PROCEEDINGS OF THE Biological Society of Washington VOLUME XV 1902 WASHINGTON PRINTED FOR THE SOCIETY 1903 COMMITTEE ON PUBLICATIONS WILLIAM P. HAY, Chairman. WILFRED H. OSGOOD DAVID WHITE vo CONTENTS. Page Officers and committees for 1902 : v Proceedings yii-xii Members xiii-xviii The Generic Names of the North American Skunks, by Arthur H. Howell 1-9 The Large Yellow Pond Lilies of the Northeastern United States, byGerrit S. Miller, Jr ... 11-13 A New ppisthoglyph Snake from Formosa, by Leonhard Stej- neger 15-17 New American Species of Chamaecrista, by Charles Louis Pollard 19-21 Ribes aureum and Ribes lentum, by Frederick V. Coville 23-29 Description of a New Swallow from the Western United States, by Edgar A. Mearns 31-32 Descriptions of Three New Birds from Cuba and the Bahamas, by William Palmer and J. H. Riley 33-34 General Notes 35-39 Note on the name Rhopocichla, 35; The technical names of two Dogbanes from the District of Columbia, 35; A fully adult specimen of Ophibolus rhombomaculatus, 36; The gen eric names Coccystes untenable, 37; Gerrhonotus cceruteus versus Gerrhonotus burnettii, 37; Ptychozoon kuhli a new name for P. homalocephalum, 37; On the proper application of the name Cambarus carolinus Erichson, 38; Note on the names of the genera of Peccaries, 38; The technical name of the Virginia Deer, 39. Twenty New Pocket Mice (Heteromys and Liomys] from Mexico, by C. Hart Merriam 41-50 A New Pig from Sumatra, by Gerrit S. Miller, Jr 51-52 Some Recent Changes in the Nomenclature of West Indian corals, by T. Wayland Vaughan 53-58 A Further Note on the Generic Names of tue Mephitince, by J. A. Allen 59-66 Five New Mammals from Mexico, by C. Hart Merriam 67-69 A New Bobcat (Lynx uinta] from the Rocky Mountains, by C. Hart Merriam 71-72 Three New Foxes of the Kit and Desert Fox Groups, by C. Hart Merriam. 73-74 Two New Shrews of the Sorex tenettus Group from California, by C. Hart Merriam 75-76 Two New Bears from the Alaska Peninsula, by C. Hart Merriam 77-79 A New Rabbit from Southern Texas, by Gerrit S Miller, Jr 81-82 Description of a New Phyllostome Bat from the Isthmus of Panama, by Marcus W. Lyon, Jr 83-84 General Notes 85-90 Ixoreus should replace Hesperocichla, 85; Two preoccupied avian genera, 85; The proper name of the arctic horned owl, 86; A new name for Buthotrephis divaricata D. W., 86; Change of name of Ficusf hesperia from vicinity of Ashland, Oregon, 86; Some names of American Cerridae, 87; The type iv Contents. locality of Ametrida minor H. Allen, 88; An early name for the northern form of Sphyrapicus ruber, 89; The color of the fully adult Ophibolus rhombomaculatus, 90. The Mammals of Margarita Island, Venezuela, by Glover M. Allen 91-97 A New Ribes from New Mexico, by T. D. A. Cockerell 99-100 Synopsis of the North American species of Sigmodon, by Vernon Bailey 101-116 Seven New Mammals from Western Texas, by Vernon Bailey.... 117-120 A List of the Batrachians and Reptiles of the District of Colum bia and Vicinity, by William P. Hay 121-145 Two New Species of Poisonous Sumachs from the States of Rhode Island and Florida, by Edgar A. Mearns 147-149 Description of a New Bat from Columbia, by Marcus W. Lyon, Jr 151-152 General Notes 153-156 The generic names of the Peccaries, Northern Fur Seal, and Sea Leopard, 153; Note on the Vespertilio incautus of J. A. Allen, 155; Note on Chilonycteris davyifulvus of Thomas, 155; Parus inornatus griseus renamed, 155; The Hooded Warbler breeding near Washington, D. C., 156. Descriptions of Two New Birds from Trong, Lower Siam, by Charles W. Richmond 157-158 Description of a New Subspecies of Stenopsis cayennensis from Curacao, by Charles W. Richmond 159-160 Six New Skunks of the Genus Conepatus, by C. Hart Merriam.... 161-165 Four New Arctic Foxes, by C. Hart Merriam 167-172 Two New Malayan Mouse Deer, by Gerrit S. Miller, Jr 173-175 Four New Plants from Mexico, by Charles Louis Pollard and T. D. A. Cockerell 177-179 Descriptions of New Species of Synaptomys and Phenacomys from Mackenzie, Canada, by Edward A. Preble 181-182 The Genus Hedysarum in the Rocky Mountains, by Aven Nelson 183-186 Descriptions of Eight New Birds from Islands off the West Coast of Sumatra, by Charles W. Richmond 187-190 Notes on Little-Known New Mexican Mammals and Species Apparently not Recorded from the Territory, by C. M. Barber 191-193 General Notes 195-199 The occurrence of certain tropical plants in Mississippi, 195; Note on the generic name of the horseshoe crab, 196; A fur ther note on the name of the Argentine Viscacha, 196; On the generic names, Notop'horus, Alces, Dama, and CepJialotes, with remarks on the "one-letter rule" in nomenclature, 197; On the distribution of Hyla emttata Miller, 199. Two New Violets from the Eastern United States, by Charles Louis Pollard 201-203 An Addition to the Coral Fauna of the Aquia Eocene Formation of Maryland, by T. Wayland Vaughan 205-206 A Redescription of the Coral Platytrochus speciosus, by T. Way- land Vaughan 207-209 A New Subspecies of the Cuban Cliff Swallow, by E. W. Nelson 211 Observations on the Conditions of Light Production in Luminous Bacteria, by Randolph E. B. McKenney 213-234 Some Generic Names of Turtles, by Leonhard Stejneger 235-238 A Salamander New to the District of Columbia, by Leonhard Stejneger 239-240 Three New Skunks of the Genus Spilogale, by Arthur H. Howell 241-242 A New Bat from the Island of Dominica, by Gerrit S. Miller, Jr 243-244 Two New Tropical Old World Bats, by Gerrit S. Miller, Jr 245-246 General Notes 247-250 The occurrence of Cardiospermum halicaeabum Linn, in Contents. v Louisiana, 247; Note on Pinaroloxias inornata (Gould), 247; The common Nyctinomus of the Greater Antilles, 248; Lophostoma Venezuela changed to Tonatia Venezuela, 248; The external characters of Brachyphylla nana Miller, 249; An overlooked specimen of CMlonycteris psilotis, 249; A second specimen of Pterygistes azoreum Thomas, 250; The status of Nyctinomus nevadensis (H. Allen), 250; The generic position of Nyctinomus orthotis H. Allen, 250. PLATES. Plate I. The two genera of Carolina Skunks in Comparison with Catesby's 'Pol-Cat'. Plate II. Nymphcea variegata and N. advena. Plate III. Seeds of four species of Rhus. TEXT FIGURES. Page 12. Section of petiole near middle, a Nymphcea variegata, b N. advena. Page 123. Topography of the salamanders and frogs. Page 133. Topography of a snake. Page 142. Topography of a turtle. Page 209. Cotypes of Platytrochus speciosus. Page 218. Diagram of culture flask and bulb for introduction of sterile reagents. ERRATA. Page 55, line 18, instead of viridus read viridis. Page 57, lines 28, 30, 31 and 32, instead of 1901 read 1902. Page 196, line 13, instead of 1826 read 1816. Page 246, line 9, instead of (skin and skull) read (in alcohol). OFFICERS AND COUNCIL OP THE BIOLOGICAL SOCIETY Of WASHINGTON For 1902. (ELECTED DECEMBER 2S, 1901.) OFFICERS President F. A. LUCAS Vice- Presidents WM. H. ASHMEAD F. H. KNOWLTON B. W. EVERMANN T. S. PALMER Recording Secretary WILFRED H. OSGOOD Corresponding Secretary T. W. STANTON Treasurer DAVID WHITE COUNCIL WILLIAM H. DALL* CHARLES L. POLLARD THEODORE GILL* GEORGE M. STERNBERG* L. O. HOWARD* H. J. WEBBER FREDERICK V. COVILLE* M. B. WAITE A. F. WOODS LESTER F. WARD* C. HART MERRIAM* CHARLES A. WHITE* WILLIAM P. HAY STANDING COMMITT^ES-1903. Committee on Communications V. K. CHESNUT, Chairman VEKNON BAILEY A. F. WOODS B. W. EVERMANN Committee on Publications WILLIAM P. HAY, Chairman WILFRED H. OSGOOD DAVID WHITE *Ex-Presidents of the Society. (vi) VOL. XV, PP. VM-XVIII JANUARY 31, 1903 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON PROCEEDINGS. The Society meets in the Assembly Hall of the Cosmos Club on alternate Saturdays at 8 p. m. Brief notices of the meetings, with abstracts of the papers, are published in Science. January 11, 1902347111 Meeting. The President in the chair and 22 persons present. F. A. Lucas exhibited a malformed mastodon tooth showing duplication of cusps. The following communications were presented: M. B. Waite: A Problem in Plant Physiology and Patho- W. H. Osgood: The Supposed Occurrence of Caribou on the Queen Charlotte Islands.* Jacob Kotinsky: Present Opinion Concerning the Home of the San Jose Scale. . ' 4 January 25, 1902 348th Meeting. The President in the chair and 34 persons present. W. H. Dall discussed the practice of rejecting zoological names preoccupied by those identical except in termination. The following communications were presented: David Griffiths: A Seed Planter, f *North Am. Fauna No. 21, pp. 26-28, 1901. fBull. Tor. Bot., 01, 29; 164-169, 1892. (vii) viii The Biological Society of Washington. F. A. Lucas: A Phase of the Blue Fox Question.* R. H. True: The Physiology of Sea Water. February 8, 1902 349th Meeting. The President in the chair and 36 persons present. The following communications were presented: Chas. A. White: The Mutation Theory of von Hugo de Vries.f H. W. Olds: Some Deductions from the Study of Bird Song. J February 22, 1902 asoth Meeting. The President in the chair and 18 persons present. V. K. Chesnut exhibited a variety of spring scales for field use. F. A. Lucas discussed a paper entitled Bird-killing as a Method in Ornithology. The following communications were presented: C. H. Townsend: The Present Status of the Carp in Ameri can Waters. C. P. Hartley: Some Effects of Pollinating Premature Flowers. || L. H. Dewey: The Identity of Prickly Lettuce. F. A. Lucas: The Armor of Stegasaurs. March 8, 1902 ssist Meeting. The President in the chair and 78 persons present. L. O. Howard exhibited photographs showing the supposed fibrous structure of artificial protoplasm. The following communications were presented: C. Hart Merriam: Protective and Directive Coloration in Animals. L. O. Howard: Some Common Instances of Protective Color ation among Insects. ^Science, XVI, 216, Aug. 8, 1902. fSmithsonian Report for 1901, pp. 631-640, 1902. JHarper's Magazine, CV, pp. 474-478, Aug. 1902. JBull. No. 22, Bureau PL Ind., U. S. Dept. Agric., pp. 1-39, Oct. 1902. Proceedings. ix March 22, 1902352011 Meeting. The President in the chair and 27 persons present. The following communications were presented: C. W. Stiles: Eleven Miscellaneous Papers on Animal Para sites.* W. C. Kendall: Notes on Sticklebacks. W. H. Dall: Notes on Trophon.\ E. S. Steele: The Vegetation of Stony Man Mountain, near Luray, Virginia. April 5, 1902 353rd Meeting. The President in the chair and 38 persons present. The following communications were presented: Frank Baker and F. A. Lucas: The Area of Muscle Inser tion as an Index to Power. William P. Hay: The Subterranean Fauna of the United States. April 19, 1902 354th Meeting. The President in the chair and 33 persons present. The following communications were presented: B. W. Evermann: Notes on Some Mexican Fishes. J W. W. Cooke: Some Untenable Theories of Migration. May 3, 1902 355th Meeting.. Ex-President Dall in the chair and 9 persons present. Adjourned without program on account of lack of attendance. May 17, 1902 356th Meeting. The President in the chair and 45 persons present. V. K. Chesnut exhibited lantern slides showing various pois onous plants. *Bull. No. 35, Bureau Animal Indust., U. S. Dept. Agric., pp. 1-59, 1902. fProc. U. S. Nat. Mus., XXIV, 533-550, March, 1902. JBull. U. S. Fish Com. for 1901, pp. 137-159, figs. 1-8, May 3, 1902. x The Biological Society of Washington. F. A. Lucas exhibited lantern slides showing a mounted skeleton of Claosaurus and a restoration based upon it. The following communications were presented: A. H. Howell: The Summer Birds of Mt. Mansfield, Vermont.* W. W. Cooke: Bird Migration Routes. June i, 1902 357th Meeting. The President in the chair and 30 persons present. The following communications were presented: C. W. Stiles: Surra, A Military Disease of Great Importance and Caused by a Protozoan, f B. W. Evermann: The American Species of Shad.]; October 18, 1902 3s8th Meeting. The President in the chair and 30 persons present. W. H. Dall discussed certain mollusks from South America and Alaska. The following communications were presented: R. E. B. McKenney: Luminous Bacteria. F. V. Coville: Plants of the Klamath Indians. November i, 1902 359th Meeting. The President in the chair and 31 persons present. H. J. Webber exhibited specimens of hybrid oranges from South Carolina. The following communications were presented: F. V. Coville: Dye Plants of the North Carolina Mountaineers. E. W. Nelson: The Evolution of Subspecies as Illustrated by Mexican Quails and Squirrels. November 15, 1902 36oth Meeting. The President in the chair and 48 persons present. *The Auk, N. Y., XVIII, 337-347, Oct. 1901. {Bull. No. 42, Bureau Anim. Indust., U. 8. Dept. Agric., pp. 1-30, 1902. JRept. U. 8. Fish Comm. for 1901, pp. 273-288, May 26, 1902. Proc. Biol. Soc. Wash., XV, pp. 213-234, Nov. 20, 1902. Proceedings. xi M. W. Lyon exhibited photographs showing certain bats and their young. M. C. Marsh exhibited samples of water contaminated by coal tar. The following communications were presented: 0. L. Pollard: Some Aspects of the Flora of Cuba. O. F. Cook: Stages of Vital Motion. November 29, 1902 36ist Meeting. The President in the chair and 38 persons present. The following communications were presented: William Palmer: Variation in Downy Woodpeckers in East ern Maryland and Virginia. Vernon Bailey: Sleepy Grass and its Effects on Horses. F. V. Coville: The Use of Sagebrush among the Klamath Indians of Oregon. O. F. Cook: The Function of Latex in the Central American Rubber Tree. December 13, 1902 3624 Meeting. Held in conjunction with the Washington Academy of Sciences, Dr. Theo. Gill in the chair and 47 persons present. The retiring President Mr. F. A. Lucas delivered an address entitled < 'Flightless Birds." December 27, 1902 363d Meeting. The President in the chair and 13 members present. The annual reports of the Recording Secretary and the Treas urer were read and approved. The following officers were elected for the year 1903. President: B. W. Evermann. Vice-presidents: W. H. Ashmead, F. H. Knowlton, T. S. Palmer, William P. Hay. Recording Secretary: Wilfred H. Osgood. Corresponding Secretary: T. W. Stanton. Treasurer: David White. xii The, Biological Society of Washington. Councillors: A. F. Woods, H. J. Webber, M. B. Waite, C. L. Pollard, J. N. Rose. The President then announced the following standing com mittees for the year 1903: Committee on Publication: William P. Hay, Gerrit S. Miller, Jr., David White. Committee on Communications: V. K. Chesnut, Vernon Bailey, A. F. Woods, A. B. Baker, M. W. Lyon, Jr. MEMBERS.* Glover M. Allen, 68 Perkins Hall, Cambridge, Mass. J. A. Allen, American Museum of Natural History, New York, N. Y. Andrew Allison, Lobdell, West Baton Rouge Parish, La. W. B. Alwood, Virginia Polytechnic Institute, Blacksburg, Va. Oakes Ames, Ames Botanical Lab., North Easton, Mass. W. H. Ashmead, U. S. National Museum. Vernon Bailey, U. S. Dept. Agriculture. Mrs. Vernon Bailey, 2019 19th St. A. B. Baker, 1845 Lanier Ave. Frank Baker, Smithsonian Institution. Carleton R. Ball, U. S. Dept. Agriculture. Outram Bangs, 240 Beacon St., Boston, Mass. Charles M. Barber, 407 N. Oregon St., El Paso, Tex. W. C. Barbour, Sayre, Pa. John Hendley Earnhardt, Tarrytown, N. Y. Paul Bartsch, U. S. National Museum. C. F. Batchelder, Cambridge, Mass. H. H. A. Beach, 28 Commonwealth Ave., Boston, Mass. F. E. L. Beal, U. S. Dept. Agriculture. Barton A. Bean, U. S. National Museum. Daniel Carter Beard, Flushing, L. I., N. Y. A. E. Beardsley, State Normal School, Greeley, Colo. W. R. Beattie, U. S. Dept. Agriculture. J. E. Benedict, Smithsonian Institution. Wm. J. Bennetts, 154 U St. N. W. E. A. Bessey, U. S. Dept. Agriculture. Louis B. Bishop, 356 Orange St., New Haven, Conn. Elliot Blackwelder, Walker Museum, Univ. of Chicago, Chicago, 111. Clarence J. Blake, 226 Marlborough St., Boston, Mass. J. Lewis Bonhote, British Museum, Cromwell Road, London, England. Julius Braunnagel, San Antonio, Tex. J. C. Branner, Leland Stanford, Jr. University, Palo Alto, Cal. Wm. L. Bray, University of Texas, Austin, Tex. William Brewster, 145 Brattle St., Cambridge, Mass. N. L. Britton, New York Botanical Garden, Bronx Park, N. Y. J. R. Bromwell, 1147 Connecticut Ave. E. Brown, U. S. Dept. Agriculture. W. L. Burnett, Fort Collins, Colo. Gary N. Calkins, Columbia University, New York, N. Y. Douglas H. Campbell, Leland Stanford, Jr. University, Palo Alto, Cal. *The town address unless otherwise stated is Washington, D. C. jfiv Names of Members* i M. A. Carleton, U. S. Dept. Agriculture. James Carroll, Army Medical Museum. James J, Carroll,. Waco, Tex. Merritts Gary, U. S. Dept. Agriculture. F. M. Chapman, American Museum of Natural History, New York, N, V. K. Chesnut, U. S. Dept. Agriculture, Wayland M. Chester, Colgate University, Hamilton, N. Y, J. W. Checkering, The Portner, 16th and U 8t. N. W. Austin H, Clark, The Harvard Union, Cambridge, Mass, V. A. Clark, Cheyenne, Wyoming. T. D. A. Cockerell, East Las Vegas, New Mexico. G. N. Collins, U. S, Dept. Agriculture, O. F. Cook, U. S. Dept. Agriculture. Charles Montague Cooke, Honolulu, Hawaiian Islands, W. P. Coues, 00 Charles St., Boston, Mass. Frederick V. Coville, U. S, Dept. Agriculture. Ulysses O, Cox r State Normal School, Mankato, Minn. Hugh W. Grouse, Valley View Hospital, Victoria, Tex, George B. Culver, Auditor's Office, War Department. W. H. Dall,- Smithsonian Institution. Chas. B. Davenport, University of Chicago,. Chicago, Ill r Charles F, Dawson, U. S. Dept. Agriculture. E. A. de Schweinitz, U. S. Dept. Agriculture, L. H, Dewey r U. S. Dept. Agriculture. Charles Wright Dodge, University of Rochester, Rochester, N. Y, A. A. Doolittle, Western High School. W. R. Dudley, Leland Stanford, Jr. University, Palo Alto, Cal, Owen Durfee. Fall River, Mass. Basil H, Dutcher, Fort Hancock, N. J. William Dutcher, 525 Manhattan Ave., New York, N, Y. Jonathan Dwight, Jr., 2 E. 34th St., New York, N, Y. L. L. Dyche, Lawrence, Kansas. Daniel G. Elliot, Field Columbian Museum, Chicago, 111. Alexander W. Evans, 2 Hillhouse Ave., New Haven, Conn, W. H. Evans, U. S. Dept. Agriculture. B. W. Evermann, U. S. Fish Commission. Walter Faxon, Museum of Comparative Zoology, Cambridge, Mass* A. K. Fisher, U. S. Dept. Agriculture. James H. Fleming, 267 Rusholme Road, Toronto, Ontario. James Fletcher, Central Experiment Farms, Ottawa, Canada, Robert Fletcher, Army Medical Museum. J. M. Flint, U. S. National Museum. C. W. Franzoni, 605 I St., N. W. Cecil French, 718 12th St., N. W. B. T. Galloway, U. S. Dept. Agriculture. Miss May Gannett, 1881 3rd St. N. W. P. D. Gardner, U. S. Dept. Agriculture, N^ames of Hfcmbers-, xv O. K. Gilbert, U. S. Geological Survey. "Theo. N. Gill, Smithsonian Institution. O. H. Girty, U. 8. Geological Survey. E. L. Goldsboro, U. S. Fish Commission, E. A. Goldman, U. S. Dept. Agriculture. Robert H. Gordon, Cumberland, Md. Edward L. Greene, Catholic University. Miss Emily Gregory, Hearst Cathedral School. David Griffiths, U. S, Dept. Agriculture. George Bird Grinnell, 346 Broadway, New York, N. Y. Chas. E, Hanaman, Troy, N. Y. Chas. W. Hargitt, Syracuse University, Syracuse, N. Y. J. Culver Hartzell, Illinois Wesleyan University, Bloomington, 111. J. B. Hatcher, Carnegie Museum, Pittsburg, Pa. William P. Hay, Howard University. Samuel Henshaw, Museum of Comparative Zoology, Cambridge. Mass* Glenn W. Herrick, Agricultural College, Miss. Robert T. Hill, U. S. Geological Survey. Albert S. Hitchcock, U. S. Dept. Agriculture. N. Hollister, Delavan, Wis. Miss Nina G. Holton, U. S.!Dept. Agriculture, A. D. Hopkins, U. S. Dept. Agriculture. Ralph S. Hosmer, U. S. Dept. Agriculture. L. O. Howard, U. S. Dept. Agriculture. Reginald Heber Howe, Jr., Longwood Ave., Brooklrne, Mass> Arthur H. Howell, U. S. Dept. Agriculture. C. H. Howland, 835 Vermont Ave. Mrs. E. B, Jones, 227 Preston St., W. Baltimore, Md. Mrs. Marion S. F. Jouy, U. S. National Museum. Sylvester D. Judd, U. S. Dept. Agriculture. T. H. Kearney, Jr., U. S. Dept. Agriculture. E. T. Keim, Am, District Telegraph Co., Denver, Colo, W. C. Kendall, U. S. Fish Commission. Harris Kennedy, 286 Warren St., Roxbury, Mass. A. F. A. King, 1315 Mass. Ave. Frank C. Kirkwood, 1811 Maryland Ave., Baltimore, Mds F. H. Knowlton, U. S. National Museum. Gustave Kohn, 14 Carondelet St., New Orleans, La. Jacob Kotinsky, U. S. Dept. Agriculture. Oscar Loew, Imperial University, Tokio, Japan. S. Ward Loper, Middletown, Conn. J. A. Loring, Oswego, N. Y. Frederic A. Lucas, U. S. National Museum. M. W. Lyon, Jr., U. S. National Museum. John M. S. McDonald, 694 Jefferson St., Milwaukee, Wis. W. J. McGee, Bureau of Ethnology. R. E. B. McKenney, U. S. Dept. Agriculture. B. P. Mann, 1918 Sunderland Place. xvi Names of Members. J. Woodward Manning, 1146 Tremont Bldg., Boston, Mass, C. L. Marlatt, U. S. Dept. Agriculture. M. C. Marsh, U. S. Fish Commission. Wm. R. Maxon, U. S. National Museum. Edgar A. Mearns, Fort Yellowstone, Wyo. W. F. Mercer, Ohio University, Athens, Ohio. C. Hart Merriam, U. S. Dept. Agriculture. W. A. Merritt, U. S. National Museum. Maynard W. Metcalf, Womans College, Baltimore, Md. Gerrit S. Miller, Jr., U. S. National Museum. H. F. Moore, U. S. Fish Commission. E. L. Morris, Central High School. James G. Needham, Lake Forest College, Lake Forest, 111. Aven Nelson, University of Wyoming, Laramie, Wyo. Elias Nelson, 720 Custer St., Laramie, Wyo. E. W. Nelson, U. S. Dept. Agriculture. J. B. Nichols, 1331 N St., N. W. Arthur H. W. Norton, San Antonio, Tex. Harry C. Oberholser, U. S. Dept. Agriculture. H. V. Ogden, 141 Wisconsin St., Milwaukee, Wis. Henry W. Olds, U. S. Dept. Agriculture. W. A. Orton, U. S. Dept. Agriculture. Herbert Osborn, Ohio State University, Columbus, Ohio. H. F. Osborn, American Museum Natural History, New York, N. Y. Wilfred H. Osgood, U. S. Dept. Agriculture. Frederick D. Owen, Room 24, War Department. Miss Juliette A. Owen, 306 North 9th St., St. Joseph, Mo. T. S. Palmer, U. S. Dept. Agriculture. William Palmer, U. S. National Museum. L. H. Pammel, Ames, Iowa. Miss Juliet Patten, 2212 R St. N. W. Mrs. Flora W. Patterson, U. S. Dept. Agriculture. A. J. Pieters, U. S. Dept. Agriculture. Gifford Pinchot, U. S. Dept. Agriculture. Charles L. Pollard, U. S. National Museum. E. A. Preble, U. S. Dept. Agriculture. D. W. Prentiss, Jr., 1100 14th St. N. W. W. W. Price, Alta, Cal. Miss M. J. Rathbun, U. S. National Museum. Richard Rathbun, U. S. National Museum. Albert M. Read, 1140 15th St. N. W. James A. G. Rehn, Academy of Natural Sciences, Phila., Pa C. W. Richardson, 1102 L St. N. W. Miss Harriet Richardson, 1848 Wyoming Ave. C. W. Richmond, U. S. National Museum. Robert Ridgway, U. S. National Museum. J. H. Riley, U. S. National Museum. William 0. Rives, 17231 St. N. W. .