky second glume and sterile lemma, as well as the narrow 124 Chase — Notes on Genera of Panicese. TV, panicle and the reverse position of the spikelets, suggest an approach to the unique North American Panicum ciliatissimum Buckl. The stipitate spikelet of Leucophrys, however, with a joint between the glumes, and the different arrangement of the silky pubescence on the second glume and sterile lemma (a dense ring of long hairs across the middle of the spikelet, an arrangement also found in Panicum nigropedatum, P. serra- tum, P. argenteum, etc., and approached in several species of Eriochloa) does not show a close affinity with P. ciliatissimum. Species intermediate between them may be found, but without such intermediate species our P. ciliatissimum finds a more natural place in Panicum. Study of more material may show that Panicum argenteum and its allies should be placed in Leucophrys. Rendle (1. c. ) places Leucophrys immediately before Tricholaena to which he compares it. The silky spikelets resemble outwardly those of Tricholaena but the indurated fruit and membranaceous, entire and awnless second glume and sterile lemma indicate but remote relationship to that genus. The two, however, probably belong closer together than is indicated by their relative position in the sequence here given, but it must be borne in mind that any lineal arrangement of that which is really irregularly radiate must fail to show all but the closest relationship. 12. Genus ERIOCHLOA H. B. K. Eriochloa H. B. K. Nov. Gen. & Sp. 1 : 94. pi. 30 and 31. 1816. Two species are included, E. dislachya H. B. K. (1. c. pi. 30) "Cresc.it in ripa Orinoci fluminis inter Santa Barbara et Esmeraldam," and E. polystac.hyaW. B. K. (1. c. pi. 31. ). Since both are illustrated and are equally covered by the generic description the first species is here taken as the type of the genus. A duplicate type of this, "Ex herb. Humboldt," was examined in the Berlin Herbarium. Helopus Trin. Fund. Agrost. 103. pi. 4. 1820. A single species, H. pilosus Trin. (op. cit. 104) is included. Since the genus only and not the species is described and since no specimen is cited we should con- sider the species based on Milium ramo- sum Retz., which is cited as a synonym, save for the fact that plate 4 shows an aristate lemma while Retzius (Obs. Bot. 6:12. 1791) states that the " valvula exterior" is ma- cronate only, and that Trinius later (Gram. Pan. 118. 1826) gives Helopus pilosus Trin. as a synonym of Paspalum punctatum Fliigge (Milium punctatum L. ), while Milium ramosum Retz. he gives as a synonym of Paspalum annulatum Fliigge. In the Trinius Herbarium is a specimen collected by Mikan in Brazil, which is marked in Trinius' hand " Helopus pilosus in. Paspalum punctatum Fliigge." Whether or Fig. 4. Eriochloa dislachya. (Two views of spikelet and fruit x 10 diam.) Chase — Notes on Genera of Paniceae. IV. 125 not this specimen be taken as the type of H. pilosus we know that the species Eriochloa punctata (L.) Hamilt. (Prod. PI. End. Occ. 5. 1825), based on Milium punctatum L., should be taken as the type of the genus Helopus Trin. Trinins places Eriochloa II. B. K. immediately after Helopus in his systematic arrangement and ( page 75) differentiates the two as follows: Racemi ad rachin eommunem. Cor. apice mucronato-aristata . Helopus. Racemi subpaniculati. Involucrum stellato-setosum ! Eriochloa. Probably Trinins had not at that time seen Kunth's species and was impressed by Kunth's description of an involucre of numerous hairs at the apex of the pedicel and by the ring of spreading hairs shown in plate 30. Later (Mem. Acad. St. Petersb. VI. Sci. Nat. 3 : - 130 and 133. 1834) the species of both Eriochloa and Helopus are placed in Paspalum in the sub- division Helopus. Oedipachne Link, Hort. Berol. 1:51. 1827. This includes a single species, 0. punctata Link (1. c. ), based on Milium punctatum L. In his additions and emendations (p. 273) Link directs that Oedipachne be ex- punged and Helopus Trin. be inserted in its place, and in the second volume (Hort. Berol. 2:199. 1833) Helopus is given as a genus with Oedipachne as a synonym. Eriochloa H. B. K. seems to have been over- looked by Link. Alycia Willd. ; Ktend. Norn. Bot, ed. 2. 1 : 66. 1840. This herbarium name is listed without description, and two undescribed species, A. coarc- tata Willd. and A. distachya Willd., are listed under it. Both names are here referred to He lop us brachystachys but on page 747 A. coarctata Willd. is referred to Paspalum polystaehyum Trin. and A. distachya to P. brachystacliyum Trin. Helopus brachystachys Trin. (upon which Paspalum bracJiystachyum is based) is Eriochloa distachya H. B. K. or a very closed allied species, while Paspalum polystaehyum Trin. is based on Eriochloa polystachya H. B. K., so that whichever species he. taken as the type of Alycia this name becomes synonymous with EriocJiloa H. B. K. Willdenow's herbarium name is spelled "Aglycia" by Steudel (op. cit. 37 ) with the same two species listed under it. Nees (Agrost. Bras. 16. 1829) recognizes Helopus Trin. as a genus, and does not mention Eriochloa. Probably since neither of Kunth's species was from Brazil Nees neglected to study the genus. Doell (Mart. Fl. Bras. 2 : 2 123. 1877) recognizes Helopus Trin. as a genus, and includes Eriochloa H. B. K. under it. Benthani (Benth. & Hook. Gen. PI. 3 : 1099. 1883) recognizes Eriochloa H. B. K. Nash (Bull. Torrey Club 30 : 374. 1903) takes up the name Monachne Beauv. for this genus, but we do not find this name tenable. It is based on M. unilateralis Beauv., an undescribed species unrecognizable from the figure, which represents a branching paniculate inflorescence, though the spikelets were evidently drawn from some species of Eriochloa, and Saccharum reptans Lam., which is a species of Panicum allied to P. urvilleanum Kunth. 126 Chase — Notes on Genera of Pan ice se,. IV. Description. — Inflorescence of one to many, usually dense, racemes along a common axis; spikelets solitary, sometimes in pairs, short pedi- celed or subsessile in two rows on one side of a narrow, usually hairy rachis, the pedicels often clothed with long stiff hairs [the "involucre " of Kunth], the back of the fertile lemma turned from the rachis; spike- lets dorsally compressed, more or less pubescent, stipitate by the length- ening of the internode of the rachilla below the second glume into a more or less ring-like, usually dark-colored callus, the first glume reduced to a minute sheath about this internode and adnate to it; second glume and sterile lemma equal or nearly so, acute or acuminate, the lemma usually enclosing a hyaline palea and sometimes a staminate flower; fruit indu- rated, less so than in Panicum, minutely papillose-rugose, the lemma mucronate-pointed or aristate, sometimes pubescent at the apex, the margins slightly inrolled. Perennial herbs with linear leaves and terminal panicles of few to many racemes, confined to the warmer temperate and tropical regions of both hemispheres. This genus approaches Leucophrys and Brachiaria through such species as Panicum nigropedatum Munro, P. argenteum R. Br., and P. serratum R. Br. in which the first glume is well developed above the more or less stipitate base. 13. Genus BRACHIARIA (Trim) Griseb. Brachiaria Griseb. in Ledeb. Fl. Ross. 4 : 469. 1853. Grisebach here raises "Panicum sect. Brachiaria Trin." (Mem. Acad. St. Petersb. VI. Sci. Nat. 3 : 2 194. 1834) to a genus, including under it a single species B. eruciformis ( Smith ) Griseb. based on Pani- cum eruciforme Smith (Sibth. Fl. Graec. 1 : 44. pi. 59. 1806). " In arvis circa Junonis teinplum in insula Sarno." The plate is a good representation of the species and fully identifies it. The enlarged portion of the raceme ( f. A) shows that the spikelets are placed with the first glume toward the axis. This species we here take as the type of the genus. Roemer & Schultes (Syst. Yeg. 2 : 42<>. 1817) misspell the name " cruciforrne," and in this form also it appears in the Index Kewensis under Brachiaria. Grisebach cites not the first work in which Trinius proposes the sub- genus Brachiaria, but a later work, in which Trinius includes a some- what different group of species from those included in his first establish- ment of the section in "De Graminibus paniceis " (pages 51 and 125. L826). Here (p. 51) the section is diagnosed as follows: Racemes simple, regularly or irregularly alternate; spikelets oblong (rarely lanceolate), either in regular series and imbricated or laxly disposed ; lower glume manifest, awnless. Under it are included "Thrasya Kth.? * * * Fig. 5. Brarhiaria eruciformis. (Two views of spikelet and fruit x 10 diam.) Chase — Notes on Genera of Panicete. IV. 127 Urochloa Pal. [de Beauv.] * * * Streptostachys Desv., Echinochloa Pal." Tn the body of the work (p. 125) the first species included under section Brachiaria is Panicum decumbens Roem. & Schult. ( Paspalum decumbens Poir.), the species which later (Mem. Acad. St. Petersb. VI. Sci. Nat. :! : - '227. 1834) is the first included under section Harpostachys of Trinius. This is followed by Panicum thrasyoides and P. cultratum (species of Thrasya), several species of Panicum (in the stricter sense) and of Echinochloa; P. holosericeum and P. argenteum, in which the spikelets are in the reversed position and which are the first species under section Brachiaria as later used by Trinius, are here only the seventeenth and eighteenth species under that section. In the " Panicearum Genera " ( Mem. Acad. St. Petersb. VI. Sci. Nat. 3 : 2 194. 1834), which Grisebach cites, Trinius makes Brachiaria the eighth section of Panicum, Har- postachys, to which are relegated the species with a single raceme earlier included in Brachiaria, being the seventh. The characters now assigned to Brachiaria are: Simple, alternate racemes, the partial axes angled ( usually 3-angled); sessile or short-pediceled, glabrous, pilose, or lanate, awnless spikelets, imbricated in 2, 3, or 4 series. No species is here men- tioned, but under "VIII Brachiaria" (pages 233-247) this purely arti- ficial division contains thirty diverse species referable to Brachiaria (as here limited), Echinolaena, and the greater number to Panicum. Since there is nothing in either work to indicate which species should be con- sidered the type, it seems best to follow Grisebach' s choice when he established Brachiaria as a genus. His choice, to be sure, was guided by the fact that Panicum eruciforme was the only one of the group which occurred in the Russian Empire, but even so, it would be unwise to reject his type and arbitrarily to choose another. Panicum eruciforme is in- cluded, under the name "Panicum Isocline Roth !" by Trinius in the first subdivision of his section Brachiaria as limited in the "Panicearum Genera." Later in the " Gramimun Supplementa " (op. cit. 4 : l 103. ls:!l5) he states that P. Isocline should be called Panicum eruciforme. Trinius' first three species, P. holosericeum R. Br., P. argenteum R. Br. and P. serratum Spreng. , are of that peculiar group of Old World species with reversed spikelets clothed with silvery hairs more or less aggregated across the middle-of the second glume and sterile lemma, and having a well-developed first glume, which, together with Leucophrys, appear to be a connecting link between Eriochloa and Brachiaria. In the present state of our knowledge it is difficult to say whether these species fall the more naturally into Brachiaria or into Leucophrys. 2. 1S29) includes in hissection " Digitariae (speculis inve'rsis)." Hackel (Engler & Prantl. Pflanzenf. 2 :235. 1887, where the name is misspelled " Lappagroslis" ) includes it in section Anastrophus. Owing to its insufficient diagnosis and the diverse species assigned to Anoxopus by Beauvois this name has been applied to different groups of species by different authors. Roemer iV- Schultes (Syst. Veg. 2 : 318. 1S17) recognize it as a genus including under it the same species as did Beauvois except A. paniceus which, they say, is Paspalum filiforme [it is a species allied to Syntherisma filiformis] . Nees, as we have seen, used it as a section of Paspalum for the species allied to A. aureus. Hooker (Fl. Brit. End. 7 : 64. 1896) says of Axonopus "A natural genus, remarkable for the small cleft palea of gl. Ill [the sterile palea]. It was established by Beauvois on Panicum cimicinum Retz, to which other grasses having no affinity with it were added." Hooker does not state why he takes P. cimicinum as the basis of Anoxopus. It fails to agree with one of the two characters of Beauvois' diagnosis (in that its spikelets are not solitary ), and it is only third in the list. Hooker here includes one other species in this genus, A. semialatus (based on Panicum semialatum R. Br. ) Stapf (Dyer, Fl. Cap. 7:418. 1898) accepts the genus as emended by Hooker. Hitchcock (Rhodora 8 : 205. 1906; Contr. Nat, Herb. 12 : 141. 1908; Gray, Man. ed. 7. 100. 1908; Contr. Nat. Herb. 12:207. 1909.) recognizes Atonopus for the congeners of A. compressus. Description. — Inflorescence of 2 to many slender racemes usually aggre- gated at the summit of the culm; spikelets solitary, sessile and alternate in two rows on one side of a 3-angled rachis, the back of the fertile lemma turned from the axis; spikelets depressed-biconvex, not turgid; first glume wanting; second glume and sterile lemma equal; sterile palea obsolete; fruit indurated, oblong-elliptic, usually obtuse, the margins of the lemma slightly inrolled. Stoloniferous or tufted perennials, with flat, condupli- cate or involute, linear leaves; species numerous in South America, a few species extending into subtropical and warm temperate regions of North America and one or two to the warmer parts of the Old World. The characters to which we here attach chief value as generic are the reversed and solitary spikelets (in which the first glume is wanting) and the racemes aggregated at the summit of the culm. The genus Axonopus as here understood, subdivides into three rather well-marked sections as follows : Rachis and spikelets glabrous or the latter softly pubes- cent, neither bearing conspicuous, stiff, spreading hairs Axonopus proper. Rachis and sometimes spikelets bearing conspicuous, stiff, spreading hairs. 132 Chase— Notes on Genera of Panicece. IV. Spikelets glabrous or minutely pubescent, tbe hairs of the rachis golden Section Cabrera. Spikelets papillose-pilose, the hairs of these and of the rachis pale Section Lappagopsis. In North America Axonopus proper is represented by the following species : Axonopus compressus (Sw.) Beauv. Milium compression Sw. Prod. Veg. Ind. Occ. 24. L788. "Jamaica; India occidentalis." No specimen of this could be found in the Swartz Herbarium, but the later detailed description of Swartz (Fl. Ind. Occ. 1 : 183. 1797) leaves no room for doubt. Paspalum platicaulon Poir. Encyc. Suppl. 5 : 34. 1S04. " Cette espece a ete recueillie a Porto-Ricco, par le citoyen Ledru. (V. s. in herb. Lam. )." The type has not been examined but the description identifies the species. Axonopus compressus Beauv. Ess. Agrost. 12. 1812. Based on Milium compressum Sw. Paspalum compressum Rasp. Ann. Sci. Nat. I. 5 : 301. 1825. Based on Axonopus compressus Beauv. Anastrophus platycaulis Schlecht. ; Ind. Kew. 1 : 118. 1893. Based on Paspalum platycaule. Closely related to this species is a narrowerdeaved form with nearly obtuse spikelets and usually few racemes which may be Paspalum tri- stachyon Lam. (Tabl. Encycl. 1 : 176. 1791. "Ex America merid. Com- munic. D. Richard"). This name has been referred by Trinius and others to P. platycaule, but Lamarck's description " spicis ternis " points to the other form. The type has not been examined. Axonopus furcatus (Fliigge) Hitchc. Paspalum furcatum Fliigge, Cram. Monog. 114. 1810. "Carolina. Bosc. Walne." The type has not been examined, but the description identifies the species. Paspalum digitaria C. Muell. Bot. Zeit. Id : .">24. 1861. "America septentrionalis, ubi forsan in Texas legit T. Drummond (coll. no. 276)." The type specimen was examined in the Berlin Herbarium. Axonopus furcatus Hitchc. Rhodora 8 : 205. 1906. Michaux's name Digitaria paspalodes and several names based upon it, Milium paspalodes Ell., Paspalum ellioltii Wats., Paspalum paspalodes Scribn., and Anastrophus paspaloides Nash, have been misapplied to this species. Michaux's type specimen is Paspalum distichum L.* Axonopus Rosei (Scribn. & Merr. ). Paspalum Rosei Scribn. & Merr. U. S. Dept. Agr. Div. Agrost. Bull. 24:9. f. 2. 1900. "Foothills of the Sierra Madre Mountains, between *See Hitchcock, Ccmtr. Nat. Herb. 12 : 146. 1908. Chase — Notes on Genera of Panicetz. W. 133 Pedro Paulo and San Blascito, L995 .1. X. Rose, August 4, L897." The type specimen is in the National Herbarium. Axonopus papillaris (Lam.). Paspalum capillnre Lam. Tabl. Encycl. 1 : 176. 1791. "Ex America merid. Comm. D. Richard." Paspalum minutum Trin. (Linnaea LO : 293. 1836), the type of which, collected by Poeppig in Peru, was examined in the Trinius Herbarium, appears to be based on depauperate specimens of A. capillaris. The only North American specimens we have seen of this species are from Costa Rica, Pvttier 508 and Jimenez L46. Axonopus laxiflorus (Trin. ). Paspalum laxiflorum Trin. Mem. Acad. St. Petersb. VI. Sci. Nat. 32 : 14S. 1834. "V. spp. Bras." The type specimen, in the Trinius Her- barium, is labeled " Paspalum laxiflorum m. In saxosis pratisque humidis S. da Lapa. Langsdorff." This species is represented from North America in the National Her- barium by Pittier 214, Alta Verapaz, Guatemala, and Xelson 273S, col- lected between Guichocovi and Lagunas, Oaxaca, Mexico. Axonopus poiophyllus sp. now Plants perennial, tufted, flattened at the base; culms erect, slender, compressed, 60 to 90 cm. high, simple, glabrous or minutely scrabrous below the appressed-pubescent nodes, the leaves mostly crowded at the base; lower sheaths much overlapping, keeled, villous, the upper pubes- cent along the margin, otherwise glabrous or minutely pubescent; ligule scarcely 0.5 mm. long, erose-ciliate ; blades erect, firm, linear, 8 to 35 cm. long (the uppermost reduced to 0.5 to 2 cm. long), 3 to 5 mm. wide, the apex boat-shaped as in Poa, the lower conduplicate at base and slightly narrower than the summit of the sheaths, usually flat above, papillose-villous toward the base on both surfaces, scabrous on the upper surface; inflorescence of about 3 slender, erect racemes, 6 to 12 cm. long, the rachis narrow, flexuous, glabrous or minutely scabrous; spikelets tinged with rose-purple, distant their own length, 3 mm. long, 1 mm. wide, oblong-elliptic, subacute, the second glume and sterile lemma slightly exceeding the fruit, minutely pubescent at the base and along the edges with appressed silky hairs, 4-nerved, the midnerve suppressed or nearly so, especially that of the glume, the lateral nerves near the margins and approximate; fruit pale, very obscurely papillose, the lemma with a minute tuft of erect hairs at the apex. Type U. S. National Herbarium no. S60024, collected in April, 1904, in the vicinity of Secanqufm, Alta Verapaz, Guatemala, by 0. F. Cook & C. B. Doyle (no. 58). This species is related to the group of cespitose South American species to which A. laxiflorus also belongs. 134 Chase — Notes on Genera of Panicecs. IV. Axonopus deludens sp. nov. Plants perennial ; culms geniculate at base, leafy, strongly flattened, rather stout, 1 to 1.5 meters high, sparingly branching, glabrous, the nodes glabrous; sheaths glabrous or minutely pubescent at the summit; lignle about 0.5 mm. long, membranaceous, erose; blades rather thin and lax, linear, 15 to 45 cm. long, 8 to 13 mm. wide, flat, sparsely papil- lose-scabrous on the upper surface and on the margin, pubescent on the narrow auricles, glabrous beneath, the midnerve prominent; inflorescence of 6 to 15 very slender, erect or rather lax racemes, 10 to 25 cm. long, the lower mostly naked at the base, the rachis narrow, subflexuous, scabrous; spikelets purple tinged, distant about their own length or, toward the base of the raceme, remote, 3 to 3.2 mm. long, 1 to 1.2 mm. wide, obtuse, glabrous, the second glume and sterile lemma covering the fruit but not exceeding it, delicate in texture, 4- or 5-nerved, the midnerve present or suppressed even in adjacent spikelets, the lateral nerves strong; fruit papillose, smooth toward the summit and margins of the lemma, the apex glabrous or with a few obscure hairs. Type U. S. National Herbarium no. 460803, collected Oct. 20, 1903, Barranca near Guadalajara, Jalisco, Mexico, by C. G. Pringle (no. 8761 ). This species is not closely related to any other we have seen. The geniculate lower nodes suggest a stoloniferous habit, but our one speci- men does not show stolons. The suppression of the midnerve in the glume or its presence is somewhat confusing, since it gives the impression of spikelets turned different ways, reverse and obverse, but turning back the glume always shows the back of the fertile lemma turned from the axis. There are some eight or ten South American species, as yet insufficiently known, that belong in Axonopus proper. Paspalum suffultum Mikan (Trin. in Spreng. Neu. Entd. 2 : 4<>. 1821) is interesting as a species inter- mediate between Axonopus proper and section Cabrera. The axis is not ciliate but the very short pedicels bear at either side a few stiff hairs nearly as long as the spikelets. Section Cabrera (Lag.). Axonopus chrysoblepharis (Lag.). Cabrera chrysoblepharis Lag. Gen. & Sp. Nov. 5. 1816. (See above under Cabrera. ) Paspalus immersus Nees, Agrost. Bras. 82. 182!). "Habitat in campis ultra 2000 pedes altis provinciae Minarum generalium passim." The type specimen was examined in the Munich Herbarium. This is, as Nees indicates, the species Kunth called Paspalum aureum, but since Kunth based this name on Axonopus aureus Beauv. , his name is a synonym of that, misapplied to this species. Panicum immersum Trin. Mem. Acad. St. Petersb. VI. Sci. Nat. 32 : 197. 1834. Based on Paspalum immersum Nees. Panicum chrysoblephare Steud. Syn. PI. Glum. 1 : 38. 1854. Based on Cabrera chrysoblepharis Lag. Chase — Notes on Genera of Pdnicees. IV. 135 Paspalum chryxoldephare Doell in .Mart. Fl. Bras. 22 : 119. 1877. Based mi Panicum chrysoblephare Steud. The North American specimens of this species in the National Her- barium are all from Costa Rica : Herb. Tnstit. physico-geogr. nat. costa- ricensis no. 4404, collected in "Savanes de Boruca," by Pittier & Tonduz and no. 4638, collected between Boruca and Terraba, by Pittier; no. 11004 of the same series, collected by Pittier, is a mixture of A. cltryso- hh phaHs and A. aureus. Axonopus aureus Beauv. Axonopus aureus Beauv. Ess. Agrost. 12. 1812. Beauvois states that the plant was given him by De Lessert, but does not say whence it came. The specimen could not be found in the Delessert Herbarium. The author's observation " Locustes [spikelets] sont garnies, en-dessous, et en forme d' involucre, de poils court et dores", points conclusively to one of the species with a cluster of golden hairs subtending the spikelets, these having a narrow rachis, not a broad one in which the spikelets are sunken as in A. chrysoblepharis. Following Trinius (Icon. 1: pi. 07. 1828) we take the common species with the smaller and glabrous spikelets to be the true A. aureus. The one with larger, pubescent spikelets, Paspalum canexcenx Xees (in Trin. Gram. Ran. SO. 1826, not Roth. 1821, Panicum chrysodactylon Trim), the type of which was examined in the Trinius Herbarium, has not been found in North America and for the present need not be transferred. The species of this group extending into North America is that called Paspalum aureum by Trinius. Paspalum aureum H. B. K. Nov. Gen. & Sp. 1 : 93. 1816. Based on "Axonopus [misspelled Axinopus] aureus Beauv. agrost. p. 12." The species described and illustrated by Kunth (1. c. pi. 27) is A. chrysoble- pharis. Digitar ia a urea Spreng. Syst. 1 : 272. 1825. Based on " Paspalum aureum Humb." Paspalum exasperatum Nees, Agrost. Bras. 81. 1829. " Habitat ad margines sylvarum prope Ferradas provinciae Bahiensis, in via Felisber- tiana, quae descendit e Minis." The type specimen was examined in the Munich Herbarium. Panicum aureum Trin. Mem. Acad. St. Petersb. VI. 32 : 196. 1834. Based on "Plaspalum] aureum Trin." In North America this species also is known only from Costa Rica. In the National Herbarium are the following: Herb, instit. physico-geogr. nat. costaricensis no. 3305, collected at Pacaca, by Pittier ; no. 3683, col- lected at Buenos Aires by Tonduz; nos. 11003 and 11004, collected at Canas Gordas by Pittier, the latter having an intermixture of A. chrysoblepharis. Section Lappagopsis (Steud.). Axonopus dissitiflorus (Trim). Paspalum, dissitijiorum Trin. Gram. Pan. 02. 1826. "V. spp. Brasil. (Langsdokpf. ) " The type specimen was examined in the Trinius Herba- rium. 136 Chase — Notes on Genera of Panicets. IV. Paspalus tener Nees, Agrost. Bras. 32. 1829. This herbarium name is given as a synonym of P. dissitiflorum Trin. This species is represented in the National Herbarium by Glaziou 15637, from Brazil. Axonopus bijugus (Steud.). Lappagopsis bijuga Steud. Syn. PI. Glum. 1 : 112. 1854. (^ee above under Lappagopsis. ) This more delicate species with smaller spikelets is represented in the National Herbarium by the following, all from Brazil: Burchell 5886, 7703; Gardner 21178. Two species described by Nees, Paspalum canaliculatum and P. fasti- giatum, the types of which, both collected by Martius in Brazil, were examined in the Munich Herbarium, belong in this section but being insufficiently known are not here transferred. 15. Genus REIMAROCHLOA Hitchc. Reimarochloa Hitchc. Contr. Nat. Herb. 12:198. 1909. "For R[eimaria] acuta and its allied species the above name is proposed with Reimaria acuta Fliigge as the type: Reimarochloa acuta (Fliigge)" Hitchc. Besides the type, two other species are here included in this genus, Reimarochloa brasiliensis (Spreng. ) Hitchc. and Reima- rochloa oligostachya (Munro) Hitchc. For the discussion of Reimaria Fliigge see this name under Paspalum. Description. — Inflorescence of few to several slender racemes, approximate at the summit of the culm, spreading or re- flexed at maturity ; spikelets strongly dorsally compressed, lanceolate, acumi- nate, solitary, rather distant, subsessile and alternate in two rows along one side of a narrow, flattened rachis, the back of the fertile lemma turned toward it ; both glumes wanting (or the second glume sometimes present in the terminal spike- let); sterile lemma about equaling the fruit, sterile palea obsolete ; fruit scarcely indurated, the lemma faintly nerved, acuminate, the margins inrolled at the base only, the palea free nearly half its length. Stoloniferous perennials with linear leaves; a small genus of but few species confined to the tropics and subtropics of the western hemisphere. The scarcely indurated, acuminate fruit, the margins of the lemma in- rolled at the base only, the palea free for its upper half, and the absence of the glumes, taken in combination, together with the spreading or Fig. 7. Reimarochloa acuta. (Two views of spikelet and fruit x 10 diam.) Chast — Notes on Genera of Panicea. IV 137 reflexed racemes approximate at the summit of the culm, are here taken as the distinguishing characters of this genus. Besides the species placed in this genus by Hitchcock (1. c. ) a third South American species belongs in the genus, Reimarochloa aberrans (Doell), Reimaria aberrans Doell (Mart. Fl. Bras. 2'z : 38. pi. 13. 1877) " I [abitat propeSantarem provinciae Paraensis (R. Spruce n. 851 et 887. )" The type specimen, Spruce 851, was examined in the Munich Herbarium. In this species the second illume is occasionally present on racemes with glumeless spikelets. 16. Genus PASPALUM L. Paspalum L. Syst. Nat. ed. 10. 855. 1759. After a brief diagnosis four species are given, /'. dimidiatum, P. virgatum, P. paniculatum and P. distichum. All agree with the diagnosis, though the last, with acute fruit, might be excluded from eligibility as type species because the gen- eric diagnosis reads: " Cor. Gluma * * * obtusa." None of the species are figured in the same work, none are economic nor indigenous from the standpoint of the author. Paspalum virgatum and P. panicu- latum are here first published, P. dimidiatum being the only one pre- viously described. For this reason and also because it is the first species under the genus it is here taken as the type. This is given as follows : "dimidiatum a P[aspalum] spicis subsolitariis, pedunculo communi membranaceo. Panicum dissectum sp.pl. 57. n. 6." There is nothing to explain why Linnaeus changed the specific name. This is discussed by Hitchcock (Contr. Nat. Herb. 12 : 115-116. 1908) who examined in the Linnaean Herbarium the type specimen, upon the sheet of which Lin- naeus wrote both " dimidiatum ", which is crossed out, and " dissectum." The plant is also marked "K" which indicates that it was collected by Kalm. In the second edition of the Species Plantarum (page 81. 1762) Linnaeus corrects himself, giving the name Paspalum dissectum L. based upon "Panicum dissectum Sp. pi. 1. p. 57." (For a full discussion the reader is referred to Hitchcock's paper. ) The specimen belongs to the species long known under the name Paspalum membranaceum Walt. The masculine form Paspalus was used by Fliigge (Gram. Monog. 51-190. 1810) Roemerand Schultes (Syst. Veg. 2 : 2! Hi-: ', 1 7 . 1 s 1 7 ) and by Nees ( Agrost . Bras. 18-S2. 1829). Digitaria Heist, in Fabr. Enum. PI. Hort. Helm. 207. 1759. This name is given as follows: "Digitaria Heist. Dactylis Rai. Gramen dactylon majus panicula longa, spicis pluribus nudiscrassis. Sloane." This phrase name in Sloane ( Vby. Jam. 1 : 112. pi. <>!». f. 2. 1707) refers to the species later published as Paspalum virgatum L. , Sloane's specimen of which was examined in the herbarium of the British Museum. This phrase Fig. 8. Paspalum dissectum. (Two views of spikelet and fruit x 10 iliain.l 138 Chase — Notes on Genera of Panicecs. IV. name Linnaeus first cites (as the third synonym) under Panicum dissec- turn (Sp. PI. 57. 1753), but erroneously, the plate being an illustration of a very different species from the type of P. dissectum in the Linnaean Herbarium.* Linnaeus later (Sp. PI. ed. 2. 81. 17(52) cites this phrase name and plate under Paspalum virgatum and also (op. cit. 1483), erroneously, under Andropogon fasciculatum. According to the Ameri- can Code of Botanical Nomenclature (Canon 10) the publication of Digi- taria Heist, as a genus would depend upon the specific description [from Sloane] "associable by citation with a previously published binomial species." While this specific description is cited under Panicum dissectum it is erroneously so; the species to which the description (as well as the figure) applies has no "previously published binomial," and therefore in the technical sense is not published. Sabsab Adans. Fam. 2 : 31, 599. 1763. No species are given. The tech- nical publication of the name is based upon the citation of " Paspalum. Lin." Cleachne Roland, in Rottb. Acta Lit, Univ. Hafn. 1:285. 1778. The name ' ' Cleachne R. " appears to be given as a synonym of Paspalum , three unpublished names of the latter being mentioned without description. Ceresia Pers. Syn. PI. 1 : 85. 1805. A brief diagnosis is given and a single species C. elegans Pers. (1. c. ) is cited. This name is based on "Paspalum membranaceum Lam. ill. gen. p. 177. t. 43. f. 2. Hab. in Peru." The generic diagnosis appears to be taken from Lamarck's specific description, but slightly rearranged. Lamarck's specimen has not been examined, but the description and figure indicate one of the species related to Paspalum stellaturn Plumb. & Bonpl. and probably that represented by Gardner's Plants of Brazil no. 4029 and Bang's Plantae Bolivianae no. 1080. Lamarck's name is preoccupied by Walter's use of P. membranaceum. Roemer & Schultes (Syst. Veg. 2 : 290. 1817) give P. elegans, based on Ceresia elegans Pers., as a synonym of P. mem- branaceum Lam., but this binomial also is preoccupied by P. elegans Fliigge (Gram. Monog. 183. 1810); P. commersonii Zucc. (in Roem. Collect, 122. 1809), which appears to refer to this species, is preoccupied by Lamarck's use of the same name. It seems probable that this species has somewhere received a name that will prove tenable, but as yet we have not found it. Trinius and Uoell use the name P. membranaceum Lam . Reimaria Fliigge. Gram. Monog. 21:!. 1810. The "character essen- tialis " given is as follows: " Caliz uniglumis, unifioris, valvae planiori appressus. Corolla plano-convexa, bivalvis," to which is added the observation " Paspalo affinis, at satis superque differt calice constanter uniglumi nee biglumi." It will be seen that the absence of both glumes is the only character given to differentiate this proposed genus from Paspalum. Three species are included, R. Candida Humb. & Bonpl., R. elegans and R. acuta, all here described for the first time. The first * For an account of the types of American grasses described by Linnaeus see Hitch- cock (Contr. Nat. Herb. 12 : 114-127. 1908). Chase — Notes on Genera of Panieea. IV. 139 two are species of Paspalum in which both glumes are wanting, the third has characters sufficiently distinct to be referred to a different genus. (See Reimarochloa Hitchc, of which R. acuta is the type.) Since the three species cited belong to two genera, it is necessary that the name Reimaria go with the larger group. (See Hitchcock, Contr. Nat. Herb. 12 : 198. 1909. ) Of these two species we take the first as the type species. This was described from a plant collected " Prope Pnanbo in America meridionali. Humboldt et Bonpland." Fliigge gives "Humboldt et Bonpland " as the authors of the species. Kunth later (Mem. Mus. Par. 2 : 68. 1815) transferred the specific name to Paspalum. A part of the type or a duplicate of it marked "ex Hb. Kunth & Hb. Humb." was examined in the Berlin Herbarium. It is the species distributed as Pas- palum candidum H. B. K. by John Donnell Smith and by the Herb. Instit. physico-geogr. nat. costaricensis, and represented in the National Herbarium by John Donnell Smith 4992, Tonduz 8-492 and several others collections from Costa Rica and Guatemala. Doell (Mart. Fl. Bras. 22 : 39. 