\'.", have already appeared.* As there stated by ('apt. Robinson, the island "lies about midway between L:i(iu:iira and Trinidad, and only some 17 miles dis tant from the nearest point of the Venezuelan coast. Its ^reat est length from east to west is l > miles, and its greatest breadth from north to south 20$ miles. It consists of two portions* connected by a narrow isthmus \\] miles in length; the western boinjx an irregular quadrilateral 1'.' miles loni; by l> miles broad. and the eastern a pentagon some 20 miles across. In the west ern portion some almost barren peaks rise to a height of 2300 feel". Of 10 species of mammals noted by ('apt. Robinson as oecurrin!'; on the island, specimens of o were preserved, and ;' of these have since been described as new. Mr. (Mark's col lections contained 1 ;" specimens of mammals, represent ins: ^ . r. s N : ,, MUS . \viii. p. r.n os;>. P i. \\\,u. 81-BIOL, 800. WAi. VOL. XV, IW. 92 Allen Mammals of Margarita Island. species, of which 4 had not previously been recorded from the island, and one, a squirrel, proves to be new. As might be ex pected, the native mammalian fauna appears to have been de rived almost wholly from South America. It is interesting to note that the influence of the insular environment has been to fade out, to a greater or less degree, the colors of several of the species as compared with the colors of their nearest repre sentatives of the continent and of Trinidad. The greater area of the island of Trinidad and the heavier growth of vegetation, with the consequent increase in humidity seem to have afforded conditions more like those of the mainland, and hence there is not so pronounced a tendency to paling out of the deeper colors such as is seen notably in the Murine opossum and the squirrel of Margarita Island, and also in several of the birds from the island, which have been described as new. From its compara tively small area, partial sterility, and greater isolation, Mar garita is more open to the bleaching effects of sun and sea- wind. Whether the bleaching of colors in the case of certain pale insular forms is due to the direct action of exposure to the elements, or is acquired as an adaptation to a more or less barren environment, is an interesting question in this connection, and it is possible that both causes often interplay to produce a com bined result. A list of the mammals at present known from Margarita follows: ' i. Marmosa robinsoni Bangs. Robinson Murine Opossum. Didelphys murina Robinson, Proc. U. S. Nat. Mus., XVIII, p. 651, 189C. Marmosa robinsoni Bangs, Proc. Biol. Soc. Washington, XII, p. 95-96, 1898. Five specimens of this opossum were collected in 1895, by Capt. Rob inson, and are characterized as forming a pallid insular species. > 2. Caluromys philander (Linnaeus). Woolly Opossum. A single specimen was brought in to Mr. Clark at El Valle, on July 12, 1901. This was an adult female within whose pouch were three young clinging to the mammas, and as yet hardly more than 3 cm. in total length. I have been unable to compare the adult specimen with Allen Mammals of Margarita Island. 93 others from the adjacent mainland, but in measurements, the skull agrees closely with the dimensions of an adult female skull from San Julian, given by Messrs. Robinson and Lyon (Proc. U. S. Nat. Mus., XXIV, p. 138, 1901); I have compared the skull also with that of Oalu- romys trinitatis (Thomas) from Caura, Trinidad, and find but little ap preciable difference, although the nasal bones and the palate are a trifle broader in the Margarita specimen, and the jaw is a trifle heavier. In color, the skins of the Margarita and Trinidad animals are nearly iden tical, though Mr. Clark's specimen is perhaps a little paler, lacking the faint chestnut tint dorsally and being a shade paler below than the Trinidad species, as might perhaps be anticipated. The material at hand, however, would not at all warrant a separation. 3. Sciurus nesseus* sp. nov. Margarita Squirrel. Sciurus cestuans hoffmanni Robinson, Proc. U. S. Nat. Mus., XVIII, p. 651, 1896. Type from El Valle, Margarita Island, Venezuela, No. 619, collection of Glover M. Allen, 9 adult. Collected July 8, 1901, by Austin H. Clark. General characters. Dorsal surface of body light ochraceous, sprinkled with black; terminal three-fourths of tail uniform ochraceous. Ven tral surface of body bright orange-rufous. Coloration. Dorsal surfaces of head, body, and proximal fifth of tail, grizzled ochraceous, or ochraceous-buff, and black; the separate hairs are plumbeous at the base for a space of about 4 mm., then comes a band of about the same width of ochraceous-buff, followed by a band of black about one-half as wide, then a second band of ochraceous-buff about 2 mm. in width, and a fine black tip. The general effect is nearly the same as that seen in a bright skin of the red squirrel (Sciurus hud- sonicus loquax) of southern New England in summer pelage. Sides of the body similar, but with less black; cheeks, and an indistinct eye ring, nearly pure ochraceous; dorsal surface of forearms and feet slightly brighter than the back. Ventral surface of body bright orange-rufous, less intense on the lower sides of the limbs. The terminal three-fourths of the tail is uniform ochraceous above, due to the entire concealment of the two small black rings, one at the base and one about half-way up on each hair; ventrally the tail shows a distinct border of orange-ru fous, with the median area grizzled black and ochraceous. The whiskers of the type are nearly all pale ochraceous. Remarks. The type is unaccompanied by measurements, but the dry skin gives the following: length, 374 mm.; tail, 159 mm.; hind foot, 56.2 mm.; ear, 16 mm. Unfortunately the skull was completely shat tered. belonging to an island, insular, 94 Allen Mammals of Margarita Island. This very distinctly characterized species appears to be nearly related to Sciurits griseogena (Gray) of the neighboring coast region, rather than to the 8. wstuans group. Specimens of 8. griseogena collected by Capt. Robinson at San Julian, Venezuela, in July, 1900, match it closely in the bright orange-rufous of the underparts, but differ very greatly in the color of the dorsal surface, in which black is the predominating color, due to the partial reduction of the ochraceous bands on the body hairs, whereas in the Margarita squirrel the opposite change has taken place, and the black has given place to the ochraceous. Moreover, the entire tip of the tail is black in 8. griseogena, and bright ochraceous in S. nescBus. .Mr. Clark obtained several of these squirrels in the woods on the mountain, which is just back of El Valle, but unfortunately was able to preserve only the single specimen collected near the town. This speci men will be deposited with the Bangs collection in the Museum of Com parative Zoology at Cambridge. 4. Mus alexandrinus Geoff roy. Roof Rat. A single specimen of this rat was preserved by Mr. Clark, who states that it was trapped in one of the houses at El Valle. Capt. Robinson mentions "a rat (Mus)" as occurring on the island, but the species is not recorded. 5. Mus musculus Linnaeus. House Mouse. Both Capt. Robinson and Mr. Clark found the house mouse common in the open dwellings on the island. Four specimens preserved by Mr. Clark, appear to be considerably smaller than the species as found in New England. Two of these, in formalin, measure about 134 mm. and 138 mm. in length, with tails 68 mm. and 72 mm. respectively; both seem to be adult, and average 25 mm. smaller in total length than six specimens from eastern Massachusetts, while the tails are also slightly shorter. 6. Loncheres sp. Spiny Rat. Loncheres Robinson, Proc. U. 8. .Nat. Mus., XVIII, p. 651, 1896. A male specimen of a spiny rat was taken by Capt. Robinson on the island, July 9, 1895. It was much injured, and in its present condition, lacks the tail and part of the skin of the head. Mr. Oldfield Thomas has pronounced it as being near L. punctatus. Allen Mammals of Margarita Island. 95 7. Lepus margaritae Miller. Margarita Rabbit. Lepus brasiliensis Robinson, Proc. U. S. Nat. Mus., XVIII, p. 651, 1896. Lepus margaritcB Miller, Proc. Biol. Soc. Washington, XII, p. 97, April 30, 1898. Capt. Robinson found these rabbits plentiful on the island in 1895 and mentions (Proc. U. S. Nat. Mus., XXIX, p. 162, 1901) that they were "split and dried like cod fish and brought by the boat load to La Guaira for sale". But two individuals were noted near El Valle by Mr. Clark, who states that the natives were in constant pursuit of them. 8. Rhogeessa minutilla Miller. Vesperugo parvulus Robinson, Proc. U. S. Nat. Mus., XVIII, p. 651, 1896 (not Rhogeessa parvula H. Allen). Rhogeessa minutilla Miller, Proc. Biol. Soc. Washington, XI, p. 139, May 13, 1897. Rhogeessa minutilla Miller, N. Amer. Fauna, No. 13, p. 125, Oct. 16, 1897. The single specimen from which this species was described was col lected on the island by Capt. Robinson, July 8, 1895. The second known specimen was captured at El Valle, late in July, 1901, by Mr. J. R. Johnston, of Mr. Clark's party. He swept it into an insect net, think ing it was a large moth, as it flew past in the early morning twilight. Mr. Gerrit S. Miller, Jr., who has examined this specimen, writes that it fully bears out the characters of the type as given in the original de scription. The measurements taken from the specimen before skinning and after an. immersion of several months in formalin, are: length, 66.5 mm.; tail, 31 mm.; ear, 11.5 mm.; forearm, 25 mm.; tibia, 11.4 mm. The specimen is now in the collection of the United States National Museum. 9. Peropteryx sp. Sac- winged Bat. A single, much shattered specimen was obtained by Mr. Clark. It was a female and was shot in a large cave on the hill southwest of El Valle, July 13, 1901. It is quite possible that it represents an unde- scribed species, but more and better material is necessary to decide this point. It is intermediate in size between P. canina and P. trinitatis, the tibia, in the dried specimen skinned out from formalin, measuring 17 mm., and the forearm 40 mm. The skull is slightly smaller than that of P. canina though slightly heavier than that of P. trinitatis. The muz zle is short and broad as compared with that of P. trinitatis and slightly 96 Allen Mammals of Margarita Island. narrower than that of P. canina. It differs notably from both in having the first premolar in contact with the second premolar on the one side and the canine on the other, instead of separate from both, as in the two other species. There are also other minor differences, and the den tition throughout is rather heavier than in the other two forms men tioned. 10. Molossus obscurus E. Geoffrey. Dusky Molossus. Three specimens were obtained by Mr. Clark, one of which, a young one and nearly naked, was found scrambling about on the roof of a house at El Valle, July 14th. The two adult examples were shot July 11, 1901, as they were hanging in a cave under a large rock. They show some slight individual variation, the male being the larger, and the fur a dark sooty color, with the bases of the hairs nearly white. The female is brighter, with a brownish cast, and the bases of the hairs a light fawn. Mr. Gerrit S. Miller, Jr., who also examined these specimens, writes that they are not referable to M. pygmceus of Curacao, but appear to be obscurus as now understood. ii. Micronycteris megalotis (Gray). Common Long-eared Bat. This species is probably common on the island. Capt. Robinson men tions taking a pair on July 12, 1895, at El Valle, from a slight hollow under a boulder. Two were secured in the large cave already men tioned, on July*13, 1901, by Mr. Clark. 12. Qlossophaga soricina (Pallas). Shrew-like Glossophaga. Two specimens of a Glossophaga obtained by Mr. Clark are referable to this species as at present understood. One is an adult male, and the other a young individual, still possessing its milk teeth. Both were shot July 13, 1901, in the large cave on one of the hills back of El Valle. A third specimen of Glossophaga, an adult female, was obtained in this cave on the same date, and though somewhat battered, seems rather distinct from the specimens of G. soricina, and may represen an unde- scribed species. Further material is necessary, however, to determine this point. It differs from G. soricina most notably, in the possession of a larger nose-leaf, a slightly longer, higher and broader skull, and in having the 1st and 2nd upper incisors separate from each other and the canines instead of touching, as in G. soricina. There are also a few other minor differences. The specimen is of a dull chestnut or tawny-brown, a color which I have not observed in the specimens of G. soricina ex amined. Allen Mammals of Margarita Island. 9*7 13. Cebus fatuellus (Linnaeus). Brown Capuchin. Cebus apella Robinson, Proc. U. S. Nat. Mus., XVIII, p. 651, 1896. A single monkey obtained on the island, July 13, 1895, seems refera ble to this species. It appears to be not uncommon, but is rather shy. Mr. Clark saw a single one in the thick growth on the mountain which rises back of El Valle, but he was unable to procure it. One or two were also seen, kept as pets by the natives. [In addition to the foregoing, both Capt. Robinson and Mr. Clark men tion a deer as occurring on the island, but no specimens were obtained. In this connection, it is of interest to recall Humboldt's statement, with regard to the small islet of Cubagua, lying between Margarita and the mainland. He says, "At Maniquarez [=Manicuaro on the mainland] we met with some Creoles, who had been hunting at Cubagua. Deer of the small breed are so common in this uninhabited islet, that a single person may kill three or four in a day. I know not by what accident these animals have got thither, for Laet and other chroniclers of these countries, speaking of the foundation of New Cadiz [a small town on Cubagua] , mention only the great abundance of rabbits. The venado of Cubagua belongs to one of those numerous species of small American deer, which zoologists have long confounded under the vague name of cervus mexicanus Its color is a brownish red on the back, and white under the belly; and it is spotted like the axis". (Humboldt: Personal narrative of Travels to the Equinoctial Regions of the New Continent, during the years of 1799-1804, p. 424 of 1815 English translation). I am informed that Cubagua at the present time would hardly support as much game as this passage implies, but doubtless a hundred years have wrought a change, or the writer's remarks may possibly have referred in part to Margarita itself.] VOL. XV, PP. 99-100 APRIL 25, 1902 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A NEW RIBES FROM NEW MEXICO. BY T. D. A. COCKERELL. Among the material collected during the past season in the neighborhood of Las Vegas I find a new variety of Ribes lep- tanthum Gray, which may be known as follows: Ribes leptanthum veganum, var. nov. Shrub about 2 to 3 m. high, with pale gray bark; on the branches the bark is almost silvery, on the twigs it becomes reddish-brown; prickles solitary on the nodes, or more abundant, long, slender, pale fer ruginous; leaves shaped about as in R. rotundifolium, the deeply crenu- late segments short and rounded; leaves and petioles with glandular hairs, petioles also with a little tomentum; peduncles short, flowers in pairs or solitary; calyx-lobes spreading, white, more or less margined with pink; calyx-tube pale greenish, cylindrical, stout (3mm. wide); tube about 6 mm. long, lobes about 5^ mm.; petals slightly over 3 mm., little more than half length of calyx-lobes, white tipped with pink; stamens not reaching to end of petals; filaments bright crimson; pistil green, only reaching to bases of stamens. Fruit glandular-hispid. Found along the Gallinas River, New Mexico, from about two miles below Las Vegas to Las Valles, from 5800 feet to 6300 feet altitude. Flowers during the last two weeks of April, and is freely visited by Andrena porterce. Ckll., and other bees. Typical specimens in U. S., Na- 22 BIOL. Soc. WASH. VOL. XV, 1J02 (99) 4- 1902 100 Cocker ell A New Ribes from JVeto Mexico. tional Museum; the material collected April 20, 1901, may be taken as the type. I thought this was a distinct species, having only a description of R. kptanthum, but Mr. Coville considers it a glandular-pubescent phase of kptanthum. The only other Ribes observed in the region of R. L vega- num is R. longiflorum, Nutt. (at Las Valles). Going north of Las Vegas towards San Ignacio, the beginning of the transition zone is marked by the abundance of R. cereum. Passing up Sapello Canon to Beulah, in the Canadian zone, R. cereum vanishes, and we have instead R. lentum, R. irriguum, and a species discovered by my wife (only obtained in fruit) which Mr. Coville thinks is R. mogollonicum, Greene. Going still up ward, on the top of the Las Vegas Range, at 11,000 feet (Hudsonian zone) we meet with only one of the Beulah Ribes R. lentum; while under the spruce trees grows one not seen before, identified by Mr. Co ville as R. wolfii. VOL. XV, PP. 101-116 JUNE 2, 1902 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON SYNOPSIS OF THE NORTH AMERICAN SPECIES OF SIGMODON. BY VERNON BAILEY. The present brief synopsis of the North American species of Sigmodon, commonly known as cotton rats, is the outgrowth of an attempt to work out the distribution of the several forms occurring in the state of Texas. In this connection