1877) proposes a section Eremaehyrion for the species of Paspalum in which both glumes are wanting. Paspalanthium Desv. Opusc. 59. 1831. This genus, which is differen- tiated from Paspalum by the loose glume and sterile lemma exceeding the fruit and by the membranaceous rachis, includes a single species P. stoloniferum Desv., based on Paspalum stoloniferum Bosc. (Trans. Linn. Soc. 2 : S3, pi. Hi. 1794 " H. in Perua"). The type specimen was examined in the Bosc Herbarium at Padua. It is the species frequently cultivated under this name, with thin, conspicuously rugose sterile lemma. Moenchia Wender in Steud. Nom. Bot. ed. 2. 2 : 153. 1841, not Roth. 1788. There is no description and a single nomen nudum, M. speciosa Wender, is given as a synonym of Panicum sacrltaroides Kunth, upon which is based Paspalum saccharoides Nees. Anachyris Xees, in Hook. Kew Journ. 2 : 103. 1850. A single species, A. paspaloides Nees "In Brasilia. Gardner, n. 4031 in herb. Lindl.," is included. A portion of a raceme from the type specimen was kindly sent by the herbarium at the University of Cambridge for deposit in the National Herbarium. The species is of that small group including Paspalum malacophyllum Trin. and P. elongatum Griseb., in which both glumes are wanting and in which the very convex fertile lemma is longi- tudinally grooved. This accounts for the mistake made by Nees in placing the proposed genus in Oryzeae. In this group the fruit is rather readily detached from the spikelet leaving the thin sterile lemma attached to the rachis. It must have been such a fruit, mistaken for the entire spikelet, that Nees had before him when he described the genus. Steudel (Syn. PI. Glum. 1 : 33. 1854) spells this name Anachyrium. Maizilla Schlecht. Bot. Zeit. 8 : 601, 605. 1850. A single species is given, " M. stolonij 'era Bosc. sub Paspalo." Cymatochloa Schlecht. Bot. Zeit. 12 : 817, 821. 1854. Two species, " C.fluitans N[o]b. (Ceresia fluitans Ell. * * *) " and " C. repens 140 Chase — Notes on Genera of Panicece. IV. Nb. (Paspalum repens Berg.)," are included, the first of which is here taken as the type. Dimorpliostachys Fourn. Compt. Rend. Acad. Sci. Paris. 80 : 441. 1875. This genus is proposed because of the presence of the first glume of the spikelet, this ghune in the lower of the pair of spikelets being larger than in the upper. The author says the group contains eleven species, but four of which he mentions, Panicum monostachyum H. B. K. Paspalum pilosum Lam., Paspalum oajacense Steud., and Paspalum pedunculatum Poir. Fournier does not here actually transfer any species to Dimorpliostachys . His first named species, which we take as the type, was, together with the others given in his posthumous work (Mex. PI. 2 : 14-16. 1S86*), published under this genus by Hemsley (Biol. Centr. Amer. Bot. 3 : 499. 1885), as I), monostachya Fourn., based on Panicum monostachyum H. B. K. The section Harpostachys of Panicum established by Trinius (Mem. Acad. Petersb. VI. Sci. Nat. :l2 : 227. L834) for species having a simple, solitary raceme, if typified by the first group of species included in it, Panicum decumbens Roem. & Schult. (based on Paspalum ihcumbens Sw., of which P. pedunculatum Poir. is a synonym) and Panicum mono- stachyum H. B. K.j is synonymous with Dimorphostachys. In this sec- tion, however, Trinius included a number of remote species, the wholly artificial character on which it is based bringing together, besides the two species of Paspalum mentioned above, Thrasya paspaloides II. B. K. and congeners; Panicum repandum Nees, a species related to P. obtusum H. B. K., Ecliinolaena hirta, and Panicum sclerochloa Trin. and other species referable to Mesosetum. Wirtgenia Nees ; Doell in Mart, Fl. Bras. 1' 2 : 40. 1877. ' ' Wirtgenia pas- paloides Nees ab Esenb. in herb. Reg. Berolin.", a herbarium name for the species Nees published as Anachyris paspaloides, is here given as a synonym of Paspalum malacoph yllum Trin. The relationship of the various species upon which are based the inn- posed genera here included as synonyms of Paspalum together with the history of these groups will be discussed in the projected revision of North American species of Paspalum. Description. — Inflorescence of 1 to numerous, simple, spike-like racemes, along a common axis; spikelets plano-convex, usually obtuse, subsessile (rarely on pedicels as long as the spikelets) solitary or in pairs, in two rows on one side of a narrow or dilated raehis, the back of the fertile lemma turned toward it ; t first glume typically wanting ( regularly present in a few species, occasionally present in others; in a few species both glumes wanting); second glume and sterile lemma subequal, the glume rarely shorter; fruit usually obtuse, the lemma and palea chartaceous- indurated (rarely but slightly so), the margins of the lemma inrolled. * For discussion of date of this work see Hitchcock & Chase, Contr. Nat. Herb. 15 : 49. 1910. + Owing to a tortion of the short pedicels the crowded, paired spikelets are often turned edgewise to the raehis or even entirely reversed, but in all species examined the spikelet is attached with the back of its fertile lemma toward the raehis. Chose — Xatcs on Genera of Ponirea. IV. 141 Mostly perennials, of various habit; a large genus of the tropics and warm temperate regions of both hemispheres, but the species much more numerous in the western hemisphere. The characters here considered as of chief generic value are the strictly racemose inflorescence, the plano-convex (sometimes slightly concavo- convex) spikelets in which the first glume is wanting, and the obtuse, indurated fruit, the margins of the lemma inrolled, taken in combination. But in this large, on the whole well-marked genus, there are many species which depart more or less from some one or two of these characters. 17. Genus PANICUM L. Panicum L. Sp. PI. 55. 1753. This genus is discussed by Hitchcock & Chase (Contr. Nat. Herb. 15 : 11-18. 1910; and the type species shown to be Panicum miliaceum L. The generic names included as synonyms under Panicum, so far as these are based on North American spe- cies, are there accounted for. The genus in re- lation to the Smith American and Old World species upon which genera have been pro- posed will be dis- cussed in a later paper. In the above mentioned work, under the genera excluded from Panicum (op. cit. 16), is given Panicum tuerckheimii Hack., "an anomalous species with spikelets in which the first glume is wholly want- ing, and in which no rudiment of a palea is found in the sterile lemma." This statement is found to be partly erroneous. There is present a small hyaline first glume, so transparent as to be invisible in the dry spikelet, which escaped the notice of Hackel and of ourselves. This species, though unique, we now include in the genus Panicum. It will be described and the spikelet figured in a forthcoming paper (by Hitchcock & Chase) on the Mexican and Central American species of Panicum, a supplement to the recent revision of the genus. * Chasea Nieuwl. Amer. Midi. Nat. 2:63, (14. L911. This name is pro- posed as " nov. nom. Panicum of the authors not of Linnaeus or only Fig. 9. Panicum miliaceum. (Two views of spikelet and fruit x 10 diam.) 142 Chase — Notes on Genera of Panicecs. IV. in part." The author states that he " restores " the " name Panicum to the group of plants to which it was applied as far back as nearly two thousand years ago," overlooking Bubani's restoration of the same name to the same group in 1901 (Fl. Pyren. 4 : 261) and the fact that Adanson, Miller and Moench, applied the name Panicum in the same way. As shown in the revision of North American Species of Panicum (Hitchcock & Chase, Contr. Nat. Herb. 15 : 13. 1910) "the historic type species of Panicum is Chaetochloa italica " , and the pre-Linn?ean name for the genus containing Panicum miliaceum is Milium. Nieuwland, while maintain- ing Panicum for the pre-Linneean genus of this name, overlooks Milium Tourn., the historic name of the genus containing Panicum miliaceum, and also the post-Linnsean Urochloa Beauv. (Ess. Agrost. 52. pi. 11. f. 1. 1812.) based on U. panicoides Beauv., which is the same as Panicum helopus Trin. , an Old World species of the Fasciculata group of Panicum ; Thalasium Spreng. (Syst. Veg. 4 : cur. post. 30. 1827), based on a South American species allied to Panicum urvilleanum Kunth; Stein- cMsma* Raf. 1827; ErioUitruin Desv. in Kunth (Rev. Gram. 2 : 217. 1830), based on a South American specimen of P. urvilleanum or an allied species; and Plianopyrum Nash. The author (op. cit. 61) states that the " other group [Panicum L. excluding Panicum italicum and its congeners] has never to my knowledge received a name," and (op. cit. 63) that " This procedure leaves the other genus hitherto called Panicum by the authors, without a name, as far as I am able to ascertain, and I propose that of Chasea." Since this is proposed as a new name for " Panicum of the authors" and no particular authors are mentioned, t we take it as based upon Panicum as used by Beauvois (Ess. Agrost. 45, 169, 170, 171. 1812) who, besides recognizing Paspahtm L., Digitaria Hall., Cyno'lon and Pennisetum Pers., segregates Setaria and Echinochloa, leaving in Panicum the remainder of the species included by Linnaeus under that genus, that is the group of Panicum miliaceum and its 18. Genus ICHNANTHUS Beauv. IcKnanthus Beauv. Ess. Agrost. 56. pi. 12. f. 1. 1812. The genus is based on a single species, I. panicoides Beauv. (op. cit. 57), " crott dans 1' Amerique meridionale : elle m' a ete communiquee par M. Desfontaines. ' ' The generic description is erroneous in that Beauvois mistook the append- ages at the base of the fruit for an abortive floret placed, he says, contrariwise to the other florets. (It was from this supposed abortive floret that Beauvois derived the name Ichnanthus.) We have not seen Beauvois' specimen. The figure, though slightly inaccurate, is a good illustration of the species well described and figured by Kunth ( Rev. Gram. 2 : 245. pi. 34. 1830) under the name Ichnanthus panicoides. ♦This and Phanopyrum are discussed in Contr. Nat. Herb. 15 : 18, 118, 327. 1910. tit might be inferred from the name proposed that Hitchcock and Chase, authors of the North American Species of Panicum, were referred to, but this work is nowhere mentioned by Nieuwland, nor is the source of the proposed name given. CIkixi — Notes on Genera of Panicece. IV. 143 Navicularia Raddi, Agrost. Bras. 38. pi. 1. f. 5. 1823. Three species are included, N. hirta, N. glabra and N. lanala (op. cit. 40. pi. 1. f. 5). The last-named species, " Invenitur in herbidis prope Rio-Inhumirim,, , is here taken as the type since it is the only one of the three figured in the same work and because Raddi, in his generic description, refers to this figure. We have not seen Raddi's specimen, but from the description and figure we judge it to he the same as Panicum leiocarpum Sprong. (Ichnanthus leiocarpus(Sjpreng.) Kunth), to which Xees (Agrost. Bras. 147. 182!>) refers it, or a closely allied species. Raddi proposes the genus be- cause of the peculiar calyx-like structure which is borne on the hermaphrodite floret. The genus Ichnanthus has been recog- nized by some authors as valid and by others has been included in Panicum. Trinius at first (Fund. Agrost. 130. 1820) accepts Ichnanthus, even to the " trans- verse" abortive floret, but later (Gram. Pan. 52,53. 1826) he includes it in his Panicum, section Jubaria, with the ob- servation that it is included in Panicum, having, like "Pan. almadense, pseudagrostis, Hofmnnseggii, melicarium" , a hermaphrodite floret appendiculate at the base with a double rudiment. Nees (Agrost. Bras. 14'.). 1820) includes Ichnanthus panicoides, which he here renames Panicum ichnanthum, in Panicum, section Virgata, placing it next after Panicum leiocarpum and P. Hoffmannseggii, species now recog- nized as belonging in Ichnanthus. Nees describes the perfect floret as being embraced at the base by an ovate, appressed, papery, two-parted lamina. Kunth (Rev. Gram. 1 : 41. 1829) accepts the genus Ichnanthus with the single original species, and describes the fertile lemma as bearing at the base oblique-oblong, obtuse, glabrous scales decurrent on the pedicel. Later (op. cit. 2 : 245. 1830), while still including but the single, species, Kunth, in his observations, states that the supposed hypogynous scales of authors are analogous to the lodicules which are ordinarily found only at the base of the interior palet (palea) but are here found also and greatly developed, on the exterior palet (lemma). In a later part of the same work (op. cit. 2:508. pi. 168. 1831) he transfers to Ichnanthus Panicum leiocarpum Spreng., a species in which the appendages are promi- nent. Still maintaining the genus solely on account of the appendages at the base of the perfect floret, Kunth (Enum. PI. 1 : 135. 1833) adds Panicum almadense Xees to Ichnanthus. Fig. 10. Ichnanih us pan icoides. (Spikelet and fruit x 7% diam.) 144 Chase — Notes on Genera of Panicecs. TV. Trinius (Mem. Acad. St. Petersb. VI. Sci. Nat. 32 : 195, 320. 1834) makes Ichnanthus (spelling it " Ichnantus " ) a section of Panicum, with the synoptical heading " Flosculus * * * hermaphroditns basi faciei utrinque canalicnlato-scrobiculatns vel (plerumque) auriculato appendicu- latus," thus indicating the group as to-day accepted, including species in which there is a scar or excavation at base as well as those having appendages. Steudel (Syn. PI. Glum. 1 : 93. 1854) follows Trinius. Bentham (Fl. Hongkong. 41 3. 1861 ) adopts Ichnanthus as a genus in this emended sense, including in it I. pallens (Sw. ) Munro. Grisebach (Fl. Brit. W. Ind. 550. 1864) gives it, in the same sense, as a section of Panicum. Doell (Mart. Fl. Bras. 2 2 : 276. 1877) maintains Ichnanthus as a genus for the group segregated as Panicum, section Ichnanthus by Trinins, making under it two divisions "I Valvula inferior ad basin utrinque auriculatus" and ';II Valvula inferior ad basin utrinque scrobiculata." Bentham (Benth. & Hook. Gen. PI. 3 : 1103. 1883) and Hackel ( Engler & Prantl, Pflanzenf. 22 : 36. 1887) maintain Ichnanthus as a genus for this larger group. (Schultes, Mant. 2 : 281. 1824, misspells the name " Ischnanthus." ) Description. — Inflorescence paniculate, the spikelets mostly short- pediceled along the usually sub-simple branches; spikelets more or less laterally compressed, the glumes and sterile lemma strongly nerved; first glume usually more than half the length of the spikelet, broad, acute; second glume and sterile lemma subequal, acute, exceeding the fruit, the lemma enclosing a membranaceous palea and rarely a staminate flower; fruit acute or subacute, indurated, the margins of the lemma usually flat, the rachilla produced below the lemma into a usually minute stipe, this bearing on either side membranaceous appendages adnate to the base of the lemma and free above, the appendages sometimes wanting and indi- cated by minute excavations only. Perennials, usually with lanceolate blades abruptly contracted into a petiole-like base; the genus mostly con- fined to the. tropics of the western hemisphere, one species extending into the Old World. Ichnanthus is closely allied to Panicum and appears to be but a loosely coherent genus, several of the species differing from each other almost as much as some of them differ from species of Panicum. A few species, such as /. lanceolatus Scribn., in which the appendages are wholly want- ing and even the scars obscure ( but in which the lemma margins are flat), are nearly as referable to one genus as to the other. 19. Genus LASIACIS (Griseb.) Hitchc. Lasiacis Hitchc. Contr. Nat Herb. 15 : 16. 1910. "The type of the genus is Lasiacis divaricata (L. ) Hitchc, based on Panicum diraricatum elms, — Notes on Genera of Panice&. IV 145 Fig. 11. Lasiacis divaricata. (Spikelet and fruit x 10 diam.) L., the type of Grisebach's section." The author raises Panicum, section Lasiacis Griseb. (Fl. Brit. W. I ml. 551. 1S64) to generic rank. The type specimen of Panicum divaricatum L. (Syst. Nat. eel. 10. 2:871. 1759) was examined in the Lin- naean Herbarium. It was col- lected by Browne in Jamaica. Description. — Inflorescence of open, rarely compact, pan- icles at the ends of the culm and branches; spikelets sub- globose, placed obliquely on their pedicels, the glumes and sterile lemma broad, abruptly apiculate, papery-charta- ceous, shining, many nerved, glabrous, or lanose at the apex only, the first glume rarely over 3-3 the length of the spikelet, somewhat inrlated- ventricose, the second glume and sterile lemma subequal or the glume slightly shorter, the lemma enclosing a membranaceous palea and some- times a staminate (lower; fruit white, bony-indurated, obovoid, obtuse, both lemma and palea bearing at the apex in a slight crateriform exca- vation, a tuft of woolly hairs, the palea concave below, gibbous above, the apex often irw at maturity. Large perennials, usually with freely branching, w ly culms, often forming tall half-shrubs; leaf-blades firm, often narrowed to a petiole-like base; a genus of some fifteen species con- fined to the tropics and subtropics of the western hemisphere. This genus i< unusually well marked, the spikelets particularly of all the Species being strikingly similar. Besides the species placed in this genus by Hitchcock (Contr. Nat. Herb. 15 : 16. 1910 and Bot. Gaz. 51:301, 302. 1911) the following North American species, taken from the manuscript revision of the genus Lasiacis, are here transferred : Lasiacis liebmanniana (Fourn. ) Hitchc. Panicum liebmannianum Fourn. Mex. PI. 2 : 33. 188(3. Lasiacis oaxacensis (Steud. ) Hitchc. Panicum oaxacense Steud. Syn. PI. Glum. 1 : 73. 1854. Lasiacis ruscifolia (H. B. K.) Hitchc. Panicum ruscifolium H. B. K. Xov. Gen. A: Sp. 1 : 110. 1816. Lasiacis rhizophora (Fourn.) Hitchc. Panicum rhizophorum Fourn. Mex. PL 2:31. 1886. Lasiacis procerrima (Hack.) Hitchc. Panicum procerrimurn Hack. Oesterr. Bot. Zeitschr. 51 : 431. 1901.. 146 Chase — Notes on Genera of Panicecz. IV. 20. Genus SACCIOLEPIS Nash.* Sacciolepis Nash in Britton, Man. 89. 1901. Based on a single species, S. gibba (Ell.) Nash, which is the same species as Hole us striatus L. , the latter name being later transferred to Sacciolepis by Nash to replace . No locality other than Brazil is given. In the Trinius Herbarium is a packet of spikelets marked "Panicum longi- florum m. Brasil," and a specimen bearing the same name and also "no. 147. Lect. — ?" Panicum longispiculum Doell in Mart. Fl. Bras. 22 : 261. 1877. 'Hab- itat in Brasilia, loco accuratius non adnotato (herb. Acad. Petropolit. )." Doell cites " Paspalum longiflorum Trinius in Act. Petrop. 1835. p. 307 non Gmelin Syst. Veg. I. 158." This must be an error for Panicum since there is no Paspalum longiflorum of Trinius nor .of Gmelin. The page reference is also erroneous. Evidently Doell's name is a typonyni of Panicum longiflorum Trin. Ichnanthus longiflorus Benth. Journ. Linn. Soc. Bot. in : 45. 1881. This is based on " Panicum longiflorum Trin"., though from his statement that "in /. longiflora (Panicum longiflorum, Trin.) they [the auricles] are very 148 Chase — Notes on Genera of Panicese. TV. small, but prominent" it seems probable that Bentham had some other species under this name. The fertile lemmas in the spikelets of this species in the Trinius Herbarium are not at all auricled nor are the spike- lets, with their villi >us-margined second glumes and villous sterile lemmas, suggestive of Ichnanthus. Both Trinius and Doell note the affinity of this species to Panicum aturense. Homolepis longispicula differs from both the other species of this genus in having a densely silky-villous margin to the second glume, and a staminate flower in the first floret. The sterile (or staminate) lemma is densely villous, the fruit but little indurated as in H. aturensis. The only collection of this of which we know other than that in the Trinius Her- barium is Glaziou 22470 from Brazil. 23. Genus SCUTACHNE Hitchc. & Chase gen. nov. Inflorescentia paniculata ; spiculae breviter pedicellatae secus paniculae ramos teneros subsimphces dispositae, fusiformes, acuminatae, basi atten- uatae, rachilla cum internodiis elongatis; gluma prima internodio rachillae imo stipitiformi adnata, quam spicula dimidio brevior, mem- branacea, marginibus basin versus connatis; gluma secunda lemmaque sterile subaequilongae, internodio rachillae manifesto separata, coriacea indurata fusca, 5-nervia; lemma sterile palea et interdum staminibus praeditum ; fructus quam gluma secunda lemmateque sterili, vix indura- tior, circumscriptione ellipticus, lemmate in mucronem pubescentem abrupte angustato, marginibus basin versus leviter involutis, sursum planis pubescentibus, parte paleae summa haud inclusa, margine dense pubescente. Herbae perennes, tenues, culmis simplicibus, laminis linearibus, planis, et paniculis angustis. Nomen a o-kvtos corium vel pellis, et ax»"7 palea. Inflorescence paniculate, the spikelets short-pediceled along the slender, simple or nearly simple branches; spikelets fusiform, acuminate, the base attenuate, the internodes of the rachilla elongated, the lowermost form- ing a stipe, the first glume adnate to it ; first glume membranaceous, about half the length of the spikelet, broad, the margins connate below; second glume and sterile lemma subequal, a manifest internode of the rachilla between them, leathery-indurated, brown or brownish, 5-nerved, the lemma enclosing a palea of similar texture, (and in the type species a staminate flower); fruit but slightly more indurated than the second glume and sterile lemma, elliptical in outline, the lemma abruptly acum- inate into a slender, densely pubescent tip, the margins slightly inrolled below, membranaceous, flat and pubescent above, the summit of the palea not enclosed, densely pubescent on the margin. Slender perennials, with simple culms, linear, flat leaves and narrow panicles; known only from Cuba. Name from :'>, 35. 1887) also assigns it to this position. Description. — Inflorescence paniculate; spikelets oho void to subglobose ; glumes membranaceous, subequal, about as long as the fruits or at maturity exceeded by these; lower floret perfect or staminate, its lemma and palea indurated and similar in form and texture to those of the upper floret (scarcely indurated and dissimilar in /. trachysperma Sri-<); both florets (or fruits) plano-convex, obtuse, equal or nearly equal in size (the lower often larger when staminate only), the pair usually remaining attached together by the minute rachilla joint below the upper floret. Perennials with simple or branching culms and flat blades, the species confined to the tropics and warm temperate regions of both hemispheres. The lower rloret often appears to be sterile (not perfecting a grain) in some and fertile in other spikelets on the same panicle. When sterile the floret is often longer and the lemma less convex than when fertile, the spikelets on the same panicle thus having a somewhat diverse ap- pearance. Several species of Isachne bear a superficial though striking resemblance to species of the North American Panicum, subgenus Dichanthelium. 25. Genus HETERANTHOECIA Stapf. Heteranthoecia Stapf in Hook. Icon. PI. 30 2 : pi. 2927. 1911. The genus is based on a single species, //. isachnoides Stapf (1. c. ) collected in "Tropical Africa: Northern Nigeria; Xupe, in swamps Barter, 1348: French Congo; Snussi Country (Chari oriental), at the sources of the Xdelle River, < 'heualier, 6825." It is not stated from which collection the illustration was made. The genus differs from Isachne in having a racemose panicle, the subsessile spikelets in short racemes, these arranged along a common axis, and in having florets with lemmas dissimilar in form and texture, though both fertile. Stapf considers the genus inter- mediate between Isachne and Coelachne, the latter an anomalous genus of the Old World at present placed in Aveneae. 152 Chase — Notes on Genera of Panieece. IV. 26. Genus OPLISMENUS Beauv. Oplismenus Beauv. Fl. d'Oware ft Benin, 2 : 14. pi. 58. f. 1. 1S09.* Under the genus is described a single species, 0. africanus (op. cit. 15), of which Beauvois says "J'ai trouve cette espece a Chamo, a Koto, a Oware et a Benin." He states that the genus is composed of several species of Panicum and notably those of which Persoon has made a divi- sion thus characterized : " Spica composita, spiculis compressis secundis." The species included by Persoon (Syn. PI. 1 : 82. 1805) in this division are Panicum hirtellum, P. setarium, P. bromoides, P. cespitosum, P. loliaceum, P. compositum and P. (latins. The type specimen of 0. africanus has not been examined. The accompanying figure is made from Zenker d' Standi 515, collected in Kamerun, Africa, which is in- cluded in the region visited by Beauvois. (The generic name is misspelled " Ophis- menus" in Poir. ; Lam. Encyc. Suppl. 4: 271. 1816.) Orthopogon R. Br. Prodr. Nov. Holl. 194. 1810. Four species are given under the genus, 0. compositus, based on " Panicum compositum P.," 0. aemulus, 0. flaccidus, 0. imbecillis. None being figured in the same work and all being equally eligible the first is here taken as the type. Cheeseman (Man. Xew Zeal. Fl. 849. 1906) refers Hekaterosachne elatior Steud. (from New Zealand), the type of the monotypic genus Hekaterosachne Steud., to Oplismenus undulatifolius Beauv. Dal la Torre and Harms (Gen. Siphonog. 14. 1000) also refer this name to Oplismenus. We have not seen Steudel's specimen. His description (Syn. PI. Glum. 1:118. 1854) does not well agree with Oplismenus', though it is possible that his description is erroneous. Hippagrostis (Rumph.) Kuntze, Rev. Gen. PI. 2 : 776. 1891. Kuntze gives Rumpf as the author of the genus, with the date 1740. Hip- pagrostis Rumph. (Herb. Amb. ii : 14. pi. 5. f. :!. 1750) is based on a single species, H. amboinica Rumph. (1. a), which the plate shows to be a species of Oplismenus, probably 0. burmannii. Since a binomial species is given under the genus, only the fact that " Botanical nomen- clature is treated as beginning with the general application of binomial names of plants (Linnaeus' Species Plantarum, 1753) " t renders Hip- pagrostis invalid for our use. In the Index Fniversalis (Herb. Amb. Auctuarium, 1755) the name Hippagrostis amboinica is given with refer- ence to the 1750 work. Perhaps this might constitute publication. Fig. 15. Oplismt nus africanus. (Two views of spikelet and fruit x 10 diam. ) * The title page is dated 1S07. The date 1809 is taken on the authority of J. H. Barn- hart as stated in a letter from G. V. Nash, to whom we are indebted for a transcript of pages 14 and 15 of the second volume of the Flore d'Oware. + American Code of Botanical Nomenclature Part 1. principle 2. Chase — Notes on Genera of Vanicece. IV. 153 Beauvois ( Ess. Agrost. 54. 1812) amplifies his earlier generic descrip- tion and transfers to the genus several species from Panieum. He places Oplismi nus immediately after Echinochloa. Kunth ( 11. B. K. Nov. Gen. & Sp. I : L06. L816) recognizes OpUsmenus as a genus, including under it as synonymous Orthopogon R. Br. and Echinochloa Beauv. The same is done in his later works (Rev. Gram. 1 : 4:\. L829; Enum. PI. 1 : 138. 1833). Desvaux (Opusc. si. 1831) follows Kunth. but also includes Panieum semialatum in the genus. Trinius (Fund. Agrost. 181. 1820) recognizes Orthopogon R. Br. as a genus, referring OpUsmenus to it as a synonym. Raddi (Agrost. Bras. 40. 1823) recognizes OpUsmenus. Trinius (Gram. Pan. 51, 153. 1826; Mem. Acad. St. Petersb. VI. Sci. Nat. 3J : 20°. 1834) reduces Orthopogon to a section of Panieum. Nees (Agrost. Bras. 255. 1829; Fl. Afr. Aust. 00. 1841* gives OpUs- menus generic rank. Steudel (Syn. PI. Glum. 1 : 44. 1854) follows Trinius in making Orthopogon a section of Panieum ami giving OpUsmenus as a synonym. Hasskarl (Cat. PI. Hort. Bogor. Alt. 16. 1844) and Schlechtendal (Linnaea 111 : 263. 1861) spell the generic name Hoplismenus. Bentham (Fl. Hongkong 400, 411. 1801) includes OpUsmenus in Pani- eum, but later (Fl. Austr. 7 : 41)1. 1878) he gives it (with Orthopogon as a synonym) generic rank " with the limits originally assigned to it by Beauvois and by Brown," that is excluding Panieum crusgalli and allies referred to it by Kunth. Grisebach (Fl. Brit. W. Ind. o44. 18(54) recognizes Orthopogon as a genus for the group thus circumscribed, and Doell (Mart. Fl. Bras. 22 : 144. 1S77) includes Orthopogon, in the same sense, as a section of Pani- eum, giving OpUsmenus as a synonym. Bentham (Benth. & Hook. Gen. PI. ."» : 1077, 1104. 1883) gives OpUs- menus generic rank as does Hackel (Engler & Prantl, Ptianzenf. 22 : 33, 36. 1887). Founder (Mex. PI. 2 : 37. 1880) maintains OpUsmenus as emended by Kunth, that is, including Echinochloa. Dalla Torre and Harms (Gen. Siphonog. 14. 1900) give the name Paniculum Anl. as asynonym of OpUsmenus. The name " PANICULUM undulatifolium " occurs in Arduini (Animadv. Bot, Spec. 2:14. pi. 4. 1764), but the generic name is an error for Panieum. The name printed on the plate is Panieum undulatifolium. Description. — Inflorescence of few to many unilateral racemes, approx- imate or distant alonga common, often flexuous axis, the racemes bearing many crowded or subdistant spikelets, or sometimes very short and bearing a single cluster ; spikelets terete or somen hat compressed laterally, subsessile, in pairs or solitary in two rows on one side of a narrow, scabrous or hairy rachis; glumes subequal, emarginate or 2-lobed (rarely entire), awned or mucronate from between the lobes; sterile lemma ex- *The second edition of this work was published in lsf>:J with the title "Agrosto- graphia Capensis." The pagination is the same in both. 154 Chase — Notes on Genera of Panicece. IV. ceeding the glumes and fruit, notched or entire, mucronate or short- awned, enclosing a hyaline palea ; fruit elliptic, acute, the lemma very convex or boat-shaped, the firm margins clasping the lemma, not inrolled. Usually weak, freely branching, creeping or ascending annuals or peren- nials, with fiat, thin, lanceolate blades, the species confined to the tropics and warm temperate regions of both hemispheres. In this genus the awns are variable in length in the same species. 27. Genus ECHINOCHLOA Beau v. Echinochloa Beauv. Ess. Agrost. 53. pi. 11. f. 2. 1812. The first of several species of Pani cum listed under this genus, and in the index (page 161) transferred to it, and the one figured is E. crusgalli ( L. ) Beauv., which is taken as the type. This species is based on Panicum crusgalli L. (Sp. PI. 56. 1753) "Habitat in Europse, Virginia cultis." The only specimen in the Linnaean Herbarium to which Linnaeus has attached the name is one of the ordinary small form of this species.* This is marked " K " indicating that it was collected by Kalm in America. Adanson (Earn. PI. 2:496. 1703) pro- poses a genus Teraa with the differentiat- ing diagnosis " Corolla obtuse or aristate on the exterior scale of the sterile flower." No binomial is cited. The authority for the genus is given as " H[ort. ] M[alab] 12. t. 79" (which would indicate that Adanson adopts the genus from that work), and "Panicum Rumph. 5. t. 76. f. 2." and " Gramen paniceum spica divisa C. B." [Caspar Bauhin] are cited. The last citation refers to Panicum crus- galli L. and is also given by Linnaeus under that species (Sp. PI. 56. 1753). The plant described and figured in Rheede (Ilort. Malabar. 12 : pi. 79. 1703) is Chaetochloa italica (L. ) Scribn. and the name there given it is Tenna, of which Tenia would appear to be an erroneous transcript. The plant illustrated in Rumphius (Herb. Amb. 5 : pi. 70. f. 2. 1747) is Eleusine indica. If Adanson meant to unite these species in a single genus, as would appear to be the case, the name Tenia would go with the first reference, and, since this is not associable with a previously published binomial the genus is not technically published. It is possible that the reference given to Rumphius is an error for plate 75, figure 2, this being an illustration of Chaetochloa italica. Fig. 10. Echinochloa crusgalli. (Two views of spikelet and fruit x ~V. 1817), Schultes (Mant. 2 : 266. 1824), and Link (Hort. Berol. 2 : 208. 1833) give Echinochloa generic rank, hut the latter adds (op. cit. 209) " < ienera Echinochloa et Panicum artincialia sunt, nee natura distincta." Of the later authors Bentham (Fl. Hongkong 411. 1861; Fl. Austr. 7 : 47s. 1878; Benth. & Hook. Gen. PI. 3 : 1102. 1883), Grisebach (Fl. Brit. \V. End. 545. 1804), Doell (Mart. Fl. Bras. 22 : 139. 1877) and Hackel (Engler & Prantl, Pflanzenf. 2 2 : 35. 1887) include Echinochloa in Pani- cum. Fournier (Mex. PL 2 : 39. 1886) includes E. crusgalli and its close allies in OpHsmenus, but OpHsmenus holciformis H. B. K., a long- awned species allied to E. spectabilis (Nees) Link, he places in Berch- toldia (misspelled " Berchtholdia" ). In his key to Paniceae (op. cit. 3) Fournier distinguishes Berchtoldia from OpHsmenus by the "remote inferior glume." (Both genera are included under " Spiculis involucra- tis; involucro constante e spiculis abortivis," as opposed to "e chaeto- cladis" including Setaria, Pennisetum, etc. What Fournier could have mistaken for an involucre of abortive spikelets is not evident.) Nash (Britton, Man. 78. 1901; Small, FL Southeast. U. S. 84. 1903) and Hitchcock (Gray, Man. ed. 7. 117. 1908) recognize Echinochloa as a valid genus. Description. — Inflorescence paniculate, the usually compact, densely dowered panicle composed of one-sided simple racemes or of subsiinple branches; spikelets plano-convex, often spiny-hispid, subsessile, solitary or in irregular clusters on one side of the panicle branches ; first glume about half the length of the spikelet, pointed; second glume and sterile lemma equal, pointed, mucronate, or the glume short-awned, the lemma long-awned, in some species conspicuously so, enclosing a membranace- ous palea and sometimes a staminate flower; fruit plano-convex, the lemma and palea smooth and shining, acuminate-pointed, the lemma margins inrolled below, flat above, the apex of the palea not enclosed. Coarse, often succulent annuals, with compressed sheaths and linear, flat blades; species of the temperate and tropical regions, two species cosmo- politan. In this genus the awn of the sterile lemma is exceedingly variable in length, sometimes even in the same plant. Echinochloa is distinguished from Panicum constantly by the plano-convex, pointed fruit, the lemma margins flat above, the apex of the palea free, and usually by the awned sterile lemma. A Mexican species described under OpHsmenus belongs to this genus: Echinochloa holciformis (H. B. K. ). OpHsmenus holciformis H. B. K. Nov. Gen. & Sp. 1 : 107. 1816. "Crescit in humidis montanis prope Cinapecuaro, alt. 970 hexap. (Regno Mexi- 156 Chase — Notes on Genera of Panicece. IV. 1825. Based on 1841. Based on Based on "Op- cano.)" A duplicate type, received from Humboldt, was examined in the Willdenow Herbarium. Of the specimens in the National Herbarium Prinyle 8622 is an excellent match for this. The other specimens mostly have longer awns. Orthopogon holciformis Spreng. Syst. Veg. 1 : 307. " Oplismenus holciformis Kunth." Panicum holciforme Stench Nona; Bot. ed. 2. 2: 257 Oplismenus holciformis H. B. K. Berchtoldia holciformis Fourn. Mex. PL 1 : 41. 1886 lismenus holciformis H. B. K." 28. Gexus CHAETIUM Nees. Chaetium Nees, Agrost. Bras. 269. 1829. This genus is based on a sin- gle species, C. festucoides Nees (op. cit. 270), "Habitat in graminosis et in cultis ad flumen S. Francesci, ad Joazeiroetc. Provinciarum Pernam- bucanae et Bahiensis." The type specimen, in the Munich Herbarium, bearing the data as published, was collected by Martius. Berchtoldia Presl, Rel. Haenk. 1 : 323. pi. 43. 1830. A single species is included in the genus, B. bromoides Presl. (op. cit. 324). " Hab. in Mexico." A part of the type specimen was examined in the Trinius Herbarium. Both Nees and Presl place their proposed genera next after Oplismenus. Kunth (Enum. PI. 1 : 146. 1833) places Chaetium festucoides in Oplismenus but to Berchtoldia he gives generic rank. After the generic description he adds "(Charact. gen. ex Presl.)" It is probable that Kunth had not seen either species, but that from the plate in Presl's work he recognized Berchtoldia as distinct from Oplismenus. Steudel (Syn. PI. Glum. 1:48. 1854) places Chaetium in Panicum, section Echinochloa under the name P. chaetium, and gives (op. cit. 117) Berchtoldia generic rank, as did Kunth, and doubt- less for the same reason. Doell (Mart, Fl. Bras. 22:150. 1877) makes Chaetium a section of Panicum with the single species P. chaetium Steud. In the observations (op. cit. 150) he mentions as belonging in this section Berchtoldia bromoides Presl, to which he wives the name Panicum berchtoldium. Under P. chaetium he mentions as an extra-Brazilian speci- men Wright 735 from eastern Cuba. This number is the type of " Perotis f cubana" Wright, Chaetium cubanum (Wright) Hitchc. Bentham (Linn. Soc. Journ. Bot. 19 : 46. 1881; Benth. & Hook. Gen. Fig. 17. Chaetium festucoides. (Two views of spikelet and fruit x f> diam.) Chase — Notes on Gaunt of Panicets. TV. 157 PI. 3 : 1077, 1104. L883) recognizes Chaeiium as a valid genus, as does Hackel (Engler & Prantl, Pflanzenf. 22:33, 36. 1887) both placing it immediately after Oplismenus. Elemsley (Biol. Cent. Amer. Hut. 3:503. 1885) transfers Berchtoldia bromoidea to Chaetium giving Bentham as authority with a reference to the Linnaean Society's Journal mentioned above, but Bentham did not there transfer the species. Fournier ( Alex. PI. 2 : 40. 1886) uives Berchtoldia as a genus, including in it Presl's species, and also two species of Echiuochloa. Description. — Inflorescence a dense, narrow panicle; spikelets short - pediceled, dorsally compressed, lanceolate and having a long, slender callus-like base from the elongation of the joint of the rachilla between the glumes, the bearded base of the first glume adnata to it; glumes bear- ing awns 3 to 4 times the length of the body of the spikelet, the first reduced to the awn or, in C. bromoides, the pair broadened and enclosing the rest of the spikelet ; sterile lemma bearing a shorter awn or awn-tipped only, the sterile palea obsolete ; fruit subindurated, lanceolate, the lemma acuminate into a scabrous awn or point, the thin margins flat, the sum- mit of the palea not enclosed. Perennials with long, narrow leaves; the genus containing but three known species, one of Mexico and Central America, one of Cuba, and one of Brazil. 29. Genus TRICHOLAENA Schrad. Tricholaena Schrad. in Schult. Mant. 2 : 163. 1824. Three species are included in the genus but the second and third are preceded by a ques- tion mark, hence the first, T. micranlha Schrad., of which Saccharum teneriffse is given as a synonym, is the type. Rhyncht lytrum Xees in Lindley, Nat. Syst. ed. 2. 446. 1836. The genus is described and a single species, R. dregeanum, given. We have not seen the type specimen but the generic description applies to Tricliolaena. Stapf (Dyer. Fl. Cap. 7 : 444. 1898) refers R. dregeanum to T. rosea Nees. Monachyron Pari, in Hook. Niger Fl. 190. 1849. A single species, M. villosum, is included. The type specimen has not been examined. Hackel (Engler c*m Prantl, Pflanzenf. 22 : 36. 1887) gives this as a synonym of Tricholaena, and Durand and Schinz (Consp. Fl. Afr. 5 : 771. 1895) trans- fer M. villosurn to this genus. Making allowance for a misunderstanding of the structure of the spikelet, owing to the remote first glume, the de- scription applies to Tricholaena. The scarcely indurated fruit, scarcely firmer than the usually 2-lobed and awned second glume and sterile lemma, together with the elongation of the rachilla joint between the glumes, serve to distinguish this Old World genus, a single species of which, T. rosea Xees, is'sparingly escaped from cultivation in the tropics and subtropics of North America. 30. Genus CORIDOCHLOA Nees. Coridochloa Nees, Edinb. New Phil. Journ. 15 : 381. 1833. This genus is based on a single species, " Coridochloa * * * cujus typus est 158 Chase — Notes on Genera of Pankece. IV. Panicum cimicinum Retz." Nees states that the genus is allied to An- thaenantia Beauv., but is distinguished by the two-flowered spikelets, the fertile floret aristate. Benthain (Fl. Austr. 7:473. 1878) gives " Coridochloa semialata, Nees in various catalogues and herbaria" as a synonym of Panicum semialatum R. Br., and adds "(the genus not published as generally quoted in Edinb. New Phil. Journ. 1832, July)." It would seem that Benthain must have cited this name and reference from memory. We can not find that Xees himself ever placed P. semialatum in Coridochloa, nor can we find any reference to the publication of the genus in 1832. It was probably a slip of memory on Bentham's part for is:;:',. The genus Coridochloa is proposed in a footnote with the statement that it will be treated of at another time, hence Bentham may have considered it as insufficiently published. But Xees states that the genus belongs in the tribe Panicex, names a type species, previously described, and gives the principal distinguishing characters of the spikelet. In Wallieh's Cata- logue (1849) no. S749 "Coridochloa fimbriate Nees ab Esenbeck " is listed, "A. Milium cimicinum HI). Heyn." and " B. Panicum cimicinum Hb. Ham." being given under it. (In a note on page 132 of the Cata- logue it is stated that " Mr. Brown * * * has had the goodness * * * to furnish the provisional list of the family" of grasses). In a criticism of Nees, Bentham (Linn. Soc. Journ. Bot. 19 : IS. 1881) again says "Brown's Australian Panicum semialatum, for instance, is raised by Nees to the rank of a genus under the name of Coridochloa in India, and that of Bluffia in South Africa." It may be that Bentham did not know Panicum cimicinum Retz., for (op. cit. 42) he says "P. semialatum Br. is widely spread over the Old World, for I am unable to distinguish the Asiatic Coridochloa, Nees, and the South African Bluffin, Xees, from Brown's Australian species." Hooker (Fl. Brit. Ind. 7 : 64. 1896) places Panicum cimicinum and P. semialatum under Axonopus Beauv. (See discussion under that genus.) Under the second species is given as a synonym " Coridochloa semi-alata, Nees, in Edinb. New Phil. Journ. XV. (is:;:',) 381." This is an error, Nees makes no mention whatever of P. semialatum nor of any species but P. cimicinum Retz. This unique species has been placed in Milium (by Linnaeus) in Panicum (by Retzius), in Axonopus (by Beauvois), in Urochloa (by Kunth), and dually in Coridochloa, based on it alone, by Xees. Coridochloa, which is perhaps rather remotely allied to the next genus, is distinguished by the concavo-convex, scarcely indurated, stipitate fruit, the lemma attenuate into an awn about as Ion- as the body of the fruit, the palea sparsely covered with stalked glandular hairs ( in appearance like minute fungi), a few of these sometimes scattered on the margin oi the lemma; by the papery glumes and sterile lemma, the second glume conspicuously stifi'-ciliate along the lateral internerves, and by the digitate inflorescence, the slender racemes naked at the base or for half their length. Chase — Notes on Genera of Panicees. IV. 159 31. Gents ALLOTEROPSIS Presl. Alloteropsis Presl, Rel. Haenk. 1 : :144. pi. 47. 1830. A single species, I. distachya, is included in the genus which is erroneously described. Scribner (.Mem. Mo. Bot. Gard. L0 : 37. pi. :;:;. 1899) and Hitchcock (Contr. Xat. Herb. L2 : 210. 1909) explain the error and emend the genus. Presl's type specimen was examined and photographed in the National Museum at Prague by Professor Hitchcock. It is the same species as Panicum sernialatum R. Br. (Prodr. Nov. Holl. L92. L810), Alloteropsis semialata Hitlchc. (1. c. ). There are two plants of the same species on the sheet, one with a ticket marked " Peruana? montanse " the other with one marked " Regio montana, Luzon." The Peruvian locality is clearly erroneous. Bluffia Xees, Del. Sem. Hort. Hamb. 8. 1834. The genus is described and a single species, B. eckloniana Xees, "ah Ecklono * * * in Africa austral i deteeti," included under it. By Hackel ( Durand & Schinz, Consp. Fl. Afr. 5 : 7<>4. L895), this species is reduced to a variety of Panicum sernialatum ; by Hooker (Fl. Brit. Ind. 7 : 04. 1896) it is referred to Axonopus sernialatus as a synonym, and by Stapf (Dyer, Fl. Cap. 7 : 418. 1898) it is reduced to a variety of that species. To us it appears to be specifically distinct. Holosetum Steud. Syn. PI. Glum. 2 : US. 1854. This is based on a single species, H. philippicum Steud. " Herhr. Cuming nr. 1363 et 1414. Ins. Philip." Cuming's no. 1363 was examined in the Kew Herbarium. It proves to be Alloteropsis semialata. This genus of two known Old World species does not appear to be closely allied to any other. The two subindurated, awn-pointed florets to some appear to suggest species of Arundinella, under which genus Bentham (Fl. Austr. 7 : 545. 1878) describes a specimen of Alloteropsis semialata, as A+undinella Schultzii Benth., though he gives Panicum sernialatum, on page 472 of the same work. In both species of this genus, as shown by specimens in the National Herbarium, there is not infre- quently found a rudiment, 0.3 mm. or more long, beyond the palea of the fertile floret. The genus is distinguished by the awn-pointed, similarly subindurated staminate and fertile florets, the margins of the fertile lemma thin, flat, the palea not enclosed at the summit, in combination with the subdigi- tate inflorescence, the short-pediceled spikelets in clusters along the racemes. The involucrate genera and the others excluded under the first and second divisions of the key will be considered in a subsequent paper now in preparation. Vol. XXIV, pp. 161-162 June 16, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A NEW BAT FROM THE CAROLINE ISLANDS. BY GERRIT 8. MILLER, JR. [Published by permission of the .Secretary of the Smithsonian Institution.] Two bats of the genus Emballonura from the Caroline Islands represent a species distinct from any hitherto described. Emballonura sulcata sp. now Type. — Adult (skin and skull), No. 151,568, U. S. National Museum. Collected on Uola Island, Truk group, Caroline Islands, February 16, 1900, by H. F. Moore (U. S. Bureau of Fisheries). Description. — Externally like Emballonura semicaudata from Samoa except that the size is appreciably greater (forearm 48.5 instead of 45, tibia 19 instead of i'7) ; general color prouts brown, slightly washed with drab below; skull decidedly larger than that of E. semicaudata (greatest length 17-17.4 instead of 14.4-15, but not peculiar in general form or in proportion of parts, except that anterior palatine emargination is somewhat deeper and mesopterygoid space wider; dorsal surface of ros- trum with conspicuous median longitudinal sulcus nearly 1 mm. wide extending from nares to front of sagittal crest and completely separating the inflated lateral areas. In E. semicaudata the inflated areas are broadly in contact along median line, so that longitudinal groove is nearly effaced. Teeth larger than those of the related species but with no appreciable peculiarities of form. Measurements. — The two specimens give the following measurements, those of the type standing first: tail, 12 (14); tibia, 19.2 (19): foot, 8.2 (8.2); forearm, 48.4 (48.6); thumb, 8.2(8.6); third finger, 78(77); fifth finger, 52 (53) ; greatest length of skull, 17.0 ( 17.4) ; condylobasal length, 15.2 ( — ); zygomatic breadth, 10.0 (10.0); lachrymal breadth, 7.0 (7.0); postorbital constriction, 3.4 (3.6); breadth of braincase, 8.0 (8.0); man- dible, 12.2 (12.2); maxillary toothrow (exclusive of incisors, 6.6 (6.8). Remarks. — In size this species lies midway between Emballonura semi- caudata and the recently described E. fura.c Thomas, of New Guinea. Its cranial characters are, however, strictly of the normal type. 2(i— Proc. Biol. Soc. Wash., Vol. XXIV, 1911. (161) Vol. XXIV, pp. 163-164 June 16, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A NEW BLUE GROSBEAK FROM CALIFORNIA. BY JOSEPH GRINNELL. Early in 1910 an expedition was sent by the Museum of Vertebrate Zoology of the University of California to the lower Colorado valley, in California and Arizona. While working up the extensive collection of birds obtained, my attention was arrested by the appearance of the blue grosbeaks from that region. Their bills are uniformly very much larger than those of the form previously familiar to me and occurring in summer on the Pacific slope of southern and central California. Examination of specimens and literature shows that the large-billed form is the one already designated and that it is the bird of western California that needs to be named. Guiraca caerulea salicarius subsp. nov. CALIFORNIA BLUE GROSBEAK. Type. — No. 3276, Univ. Calif. Mus. Vert. Zool. ; Santa Ana River bottom, near Colton, San Bernardino Co., Calif.; July 21, 1908; C. H. Richardson, Jr., collector. Diagnostic characters. — Similar to Guiraca caerulea lazula, of Arizona and Mexico, in coloration and general size, but bill much smaller and proportionally less tumid, that is, outlines straighter; compared with Guiraca caerulea caerulea of the South Atlantic States, blue color of the male paler throughout, bill smaller, and wing and tail longer. Measurements of type. — Wing, 90.7 mm. ; tail, 72.6; tarsus, 20.0; culmen, 15.0; bill-from-nostril, 11.4; depth of bill at base, 11.9; gonys, 9.5; greatest outside width of corneous portion of lower mandible, 10.4. Habitat. — In summer, the Lower Sonoran zone of central and southern California west of the Sierran divide; in winter, unknown. Remarks. — This form was characterized, but not named, by Ridgway (Birds N. ct Mid. Am., I, 1901, p. 610). It is paralleled in many other passerine genera of the southwest, and since the characters are conspicu- ous to the trained eye and are fairly constant, the only wonder is that the race has not been provided with a name before. 27— Proc. Biol. Soc. Wash., Vol. XXIV. 1911. (163) Vol. XXIV, pp. 165-166 June 16, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A NEW .MOUSE-DEER FROM THE RHIO-LINGA ARCHIPELAGO. BY GERR1T S. MILLER, JR. Published by permission of the Secretary of the Smithsonian Institution.] When describing the napu of Pulo Bakong* I supposed that the animal of Pulo Sebang was identical with it. Further com- parison of the material shows, however, that the forms occurring on tbf two islands must be regarded as distinct. Tragulus pretiellus parallelus subsp. nov. Type.— Adult male (skin and skull), No. 123,051, U. S. National Museum. Collected on Pulo Sebang, Rhio-Linga Archipelago, July 30, 1903, by Dr. W. L. Abbott, Original number, 2694. Diagnosis. — Like Tragulus pretiellus pretiellus of Bakong Island, but back loss suffused with black and middle region of underparts less tinged with huffy and never with any decided grizzle due to the presence of black-tipped hairs. Measurements. — Type: head and body, 513; tail, 67; hind foot, 129 (125); condylobasal length of skull, 100.6; zygomatic breadth, 48.6; mandible, 84.8; maxillary cheek-teeth (alveoli), 34.2; mandibular cheek- teeth (alveoli), 39.6. Specimens examined. — Sixteen, all from Pulo Sebang. Remarks. — Though readily distinguishable from Tragulus pretiellus pretiellus when both forms are considered as a whole, the Sebang animal is not completely differentiated from thai of the nearbyisland of Bakong. Of the sixteen specimens of parallelus eleven differ from all pretiellus in the absence of a definite huffy area on middle region of underparts. Of the nineteen specimens of pretiellus fifteen differ from all parallelus in the presence of a noticeable blackish grizzle on buff of upperparts. Four- teen -kin- of pretiellus show a heavy black suffusion on underparts, while in three the suffusion is not conspicuous. Five skins of parallelus show a heavy suffusion of black on underparts, while in thirteen it is not con- spicuous. *1906. Tragulus pretiellus Miller, Proe. I". S. Xai. Mus., Vol. XXXI, p. 253, Sept. 11, 1 i 28— Proc. Biol. Soc. Wash., Vol. XXIV, 1911. (165) Vol. XXIV, pp. 167-170 June 16, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON DESCRIPTIONS OF FOUR NEW TREESHREWS. BY MARCUS WARD LYON, JR. [By permission of the Secretary of the Smithsonian Institution.] Daring his recent Zoological explorations of the Malayan Archipelago, Dr. W. L. Abbott collected, among numerous other mammals, representatives of four species of treeshrews which have not been previously described. They may be known by the following descriptions. Tupaia raviana sp. nov. Type.— Skin and skull, old adult male, Cat. No. 104,355, U. S. N. M., Pnlo Rawi, Butang Islands, off west coast Malay Peninsula, Dec. 18, 1899, Dr. W. L. Abbott. Grig. No. 172. Diagnostic characters. — Related to T. belangeri of the neighboring mainland and T. lacernata of Pulos Langkawi and Terutau, but skull generally wider, and rostrum especially thicker and wider, interorbital region wider, zygomata more spreading; skin rather intermediate between the two, but more like that of T. belangeri. Color. — Type: Upperparts of head, neck, body, and outer side of legs a grizzle of a color between buff and olive-buff, and blackish, the two colors about equally mixed, and the huffy color inclining toward ochraceous buff in the region of the thighs; underparts generally huffy or an extremely pale gallstone yellow; tail a hue grizzle of cream color and blackish, the latter predominating above and both about equally prominent below except in the middle line where the lighter color is in excess; shoulder stripe not conspicuous, dull cream color. Skull. — In general like that of T. belangeri and T. lacernata, but the skull is generally wider, with thicker anil wider rostrum and more spread- ing zygomata. This shows more differences from the related forms than does the skin. Measurement*. — Type, and Cat. No. 123,984, V. S. N. M., Tupaia lacernata, an old adult male from Pulo Terutau: Head and body, 184, ISO; tail, 165, 145; hind foot (dry), 40, 44; condylobasal length of skull, 46.5, 46; zygomatic width, 25.5, 24.5; width of braincase above 29— Proc. Biol. Soc. Wash., Vol. XXIV, 1911. (167) 168 Lyon — Descriptions of Four New Treeshrcws. zygomata, 19.19; maxillary toothrow, alveoli, 17.5, 18; width of rostrum back of incisors, 7, 6.5. Specimens examined. — Two. the type and Cat. Xo. 104,354, U. S. X. M., from Pulo Adang, Butang Islands. Tupaia pemangilis sp. nov. Type.— Skin and skull, adult female, Cat. No. 112,499, U. S. N. M., Pulo Pemangil, east coast Malay Peninsula, June 12, 1901, Dr. W. L. Abbott, Orig. No. 1064. Diagnostic characters. — A member oi the ferruginea group closely" related to T. sordida from Tioman, but slightly duller in color, with slightly grayer feet, and with anterior portion of underparts more buffy, much like that of T. pulonis from Pulo Aor. ; skull like that of sordida, but slightly narrower, and with m3 rather smaller. Cotor. — Type: Upper parts a grizzle of dull tawny ochraceous and blackish, brightest on posterior parts of body, and lightening anteriorly on neck and head to a grizzle of raw sienna (or gallstone yellow) and blackish; outer side of forelegs like top of neck; outerside of hindlegs, like lower hack; shoulder stripe moderately conspicuous, between bun and olive buff; throat and chest pale dull buffy; inner sides of legs dull buffy with darker bases of the hairs showing through; tail above a uniform grizzle of cream color and blackish, both colors about equally prominent; below a coarse grizzle of dull buff and black, the buff color predominating along central line ; hands and feet a fine grizzle of the colors of the adjacent parts. Skull and teeth. — Very similar to those of Tupaia, sordida from the nearby Pulo Tioman, but skull slenderer, and with to3 rather smaller, and with slightly smaller bullae. Measurements. — Type and the type of Tupaia sordida. Cat. Xo. 101,747, U. S. X. M.: Head and body, 185, 171; tail, 150, 165; hindfoot (dry), 40, 41 ; condylobasal length of skull, 45, 46; zygomatic width, 23, 25.5; width of braincase above roots of zygomata, 19, 20; maxillary toothrow, alveoli, 17.5, 18.5. Specimens examined. — One, the type. Remarks. — Tupaia pemangilis is only a slightly differentiated form, and were it not for its insular isolation could not he considered more than a subspecies. It is most closely related to T. sordida of Pulo Tioman and is not distantly removed from T. pulonis of Pulo Aor. The three islands Aor, Tioman, and Pemangil are thus inhabited by Tupaias of the ferru- ginea group differing slightly from each other and as a whole from the mainland form. Tupaia natunae sp. nov. Type. — Skin and skull of rather old adult female, Cat. Xo. 104,714, U. S. X. M., collected on Bunguran, Xatuna Islands, June 27, 1900, by Dr. W. L. Abbott, Orig. No. 514, Lyon — Descriptions of Four New Treeshrews. 1G9 Diagnostic characters. — Related to Tupaia splendidula of Borneo from which it differs in a generally brighter and more reddish coloration of the upperpartSj sides, legs and tail, and more inflated braincase. Color. — Type: General color effect of upperparts of neck and body, in fresh pelage, most like Ridgway's burnt sienna, but rather brighter, this color being produced by a wide band on most of the hairs, of a bright ferruginous burnt sienna mixture, with blackish bases and a considerable number <>f long blackish hairs; on rump, where an old pelage persists, the general color darker and duller; sides of body and outer side of legs similar to upperparts, but showing a tendency to grizzling; top of nose a grizzle of raw sienna and blackish gradually blending in on top of head with color of upperparts; cheeks and sides of neck an indistinct grizzle of blackish and buff olive; shoulder-stripe buffy to ochraceous buff; an indistinct huffy ochraceous eye ring; underparts generally a color between buff and olive-buff; inner side of legs essentially like sides of body, but rather lighter; tail above generally like back, but in dull worn pelage; underside of tail and bases of tail hairs generally, tawny ochraceous, with the outer and terminal margins of tail dark tawny. A paratype, a barely adult female, is generally like the type in color, but shows distinct griz- zling on sides of body, on neck, shoulders and thighs, and indistinct grizzling on upperparts generally. Skull and teeth. — These are of the same general form as they are in T. splendidula, the skull averages longer, however, has a more inflated braincase, and the teeth are distinctly larger. Measurements. — Type and those of Cat. No. 151,883, T. splendidula, from Klumpang Bay, southeastern Borneo; head and body, 184, 189; tail, 140, 130; hindfoot, 40, 41; condylobasal length of skull, 4(i, 44; zygomatic width, 25, 25; width of braincase above roots of zygomata, 19, 17.5; maxillary toothrow (alveoli), 18.5, 17. Specimens examined. — Two, the type, and skin and skull, Cat. No. 104,715, a young adult female, also from Bunguran Island. Tupaia sincepis sp. now Type.— Skin and skull of adult male, Cat, No. 123,105, U.S. N. M., collected on Pulo Singkep, Rhio-Linga Archipelago, August 8, 1903, by Dr. W. L. Abbott, Orig. No. 27:12. Diagnostic characters. — A slightly differentiated form of Tupaia malac- cana with a more rufescent color on the back, and with a darker and more blackish tail. Color. — Type: Upperparts of head, neck, and body, and outer side of legs, a line grizzle of buff and blackish, the buff color being gradually replaced by tawny ochraceous on the lower back, rump, and base of tail; underparts varying from pale buff, or cream color anteriorly to dirty whitish posteriorly; inner side of legs similar to adjacent portions of underparts; tail, above, an indistinct grizzle of ochraceous buff and black, the latter color in excess especially along margin and tip of tail; middle portion of underside of tail cream buff finely lined with blackish, outer 170 Lyon — Descriptions of Four New Treeshrews. portion and tip generally blackish with a slight admixture of buff or ochraceous buff; upper surfaces of feet ochraceous buff with the dark bases of the hairs showing through ; eye ring, fairly well defined buffy ; shoulder stripe, well marked, whitish. The rest of the series are essen- tially like the type, but three of them are more tawny about the rump and base of tail. Skull and teeth. — These show no special characters by which Tupaia sincepis can be distinguished from T. malaccana. Specimens examined. — Six, the type and five others, all from the island of Singkep. Measurements. — Head and body, 140; tail, 165; hind foot, 37; condylo- basal length of skull. 36.5; zygomatic width, 20.5; width of braincase above roots of zygomata, 16.5, maxillary toothrow (alveoli), 13. Remarks. — Tupaia sincepsis is very closely related to T. malaccana from the Malay Peninsula. Specimens from the islands of Linga and Sumatra appear to be rather intermediate between the two forms, those from Sumatra being more like the Peninsular animal. The Linga speci- mens resemble T. sincepsis in having a tawny color on the back, but lack the darker tail. Vol. XXIV, pp. 171-174 June 16, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON RECENT ADDITIONS TO THE FISH FAUNA OF THE DISTRICT OF COLUMBIA. BY BARTON A. BE AX AND ALFRED C. WEED. [Published by permission of the Secretary of the Smithsonian Institution.] Since the publication of the list of fishes of the waters of the District of Columbia and vicinity by Smith and Bean* several species not there listed have been taken. Some of these have been introduced by agents of the Bureau of Fisheries, others which were present were overlooked in the making of the col- lections previously listed. In view of the changes in fish fauna which are being made by the introduction of new species in various waters it seems advisable to publish, as soon as possible, notes on the appear- ance of species in waters foreign to their natural habitat. According to U. S. National Museum records the large mouth black bass was present in the Potomac long before the elate given by Smith and Bean (1889). No. 16,841 of the Museum register reads: ' Black bass, M. salmoides, upper Potomac, weight 4 lbs. 11 ozs., Major Hobbs, Washington, D. C." This fish was cast in plaster by Mr. Jos. Palmer, cast No. 602, date November 1, 1876. We have recently seen the mold of this specimen and there is no doubt as to its identity. Ictalurus furcatus (La Sueur). Dr. Hugh M. Smith kindly furnishes the following note on the occur- rence of this species in the District of Columbia: ' In 1905 the fishermen about Washington began to catch this species, and samples were submitted to the Bureau of Fisheries for identification. There is no record of the introduction of the species and it is evident that it was not distinguished from /. punctata* at the time the young were planted. Specimens weighing over thirty pounds have been reported." •Bull. IT. S. Fish. Com., vol. 18 (1898), pp. 179-187. 30— Proc. Biol. Soc. Wash., Vol. XXIV, 1911. (171) 172 Bean and Weed — Fish Fauna of the District of Columbia. Schilbeodes gyrinus (Mitchill). A specimen of this little Mad Tom was taken in Little Beaver Dam Branch, Anacostia River, May 25, 1911, by Mr. Weed. Pimephales notatus ( Rafinesque). BLUNT-HEAD MINNOW. Exceedingly common in the pools of all the swifter streams about Washington. Mention of this species was omitted by error from the pre- vious list. Notropis arge (Cope). Many specimens provisionally identified as this species have been taken in the lower portion of Cabin John Run. These specimens may prove to be Notropis photogenis if the two species are finally proven to be specific- ally separable. These fishes may have been introduced from the Ohio River drainage. A very large number of fishes of many species were taken in Cabin John Run December 4, 1909, and March 19, 1910. These hauls were made with a twenty-foot bag seine (Baird collecting seine) under the Chesapeake & Ohio Canal. The haul that was made in December took many thousand fish, about half of which were Pimephales notatus and the rest various small minnows. All the fish were unusually silvery in color. Notropis amoenus (Abbott). This species has been taken in moderate numbers in Cabin John Run and in some of the other streams flowing into the Potomac River. It is listed by Evermann and Hildebrand from the Mattapony River, Virginia.* Ericymba buccata Cope. This common western minnow occurs in fairly large numbers in the lower portion of Cabin John Run. It was probably introduced in the upper portion of the Potomac River by the U. S. Bureau of Fisheries. Lepomis cyanellus Rafinesque. This western sunfish occurs in very large numbers in pools in the river in the neighborhood of Chain Bridge. Few large specimens are seen but very many ranging from one to three inches in length. Ten specimens, each about three inches in length, were caught in considerably less than ten minutes, in a small pool. All were taken on hook and line, using a single small piece of angleworm as bait. This species was probably intro- duced into the upper portion of the Potomac River. t * Proc. Biol. Poc. Wash., vol. XXIII. p. 158. t First published record of L. cyanellus, Bean ami Weed. Proc. V. S. Nat. Mus.. 40. p. 376, plate 50. Bran and Weed — Fish Fauna of the District nf Columbia. 173 Lepomis auritus ( Linnaeus). Tt is probable that there are two species confused under this name in the waters of the District of Columbia and vicinity. Living specimens of both may be seen in the aquarium of the Bureau of Fisheries at Wash- ington, D. C. Botli are distinguished by the great length of the opercular nap but otherwise are very dissimilar in color. One is a rather slender fish having the body-color and color changes of Lepomis cyanellus. It is very difficult to distinguish this form from Lepomis cyanellus except by the character of the < ipercular flap which is long and pointed, extends obliquely upward and has a straight light colored line along its upper and its lower edge. These lines extend well on to the operculum. The other form is much deeper bodied and has the body-color of Lepomis gibbosus.* The opercular flap extends horizontally, is about as wide at its distal end as at its proximal end and appears rather truncate. Two forms have previously been described under the names Lepomis auritus and Lepomis auritus soils. It appears rather probable that the slender form should stand as a separate species under the name Lepomis soils. Lepomis pallidus (Mitchill). This species is taken in the Potomac River, in the Tidal Basin and in the lower portion of the Eastern Branch. This species may be native rather than introduced, as it is an inhabitant of the deeper waters where ordinary scientific collecting might easily overlook it. Doctor Smith furnishes the following note on the occurrence of this species in the Potomac River as early as 1900: " Extract from letter from R. Hessel, May 2, 1900. "In regard to the Lepomis pallidus I sent the other day to Central station, I have the following to state: Four years ago, in June, 1896, I went with some of the men of the station to the Potomac to get some liv- ing food for the black bass. Hauling the large seine I caught about 1,600 young fish — sunfish, catfish, roaches— and amongst the sunfish — about 500 of % of an inch in length — I noticed three small sunfish of the same size, but of a little different shape from the common sunfish ; they seemed to be of more slender form and had a little darker coloration. I exam- ined all the fish I had in the tub and found 14 in all. I tried to get some more of them and made a few more hauls, but could not find any more. "I put the 14 young fish in a small pond to see their development. Four of them died and the others developed very well. I feed them on mashed fish, and last year they spawned and I have now about 2,000 on hand." These specimens were identified by Dr. W. C. Kendall. Stizostedion vitreum (Mitchill). The Bureau of Fisheries has planted a large number of pike perch fry in the Potomac (during the years 1901 to 1904) and a number of speci- mens have been taken. ♦See: Bean and Weed, Proc. Biol Soc. Wash., Vol. XXIV, p. 73. 