it became necessary to make a critical study of the group, which has re sulted in the recognition of 21 apparently valid forms, 10 of which are here described for the first time. A long discarded name, berlandieri, is reinstated, and two recent names are placed in synonymy. Of the 21 species and subspecies, 9 occur in the southern United States; the others are scattered over Mexico and Central America. Of the four species described from northern South America I have seen no specimens. The names and type localities are as follows: Sigmodon sanctcemar- tce Bangs, from Pueblo Viejo, Colombia; Sigmodon bogotensis Allen, from Plains of Bogota, Colombia; Sigmodon peruanus Allen, from Trujillo, Peru; Sigmodon simonsi Allen, from ]Eten, coast of northwest Peru. The material on which the conclusions are based consists of 23-BioL. Soo. WASH. VOL. XV, 1903, (101) 102 Bailey North American Species of Sigmodon. about 1000 specimens in the Biological Survey collection and the private collection of Dr. C. Hart Merriam, both in the United States National Museum, besides a number of types and topotypes from other sources, for the use of which I am in debted to Dr. F. W. True, executive curator, and Mr. Gerrit S. Miller, Jr., assistant curator of mammals, in the U. S. National Museum; to Dr. J. A. Allen, curator of mammals and birds, American Museum Natural History; and to Mr. Outram Bangs, curator of mammals, Museum of Comparative Zoology. The North American species of Sigmodon fall naturally into two groups, which may be designated the hispidus group, and the hairy- tailed group. The former, of which hispidus was the first named species, is characterized by a semi-naked tail, the coarse annulations of the epidermis being scarcely obscured by the short scattered hairs. The forms of this group inhabit the Lower Sonoran and Tropical zones, are widely distributed, with an almost continuous range, and, with the one exception of an overlapping species (alleni) are here treated as subspecies of hispidus. In general the pelage is coarse and hispid, usually much more so in summer than in winter, but this character varies considerably with the various forms. The group includes hispidus, littoralis, spadicipygus, texianus, berlandieri, eremi- cus, arizonce, mascotensis, tonalensis, major, toltecus, satura- tus, microdon, borucce, and alleni. The hairy-tailed group, of which fulviventer was the first known species, is characterized by a moderately hairy tail, the short, bristly hairs almost or quite concealing the underlying annulations. In most of the species the pelage is softer than is usual in the hispidus group. The group is not distinguished by cranial characters and the species differ as widely among themselves as they do from those of the hispidus group. So far as ascertained they inhabit the Upper Sonoran and Transi tion zones in widely scattered mountains in Mexico and adjoin ing parts of the United States. In most cases they appear to be isolated by surrounding areas belonging to lower zones. Most of the species seem to have a very limited distribution and are distinguished by strongly marked characters. The group includes fulviventer ', melanotis, minimus, ochrognathus, leucotis, alticola and amoles, all but the last of which are treated as full species. Bailey North American Species of Sigmodon. 103 Key to Species. 1. Tail thinly haired; skull relatively long and narrow. 2. Size large, hind foot averaging 32 or more. 3. Upperparts mainly warm brown. 4. Upperparts uniformly brown, size larger - - Mspidus. 4 / '. Upperparts brownest on rump, size smaller - spadicipygus. 3'. Upperparts light gray, blackish or yellowish gray. 4. Upperparts dark gray; much black and no brown littoralis. 4'. Upperparts light buffy or yellowish gray. 5. Upperparts pale buffy gray. 6. Pelage fine and not very hispid - -.' - eremicus. 6'. Pelage coarse and very hispid, especially in summer. 7. Size large, hind foot averaging 40 major. 7 / . Size medium, hind foot not over 36. 8. Hind foot about 33 - * : "- - berlandieri. 8'. Hind foot 35-36 - . - "' - - '..'-. arizonw. 5'. Upperparts dark buffy or yellowish gray. 6. Upperparts buffy gray; hind foot about 34 mascotensis. 6'. Upperparts yellowish gray ; hind foot 41 - tonalensis. 2'. Size small, hind foot averaging less than 32. 3. Upperparts buffy gray - - . - - : " texianus. 3'. Upperparts tawny or dull brownish gray. 4. Upperparts tawny or dull yellowish. 5. Audital bullae wide and flattened - - - borucce. 5'. Audital bullae high and narrow, rounded or elongated. 6. Tail short, averaging 103; audital bullse narrow saturatus. 6'. Tail longer, averaging 112; audital bullae short and rounded alleni. 4'- Upperparts dull brownish gray. 5. Belly white; feet clear gray .... toltecus. 5'. Belly grayish buff; feet yellowish brown - microdon. V. Tail well haired; skull short and wide. 2. Upper half of outer surface of ear black. 3. Size large, hind foot 30-33. 4. Upperparts grizzled buffy; lowerparts ochraceous ochrogaster. 4'- Upperparts grizzled tawny; belly rich ferruginous melanotis. 3'. Size small, foot 28; upperparts gray; belly buffy minimus. 2'. Outer and inner surface of ears whitish or buffy gray. 3. Ears buffy gray; nose conspicuously ochraceous ochrognathus. 3'. Ears light gray or whitish; nose not conspicuously ochraceous. 4. Ears silvery gray; belly white ... - leucotis. 4'. Ears darker gray; belly buffy. 5. Upperparts rich tawny gray - alticola. 5'. Upperparts dull tawny gray - - - amoks. 104 Bailey North American Species of Sigmodon. Sigmodon hispidus Say and Ord. Sigmodon hispidus Say and Ord, Jour. Acad. Nat. Sci. Phila., IV, pt. ii, 354, 1825. Type locality. St. Johns River, Florida. General characters. Pelage full and soft in winter, coarser and slightly hispid in summer. Color. Upperparts rich umber brown, darkened by black underfur and long black hairs; belly whitish, smoky gray or dull buffy brown; feet dull rusty brown; tail blackish, slightly lighter below. Skull. Relatively short and wide with well arched upper outline. Measurements. Average of 8 adults from Riceboro, Georgia: total length 256; tail 106; hind foot 32. Skull (No. 45,169, from Riceboro, Georgia): basal length 32; nasals 14; zygomatic breadth 21; mastoid breadth 15; alveolar length of upper molar series 6.7. Distribution. North Carolina to northern Florida and west to southern Louisiana, in Austroriparian zone. Specimens examined. Total number 125, from the following localities: North Carolina: Raleigh 33, Elkin 1. South Carolina: Georgetown 6, Beaufort 27, Society Hill 1. Georgia: Augusta 5, Butler 5, Riceboro 20, Savannah 1, Nashville 1. Florida: San Mateo 1, Gainesville 3, Chattahoochee 3, Milton, 2. Alabama: Greensboro 5, Catherine 1, Gallion 4. Mississippi: Bay St. Louis 3. Louisiana: Houma 1, Lafayette 3, Avery 2. Sigmodon hispidus littoralis Chapman. Sigmodon hispidus littoralis Chapman, Bull. Am. Mus. Nat. Hist. II, 118, June 7, 1889. Type locality. East Peninsula, opposite Micco, Florida. General characters. Slightly larger and darker than hispidus, with coarser more hispid pelage. Color. Upperparts dark sooty gray coarsely lined by the dark and light hairs, buffy wash of the main cpat obscured by the long black hairs and black underfur; feet and tail dark gray or blackish; upper surface of tail quite black. Skull. Relatively larger, narrower and less arched than in hispidus. Measurements. Average of 6 adult topotypes: total length 284; tail 114; hind foot 32. Skull (No. 70,808, $ ad.): basal length 33; nasals 14; zygomatic breadth 21; mastoid breadth 14.6; alveolar length of upper molar series 6.8. Distribution. Eastern part of the peninsula of Florida, from Lake Harney to the Everglades. Specimens examined. Total number 119 from the following localities in Florida: Peninsula opposite Micco (type locality) 42, Lake Harney 29, Bailey North American Species of Sigmodon. 105 Titusville 17, Sebastian 9, Lake Kissimmee 2, Kissimmee River 5, Mul let Lake 2, Eden 4, Miami 7, Everglade 2. Sigmodon hlspidus spadicipygus Bangs. Sigmodon hispidus spadicipygus Bangs, Proc. Bost. Soc. Nat. Hist. XXVIII, 192, March, 1898. Type locality. Cape Sable, Florida. General characters. Like Jiispidus but smaller, loss uniformly brown, and with lighter dentition; pelage soft as in hispidus. Color. Upperparts dark brownish gray, becoming warm brown only on rump; belly grayish brown or whitish; feet dull brownish; tail black ish, slightly lighter below. Skull. Similar to that of hispidus but smaller with conspicuously lighter dentition. Measurements. Type: total length 279; tail 98; hind foot 31-33 [from original description]. Skull (No. 4485, $ ad., Bangs Collection, from Flamingo, Fla.): basal length 30; nasals 12; zygomatic breadth 19.5; mastoid breadth 14.5; alveolar length of upper molar series 6.2. Distribution. The extreme southern part of the peninsula of Florida. Specimens examined. Total number 8 from three localities: Cape Sable, Flamingo and Planter, Florida. Remarks. In external as well as cranial characters this form differs more from littoralis than from true hispidus with which Mr. Bangs sup posed it to intergrade along the west coast of the peninsula of Florida. Sigmodon hispidus texianus (Aud. and Bach.). Armcola texiana Aud. and Bach. Quad. N. Am. Ill, 229, 1853. Type locality. Brazos River, Texas. General characters. Slightly smaller and paler than typical hispidus. Color. Upperparts grayish brown or dark buffy gray; belly grayish or buffy white; feet clear gray; tail gray with blackish line along the top. Colors darkest in fresh fall pelage, lightest in faded spring pelage. Skull. Differing from the skull of hispidus mainly in smaller size, less arched outline, and smaller, more rounded audital bullae. Measurements. Average of 4 adults from Richmond, Texas: total length 248; tail 103; hind foot 30. Skull (No. 97,014, $ ad.): basal length 30; nasals 13; zygomatic breadth 21; mastoid breadth 14; alveolar length of upper molar series 6.4. Distribution. Approximately the eastern half of Texas, westward to Vernon and San Antonio, and northward to Cairo, Kansas. Specimens examined. Total number 77 from the following localities: Texas: Richmond 4, Matagorda 20, Velasco 1, Port Lavaca 6, Talley Island 6, Corpus Christi 4, San Antonio 18, Boerne 1, Fisher Co. 1, Ver non 2, Gainesville 1. 106 alley North American Species of Sigmodon. Oklahoma:: Ft. 'Reno 2, Ft. Cobb 1, Orlando|2, Alva 1. Kansas: Cairo 1. Remarks. 8. texianus is an intermediate form between the dark, rich hispidus and the very pale berlandicri, grading both ways in accordance with the change from humid to arid climate. Specimens from Okla homa and Kansas are almost equally near to hispidus, while San Antonio and Corpus Christi specimens could just as well be referred to berlan- dieri. Sigmodon hispidus berlandieri Baird. Sigmodon berlandieri Baird, Proc. Acad. Nat. Sci. Phila., VII, 333, 1855. /Sigmodon hispidus pallidus Mearns., Proc. U. S. Nat. Mus., XX, 504, March 15, 1897. From El Paso, Texas. Type locality. Rio Nazas, Coahuilla, Mexico. General characters. About the size of hispidus or averaging a little smaller and much paler; pelage full and soft in winter, thin and very hispid in summer. Color. Upperparts light buffy or ashy gray; nose not conspicuously yellowish; belly white; feet light gray; tail distinctly bicolor, light gray below, blackish above. Skull. Similar to the skull of texianus; differing from that of hispidus in less arched outline and smaller, more rounded bullae. Measurements. (No. 58,074, $ ad., from Jimulco, Coahuila, Mexico): total length 232; tail 105; hind foot 33. Average of 5 adults from Del Rio, Texas: total length 256; tail 113; hind foot 32.5. Skull (No. 58,073, $ ad., from Jimulco, Coahuila): basal length 30.5; nasals 12.3; zygo- matic breadth 19; mastoid breadth 13.4: alveolar length of upper molar series 6.5. Distributian. Rio Grande and Pecos valleys from Brownsville to El Paso, Texas, and Carlsbad, New Mexico; south to southern Jalisco, Mexico. Specimens examined. Total number 200, from the following localities: Texas: El Paso and vicinity 7, Guadalupe Mts. 4, Langtry 3, Painted Caves 3, Del Rio 7, Sycamore Creek 3, Eagle Pass 7, Santa Tomas 27, Sauz 2, Brownsville 20, Ft. Clark 3. New Mexico: Carlsbad 3. Chihuahua: Juarez 1, Chihuahua City 18, Santa Rosalia 8, Presidio del Norte 1. Coahuila: Rio Nazas 1, Jimulco 2. Nuevo Leon: Doctor Arroyo 1. Tamaulipas: Nuevo Laredo 7, Matamoras 11, Camargo 26, Victoria, 2, Jaumave 6, Soto la Marina 9. San Luis Potosi: Hda. La Parada 3, Villar 1, Jesus Maria 1, Molino 1. Queretaro: Tequisquiapam 1. Zacatecas: Valparaiso Mts. 3. Bailey North American Species of Sigmodon. 107 Jalisco: Huejuquilla 1, Lagos 2, Atemajac 1, Ameca 1, Chacala 1, Zacoalco 3, Zapotlan 2. Remarks. 8. h. berlandieri is a pale desert form of the hispidus group ranging over a wide extent of extremely arid country in western Texas and northeastern Mexico and intergrading with forms to the east, south and west. Prof. Baird described it from specimens "collected between San An tonio and El Paso by Mr. Clark and in Northern Mexico by Lieut. Couch". The skulls of these two specimens still in the U. S. National Museum show no important subspecifiyc characters but are of importance geographically. The one from between San Antonio and El Paso ap parently, but not positively, (see old museum catalogue) came from Presidio del Norte, and the other, No. 566, from Rio Nazas, southern Coahuila (Baird, Mamm. N. A. p. 505). No type was designated but as Baird gave full measurements of No. 566 this specimen is generally con sidered the type. Specimens in the Biological Survey collection from Jimulco (about 30 miles south of the short Coahuila section of the Rio Nazas), from near the mouth of the Pecos and from El Paso, Texas, are indistinguishable and are all clearly referable to this 1 pale gray form. Typical specimens of berlandieri are no nearer to typical texianus than that species is to hispidus, so that if texianus is to be recognized, berlan dieri must also be. Sigmodon hispidus eremicus Mearns. Sigmodon hispidus eremicus Mearns, Proc. U. S. Nat. Mus. XX, 504, advance sheets, March 15, 1897. Type locality. Cienega Well, Sonora, Mexico, on east bank of Colora do River, 30 miles south of United States and Mexican boundary line. General characters. Size slightly larger than hispidus; pelage fine and rather lax and soft, colors pale. Color. Upperparts pale yellowish gray; belly whitish; feet light gray ; tail brown above, gray below. From berlandieri it differs in slightly more buffy upperparts and less of the mixture of black hairs. Skull. Heavy, rough and much ridged, even in specimens that are not very old; coronoid process of jaw short and wide; bullae short and rounded as in berlandieri. Measurements. Type: total length 280; tail 128; hind foot 34. Skull of type: basal length 30.6; nasals 13.3; zygomatic breadth 20.3; mastoid breadth 14; alveolar length of upper molar series 6.6. Distribution. Along both sides of the lower Colorado River. Specimens examined. Total number 49 from the following localities: Sonora: Cienega Well, south of Mexican boundary on east side of Col orado River 19, Colorado River at mouth of Hardee River 11. California: Ft. Yuma 19. 108 Bailey North American /Species of Sigmodon. Sigmodon hispidus arizonae Mearns. Sigmodon hispidus arizonae Mearns, Bull. Am. Mus. Nat. Hist. II, 287, Feb. 21, 1890. Type locality. Ft. Verde, Arizona. General cliaracters. Size large; pelage coarse and hispid; colors pale. Color. Upperparts light buffy gray as in berlandieri; nose slightly yellowish; belly white; feet and tail as in berlandieri and eremicus. Skull. I have not been able to get a skull of this species for exam ination, but the original measurements show it to have been larger than the average eremicus skull. Measurements. Type (from original description): total length 320; tail 121. Hind feet of two topotypes taken from dried specimens, 35 and 36. Skull (from original description): total length 40; nasals 15.9; zygo- matic breadth 23; alveolar length of upper molar series 7.1. Remarks. Without seeing skulls of this form I am in doubt as to its status and affinities, but the two skins from the original topotype series, kindly sent for examination by Dr. Allen, indicate a larger form than eremicus with more hispid pelage. Sigmodon hispidus mascotensis Allen. Sigmodon mascotensis Allen, Bull. Am. Mus. Nat. Hist. IX, 54, March 15, 1897. Sigmodon colimae Allen, Bull. Am. Mus. Nat. Hist. IX, 55, March 15, 1897. From Plains of Colima, Colima, Mexico. Type locality. San Sebastian, near Mascota, Jalisco, Mexico. General characters. Slightly larger than hispidus or berlandieri, with especially longer tail and larger hind foot; pelage in summer very coarse and hispid. Color. Upperparts dark buffy gray with bright buffy or ochraceous nose; belly yellowish white; feet dull brownish gray; tail brownish gray below and brownish black above, less distinctly bicolor than in berlandieri. SkuU. Averaging larger than in berlandieri with relatively smaller bullae and the same concavo-convex interparietal and posteriorly trun cate nasals. Measurements. Type (from skin): total length 272; tail 117; hind foot 32. Average of 6 adults from Colima: total length 253; tail 116; hind foot 34. Skull of type: basal length 30.7; nasals 13.3; zygomatic breadth 19.4; mastoid breadth 13.2; alveolar length of upper molar series 6. 3. Distribution. West coast of Mexico from western Jalisco to southern Oaxaca. Specimens examined. Total number 62 from the following localities: Jalisco: San Sebastian 2. Colima: Colima 10, Armeria 12. Bailey North American Species of Sigmodon. 109 Guerrero: Chilpancingo 5, Acapulco 1, near Ometepec 2, Tlaxtoquilla 1. Oaxaca. Puerto Angel 5, Oaxaca 5, Tehuantepec 6, Tamazulapam 2 im., Huajuapam 3 im., Cuicatlan 1 im. (sp?). Morelos: Yautepec 4 im., Quernevaca 2 im. Michoacan: Querendaro 1 (very large). Sigmodon hispidus tonalensis subsp. nov. Type from Tonala, Chiapas, Mexico, No. 75,144, U. S. National Mus eum, Biological Survey Collection. $ ad. Collected August 9, 1895 by E. W. Nelson and E. A. Goldman. Original No. 8312. General characters. Similar to mascotensis but larger and slightly browner; pelage very hispid. Color. Upperparts yellowish brown, more yellowish about nose; belly creamy white; feet and tail dull brownish gray, tail not distinctly bicolor. Young, duller and slightly darker colored. Skull. Long and narrow, heavily ridged in adults, interparietal pointed at ends and wide in middle, pushing supraoccipital well back. Measurements. Type: total length 350; tail 166; hind foot 41. Skull of type: basal length 34.5; nasals 15; zygomatic breadth 22; mastoid breadth 14.5; alveolar length of upper molar series 6.8. Distribution. Western Chiapas and eastern Oaxaca. Specimens examined. Total number 5, from 3 localities: Chiapas: Tonala 1; Mountains near Tonala 1. Oaxaca: Santo Domingo 3 (not typical). Remarks. This is merely a large form of mascotensis, from which there is no sharp geographic separation. Specimens from Chilpancingo, Guerrero, could be referred to it as well as to mascotensis. Sigmodon hispidus major subsp. nov. Type from Sierra de Choix, 50 miles northeast of Choix, Sinaloa, Mexico. No. 96,275, U. S. National Museum, Biological Survey Collec tion. $ ad. Collected October 20, 1898 by E. W. Nelson and E. A. Goldman. Original No. 13,154. General characters. Very large, almost equalling Mus norvegicus; feet stout; tail with coarse annulations; pelage coarse and hispid. Color. Upperparts light brownish gray; nose yellowish; belly white or yellowish white; feet light gray; tail blackish above, dark gray below. SkuU. Massive and heavily ridged in adults; interparietal strap-shaped with wide, rounded ends; nasals notched posteriorly; audital bullae relatively as well as actually large and elongated. Measurements. Type: total length 365; tail 156; hind foot 40.5. Av erage of 5 adults: 351; 160; 40. Skull of type: basal length 36; nasals 16; zygomatic breadth 23.5; mastoid breadth 16.4; alveolar length of upper molar series 7.3. 