174 Bean and Weed — Fish Fauna of the District of Columbia. Boleosoma longimanus (Jordan). Many specimens of this species, which was previously known only from the headwaters of the James River in Virginia, have been taken in the Northwest Branch about one mile above Hyattsville, Maryland. It occurs in company with Boleosoma olmstedi but seems to prefer rather deeper water. All the specimens were taken in a stretch where the water, about two feet deep, hows rapidly and without ripples over a bottom of medium sized gravel (stones one to four inches in diameter). Etheostoma flabellare Rafinesque. Two specimens have been taken in Cabin John Run. Cat. No. 64,398, U. S. Nat. Mus., was taken December 2, 1009, and No. 06,330 March 19, 1910. Vol. XXIV, pp. 175-178 June 16, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON THE LOUISIANA PUMA. BY N. HOLLISTER. [Published by permission of the Secretary of the Smithsonian Institution.] An attempt to identify the specimens of pumas in the collec- tion of the United States National Museum has resulted in the discovery that the form found in Louisiana is very different from any yet described. This new form is represented in the museum collections by three specimens; two skulls from Mer Rouge, Morehouse Parish, collected by J. Fairie many years ago, and a skin and skull in the Biological Survey collection from Vidalia, Concordia Parish. The Louisiana puma is much more closely related to the puma of Florida than to the northern form or to Felis oregonensis azteca. It is readily separable from both the first and the last species by very marked color differ- ences. The museum is fortunate in possessing two skulls each of Felis couguar from New York and Felis corgi from Florida, as well as good series of the western species. The pelage of the Florida puma has been carefully described by Mr. Chas. B. Cory,* by Mr. Outram Bangs,! and by Dr. C. Hart Merriam,+ and Mr. Bangs has proved that it varies little in color with the season. From the records of the occurrence of the puma in the Southern States collected and published by Dr. F. W. True § it seems certain that the distribution between Florida and Louis- iana was at one time continuous. As so little is known of the animal formerly inhabiting that region, however, and as the Louisiana puma differs so greatly from the Florida animal in color it seems best to regard it at present as a distinct species. * Hunting and Fishing in Florida, p. 109, 1896. t Proc. Biol. Soc. Washington, XIII, pp. 15-17, 1899. tProc. Washington Acad. Sci., Ill, pp. 583-585, 1901. § Report U. S. Nat. Mus., 1888-'89, pp. 591-608, 1891. 31— Pkoc. P.iol. Soc. Wash., Vor.. XXIV, 1911. (17.-,) 176 Hollister — The Louisiana Puma. Felis arundivaga sp. nov. Type from 12 miles southwest of Vidalia, Concordia Parish, Louisiana. No. 137,122, U. S. National Museum (Biological Survey Collection), skin and skull of d\ adult ( occipito-sphenoid suture entirely closed); collected June 17, 1905, by B. V. Lilly. General characters. — A large puma of the Felis coryi type but lacking the bright ferruginous color; skull much larger than in Felis couguar, with the very broad nasals and highly developed saggital crest of F. coryi. Differs externally from Felis o. azteca in its darker coloration and very pronounced caudal stripe. Color of type. — Upperparts, including outer sides of limbs, grayish fawn color with a decided cast of ecru drab, especially on flanks and legs; nape, withers, and an indefinite stripe down back to rump brighter, pale rufous; the entire upperparts finely mixed with dusky, the darker color of the hair tips. Face from crown to nose darker, blackish, with clear black spot each side of nose; a white streak over eye; ears blackish out- side, with edge of gray and lining of white hairs. Lips and throat almost pure white, shading through creamy white to fulvous on breast. Insides of legs grayish, mixed brown and white hairs; fur of foot pads dark brown; feet like legs and thinks. Tail much darker than back with a sharply defined stripe of dark brown along the entire length of upper surface from rump to the short black tip. Skull and teeth. — Skull large, massive, with the large, broad nasals and high saggital crest as in F. coryi; but with larger audital bullae. Much larger than the skulls of F. couguar, with well developed crest and much larger nasals ; bullae much larger ; opening of anterior nares very much larger ; teeth, especially the second and third upper premolars, larger. Measarements. — Skin of type: Total length, 2100 nun.; tail, 735 (meas- urements from tanned skin). Skulls of type and an adult male, very slightly younger, from Mer Rouge, Louisiana; the latter in parentheses: Condylobasal length, 193.5 (192); basal length, 180 (178); zygomatic breadth, 149 ( — ); palatal length, 87 (88); postpalatal length, 95.5 (97) ; least interorbital breadth, 43.5 (4ii); greatest length of nasals, 63 (60); greatest breadth of nasals, 36.3 (35.4); length of upper premolar row, 46.1 (46); length of audital bulla, 36 (37); length of lower molar-pre- molar series, 45.7 (46). Remarks. — The rich ferruginous or intense rusty red back and other color characters of Felis coryi, and the pale uniformly colored back and tail of Felis oregonensis azteca, are enough to distinguish readily the Florida and Texas forms from the Louisiana puma. Pumas are still fairly common in the wilder parts of the cane brake region of eastern Louisiana. In the early spring of 1904 while Mr. Waldo E. Forbes of Boston, Mr. B. V. Lilly of Louisiana, and I were hunting in the Bear Lake Cane we heard pumas calling in the forests at night, and sev- eral times succeeded in starting the animals with the hounds, but they were invariably too fast for our slow bear dogs and always quickly left Hollifti i — 77(6' Louisiana Pinna. 177 them in the distance. The following notes are taken from my field note- book of this trip: February 23 < 1904). Heard panthers crying about nine o'clock last night. There were probably two of them as the calls were sounded at short intervals, some times only about a minute apart, and one seemed a little farther away. The animals were evidently moving along to the north. The cry is a long drawn out, shrill trill, weird and startling. It commences low on the scale, gradually ascends, increasing in volume, and then lowers at the end. Forbes and Lilly put the hounds after them early this morning, but without success. Heard panther again in evening. February 24. Made a drive for the panther again this morning but did not run him hard enough to tree. February 26. Heard panthers to-night on both sides of Bear Lake. February 27. Crossed the lake early this morning and made a drive for the panther but the dogs seemed to lose trail at lake. Lilly says that panthers take readily to water and be thinks this one crossed to our side before morning. Made a drive on our side and soon started a panther in the cane. The animal promptly outdistanced the dogs after badly slitting an ear for one of them. February 28. Heard the panther's trilling wail across the lake to-night. We went out in the boat and silently paddled quite near him in an effort to "shine" his eyes, but failed to get a shot. I believe the pumas call more in the early spring than at any other season, and we were probably very fortunate in the time of our visit to the cane, as the experience was one of very great interest. Mr. Lilly's efforts to obtain a good specimen for the Biological Survey resulted in the capture of the type the following year. Vol. XXIV, pp. 179-184 June 16, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON SOME OBSERVATIONS ON A PHOTOGENIC MICRO- ORGANISM, PSEUOOMONAS LUCIFER A MOLISCH. BY F. ALEX. McDERMOTT, [Hygienic Laboratory, U.S. Public Health and Marine-Hospital Service. Washington, D.C.] Some years ago Dr. R. E. B. McKenney published in the Proceedings of the Biological Society a very interesting paper on luminous bacteria (Proc. Biol. Soc. Wash., 1902, Vol. 15, pp. 213-231). The form which serves as the basis for what is to follow has been isolated since Dr. McKenney's paper was published, and appears to present some points of possible interest. This organism is Pseudomonas lucifera Molisch, iso- lated by Prof. Hans Molisch, of the Plant Physiology Institute of the University of Vienna, and it is through the courtesy of Professor Molisch that the parent culture for this work was obtained. The organism was isolated from sea-water, and is claimed by the discoverer to give the brightest light of any bac- terial form so far isolated. Like most other luminous micro-organisms, Ps. lucifera will grow on the ordinary culture media, under aerobic conditions, but for luminescence there must be present 2.5 to 3.0 per cent of sodium chloride, or some one of certain other mineral salts. The use of media made from fish-meat is unnecessary. The light given by ordinary bouillon-gelatin-salt cultures is a soft and beau- tiful green, which after the eyes become accustomed to it, appears of considerable intensity. A veritable " living lamp,'' as Dubois has called it, may be made by coating the inside of a sterile flask with the bouillon-gelatin-salt medium, and then inoculating the surface of the gelatin with a liquid culture. Such a lamp will continue to give light for about a week, though the light 32— Proc. Hior . Soc. Wash., Vol. XXIV, 1911. (179) 180 McDermott — Observations on a Photogenic Micro-organism. gradually weakens, owing to the drying of the medium, and finally dies out; the maximum intensity of light is about 48 hours after inoculation. Agar cultures appear to grow about as well as gelatin cultures but luminesce weakly or not at all, even in the presence of the proper amount of mineral salts. Liquid media of very simple composition serve for the growth of the organism ; the simplest medium serving for luminosity is a 3.0 per cent solution of salt, containing 1.0 per cent of asparagin.* This medium is of course not the best; cultures in it are short-lived, and glow only feebly; a better medium is — Sodium chloride 2.5% Magnesium chloride 5 Peptone 1.0 Asparagin 1.0 Glycerin > - .5 Larger amounts of peptone and asparagin act adversely to growth; the organism is apparently sensitive to " over-feeding." It will grow and luminesce in milk containing 3.0 per cent of common salt; the light on the surface is quite bright. As a rule, however, the light from luminous cultures is much less intense than that from gelatin, though when shaken gently, the light perceptibly increases in intensity. The mineral salt present may be any one of several, but the brightest cultures have been obtained by the use of 2.5 per cent of sodium nitrate, and 0.5 per cent of magnesium chloride. All media must be faintly alkaline to litmus. The spectrum of the light emitted by a gelatin culture is of very limited range, extending from the yellow-orange to the indigo, with a decided maximum intensity in the green. This spectrum is of less extent than that of our local fireflies (Photi- nus pyralis, Photuris pennsylvanica, etc.) although the appear- ance of the light to the eye does not differ so very much from that of the last-mentioned Lampyrid. In liquid cultures the light appears to the eye to be more whitish, though the range of the spectrum is the same; probably the maximum intensity is shifted to a different point. * In this connection it is of interest to note that Wood (Journ. Amer. Med. Assn., 1911. Vol. 56, pp. 1094-6), has recently recommended " normal " (physiologic) salt solu- tion as an emergency culture medium for many bacteria. McDermott — Observations on a Photogenic Micro-organism,, 181 Forsyth (Nature, 1910, Vol. 43, p. 7) has reported the dis- covery, by spectrophotography, of ultra-violet rays in the light of PhotobacteHum phosphorescens Fischer. This seems very remark- able, especially in view of the fact that ultra-violet light is used as a batericide, and the further fact that previous photographs of the spectrum of the light of these organisms had failed to show the presence of such rays. In this connection solutions and crystals of para-amino-ortho-sulpho-benzoic acid* have been exposed to the light from large cultures of Ps. lucifera, and have failed to show the least trace of fluorescence, although this substance is used to detect ultra-violet radiation by means of its fluorescence. It seems very unlikely that this organism emits ultra-violet radiation. Luminous cultures of Ps. lucifera on the usual culture media do not appreciably aflect a charged electroscope. For the conduct of the experiments leading to this conclusion, I am indebted to Drs. F. W. Clarke and R. C. Wells, of the Geological Survey. In its chemical conduct this micro-organism exhibits only certain slight similarities to the firefly. Professor Kastle and the author have shown that the luminous tissue of the Lampyridae which has been dried in the absence of oxygen — preferably in a hydrogen vacuum — and sealed in hydrogen, will exhibit lumi- nosity when moistened , for at least thirteen months after prepara- tion (Amer. Journal Physiol., 1D10, Vol. 27, pp. 122-151, November), and since the above paper was written, specimens prepared at the same time have glowed on moistening after eighteen months; in fact there seems to be no good reason why they should deteriorate at all . The author has found that if com- mercial hydrogen peroxide solution is used in place of water for moistening the dry material, a much brighter light is obtained, whose spectrum is of about the same range as that of Pi. lucifera, though with its point of maximum intensity nearer the yellow. (Canadian Entomologist, November, 1910, Vol. 42, pp. 357- 363.) Somewhat similar observations may be made upon this micro-organism. An open vessel tilled with sulfuric acid was placed upon the bottom of a flask, the walls of which were coated with a luminous gelatin culture of Ps. lucifera, the flask filled with hydrogen, tightly stoppered, and set away. After two * Kindly sent me by Prof. . I. H. Kastle of the University of Virginia (See American Chemical Journal. January. 1911 >. 182 McDermott — Observations on a Photogenic Micro- organism. days the gelatin showed signs of being drier, and the culture did not glow. A small portion of the gelatin was removed by means of a spatula and placed in a Petri dish; it glowed feebly in the air, the glow being just perceptible in a dark room. A few cubic centimeters of 2.5 per cent hydrogen peroxide solu- tion were then run upon the gelatin, when a number of bright points showed for an instant, after which no glow was visible. Attempts to completely dry liquid cultures, and to harden gela- tin cultures in a hydrogen vacuum have so far been unsuccessful on account of the leakage of the apparatus. Nitrobenzol, which Dr. Kastle and the author found to be a powerful stimulus to activity on the part of the luminous tissue of the firefly, was without effect in stimulating the luminous activity of gelatin cultures of this organism whose luminosity was on the wane, and when added to liquid cultures promptly extinguished them. A solution of sodium nitrite added to a liquid culture of Ps. lucifera extinguished the light instantly. Both of these sub- stances are germicides, and the results obtained are those which would naturally be foretold. The addition of a few drops of 1 : 10,000 adrenalin hydrochloride solution to 20 c. c. of a liquid luminous culture of this organism produced no immedi- ate effect, but after 18 hours, the culture was apparently dead; adrenalin was found to be a powerful stimulant of photogenic activity in the firefly, when injected into the living insect. It would appear, therefore, that in the firefly, these exciters act upon the nervous system, and not directly upon the luminous tissue. Oxygen under a maximum pressure of two or three atmos- pheres was applied to a liquid culture in a closed bulb; the light emitted became much stronger as long as the oxygen pressure was maintained; sudden release of the oxygen pressure was followed by a slow diminution of the intensity of the light. A liquid culture placed in a desiccator rilled with hydrogen gave no light after about five minutes; it also failed to give light when treated with hydrogen peroxide at the end of three days, when it had dried out to the point of crystalization. But little can be said as to the chemical processes by which these organisms produce light. The process is certainly one of oxidation, or at least one requiring the presence of oxygen, as is the case with the firefly. Probably the actual use made of McDermott- -Observations on a Photogenic Micro-organism. 183 the oxygen is somewhat similar to use made of it in the luminous insects, except that the gas is taken by the cells directly from the air to which the organism is exposed, whereas in the fire- fly an intricate network of trachea' supply the air for the process. But while the nature' of the light emitted by these organisms, and some of their chemical properties suggests a general simi- larity of the photogenic processes in the two forms, it is by no means necessary to assume that it is identical. It may even prove that in organic chemistry we may have photophore group- ings, as we now have fluorophore groups and chromophore groups, and that biologic oxidations producing very similar luminous manifestations may actually involve very different active substances. The fact that these organisms will luminesce in a medium consisting of water containing three per cent of salt and one per cent of asparagin — <*-amino-succin amidic acid — HOOC.CH2.CHNH2.CONH,, suggests that the process may really be less complicated than would at first seem. So far as I know, Ps. lucifera is not pathogenic for man. It appears to lie very small, and in common with most photogenic micro-organisms, it is rather delicate; it grows and luminesces best at 18-22° C. ; 28 to 30° C. will kill the cultures. The whole subject of the photogenic bacteria has been pretty thoroughly covered in Professor Molisch's book, Leuchtende Pflanzen (Jena, 11)04), and a very exhaustive and interesting review of the entire subject of biophotogenesis, — regretably in German again, — has been published by Prof. Ernst Mangold of Greifswald, under the title Die Produktion von Licht, as the 2nd half of the third volume of Hans Winterstein's Handbuch der vergleichende Physiologie. (Jena, 1910.) I have to thank Mr. Wm. Lindgren, of the Hygienic Labor- atory, for propagation of this organism from the culture sent me, and for advice in handling the same. Vol. XXIV, pp. 185-186 June 23, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON DESCRIPTION OF A NEW PHILIPPINE FLYING- SQUIRREL. BY N. HOLLISTER. Among some mammals submitted for identification to the United States National Museum by the authorities of the Phil- ippine Bureau of Science, is a flying-squirrel from Basilan Island, which proves to be new. Sciuropterus crinitus sp. nov. Type from Basilan Island, Philippines. No. 101, Collection of Philip- pine Bureau of Science, c? adult; skin and skull. Collected by Richard ('. McGregor and A. Celestino, January 12, 1907. General characters. — A large member of the subgenus Petinomys; ex- ternally somewhat like Sciuropterus (Hylopetes) nigripes -of Palawan, but size smaller ; ears very small, with three long tufts of hair, one from below, and one from each side; feet grayish-brown instead of black, and tail brown. Fur of back and upper side of parachute long and heavy; belly thinly haired. Color of type. — General color of upperparts pale chestnut, finely mixed with black. Nose and lips gray; crown, back, and upper surface of parachute and tail pale chestnut, finely lined with black; tail blackish near tip ; underfur slate color, the hairs tipped with the chestnut and black. Ear tufts brown at base, black at terminal half ; whiskers long and black. Limbs and margin of parachute blackish and buffy; feet grayish-brown. Underparts of body and limbs buffy, the underfur gray; under side of tail pale chestnut, lighter along median line. Skull and teeth. — Skull resembling; those of the other larger members of the subgenus Petinomys ; * but bullae slightly less flattened and cheek teeth with the small accessory cusps at outer exit of the valley between the two main transverse ridges very minute. Compared with a skull of Sciuropterus (Petinomys) hageni from Sumatra, which is of the same essential size, it has a narrower zygomatic breadth, with a slightly wider * See Thomas, Annals and Mag. Nat. Hist., 1908, I, p. 6. 33— Proc. Biol. Soc. Wash., Vol. XXIV, 1911. (185) 186 Hollistei — A New Philippine Flying-Squirrel. rostrum; interorbital breadth less; all teeth slightly larger; zygomata lighter; and bullae slightly less flattened (but still very broad and low and not inflated to anywhere near that degree found in species of the other subgenera). Mastoids not inflated. Small premolar robust, and other cheek teeth slightly increasing in size backward; crowns low; enamel finely sculptured; ridges low, with transverse ridges somewhat broken, the supplementary cusps between them small, but plainly visible on pm4, and less on m1; obsolete on m2. Measurements of type. — Total length, -570 mm. ; tail, 260; hind foot, 42. Ear from notch (relaxed on dry skin), 13.5; length of ear tufts, 36. Skull.— Greatest length, 53; condylobasal length, 48.6; basal length, 45.1; zygomatic breadth, 32; least interorbital breadth, 10.1; greatest breadth of nasals, 9.2; mastoid breadth, 22; length of maxillary tooth row (alveoli), 11.6; length of mandible from anterior surface of symphy- sis to condyle, 32.5. Remarks. — It is interesting that this second known Philippine flying- squirrel belongs to a different subgenus from the Palawan species, S. nigripes. Though the subgeneric characters in »S'. crinitus are less pro- nounced than in some other species, there seems no question as to its place in Petinomys. Vol. XXIV, pp. 187-190. June 23, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON DESCRIPTIONS OF NEW AMERICAN BIRDS. BY OUTRAM BANGS. The American birds described as new in the following pages are from various sources, and are such as have turned up from time to time during the process of arranging and identifying the collections in the Museum of Comparative Zoology. Rupornis magnirostris occidua subsp. nov. Type from Rio Tambopata, eastern Peru, adult (d1?), no. 47,."><>2, Coll. Mus. Comp. Zool. Collected by Dr. W. C. Farrabee, in May, 1907. Characters. — Similar to true R. magnirostris (Gmelin) of Colombia, Guiana, and Amazonia, with the upper parts pale gray and the belly and flanks barred with white and pale cinnamon-rufous, but with the chest bright cinnamon-rufous as in R. magnirostris natteri (Gel. & Salv. ) of Brazil, not gray as in true R. magnirostris. Measurements. — Type, adult ( I < 67 54 26 17 Remarks. — As Thamnophilus doliatus has now, by general consent, been divided into a number of subspecies, 1 feel little hesitation in adding an- other quite as good, it seems to me, as those already recognized. Indi- vidual variation in this species must, however, always be reckoned with, and series, not individuals, compared. The pale colors of the female of the Margarita Island form have already been mentioned by Cory in his list of Birds of the Leeward Islands, based upon Ferry's series of fourteen skins, of which eight were females. Pyrocephalus rubineus blatteus subsp. nov. Type from Sabune Dist., British Honduras, adult d\ Mus. Comp. Zool., no. 19,812, Bangs Coll. Collected May 2, 1906, by Morton E. Peck. Characters. — Similar to P. rubineus mexicanus Sel. of Mexico and the southwest border of the United States, but smaller, with an actually as well as relatively broader bill; adult male with red of under parts and pileum more crimson, less orange — about geranium red. Measurements . — No. Sex. Locality. Wing. Tail. Tarsus. Culmen. L9,812 c?ad. 40.145 ! c? ad. 40.146 c?ad. Br. Hon. Sabune Dist. . 1 1 73 73.5 74 53.5 55. 54 15.5 16 16 13 12.5 12.5 190 Bangs — Descriptions of New American Birds. Remarks. — This new form of the vermilion tyrant-bird occupies Yuca- tan, British Honduras and parts of Guatemala. The specimens which I have examined are conspicuously different from the more northern race. Its characters were fully pointed out by Ridgway in Birds of North and Middle America, but it was not given a name by that distinguished orni- thologist. Leistes superciliaris petilus subsp. nov. Type from Concepcion del Uruguay, adult d\ no. 31,023, Mus. Comp. Zool. Collected November 27, 1880, by W. B. Barrows. Characters. — Similar to true L. superciliaris Bp. of Brazil in color, but decidedly smaller, with a shorter bill, wing in adult c? about 98 (in adult c? of L. superciliaris superciliaris about 105) ; culmen in adult cf about 19.5 (in adult (J1 of L. superciliaris superciliaris about 22.5). Measurements. — No. Ses Locality. Wing. Tail. Tarsus. Culmen. 31,019 cTad. Conception del Uruguay 96 57 31 19.5 31,020 rjTad. it it a 96 58 30 19 31,021 tfad. a a a 97 (in 20.5 19 31,022 tfad. It It it 98 56 30 20 31,023 C?ad. i i a a !KS 57 30.5 19 Remarks. — I would also suggest the necessity of subdividing Leistes militaris (Linn.) and recognizing Leistes militaris militaris (Linn.), Guiana and eastern Peru north to Panama, smaller, with a shorter bill, wing in adult c? about 93, culmen about 20, and L,eistes militaris enjtli- rothorax Pelzeln, Lower Amazon Valley, larger, with a longer bill, wing in adult c? about 100, culmen about 22.5. Vol. XXIV, pp. 191-194 June 23, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON THE GENERIC NAME OF THE AFRICAN BUFFALO. BY N. HOLLISTER. [Published by permission of the Secretary of the Smithsonian Institution.] Aside from conspicuous and well known differences in the general appearance of the animals and in the types of their horns, the buffaloes of Africa and of the Indian region are dif- ferentiated by constant characters of real generic value. At present these buffaloes are commonly combined in the genus Bubalus, with Bubalus coffer of South Africa as the type, or even loosely thrown in the old Linnsean genus Bos. The type of Bubalus H. Smith is, however, by tautonymy Bos bubalis Lin- naeus, the Indian Buffalo. The Cape Buffalo and its allies should stand as a separate genus, the correct name for which appears to be Syncerus Hodgson. The synonymy and general characters of the two genera follow7: Genus Bubalus Smith. 1827. Bubalus H. Smith, Griffith's Cuvier's Animal Kingdom, V, p. 371. (Subgenus of Bos.) Type by tautonymy, Bos bubalus Smxth= Bos hubalis Linx^eus. Skull much less massive than in Syncerus ; more narrow and elongated ; facial profile nearly straight ; rostrum relatively long and slender. Nasal bones elongated and slender, narrowed in middle, and projecting much beyond end of median suture; thus longest laterally. Vomer fused with palatine bones and palatine plate of maxilla the entire length of median suture. Audital bullae very small, reaching to about plane of ventral surface of basi-occipital. Molars short and high, crowns almost square. Hair of dorsum reversed, directed forward from haunches to head; ears comparatively small, without conspicuous fringes. Indian Region. The Tamarau (B. mindorensis Heude), from Mindoro, agrees with B. bubalis in the above general characters and can not be subgenerically separated. I have not seen a skull of Anoa depressicornis from Celebes. 35— Proc. Biol. Soc. Wash., Vol. XXIV, 1911. (191) 192 Hollistei — The Generic Name of the African Buffalo. Genus Syncerus Hodgson. 1847. Syncerus Hodgson, Journ. Asiatic Soc. Bengal, XVI, new series, no. 7, p. 709, July. (Genus.) Type by selection, Bos brachyceros Gray.* 1872. Planiceros Gray, Cat. Rum. Mamm. in Brit. Mus., p. 10. (Sub- genus of Bubalns.) Type by tautonymy, Bubalus centralis a. Bos planiceros Blyth. 1872. Synceros Gray, Cat. Rum. Mamm. in Brit. Mus., p. 12. (Sub- genus of Bubalus.) Type by monotypy, Bubalus caffer Gra\= Bos caffer Sparrman. Skull massive, short and broad, facial profile concave; rostrum rela- tively short and broad. Nasal bones short and wide; greatest length mesially. Vomer not attached to palatine bones. Audital bullfe large, reaching far beyond plane of ventral surface of basi-occipital. Molars long, crowns relatively narrow. Hair of dorsum directed uniformly back- ward from neck to rump; ears large, heavily fringed with long hairs. Ethiopian Region. The following African buffaloes were recognized by the last monographer, Matschie, 1906, t or have been described since the publication of his revi- sion. The specific or subspecific rank of these forms is still very uncer- tain. I use trinomials where the original describer has done so. 1. Syncerus azrakensis (Matschie). Dar Roseres, on the Bahr el Azrak, Egyptian Sudan. 2. Syncerus brachyceros (Gray ). Central Africa (shores of Lake Tchad, Matschie ) . 3. Syncerus caffer caffer (Sparrman). Sunday River, Eastern Capeland. 4. Syncerus caffer aequinoctialis (Blyth). White Nile. 5. Syncerus caffer cottoni (Lydekker). Semliki Valley. 6. Syncervs caffer matthewsi (Lydekker). Mfumbiro, Ruanda, German East Africa. 7. Syncerus caffer radcliffei (Thomas). Burumba, Ankole, S. W. Uganda. 8. Syncerus centralis (Gray). ?Atbara, Egyptian Sudan. 9. Syncerus gariepensis (Matschie). Ligua River, upper Orange. 10. Syncerus limpopoensis (Matschie). Lembobo Mountains, Southern Swasiland. 11. Syncerus rnayi (Matschie). Bengo, Loanda. 12. Syncerus nanus (Boddaert). ' Probably somewhere between the Congo and the mouth of the Niger," Lydekker, Wild Oxen, Sheep, and Goats of All Lands, p. 111. ]?>. Syncerus neumanni (Matschie). Chagwe, Uganda. * Hodgson's genus contains two species, " Bornouensis and Brachycerus." The first has been overlooked by all recent workers. It dates from 1858, and first appears in Chas. Hamilton Smith's Synopsis of the Mammalia, Jardine's Naturalists' Library, XV, p. 290, as -Bos bornouensis. with description ; type locality, Bornou. t Site.— ber. ges. nat. freunde Berlin, 1906, no. 7. pp. 161-179, July, 1906. Hollister — The Generic Name of the African Bvffnln. 193 ll. Syncerus planiceros (Blyth). Gambian Region. 15. Syncerus ruahaensis (Matschie). Rnaha River, German East Africa. K». Syncerus schillings! (Schillings). Pangani, German East Africa. 17. Syncerus thierryi (Matschie). Togo, West Africa. 18. Syncerus wiesei (Matschie). Between Loangwa and Revuga rivers, north of Zambese River. L9. Syncerus wembarensis (Matschie). Tschaja Swamp, southern Wem- bere steppe. Vol. XXIV, pp. 195-196 June 23, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A NEW SWIFT FROM PALESTINE. BY OUTRAM BANGS. The well-known Selah Merrill collection of birds from Pales- tine was acquired by the Museum of Comparative Zoology in July, 100!). Tn the species and subspecies of the region, except for a few very rare or local forms, it is complete, and most of the species are represented in long series. Though many skins are from other places, and a number were got for him by Arabs, the collection was principally made in the vicinity of Jerusalem during the years 1884-1886 by the late Rev. Dr. Merrill him- self. Some day I hope to publish a catalogue of it, with such notes as were kept by Dr. Merrill in his register, though these in many cases, I am sorry to say, are not entirely satisfactory. The forms that have long been recognized as peculiar to Pal- estine and the ones recently described by Hartert during the progress of his " Vogel der Palaarktischen Fauna," appear to be all represented in the Merrill collection . In the collection the swifts are represented by smaller series than usual, there being three skins of Apus apus apus (Linn.), three of Apus affinis galilejensis (Autin.), and two of the white- bellied swift. The latter are so very pallid and otherwise dif- ferent from European specimens that I propose for the Palestine form the name, Apus melba petrensis subsp. nov. Type from the Jordan Valley, Palestine, no. 59,534, Coll. Mus. Comp. Zool. From the Selah Merrill Coll., taken in April, 1886, original no. "a. 108." Characters. — Similar to true A. melba (Linn.) of Europe, but much 36— Proc. Biol. Soc. Wash., Vol. XXIV, 1911. (195) 196 Bangs — A Neio Swift from Palestine. paler above and upperparts pale drab, lighter still on forehead and crown, the dusky marking in front of eye consequently very conspicuous; brown band across chest very narrow and the white throat patch large. Measurements. — No. Wing. Tail. Tarsus. Oilmen. 