110 Bailey North American Species of Sigmodon. Distribution. West coast of Mexico from Tepic to southern Sonora. Specimens examined. Total number 16, from the following localities: Sinaloa: Sierra tie Choix 6, Culiacan 1, Rosario 1, Plomosas 1. Tepic: Acaponeta 2, Tepic 1, San Bias 1. Sonora: Alamos 3 (not typical). Remarks. This largest North American species of Sigmodon occurs with alleni at San Bias, Tepic, and while it may be quite distinct from its nearest congeners, berlandieri and mascotensis and does not appear to overlap the range of either, it unquestionably belongs to the same general group and should stand as a subspecies. Sigmodon hispidus toltecus Saussure. Hesperomys toltecus Saussure, Rev. et Mag. de Zool. XII, p. 98, 1860. Type locality. Mountains of the state of Vera Cruz, Mexico. General characters. Conspicuously smaller and darker colored than berlandieri; molars and audital bullae relatively smaller. <70for. Upperparts dull brownish gray; belly whitish; feet dark gray, not yellowish or brown; tail bicolor, gray below, blackish above. Skull. Smaller than that of berlandieri with much smaller bullae and molars. Measurements. Average of 8 adults from Orizaba, Vera Cruz: total length 235; tail 100; hind foot 28. Skull (No. 58,226, $ ad. from Ori zaba): basal length 29.5; nasals 13; zygomatic breadth 18; mastoid breadth 13; alveolar length of upper molar series 6. Distribution. Eastern Mexico from Alta Mira, southern Tamaulipas to Orizaba, Vera Cruz. Specimens examined. Total number 54 from the following localities: Vera Cruz: Orizaba 17, Chichicaxtle 4, Mirador 1. Tamaulipas: Alta Mira 12. San Luis Potosi: Valles 7. Puebla: Metlaltoyuca 13. Remarks. This is the only known species of North American Sigmo don of which I have not been able to examine the type or topotypes. As no more definite type locality was assigned than the mountains of Vera Cruz, I am assuming that the Orizaba specimens are typical, an assumption that seems safe in view of the wide range of the form, and the fact that no other species is known to occur in the general region, and that the characters assigned in the original description do not dis agree with those of the present series of specimens from localities cov ering most of the state of Vera Cruz. Specimens from as far north as Alta Mira, Tamaulipas, are indistinguishable from the Orizaba series, but farther north they seem to grade into berlandieri, while in the lower country of southern Vera Cruz they run into saturatus. Bailey North American Species of Sigmodon. Ill Sigmodon hispidus saturatus subsp. nov. Type from Teapa, Tabasco, No. 99,998, U. S. National Museum, Bio logical Survey Collection $ ad. Collected April 5, 1900, by E. W. Nel son and E. A. Goldman. Original No. 14,108. General characters. About the size of berlandieri but with much shorter tail and richer coloration; larger and more fulvous than toltecus. Color. Upperparts dark reddish brown; belly dull cinnamon brown, rarely whitish; feet yellowish brown; tail black, but little lighter below. Young, darker and duller colored. Skull. Short and compact with heavy rostrum; molars and audital bullae as in berlandieri. Measurements. Average of 7 adult topotypes: total length 250; tail 103; hind foot 31.2. Skull of type: basal length 29.5; nasals 12.7; zygo- matic breadth 19; mastoid breadth 13.5; alveolar length of upper molar series 6. Distribution. Mainly in Chiapas, Tabasco, and southern Vera Cruz. Specimens examined. Total number 101, from the following localities: Tabasco: Teapa 18, Frontera 4, Monte Cristo 2. Chiapas: Palenque 1, Yajalon 1, Chicharras 4, Huehuetan 7, Mts. near Oomitan 7, Comitan 8, Ocuilapa 2, Tenejapa 2. Guatemala: Zunil 2, Hda. Chancol 2, Jacaltenango 6, Nenton 7. Oaxaca: Tuxtepec 4, Cuicatlan 1 (im.), Mts. near Santo Domingo 3. Vera Cruz: Tlacotalpam 2, Otatitlan 3, Coatzacoalcos 6, Motzorongo 9. Remarks. While the rich coloration is unusually marked in specimens from Teapa, the type locality, considerable variation is shown through out the wide range assigned to this form. Specimens from the west slope in southern Chiapas and western Guatemala are slightly larger and duller colored without showing any difference to warrant separation or agreeing with any other described form. In general appearance they resemble S. borucae Allen, of Costa Rica, more than anything else, but in cranial characters are nearer to saturatus. To the northwest the in- tergradation with toltecus is shown by specimens from Motzorongo. Sigmodon hispidus microdon subsp. nov. Type from Puerto Morelos, Yucatan, No. 108,467, U. S. National Mu seum, Biological Survey Collection, $ ad. Collected March 13, 1901 by E. W. Nelson and E. A. Goldman. Original No. 14,581. General characters. Size small; tail short; colors dark and dull; teeth smaller than in any other North American Sigmodon. Color. Upperparts dark, dull brown; belly grayish white or buffy; tail black above, brownish black below. Skull. Small, slender, and narrow posteriorly with small audital bullae and very small molars. Measurements. Type: total length 243; tail 96; hind foot 32. Skull 112 Bailey North American Species of Sigmodon. of type: basal length 28.4; nasals 12.5; zygomatic breadth 18; mastoid breadth 13; alveolar length of upper molar series 5.5. Distribution. Northern Yucatan and Campeche. Specimens examined. Total number 45, from the following localities: Yucatan: Puerto Morelos 2, La Vega 5, Tunkas 5, Chichen Itza 20. Campeche: Apazote 17. Remarks. From toltecus this form is readily distinguished by darker coloration and brown instead of gray feet, from saturatus by duller colors and smaller size, and from both by its very light dentition. Specimens from Apazote and Chichen Itza are grading toward saturatus. Sigmodon hispidus borucae Allen. Sigmodon borucae Allen, Bull. Am. Mus. Nat. Hist. IX, 40, March 11, 1897. Type locality. Boruca, Costa Rica. General characters. Size medium, about as in berlandieri, pelage rather soft. Color. Upperparts dull yellowish brown; belly grayish or buffy white; feet grayish brown; tail blackish above, grayish brown below. Skull. Nearest to that of saturatus but distinguished by the smaller, flatter audital bullae, and other slight characters. Measurements. Type (from original description): total length 275; tail 115; hind foot (measured dry) 30. Skull: basal length 29; nasals 12; zygomatic breadth 18; length of upper tooth row 5.5. Distribution. Costa Rica; specimens examined from the type locality only. Remarks. Sigmodon h. borucae belongs to the toltecus group of rather small, dark colored forms, but shows good subspecific characters, differ ing from toltecus in darker coloration, brown instead of gray feet, heavier dentition and other slight skull characters; from microdon in larger size, longer tail, heavier dentition and flatter audital bullae; from typical saturatus in duller coloration, longer tail, flatter audital bullae; but from Guatemala specimens of saturatus it does not differ in color. Sigmodon alleni sp. nov. Type from San Sebastian, Mascota, Jalisco, Mexico, No. 88,227, U. S. National Museum, Biological Survey Collection, $ ad. Collected March 15, 1897 by E. W. Nelson and E. A. Goldman. Original No. 10,708. General characters. Size of berlandieri but with slenderer feet and tail, smaller molars, brighter colors, and softer pelage; pelage fine and not hispid in winter or summer. Color. Upperparts dull tawny; belly white or buffy; feet yellowish brown; tail black above and brownish below. Young, dark umber brown, soon becoming tawny. Bailey North American Species of Sigmodon. 113 Skutt. Similar to that of berlandieri but with narrower braincase, double convex outlines of interparietal, smaller molars, and smaller, shorter, more rounded bullae. Measurements. Average of 8 adults from type locality: total length 244; tail 112; hind foot 31.6. Skull of type: basal length 29.3; nasals 13; zygomatic breadth 19.3; mastoid breadth 13; alveolar length of upper molar series 6.4. Distribution. Western Jalisco and southern Tepic in Western Mexico. Specimens examined. San Sebastian, Jalisco, 10, San Bias 1, Tepic 2, and Valle de Banderas 1. Remarks. Although belonging to the hispidus group, alleni is quite distinct from all neighboring species, occurring at the type locality with mascotensis, the west coast form of berlandieri, and at San Bias with major. It is the species Dr. Allen had in mind in writing his description of mascotensis from which he intended to separate the Colima form, but as his two specimens of the species here named alleni were not fully adult he unfortunately selected as his type another specimen that proves to be the same as the form to which on the following page he gave the name of colimae. As these two names apply to the same species I have placed colimae as a synonym under mascotensis and taken the opportunity of associating Dr. Allen's name with the species which he intended to describe. Sigmodon fulvi venter Allen. Sigmodon fulviventer Allen, Bull. Am. Mus. Nat. Hist., II, p. 180, October 21, 1889. Type locality. Zacatecas, Zacatecas, Mexico. General characters. Size largest of the hairy tailed group; tail cov ered with short, bristly hairs that almost conceal the annulations; pe lage rather hispid. Color. Upperparts yellowish brown, lightened on sides by white bristles; upper half of outer surface of ear black; lowerparts and feet clear rich fulvous; tail brownish black above and below. Skull. Short, wide, and heavily ridged, not much arched, not bulged interorbitally ; interparietal, about 2 mm. wide and not divided; supra- occipital with a slight median ridge; nasals short, wide and rounded at ends; jaw short and heavy with short, wide coronoid. Measurements. Type: total length 270; tail 108; hind foot 33. Skull of type: total length over incisors 35.5; nasals 13; zygomatic breadth 21.5; alveolar length of upper molar series 6.3. The broken base of the skull makes the usual set of measurements impossible. Distribution. Known only from two localities; from about 8000 feet altitude near the city of Zacatecas and from Durango, Durango. Speci mens examined 2. 114 Bailey North American Species of Sigmodon. Sigmodon melanotis sp. nov. Type from Patzcuaro, Michoacan, Mexico, at 7000 feet altitude. No. 50,190, U. S. National Museum, Biological Survey Collection, 9 ad. Collected July 15, 1892 by E. W. Nelson. Original No. 2834. General characters. Size nearly equalling that of fulviventer; colors darker and richer; pelage softer; tail equally hairy. Color. Upperparts dark, rich ochraceous, heavily lined with black; outer and part of inner surface of ears black; face and nose blackish; belly clear, dark, rusty ochraceous; feet yellowish brown; tail mainly black, yellowish below at base. Skull. High and heavily ridged in adults with zygomatic arches falling low at sides; interorbital space narrow and bulged on top; nasals rounded at ends: interparietal about 2 mm. wide and not divided; supraoccipital with distinct median ridge. Measurements. Type: total length 275; tail 100; hind foot 31. Aver age of 4 from type locality: 257; 102; 31. Skull of type: basal length 31.4; nasals 12.2; zygomatic breadth about 20.5; mastoid breadth 14.5; alveolar length of upper molar series 6.3. Distribution. Known only from a series of 15 specimens from Patz cuaro. Sigmodon minimus Mearns. Sigmodon minima Mearns, Proc. U. S. Nat. Mus., XVII, 130, July 19, 1894. Type locality. Upper Corner Monument, Grant Co., New Mexico, on the Mexican boundary line 100 miles west of the Rio Grande. General characters. Size small, hind foot about 28; tail hairy as in fulviventer; pelage not hispid. Color. Upperparts clear gray, with coarsely grizzled appearance, and without yellowish on nose or rump; belly buff y; feet gray; tail brownish black above and below. Skutt. Short and wide with abruptly spreading zygomata; interparie tal narrow; supraoccipital with slight median ridge; audital bullae small and basioccipital narrowed between them. Measurements. Type: total length 223; tail 94: hind foot 28. Topo- type: 223; 91; 27. Skull of type: basal length 28.5; nasals 11.3; zygo matic breadth about 19; mastoid breadth 14; alveolar length of upper molar series 5.9. Distribution. Mountains of southern New Mexico, Arizona and north ern Mexico. Specimens examined. Total number 14 from the following localities: New Mexico: Grant Co., near Mexican line, 2, (including type). Arizona: Fort Huachuca 2. Chihuahua: Casas Grandes 1. Sonora: Santa Cruz River 9. Remarks. This is the northernmost form of the hairy tailed group. Bailey North American Species of Sigmodon. 115 Sigmodon ochrognathus sp. nov. Type from the Chisos Mts., Texas, 8,000 feet altitude No. 110,333, U. S. National Museum, Biological Survey Collection, 9 ad. Collected June 13, 1901, by Vernon Bailey. Original No. 7681. General characters. Size small; tail long and hairy; pelage but slightly hispid. Color. Upperparts yellowish gray, strongly tinged around ears, face and rump with ochraceous; nose, orbital ring and base of tail clear, bright ochraceous; belly white; feet buffy gray; tail blackish above, buffy gray below. Skull. Similar to that of minimus but interparietal with a posterior indentation, bullae narrower and basioccipital shorter and wider; nasals short, wide, and truncate posteriorly. Measurements. Type: total length 260; tail 117; hind foot 29. Skutt of type: basal length 28; nasals 11.6; zygomatic breadth 19; mastoid breadth 13; alveolar length of upper molar series 5.5. Distribution. The transition zone top of the Chisos Mts., Texas and 6,700 feet altitude near Parral, Chihuahua. Four specimens examined from the type locality, and one from near Parral. Sigmodon leucotis sp. nov. Type from the Valparaiso Mts., Zacatecas, Mexico, 8,700 feet altitude, No 92,001, U. S. National Museum, Biological Survey Collection, 9 ad. Collected December 2, 1897, by E. W. Nelson and E. A. Goldman. Original No. 11,812. General characters. Size medium; tail short and hairy; pelage rather soft; colors dull, with conspicuously light ears. Color. Upperparts dull brownish gray; ears whitish gray; belly whitish; feet brownish gray; tail black, becoming brownish at the base below. Skutt. Short, wide and heavily ridged along sides; interparietal narrow, with a dividing suture across middle in 11 out of 13 specimens; supraoccipital sloping, without trace of median ridge; nasals short, narrow and notched at posterior end; audital bullae full and oval. Measurements. Average of 8 adult topotypes: total length 234; tail 91; hind foot 29.4. Skull of type: basal length 31; nasals 11.4; zygo matic breadth 20.5; mastoid width 15; alveolar length of upper molar series 6. Distribution. Known only from the type locality. Number of speci mens examined 13. Remarks. The divided interparietal occurs in one other species from Oaxaea, but as it is not entirely constant it is probably of little signifi cance. 116 Bailey North American Species of Sigmodon. Sigmodon alticola sp. nov. Type from Cerro San Felipe, Oaxaca, Mexico, at 10,000 feet altitude, No. 68,231, U. 8. National Museum, Biological Survey Collection, $ ad. Collected August 24, 1894, by E. W. Nelson and E. A. Goldman. Orginal No. 6624. General characters. Size medium; tail hairy; pelage long and soft both in winter and summer; colors dark and rich except gray ears. Color. Upperparts dark ochraceous or light umber brown; ears clear gray, a little darker than in leucotis; belly pale cinnamon brown; feet yellowish gray; tail black, becoming yellowish brown below at base. Skutt. Similar to that of leucotis in form but slenderer, less heavily ridged and narrower interorbitally; interparietal narrow with a dividing suture in middle; supraoccipital without median ridge; nasals narrow -and rounded at ends; lateral pits of palate very shallow. Measurements. Type: total length 230; tail 101; hind foot 28.5. Skutt of type: basal length 28.5; nasals 11; zygomatic breadth 17.8; mastoid breadth 13.5; alveolar length of upper molar series 6.3. Distribution. Mountains of Oaxaca. Specimens examined. Cerro San Felipe 1, 15 miles west of Oaxaca 2. Sigmodon alticola amoles subsp. nov. Type from Final de Amoles, Queretaro, Mexico, at 7,000 feet altitude. No. 81,430, U. S. National Museum, Biological Survey Collection, $ ad. Collected September 18, 1898, by E. W. Nelson and E. A. Goldman. Original No. 10,161. General characters. Similar to alticola but upperparts duller and less tawny, skull wider interorbitally, with slightly smaller bullae and deeper lateral pits of palate; interparietal narrower but not divided. Measurements. Type: total length 252; tail 105; hind foot 29.5. Skutt of type: basal length 29.5; nasals 5.7; zygomatic breadth 19.6; mastoid breadth 14; alveolar length of upper molar series 6.3. Distribution. Known from only 2 specimens from type locality. Remarks. While there seems hardly a possibility of continuity of range across the wide gaps of low country separating the type localities of the two forms, the slight difference of the present form from alticola does not admit of more than subspecific rank. VOL. XV, PP. 117-120 JUNE 2, 1902 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON SEVEN NEW MAMMALS FROM WESTERN TEXAS. BY VERNON BAILEY. In working out the distribution of the various species of mammals inhabiting the state of Texas, I find the following forms that need subspecific recognition. Two of the genera, Eutamias and Microtus were not previously known to occur within the state. The muskrat, while described from southern New Mexico, occurs also along the lower Pecos and Rio Grande. Eutamias cinereicollis canipes subsp. nov. Type from Guadalupe Mts., Texas. Altitude 7,000 feet in Transition Zone. No. 109,229, U. S. National Museum, Biological Survey Collec tion, 9 ad. Collected August 24, 1901, by Vernon Bailey. Original No. 7827. General characters. Similar to cinereicollis in fresh postbreeding pelage, but grayer throughout, with paler and duller ochraceous on sides; tips of the long hairs on rump and upper surface of tail white instead of yellowish; feet clear gray without a tinge of yellowish; black dorsal stripe reaching forward to between the ears. Measurements. Type: total length 230; tail vertebrae 104; hind foot 35. Skull of type: basal length 30; nasals 11; zygomatic breadth 19.3; mastoid breadth 15.8; alveolar length of upper molar series 5.8. Remarks. The present form needs no comparison with its nearest neighbor, the little light colored gracilis. 24 BIOL. Soc. WASH. VOL. XV, 1902. (117) 118 Bailey New Mammals from Western Texas. Spermophilus spilosoma marginatus subsp. nov. Type from Alpine, Texas, No. 108,927, U. S. National Museum, Biolog ical Survey Collection, $ ad. Collected July 5, 1901, by Yernon Bailey. Original No. 7702. General characters. Similar to the dark, typical form of spilosoma from northeastern Mexico, but smaller and brighter colored. Color. Upperparts bright cinnamon brown, the whole back from ears spotted with whitish, the spots conspicuously edged with black. .Skull. With relatively small bullae and wide basioccipital. Measurements. Type: total length 225; tail vertebrae 67; hind foot 32. Skull of type: basal length 33.7; nasals 13.3; zygomatic breadth 23.5; mastoid breadth 19; alveolar length of upper molar series 7.2. Specimens examined. Alpine 1; Toyahvale 1; Presidio Co. 1 (im). Remarks. From the El Paso form, marginatus differs in darker color and heavier spotting; from major in finer, sharper spotting, and from both in the black edging of spots as well as in cranial characters. Spermophilus spilosoma arens subsp. nov. Type from El Paso, Texas, No. 64,977, U. S. National Museum, Biolog ical Survey Collection, $ ad. Collected May 10, 1894, by Dr. A. K. Fisher. Original No. 1446. General characters. Similar to cryptospilotus but larger with coarser spotting. Color. Upperparts light clay color or dark buffy, finely spotted over back and rump with white, the spots without black edges. Distin guished from neighboring forms by its pale colors. Skull. Like that of cryptospilotus but more elongated, with wider nasals and heavier dentition. Mersurements. Type: total length 237; tail vertebras 85; hind foot 35. Average of 5 adults: 230; 78; 34. Skull of type: basal length 33; nasals 12; zygomatic breadth 22; mastoid breadth 18; alveolar length of upper molar series 7. Specimens examined. El Paso, Texas, 5; Casas Grandes, Chihuahua 4. Microtus mexicanus guadalupensis subsp. nov. Type from Guadalupe Mts., Texas, altitude 7,800 feet. No. 109,191, U. S. National Museum, Biological Survey Collection, $ ad. Col lected August 21, 1901 by Vernon Bailey. Original No. 7807. General characters. Slightly larger and lighter colored than mexicanus, with skull characters distinguishing it from both mexicanus and mogollo- nensis. Color. Upperparts dull umber brown; belly buffy gray; feet and tail brownish gray. Scarcely distinguishable from mogollonemis in color. Bailey New Mammals from Western Texas. 119 Skull. Comparatively long, with short, wide nasals, rounded at pos terior ends; incisors dark orange instead of light yellow; middle upper molar with projecting inner point at base of posterior triangle as in mogo llonensis. Measurements. Type: total length 152; tail 34; hind foot 20. Aver age of 9 adults: 144; 34; 19. Skull of type: basal length 24.5; nasals 7.5; zygomatic breadth 16; mastoid breadth 12.4; alveolar length of upper molar series 7.2. Fiber zibethicus ripensis subsp. nov. Type from the Pecos River at Carlsbad (Eddy), New Mexico, No. 109,012, U. S. National Museum, Biological Survey Collection, $ ad. Collected July 25, 1901, by Vernon Bailey. Original No. 7757. General characters. Size small; color paler than in zibethicus but darker than in pattidus; skull narrow; incisors heavy. Color. Adults in July: Upperparts light brown without any distinct black; belly buffy brown; throat gray, inclosing a sharp, dusky chin stripe; a small spot over each nostril whitish. Skull. Small and narrow; nasals short and rounded or pointed at pos terior end, never notched; incisive foramina widest at posterior end; incisors heavy, molars light. Measurements. Type: total length 470; tail 202; hind foot 67. Aver age of 6 adults: 463; 204; 68. Skull of type: basal length 55; nasals, 18; zygomatic breadth 35; mastoid breadth 25; alveolar length of upper molar series 15. Remarks. This form is distinguished from zibetMcus by paler color, smaller size, heavier incisors, unnotched nasals, and form of incisive foramina; from paUidus by darker color, narrower skull, form of nasals and incisive foramina, and slightly heavier dentition. No other compar isons are necessary. Thomomys fulvus texensis subsp. nov. Type from the head of Limpia Creek at 5,500 feet altitude in the Davis Mts., Texas, No. 2251, U. S. National Museum, Biological Survey Collection, $ ad. Collected January 7, 1890, by Vernon Bailey. Original No. 876. General cJiaracters. Similar to Thomomys fulvus but smaller, less dusky, and with narrower skull. Color. Upperparts dusky fulvous with blackish ear patch and nose ; lowerparts plain bright fulvous; feet and lips gray; lining of pouches white. Skull. Similar to that of fulvus but with conspicuously narrower braincase, shallower lateral pits of palate, and shorter pterygoids. Measurements. Type: total length 204; tail 63; hind foot 26. Aver- 120 Bailey New Mammals from Western Texas. age of 10 adults: 193; 60; 26. Skull of type: basal length 34.5; nasals 13; zygomatic breadth 22; mastoid breadth 18; alveolar length of upper molar series 7. Remarks. The affinity of the present form with the fulvus group is shown by the general form of skull and by the wide, flat basioccipital with sharp instead of rounded lateral edges as well as by the similarity of color. Thomomys aureus lachuguilla subsp. nov. Type from arid foothills near El Paso, Texas, No. 110,336, U. S. National Museum, Biological Survey Collection, $ ad. Collected September 24, 1901, by Vernon Bailey. Original No. 7858. General characters. A small, light yellowish form of the aureus group with big bullae and narrow basioccipital. Color. Summer pelage: Upperparts dull ochraceous; belly buffy or pale cinnamon. Winter pelage: Brighter ochraceous above and light buffy below. /Skull. Slender, with conspicuously slender rostrum; bullae full and rounded with basioccipital narrowly constricted between; interparietal small and rounded or quadrate; premaxillae slender and projecting well back of nasals; upper incisors bent down at right angles to axis of skull. Measurements. Type: total length 215; tail 65; hind foot 29. Aver age of 4 adults: 201; 61; 27.4. Remarks. Named for the Agave lachuguilla which furnishes its favorite food. VOL. XV, PP. 121-145 JUNE 20, 1902 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A LIST OF THE BATRACHIANS AND REPTILES OF THE DISTRICT OF COLUMBIA AND VICINITY. BY W. P. HAY. The following list which completes the series, published by various authors in various papers, upon the vertebrata of the District of Columbia, represents the work of many collectors extending over a long series of years. The compilation of records was begun in 1890 by Mr. Frederick C. Test, at that time aid to the curator of Reptiles in the United States National Museum, and since 1895 has been continued by the author. Extensive collecting by Mr. Test and myself with the coopera tion of many friends, among whom I might mention several of my pupils at the Central High School, has been the means of adding new species to our known fauna and of establishing numerous additional stations for those already known. In the effort to make my paper of greater value than a mere list of species I have prepared keys to all the species known to occur, or likely to occur within our limits, and have given brief descriptions and such notes on the habits as my very limited space makes possible. I have drawn the descriptions of form and color largely from the paper of my father, Dr. O. P. Hay, on the Batrachians and Reptiles of the State of Indiana, but have been at some pains to modify them to suit the peculiar require- 25-BIOL. SOC. WASH. VOL. XV, 1902. (131) 122 Hay JSatrachians and Reptiles of D. C. ments of the present case. The notes on the habits are such as I have been able to make during some fifteen years of collect ing in the central states and in this locality, but most of those on the cries of the frogs have been furnished me by Mr. Gerrit S. Miller, Jr., who has enjoyed exceptional advantages for the observation of these animals. It is a matter of regret to me that my paper could not have been expanded into many times its present volume so that these extremely interesting and much neglected groups of animals might have been given an adequate exposition, but under the circumstances it has been impossible. My thanks are due to the many friends who have assisted in the compilation of records, but perhaps most of all to Dr. L. Stej- neger, of the United States National Museum, who has been kind enough to review my manuscript, and call my attention to errors which otherwise would have appeared in the text. Key to the Classes and Orders. a. Body scaleless in all our species; covered with a skin which is usually smooth, soft, and more or less slimy, but sometimes rough and warty; eggs laid in or near the water and giving origin to tadpoles (Water-dogs, salamanders, toads, and frogs). Class BATRACHIA. b. Body with a distinct tail throughout life; hind limbs, if present, not especially enlarged. Order CAUDATA, p. 123. bb. Body without tail except during immature stages; all four limbs present, the hind pair much the larger. Order SALIENTIA, p. 127. aa. Body covered with yielding, overlapping scales, or enclosed in a bony, box-like shell; eggs, when laid, deposited on land and giving origin to young which are similar in form to the adults (alligators, lizards, snakes, and turtles). Class RBPTILIA. c. Body not enclosed in a shell. Order SQUAMATA, p. 132. cc. Body enclosed in a box-like shell (turtles). Order CHBLONIA, p. 142. CLASS BATRACHIA. Up to the present time the known species of jBatrachia of the region about Washington number 23. These are dis tributed as follows: Caudata 11, Salientia 12. The distinguish ing characters of these groups having been given above; there follows the key to the order Caudata. Hay Batrachians and Reptiles of D. C. 123 FIG. 1. Topography of the salamanders and frogs, eg, costal grooves; fi, fingers; to, toes; gf, glandular folds; , vomerine teeth; pa, parasphe- noid teeth; c, choanae; d, adhesive disks; h, heel; t, tympanum. Key to the Species of Caudata. a. Body eel shaped; hind limbs absent; Siren lacertina, 1. aa. Body salamander-shaped; all 4 limbs present. b. Toes 4 in front and 4 behind. - - Hemidactylium scutatum, 3. bb. Toes 4 in front and 5 behind. c. Parasphenoid teeth present. d. Tongue mushroom-like, attached by a central stalk only. e. Costal grooves 13 or 14. /. Tail not much, if any, longer than rest of animal; yellow ish, back with median row of dots, and two dusky bands. Spelerpes bislineatus, 6. ff. Tail considerably longer than rest of animal. g. Yellow, with many black spots; tail with black cross bars. \\- ...-' -.. v. - Spelerpes longicaudus, 8. gg. Yellow, back with a broad median and two lateral bands; tail with yellow cross lines. Spelerpes guttolineatus, 7. ee. Costal grooves 15 to 17; red, with many small black dots. Spelerpes ruber, 9. dd. Tongue attached by a median longitudinal band running from the central stalk to the tip, therefore extensively free only at the sides. Ji. Costal grooves 16 to 19. Color dark ashy or brown, with or without a rusty dorsal band. - Plethodon cinereus, 4. hh. Costal grooves 14. i. Black above, with small white dots. Plethodon glutinosus, 5. ii. Brownish, variously mottled and marbled with darker and ocraceous. - Desmognathus fusca, 10. ee. Parasphenoid teeth absent. j. Tongue small, free at the sides; a medium sized species, green ish or red, belly yellow with many black specks. Diemictylus viridescens, 11. 124 Hay Batrachians and Reptiles of D. C. jj. Tongue large, attached by nearly the whole posterior surface and almost filling the mouth; large clumsy species, brown or black, variously marbled or spotted with whitish or yel low. k. Costal groves 11. Ambystoma opacum, 2. kk. Costal grooves 12. I. Sole with 2 distinct tubercles. Ambystoma tigrinum, 2a. tt. Sole with 1 indistinct tubercle or none. Ambystoma jeffersonianum, 2b. 1. Siren lacertina Linnaeus. Siren; Mud Eel. Body elongate, eel-like; head flat, tapering to the blunt snout; three pairs of external gills; eyes very small; anterior limbs short, with 4 toes; hind limbs absent. Bluish black, paler below. Length 2 to 3 feet. A single record from the Potomac Flats testifies to the presence of this species within our limits. It seems to spend most of its time in the soft mud of swamps, and is, therefore, seldom observed even where it is common. 2. Ambystoma opacum (Gravenhorst). Marbled Salamander. Body short, stout, and swollen; tail short; limbs weak. Dark brown or black above, with about a dozen bluish gray cross bars; an irregular splotch on head and nape of the same color. Length 3^ inches. Reported from Takoma Park, Laurel, Md., and Mt. Vernon, Va. At the latter locality I took two specimens from under an old stump in a very dry locality. aa. Ambystoma tigrinum (Green). Tiger Salamander. Body large, heavily built, with long tail, and stout legs. Bluish or brownish black, more or less spotted or blotched with bright yellow. Length, 5-10 inches. There are no definite records for this species within our limits but it will doubtless be found to occur here. 2b. Ambystoma jeffersonianum fuscum (Green). Jefferson Sala mander. Body and head rather elongated; head broad and somewhat depressed; tail rather long, compressed. Dark bluish or brownish, or black, paler below; sides sometimes with small white spots. Length 5-8 inches. As with the preceding there is no record of the capture of this species within our limits, but careful collecting will probably add it to our fauna. 3. Hemidactylium scutatum (Schlegel). Scaly Salamander. Body slender; head depressed, broadest just behind the eyes; snout short, truncate; base of tail with a decided constriction, beyond which the tail swells out and then tapers to a sharp point; legs weak; toes flay JBatrachians and ^Reptiles of D. C. 125 short, almost rudimentary. Brown, chestnut, or purplish, bluish white below; above there are numerous pale spots and specks; below there are specks and spots of black. Length 3.5 to 4 inches. In the collection of the United States National Museum there is a specimen labelled, "D. C.", but with data so insufficient as to throw the locality in some doubt. During the collecting seasons of 1892 and '93 I found the species occasionally under logs in the heavy woods between Little Hunting Creek and the Mt. Vernon estate. 4. Plethodon cinereus (Green). Ashy or Red-backed Salamander. Body elongate, slender, cylindrical; head small and short, snout rounded; eyes large and prominent; limbs short and weak; toes short, the inner ones rudimentary. Brownish black above; whitish or yellow ish below, finely marbled with black; back often with a broad, brownish red longitudinal stripe. Length 3.5 to 4 inches. Both the red-backed and brown varieties of this species are found in abundance throughout the region covered by this list. They are to be looked for under bark, leaves, logs, or flat stones in damp situations in the woods. 5. Plethodon glutinosus (Green). Slimy Salamander. Body cylindrical or somewhat depressed; skin smooth, shining and very slimy; legs moderately developed, the toes short and depressed, the inner on fore and hind legs small, but distinct. Black or blue-black above, paler below; sides with numerous whitish spots about as large as the eye. Length 3 to 7 inches. Rather common within our limits. I have found it most abundant about Takoma Park and Mt. Vernon. 6. Spelerpes bislineatus (Green). Two-lined Triton. Body elongate and slightly depressed; tail forming about one half the total length; head rather narrow, snout short and rounded. Color yel lowish brown or clear yellow above, pale yellow below; along each side of the back there is a brown line extending from the eye to the tail. Length about 3 inches. A very common species throughout the region. It may be sought un der the stones or leaves near springs or small streams. 7. Sperlepes guttolineatus (Holbrook). Holbrook Triton. Body somewhat flattened, elongate and slender, but stouter than in 8. longicaudus; head flat, snout rounded; tail low and flattened, forming more than one half the total length, tapering to a sharp point. Brown ish yellow above, with three stripes of dark brown, one along the me dian line and one on each side; tail dark brown with vertical, yellow lines; the under surface is yellow mottled with brown. Length about 5 inches. 126 Hay Batrachians and Reptiles of D. C. This triton has been taken at Dun Loring and Munson Hill, Va. It is a southern species and its occurrence in our fauna is probably very rare. 8. Spelerpes longicaudus (Green). Long-tailed Triton. General appearance and proportions very similar to the preceding. Above, bright yellow or orange with many black spots and specks, some of which may form an interrupted band along the sides; tail with ver tical black bars; below, the color is paler and there are no spots. Length 5 to 6 inches. Until the fall of 1896 this salamander was known from our region by a single, very indefinite record which was all the more unsatisfactory that the specimen had been lost. In September of that year I secured a fine, well developed individual from among some damp leaves, near Great Falls, on the Maryland side of the river. 9. Spelerpes ruber ruber (Daudin). Red Triton. Body varying with age but rather heavy and clumsy; head depressed and short, snout rather pointed ; tail about two fifths total length, strong ly keeled. Color varying from clear coral red in the young to brownish red in the fully adult; above, there are numerous small black spots, dis tinct in the young but much diffused in old individuals. Length 3 to 6 inches. The typical form alone occurs here. It is very abundant and has been recorded from nearly every spring in the vicinity. From the muddy margins of a spring near Brookland, I once secured over twenty speci mens and from a spring near Accotink, six. I have also seen it crawling about on land after a rain. 10. Desmognathus fusca fusca (Rafinesque). Brown Triton. Body rather heavily built and somewhat depressed; head flat, snout rounded; limbs feebly developed; tail forming about one half total length, circular at base, but much flattened posteriorly. Dark brown, belly paler; back and sides often variously mottled with ocraceous or grayish. Length 3 to 4 inches. This very active species is quite common about Washington. It is usually found in springs and rapid streams where it can conceal itself under stones and drift, but it is often to be observed lying in a sort of burrow in moss or grass roots near the water. n. Diemictylus viridescens Rafinesque. Green Triton; Newt. Body rounded and full, with a sharp vertebral ridge; head tapering to the rounded snout, and with three longitudinal grooves above; tail form ing about one half the total length, flattened, tapering and with a more or less developed membrane-like fin. Front legs weak, hind legs stout. Above olive green or reddish of varying shades; lemon yellow below, Hay Batrachians and Reptiles of D. C. 127 with numerous scattered black specks; each side with a series of scarlet spots each surrounded by a black ring. The young (var. miniatus] are bright vermillion red with the spots on the sides rather indistinct. Length about 3| inches. While this species is not often collected, on account of its retiring habits and its very perfect color protection, it is common in most quiet ponds where there is growing vegetation. I have found the immature forms most abundant near Mt. Vernon. Key to the species of Salientia. a. Upper jaw toothless. r - - ':- - Bvfv lentiginosus, 12. aa. Upper jaw with teeth. b. Fingers and toes more or less dilated at the tips, this dilation form ing a viscid disk; small species, not over 2 inches long (body measurement). c. Disks small; fingers not webbed. d. Toes broadly webbed. .... Acris gryllus, 14. dd. Toes scarcely webbed. ChoropJiilus nigritus, 15. cc. Disks well developed; fingers somewhat webbed. e. Color above clear green, with usually a few golden yellow specks. - -- . "' * .- - - Hyla evittata, 18. ee. Color above grayish, greenish, or brown variously striped or mottled /. Back with irregular lines forming an X, outside of which are other lines; general color brown. - Hyla pickeringi, 16. ff. Back uniform or mottled, general color grayish-green, gray, or brown. "'', 1 ^- - '''-.* - " -' ' - Hyla versicolor, 17. bb. Fingers and toes not specially dilated at the tips and not provided with adhesive disks; larger species, 2| inches long or more. g. Sole of foot with a horny spade-like process; form squat and toad- like; hind legs weak. - - - Scaphiopus holbrooki, 13. gg. Sole of foot without a spade-like process; body frog-like; hind legs powerfully developed. li. Sides of back nearly smooth, the glandular thickening of the skin being obsolete. - . Rana catesbeiana, 23. ?ih. Side of back with one or more well developed, thickened ridges of skin. i. Back with large distinct black spots, more or less regularly arranged. j. Outline of dorsal spots irregularly roundish. Rana pipiens, 19. jj. Dorsal spots squarish in outline and arranged in two fairly regular rows. .... Rana palustris, 20. ii. Back with small, indistinct spots or none. k. A conspicuous dark stripe from snout to front of eye. Rana sylvatica, 22. kk. No conspicuous dark stripe on side of head. Rana clamata, 21. 128 Hay Batrachians and Reptiles of D. C. 12. Bufo lentiginosus americanus (Le Conte). Toad. Body heavy and awkward, the skin everywhere rough and warty ; snout short and blunt; limbs shorter than in the frogs. Color sometimes ash gray, black, or even brick red but usually olive or brown with ir regular blotches of dark brown; below, dirty yellow. Length 2} to 5 inches. The toad is an extremely common animal in our region. It appears early in the spring when its long continued, loud trill is one of the first evidences of the renewed activity of animal life. I have found them breeding in early March. 