59.534 Type 59.535 Topotype 215 219 87 87.5 15.5 L6 9 9 Remarks. — Our two specimens, neither of which bears any sex mark, were in fully adult plumage, fresh and unworn, and their pallid coloring is not the result of fading. Hartert, in Catalogue of Birds in the British Museum, Vol. XVI, p. 439, mentions some very pallid skins from the western Himalayas, and from Belgaum, and possibly the pale form described above lias a wide breeding range. Vol. XXIV, pp. 197-208 June 23, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON SOME NEW NORTH AMERICAN IXODIDAE WITH NOTES ON OTHER SPECIES. BY F. C. BISHOPP. [Published by permission of the Chief of the Bureau of Entomology.] In studying the collections of ticks that have accumulated in the collection of the Dallas Laboratory several new species have been discovered, and in order that the biological notes on these forms to be published later may not contain descriptions of new species these descriptions are presented here. I wish to express my sincere thanks to Mr. Nathan Banks for the many helpful suggestions given and for his kindness in placing at my disposal the collection of the Bureau of Ento- mology and of the National Museum, and the collection of the late Doctor Marx, as well as specimens from his private collec- tion. My thanks are also due to Doctor L. 0. Howard and Mr. W. D. Hunter for courtesies extended in the course of my work and to Messrs. G. N. Wolcott and H. P. Wood, the former for making the drawings used in illustrating this paper, the latter for assistance in the rearing of specimens received at the laboratory in the immature stages and in the care of all collected material received by him. Ixodes cookei var. rugosus n. var. Female. — Capitulum (Fig. 1), length TfiOjU. (from tip of hypostome to a line ilrawn between tips of postero-lateral angles of basis capitnli); basis capituli dark reddish brown, borders almost black, rather wide (560 /O; posterolateral angles not prominent, basal border incurved at center; surface, especially around pnrose areas, roughened; porose areas large, somewhat triangular, reaching very near to posterior margin of basjs 37— Proc. Biol. Soc. Wash., Vol. XXIV, 1911. (197) 198 Bishopp — Some New North American Ixodidas. capituli; slightly broader than long and separated by about half their length ; the space between them with a pronounced rather irregular groove ; palpi 574 // long, rather broad and heavy, entire surface rough- ened ; third segment slightly longer than broad. Scutum, length 1.44 mm., width 1.41 mm.; reddish brown, broadest slightly in front of the middle, broadly rounded behind; lateral carinae present but not well denned, extending back a little behind widest part of shield but not reaching the shield margin ; cervical grooves moderately distinct, extending back to middle of scutum ; surface of scutum closely covered with large deep punctures, more scattered toward tip, in regions of cervical grooves and lateral carinae they are somewhat confluent, form- ing rugose areas. Legs light reddish brown, translucent, rather long and stout, all tarsi abruptly narrowed near tips, all coxae with a short posterior apical tooth and with a trace of a tooth at the anterior apical angle; coxae I with a long spine at base, slightly shorter than in I. cookei. Stigmal plates rather small, transversely oval, 330// x 287// and 344 /j. x 273/^; about 108 goblets per plate, irregular in size, some rather large, not closely crowded together, six rows between the macula and marginal cells at the widest point; macula nearly circular, located slightly antero- ventrally from center; marginal cells large, nearly circular. Body pale yellow (slightly engorged), finely striate and with a few very short pale hairs; postero-median groove shallow, running from near pos- terior margin to a point opposite anus ; accessory grooves deep and broad, running from near posterior margin nearly to shield, almost straight be- hind, irregular in front; anal frame about two and a half times as long as broad, sides nearly parallel, somewhat pointed in front of anus. Male. — Capitulum (Fig. 2), length 502 //(from tip of chelicerae (closed) to line connecting postero-lateral corners of basis capituli); basis capituli very dark brown, posterior and lateral edges black, greatest width at base of palpi (358//) only slightly narrowed toward basal margin where the width is 301 m ; the postero-lateral angles are not produced; the basal margin only very slightly concave; dorsal surface of basis capituli dis- tinctly rugose and with a few punctures; ventrally it has a semi-circular ridge curving backward from near the base of each palpus, behind this it is distinctly narrowed; palpi very short and broad (330// x 187//) ; third segment but very little shorter than the second. Scutum, length 2.75 mm., width 1.54; very dark brown, some parts almost black; sides nearly parallel, broadly rounded behind, less so at anterior end ; the entire surface, with the exception of a few small areas along the median line, closely covered with large punctures, at anterior end these are somewhat confluent, producing rugosity ; a few very short pale hairs near edges of scutum ; cervical grooves deep and short ; accessory grooves shallow ; a series of broad depressions at intervals from the an- terior ends of accessory grooves to near the cervical angles where surface of scutum is decidedly uneven. Bishopp — Some New North American Ixodidss. 199 Legs moderately long and stout, reddish brown, somewhat punctate; coxae I with a basal spine of medium length, shorter than in I. cookei ; all coxae with minute apical teeth. Stigma] plates nearly round, 330m wide by 301 m long; about 135 gob- lets, of medium size, closer together and more uniform in size than in female; about seven rows between macula and marginal cells at widest point; macula broadly oval, located ventro-anteriorly from center; adis- tinct depression running around macula on antero-ventral side; marginal cells circular, smaller than in female. Body with marginal strip on dorsum, somewhat lighter than scutum; finely punctate and with many very short yellow hairs; venter dark In-own somewhat mottled with light brown, punctate over its entire sur- face ; the chitinized plates exceedingly closely covered with very large punctures and with a few very short hairs; edges of post-genital plate rugose; anal groove straight (transverse) in front of anus; length of this portion of groove 215 /x, length of anal frame 932 y. ; sides of anal frame slightly curved and strongly divergent behind anus. Type, Cat. No. 13,973, U. S. National Museum. Type host. — Dog. Type locality — Tiller, Oregon. Described from a female and male of a lot of two females and six nymphs (one of which was engorged and molted to a male) collected March 28, 1910, by Mr. S. S. Stevens. Dallas Ace. No. 1316. The collection of the Bureau of Entomology at Dallas, Texas, contains the following lots of specimens, taken on dogs, which are referred to this variety : Two 9, Beatrice, Cal., March 2, 1910 (J. Sawyer); 1 9 and 2 nymphs, Hemlock, Oregon, March 28, 1910 (Chas. Desmond); 4 ?,1 nymph, 1 larva, March 27, 1910; 5 9, 4 nymphs, May 5, 1910; 1 ?, May 23, 1910; 1 $, July 1, li)10; 1 ?, 1 nymph, November 29, 1910; 10 nymphs, De- cember i), 1910; all from Pysht, Washington, collected by Mr. Guy Decker; 19,9 nymphs, Sightly, Washington, September 21, 1910 (B. A. Bruce) ; the series of females shows but slight variations from the type. In some, the punctures over the entire scutum are more or less confluent, making the surface longitudinally rugose. In some, the capitulum and palpi are more distinctly roughened than in the type. The females of this variety are separated from those of I. cookei proper by the porose areas being as long as broad, more triangular in outline, and somewhat closer together ; also by the roughened capitulum and shield. The latter is rather less narrowed behind the middle than in cookei. The male is easily distinguished from the males of cookei and other species by the large punctures which closely cover almost the entire body and scutum. As compared with I. cookei, the male is darker in color, sides of body more nearly parallel, basal spines on coxae I shorter, and stigmal plates slightly smaller. 200 Bishopp — Some New North American Ixodidse. Ixodes banksi n. sp. Female. — Capitulum (Fig. 3), length 689m (from tip of hypostome to posterior edge of basis capituli on dorsum); basis capituli narrow, greatest width (452 m) at base of palpi where the basis capituli is some- what extended to form a socket for their reception ; dark yellowish brown, posterior margin almost black, postero-lateral angles scarcely visible; ventrally, basis capituli shows two broad flat processes at the base of each palpus; porose areas rather large, well defined, about as long as broad, extending close to posterior margin, separated by about one-half their length; between them is a narrow groove or depressed line; palpi lighter in color than basis capituli; length 474/;, greatest width 158m; third segment slightly longer than broad; second segment nearly twice as long as third ; first segment rather long and without prominent ventro-lateral process. Scutum (Fig. 4), length 1.14 mm., width 1.01 mm., reddish brown, moderately narrowed behind, postero-lateral margins not concave; lateral carinae present but not strong; cervical grooves moderately distinct beginning some distance behind cervical angles, first converging then diverging so as to form an angle at the point where they are nearest together; surface irregularly covered with punctures of different sizes, mostly rather large; also with a few yellow hairs. Legs amber in color, long and rather slender; tarsi not abruptly nar- rowed at tips; all coxae (Fig. 5) with a distinct blunt spine at apex; coxae I with a rather long and moderately slender basal spine; coxae II with a trace of a basal spine. Stigmal plates (Fig. 6) exceedingly large, slightly broader than long (509 m x 466 m); about 440 goblets per plate, small and very closely crowded together, even near macula, around which there is but a small area without goblets; about 15 rows of goblets between macula and marginal cells at widest point; marginal cells small and close together, only slightly flattened laterally; macula oval, located comparatively close to the antero-ventral border, slightly elevated above area immediately surrounding it, which area is concave. Body dark reddish brown, finely striate and minutely punctate, well covered with moderately short, yellow hairs; marginal groove distinct, running from near the lateral angles of shield well around curve of body at posterior end; postero-median groove broad and shallow, extending from near posterior margin half way to posterior end of scutum ; acces- sory grooves deep and broad, curved, slightly longer than the postero- median ; venter with hairs as on dorsum ; anal groove deep ; anal frame slightly pointed in front of anus, sides nearly parallel; entire frame one- fourth as wide as long. Type, No. 13,074 U. S. National Museum. Type host. — Muskrat (Fiber sibethicus Linn. ). Type locality. — Mammoth Springs, Arkansas. A female described from a lot of 15 females and 7 nymphs collected June 14, 1910, by Mr. A. H. Howell ( Dallas Accession No. 1833); named Bishopp — Some New North American Ixodidse. 201 for Mr. Nathan Banks, the principal contributor to onr systematic knowledge of the ticks of North America. This species would run to I. cooked or /. sculptus in Mr. Banks' table in his Revision. However, it is easily separated from those species by the much larger stigmal plates and somewhat shorter basal spine on coxae 1. The dimensions of the stignial plates of I. cookei .vary from 330/* x 287 tt to 4<>7 m x 412 m. There are six or seven rows containing 120 to 150 goblets. The dimensions of the stigmal plates of I. sculptus are much smaller than in I. cookei and the number of goblets fewer. The shape of the basis capituli and porose areas is markedly different from 7". sculptus and the lateral carinae are not distinct and incurved as in that species. The basis capituli is much narrower than in 7". cookei and the postero-lateral angles are less prominent than in that species. The specimens at hand are very uniform in most characters; the width of the basis capituli varies from 41(>m to 439 M> and the size of the stigmal plates from 502 m x 445 m to 560 m x 504 m, and the number of goblets from about 440 to 500 per plate. The length of the first tarsi varies from 574 m to 653 m. Some of the females in the lot were about one-third engorged; the bodies of these are light yellow and elongate. Ixodes kingi n. sp. Female. — Capitulum (Fig. 7), length 78U m ( froin tip of hypostome to line connecting postero-lateral angles of basis capituli ) ; basis capituli dark reddish brown, of medium size, width 574 m; postero-lateral angles promi- nent; porose areas small, slightly broader than long, separated by nearly their width; outline well defined, pits large and deep; palpi very short and broad (488 m x 244 m), the greatest width occurring at the apex of the second segment; this segment is only slightly narrowed back to its basal articulation where it is abruptly constricted; first segment broad and deeply cut out dorso-laterally and bears a prominent, broadly rounded projection on its anterior margin ventro-laterally ; ventrally the basis capituli is smooth and somewhat narrowed posteriorly. Scutum (Fig. 8), length 1.34 mm., width 1.2 mm., reddish brown, darkest anteriorly, greatest width at about one-third of the distance from anterior angles, distinctly narrowed behind, the postero-lateral margin being slightly concave ; lateral carinae strong, running nearly to margin, slightly behind middle of scutum; scutum distinctly depressed for the entire length of the lateral carinae immediately mediad of those carinae; surface of scutum densely and rather evenly covered with coarse punc- tures. Legs amber in color, short and rather slender; tarsi abruptly narrowed near apex; all coxae (Fig. 10) with a distinct apical tooth; coxae I with a moderately long, stout basal spine. Stigmal plates (Fig. 9) rather small, 287m x 244m, transversely oval, about 9b' goblets per plate, medium sized, not crowded together; a con- siderable area postero- ventrally from macula without goblets; goblets 202 Bishopp — Some New North American Ixodidfe. arranged in irregular concentric rows ; five rows between the macula and marginal cells at the widest point and two rows at the narrowest; mar- ginal cells much smaller than goblets, close together and somewhat rectangular; macula oval, slightly elevated above smooth area surround- ing it ; this area is depressed slightly below the goblet covered portion. Body pale yellow, punctate, rather densely covered with moderately short pale yellow hairs ; genital aperture opposite coxae III; anal groove broadly round in front of anus and divergent behind it; anal frame about one-half as broad as long. Male. — Capitulum (Fig. 11), length 517m (from tip of hypostome to a line connecting postero-lateral angles of basis capituli) ; basis capituli dark brown, greatest width at base of palpi (41(>m) slightly narrowed from this point to posterior angles which are 301 ;"■ from center to center; these angles prominent; on the ventral side of the basis capituli is a nearly semi-circular ridge curving backward from near the base of each palpus where it is very prominent ; palpi very short and broad (length 351 a*, width 194 m); the third segment broader than long, the second nearly as broad as long. Scutum, length 2.58 mm., width 1.28 mm., brownish yellow, darker in front over an area corresponding to the shield in the female, sides nearly parallel, broadly rounded behind; surface shiny with a very few short pale hairs, closely covered with large deep punctures at anterior end, these become small and more scattered posteriorly to the middle of the shield where they disappear; the posterior half of the scutum has a few exceedingly minute punctures; lateral carinae strong, running from angles of capitular emargination to edge of scutum one-fifth of its length back ; as in the female, shield is depressed immediately mediad of the lateral carinae ; cervical grooves indistinct, shorter than lateral carinae, only distinctly visible toward their posterior ends. Legs (Fig. 12.) moderately short and rather slender, yellowish brown; tarsi III and IV gradually tapering at tip, others rather abruptly nar- rowed ; all coxae with a distinct apical tooth and coxae I with a short stout spine (about 86 /j. long ). Stigmal plates Longitudinally oval (287/* x 251 m); 90 goblets per plate, medium sized and rather more scattered than in female; arranged in concentric rows, five rows at the widest and two at the narrowest point between the macula and marginal cells ; a much smaller area is left with- out goblets around the macula than in the female; marginal cells about one-half the size of the goblets and somewhat rectangular; macula oval, situated ventro-anteriorly from center. Body (Fig. 12), marginal strip around scutum extends forward to opposite second pair of legs, pale yellow, punctate and with numerous short pale hairs ; venter yellow, plates highly chitinized and without punctures except the lateral ones (which have the stigmal plate located at their anterior end), the punctures on these lateral plates are moder- ately large and close together along their outer halves, the inner halves are free from punctures except a few ventrally from the stigmal plates ; anal groove straight (transverse) in front of anus, the length of this Bishopp — Some New North American Ixodidse. 203 portion of the groove being 215m; behind the anus the anal groove is nearly straight and strongly divergent: length of anal frame 789 /*. Type, Cat. No. 13,975, U. S. National Museum. Type host. — Badger ( Taxidea taxvs Schreber). Type loculit;/. — Meeteetse, Wyoming. A female and male described from a lot of L6 females, 13 males, 9 nymphs and 14 larvae collected July 10, L909, by Mr. \V. V. King (Dallas Accession No. 621) for whom this species is named. The collection of the Bureau of Entomology contains 1 9 on Thomomys, Walker Pass. Cal., July 1, 1891 (Dr. A. K. Fisher); 2 $ on prairie dog, Sherwood, Texas, November 2, 1906 (F. ('. Pratt); 1 tf, 1 9 on badger, 1 9 on wolf, 1 9 on skunk, Sabinal, Texas (F. C. Pratt and C. T. Atkinson); 5 J1, 5 $ on badger, Fairview, New Mexico, October 2, 1909 (E. A.Goldman); 6 <■?', 13 9 on dog, Cedarview, Utah ( W. F.Fishback); 2 9 on mink, Randlett, Utah (E. P. Ford); 7 9 on dog, Sunnyside, Idaho (E. H. Raymond); 1 9 on Thomomys c. ocius and 1 9 on Perodipus richardsoni, both from Sun, Wyoming (M. Cary); 1 9, Sheri- dan, Wyoming (Cecil Lowman); 1 tf, '2 9, Lost Cabin, Wyoming (Dr. P. H. Shallenberger) ; 1 <$, 13 9 on dog, Hyattsville, Wyoming (J. R. Matthews); 1 9, Mt. Shaw, Montana; four lots, one of which contained both sexes, on dog, Garneill, Montana (H. McLaughlin); 1 9 on a Spermophile, Garneill, Montana (H. McLaughlin); two lots of 1 9 each on dog, Chauteau, Montana (G. M. Carson); two lots of 1 9 each on dog, Conrad, Montana; 1 9, Townsend, Montana ( W. D. Neild); 1 9 on Marmota flaviventer , Florence, Montana (W. V. King). The first two lots (from Walker Pass, Cal., and Sherwood, Texas) were placed by Mr. Banks with Ixodes pratti Banks (A Revision of the Ixo- doidea, puis, p. 28). I have examined the type of I. pratti and find it to be distinct from the specimens from Walker Pass, Cal., and Sherwood, Texas. The tooth beneath the rostrum at the base of each palpus is suffi- cient to separate the species. The first coxae spine is shorter in pratti and no posterolateral angles are present on the basis capituli. Mr. Banks' drawings of/, pratti appear to have been made in part from the speci- men from Sherwood, Texas ( /. kingi) as the drawing of the capitulum does not agree with the type. The female from Sherwood was probably drawn on account of the poor condition of the type. Ixodes kingi is most closely related to I. cookei, but is readily distin- guished from that species by the following characters: siiorter, more slender legs; smaller capitulum ; much smaller and more widely separated porose areas; distinct lateral carinae; shield more narrowed behind; spine at base of coxae I less than one-half the length of that in /. cookei; stigmal plate much smaller than in cookei . In a large number of females of I. kingi the stigmal plate averages 294 m x 258/* while in a number of females of cookei, the stigmal plate averages 418/* x 360/*. In kingi the goblets are more widely separated and do not exceed five rows at the widest place, the total number not exceeding 96, while in i". cookei there are seven or eight rows and at least 122 goblets per plate. 204 Bishopp — Some New North American Ixodidse. The engorged females are almost globular, the length and width differ- ing but very little. In this respect this species is unlike other members of this genus. The males may be distinguished from those of cookei by the basal spine on coxae I being much shorter (less than one-half tbe length); the entire tick smaller; sides of shield parallel ; absence of punctures on posterior half of shield; presence of well-defined lateral carinae; shorter and broader palpi; shorter, smaller legs; smaller stigmal plates; absence of punctures on the ventral plates; width of anal frame in front of the anus much greater. Nearly all of the other specimens at hand have rather shorter and more slender legs and shorter basal spine on coxae I than the type and para- types. The scutum, in a number of specimens, is somewhat smaller than the type. The lateral carinae are distinct in all specimens. Ixodes angustus Neum. This species is abundant in various parts of the Northwest. The spec- imens at hand show no decided variation, although some have the porose areas less triangular and the scutum rather shorter and broader than typical. It is worthy of note that out of thirteen lots of adults of this species collected from hosts, only one contained male specimens. The thirteen lots contained forty females and two males. Certain data relating to specimens obtained by the Bureau in connection with the work con- ducted on the Rocky Mountain spotted fever tick may be of interest on account of their bearing on host relationship, distribution and seasonal occurrence of the species. The material is contained in the collection of the Bureau of Entomology at Dallas, Texas. During 1009 the following collections were made: 2 9? 3 nymphs and 3 larvae from Ochotona princeps, Meeteetse, Wyoming, July 7 ( W. V. King); 1 9 from Neotoma orolestes, Casper, Wyoming, August 30 (M. Cary); 2 9 IromEutamias townsendi, Detroit, Oregon, October :! ( V. Bailey) ; 6 9 and 3 nymphs from Sciurus mollipilosus, Empire, Oregon, October 12 (V. Bailey); 14 9 irom Sciurus mollipilosus, Empire, Oregon, October 1(>(D. D. Streeter). The follow- ing collections were made during 1910: 2 9 from (Sciurus h. douglassi) Mt. Lehman, B. C, March 12 (Dr. Seymour Hadwen); 3 9 and :'> nymphs from (Sciurus h. douglassi) Silver Lake, Washington, April 7 (B. A. Bruce); 1 9 from Sciurus sp. Pysht, Washington, March 23 (Guy Decker); 2 9 from rabbit, Mora, Washington, June 13 (A. W. Smith); 2 nymphs from Citellus columbianus, Florence, Montana, June 16 (W. V. King); 1 9 , 20 nymphs and 02 larvae from Ochotona princeps, Florence, Montana, June 16 (W. V. King); 1 9 from Sciurus h. rich- ardsoni, Lo Lo Hot Springs, Montana, June 23 (W. V. King); 2 9,7 nymphs and 2 larvae from Ochotona princeps, Lo Lo, Montana, June 29, (W. V. King); 1 9 from cat, Pysht, Washington, July 6 (Guy Decker); 2 cT, 4 9 and 4 nymphs from rabbit, Pysht, Washington, August 30 (Guy Decker); 1 nymph from dog, Pysht, Washington, August 31 (Guy Decker). Bishopp — Some New North American Ixodidse. 205 It should be stated that the determinations of the immature stages are not absolutely certain owing to the difficulty of positively identifying the immature stage.-- of Ixodes. In a few instances nymphs were bred to adult, and in the other case- I fed reasonably sure of the correctness of the determinations. Ixodes angustus var. woodi n. var. Female. — Caftitulum (Fig. 13), length 680 p. ( from tip of hypostome to a line drawn between posterolateral angles of basis capituli), basis capituli dark brown, greatest width 416^; postero-lateral angles rather less prominent than in I. angustus proper; the basal border of capitulum slightly concave; porose areas small and not well defined, nearly semi- circular with the flat side along the strong carinae which run toward the rostrum from the posterolateral angles of the basis capituli; length of porose areas along these carinae is distinctly greater than their width; widely separated, the distance between them being nearly equal to their length; ventrally the basis capituli is very long, comparatively narrow, and snu >oth ; distinctly longer and more narrow than in typical i". angustus; palpi similar to angustus, but the anterior portion of the flrst segment is distinctly produced ventrodaterally ; hypostome less pointed at tip than in angustus proper. Scutum, length 1.44 mm., width 1.07 mm., very similar to angustus, but rather more coarsely punctured and slightly more narrowed behind. Legs as in angustus, but heavier; all tarsi are rather abruptly narrowed near tips; coxae I with a moderately long hasal spine (longer than in /. angustus) and a large blunt apical spine; other coxae witli distinct broad apical spines, smallest on posterior coxae. Stigmal plates very small (179 fi x 158/*), transversely oval, about 49 goblets per plate, medium sized ; four rows at widest and one row at narrowest point between macula and marginal cells. Body with numerous, moderately short, pale hairs; anal frame widest near posterior margin, narrowed anteriorly to a blunt point in front of anus. Type, Cat. No. 18,976, U. S. National .Museum. Type host. — Bairn's wood rat (Neotoma micropus Baird). Type locality. — Sabinal, Texas. A female described from a lot containing two females and three nymphs collected May IS, 1910, by Messrs. F. C. Pratt and C. T. Atkinson (Dallas Accession No. 1641 ). Named for my associate, Mr. H. P. Wood. The collection of the Bureau of Entomology at Dallas, Texas, contains the following material which I have referred to this variety : 2 nymphs, 2 larvae, February 7, 1910; 2 ? May 10, 1910; 2 ? May 20, 1910; 3 nymphs May 31, 1910. All collected at Sabinal, Texas, on Neotoma micropus Baird, by Messrs. F. C. Pratt and C. T. Atkinson. No males were collected. 206 Bishopp — Some New North American Ixodidx. Ixodes sculptus Neum. But a few specimens of this species have been collected and these come from widely separated localities. The very narrow basis capituli, the incurved lateral carinae and small stigmal plates, differentiate it from the Ixodes cookei group. The some- what elongate scutum, narrow capitulum, rattier slender palpi and small stigmal plates indicate a relationship with I. angustus. In the type specimen the greatest width of the basis capituli (at base of palpi) is 467m- The width between the centers of the postero-lateral angles, which are distinct and incurved, is 373 m. The area bearing the porose areas is distinctly elevated above the lateral portions where the palpi articulate. This gives the appearance of still greater narrowness. The length of the capitulum from tip of palpi to postero-lateral angles is 747 m. The stigmal plates are transversely oval and measure about 233m x 1S7 m. Mr. Banks' collection contains two partially engorged females collected by Prof. J. M. Aldrich at Brookings, South Dakota, which agree very closely with the type. In these two specimens the maximum widths of the Oasis capituli are 420m and 439 m and the widths between the centers of the postero-lateral angles 327 m and 355 m respectively. In one of these specimens the stigmal plate measure 201 m x 233 m. The scutum of each is finely punctate and the lateral carinae strong and incurved at tips. But one of the specimens shows the depression between the porose areas. On account of the fact that many specimens of cookei have a more or less distinct scar between the porose areas, this character is of limited syste- matic value. The collection of the Bureau of Entomology at Dallas, Texas, contains 1 9 from Citellus columbianus, Bozeman, Montana, March 20, 1910 (Prof. R. A. Cooley); 1 $,4 nymphs and 5 larvae from wild cat, AVichita Mts., Oklahoma, December 21, 100'.) (Frank Rush); 1 9 from striped ground squirrel, Luther, Wyoming, September 24, 1910 (C. 0. Lyon); 1 9 from rock squirrel {Citellus variegatus couchi Baird) Devils River, Texas, May 4, 1007 (F. C. Bishopp). The specimen from Montana agrees with the type very well, but the lateral carinae are not strong and are but little incurved at their tips. The greatest width of the basis capituli is 445 m; the width between centers of postero-lateral angles is 337 m. Stigmal plates have about 36 goblets each and measure 201 m x 144 m. The female from Oklahoma has the capitulum pulled off, hence the determination is doubtful. The scutum and legs agree closely with the type. However, the stigmal plates are larger (373 m x 316 m and 395 mx 316 m) and have many more goblets. The specimen from Wyoming is typical of the species, although the basal spine on coxae I is slightly shorter than the type. The greatest width of the basis capituli is 457 m and the width between the centers of the postero-lateral angles 373 m- The stigmal plates are very small, measuring 177 MX 168 m- The female from Texas is from the same lot which Mr. Banks examined and placed under /. sculptus in his Revision. This Bishopp — Some Neiv Xorth American Ixodidsb. 207 specimen varies more Erom the type than the others herein listed. The greatest width of the basis capituli is 660/* and the width between the centers of the postero-lateral angles is 502/*. The porose areas are rather smaller than in the type and do not reach to the posterior border of the basis capituli. The leg's are longer and stouter than in the type. The stigmal plates measure about 316 m x 27.">/*. Haemaphysalis chordeilis Packard. In the course of the studies of the life histories of ticks being conducted by the Bureau of Entomology three lots of ticks were collected which con- tain males of this species. In three instances we have reared to adult males, specimens collected as nymphs. The three above mentioned collec- tions which contained males were made by Mr. J. I). Mitchell in Victoria County, Texas. The collections were made on meadow larks on the fol- lowing dates; November 12, 1909 ( 1 c?, 4 $ ), November 25, 1909 (5 44 /* wide, amber in color, rectangular, postero-lateral angles short but well defined, center of dorsal surface with a number of punctures; palpi amber in color, length from 258/* to 316/*; first and fourth segments very small; second segment produced laterally and ventrally to form a promi- nent but not acute angle; a few scattered yellowish bristles on palpi; the infra-internal edges bear a number of feather-like bristles; hypostome rather blunt with five rows of moderately small teeth. Scutum covers all of dorsum but a narrow marginal strip, light gray in color, shading into amber anteriorly; glaborous, shining rather thickly and coarsely punctured, some specimens showing a small rugose area at antero-lateral angles; fovae nearly black opposite or slightly posterior to fourth pair of legs; cervical grooves deep anteriorly, first converging, then diverging and disappearing posteriorly; postero-median groove dis- tinct, deepest at posterior end; accessory grooves shallow and broad; marginal groove deep, extending from a point opposite the second pair of legs to the fourth festoon. Legs yellowish brown, translucent, with numerous light yellow hairs; trochanters I with prominent blade-like process projecting backward; tarsi all tapering rather abruptly to end; coxae slightly darker than legs, with numerous yellow hairs; coxae I, II and III with very short, blunt basal spurs; coxae IV with a rather long, blunt spur. Stigmal plates broadly oval, sides nearly parallel for some distance; with a short, broad dorso-posteriad prolongation; size from 358/* long x l's7 /* wide to 373 /* long x 230 /* wide (width includes prolongation); macula small, situated toward antero-internal angle, rather closely cov- ered with many moderately small goblets of nearly uniform size. Body ellipsoidal in contour, widest posteriorly; marginal strip of dor- 208 Bishopp — Some New North American IxodMce. sum milk color; this color also occurs along the festoon grooves; festoons brownish slate color with punctures similar to those on scutum ; venter gray blue; festoons brownish with yellow centers ; genital groove distinct, almost parallel anteriorly, diverging behind posterior coxae and extending to a point on the marginal furrow between the second and third festoons; post-anal groove straight, forming aY at its anterior extremity, the forks of which extend forward on either side of the anus. Total length of specimens from 2.8 mm. to 2.9 mm., width from 1.5 mm. to 1.6 mm. Described from several living males collected in Victoria County, Texas. Dallas Accession Nos. 886 and 1012. The specimens at hand show a slight variation in the shape of the stigmal plates and basis capituli. Proc. Biol. Soc. Wash., Vol. XXIV. L911. Plate II. Explanation of Plate Ixodes cookei var. rugosus, Fig. 1, capitulum of female; Fig. 2, capitulum of male. Ixodes banksi, Fig. :'■. capitulum of female; Fig. 4, scutum of female; Fig. 5, coxae of female; Fig. 6, stigma 1 plate of female. Ixodes kingi. Fig. 7, capitulum of female; Fig. 8, scutum of female: Fig. 9. stigmal plate of female; Fig. 10, coxae of female; Fig. 11, capitulum of male; Fig. 12, ventral of male. Ixodes angustus var. woodi. Fig. 13, capitulum of female. Prawn by G, N. Wolcott. (Original.) Vol. XXIV, pp. 209-214 October 31, 1911 ' PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON SOME NOTES ON FISH SCALES. BY T. D. A. COCKERELL. Last summer, thanks to the Bureau of Fisheries, I was enabled to continue my work on fish scales at Woods Hole, where I obtained much interesting material not previously available. Thanks to Dr. F. B. Sumner, Dr. Hugh M. Smith, Dr. B. W. Evermann and others, I have before me a remarkable collection of fish scales, which will form the subject of a report later on. In the meanwhile, the following notes are offered. They were written and sent for publication before I knew how much mate- rial I was to obtain, or that means would be found for the publication of a detailed illustrated report. PLEURONECTID.E AND SOLEID.E. The following table will serve for the separation of the scales of a -eries of flat-fishes, all in the collection of the Bureau of Fisheries Station at Woods Hole, except Platophrys constellatus, which is from the collection of the Bureau in Washington. A. Scales ctenoid on both side- (Soleidse). A.pical teeth large and few (6 or 7 on upper, 3 to 5 on lower side), with a well-marked zone of large reticulations at their base . Achirus I 'meatus (L. ). Xo such zone of reticulations at the base of the teeth 1. 1. Scales narrow, with 8 to 20 basal radii Solea solea (L. ). Scales larger and broader, with over 30 basal radii Symphurus piger (Goode and Bean). Gulf Stream. < If these, the Achiras must be considered the most modified. The three genera represent the subfamilies Achirime, Soleinee and Cynoglossinae. B. Scales ctenoid above, cycloid on the lower (blind) side (Pleuronectidse). 3S— Peoc. Biol. Soc. Wash., Vol. XXIV, 1911. (209) o 210 Cockerell — Some Notes on Fish Scales. Scales large, minutely ctenoid above 1. Scales minute 2. 1. Lateral areas with very fine radiating stri;e . Syacium papillosum (L. ). Albatross Sta. 2405. Lateral areas without such stripe . . . Citharichthys arctifrons Goode. 2. Basal radii so many and close as to produce the effect of a fine striation ; scales broad 3. Basal radii not thus numerous 4. 3. Apical teeth of scales on upper side of fish projecting beyond margin Platophrys constellatus Jordan. Magdalena Bay (Albatross). Apical teeth of scales on upper side submarginal; so that the scales seem cycloid Platophrys ocellatus Agassiz. Albatross Sta. 2414. 4. Basal radii many, about 15-20 on lower, over 20 on upper side; teeth of ctenoid scales few (about 5 to 7) and very sharp . . . Pseud opleuronectes americanus Wall). Basal radii fewer 5. 