13. Scaphiopus holbrooki (Harlan). Spadefoot Toad. Body toad-like, but a little less squat and clumsy; skin pustular but lacking the large warts of the preceding species; both pairs of legs short and stout; hand with two black callosities; foot with an elongated, com pressed, spade-like process, black in color, with a sharp horny edge, and situated at the base of the inner toe. Brownish with a slightly paler band extending backward from the eye; sides sometimes mottled with ashy. Length about 3 inches. There is a single record for this rare species within our limits. It is almost altogether subterranean in its habits, coming forth very rarely, and then at night. Its cries during the breeding season are said to be almost as loud and shrill as a steam whistle. 14. Acris gryllus (Le Conte). Cricket Frog. Form frog-like; snout pointed; hind legs long, the heel reaching near to or beyond the snout when the leg is turned forward; fingers without web; toes webbed to near their tips; skin of back smooth or with large or small warts. Color variable and changeable; usually the upper sur face is gray or brownish, there is often a large green dorsal patch and occasionally this color predominates; often there is considerable reddish; there is a dark triangle between the eyes and a white line from eye to arm; legs with cross bars; below, the color is whitish. Length 1| inches. The species has been divided by Cope into two subspecies, A. gryllus gryllus and A. gryllus crepitans, which are of very doubtful validity. Both are found in abundance along the banks of streams and ponds, whence their continuous metallic cry comes almost without intermission during the season of activity, particularly during hot, sunshiny days. 15. Chorophilus nigritus feriarum (Baird). Striped Treefrog. Body and legs rather slender; head rather pointed; fingers with at most only a rudiment of a web; toes with a slight web at their basal joints. Color varying from light ash to fawn and purplish brown; below, cream color; on the back there are three dark stripes, extending from the head to near the end of the body, of these the median sometimes Hay Batrachians and Reptiles of D. C. 129 forks near the middle of its length; upper surface of limbs barred or blotched. Length, H inches. This species is rarely seen except during its breeding season, when it is quite common. Its cry consists of two or three clear whistle-like chirps, like those of a young turkey, and, after one has learned to distinguish it from that of H. pickeringi, serves to betray its presence at once. In the spring of 1896 I collected thirty specimens in less than an hour from the gutters along the Conduit Road near Cabin John's Bridge. 16. Hyla pickeringi (Storer). Pickering Treefrog. General form like the preceding but the body somewhat stouter; fingers and toes with well developed disks; fingers entirely unwebbed; toes half webbed; color ashy, brown, or reddish above; grayish or whitish below; on the back there is an irregular X of dusky; on the sides another dusky line runs parallel with the posterior limbs of the X, and on the head is a broad V-shaped mark; a dark line from snout, through eye, to back of head; limbs barred or mottled. Length about 1 inch. This species is more common than the preceding and is occasionally met with throughout the summer. Its habits during the breeding season are easily studied, if one will go to some secluded woodland pool with a lantern, and the sight is well worth the trouble. Like many other species, this one has a large gular sac which it inflates when it sings. This sac is distended until it is almost as large as the body, and the skin is so stretched that it is nearly transparent. The muscles of the sides contract tremendiously and a clear whistle-like cry is produced, out of all proportion to the diminutive size of the frog. This cry differs from that of the preceding species in that the chirp is repeated monotonously time after time. 17. Hyla versicolor Le Conte. Common Treefrog, Chamasleon Tree- frog. Body rather stout and almost toad-like; head broader than long, snout rounded; fingers and toes with large disks; fingers with an evident web; toes webbed to near their tips; skin with numerous small warts above, granulate beneath. Gray-green, gray, or brown, with irregular, indistinct dark blotches and lines; below, white with bright yellow on groin and front and back of thighs. Length 1 to 2 inches. Common but not often collected. Its "short, loud, trilled rattle" is heard throughout the summer in the evenings or during damp weather, individuals calling and answering from their perches on trees or fences. 1 8. Hyla evittata Miller. Swamp Treefrog. Form similar to ChoropTiilus nigritus but with blunter snout and slenderer legs: fingers and toes with well developed disks; fingers dis tinctly webbed; toes webbed to the disks; skin nearly smooth. Color of 130 Ruy llatrachians and Reptiles of D. C. adults varying from olive brown to bright leaf green; below, white slightly tinged with j'ellow; on the back there are often a few small yellow specl. Small specimens will often be found to have a more or less distinct white line extending about half way down the sides of the body from the snout, and both pairs of legs with a whitish band on the anterior surface. Length H inches. This frog which I first discovered near Mt. Vernon and regarded as H. ciner ea, has since been shown by Mr. G. S. Miller, Jr., to be a dis tinct species. My specimens were found along the shore of Little Hunt ing Creek, October 15th, 1892. They were immature and were evidently looking for winter quarters and were hopping about everywhere among the damp leaves. I collected about fifty specimens, and on October 29th secured another lot. On visiting the locality late in the following spring, I found them among the cat-tails and other large aquatic plants some distance from the shore. Since then it has been observed in Swan Creek and Messrs. Miller and Preble secured full grown specimens from Four- mile Run and at Dyke. Its note is quite unlike that of any of our other frogs and consists of a series of high pitched, rattling croaks, which may remind one slightly of the shrill grunts of a very young pig. 19. Rana pipiens Schreber. Leopard Frog. Form typically frog-like; head rather pointed, its length contained about three times in the body; tympanum about as large as eye; the dorso-lateral folds are rather prominent and between the two is a pair of indistinct, much broken, dorsal folds. Color ashy, olive, or bright green above, uniform white or yellowish below; upperparts with irregular, roundish, black blotches, margined with whitish, and arranged in two irregular rows; usually two spots between the eyes; legs with two rows of spots above. Length 3 to 4 inches. A common inhabitant of the marshes about Washington and well known to the small boy as the ' ' bloodynoun " or " bloodney. " It is one of the first species to come forth in the spring and and one of the last to seek its winter quarters. Its notes appear to be somewhat variable, but during the breeding season the cry is a highly characteristic snoring croak or rattle. 20. Rana palustris Le Conte. Swamp Frog. Very similar to the preceding species but with longer head and larger glandular folds; of the latter there are distinctly two pairs, the dorso- lateral and the dorsal; the dorsal fold is continuous as far as the pelvic hump; the snout is usually obtuse and the form broader and flatter than in the preceding species. Color pale greenish brown or ashy above, yellowish white below; the blotches on the back are squarish and are arranged in two quite regular rows; sides with two rows of spots smaller than those on the back; hind legs conspicuously barred from the groin to the toes. Fairly common, but usually living singly in cool pools or in the well- Hay Batrachians and Reptiles of D. C. 131 shaded grass along the banks of some spring-fed stream. Note probably similar to that of the leopard frog. 21. Rana clamata Daudin. Green Frog. Head broad and flattened, snout rounded; eyes large and protruding; skin of back and sides more or less rough; a dorso-lateral glandular fold from eye to pelvic region, this sends a branch behind tympanum to front of arm. Color from bright green to brown, pale below; chin marbled with brown; back usually with indefinite small blotches of dark brown and similar, but larger, blotches on the sides. Length 3 to 4 inches. This species bears a very close resemblance to young individuals of the bullfrog but may be distinguished at once by the presence of the gland ular folds. It is quite common through the spring and early summer. I have found it in abundance in the series of pools between the canal and the river about 2$ miles above Georgetown. Its active movements and its strictly aquatic habits makes it a more difficult species to collect than most other species of this genus. Note, a series of three or four croaks irregularly descending in pitch. 22. Rana sylvatica Le Conte. Wood Frog. Body slender and graceful; head broad; legs long; a dorso-lateral glandular fold from eye to end of body and another from mouth to front of arm. Color grayish or light chocolate brown, sometimes with green ish; beneath, whitish; a light streak from snout to vent, glandular folds often golden yellow and tympanum coppery, a black stripe from snout to front of eye, sides and back sometimes obscurely blotched and limbs barred. An inhabitant of woodlands exclusively, this species is not often seen except during the breeding season when it sometimes occurs in great abundance in secluded ponds. During the remainder of the summer it prefers to live among the fallen leaves, where its colors correspond so perfectly with its surroundings that it is very effectually concealed. Note, a series of rapid, sharp croaks, all of abcut the same pitch. 23. Rana catesbeiana Shaw. Bull Frog. Body large and heavy; head large and broad, snout rounded; legs long and strong; skin of back nearly smooth; no dorso-lateral folds; a small glandular fold from eye to front of arm, between this and the tympanum a sharp groove. Color yellow, green, brownish or deep brown, sometimes uniform but usually with indefinite blotches of brown; sides usually with marblings of brown; underparts white or yellowish. Length 4 to 8 inches, or total length 9 to 18 inches. This, our largest frog, is very common along the larger and deeper watercourses about the city, wherever there is vegetation close to the waters edge to offer it a secure retreat. The thickets along the canal and the Eastern Branch are congenial lurking places from which it is to be taken only by frog hunters who go out at night and spear it from boats. Its roaring bass note is well known to all. 132 Hay Batrachians and Reptiles of D. C. CLASS REPTILIA. The reptilian fauna of this region includes, according to our present information, about forty species and subspecies. Of these, four are lizards, eight are turtles and the remainder are snakes. For the distinguishing characters of the two orders represented in our fauna the reader is referred to page 122. Below is given the Key to the Species of Squamata. a. LIZARDS Sides of lower jaw not independently movable, united by a bony suture in front; shoulder girdle present; eyelids and tympan um usually evident; limbs 4, in all our species. Suborder SAURIA. b. Scales on tail, or body, or both, keeled. c. Keeled scales on tail only. Cnemidophorus sexlineatus, 25. cc. Both tail and body with keeled scales. Sceloporus undulatus, 24. bb. Body and tail with smooth scales only. d. Limbs overlapping when pressed to side. Eumeces fasciatus, 26. dd. Limbs not meeting when pressed to side. Lygosoma later ale, 27. aa. SNAKES Sides of lower jaw independently movable, united by liga ment only; no shoulder girdle; no eyelids; no tympanum; limbs entirely wanting. - - ... Suborder SERPENTES. e. Non venomous species ; no pit between eye and nostril ; no fangs. /. None of the dorsal scales keeled. g. Anal plate divided. h. Scales on tail feebly keeled. - Virginia Valeria, 46. 7ih. Scales on tail smooth like those on body. i. Scales in 13 rows, color brownish. Carphophis amcenus, 28. ii. Scales in 15 to 17 rows. j. Color grass green. - - Liopeltis vernalis, 31 jj. Color black or blotched light and dark brown. k. Bluish black, with yellow collar. DiadopMs punctatus, 29. kk. Black, no collar; young blotched. Bascanion constrictor, 33. gg. Anal plate not divided. I. Black or brown, with narrow yellowish cross-lines. Lampropeltis getulus, 38. II. Brown or red, with large, dark-margined blotches. m. Chestnut or olive brown, with about 50 rhombic blotches along the back, others on sides. Lampropeltis rhombomaculatus, 37. mm. Red or brownish red, with a series of triple rings or dark bordered spots; head with yellow or red. Lampropeltis doliatus, 36. Hay Batrachians and Reptiles of D. C. 133 P. Dorsal scales more or less keeled. n. Anal plate divided. o. Scales in less than 18 rows. p. Scales in 15 rows. - Storeria ocdpitomaculata, 44. pp. Scales in 17 rows. q. Color grass green. - - Opheodrys aestivus, 32. qq. Color brownish. - - Storeria dekayi, 45. oo. Scales in more than 18 rows. r. Scales in 19 rows. - . * Regina leberis, 41. rr. Scales in 23 to 29 rows. s. Snout recurved and keeled. Heterodon platirhinos, 30. ss. Snout normal, without recurved keel. t. Ventral plates 130 to 160. u. Upper labials 8; preocular 1. Natrix sipedon, 39. uu. Upper labials 9; preoculars 2. Natrix bisecta, 40. tt. Ventral plates more than 200. v. Black, with obscure lighter blotches. Callopeltis obsoletus, 35. vv. Red, with dark bordered blotches. Callopeltis guttatus, 34. nn. Anal plate not divided. ID. Lateral stripe on 3d and 4th rows of scales. Thamnophis sauritus, 42. ww. Lateral stripe on 2d and 3d rows of scales. Thamnophis sirtalis, 43. Venomous species; a deep pit between the eye and nostril; upper jaw in front, with large, erectile, perforated fangs. x. Tail short, without a rattle, ending in a horny point. Agkistrodon contortrix, 47. xx. Tail provided with a rattle. Crotalus horridus, 48. FIG. 2. Topography of a snake, r, rostral plate; , anal plate; tral plates; #, subcaudal plates; 1, upper labials; 2, lower labials; tical plate; 4, occipital plates; 5, postf rental; 7, anteorbitals. ver- 134 Hay Batrachians and Reptiles of D. C. 24. Sceloporus undulatus (Bosc). Pine-tree Lizard. Head broad and flat, tapering to the short and rounded snout; tail slender, forming considerably more than one-half total length; plates on top of head rather large, a series of these above each eye; back of head with three plates, the median larger than the lateral and with a central translucent spot; scales of back and sides strongly keeled and pointed; those of the ventral surface smaller and smooth. Olive, brownish or nearly black, with undulating lines of dark brown across the back; a row of whitish spots on each side of the back; the tail is crosslined with black and the sides mottled with black and whitish. The males and, possibly sometimes the females, with brilliant green or blackish throat patches which are prolonged onto the arm and sides of the body. Length, 4 to 6 inches. In our region this harmless and interesting lizard is very common in the higher and drier situations. It is often to be seen during the warm days of summer basking in the sunshine on some exposed rail fence, or log. The eggs are laid early in the summer and hatch in July. By the latter part of August the young begin to shift for themselves and leave the company of the adults. 25. Cnemidophorus sexlineatus (Linn.). Six-lined Lizard. Long and slender, head pointed; top of head with a large median plate, on each side of which, over the eyes, are four smaller plates; ear about as large as eye; neck with two loose folds of skin below; front of thigh and under surface of hind leg with enlarged scales; scales of upper surface of body small, those below large and arranged in eight longitudinal rows; tail nearly twice as long as head and body and covered with large keeled scales. Olive or brownish or green, sides black, with three narrow stripes of 'yellow; a greenish stripe on back of thigh prolonged on sides of tail. Length, 6 to 10 inches. As to the occurrence of this species within our limits there is some doubt, but it has been taken at no great distance both north and south so it will doubtless be found here sooner or later. It runs with great swiftness, is difficult to capture, and very little is known of its habits. 26. Eumeces fasciatus (Linn.). Blue-tailed Skink, "Scorpion." Body slender or stoutish, head short, broad behind, snout rather blunt; ear smaller than eye; neck very short and with no loose folds of skin; tail longer than head and body; scales of back, sides and ventral surface about equal in size and smooth, but a row of enlarged scales on under side of tail.' Young and medium sized specimens nearly black above, with five yellow lines running from the head to the middle of the tail; the median line forking on the head; the extremity of the tail is often bright blue. As the animal grows older, the stripes becomes obscure, the color of the body becomes olive or brownish, and the head becomes bright red. Length, 6 to 12 inches. Hay Batrachians and Reptiles of D. C. 135 Under the name of "scorpion " this lizard is well known to the rural residents of our region, and by them is regarded as worthy of suspicion if not actually venomous. It is needless to say that there is no foundation for such a notion and the pretty animal is worthy of a better fate than usually is meted out to it. It is fairly common but its agility makes its capture difficult. They are very shy and timid and spend much of their time hidden under leaves and bark or in trees. 27. Lygosoma laterale (SayJ. Brown-back Lizard. Body elongate, tail long and slender, limbs feebly developed; head short, nearly flat above, the perpendicular sides tapering to the snout; lower eyelid with a transparent spot; ear opening as large as eye; no folds of skin on neck ; body covered with smooth scales. Olive or brownish above; abdomen greenish white; sides with a brown, white margined stripe running from eye to tail, below this, between the fore and hind legs may be a similar but narrower stripe. Length, 4 to 6 inches. A rare species within our limits, only a few specimens having been col lected in the neighborhood of Mt. Vernon and Marshall Hall. At John son's Gully I have taken three specimens, all of which were found under logs in rather damp situations. 28. Carphophis amcenus (Say). Ground Snake. Head small, snout moderately elongated and rounded; vertical plate hexagonal; prefrontals small or wanting; postorbital single; upper labials, 5, eye over third and fourth; lower labials, 6; ventral plates, 112 to 131; scales smooth and glossy, in 13 rows. Above, rich chestnut brown; below, yellow or salmon. Length, less than one foot. A not uncommon species, most often discovered by accident under some decaying log or sunken stone. It seems to prefer the rather cool and damp recesses of the thick woods, and so far as I know, is never found crawling about. I have found it frequently in Johnson's Gully and near Mt. Vernon. 29. Diadophis punctatus (Linn.). Ring-neck Snake. Head rather distinct from body, flat; snout rather broad and projecting beyond lower jaw; upper labials, 7 or 8; lower labials, 8, the fifth the largest; ventral plates, 148 to 203; scales smooth, in 15 rows. Blue-black above, orange-red below; a cream colored ring around the neck just behind the head. Length, about 10 inches. Not as common, perhaps, as the preceding, but to be looked for in the same localities. It has been taken at Mt. Vernon, Takoma Park, near Cabin Johns Bridge, etc. It can be recognized at once by its coloration. 30. Heterodon platirhinos Latreille. Hognose Snake; Spreading Adder. Body thick and heavy; head broad and short; mouth large and much 136 Hay Batrachians and Reptiles of D. C. curved; rostral plate greatly developed, trihedral, pointed and upturned; upper labials, 8; lower labials, 11; ventral plates, 120 to 150; scales keeled, except those of the outer row which are smooth, arranged in 25, or rarely 23 rows. General color, brownish, reddish, or sometimes uniform gray or black ; when not uniform there are from 20 to 30 dark dorsal blotches with an alternating series of dark blotches on the side and sometimes a third series low down on the sides; tail with bands of dark brown ex tending nearly around. A common species, well known throughout the rural districts under the names, ''black adder, spreading adder, blowing viper," etc. Its habits, which have given rise to the last two names, are exceedingly interesting and, as the snake is perfectly harmless, may be easily in vestigated. The various color phases, mentioned above, have been ob served in our region. While they are regarded by the layman as indi cating very distinct species they are known, by the herpetologist, to be nothing more than variations of the same thing. 31. Liopeltis vernalis (De Kay). Smooth Green Snake; Grass Snake. Body and tail rather long and slender, the tail forming about one-third the total length; upper labials, 7, the eye over the third and fourth; lower labials, 8, the fifth the largest; scales smooth, arranged in 15 rows; ventral plates, 125 to 140; subcaudals, 69 to 95. Grass green above, greenish yellow below, throat and lower labials yellowish white. Length, 20 inches. v There are several records for the capture of this beautiful and inoffen sive snake within our limits. In its habits it differs considerably from the next in that it seems to prefer to live among the grass and seldom if ever climbs above the ground. 32. Opheodrys cestivus (Linn.). Keeled Green Snake. Body and tail very slender, the tail forming, usually, more than one- third the total length; neck slender; head narrow and high and somewhat swollen behind, the snout projecting considerably beyond the lower jaw; eyes large; ventral plates, 150 to 165; subcaudals, 110 to 135. Scales in 17 rows, keeled, except those of outer row and some of those of the second row. Grass green above, greenish white below. Length, 30 inches. This species resembles the preceding very closely in color but may readily be distinguished by its more slender form and keeled scales. I have kept several specimens in confinement and have never seen one attempt to bite. They are often found twined about the limb of some bush, or running rapidly about through the limbs of trees or shrubbery in search of their food. 33. Bascanion constrictor (Linn.). Black-snake; Blue-racer. Body long and slender, the tail constituting about one-fourth the total length; head distinct, long, pointed, high, the crown flat; eye in a groove Hay JSatracMans and Rep tiles of D. C. 137 which runs forward to the nostril; snout rather projecting and pointed; rostral plate high, upper anteorbital large, the lower very small; upper labials 8 to 10, the sixth very large; scales in 17 (15 to 19) rows, all smooth; ventral plates, 171 to 190; subcaudals, 80 to 110. Uniform black above, greenish white or lead color below; chin, lower jaw and upper labials with more or less white. The young are grayish olive, with a row of reddish brown, black bordered spots along the middle of the back, disappearing on the tail; sides with many specks and spots of brown; head mottled and specked; below, the color is greenish white with three or four specks of brown on each scale. They begin to assume the color of the adults when about 18 inches long. Length 6 feet. This snake is very common with us and is often seen, especially among the undergrowth near streams and the margins of thick woods. It is perfectly harmless but when cornered will fight viciously. Its bite although severe is no worse than a bad scratch and will produce no more serious results. 34. Callopeltis guttatus guttatus (Linn.). Spotted Coluber. Body rather elongate and slender, the tail forming about one-sixth the total length; head narrow, snout rounded; upper labials, 8; lower labials, 11; ventral plates, 214 to 236; subcaudals, 63 to 79; scales in 27 rows, those of the five central rows faintly keeled. Color above, light red becoming paler on the sides; back with about 40 irregular dark red spots each with a darker border; below these, on the side, is another series of spots, alternately larger and smaller, the larger alternating with the dorsal spots; lower surface checkered with black and yellow; head with dark bands, one through the eye to the corner of the mouth and onto the neck, another in front of this, and a third from the back of the head to the neck, Length, 4 feet. There are two records of the capture of this snake within our limits, one at Wheaton, Md., the other at the U. S. Insane Asylum grounds. It is a southern species and is doubtless very rare here. 35. Callopeltis obsoletus obsoletus (Say). Allegheny Blacksnake. Body moderately slender, the tail forming about one-fifth the total length; head rather broad, snout blunt; rostral plate broad, snout pro jecting; upper labials, 8 (rarely 9), the sixth and seventh the largest; lower labials, 11, the sixth the largest; scales in 27 (rarely 25) rows, all but the outer ones feebly keeled; ventral plates, 230 to 250; subcaudals, 53 to 86. Black or bluish black above, usually with obsolete lighter blotches, showing principally between the scales; below, the color is yellowish becoming darker posteriorly; lower jaw and throat white. Length, 3 to 8 feet. This species is quite common, especially along Rock Creek and the Potomac River above Washington, I have seen it frequently on High Island and have taken two specimens at Takoma Park. 138 Hay Batrachians and Reptiles of D. C. 36. Lam propel t is doliatus (Linn.). House-snake; Milk-snake; Chicken- snake. Head small; upper labials 7; lower labials 9 (occasionally 10), eye over third and fourth; scales smooth, in 21 rows (rarely 19, 20, or 22); ventral plates, 184 to 214. Length, 2 to 4 feet. The coloration is variable and on it numerous subspecies have been based; three of these occur within our limits. a. Lampropeltis doliatus doliatus. (Linn.). Bright red above, with 20 to 30 triple rings, of which two black ones enclose one of white or yellow; the black rings do not pass around the body, but the anterior of one set turns forward and joins the posterior of the preceding set on the ends of the ventral plates; sides with spots alternating with the dorsal blotches. Common in all parts of the District. b. Lampropeltis doliatus triangulus. Cope. Brownish red or gray above with a dorsal series of about 55 grayish brown or chocolate colored, black bordered blotches which do not reach the ventral plates; on each side are two, more or less confluent, rows of spots, blue black in color, the lowest row situated on the first and second rows of scales; a spot of yellow on the back of the head and a band of black bordered above by yellow running from in front of the eyes back onto the neck. There are seven records of the capture of this form within our limits. c. Lampropeltis doliatus collaris. Cope. Yellowish red with a series of reddish brown, dark bordered 'spots which do not extend further down than to the second row of scales; alternating spots black, extending from the second row of scales to, and including, the ends of the ventral plates; top of head with cross bands of yellow and brown or black; sides of head with black and yellow bands. This form seems to be rarer here than either of the preceding as there are but few records of its capture. 37. Lampropeltis rhombomaculatus (Holbrook). Brown Snake. Head small, hardly distinct from the body, snout blunt; upper labials, 7; lower labials, 9; scales all smooth, in 21 rows; ventral plates about 205; chestnut brown above, becoming lighter on the sides; below, light reddish yellow, obscurely blotched with brown; on the back is a series of from 52 to 60 rhomboidal, dark brown blotches and on the sides a series of about 42 smaller, rather indistinct spots mostly alternating with those of the dorsal series; neck and back of head with a longitudinal stripe of dark brown on each side. Length, 30 to 46 inches. Until within the last few years this snake was regarded as a great rarity but it is apparently becoming quite common in our region. It was noticed first near Alexandria, then at Rosslyn, and a few seasons later began to be taken on the northern side of the Potomac. It now seems to have extended its range as far north as Woodside. It is a very handsome snake, graceful, and usually quite good tempered. Hay Batrachians and Reptiles of D. C. 139 38. Lampropeltis getulus getulus (Linn.). Chain Snake; King Snake. Body rather slender, head small; snout rather compressed and pro jecting; upper labials, 7; lower labials, 10, the fourth and fifth the largest; ventral plates, 210 to 225; scales in 21 to 23 rows, all smooth. Black above, with about 33 continuous yellow lines running across the back and forking on the sides to join the branches from the lines in front and behind; a series of small black spots on the sides alternating with the large dorsal spots. Belly checkered with black arid yellow. Length, 2 to 4 feet. Judging from the records this snake must be quite common in this vicinity, but in the course of 10 years collecting I have never been so fortunate as to secure a specimen. 39. Natrix sipedon (Linn.). Water Snake. Head rather narrow, pointed in front; upper labials usually 8, the sixth and seventh large, the eye over the fourth and fifth; lower labials, 10; scales strongly keeled, in 23 or 25 rows; ventral plates, 135 to 150; subcaudals, 60 to 75. Color variable, brownish; back and sides each with a series of large, square, dark blotches, transverse and alternating with each other; belly with brown blotches. Length 1 to 4 feet. This disagreeable but perfectly harmless snake is very abundant along the banks of our streams and small ponds. The margin of the canal in the neighborhood of High Island is a favorite haunt. 40. Natrix bisecta Cope. Washington Water Snake. . Head rather narrow, the snout bluntish; upper labials, 9, eye over third and fourth; lower labials 9; scales all keeled, in 25 rows; ventral plates, 143; subcaudals, 67. Olive brown above, with a row of small longitudinal blackish spots on the sides and two very indistinct longitudi nal stripes on the anterior half of the body; head with three light spots above; belly nearly plain. Length, 13 inches. The type specimen, which remains unique, was collected about 15 years ago in the grounds of the central station of the U. S. Fish Com mission. In this specimen the rostral plate is divided into two by a vertical fissure, a character which must be regarded as accidental. 41. Regina leberis (Linn.). Striped Water Snake. Head small, little distinct from body, snout rounded; upper labials, 7, eye over third and fourth; lower labials, 10, the fifth and sixth largest; s cales all keeled, in 19 rows; ventral plates, 140 to 150; subcaudals, 70 to 80. Chestnut brown above with three longitudinal narrow stripes of brownish black, one on the median line and one on the fifth row of scales; belly yellow with two brown bands. Length, 2 feet. Quite common along the banks of our shallow streams and ponds. It appears very early in the spring and is never found far from the water. 140 Hay Batrachians and Reptiles of D. C. 4*. Thatnnophls sauritus (Linn.)' Riband Snake. Body rather elongate, head distinct; tail forming about one-third total length; upper labials, 7 or 8; lower labials 10, the fifth and sixth large, scales distinctly keeled, in 19 rows; ventral plates, 150 to 180; subcaudals, 100 to 120. Above, the color varies from light chocolate to almost black, relieved by three stripes of greenish white or yellow; the dorsal stripe lies on the median row of scales and the adjacent half of the next row on each side; the lateral stripes are on the third and fourth rows above the ventral plates. The lower surface is greenish white without markings. Length, 2 to 3 feet. This snake is quite abundant about Washington and is to be looked for in almost any locality. It is most common, however, in low meadows or near the edges of streams. 43. Thamnophis sirtalis (Linn.). Common Garter Snake. Body varying from slender to quite stout, head distinct; tail forming about one-fourth or one-fifth the total length; upper labials, 7 or 8, eye over third and fourth; lower labials, 10; scales keeled, in 19 rows; ven tral plates, 140 to 180; subcaudals, 50 to 90. Upper surf ace varying from olive to dark brown as a ground color; a narrow, indistinct, greenish yellow line along the back, on each side of which are three series of small indistinct dark brown spots, about 70 in each row from head to vent; sides and belly greenish, but there is a lighter tint on the second and third rows of scales; ends of ventral plates each with a black blotch. Length, 2 to 3 feet. Thamnophis sirtalis dorsalis (Baird and Girard). A form of the above which has been reported from this Ipcality differs in having the dorsal stripe broad and two rows of small distinct spots on each side. The garter snakes are perhaps the most familiar and well-known of all our snakes and are to be met with in almost every part of the region. Frogs form an important part of their diet and consequently they are most abundant in moist situations and near the water. When cornered they will fight savagely but their bite is perfectly harmless. 44. Storeria occipitomaculata (Storer). Red-bellied Snake. Snout short and blunt; upper labials, 5 to 6, growing larger posteriorly, eye over third and fourth; lower labials, 6 or 7; scales in 15 rows, all keeled; ventral plates, 117 to 128; subcaudals, 43 to 50. Olive brown to reddish gray, sometimes uniform, but usually with a paler dorsal band bordered by blackish dots; sides with obscure dots; back of head with three pale blotches; belly salmon red. Length, 12 inches. Not uncommon; I have collected it at Johnson's Gully, Mt. Vernon and at various points along the Potomac above Georgetown. Hay Batrachians and Reptiles of D. C. 141 45. Storeria dekayi (Holbrook). DeKay's Snake. With the general form and proportions of the preceding; upper labials, 7, eye over third and fourth; lower labials, 7, fourth and fifth large; scales in 17 rows, all keeled; ventral plates, 120 to 145. subcaudals, 40 to 60. Grayish brown, with a clay colored dorsal band, bordered by a row of brown or black dots (occasionally a second series lower down) ; a dark patch on each side of the back of the head; belly grajash, yellowish or reddish. Length, 12 inches. This species is perhaps more common than the preceding and is less retiring in its habits. I have taken it in some numbers at the localities just mentioned. 46. Virginia valeriae (Baird and Girard). Valeria's Snake. Head small, narrow, and relatively high; snout pointed and the sides of the head perpendicular; upper labials, 6, the fifth largest, eye over third and fourth; lower labials, 6; scales in 15 rows, those on the tail feebly keeled, all the others smooth; ventral plates, 111 to 128; sub caudals, 24 to 37. Yellowish or grayish brown above, with usually scattered dots forming a faint line on each side of the back; a faint light line along the middle of each scale; beneath, the color is uniform dull yellow. Length, 8 to 10 inches. Not an uncommon species in thick woods, 47. Agkistrodon contortrix (Linn.). Copperhead. Stoutish but with rather slender neck. Head large, flat, triangular, and with the sides in front of the eyes perpendicular; upper labials, 7 or 8, none of them reaching the eye; lower labials, 10; scales in 23 rows, all keeled; ventral plates, 150 to 155; subcaudals, 42 to 52, all entire except the last 8 to 18. Hazel brown, sides with a series of 15 to 25 A shaped chestnut brown blotches which extend to or across the vertebral line; top of head coppery-red; sides of head and end of snout cream colored; belly yellowish, with 25 to 45 dark spots on each side. Length, 1 to 3 feet. This serpent is still not uncommon notwithstanding the efforts to ex terminate it. It is a frequenter of cool, damp woods and generally chooses a place where rock debris will afford it a ready refuge in time of danger. Although its bite is extremely dangerous, it is not often inflicted, the reptile seeming to prefer to run away rather than fight. I have taken two or three large specimens alive which showed no disposition to bite until after their capture. 48. Crotalus horridus (Linn.). Common Rattlesnake. Form of head and body like the preceding species but with the end of the tail provided with a jointed rattle. Upper labials, 12 to 16, none of them touching the orbit; lower labials 13 to 18; scales in 23 to 25 rows, all, except those of outer row, keeled; ventral plates, 165 to 175; 142 Hay Batrachians and Reptiles of D. C. subcaudals, 19 to 25. Ground color above, cream color to yellowish or very dark brown, with three rows of confluent irregular brown spots forming zigzag cross blotches, bordered with sulphur yellow; tail black; upper lip sulphur yellow; lower lip lighter; belly yellow with some mottlings and sprinklings of black. Length, 2 to 5 feet. The evidence as to the occurrence of the rattlesnake within our limits is not altogether satisfactory, yet it is of such a character as to make it seem extremely probable that the species is still to be found about the rock slides along the Potomac River above the city. Key to the species of Chelonia. \ a. Tail long, about equal to the plastron. - Chelydra serpentina, 49. aa. Tail much shorter than the plastron. b. Marginal plates, 22. c. Hinder lobe of plastron considerably more than as wide as the carapace. - - - - Kinosternon pennsylvanicum, 50 cc. Hinder lobe of plastron not more than as wide as the carapace. Aromochelys odwata, 51. bb. Marginal plates, 24. d. Plastron without hinge, immovably joined to the carapace. e. Plates of carapace nearly or quite smooth. /. Carapace without round yellow spots. g. Neural and costal plates alternate. Pseudemys rubriventris, 52. gg. Neural and costal plates arranged in 4 nearly straight rows across the back. _-- : ;.',*. . Chrysemys picta, 53. ff. Carapace with many round yellow spots. Ckmmys guttatus, 54. ee. Plates of carapace rough, each one with low, concentric ridges. Malaclemmys centrata, 55. dd. Plastron with a hinge across its middle and united by a movable suture with the carapace. - - Terrapene Carolina, 56. FIG. 3. Topography of a turtle, e, nuchal plate; n, neural plates; co, costal plates; m, marginal plates; g, gular plates; h, humeral plates; pe, pectoral plates; ab, abdominal plates; /, femoral plates; an, anal plates; c, carapace; p, plastron; al, alveolar surface; t, tongue. Hay Batrachians and Reptiles of D. C. 143 49. Chelydra serpentina (Linn.). Snapping Turtle. Shell high in front, low and notched behind, and with three keels, a median and two lateral, which become obsolete with age; body heaviest forward; head and neck very large, the snout narrowed forwards; jaws strongly hooked and very powerful; tail long, its upper margin with a crest of horny compressed tubercles, its ventral surface with two rows of moderate scales; plastron small, cross shaped, with nine plates be sides the very narrow bridge; claws, 5-4, strong; soft skin everywhere with wrinkles and warts; fore-arm, hands, and feet with large scales. Carapace dusky brown or black, head and neck brown, plastron and soft skin whitish or yellow. This species, it is said, sometimes attains a length of 4^ feet, but the largest specimen which I have seen in this locality was 2 feet long. It is fairly common in the marshes, ponds, and shallow waters about Washington. 