5. Lateral areas, free from radii, extensive, their circuli dense . . Notosema dilectum Goode and Bean. Lateral areas, free from radii, small <>. (i. Scales smaller, radii more parallel .... Limanda ferruginea Storer. Scales larger, radii less parallel Hippoglossoides platessoides Fabricius. The last two are nearly alike. C. Scale cycloid on both sides (Pleuronectidse). Scales very small, long-oval, basal radii few Glyptocephalus cynoglossus L. Albatross Sta. 2546. Scales much broader, basal radii many 1. 1. Scales smaller; those of lower side about half size of those on upper Lophofjsetta maculata Mitch. Scales larger; those of lower side not so much smaller than those of upper Paralichthys oblongus Mitch. The Glyplocephalus scale is the most modified (degenerate) of the whole series. Some Achirinre (not seen by me) are scaleless. With regard to the subfamilies of Pleuronectidse, it will be noted that each subfamily has produced a genus with cycloid scales on both sides. It is is obvious that the flat-fishes are descended from fishes with the typical Acanthopterygian (ctenoid) type of scales, and that the cycloid scales have arisen through loss of the ctenoid character. From all we knowabout inheritance, it appears practically certain that this character, once completely lost on the lower side, would not return in the well- developed condition in which it occurs in the soles. It must, therefore, be held that the Soleidfe are not derived from the ordinary Pleuronectidse, but rather from some ancient type of flat-fish, probably now without liv- ing representatives. There is, however, a living genus of Pleuronectids Cockerell — Some Notes on Fish Scales. 211 having ctenoid scales on both sides like the soles, namely Ancylopsetta, said to be very close to Paralicldhys.* I had readied these conclusions from the scales, when I fortunately mel Professor G. II . Parker, and mentioned them to him. He at once kindly directed my attention to his paper on the optic nerves of flat-fishes, pub- lished in the Bulletin of the Museum of Comparative Zoology, Vol. XL ( L903). In this work he shows that the condition of the optic chiasma in the Soleidse is as in ordinary Teleosts, whereas in the Pleuronectidse it is modified. He therefore concluded that " the Soleidse are not degraded Pleuronectidse, but degenerate descendants of primitive flat-fishes, from which the Pleuronectidse have probably been derived." The Soleid genera examined by Professor Parker were the same as mine; his Pleuronectids were Atheresthes, Eppsetta, Hippoglossoides, Pseltichthys , Paralichthys, Hypsopsetta, Parophrys, Isopsetta, Oncopterus, Limanda, PseudopJeuronectes, Pleuronectes, Liopsetta, Olyptocephalus, Lophopsetta, Platophrys, Syacium, Azevia, Citharichthys , Etropus. Unfortunately, he did not see Ancylopsetta, but it seems very unlikely that it would de- part from the condition he found in all the other genera. GADID.E AND ALLIES. The table below separates the scales of a number of Gadoids, the speci- mens all from the Bureau of Fisheries Station at Woods Hole, except the Macrouridse, which are (except Macrourus sp. ) from the U. S. National Museum. A. Apical field of scales with long spines (Macrouridse). Basal circuli sharply angulate in the middle 1. Basal circuli not sharply angulate 2. 1. Apical teeth shorter, densely set, no circuli visible between them Macrourus bairdii Goode and Bean. Lat. 39.53 N., Long. 70.9 W. (Albatross). Apical teeth longer, not densely set, circuli very conspicuous be- tween them Macrourus sp. Albatross Sta. 2426. 2. Scale very broad and short; teeth very long, many projecting far beyond margin Coryphxnoides rupestris Gunner. (Albatross. ) Scale not greatly broader than long; teeth shorter, not or hardly projecting beyond margin . Coelorhynchus caribbeeus (Goode & Bean). These scales must be called ctenoid, yet the margin itself is not dentate, except slightly in Macrourus sp. The scales of Capros, as figured by Goodrich, have similar teeth; Capros is, of course, an entirely different fish. "Jordan & Evermann, Bull. 47, U. S. N. M., Part III, p. 2634. In this work the scales of Paralichthys oblongus are said to be " weakly ctenoid or cycloid.'' Those examined by me are strictly cycloid. 212 Cocker ell — Some Notes on Fish Scales. B. Scales cycloid, wholly without spines (Gadidse and Merlucciidse). a. Scales without radii (Merlucciidse, Phycinse, Gaidrop- sarinse). Scales elongate, with a strong median transverse furrow, variously developed, sometimes only near the margins, never crossing the nucleus; circuli not angled . . . . Enchelyopus cimbrius (L.). Eel Pond, Woods Hole. Scales larger, rather broad, without such a distinct median fur- row, and with the basal circuli variously angled 1. 1. Region above the nucleus (except in latinucleate* scales) con- spicuously finely pitted Merluccius bilinearis (Mitchell). Woods Hole. Mass. Region above the nucleus not or hardly pitted Urophycis regius ( Walb. ). Woods Hole. .Mass. 6. Scales with numerous radii (Gadinse, Brosminse). Scales very long and narrow (about 4 nun. and 1% broad), the central area usually free from sculpture and the tine radii broken Brosme brosme (Mull.) Scales not so long and narrow, usually with a fine reticulated sculpture all over, though latinucleate scales of Microgndus re- semble those of Brosme in structure though not in shape . . . 1. 1. Scales comparatively large, about 7 to 8 nun. long and -I to 5 broad Mel anogr animus seglefinus (L.). Scales smaller, less than 5 mm. long 2. 2. Scales smallest (about 2 mm. long), narrow . Pollachius virens (L. ). Woods Hole. Scales medium (between 3 and 5 mm.) .... Gadus callarias L. and Microgadus tomcod ( Walb. ). The scale- of the Macrouridse, except for the shape (round or trans- versely elongate) and the spines, agree with those of the Merlucciidse and Phycinse, having more or less angled circuli and no radii. Those of the Gadinse and Brosminse are entirely different, and could not have been derived from those of the others. According to the scales, some type allied to the Macrouridse may have given rise to the different lines repre- sented by Merluccius, Urophycis and Enchelyopus; but the Gadinse with the Brosminse must have another origin, although no doubt all can be referred to a common ancestor by going back far enough. There is no essential difference between the Gadine and Brosmine scales, but in my preparation of Brosme all the scales are latinucleate except one. * Latinucleate scales are those in which the nuclear region is broadly modified, the normal sculpture largely suppressed or altered, and the radii more parallel. This dimorphism is of general occurrence among fish scales, and appears to be due to regen- eration after loss of the original scales. Cockerell- Some Notes on Fish Scales. 213 CYPkiMD.i:. The following interesting species arc described from the collection of the Bureau of Fisheries at Washington and {Algansea and Lepidomeda) the National Museum : (1.) Algansea sallaei (Giinther). Scales oblong-, subquadrate, nearly parallel-sided, slightly over 2 nun. long, about 1| wide; nucleus subbasal; radii all around; apical radii about 10 or 12, the inter- radial circuli very widely spaced. This is one of the few American Cyprinids having basal radii; the scales are not unlike those of Temeculina. The fish is from Guanajuato, Mexico (A. Duges). (•_'.) Mylocheihts caurinus (Richardson). Snake River, Idaho (Ever- mann). Large scales, about b% mm- l°ng an^ 5 broad; nucleus subbasal; no basal radii; apical radii about 25 or more; apical circuli widely spaced, not at all angulate in middle; very obtuse hiterobasal angles. Except for the non-angulate apical circuli, this reminds one of Semotilus. (3. ) Richardsonius siuslawi (Evermann&Meek). Siuslaw River, Oregon (S. E. Meek). Scales quite normal for the typical subgenus Rich- ardsonius ; apical radii about 18, some imperfect. (4.) Exoglossum maxillingua (Le Sueur). Roanoke, Ya. Scales about 2% to 2% mm. long and broad; nucleus subbasal; no basal radii; apical radii about 20; apical circuli widely spaced, not angled; laterobasal angles evident but obtuse. The structure is almost as in Mylocheilus. (•">.) Lepidomeda vittata Cope. Colorado Chiquito River. Transversely oval scales, with the usual subbasal nucleus; apical radii about 7, strong; no basal radii; apical circuli extremely widely spaced. These scales are quite different from those of Gila, a fact useful for the separation of the fishes from juvenile Gila, occurring in the same region. Vol. XXIV, pp. 215-216 October 31, 1911 PROCEEDINGS OK THE BIOLOGICAL SOCIETY OF WASHINGTON A NEW TEAT, FROM THE ANDAMAN ISLANDS. BY J. H. FLEMING. While cataloguing a number of birds from the Andaman Islands, collected in 1905, by Mr. B. B. Osmaston, and now in my collection, I noticed that in a series of twelve Andaman Teal, only five agreed with the description and plate of Nettion albigulare ( — Poliowetta albigularis) in Vol. '27 of the Catalogue of Birds. Six from North Reef Island (to the westward of North Andaman), and one from Middle Andaman differ suffi- ciently from the five from South Andaman to be described as a new subspecies, which may be known as — Polionetta albigularis leucopareus subsp. nov. Coti/pes from North Reef Island, Andaman Islands, <$• and 9 , collected by B. B. Osmaston, December 25, 1905. Fleming Coll. Nos. 19,064 (d1), and 19,069 ( $ ). Character*. — Similar to P. albigularis albigularis (Hume), but averaging larger, with a great deal of white on head and neck; margins of body feathers cinnamon brown and better defined, producing strongly defined splits, especially on the sides of the breast. c?. Pileum, except frontal apex, very dark brown, this color descend- ing the sides of the face behind the auricular region, and partially sur- rounding it below ; sides of upper neck light brown. Frontal apex and whole face, including eyebrows and auricular region, pure white, this color extending over the throat and lower neck, and continuing in a wide collar around the hind neck. $ . Entire pileum dark brown becoming lighter on the sides of the head and neck, confining the white collar to the hind neck, and reaching for- ward below the auricular region. Auricular region, eyebrows, front of face, throat and lower neck as in the male. 39— Proc. Biol. Soc. Wash., Vol. XXIV. 1911. (215) 216 Fleming — A New Teal from the Andaman Islands. Table of Measurements. (Measurements in inches.) Flem- ing Col. No. Sex. Date. s is '3 'Si w u a a o Polionetta albigularis leuco- pareus. 19,065 male ad. Dec. 25, 1905 North Reef Island, Andamans 8.20 3.23 1.45 1.35 *19,064 " " " " ' ' 8.00 3.16 1 .42 1.4S 19.066 ,1 i . .. ,i 8.00 3.12 1.40 1.43 19,067 •* " " " *' " t i.n 3.03 1.45 1.49 19,068 " " " '* " ;• t 7.78 2.83 1.34 1.42 *19,069 female " * ' * * " 7.91 2.88 1.40 1.50 19,075 ( male) Jan. — North Andaman. 7.80 2.78 1.40 1.52 Polionetta albigularis albi- gularis. 19,073 male ad. Dec. 10, 1905 Port Blair, South Andaman 7.60 2.80 1.32 1.47 19,071 '.! * * * ■ * * . . ,i • . " 7.38 2.88 1.37 1.37 19,072 female * i ** ' * ii .« i. " 7.7(1 2.57 1.40 L.37 19,070 " " * ' '* ** 7. IS 2.61 1.31 1.35 19,074 .iuv. Oct. 10 " Andamans 7.20 2.83 1.25 1 :ir< * Types. + All primaries in quill. t First primary in quill. Remarks.— I am indebted to Dr. L. B. Bishop, of New Haven, Conn., and Air. W. R. Ogilvie-Grant for comparing two of my North Reef Island birds (the types) with the series of Andaman Teal in the British Museum and the types of P. albigularis, and both gentlemen agree that my form is at least a good sub-species. I have some hesitation in including the single bird from North Andaman in the new form ; it is intermediate in coloring and is probably the connecting link between the forms. The white on the frontal apex varies, being rednced to a line in that of the Reef Island birds, and one ( 19,065) lias a great deal of white on the crown, 19,066 has a little, and 10,064 a small spot. Vol. XXIV, pp. 217-218 October 31, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON DESCRIPTION OF A. NEW SUBSPECIES OF THE PAINTED BUNTING FROM THE INTERIOR OF TEXAS. BY EDGAR A. MEARNS. The Nonpareil, or Painted Bunting, of western Texas proves to be an undescribed race which may be named — Passerina ciris pallidior subsp. nov. PALE PAINTED BUNTING. Type-specimen.— Adult male. Cat, No. 163673, U. S. X. M. Collected at Fort Clark, Kinney County, Texas, May 7, 1898, by Edgar A. Mearns. Original number, 11800. Characters. — Larger than Passerina ciris ciris ; adult male with under parts considerably paler vermilion red; rump and upper tail-coverts paler, more purplish red. Female paler throughout, with upper parts paler, grayer, less yellowish green, and under parts buffy instead of decid- edly yellow. Measurements of type (taken from the fresh specimen by the author). — Length, loo mm. ; alar expanse, 240 ; wing, 75; tail, 61; culmen (chord), 11.2; tarsus, 18.3; middle toe with claw, 18.3. Geographical range. — Breeds in the interior of Texas, migrating to Ari- zona, Mazatlan, and the plains of Colima. Hi marks.— Topotypes of Passerina ciris ciris from Vera Cruz, eastern Mexico, are indistinguishable from specimens from the southeastern United States, and are probably migrants from the latter region. The breeding range of typical ciris extends westward from the South Atlantic States around the Gulf coast to Mexico. A table of comparative measurements appears on following page. 10— Proc. Biol. Soc. Wash., Vol. XXIV, 1911. (217 218 Mearns — A New Subspecies of Painted Bunting from Texas. CC M - F § 0) o t 2 0J po jH X S =0 ■ P Q rn 3 ^s a a> m S ^ x ~ P»-53 ^j f-HIV^T-H*^ o ^""•^OOOOCOC . ts . T3 03 . 73 ^ JOB O ° • r^rj,j"3",tn: O 3wa ^ d d •M -t N fM C-l i-H 0 co go (N ci x r; 7- a '-f- 5 P o3 o'oo — oooooo 7J *-H — o HHHri! — It-Hi — 1 l — 1 ^ I — i t-Mt— r— — * t— It— 1 t-h t— i — i 'c3 NCIHhOt-iQOOJ O "M -v -v -~ i - -o -o t^ IO n N H iOiOCOCDCOCOiOHC, OvD io wi io ic «i io io io io io bo •M -t- 71 CO CO iC 7-1 01 77 -f - 1 - — X 02 1^ t~- X X ~. £ 1^ 1 - 1^ [^ 1^ |^ l^ t^ l^ 1^ oi x ~ co x x x ■— 1 I— 1 T— 1 T— 1 1 Til— Itl- ~ x oa a: 3i 32 i— It— li— It— It— IOt— c^t- 03 X X X X X X i-^ 1— rH ■1 loToTira x~'cTr^ o o o o r-T co o co cc x" : r ad 3 G C I i— 7 1 T3 -c3 73 "z t— 1 C-l t—i 7 1 H , , , — — 1 •— - ^i t_, •>} J^ t*3 K*} >-:£ >> >3 >•- 2 !>» "H 7: ~ 7: ~ Zl 73 S "1 S "J S ~ ?3 o . 03 C3 73 — 73 pt S 3 . >~. Z CO 7- C3 3-i — L. '*-. a. rR • 3 CS '-? 2 73 ~ o o *c^ 0 ^ , , bO "3 o _S- . • ooooooc CJ ..oooo.s - • 5 ^Z "Z - 'TH X X T3X t: '0^3 ^ 73 O _, X.2T.— g ?~ -* - -3 - ~ " 5^, l2 =*3 &3 Ph -" 0 s« 3 so : = o p 0 CO 0 B h - -/ ^ 0 3 -S ■>, 2 +-* ^3 73 »*» 73 ?? 13 3 i: N OJ JJ " C3 r; TJ 2 i3 ^ 03 Is ro co *f X 3: O >0 "-D t- c: "O CC CO -P ^^ CO 5 — coooaooooor- CM O CC X 10 iC •- § l^- i— l t^ I - X X X X X 1— ri x x 1—1 1—1 3Sl— It— It— It— ll— It— It— 3: 3^ rH i—i w^ S-i S T— < T— ( T— 1 ~H T— 1 T— 1 1— r^ t-h 1—1 — -J t^ lo cc o h ec >o cm -+ t> •t f 1 -t< N X lOMHC ~ »l — £ ©t^t^t^t^t~t~I> -+ 10 10 10 lo n 10 r- n y; x> ■* io co -o -o co -o vr -o 35 1—1 1—1 CO CO CM l-i-O *a S i — i ^c - ^ *~ cc cc co co co ~* *— C-l CI T 1 71 X ~ 1 T 1 •-. CO cS 3 0 1 To CO ~ — ~ — "O "O cc O u r/1 0 - M ti; - H — 0 ^H cS 1 L / O "3 O / ; - — O 0 =1-1 - '/: '- - bo -T" a - | jj O 1 IS | 00 a. «J 2 2 § ^ *- b W U. S. National Museum. Collected at head of Lari River, near base of Pico Blanco, Talamanca, Costa Rica, by Win. M. Gabb. Altitude about 6,000 feet. Diagnosis. — Size and general appearance as in the Cryptotis mexicana group (including the Costa Rican C. orophila) ; skull narrow and elon- gated, this peculiarity more pronounced than in the small G. tropicalis and C. soricina. Color. — Entire animal a dark smoke-gray, the upperparts heavily washed with blackish brown, the general effect darker and less plumbeous than in a skin of C. orophila taken in February; under-color slate-gray. Skull. — The skull about equals that of Cryptotis mexicana or C. orophila in length, but the width throughout is noticeably less, this peculiarity equally appreciable in braincase, interorbital region, rostrum, and palate. Braincase nearly circular in outline when viewed from above, the lateral angles barely indicated; viewed from behind, it is conspicuously narrower in proportion to its depth than in C. mexicana and C. orophila and appreciably more so than in C. soricina and C. tropicalis. While the braincase retains essentially the same depth as in the other species of similar size, the rostrum is lower, so that the general outline of skull in lateral view is more strongly cuneate. Antorbital foramen smaller than in C. mexicana or C. orophila, but in same position, its posterior border over space between in1 and m2. Mandible decidedly more slender than in mexicana and orophila, but not peculiar in form. Teeth. — Upper incisors and unicuspids similar in general to those of 42— Peoc. Biol. Soc. Wash., Vol. XXIV, 1911. (221) 222 Miller — Three. New Shrews of the Genus Cryptotis. C. mexicana, the unicuspids showing no tendency toward the con- spicuously concave posterior border characteristic of these teeth in C. orophila ; inner tubercle of unicuspids faintly developed, much as in C. mexicana per egrina. Upper cheek-teeth sharing in the general reduc- tion of width characteristic of entire skull, this tendency especially noticeable in the large premolar, the posterior border of which is decidedly shorter than outer bonier, while in C. mexicana and C. orophila they are about equal; emargination of posterior borders slight, essentially as in mexicana and orophila ; hypocnes somewhat better developed than usual ; m3 with well developed metacone and third commissure. Mandibular teeth less robust than in C. mexicana and C. orophila but with no special peculiarities of form. Measurements. — Head and body,* 70; tail,* 35; hind foot, 13.6 ( L2.4); condylobasal length of skull. 18.6 ( L9.4)t; lachrymal breadth, 4.2 (5.0); zygomatic breadth, 5.2 (6.0); distance between zygomatic root and gnathion, 8.2 (8.2); breadth of braincase, 9.0 (10.0); depth of braincase, 5.4 (5.4); mandible, 10.4 (11.0); maxillary toothrow (entire), 8.8 (9.0), mandible tooth row (entire), 7.8 (8.2). Remarks. — Three very distinct species of Cryptotis are now known to occur in Costa Rica: the medium sized C. orophila on the Volcano of Irazu, C. gracilis in Talamanca, and the small C. nigrescens at San .lose. The last is further distinguished from the two others by the reduced con- dition of the posterior molar both above and below; m3 with mesostyle, metacone, and their commissures very small and barely differentiated, m3 with posterior V reduced to a loop scarcely more than one-fourth as large as anterior V. Cryptotis frontalis sp. nov. Type— Adult female (in alcohol) No. 123,429, U. S. National Museum. Collected near the City of Tehauntepec, Mexico, by F. Sumichrast. Diagnosis. — Size and general appearance as in C. mexicana mexicana. Skull with more robust rostrum and with braincase unusually deepened posteriorly, and rising anteriorly at a much more conspicuous angle (in this respect noticeably surpassing C. nelsoni); teeth larger than in true mexicana but inner tubercles of upper unicuspids less developed, though retaining distinctly pigmented tips. Measurements. — Head and body, 06; tail, 27; hind foot, 12.2 (11.0). Skull: condylobasal length, 19.0(18.6)1; zygomatic breadth, 6.2 (6.0); lachrymal breadth, 5.0 (5.0); breadth of braincase, 10.2 (10.0); depth of braincase, 6.2 (5.6); mandible, 10. (i (10.0); maxillary toothrow, 8.6 (8.6); mandibular toothrow, 8.0 (8.0). *Approximate ; from skin. t Measurements in parenthesis are those of an adult female topotype of C. orophila (No. 116,649) . t Measurements in parenthesis are those of an adult female topotype of C. mexicana mexicana. Millet — Three New Shrews of the Genus Cryptotis. 223 Cryptotis pergracilis macer eubsp. nov. Type. — Adult female (in alcohol) No. Iffff, near Guanajuato City, Mexico, by A. Duges. Characters. — Similar to Cryptotis pergracilis pergracilis Elliot, but with third upper unicuspid relatively larger, its area when viewed from the side distinctly more than half that of first unicuspid. Measurements. — Head and body, 50; tail, 17.4; hind foot, 10 (9.2). Skull: condylobasal length, — (16.0)*; from gnathion to posterior border of parietal in median line, 14.0 (13.8); zygomatic breadth, 5.0 (5.0); lachrymal breadth, 3.6 (3.6); mandible, 8.4 (8.4); maxillary toothrow, 7.2 (7.0); mandibular toothrow, 6.6 (6.4). Remarks. — In three specimens of Cryptotis pergracilis pergracilis from Ocotlan, Jalisco, the type locality, the third upper unicuspid is uniformly less than half as large as first when viewed from outer side. * Adult female, Topotype of pergracilis. Vol. XXIV, pp. 225-230 October 31, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON GENERAL NOTES. THE NOMENCLATURE OF THE CHEETAHS. Two forms of cheetahs are commonly recognized, Cynailurus jubatus of India and C. guttatus of Africa. A review of the nomenclature of the genus results in an unavoidable and entire change of names. The generic name Acinonyx Brookes, Catalogue of the Anatomical and Zoological Museum of Joshua Brookes, Esq., p. 16, 1828, proves, on examination of this work, t'> lica valid name, and it has priority over Cynailurus Wagler, 1830. The first name to be applied to a cheetah was Felis jubata Schreber, pi. cv, I>ie Siiugthiere, 177<>. From the text of this plate, published in 1 777, the typedocality may be settled definitely as the Cape of Good Hope. Schreber says: " Das Vaterland dieses Thieres ist das siidliche Afrika; man bekommt die Felle voni Vorgebirge der guten Hofnung" (vol. Ill, pp. 392-393). This is followed by a statement that Pennant says it is also found in India, and an account of its use for hunting. The name jubata must, therefore, be transferred to the African cheetah. The earliest name for the Indian form is Felis venalica Hamilton Smith, 1S27. At the time this name was proposed, jubata was restricted to Africa. Of the remaining names which have, at various times, been used for cheetahs, four apply to African animals, one to the Indian species, and three may be disregarded as not originally applying to cheetahs or as indeterminable. Felis guttata Hermann, Obs. Zooh, p. 38, 1804, is based primarily on " Prosp. Alpini Aegypt. tab. xv, fig. 1, p. 238" (really fig. 2). The editor of Hermann's work, Frid. Lud. Hammer, has supplied a description in brackets. Alpinus' plate proves to be a drawing of a spotted cat, which the text explains was one of a litter of five seen in Cairo in the possession of a Greek courtesan. These kittens were said to have recently been born of a panther and were purchased of Arabs. The description and account which follows is very confusing and it is prol >ably impossible to identify the animal. The name can not, at any rate, be applied to a cheetah. In Griffith's Cuvier, V, p. 175, I find the statement that " F. guttata, of Hermann, figured by Schreber, is stated by M. F. Cuvier to be a young panther." Felis chalybeata Hermann, Obs. Zooh, p. 36, is likewise not determinable, andean be ignored as a cheetah name. It probably really refers to an American spotted cat, though the name later applied to such an animal by various authors is said not to be the 43— Pkoc. Biol. Soc. Wash., Vol. XXIV, 1911. (225) 226 General Notes. chalybeata of Hermann. I am not able to satisfy myself that F. megabalia Hartmann, Zeitschr. ges. Erdkund Berlin, 1868, III, no. 13, p. 55, applies to a cheetah. The animal is said to live in central Africa and to resemble the F. jubata of Schreber's plate. From an examination of the specimens in the National Museum it seems probable that Acinonyx soemmeringii (Fitzinger) will have to be recognized as a distinct form, but the material is at present too scanty for conclusive results. The two currently recognized forms will stand as follows : Acinonyx jubatus (Schreber). AFRICAN CHEETAH. 177(i. Felis jubata Schreber, Die Saugthiere, pi. cv. (Cape of Good Hope). 1855. Ci/nailurus soemmeringii Fitzixger, Sitz.-Ber. Math. nat. cl. d. K. akad. Wiss., xvi, lift. 2, p. 245. (Bajuda Steppe, Kordofan.) 1868. Fel 'is jubata var. africana Hartmann, Zeitschr. ges. Erdkund Ber- lin, III, no. 13, p. 56. (New name for the combined C. guttatus and C. soemmeringii of Fitzinger. ) L869. Felis fearonis Fitzixger, Sitz.-Ber. Kais. Akad. Wiss., lix, 1 abth., p. 664. (Cape of Good Hope, 1. c, p. (567. ) 1S77. Felis lanea Sclater, Proc. Zool. Soc. London, p. 532. (Beaufort West, Cape Colony. ) Acinonyx venaticus (Smith). INDIAN CHEETAH. 1827. Felis venatica Smith, Griffith's Cuvier, V, p. 166. (India.) 1828. Acinonyx venator Brookes, Cat. Anat. & Zool. Mus. of Joshua Brookes, p. 16. (India.) — JV. Hollister. ON THE CORRECT NAME FOR THE RED-WINGED BLACKBIRD OF THE NORTHEASTERN UNITED STATES. Examination of the series of breeding Red-winged Blackbirds recently collected by Messrs. E. J. Brown, J. H. Riley, and E. A. Mearns, in Charleston County, South Carolina, develops the fact that the form resi- dent on the coastal plain of South Carolina is identical with the Florida bird at present known as Agelaius phozniceus floridanus Maynard, with which it exactly agrees in size and coloration; and resident South Caro- lina birds also have the longer and more slender bill of the Florida form. Linnaeus based his [Oriolus] phozniceus on Catesby's Sturnus niger, alls supeme rubentis (Nat. Hist. Carolina, I, p. 13, pi. 13). Catesby's figure of an adult male indicates, by the complete nuptial plumage and slenderness of the bill, that the form portrayed was the resident bird of South Carolina, and not a migrant or winter resident from the north. The text also shows that the author was very familiar with the breeding haunts and nesting habits of the bird, with relation to the ocean tides, making it very probable that the individual figured was taken on the coast. General Notes. 227 Agelaius phoeniceus floridanus Maynard (1896) therefore becomes a synonym of Agelaius phoeniceus phoeniceus (Linnaeus, L766), and the northern subspecies must be known as Agelaius phoeniceus predatorius ( Wilson ). NORTHEASTERN RED-WING. Sturnus predatorius Wilson, American Ornithology, IV, 1811, p. 30, pi. 30, figs. 1 and 2. Characters. — Larger than Agelaius phoeniceus phoeniceus* with a shorter and stouter bill. Female darker, especially below, where the black stripes are much broader, t Geographical distribution. — This form breeds from Fort Macon, North Carolina, and Smith's Island, Virginia, north to Nova Scotia, and west- ward, wintering as far south as ( ieorgia and the Gulf States. % Remarks. — Wilson's figures, descriptions and measurements all rep- resent the northern subspecies; and his reference, in synonymy, to specimens in Peale's Museum ( " No. L466, 14G7") make it probable that his material came from eastern Pennsylvania, which region 1 therefore fix as the type locality of his Sturnus predatorius. — Edgar A. Mearns. NOTE OX THE MEXICAN BATS OF THE GENUS DASYPTEEUS^ Mr. Geo. F. Gaumer has recently presented to the U. S. National Museum some specimens of Dasypierus taken at Izamal and Yaxcash, Yucatan, which represent two very distinct species. One is a large animal witli skull about 18 mm. in condylobasal length, while the other is noticeably smaller; condylobasal length of skull about 15 mm. The difference in size between the two animals is thus about the same as that separating the European Nyclcdus noctula and N. leisleri. Apparently tins is the first instance on record of the occurrence of two members of the Deisypterus ega group at one locality. The smaller animal, represented by adults only, appears to be the D. ega panamensis of Thomas; unquestionably it is a local representative of true ega, a species which in its various geographic forms ranges from Argentina to Lower California. It maintains a uniformly small size throughout a very extended area, the range of individual variation in condylobasal length of skull in specimens from Paraguay, Brazil, Bolivia, Yucatan and Lower California being from 14 to 15.6 mm. The larger species, represented by both adults and young from Yucatan, is equally constant, as the range of variatii >n in adult skulls from Yucatan, Chiapas, || *See Ridgway's Birds of North and Middle America, vol. II, 1902, pp. :;:;i and 333. iThe darkest female specimens in the collection of the United states National Museum were collected at Plum Island Marsh. Kssex County. Massachusetts, by Mr. William D. Carpenter, in June, 1911. X Numerous winter specimens of both sexes, from South Carolina, in the U. 8. National Museum collection, are all predatorius. 5 By permission of the Secretary of the Smithsonian Institution. [Adult male (skin and skull). No. 133,030 (Biological Survey Collection) , San Barto- lome, Chiapas, March 15, 11)04, Nelson and Goldman. 228 General Notes. southern Texas, Louisiana and Florida is only from 17.11 to IS. 4 mm. This larger animal is Dasypterus intermedins (II. Allen), originally de- scribed from Matamoras, Tamaulipas. While it is not possible, in the absence of a better series of skins than that now available, to find any appreciable differences between the Yucatan and Chiapas specimens and true intermedins, it is evident that the form occurring in Louisiana and Florida ( D. floridanus Miller) is sufficiently characterized by its shorter forearm and third ringer to be regarded as distinct. — Gerrit S. Miller, Jr. THE VOLCANO RABBIT OF MOUNT IZTACCIHUATL.* In the catalog of the objects exhibited by the Mexican "Comision Geografico-Exploradora " at the Columbian Exposition in Chicago t Pro- fessor Augustin Diaz figures (pi. -L' ) a rabbit of the genus Romerolagus, under the name " Conejo del Volcan (Lepus diazi Ferrari -Perez, sp. nov)." This plate is not mentioned by Merriam in the original descrip- tion of Romerolagus nelsonii from Mt. Popocatepetl, or by Nelson in his monograph of the Rabbits of North America, § though it is alluded to by an anonymous editorial writer in Natural Science for March, 1897. || The validity of the specific name diazi is unquestionable under the Internationa] Code of Zoological Nomenclature (Art. 25, and Opinion 1). According to the same Code the authority for the name is Diaz, not Ferrari-Perez, since Lepus diazi remained in MS. until the publication of the catalog by Diaz, and no statement is anywhere made that " some other person is responsible for [the publication of] said name and its indi- cation [= illustration]" (Art. 21, and Opinion 4). Nonienclatorily the case presents no difficulties, but zoologically it has hitherto been obscured by the uncertainty as to whether the " Volcano Rabbit" was identical with Romerolagus nelsoni. This doubt has now been removed by the kindness of Professor Ferrari-Perez, through which I have been enabled to examine the type and a second specimen of Romerolagus diazi. Both were taken in Puebla, on the eastern slope of Mt. Iztaccilniatl, a region about fifteen miles from the type-locality of Romerolagus nelsoni. As compared with four skins of Romerolagus nelsoni the two specimens of R. diazi (both marked female, the type taken in March, 1885) show slightly less black in the grizzle of dorsal surface and slightly more gray on cheeks, characters probably due to the fading usual in mounted spec- * By permission of the Secretary of the Smithsonian Institution. -rExposicion Internacional Columbina de Chicago en 1893 | ComisiYm Geografico- Exploradora | de la I Reptiblica Mexicana | Catalogo | de los objectos que componen el eontingente de la Comision, precidido de algunas notas | sotare su organization y trabajos I por el ingeniero director I Agusttn Diaz | Coronel de Estado Mayor Especial, Ex-Profesor en el Colegio Militar y en la | Escuela Especial de Ingenieros I Xalapa- Enriuuez. Marzodel893 I Tipografia de la Comisi6n Geografico-Explnntdora. | t Proc. Biol. Soe. Washington, X, pp. 109-174. December 29, 1896. § North American Fauna No. 29. August 31, 1909. II X, p. 151. General Notes. 220 imens. The skull of the type (that of the other .specimen is missing) lacks the mandible and the entire basal part of braincase, though the remaining portion is in fair condition. It has certain peculiarities not shared by any of the live skulls of R. nelsoni: the postorbital processes are broader and heavier, the palate is more arched between toothrows, and the dorsal profile of braincase is less bent downward posteriorly. These cranial characters are, however, quite within the range of normal variation in members of the family. I consequently have no hesitation in regarding the later name nelsoni, as a synonym of diazi. Mr. E. W. Nelson, who examined the specimens with me, is of the same opinion.* Measurements of the type specimens (those of R. "nelsoni" in paren- thesis): hind foot, 48.6 (53); ear from crown, 38.6 (37); greatest length of skull, 60.4 (60.6); zygomatic breadth, 2!)+ (30.4); breadth of brain- case, 22.6 (23.4) ; postorbital constriction, 10. 0 (9.8); interorbital con- striction 9.8 (10.0); nasal, 24.4+ (24.8); greatest breadth of both nasals together, 9.8 (10.0); diastema, 16.0 (15.8); length of palate (lateral), 7.4 (7.S); width of palate between anterior premolars, 8.0 (7.6) ; width of palate between posterior molars, 10.4 (10.6); depth at middle of palate, 13 6 (14.0); maxillary toothrow (alveoli), 11.6 (12.0). Note. — Four specimens (Nos. 1 74-">:>>l — 1 ) collected at Textlananquila on Mount Iztaccihuatl, above Amecameca, and presented to the U. S. National .Museum by Professor Ferrari -Perez, arrived in Washington after this article was in type. They entirely confirm the identity of Romerolagus dinzi and R. " nelsoni " . —Gerrit S. Miller, Jr. * I ii view of Mr. Nelson's personal acquaintance with the region where the two types were collected I requested him to give me a brief accountof the geography. This he lias kindly done as follows: " The total area occupied by Rnmerolagus appears to be limited to the middle slopes of the volcanos Popocatepetl and Iztaccihuatl. The boundary line between the states of Mexico and Puebla runs along the crests of these mountains in a northerly and southerly course, thus placing the western slopes in Mexico and the eastern ones in Puebla. These mountains are joined by a broad ridge about 12,000 feet high, at each end of which rise the high peaks. The middle slopes, where Roirn rolagus lives, are continuous around both mountains and lie in an area 20 miles long, in a north and south direction, and ten miles broad, from east to west. Romerolagus lives permanently only where it has the shelter of the coarse saccaton grass and the areas occupied by this plant are broken by hot slopes on which it can not maintain itself. These breaks are not wide and can be readily crossed by small mammals. The extremely limited area which Romerolagus occupies, and the absolute identity of climatic and other physical conditions within it, appear to preclude the possibility of the existence of more than one local form." Vol. XXIV, pp. 231-232 November 28, 1911 PROCEEDINGS OF THE BIOLOGICAL- SOCIETY OF WASHINGTON A NEW ROE-DEER FROM CHINA. BY GERRIT S. MILLER, JR. [By permission of the Secretary of the Smithsonian Institution.] While exploring portions of western China during the summer of 1909, Mr. Arthur de C. Sowerby procured skins and skulls of ('■i/in ol us in the provinces of Shensi and Kansu.* Those from Shensi are essentially topotypes of C. bedfordi Thomas; the single Kansu specimen represents a species readily distin- guishable from the members of the genus hitherto described. Capreolus melanotis sp. nov. Type. — Immature female (permanent premolars in place, but crown of in ' not yet in line with cutting surface of other teeth) No. 155,220 U. S. National Museum. Collected thirty miles east of Ching-yang-fu, Kansu, China, August 14, 1909, by Arthur de C. Sowerby. Original number, 247. Diagnosis. — Resembling Capreolus bedfordi in essential characters, including the hypsodont form of the teeth, but color of summer pelage more reddish, and outer surface of ear mostly clear Mack in striking con- trast with surrounding parts. Color. — Qpperparts a clear tawny, brighter and richer than that of Ridgway, becoming duller on sides and fading on underpartsto ochraceous- Imff or cream-buff. < >n sides of body the change is gradual, but on sides of neck it is so abrupt as almost to produce a contrasted dorsal hand. Lower half of underside of neck variegated by faint grayish white blotches about the size of finger-prints. Markings on muzzle and chin normal, apparently rather better defined than in C. bedfordi. Inner surface of ear light cream-buff; outer surface black, except at extreme base, which is like sides of crown, and along outer border where, there is a wash of cream-buff extending upward as a band about 15 mm. wide to within 20 mm. of tip. * Descriptions of four now mammals obtained during the same expedition have already been published U'roc. Biol. Soc. Washington, XXIV, pp. 53-56. February 24, 1911). U— Pnoc. Biol. Soc. Wash., Vol. XXIV, 1911. (231) 232 Millet — A New Roe-Deer from. China. Skull and teeth. — Both skull and teeth appear to be more robust than in C. bedfordi, but the material is insufficient to show whether this char- acter is constant. As compared with a male C. bedfordi of same age and a female slightly younger, the type of C. melanotis shows the following peculiarities: lachrymal vacuities decidedly larger; lachrymal pit barely indicated; mesopterygoid fossa extending forward as an acute angle to level of lateral palatine emarginations. Measurements. — Type: head and body, 1110; hind foot, 350 (310); ear from crown, 140; condylobasal length of skull, 186 (186);* zygomatic breadth, 86.4 (80); least orbital breadth, 53.6 (51.4); breadth of brain- case, 61.4 (57.6); depth of brainease at middle, 57.4 (53); nasal, 59.6 (61.6); greatest breadth of both nasals together, 19.8 (19); mandible, 156 (152); maxillary toothrow, 69 (67); mandibular toothrow, 77± (73.6); crown of m1, 14.6 x 13.8 (14.0 x 12.8); crown of m2, 15.6 x 14.0 (14.6 x 13.2); crown of m2, 15.2 x 10.0 (14.0 x 9.0). Remarks. — Together with Capreolus bedfordi this animal differs from C. capreolus in the conspicuously hypsodont character of the teeth. Though this is shown by both molars and premolars it is most noticeable in the latter. Height of inner border of crown of middle upper premolar in type, 12.8 mm., diameter of crown in line of toothrow, 9.2. The same measurements in a male C. bedfordi, 12.4 and 10. In a female C. capre- olus from Bavaria they are 7.2 and 8.8, and in a male from Italy, 7.4 and 8.2. * Measurements in parenthesis are those of a male C. bedfordi of approximately the same age. Vol. XXIV, pp. 233-236 November 28, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON DESCRIPTIONS OF THREE NEW BIRDS FROM CANADA. BY J. H. RILEY. [Published l>y permission of the Secretary of the Smithsonian Institution.] While working up the collection of birds taken on the 1911 Alpine Club of Canada's expedition to Jasper Park, Yellowhead Pass, and the Mount Robson region, the following birds have been found to be apparently unnamed and are herewith de- scribed . Lagopus lagopus ungavus subsp. nov. Type, U. S. National Museum, No. 101,068, adult male, Ft. Chimo, Ungava, July 22, 1884. Collected by L. M. Turner (original No. 5S23). Like Lagopus lagopus albus, but with a heavier bill. Measurement of type: Wing, 205; tail, 128 ; culmen, 21.5; depth of bill at base, 15. Remarks. — Leaving out of consideration Lagopus lagopus alexandrx, which is recognized as a perfectly distinct form, I have gone over the same ground as Mr. A. H. Clark,* with some additional material, and reached the following conclusions: That the name Lagopus lagopus lagopus should be restricted to the small-billed European bird and that the bird from the west side of Hudson Bay should be known as Lagopus lagopus albus.y Specimens from the mountains of western Alberta and eastern British Columbia are slightly smaller than birds from Hudson Bay and northern Alaska. There appear to be no color differences except in Lagopus lagopus alexandrse, which is darker. In my opinion, the willow ptarmigans of the North American mainland can be separated into at least three recognizable races, as follows : 1. Lagopus lagopus alexandrn Grinnell, from the southwest coast of Alaska. 2. Lagopus lagopus albus Gmelin, from the west side of Hudson Bay, west through northern Alaska to eastern Siberia. :;. Lagopus lagopus ungavus subsp. nov., from L'ngava and probably the eastern shore of Hudson Bay south. •Proc. r. s. Nat. Mus., 38, No. 1727, April 30, 1910,51. t [Tetrao] albus Gmelin, Sys. Nat., I, pt. 2, 750. ■15— Pkoc. Biol. Soc. Wash., Vol. XXIV, 1911. (233) 234 Riley — Descriptions of Three New Birds from Canada. The average measurements of a series of willow ptarmigans taken in tlic breeding season, except Lagopus lagopus alexandrse, from the various parts of its range are as follows: cul- Depth men of Bill One male, Norway Ten males, Ungava Five males, west side Hudson Bay . . . Eight males, mountains Alberta and B. ('. Six males, northern Alaska One male, Lake Iliamna, Alaska . . . . One male, Petropaulski, Kamchatka . . Ten females, Ungava Three females, west side Hudson Bay . . Seven females, nits. Alberta and B. C. Melospiza melodia inexspectata subsp. nov. Type, U. S. National Museum, No. 222,829, adult female, three miles east of Moose Lake, British Columbia, August 21, 1911. Collected by J. H. Riley (original number 2268). Similar to Melospiza melodia rufina, but the browns of the upper parts lacking the reddish tinge, thus giving to the hack a gray cast; below not so heavily streaked; averaging smaller. Measurement of type: Wing, 65; tail, 63.5; culmen, 12; tarsus, 22 ; middle toe, 15.5. Remarks. — This race is founded on four specimens, as follows : the type and a young male still in the juvenile plumage from the type locality; one adult female from Henry House, Alberta, September 14, 1911 ; and an adult female from near Telegraph ('reek (25 miles east, near Buckley Lake), British Columbia, July 26, 1910, the latter specimen taken by E. A. Preble and in the Biological Survey Collection. Average of the three adult females: Wing, 63.8; tail, DO.:;; culmen, 12; tarsus, 21.:!; middle toe, 15.7. Average of three adult females of Melospiza melodia rufina from the Sitkan District taken at about the same season of the year: Wing;<>7; tail, 63.8; culmen, 12.8; tarsus, 22.8; noddle toe, 16. One of the interesting results of our trip was the discovery of this song sparrow, which instead of resembling Melospiza melodia merrilli, as one would naturally suppose, is more nearly related to the Pacific Coast, bird. Its range is probably the interior mountain valleys from northern British Columbia at least, south to Henry House, Alberta, and in migration probably further. The bird taken at Henry House may have been a migrant, hut the type was probably a breeding bird as the young male taken at the same locality seems to indicate. Passerella iliaca altivagans subsp. nov. Type, IT. S. National Museum, No. 222,832, nearly adult male, Moose Branch of the Smoky Liver, Alberta (about 7000 feet altitude), July 31, 1911. Collected by J. H. Riley (original number, 2175). Riley — Descriptions of Three New Birds from Canada. 235 Similar to Passerella iliaca schistacea, but middle of the back mars brown instead of mouse gray ; wings and tail with more red in the brown (near burnt umber). Measurement of type: Wing, 81.5; tail, 7(i; cul- nuMi, ll.o; tarsus, 21; middle toe, 14.5. Remarks. — This race is founded on two slightly immature birds from the type loeality; an immature male from Moose Pass, British Columbia; and an immature specimen from Thudade Lake (source of Finlay River), British Columbia. An adult male from Columbia Falls, Montana, Sep- tember lM. L896; an adult male and female, Ft. Klamath, Oregon, April it and 28, 1883; and an adult female, Ft. Crook, California, April 12, L860, all in the collection of the U. S. National Museum, are apparently migrants of this form. They differ from the type in having the upper parts more rusty, but from typical Passerella iliaca schistacea, to which they have been referred, they are quite distinct, having the mouse gray of the upper parts of that race replaced by broccoli brown and burnt umber and the spotting of the under parts prout's brown, not sepia. Ap- parently there is no difference in size. Vol. XXIV, pp. 237-240 November 28, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON THREE NEW MAMMALS FROM CENTRAL AND SOUTH AMERICA. BY E. A. GOLDMAN. While engaged in the identification of a collection of mam- mals for the Biological Survey the following unnamed species have been detected among the undetermined material in the l". S. National Museum. For the privilege of describing them, and for other courtesies, I am indebted to Mr. Gerrit S. Miller, Jr., Curator, and Mr. N. Hollister, Assistant Curator of the Division of Mammals. Marmosa ruatanica sp. nov. Type from Ruatan Island, off the north coast of Honduras. No. 7785, adult, I'. S. National Museum, collected by J. Akhurst. General characters. — Size large — about as in chapmani, of Trinidad; general color similar, but duller; black orbital markings extending for- ward to nose; molariform teeth smaller. Color. — Type (in worn pelage): General color above between cinnamon and isabella color, becoming smoky-brown on top of head, and lighter again on broad line down middle of face; blackish orbital markings extending to nose; under parts cream color, tinged with pale yellow; ears dark brownish ; feet soiled white; tail dusky on basal half, becoming somewhat lighter terminally. Skull. — Similar in general to that of chapmani, the rostrum rather broad ami heavy as in that species; supraorbital ridges less developed; nasals similar in shape, but less prolonged posteriorly; molariform teeth smaller. Measun merits. — Dry skin : Total length, 338; tail, 188; hind foot, 24.5. Skull: Occipito-nasal length, 40.5; zygomatic breadth, 22.7; nasals, 17.7 x 4.9; interorbital breadth, 5.8; palatal length, 21; three anterior molariform teeth, 6. Remarks. — In general characters this insular species agrees more closely with chapmani of Trinidad, than with any mainland form available for 46— Proc. Biol. Soc. Wash., Vol. XXIV, 1911. (237) 23S Goldman — New Mammals from Central and South America. direct comparison. It is well marked externally by the greater extension of the black orbital areas. The type, and only known specimen is in a much worn pelage, and would doubtless lie brighter colored in the fresh coat. Marmosa zeledoni * sp. nov. Type from Navarro, Costa Rica. No. ifftf> c? adult, U. S. National Museum, collected by Juan Cooper, for Jose C. Zeledon, March 1, 1878. General characters.- — Allied to mexicana, but color darker, more rufes- cent, the line of demarcation between whitish under parts and huffy sides very sharp; interparietal narrower. Somewhat similar to mitis, but color decidedly darker, and skull quite different. Color. — Upper parts cinnamon-rufous, becoming lighter on middle of face, and paling to ochraceous huff on cheeks, sides of neck, and flanks, the latter color encroaching on under parts along sides of belly where the fur is basally plumbeous; throat, chest, median line of belly, inguinal region, and inner sides of limbs yellowish white, the hairs whitish to roots; ears brown; orbital area black; fore feet brownish; hind feet soiled white; tail light brownish. Skull. — Similar to that of mexicana, but interparietal narrower, less extended laterally along lambdoid crest. Somewhat like that of mitis, but braincase larger; interorbital region broader; supraorbital and temporal ridges more widely separated. Measurements. — Type (dry skin): Total length, 345; tail, 168; hind foot, 23. Skull (type): Greatest length, 36; zygomatic breadth, 19.2; nasals, 16.2 x 4.4; interorbital breadth, 6.5; palatal length, lS.-_'; three anterior molariform teeth, 5.2. Remarks. — This form is probably most closely allied to mexicana, but is readily distinguished by richer color. The supraorbital and temporal ridges in mexicana and zeledoni are widely separated, and especially in the latter, extend backward across the parietals in nearly parallel lines. Specimens examined. — Five, from localities as follows: Costa Rica, Na- varro (type locality), 2; Nicaragua, Escondido River, 3. Proechimys steereit sp. nov. Type from Rio Purus, a southern tributary of the Amazon, in north- western Brazil. No. 105,535, cT adult, U. S. National Museum, collected by J. B. Steere, in 1901. General characters. — Size medium ; color dark and rich ; ankles dusky all around; tail rather short, thinly haired; skull without temporal ridges. Closely resembling semispinosus in color, and in arrangement and stiffness of dorsal spines, but skull more like that of trinitaiis. Color. — Upper parts between orange rufous and ferruginous, lined with black, the color richest and darkest along median line of back, fading to huffy grayish or brownish on cheeks, flanks, and outer sides of limbs; under parts and inner sides of limbs pure white, abruptly interrupted at * Named for Mr. Jos6 C. Zeledon, the well-known Costa Rican naturalist. + Named for Dr. J. B. Steere, the naturalist and traveler. Goldman — New Mammals from Central and South America. 239 ankles by a dusky band below which the white is continued along inner side of metatarsus to toes; fore feet and outer side of metatarsus pale brownish; nose and ears blackish, the latter nearly naked as usual in the group; tail brownish above, grayish below. Skull. — Similar in general form to that of trinitatis, hut rostrum broader above, less compressed over roots of incisors; supraorbital ridges heavier; audital bullae decidedly larger, more inflated; temporal ridges absent as in trinitatis; dentition about the same, except that there are four transverse furrows or enamel islands in the lower premolars of rive skulls examined, instead of three as in the three skulls of trinitatis availa- ble for comparison. Measurements. — Type (dry skin): Total length, 347; tail vertebrae, 122; hind foot, 49. Skull: Greatest length, 53.8; eondylobasal length, 48; zygomatic breadth, 26; length of nasals, 19.3; interorbital breadth, 11.9; length of palatal bridge, 10.5; alveolar length of upper molariform toothrow, S.5. Remarks. — This spiny rat is a member of the same group as trinitatis and mincae, hut differs in important respects from both and 1 am unable to associate it very closely with any of the other species described. In dentition it agrees with trinitatis, the type species of the genus, in the possession of three transverse furrows, or elongated enamel islands in most of the molariform teeth. This character is somewhat variable, how- ever, in trinitatis and other species, the number in the lower molars being sometimes reduced to two. The furrows may also become divided through wear and the normal number thus obscured through the multiplication of enamel islands. Specimens examined. — Five, all from the type locality. Vol. XXIV, pp. 241-242 November 28, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON TWO NEW SHREWS FROM KASHMIR. BY GERRIT S. MILLER, JR. [By permission of the Secretary of the Smithsonian Institution.] In a collection of mammals from Kashmir recently presented to the United States National Museum by Dr. W. L. Abbott are specimens of two shrews which have not hitherto been described. Crocidura pullata sp. now Typt .— Adultmale (skin and skull ) No. 173,213 V. S. National Museum. Collected at Kotihar, Kashmir (altitude 7000 feet) October 9, 1910, by by Dr. W". L. Abbott. Original number, 7421. Diagnosis. — General characters much as in Crocidura fumigata as de- scribed by de Filippi and figured by Dobson, but size greater (head and body 80-90 instead of about 60), third upper unicuspid about equal to second, and basal lobes of i1 low and broad as in C. russula. Color.— Upperparts a uniform, dark hair-brown with a slaty tinge and sometimes a wash of wood-brown, the hairs showing rather noticeable silvery reflections in certain lights; basal portion of hairs slate-color; underparts between smoke-gray and ecru-drab, the slaty under color showing through irregularlj'. Skull and teeth.— The skull is about the same length as that of Crocidura russula, but the general structure is decidedlj more robust, and the out- line of braincase is noticeably longer than broad; distance from front of glenoid surface to back of condyle equal to greatest breadth of braincase instead of decidedly less as in C. russula; mandible more heavily built, both in ramus and postdental region. The teeth differ from those of Crocidura russula in relatively greater size of first unicuspid, less development of minute cusp at inner base of large upper incisor, and greater crown area of molars. Measurements.— Type: head and body, 90; tail, 43; hind foot (dry) 15 (14). Average and extremes of nine adults from the type locality: head and body, 86.1 (82-90); tail, 42.6 (40-47). Skull of type: condylobasal length, 19.8; zygomatic breadth, 6.2; breadth of braincase, 9.0; condylo- 47— Proc. Biol. Soc. Wash., Vol. XXIV, 1911. (241) 242 Miller — Two New Shrews from Kashmir. glenoid length, 9.0; depth of hraincase, 4.8; mandible, 10.2; maxillary tooth row( entire) 8.8; mandibular tooth row (entire), 8.2, Specimens examined. — Sixty-three, from the following localities in Kashmir: Kotihar Valley , 27 ; Sogam, 2; Nowboog Valley, 33; AVangat Valley, 1. Sorex planiceps sp. now Type.— Adult male (skin and skull) No. 173,915, United States National Museum. Collected at Dachin, Khistwar, Kashmir (altitude, 9000 feet), .May 30, 1911, by Dr. AY. L. Abbott, Original number, 7714. Diagnosis. — Color and general external appearance as in Sorex minutus, but size less diminutive; skull as in the largest specimens of S. minutus, but mst rum, palate and interorbital region wider, and braincase more flattened (ratio of depth to width about 50); teeth relatively larger than in S. minutus though of the same general form, except that basal cusp of anterior upper incisor is lower and more robust, somewhat as in S. araneus. Measurements. — Type: bead and body, 69; tail, 44; hind foot, 11; condylobasal length of skull, 16.8; zygomatic breadth, 4.8; lachrymal breadth, 3.2; breadth of braincase, 8.0; depth of braincase at middle, 4.0; mandible, S.4; maxillary toothrow, 7.0; mandibular toothrow, 6.8. Specimens examined. — Seven, from the following localities in Kashmir: Dachin, Khistwar, 3; Sind A'alley, 1; Gadasar, 1; Sogam, 1; Nowboog Valley, 1. Vol. XXIV, pp. 243-250 December 23, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON THE AMERICAN SPECIES OF FAGONIA. The representatives of the genus Fagonia are confined chiefly to two parts of the earth, the more arid regions bordering the Mediterranean in Europe, Asia, and Africa, and the deserts aboufthe Gulf of California in North America. Two species occur on the western coast of South America in Chile and Peru, while another has only recently been described from the interior of Mexico. The genus is composed of remarkably uniform species that are similar in general appearance and differ only in minor characters such as pubescence, shape of leaflets, and form of stipules. In the Kew Index but two names are recog- nized as valid. All the Old World species, of which a consid- erable number have been described, alongwith Fagonia chilensis and F. californica, are referred to the type species, F. cretica L. After an inspection of the Old World material in the National Herbarium, consisting of many more sheets than there are of American collections, the writer is inclined to believe that there are several species in the Eastern Hemisphere although possibly not so many as in America. Our American forms certainly are more diversified than those of Europe and Africa. The latter, for example, exhibit no such distinct types as Fagonia califor- nica, F. palmeri, and F. scoparia. The genus is ably treated by Dr. P. A. Rydberg in the part of the North American Flora dealing with the Zygophyllaceae, five species being recognized for North America. Before the publication of that monograph only two species had been de- scribed from the region, F. californica and F. palmeri. Miss A. M. Vail in 1895 named a subspecies of Fagonia californica from the Southwest. ■IS— Proc. Biol. Soc. Wash., Vol. XXIV, 1911. (243) A BY PAUL ('. STANDLEY. [By permission of the Secretary of the Smithsonian Institution.] 244 Standley — The American Species of Fagonia, Recently the writer had occasion to determine several collec- tions of this group. One of them seemed unlike any of the species described by Dr. Rydberg, and upon examination of other material in the herbarium several forms were found which seem to merit description. It was discovered, besides, that Mr. T. S. Brandegee has described a very remarkable species from Coahuila within the present year. It seems worth while to prepare an account of these newest discoveries and to coordinate them with previously described species by a key. Only two species have been described from South America and since they are closely related to ours and we have herbarium material of both, tbey may be included as well. The authors of the Kew Index considered that both Fagonia californica and F. chilensis were synonymous with F. cretica of the Old World. The most casual comparison of our plants with Fagonia cretica compels a different conclusion. That species has much larger flowers than any of the American ones, its leaflets are larger, the pubescence different, the beak of the fruit is conspicuously thickened at the base, a condition not existing in the plants of the western world, while F. cretica is much stouter than our plants that are nearest related to it. Fagonia chilensis is almost like F. californica but the differences are such that the two can be distinguished, and because of their different ranges it seems as well to hold them apart. The writer is under obligations to Dr. Wm. Trelease who courteously loaned him all the American material of the genus in the herbarium of the Missouri Botanical Garden. KEY TO THE SPECIES. Leaves wanting; stipules triangular-subulate, not more than 1 nun. long; branches erect; sepals persistent 1. F. scoparia. Leaves present ; stipules subulate, averaging much more than 1mm. long; branches spreading; sepals caducous. Leaflets 5 2. F. palmeri. Leaflets 3. Ovaries glabrous 11. F. laeris. Ovaries and mature fruit pubescent. Pedicels twice as long as the fruit or more, glabrous or slightly scaberulous . . . 12. F. longipes. Pedicels not longer than the fruit, mostly shorter, or when longer glandular. Leaflets glabrous. Standley — The American Species of Fagonia. 245 Stems densely glandular 3. F. viscosa. Stems not glandular, sparingly scaberulous on the angles or glabrous. Beak of the fruit 1 to L.5 mm. long; steins usually scaberulous; North American ... S). F. californica. Beak 2 to 3 mm. long; stems glabrous; South Amer- ican 10. F. chilensis. Leaflets pubescent, often glandular. Plants not glandular; South American 7. F. aspera. Plants more or less glandular; North American. Pedicels shorter than the fruit. Stipules long and stout; leaflets 8 to 13 mm. long; stems densely glandular . . . 4. F. pachyacantha. Stipules short, slender; leaflets 3 to 10 mm. long; stems sparingly or often scarcely at all glandular. 8. F. bar clay ana. Pedicels longer than the fruit. Leaflets linear ; stipules 4 to 5 mm. long, spreading; stems sparingly soft-villous 5. F. insularis. Leaflets lanceolate or linear-oblong; stipules 2 mm. long, reflexed ; stems scantily scaberulous . 6. F. rosei. 1. Fagonia scoparia Brandegee, Univ. Calif. Publ. Bot. 4 : 181. 1911. Type locality, "On Cerro del Macho, Coahuila." Perennial with numerous very slender, broom-like, erect, glabrous, striate, not angled branches; leaves unknown, apparently wanting; stip- ules 1 mm. long or less, triangular-subulate, spinescent tipped; flowers terminating the branches or on slender pedicels mostly about 15 mm. long; sepals persistent, lanceolate or lance-ovate, acute or somewhat acuminate, 4 mm. long or less; petals rose purple, about 5 mm. long, acute or acuminate; fruit 6 mm. long, hirsutulous, the beak not seen. Specimens examined: Coahuila: Cerro del Macho, June, 1910, Purpus 4495, type collection. This is a very remarkable species and comes from a locality far distant from the usual range of the genus in North America. Mr. Brandegee reports that although some of the specimens bear young branches there are no leaves on any of them. The habit of the plant is different from that of our other species, the flowers are smaller, the petals have scarcely any claws, and the sepals are persistent instead of caducous. Doctor Rydberg's diagnosis of the genus in the North American Flora must be changed to include this plant, especially that part dealing witb the leaves and the persistence of the sepals. Perhaps when more complete material is secured the plant may prove to be the type of a new genus. i'. Fagonia palmeri Vasey & Rose, Contr. Nat. Herb. 1 : 82. 1S90. Type locality, "Santa Rosalia," Lower California. A stout, rigid, much branched plant, growing in dense clumps 30 to 45 cm. high ; stems yellowish, angled and striate, densely glandular, some- 246 Standlcy — The American Specie* of Fagonia. times woody at the base; petioles 4 to 10 mm. long; stipules stout, rigid, spinescent, densely glandular, one-third to one-half as long as the petioles; leaflets five, nearly linear, the principal ones of each leaf 3 to 10 mm. long, spinescent tipped; pedicels :\ to 5 nun. long, equaling or shorter than the fruit ; sepals oblong or oblong-lanceolate, acute, 4 or 5 mm. long; petals pinkish, 6 to 8 mm. long; fruit minutely pubescent and glandular, the beak about 3 mm. long. Specimens examined: Lower California : Santa Rosalia, 1889, Palmer I'll'.), type; San Francisquito Bay, April (.», 1911, Rose 16,729; Tiburon Island. April 11, 1911, Rose 1.6,779. The species is one of the most distinct of the entire genus. No other is known to have more than three leaflets. Its range, so far as known, does not extend beyond Tiburon Island in the Gulf of California, and the middle part of the east coast of Lower California. 3. Fagonia viscosa Rydb. N. Amer. Fl. 25 2 : 104. 1910. Fagonia californica glutinosa Vail, Bull. Torrey Club 22 : 220. 1895. Not F. glutinosa Delile, 1813. Type locality, " Sonora, Mexico." Type collected by C. G. Pringle in ISS4. Stems stout, densely covered with large, yellowish glands, glabrate in age; petioles stout, 4 to 10 nun. long, glabrous; stipules thick, rigid, glabrous or sparingly glandular, much shorter than or even equaling the petioles; leaflets 3, 8 to 15 mm. long, 3 to S mm. wide, the terminal one rhombic-obovate to oblanceolate, the lateral ones lanceolate and oblique, spinulose tipped, glabrous, thick and fleshy; pedicels stout, ."> to 5 mm. long, glandular; sepals oblong or oblong-lanceolate, acute; petals purple, 6 to 8 nun. long; peduncles 3 to 5 mm. long, shorter than the fruit; this strigillose and glandular, the beak 1.5 to 2 mm. long. Specimens examined: California: Southwestern part of the Colorado Desert, San Diego County, April. 1887, Orcutt; Signal Mountain, April 2, 1903, Abrams, .".158. Lower California : Santa Rosalia, 1889, Palmer 180; Los Angeles Bay, 1887, Palmer 546a; Signal Mountain, May 6, 1894, L. Schoenfeldt 2950. 4. Fagonia pachyacantha Rydb. X. Amer. Fl. 25 2: 105. 1910. Type locality, " Lower California." Type in the New York Botanical Garden, collected by Leon Diquet. An undershrub, 3 to 6 dm. high ; branches yellowish or straw-colored, viscid, with glandular hairs, striate and somewhat angled ; stipules stout, subulate, 6 to 10 mm. long, spreading or somewhat reflexed, longer than the petioles; petioles 5 to 8 mm. long, spreading; leaflets 3, linear, 8 to 15 mm. long, about 1 mm. wide, thick, glandular; peduncles 3 to 5 mm. long; sepals ovate-lanceolate, 3 mm. long, with very short spinulose tips; petals rose-purple, (i to 8 mm. long; blades obovate-spatulate ; fruit 5 mm. long, reticulate, pubescent, the hairs with thickened bases; beak about 2 mm. long. SStandley — The American Species of Fagonia. 247 The writer lias seen no materia] of this species. The description is a transcript of the original one. 5. Fagonia insularis Standley, sp. nov. Stems slender, nmeli branched, 30 cm. long or less, sparingly soft-villons with short, white hairs; petioles 5 to 12 mm. long, short-villons and slightly viscid ; stipules slender, spinescent, 4 to 5 mm. long, short-villous and viscid, spreading; leaflets 3, linear, spinescent tipped, 1 to 9 mm. long, pubescent and glandular; pedicels about 8 mm. long, reflexed in age, hearing numerous short, gland-tipped hairs; sepals 2 mm. long, oblong, acuminate; petals 5 mm. long; fruit finely pubescent and spar- ingly glandular, :'> to 4 mm. high, with a slender beak 2.5 to 3 mm. long. Type in the U. S. National Herbarium, No. 14,216, collected on Carmen Island. Lower California, in November, 1890, by Dr. Edward Palmer ( No. 830). Also collected on the same island by Doctor Palmer 20 years earlier (1870), No. 13. < )f the previously described species this appears to be nearest Fagonia pachyacanlha, but is a more slender plant with smaller leaflets, slender, shorter stipules, and has a longer beak on the smaller fruit. 6. Fagonia rosei Standley, sp. now Perennial with slender, scaberulous and somewhat glandular sterns about 30 cm. long; petioles 4 to 7 mm. long, glandular; stipules stout but short, 1.5 to 3 mm. long, mostly 2 mm., spinescent, prominently reflexed, glandular; leaflets 3, lanceolate or linear-oblong, glandular, spinescent tipped, small, 2 to ?> mm. long; pedicels usually slightly longer than the fruit, deflexed, glandular; sepals lanceolate or lance-oblong, rarely more than 2 mm. long, acute; petals pale purplish, 7 mm. long; fruit 4 mm. high, minutely pubescent and glandular, with a slender beak •_' to 2.5 mm. long. Type in the U. S. National Herbarium, No. (>19,744, collected on Tiburon Island in the Gulf of California, April 11, 1911, by J. N. Rose (No. 16,779a). This stands nearest Fagonia insularis but is a greener plant with broader leaflets and different pubescence and stipules. 7. Fagonia aspera C. Cay, Hist. Chile Bot. 1 : 470. 1S45. Type locality, Chile. Low, rather stoutly branched plant with angled, scaberulous or puber- nlent branches; stipules stout, L' to 3 mm. long, spinescent, short-villons; petioles stout, about equaling the leaflets, these obovate to oblong-lanceo- late, nearly obtuse, abundantly pubescent; pedicels 2.5 to 5 mm. long, pubescent; flowers not seen ; fruit :! to 4 mm. high, conspicuously strigose with rather long hairs, noticeably tapering at the apex, with a slender beak 1.5 to 2 mm. long. The description is drawn from poor specimens collected in Peru by the Wilkes Exploring Expedition. This is the only collection of the species seen and seems to agree well with the original description. 248 Standley — T7ie American Species of Fagonia. 8. Fagonia barclayana (Benth.) Rydb. N. Amer. Fl. 25 2: 104. 1910. Fagonia californica barclayana Benth. Bot. Voy. Sulph. 10. 1844. Type locality, "Bay of Magdalena," Lower California. A dichotomously branched undershrub; stem with rather slender green- ish branches, 3 to 7 dm. long, finely villous-pubernlent ; stipules subulate, about 5 mm. long, spinulose, reflexed-spreading ; petioles 3 to 5 mm. long; leaflets ,">, lanceolate, 8 to 20 mm. long, 2 to 5 mm. broad, finely pubescent, spinulose-tipped ; peduncles 2 to 5 mm. long; sepals narrowly lanceolate, 3 mm. long, spinulose tipped; petals rose-purple, about 5 mm. long; blades ovate-spatulate, acutish ; fruit 4 to •"> mm. long, finely pubes- cent, slightly reticulate; beak about 1 mm. long. Here apparently belong the following specimens from Lower California: San Jose del Cabo, < >ctober 17, 1890, Brandegee SI ; Lagoon Head, March, 1889, Palmer 827; Agua Verde, 1911, Rose 16,604. Pemarks upon the species may be found under Fagonia californica. The description is that of Doctor Rydberg in the North American Flora. 1). Fagonia californica Benth. Bot. Voy. Sulph. 10. 1S44. Fagonia californica hindsiana Benth. loc. cit. Type locality, " Bay of Magdalena," Lower California. Densely branched, usually about 30 cm. high, with slender, angled stems scaberulous along the angles; petioles 4 to '.) mm. long, glabrous or scaberulous; stipules short, slender, spinescent, half as long as the petioles or shorter; leaflets 8, glabrous, narrowly lanceolate to ovate-lanceolate, acute, spinescent tipped, the lateral ones oblique, sometimes longer than the petioles and sometimes much shorter; pedicels 2 to •"> mm. long, usually much shorter than the fruit, deflexed in age; sepals lanceolate or ohlong-lanceolate, acute; petals purplish, •"> to s mm. long; fruit puberulent, densely so when young, often nearly glabrous in age except along the angles, the beak short, about 1 mm. long. Specimens examined : Utah: St. George, 1879, Palmer. Arizona: Gila City, Gila Mountains, March 1, 1894, Mearns 2820. California: Sierra Prieta near Fort Yuma, 1855, Schott; Tia Juana, May 15, 1903, Abrams 3500; canyon west of Borrego Spring, April 19, 1906, M. E. Jones; without locality, Mexican Boundary Survey; San Bernardino, 1880, S. B. Parish; Ogelby, San Diego County, March, 1901, A. F. Eby; Cargo Muchacho, September 20, 1890, Orcutt 2070; Coyote Canyon, altitude 1350 meters, April, 1902, H. M. Hall 2704. Lower California : Los Angeles Bay, 1887, Palmer 546 ; Valley of Palms", April 15, 1882, M. E. Jones 3691 ; Santa Rosalia, 1889, Palmer 196; San Bartolome Bay, March 14, 1911, Rose 16,235; Lagoon Head, March, 1889, Palmer 818. The species probably reaches the northwest corner of Sonora but I have seen no specimens from that State, unless the Schott specimen may be Sonoran rather than California!!. Slandley — Tlie American Species of Fagonia. 249 The specimen from Utah differs from the others in having very small stipules and Leaflets and remarkably long petioles. Probably it is a different species for it is from a region well removed from the usual range of Fagonia californica. As originally published the species consisted of two forms, a hindsiana, and |8 barclayana. The first, of course, is to be taken as the type. The other is /•'. barclayana( Benth. ) Rydb. The form hindsiana was described as being "glabra, stipulis brevissimis." The second form was described as " puberula, stipulis setaceo-spinescentibus petiolo paullo brevioribus." Personally the writer is inclined to believe that these two are. the same species. This opinion, however, may be altogether wrong. A few of the plants here listed under the species are nearly perfectly glabrous while others are conspicuously scaberulous. Judging from Bentham's meager descriptions alone it seems likely that these are the two forms he had before him. 10. Fagonia chilensis Hook. l*c Arn. in Hook. Bot. Misc. 3 : 165. 1833. Type locality, " Coquimbo," Chile. Type collected by Cuming, No. 907. Rather slender, 25 cm. high or less, with glabrous, angled branches; petioles equaling or shorter than the leaflets, glabrous; stipules stout, spreading or slightly refiexed, :'> to 4 mm. long, spinescent; leaflets obovate to linear-lanceolate, glabrous, acute or acuminate, spinescent tipped, 2 to S mm. long, the lateral ones oblique; pedicels 3 to 4 mm. long, refiexed in age, shorter than the fruit: sepals lanceolate or oblong- lanceolate, 3 to 4 mm. long, acute; petals rose purple, 7 or S mm. long^ fruit about 5 mm. high, scaberulous, often nearly glabrous in age, with a beak 2 to .". mm. long. Specimens examined: Chile: Atacama, 1890, Morong 1180; prov. Coquimbo, R. A. Philippi. 11. Fagonia laevis Standley, sp. now Low, densely branched perennial 20 to 40 cm. high; stems rather slender, green, glabrous, angled and striate; petioles 4 to 10 mm. long, glabrous; stipules slender, 1.5 to 2.5 mm. long, spinescent, spreading; leaflets 3, linear-lanceolate, 2 to s mm. long, glabrous, petidlulate, spines- cent tipped, the lateral ones somewhat oblique; pedicels glabrous, shorter than the fruit, derlexed in age; sepals oblong-lanceolate, 2 to 2.5mm. long, acute, spinescent tipped; petals rose purple, about 5 or 0 mm. long, narrow, long-clawed; ovaries and fruit glabrous, the latter :'>.."> mm. high, the slender beak 1.5 to 2 mm. long. Type in the U. S. National Herbarium, No. 855,582, collected near Yuma, Arizona, April 25, 1906, by Marcus E. Jones. Additional specimens examined: California: 10 miles west of Coachella, Riverside County, altitude 150 meters, April, 1905, H. M. Hall 5806. Lower California : Tia Juana, June 30, 1884, Orcutt. 250 Standley — Tlie American Species of Fagonia. From all our other species this is distinguished by the glabrous ovaries and the completely glabrous stems. Aside from these differences it is not far removed from Fagonia calif ornica. 12. Fagonia longipes .Standley, sp. now Stems very slender, glabrous, angled, about 30 cm. lung, abundantly branched; petioles 4 to 10 mm. long, stout, glabrous; stipules stout, 1.5 to 2 mm. long, slightly reflexed; leaflets 3, glabrous, linear-oblong or linear-lanceolate, the lateral ones oblique, :i to 10 mm. long, minutely spinulose tipped; pedicels 10 to 12 mm. long, slender, glabrous or ob- scurely scaberulous, deflexed in age; sepals lanceolate or oblong-lanceo- late, acute or acuminate; petals 7 to 9 mm. long, rose purple; fruit 4 mm. high, sparingly and finely pubescent and slightly glandular, with a beak 1.5 mm. long. Type in the XLS. National Herbarium, No. 14, 222, collected in Arizona in 1876 by Dr. Edward Palmer. No other data are given on the sheet in the National Herbarium but on one in the J. II. Redfield Herbarium in the collection of the .Missouri Botanical Garden, the label gives the locality as Bill Williams Fork, the date of collection as March II, and the collector's number as 58. The plant is related to Fagonia californ.ica but may be distinguished at a glance by the very long pedicels. The stems, too, are nearly glabrous, the stipules shorter, and the whole plant more slender. Vol. XXIV, pp. 251-252 December 23, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON DESCRIPTION OF A NEW MELOSPIZA FROM CALIFORNIA. BY HARRY C. OBERHOLSER. A scries of song sparrows from eastern California, in the Biological Survey Collection, United States National Museum, appear to represent an undescribed race. Since several of these specimens were obtained l>y Dr. A. K. Fisher, in Owen Valley, California, during the Death Valley Expedition of 1891,* this new form may appropriately be named Melospiza melodia fisherella subsp. nov. ('liar*. subsp. — Similar to Melospiza melodia heermanni Baird, but larger; upper surface paler, less rufescent; streaks on lower parts less blackish ( more brownish). Type. — Adult male. No. 203,507, U. S. Nat. Mus., Biological Survey Collection; Honey Lake, near Millford, California, June 18, 190t>; A. S. Bunnell. Description of type. — Upper surface somewhat rufescent hair brown; the crown with broad lateral stripes of burnt umber, and throughout with streaks of clove brown; the back and scapulars broadly streaked with black and burnt umber; the rump and upper tail-coverts somewhat streaked with dark brown ; tail sepia, edged externally with wood brown and dull russet; wing-quills grayish sepia, margined exteriorly with wood brown and dull russet; superior win ^-coverts mostly burnt umber, margined more or less with pale hair brown, the median and greater series and the tertials with terminal shaft markings of black or clove brown; superciliary stripe, auriculars, and sides of neck ash gray, the lirst mentioned paler than the others; postocular and malar stripes burnt umber; entire lower surface, including inferior wing-coverts, dull white, the sides, thinks, and crissum washed with brownish or buffy; breast, jugulum, sides, Hanks, and crissum streaked with burnt umber and clove brown. * Recorded by Dr. Fisher (North American Fauna No. 7, 1S93, p. 100) as Melospiza fasciata In < rmanni. 49— Proc. Biol. Sue. Wash., Vol. XXIV, 1911, (251) 252 Oberholsei — A New Melospiza from California. Measurements of type. — Wing, 66; tail 66; exposed culmen, 12.2; tarsus, 22.7 mm. Geographical distribution. — Eastern California, south to Owen Valley; western Nevada; and central southern Oregon. This new race differs from Melospiza melodia fallax Baird ( = Melospiza melodia montana Henshaw) in its darker upper parts, more blackish brown streaks of under surface, heavier bill, and shorter wing. Vol. XXIV, pp. 253-254 December 23, 1911 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A NEW JUMPING-MOUSE FROM NEW MEXICO. BY GERRIT S. MILLER, JR. [By permission of the Secretary of the Smithsonian Institution.] In 1858 a specimen of Zapus was collected by Dr. Anderson at Fort Burgwyn, New Mexico. It was entered in the catalogs of the U. S. National Museum on October 22 of the same year. Coues does not mention this skin in his monogra ph of the genus,* but Preble f refers it to Zap us princeps Allen. In both color and cranial characters, however, it differs widely from Zapus princeps and approaches the Californian Z. alleni Elliot. On examining the specimens of Zapus in the Biological Survey collection I find six representatives of the same species, collected in New Mexico since the publication of Preble's revision. As these are in better condition than the original specimen one of them has been taken as type of the long overlooked form. Zapus luteus sp. nov. Type.— Adult female (skin and skull) No. 133,601, V. S. National .Museum (Biological Survey Collection). Collected at Espauola, New Mexico, June _4, 1904, by M. Surber. Original number, L62. Characters. — Resembling Zapus trinotatus alleni (Elliot); ground color of upperparts brighter and more ochraceous I less buffy); dark dorsal area much less evident, usually, in adults, passing insensibly into color of sides, without the line of demarcation nearly always well defined in alh ni : skull smaller and more slender but without special peculiarities of form, the antorbital foramen, as in alleni, relatively larger than in princeps : teeth without evident peculiarities, but anterior surface of incisors darker in color. Measurements. — Type: head and body, 86; tail, 138; hind foot (dry), 33 ; condylobasal length of skull in type and in adult male ( No. 133,604) from type locality (both with much worn teeth), 21.2 and 21.0 (22.0 and •Monogr. X. Amer. Rodentia, pp. 461-479. 1877. t Xorth Amer. Fauna . Xo. 15, p. 23. August 8, 1899. 50-Proc. Biol. SOC. Wash., Vol.. XXIV, 1911. 254 Miller — A Neiv Jumping- Mouse from Nor Mexico. 22.(1) ;* basilar length, 16.6 and 16.4 ( 18.0 and 18.0); zygomatic breadth, 11.2 and 11.2 (12.2 and 12.6); breadth of braincase, 10.0 and 9.8 (10.8? and 10.4) ; depth of braincase at middle, 7.6 and 7.4 (8.0 and 7.8) ; nasal, 9.6and 9.2(9.6and 9.2); diastema, 5.2 and 5.2 (6.0 and 6.0); mandible, 12.2 and 12.2 (13.0 and 13.0); maxillary toothrow, 3.8 and 3.8 (4.0 and 4.0); mandibular toothrow, 3.4 and 3.6 (3.8 and 3.S). Specimens examined. — Seven, from the following localities in New Mexico: Cloudcroft, Otero County, 3; Espahola, Santa Fe County, 3; Fort Burgwyn, near Taos, Taos County, 1. * Measurements in parenthesis an' those of an adult female and male (both with moderately worn teeth) Z. trinotatus allehi from Donner, California (Nos. 100,469 and 100,244). INDEX New names are printed in heavy type Ablephararus boutoni ........ L8 keiensis ' Is Achirus lineatus 209 Vcicarpa 109, 120 sacchariflora 109 Acinoivyx jubatUS 226 venaticus 226 vena tor 226 Acosmetira pilosella *■_> Acritochjete 112 volkensii 112 .Egina citrea 86 /Eginidae -85 .Efdnodiscus actinodiscus 86 .Eyinopsis laurentii 86 /Eginura myosura 86 .Enuorea aequorea s^ /Equoreidse . . 83 Agarna caiinata 98 cumulus .Is Agastra mira 82 Agelaius floridanus 226 phreniceus 227 predatorius 227 Aglantha digitalis sr' Aglaura hemistoma 85 Aglaurinse 85 Aglauropsis agassizii 84 Algansea sallaei Allactaga longior Alloteropsis . . distachya . Alsberg, Cart L. cell activity . Alutera scripta The 213 . 54 120, 159 159 mechanics of ... . viii 74 Alycia 125 coarctata 125 distachya 125 Amalthaea uvifera 77 Amalthseini . . 77 Ameiurus 75 Amphogona apsteini s"> Anachyris 139 paspaloides 139 Anastrophus 130 Ancylopsetta 211 Andropogon insulare 109 Anodonta grandis 68 Anthaenantia 109 villosa 109 Anthbmedusae 77 Anthornis incoronata 23 Antrostomus otiosus 188 Apus petrensis 195 Arvicola agrestis 39 fuscus 39 selysii 39 Aspidogaster conchicola 67 Atolla wyvillei 9 Atollidse 9 Atorella subglobosa 10 Atorellidse Aulaxanthus ciliatus rufus . . . . Aulaxia Aulaxis Aurellia aurita Aurelliidae Aurelliinse Aurosa furcata Axonopiis aureus 130, bijugus capillaris chrysoblepharis compressus 129, deludens dissitiflorus furcatus laxiflorus poiophyllus rosei in 109 109 109 109 109 11 10 11 11 129 135 136 133 134 132 134 135 132 133 133 132 B Balanosphyra 34 aculeata 35 albeola 35 angustifrons 3.") l)airdi 35 flavigula 35 formicivora 35 striatipectus 35 xantholarynx 35 Hangs. Outram. A new bell-bird from Auckland Island 23-24 Two new birds from the island of Molokai ■ ■ 29-30 A new fantail from the Chat- ham Islands 41-42 Descriptions of new American birds 187-190 ■ A new swift from Palestine 195-196 Batchelder, C. F. A new name for Sorex macrurus Batchelder .... 97 Barbour, Thos. New lizards and a new toad from the Dutch East Indies, with notes on other species 15-22 Notes on some birds from the island of Grenada 57-60 A new race of chameleons from British East Africa . 219-220 Bartsch, P. The fauna of Lower Cali- fornia ... ix A study of distribution based upon the" family Pyramidellidae . x Bean, Barton A., and Weed, Alfred C. Notes on the coloration of fishes 69-76 Recent additions to the fish fauna of the District of Colum- bia 171-174 Berchtoldia 156 51— Proc. Biol. Soc. Wash., Vol. XXIV, 1911. (255) 256 The Biological Society of Washington. Berchtoldia bromoides 15G holciformis 156 Bifaria bifaria 123 Bishopp, P. < '. Some new North Amer- ican Ixodidse with noteson other species 197-208 Blacieus flaviventris 58 Blackfordia manhattensis 82 Blarina brevicauda 99 Blnttia 159 eckloniana 159 Boleosoma longimanus 174 Bos bubalus 191 Botrynema brucei 85 Bougainvillia bougainvillii - • . . so macloviana 80 Bougainvilliini so Brachiaria 126 eruciformis 126 Brosme brosme 212 Bubalus caffer 191 Bul'o cavator '21 Busck, August. Recent explorations in Panama ix Bythotiara murrayi so Bythotiarini so c Cabrera L29 chrysoblepharis ......... 129 Callitiara 80 Calycopsis typa 80 Cannota dodecantha si Cannotidse 81 Cannotinse 81 Capreolus melanotis 231 Capria sturdzii 9 Cary. Merritt. A new pocket mouse from Wyoming 61-62 Carybdea marsupialis 8 Carybdeida S ('aryl)'leidie 8 Cassiopea andromeda . . 11 • lassiopeidae 11 Castor moschatus 1 :; zibethicus 13 Catostylidse 11 < latostylus mosaicus 11 Cenchrus inflexus 117 marginalis 117 Centurus brewsteri 32 chrysogenys 32 elegans .... 32 flavinuchus 32 Ceresia 138 elegans 138 Cephea 8 cephea 11 Cepheidse 11 Chaetium 156 festucoides 156 Chsetochloa italica 142 Chamaeleo excubitor 219 Chase, Agnes. Notes on genera of Panicse. IV 103-160 Chasea ill Chiarella centripetalis so Chirodropus gorilla 8 Chloridion Ill cameroni Ill Chloronerpes meridensis 33 tobagensis :;;; trinitatis 32 Chrysaora hysoscella 10 Chyropsalmus quadrumanus ..... 8 Citharichthys arcti irons 210 Cladonema radiatum 79 Cladoneniidie 78 Cladonemiiiffi 70 < 'lark. H. Walton, and Gillette, Geo. H. Some observations made on Little River, near Wichita, Kansas, with reference to the Unibnida? . . .63-68 Clark, Austin II. A new crinoid genus from the Indian Ocean ... . 87-88 ■ Owen's recent encrinite identi- fied '.IS Clavula gossei 79 neglecta 7o pileata 82 1 lleachne 138 Clytia volubilis 82 ClytiinsB 82 ( lockerell, T. I). A. The nomenclature of the Scyphomedusse 7-12 Note on the scales of the Oste- oglossid fishes 39 The nomenclature of the Hy- dromedusse 77-86 Some notes on rish scales .209-211 Codonidse • • - 77 Coelorhynchus caribbseus 211 Colaptes borealis .",1 brunnescens 32 chrysoides 32 mearnsi 32 Collaspidae 0 ( 'onis mitrata 70 Cooke, W. W. Peculiar migration of the evening grosbeak x Coregomis quadrilateralis 49 Coridochloa .... 120, 157 semialata 158 Coronatse ... 9 Corynitis 78 Coryphsenoides rupestris 211 Cosmetira 82 Cotylorhiza tuberculata 11 Coville, F. V. Exhibition of a pot of trailing arbutus grown from seed . ix Raising trailing arbutus from the seed viii 1 'tenaria ctenophora .... ... 70 ( lubaia aphrodite . si Cunina globosa 86 Cunissa polyporpa 86 Cunoctantha octonaria 86 Cuvieria carisochroma 81 Crambione mastigophora 11 Craspedacusta sowerbii si Cristivomer namaycush 19 Crocidura pullata 241 1 Irossota brunnea 85 Cryptoblephnrus balinensis . ... is boutonii 17 cursor 18 keiensis is peronii is Cryptotis frontalis 222 gracilis 221 macer 223 nigrescens 222 orophila 222 Cyanea capillata 10 Cyaneidse 10 Cyclometra flavescens 87 clio sv Cymatochloa 139 liuitans ... 139 repens 139 Cynailurus guttatus 225 jubatus 225 soemmeringii 226 1 lytseidse 70 < lytseinse 70 Cytseini 70 CytaMs tetrastyla 70 Index. 257 D Daetylometra lactea 10 Dapedoglossus testis 39 I >:i siii moluccarutn 1< smaragdinum 16 Dasypterus floridanus 228 intermedius 228 pana mensis 227 Dearborn, N. Some nesting habits of the water ousel and the cuckoos . vii l (elphinodi m mento 38 wymani :js Dendronema stylodendron 79 Depastrum cyathiforme 8 I tesmana moschata 13 Desmonema gaudichaudii in Dichotomia cannoides si Dicodoniini 78 Dicodonium cornutum 7s Digitaria 111. 137 Dimorphostachys 140 monostachya -140 Diplaria 120 Dipleurosoma typicum si Dipluhnaris antarctica 10 Dipurena strangulata 78 Discomedusa lobata in Discomedusinse in Dissonema saphenella 79 Drj monema dalmatina in Dryonastes shensiensis 13 E Echinochloa 154 crusgalli 154 holciformis 155 Echinolaena 116 hirta 116 inflexa • • 11" madagascarensis 119 polystachya in;, us scabra 116 Ectopleura diunortieri 78 Eirene viridula s-'> Eireninae • • 83 Eleutheria dichotoma . . . 79 Emballonura sulcata 161 Empidonax johnstonei 58 Enchelyopus cimbrius 212 Enneacanthus gloriosus t:; Epimys barussanus 26 batus 27 calcis 89 majrens 26 mengurus 21 querceti 90 Eptesicus pallens 53 Eremachyrion L39 Eriachne 120 Ericymba buccata 172 Eriocbloa 124 distachya 124 polystachya 124 Esox reticulatus 74 Kil stoma Habellare 174 Exoglossum maxillingua 213 Eucheilota ventricularis 82 Eucodonium brownei 7s Eucopella bilabiata v- Eucopiidae 82 Eucopiinae 82 Eucopium pictum 82 Eupilema scapulare 12 Eutima mira 83 Eutiminae 83 Eutonina socialis 83 Evermann, B.W. Observation on fly- catching by brown rat .vii E\ ermann, B.W. On the shipment to the U. S. of young fur seals from the Pribilof Islands x F Fagonia aspera 247 barclayana 248 californica 248 chilensis 249 glutinosa 246 hindsiana 248 insularis 217 laevis 249 longipes 249 pachyacantha 246 palmeri 245 rosei 247 scoparia 245 viscosa 247 Felis arundivaga . 176 azteca 175 chalybeata 225 coryi 175 couguar 175 fearonis 226 guttata 225 jubata 225 venatica 225 Fiber zihethicus 13 Fleming, J. II. A new teal from the Andaman Islands 215-216 Q Gadus callarias ... 212 Gambusia holbrookii 91 patruelis 91 Gastroblasta timida 82 Geryonia proboscidalis 85 Geryoniidse 85 Gill, Theodore. A new translation of Aristotle's " History of Animals" . viii Gillette, George II. and Clark H.Walton. Some observations made on Little River, near Wichita. Kansas, with reference to the Unionidse ■ • ■ 63-68 Glyptocephalus cynoglossus 210 Goldman, E. A. Three new mammals from Central and South America 2: 17-240 Recent explorations in Panama x Gonionemus vertens 84 Gossea corynetes 84 Gramerium Ill convolutum Ill Grinnell, Joseph. A new blue grosbeak from < 'alifornia 163-164 Guiraca salicarius 163 H Haemaphysalis chordeilis 207 Haliclystus auricula 9 Halicreas minimum 85 Halimocyathus platypus 9 Halitrephes maasi 85 Halopsis ocellata 83 Harpostachys 140 Hay, O. P. Exhibition of a fossil skull of a snapping turtle . . . ix The ice-age and its extinct ani- mals . . ix Hay, W. P. Exhibition of slides show- ing amphioxus and the eggs of the blue crab x Hekaterosachne 152 elatior 152 Heller, Edmund. Collecting big game with Roosevelt in Africa vii 258 The Biological Society of Washington, Helopus 124 brachystachys 125 pilosus 124 Hesperomys arcticus 101 maniculatus 101 nebrascensis 102 Heterandria formosa 91 Heteranthoecia 151 isachnoides 151 Heterotiara anonyma 79 Hippagrostis 152 amboinica 152 Hippoglossoides platessoides 2lo Hitchcock, A. S. Review of N. A. Flora and remarks on the type of Pani- cum ix Hollister, N. A new musk-deer from Korea 1-2 The generic name of the musk- rat 13-14 The type locality of Mellivora abyssinica 37 Two new species of Epimys from Luzon 89-90 The Louisiana puma • . 175-178 Description of a new Philippine flying-squirrel 185-186 The generic name of the African buffalo 191-194 The nomenclature of the cheetahs 225 Holosetum 120, 159 philippicum 159 Homoeonema typicum 85 Homolepis 146 aturensis . 1 jr. isocalycina 14r> longispicula 14(1 Howard, L. O. The house-fly and typhoid fever . . ix On the arrival in the U. S. of living parasites of the white fly . . x Howell, A. H. ( aptureof Sorex dispar in West Virginia '.is Hunter, W. 1). Some cactus insects with special reference to cochineal viii Hybocodon prolifer 78 Hybris noctividus :>7 Hydrichthys minis .... . . . . 7s Hydroctena salenskii 86 Hydrosaurus amboinensis 19 microlophus 19 pustulosus 1!) weberi 20 Hymenachne 146 amplexicaulis 140 I Ichnanthus 142 neinorosus lis panicoides 142 Ictalurus f meatus 171 Inca 38 Irenopsis hexanemalis 83 Isachne lt;i australis 150 trachysperma 1">1 Ixodes angustus 204 banksi 200 kingi 201 rugosus 197 sculptus 206 woodi ... 205 K Kendall, W. C. Notes on Percopsis guttatus Agassiz and Salmo omis- cumaycus YValbaum 15-52 Kishinouyea nagatensis 9 Kuragea depressa in L Lagodon rhomboides 71 Lagopus albus 233 alexandne 233 ungavus 233 Lampsilis anodontoides 68 Lantz, D. E. Domestication of wild animals viii Laodicea cruciata 81 Lappagopsis 131 bijuga 136 Larosterna inca 38 Lasiacis ... 144 divaricata 144 liebmanniana 1 45 oaxacensis 14."> procerrima ]4."> rhizophora 145 ruscifolia 145 Leiolepisma pullum 15 Leistes erythrothorax 190 militaris 190 petilus 190 Lepidoineda vittata 213 Lepidosteus osseus 73 Lepomis auritus 72, 173 cyanellus 71,172 gibbosus 71 pallidus 73, 173 solis 17:! Leptobrachia leptopus 12 Leptobrachiidse 11 Leptocoryphium iu9 lanatum 109 molle 109 Leptoloma 111 cognatum 111 Leptomedusie si Leucophrys 123 mesocoma 123 Limanda ferruginea 210 Limnoenida tanganjicte 84 Limnocnididse s4 Limnoreini So Linuche unguiculata 9 Linuchida? .... 9 Linvillea agassizi 7s Lipkea ruspoliana 9 Lipkeinre 9 Liriope cerasiformis 85 exigua 85 Livoneca cumulus . . . .' vis Lobonema smithii 11 Lophopsetta maculata 210 Lorifera lorifera . . 12 Lucernaria quadricornis 9 Lucernariidie s Lucernariina: s Lychnorhiza lucerna 11 Lymnorea triedra 80 Lyon, M. W., Jr. Seiuruspoliopas Lyon changed to Sciurus conipus . ... 98 Descriptions of four new tree- shrews 167-170 M McDermott, F. Alex. Some observa- tions on a photogenic micro-organ- ism, Pseudomonas lucifera Molisch 179-1S4 Macrourus bairdii 211 Mseotias inexpectata 84 Maizilla 139 stolonifera 139 Margelopsinae 78 halt J 259 Margelopsis haeckelii 78 Marmosa ruatanica 237 zeledoni ... 238 Marsh, CD. <>n Aars' biological in- vestigations in Lake Tanjanyika . vii Review of Hollinger's paper on the geographical distribution of diatoms viii Mastigias papua n Mearns, E.A, New names for two sub- species of Peromyscus manieulatus (Wanner) .... , .... 101-102 Description of a new subspecies of the painted bunting from the interior of Texas 217-218 On the correct name for the red-winged blackbird of the north- eastern United States 226 Meek, S. E. Recent explorations in Panama x Megamastigias punctata 11 Melanogrammus seglefinus 212 Melicertinse sl Melicertissa clavigera si Melieertnm campanula 81 Mellivora abyssinica 37 Melospiza fisherella 251 inexpectata 234 Merluccius bilinearis 212 Mesosetum 119, 120 bifarium 123 cayennense 119 chlorostachyum 122 exaratnm 121 ferrngmeum 122 loliiforme 122 rottboellioides 121 sclerochloa 122 wrightii 121 Mieroeampana conica 78 Microgadus tomcod 212 Microhydra ryderi 84 Micropterus dolomieu 69 salmoides 69, 171 Microtus mandarinus 54 pullus 53 Milium punctatum 124 ramosum ■ . 124 Miller, G. S.. Jr. Descriptions of two new raccoons ... 3-0 Descriptions of six new mam- mals from the Malay Archipelago 25-28 ■ Note on the Mus commissar ius of Mearns 38 New names for two European voles 39 Four new Chinese mammals . 53-56 A new bat from the Caroline Islands 161-162 A new mouse-deer from the Rhio-Linga Archipelago . 165-160 Three new shrews of the genus Cryptotis .... 221-224 Note on the Mexican bats of the genus Dasypterus 227 The volcano rabbit of Mount Iztaccihuatl 228 A new roe-deer from ( 'hina 231-232 Two new shrews from Kash- mir 241-242 A new jumping-mouse from New Mexico 253-254 Mitrocoma annse 82 Mixter, George. Afoot across the British Columbia Rockies ... viii Mixter. W. J. The desert country of Lower California ... . . . viii Mnestra parasites 79 Modeeria formosa su Moenchia speciosa .... Moerisia lyonsi . . . Monachne . . unilateralis . . Monachyron villosum . ■ Moschus parvipes . . Mus commissarius • • coypus Mylocheilus caurinus N 139 139 78 125 125 157 157 1 38 13 213 Naenia 38 Nannus idius II Narcomedusae ... 85 Nausithoe punctata 9 Navicularia 143 glabra 143 hirta • 143 lanata 143 Nemopsis bachei 80 Netocertoides brachiatum 81 Niobia dendrotentacula 80 Notosema dilectum 210 Notropis amcenis 172 arge 172 photogenis 172 O ( Ibelia marina 82 Oberholser, II. c. Description of a new Melospiza from California 25H252 Oceania coccinea 82 Ochotona annectens 54 bedfordi 55 Octocanna polynema 83 Octogonade mediterranea 83 ( (edipachne 125 punctata 125 Olindias sambaquiensis ,H4 Olindiidse . -84 Olindioides formosa *4 Omisco maycus 47 Ondatra coypus 13 zibethicus 13 Oplismenus 123.152 africanus 152 holciformis 155 undulatifolius 152 Orchistoma pileus si steenstrupii 81 Orthopogon 152 aemulus 152 compositus 152 flaccidus 152 holciformis 150 imbecillis 152 Orthopristis chrysopterus 71 Osteoglossum formosum 39 Pachycordyle napolitana 77 Pachycordylini 77 Palephyra antiqua 9 Palmer, Wm. Notes on Javan natural history and the salt makers of Tjihara, Java viii Pandea conica 79 Pandeinae 79 Pandeini 79 Paniceae 103 Panicum ill almadense -143 amphistemon 149 argentium 124 260 The Biological Society of Washington. Panicum aturense 149 bifarium 123 brachystachyum ill) campylostachyum 114,115 caudiculatum 123 chlorostachyum 122 chrysoblephare 134 eiliatissimum 124 cimicinum 158 cognatum Ill cultratum 115 digitarioides 128 durum 148 echinolama 117 elytrochaetum 123 eriochryseoides 122 eruciforme 126 exaratum 120, 121 ferrugineum 120, 122 geminatum 127 gland ulosum 118 hemitomon 128 holciforme ... 156 isachne 127 leiocarpum 143 leucophaeum 119 loliiforme 122 lolium 122 mesocomnm 123 miliaceum 141 monostachyum 114 nemorosum Us nigropedatura 124 obtusum ■ 128 pappophorum I'.'n petrosnm 113, 115 prostratum 127 rottboellioides 119 sanguinale Ill schumannii . . 116 sclerochloa 122 semialatum 158 serratum . 124 tenerrimum 110 texanum 12!) thrasya 114 thrasyoides 112, 114 tuerckheimii 141 uncinatum . lis urvilleanum 142 viridiflorum 146 Pantachogon haeckelii 85 Paralichthys oblongus 210 Paraphyllina intermedia 9 Paraphyllinidse 9 Paruinbrosa polylobata 10 Paspalus 137 tener 136 Paspalanthium 139 stoloniferum 139 Paspalusn 137 annulatum 124 canaliculatum 136 capillare 133 chrysoblephare 135 digitaria 132 dimidiatum 137 disseetum 137 dissitiflorum 135 distichum 137 elongatum L39 exasperatum 135 fastigiatum L36 furcatum 132 immersus 130 lanatum 109 laxiflorum 133 malacophyllum 139 membranaceum 137 Paspalum minutum I33 paniculatum 137 platycaule 113, 132 punetatum 124 rosei 132 stel latum 138 stoloniferum 139 virgatum 137 Passerella altivagans 234 Passerina ciris . . . .217 pallidior 217 Pegantlia pantheon 85 1'elagia noctiluca 10 Pelagiidse 10 Pelagohydra mirabilis 78 Penelope perspicax is; Pennaria pennaria 77 Pennariini . 77 Pentarrhaphis 120 Perca flavescens 72 Percopsis guttatus 46 hammondi 51 omiscomaycus 51 Perieolpa quadrigata 9 Periphylla hyacinthina 9 Periphyllidse . . 9 Periphyllopsis braueri 9 Perognathus litus (31 Peromyscus arcticus 102 borealis 102 lutens 102 nebrascensis 102 osgoodi 102 Persa ineolorata 85 Petasidse si Petasus atavus 84 Phacellophora camtschatica 10 Phialidium hemisphaericum 82 Phialopsis diegensis 82 Phialucium virens 82 Phloeotomus abieticola 33 floridanus 33 picinus :;] pileatus 33 Phortis gibbosa 83 Pliotobacterium phosphorescens . . .181 Phyllorhiza 8 Picumnus dimotus 34 flavotinctus 34 granadensis 34 olivaeeus 34 panamensis 34 Pimephales notatus 172 Pipistrellus curtatus 25 Pitymys druentius 39 fuscus • • 39 pascuus 39 savii 39 selysii 39 subterraneus 39 Planesticus personus 58 Planiceros 192 Platophrys constellatus 210 ocellatus 210 Podocorj in' earnea 80 Polionetta leucopareus 215 Pollachius virens 212 Polyorchidinae 81 Polyorchis penicillata 81 Polyrhiza vesiculosa 11 Pomoxis annularis 73 sparoides 73 Poralia rufescens 10 Preble, E. A. Exhibition of bird and mammal skins from British Co- lumbia viii The hooting of the blue grouse x Prionotus evolans 74 Proboscidactyla flavicirrata 80 Index. 261 Proboscidactylinse 80 Proboscidactylini su Procyon hernandezii 6 1 1 1 ; i > nardi 6 minor I pumilus 3 pygmaeus 6 Proechimys steerei 238 Protiara tetranema 79 Protiarini 79 Pseudoclj tia pentata 82 Pseudomonas lucifera L79 Pseudopleuronectes americanus . . .210 Pseudorhiza aurosa n Psittdrostra oppidana 30 Pteronema darwinii 79 Pteroneminae 7s Ptychogastria polaris 84 Ptychogastriidae 84 Ptychogena Iactea 81 Pyrocephalus blatteus 189 Q Quadrula lachrymosa pustulosa • ■ . 66 66 R Rathkea blumenbachii so Ratufa major 28 Reimaria .... i:;s acuta 136, 138 Candida 138 elegans 138 Reimarochloa 136 aberrans 137 acuta 136 brasiliensis 136 oligostachya . 136 Rhacostoma atlanticum 83 globulare .... . . 83 Rhinodon typicus 97 Rhipidura penitus ... II Rhizostoma pulmo 12 Rhizostomata 11 Rhizostomidae 12 Rhopilema rhopalophora \i Rhynchelytrum 157 dregeanum . . ... 157 Richardson, Harriet. Noteon Agarna carinata Schioedte and Meinert . . 98 Richardsonius siuslawi 213 Ridgway, R. Diagnoses of Mime new forms of Picidae • . . . . 31-36 Riley, J. II. On the correct name of the Inca tern 38 Description of a new Dryon- astes from China .43-44 Descriptions of three new birds from ( lanada 233-236 Romerolagus diazi 228 nelsoni 228 Rose, J. N. The deserl vegetation of Lower California ix Rupornis occidua 187 gabsab 138 Saccharum reptans 125 Sacciolepis . 146 gibba 146 Safford, w. E. A day in the Galapagos Islands ix Salmo omisco maycus 51 salar 49 Sahnoperca pellucida 47 Sanderia malayensis 10 Sanguinaria ill nevena rae ill Sanguinella Ill Saphenia bitentaculata 83 Sarsia 78 tubulosa 78 Sarsiidse 77 Sarsiinae 77 Sarsiini 7s Scapaneus nelsoni 34 guatemalensis 34 renins 34 Schilbeodes gyrinus 17J Schwarz, E. A. Recent explorations in Panama x Sciuropterus crinitus 185 Sciurus conipus 98 poliopus 98 Scleropages formosus 39 Scrippsia pacifica sl Scutachne 148 amphistemon 149 dura 149 Seal, William P. Breeding habits of the viviparous fishes Gambusia hol- brookii and Heterandria formosa 91-96 Semaeostomata 10 Semseostomeae in Setaria 123 Slabberia • . . 78 Sminthea eurygaster si Smith, II. M. Noteon the occurrence of the whale shark, Rhinodon typi- cus, in the Philippine Islands ■ . 97 Snyder, T. E. The insect t'auna of tele- graph poles viii Solea solea 209 Solmaridae 85 Solmaris leucostyla 85 Solmissus albescens 86 Solmundella bitentaculata 86 Sinvx dispar 97,98 fumeus 99 maenrrus 97 planiceps 242 Rpheroides maculatus 74 Spirocodon saltatrix 81 Standley, Paul <'. The American spe- cies of Fagonia 243-250 Stauraglaura tetragonima ...... 85 Staurodiscus tetrastaurus si Stauromedusida . 8 Staurophora mertensii 83 Steenstrupia rubra 7s Steenstrupiini ... 78 Stenoscyphus inabai ■ ■ '-1 Stereochlaena 112 jeffreyssii 112 Sthenonia albida 10 Sthenoniinae . . . .... 10 Stiles, O. W. Note on the progress of the International Committee on Zoological Nomenclature ix Stizostedion vitreum 17:'. Stomobrachium lenticulare 83 Stomolophus meleagris 12 Stomotoca dinema 79 Stomotocini 79 Sturnus predatorius 227 Stylactis fuciola • .80 Syacium papillosum 210 Symphurus piger 209 Symphynota complanata 67 Synceros 192 Syncerus aequinoctialia 192 azrakensis .... 192 brachyceros 192 caffer 192 262 The Biological Society of Washington. Syncerus centralis 192 cottoni 192 ga riepensis 192 limpopoensis 192 matthewsi 192 mayi 192 nanus 192 neumanni 192 planiceros 198 radcliffei 192 mahaensis 193 schillingsi 193 thierryi 193 wiesei 193 wembarensis 193 Syntherisma Ill sanguinalis ill Tamoya haplonema 8 Tautoga onitis 70 Tesserantha connectens . 8 Tesseranthinse 8 Tetranema europium 81 Tetrorchis erythrogaster sn Thamnitis tetrella 80 Thamnophilus catus 189 Thamnostylus dinema so Thaumantias 82 Thaumatoscyphus distinctus 9 Thrasya 112 campylostachya 11") cultrata 114, 115 hirsuta 112. 115 paspalnides 112, ill petrosa 115 tbrasyoides lit Thysanostoma 8 brachyura 11 Tiaropsis multicirrata 82 Tidestrom, I. Notes on the aspens . . vii Tima flavilabris 83 Timoides agassizii 81 Toreuma dieuphila 11 Townsend, C. H. On the Albatross Ex- pedition to Lower California ■ . . ix Toxorchis arcuatus 81 Tragulus parallelus 165 Trachymedusae 84 Trachynema ciliatum 84 Trachyneniidte 84 Trachyneminse St Tricachne Hi'.i ferruginea 110 insularis 109 recalva 110 sacchariflora 109 tenuis 110 Tricholaena 1"i7 micrantha 157 Trichorhiza brunnea 78 Trichorhizini 78 Tripedalia eystophora 8 Tritogonia tuberculata . 68 True, F. W. Discovery of a fossil delphinoid cetacean with tubercu- late teeth 37 Tupaia natuna? 168 pemangilis 168 raviana 167 sincepis 169 Turritopsis nutricula ... 80 Tylothrasya petrosa 115 u Ulmaridse L0 Umbellularia caTpenteri 98 I'ndosa undulata 10 t"nio schoolcraftii 66 Urashimea globosa 79 Urochloa panicoides • . . . 142 I'rophycis regius 212 V Vallentinia falklandiea 84 Valota 109 hitchcockii 1 In insularis 109, 1 19 pittieri 1 in Vaughan, T. W. The keys, corals, and coral reefs of Florida ix Veniliornis darienensis ....'.... 33 Versura palmata 11 Vestiaria suavis 29 W Webster, F. M. A recently imported enemy of alfalfa vii Weed, A. C. Noteson the fishes of the District of Columbia x and Bean, Harton A. Notes on the coloration of fishes 69-76 Recent additions to the fish fauna of the District of Columbia 171-171 White. David. Exhibition of coal showing carboniferous plants . . ix <>n another supposed fruit- bearing, fern-like plant from the American Permian x Wilcox, T. E. Note on migrating wild geese and on the transplanting of trailing arbutus viii On the occurrence of white quartz in the stomach of blue grouse x Willsia sellata 80 Wirtgenia 1 H » paspaloides 1 In z Zanclea costata 78 Zancleopsis dichotoma 79 Zapus alleni 253 lutens 253 Zygocanna pleuronota 83 ZygocannuU undulosa 83 MBL WHOI LIBRARY UH 11NK G