50. Kinosternon pennsylvanicum (Bosc). Eastern Mud Turtle; Skillpot. Body oval; carapace with three indistinct keels, evident in the young but entirely disappearing in the adult; plastron large, almost filling the opening of the carapace, its anterior and posterior lobes movable on the fixed central portion, the anterior lobe rounded and with a single gular scute, the posterior lobe notched behind; head of moderate size. Males with two patches of sharp edged scales on the hind legs and the tail ending in a horny point. Carapace horn color or brown, plastron yellow or brown; soft skin above, brownish with yellow spots; head and neck with yellow stripes; skin of lower surfaces yellow. Length, about 5 inches. Fairly common in the marshy parts of the District. 51. Aromochelys odoratus (Bosc). Marsh Turtle; Stink Pot. Body oval but much narrower in the adults than in the young; the carapace with a more or less prominent median keel; plastron narrow, lacking much of filling the opening of the carapace, only its anterior lobe movable; gular scute single; posterior lobe of plastron notched behind; head large, snout projecting, jaws strong, the lower one hooked. Tail of males coiled at tip and furnished with a small nail. Carapace brownish or horn color, often spotted or striped with dark brown; upper surfaces of head, neck, and limbs brown, the lower surfaces paler; plastron yellow; head with two yellow stripes on each side. Length, 5 inches. This disagreeable and repulsive turtle is often met with in the marshes about the city and can usually be obtained with ease at Four Mile Run or the Bennings Swamp. It is sometimes taken by the angler on his hook. 52. Pseudemys rubriventris (LeConte). Red-bellied Terrapin; Slider. Body oval, less rounded and less elevated in the adults than in the young; alveolar surface of jaw broad and with a longitudinal, tuberculated 144 Hay Batrachians and Reptiles of D. C. ridge; jaws coarsely serrated, the upper with a prominent hook on each side of the median notch; a row of large scales along the outer edge of the fore feet; neck with scales; loose skin "between legs scaleless. Color exceedingly variable, dusky with irregular red markings above; marginal plates with much red; plastron red or partly yellowish; head and neck brown with reddish lines. Length, 6 to 8 inches. A rather common species and probably the best known, as it is brought in numbers to our markets and sold for food. It is to be looked for in the same localities as the preceding. 53. Chrysemys picta (Hermann). Painted Turtle. Shell broad and low, the surface very smooth, and with no traces of a keel; plates of carapace arranged in four, nearly straight, transverse rows; upper jaw with a small tooth on each side of the median notch, alveolar surface narrow and with a low, smooth ridge on each side for about half its length; plastron broad and flat, truncated behind. Greenish black above, the plates margined with paler; marginal plates marked with bright red; plastron yellowish, often blotched with brown; head and neck with yellow stripes. Length, 4 to 5 inches. Our most common species. It occurs in great numbers in the low marshes along the river and the Eastern Branch. An energetic collector once brought me fifty specimens of this turtle which he had secured in one afternoon in the marsh near Bennings. 54. Clem my s guttatus (Schneider). Speckled Tortoise. Shell low, broad, and without traces of keel except sometimes in the young; nuchal scute very narrow; plastron large, its posterior border slightly emarginate; upper jaw slightly notched in front, alveolar surface very narrow; legs and feet all covered with scales, those of the front limbs large and overlapping; tail long, that of the male bringing the vent beyond the end of the carapace. Color black above, often with reddish-brown tints; each scute with from one to twelve round orange spots; plastron red, orange, and black, the latter usually predominating; head black, with orange dots, neck and soft skin about legs with more or less red. Very common in about the same localities as all the preceding. 55. Malaclemmys centrata (Latreille). Diamond-back Terrapin. Carapace broad, rather depressed and rising gradually to the central ridge, the plates which cover it always with numerous concentric, low ridges; upper jaw not notched in front, alveolar surface smooth. Greenish or almost black, plates of both carapace and plastron usually with dark concentric lines. Length, 4 to 12 inches. There is a single record for this turtle within our limits and was prob ably based upon some individual which had escaped from the market. Hay Batrachians and Hep tiles of D. C. 145 56. Terrapene Carolina (Linn.). Box Turtle. Shell broadly oval, high and very convex, heavy and solid ; posterior portion of carapace with a well-defined keel in the young, this more or less'evident in the adults; plastron large, tightly closing the opening of the carapace, consisting of two lobes movable by hinges on each other and the carapace; upper jaw with a median hook but no notch, alveolar surface narrow; limbs and feet scaly; tail short. The colors of the cara pace are yellow and brown or black, variously arranged and with some times one and sometimes the other color predominating; the young have a single yellow spot on each scute; plastron with black and yellow; head, neck, limbs, and tail black, with yellow spots. Iris of males red, and the skin of the neck and legs more or less suffused with orange. Length, about 5 inches. Abundant in the woods everywhere about the city. On a single short walk in the early morning I have found 8 or 10 of this species. VOL. XV, PP. 147-149 JUNE 20, 1902 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON TWO NEW SPECIES OF POISONOUS SUMACHS FROM' THE STATES OF RHODE ISLAND AND FLORIDA. BY EDGAR A. MEARNS. Mr. Gerrit S. Miller, Jr., recently directed my attention to marked variations in the JRhus, known as Poison Oak and Poison Ivy growing in the vicinity of his residence in Virginia. As I justifiably considered myself to be practically immune to the poison of these plants, comparisons were commenced, in the course of which it developed that the Rhus radicans of Linnaeus (Poison Ivy), an extremely variable species of the North, meets the range of JRhus toxicodendron of Linnaeus (Poison Oak), a southern species, in Virginia and the District of Columbia. Rhus toxicodendron Linnaeus appears to be confined to the Southern States. It is a shrub having crenately-lobed, very pubescent leaflets. Seeds of Khus radicans which I collected in Virginia, in No vember, 1901, were found to differ from those gathered on the island of Rhode Island, near Newport, during the same month, as well as from those labelled Rhus toxicodendron Linnaeus by Mr. E. S. Steele, whose specimens (Nos. 364,079 and 364,080, U. S. National Herbarium) were collected in the District of Columbia, and kindly lent me, together with other specimens of JRhus, by the authorities of the United States National 26-BIOL. SOC. WASH. VOL. XV, 1902. (147) 148 Mearns New Species of Poisonous Sumachs. Museum. To others I am indebted for much additional material from private herbaria; also to Mr. Gerrit S. Miller, Jr., and Mr. Charles Louis Pollard, for special assistance. In the Na tional Herbarium is another entire-leaved form, from Florida, related to the Rhode Island species. Prof. Amos Eaton, in the 4th edition of his Manual,* de scribes the "poison vine" or "poison ash" under the specific name toxicodendron, and gives the following: "Var. radicans, (poison ivy) stem climbing and rooting: leaflets broad, entire, or with scattered teeth. Var. quercifolium, (poison oak) erect, low: leaflets variously sinuate-lobed. Var. microcarpon, leaflets ob long-oval, long-acuminate, sub-rhombic: fruit very small." Neither these names and descriptions nor those of later writers, Toxicodendron vulgare Miller included, are applicable to the entire-leaved littoral forms occurring on Rhode Island and in Florida, which are described below. Rhus littoralis sp. nov. RHODE ISLAND SUMACH. Type. From Newport, Rhode Island. No. 403,300 U. S. National Her barium. Specimen consisting of three parts: (1) a botanical sheet with branch, leaves, and fruits (September 10, 1901); (2) a quantity of drupes; (3) central stem (all from the same plant), 2 meters in height, cut in lengths of a little less than 4 decimeters each (November 28, 1901). Collected by the author. Original number, 233. Description. An erect, woody shrub, 2 to 4 meters in height, with stem 1 to 2 dm. in circumference. Wood fine-grained and white. Stem terete, with scattered aerial rootlets near the ground ; branches longi tudinally ridged; bark gray or reddish brown on new growth; young shoots and buds covered with a brown pubescence. Leaves thick, petioled, 3-foliate, resinous spotted, and with brownish pubescence on veins; petiole of leaf 50 to 200 mm., of lateral leaflets I to 10 mm., and of the terminal leaflets 10 to 30 mm.; leaflets oblong-lanceolate, inequi lateral, rounded at base, and acute at apex, with margin entire, 50 to 100 mm. in length by 20 to 60 mm. in breadth. Flowers green, in loose axillary panicles. Fruit a globose drupe, consisting of a flattened, striated, 1 -celled stone; a persistent, white, lobed pericarp; with a greenish or yellowish fugacious outer investment which at first is very pubescent and afterwards smoothish. In plate III are shown the dessicated pericarps of four species as they appear in winter after the epicarp has separated *A Manual of Botany for the Northern and Middle States of America, etc., Albany, 1824, p. 428. PKOC. BIOL. Soc. WASH., Vox,. XV, 1902. SEEDS OF FOUR SPECIES OF RHUS. Mearns New Species of Poisonous Sumachs. 149 from the sarcocarp and disappeared. The surface of the indurated sar- cocarp conforms to that of the endocarp, appearing quite smooth in Rhus toxicodendron (fig. 4) globosely lobed (like a canteloupe) in JR. lit- tvralis, (fig. 3) and R. floridana (fig. 2), and flattened (like a peach-stone) in R. radicans (fig. 1). Rhus floridana sp. nov. FLORIDA SUMACH. Type. No. 387,083 U. S. National Herbarium; collected by A. 8. Hitchcock, at Alva, Lee County, Florida. Sheet consisting of two twigs with leaves and fruitage. Description. Similar to Rhus littoralis, but with leaves much thinner and smoother, and with drupe (pi. Ill, fig. 2) much smaller, with epicarp shining and glabrous. Flowers small, green, in loose axillary panicles. Petioles as in R. littoralis. Entire plant not seen. VOL. XV, PP. 151-152 JUNE 20, 1902 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON DESCRIPTION OF A NEW BAT FROM COLOMBIA. BY MARCUS W. LYON, JR. The writer has recently had opportunity to examine a fine series of skins and skulls of Vampyrops lineatus collected by Mr. Wm. T. Foster, in Paraguay, the type locality of the species. A comparison of the Colombian bat, previously as sumed to be V. lineatus* with these topotypes, shows the former to be specifically distinct, though closely related to true lineatus. Until now, apparently no name has been proposed for the Colombian bat. Vampyrops umbratus new spepies. Type from San Miguel, Colombia. No. 8180, Bangs' Collection in Museum of Comparative Zoology. Skin and skull of adult male, collected June 8, 1898, by W. W. Brown, Jr. Original No. 234. General characters. Allied to Vampyrops lineatus (E. Geoffrey); differs in having a larger skull, darker coloration and with the light markings on the face and back less clear and less sharply defined. Description of type. Color above, seal-brown (Ridgway), bases of the hairs somewhat lighter, approaching a dark hair- or broccoli-brown; below, practically the same, but duller and a little lighter. A fine whitish line about 1 mm. in width runs from the back of the neck along the middle of the back to the edge of the interf emoral membrane. The *Robinson and Lyon, Proc. U. S. Nat. Mus., XXIV, p. 149, 1901. 27 BIOL. SOC. WASH. VOL. XV, 1902. (151) 152 Lyon A New Bat from Colombia. face is marked by two stripes of light broccoli-brown 1-1.5 mm. in width, one on each side extending from the posterior base of the nose-leaf, up over the eye, to nearly as far back as the occiput. The two lower face markings from each angle of the mouth to each ear, are scarcely definable. Hair covers the forearm above for about two-thirds its length, extend ing out on the membrane for about 5 mm. The underside of the forearm is scantily covered with short hairs for about the same distance and the hair extends about 10 mm. on the membrane. Interfemoral membrane scantily covered above and below, with hairs like those on the back and belly, the hairs forming a fringe on the posterior border. Length of hair on back 8 mm., on belly 5 mm. Skull. Like that of Vampyrops lineatus, but larger in every way and with heavier teeth. The last upper molar is different in shape, being disproportionately wider, from side to side, than the same tooth in V. lineatus. Measurements of type. (The figures in parenthesis are those of an adult specimen from Paraguay: Wm. T. Foster, No. 361). Forearm, 43 (46): longest finger, 104 (102); tibia, 15 (17); calcar, 5 (7); foot, 12 (12); greatest length of skull, 27.2 (25); zygomatic breadth, 15.6 (15); front of upper canine to posterior edge of last upper molar 10.4 (8.2); width of last upper molar, 1.2 (0.8); distance between outer angles of second upper molars at alveoli, 11.2 (10.4); greatest length of mandible, 18 (17). Remarks. Vampyrops umbratus needs comparison with no other species than lineatus. It is darker throughout, has a slightly smaller body and a distinctly larger skull. The color of lineatus is a broccoli- or hair-brown, in contrast to the seal-brown of umbratus. The stripe down the back of lineatus is pure white and from two to three times the width of the whitish stripe down the back of umbratus. The face stripes in lineatus are white and distinct, in umbratus they are a light broccoli- brown and ill-defined, especially the lower markings. The nose-leafs and ears show no apparent differences in the dried skins of the two species. Vampyrops umbratus is represented by three specimens, all in Bangs' Collection in Museum of Comparative Zoology; the type, No. 8180, from San Miguel, No. 8300, from San Antonio, and No. 8301, from Palamina, Colombia. The two latter are entirely uniform with the type in point of size and coloration. Unfortunately they are not accompanied by skulls. VOL. XV, PP. 153-156 JUNE 20, 1902 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON GENERAL NOTES. The Generic names of the Peccaries, Northern Fur Seal, and Sea Leopard. Dr. Gill's ruling* about the names of the Peccaries surely introduces a very dangerous principle^ which might carry us a great deal further than at first appears. Tayassu (1814) and Dicotyles (1817) each contain exactly the same species, and the later one should therefore be con sidered as a full synonym of the earlier. To permit any later restric tion of the synonymic name to reinstate it as a valid one side by side with the earlier would be a plan quite opposed to the usual custom, and one fraught with the danger of much further disturbance to nomen clature. Nothing is more likely to introduce confusion and doubt into nomen- clatural decisions than any paltering with the good old principle of "once a synonym always a synonym." Consequently I think Dr. Merriam perfectly correct in his contention that Tayassu and Dicotyles are strictly synonymous, but Fischer's Notophorus f introduces a new element into the case. In giving this name, Fischer no doubt intended merely to replace his earlier but barbarous Tayassu by a classical term, but he happened to mention only one species as belonging to it, viz: "Sus tajassu Lin. Gmel." *General Notes, supra p. 38. fG. Fisch. Mem. Soc. Moscow, V, p. 418, 1817. 28-BIOL. SOC. WASH. VOL. XV, 1902. (153) 154 General Notes. This therefore would not only be its type, but would by elimination re strict Tayassu and DicotyUs to the other species mentioned in the re spective original descriptions of those genera. Dr. Merriam's Olidosus would thus not be required. But while Linnaeus's Sus tajacu was the collared Peccary, Fischer's Sus tajassu was, as is clearly shown by the synonymy in his Zoognosia,* the whitelipped species, and that, therefore, must count as the type of Notophorus, and by elimination bring Tayassu on to the true T. tajacu. We should thus have Tayassu, Fisch. Subgenus Tayassu, Fisch. (Syn. Dicotyles, Cuv.) Species T. tajacu (type), angulatus, torvus, etc. Subgenus Notophorus, Fisch. (Syn. Olidosus, Merr.) Species T. (JV.) albirostris (type) and T. a. ringens. Doctor Allen has taken the same view about the principle of full syno nyms in the case of the Eared Seals, when showingf "that the genus Otoes, Fischer, 1817, is an exact synonym of Otaria Pe"ron, 1816," and so far I fully agree with him. But I would venture to ask if the principle on which the name Cal- lorliinus Gray, 1859, is rejected ("preoccupied by CallirMnus, Blanch- ard, 1850 ") is not now an obsolete one, and should we not therefore adopt that name. The logical outcome of the non-emendation system, nearly universally adopted in America, and now unreservedly accepted here by myself and others, is that unless two names are entirely alike, the earlier does not preoccupy the later, however closely they may resemble each other. Even if I did not go so far as in the case next following, I should still consider Callorhinus and Callirhinus as " permissible variants " of one compound, and each as valid as Picus and Pica. The extreme of this principle is reached in the case of the Leopard Seal, now commonly, but for two reasons erroneously, called Ogmorhinus, (Peters, 1875). Firstly, its original and still well-known name was spelt Stenorhinchus on its first publication in 1826,^: and I hold that the name is not invali dated by the StenorJiynchus or Stenorynchus of earlier authors. Secondly, it may be pointed out, for the benefit of those who might refuse to accept Stenorhinchus that Peter's Ogmorhinus of 1875, can in no case stand, as the genus had already been renamed Hydrurga by Gistlinl848.|| To recapitulate, I consider that the two subgenera of the Peccaries should bear the names of Tayassu, Fisch., and NotopJiorus, Fisch., the Northern Fur Seal that of Callorhinus, Gray, and the Sea Leopard, Stenorhinchus, F. Cuv. Oldfield Thomas III, p. 285. fBull. Am. Mus. N. H., XVI, p. 116, 1902. {F. Cuv. Diet. Sci. Nat. XXXIX, p. 549, 1826. llNaturg. Thierr, p. XI, 1848. General Notes. 155 Note on the Vespertiiio incautus of J. A. Allen. In my revision of the North American Bats of the family Vesper- tilionidas (North American Fauna, No. 13, October 16, 1897), I consider the Texan Vespertiiio incautus of J. A. Allen (Bull. Amer. Mus. Nat. Hist., VIII, p. 239, November 21, 1896), as identical with the Mexican bat previously described by the same author as Vespertiiio velifer. Mr. Vernon Bailey has recently asked me to identify eight bats of this group which he collected in the summer of 1901, at Carlsbad, New Mexico. They prove to be the same as incautus; and, after examining all the material now available (including the original series of incautus, kindly loaned by Dr. Allen), I regard this form as worthy of recognition by name. It should stand as Myotis incautus (J. A. Allen), and may be distinguished from M. velifer by its more pallid coloration. Oerrit S. Miller, Jr. Note on the Chilonycteris davyi fulvus of Thomas. In November, 1892, Mr. Oldfield Thomas separated the Mexican bare backed bat from the South American form under the name Chilonycteris davi/i fulvus (Ann. and Mag. Nat. Hist., 6th ser., X, p. 410) on account of the small size and "briliant fulvous chestnut " color of some specimens from Las Penas, Jalisco. Two years later (Bull. Amer. Mus. Nat. Hist., VI, p. 248, August 3, 1894), Dr. J. A. Allen showed that these bats are dichromatic, and that the red phase is not characteristic of the Mexican form. He proposed, therefore, to place the name fulvus as a synonym under davyi. Having recently examined a considerable number of specimens of Dermonotns from Mexico, as well as a topotype of D. davyi and severaL individuals of the same species from Dominica, I find that D. fulvus is a well-marked form, characterized by its noticeably smaller, narrower skull. In the Trinidad specimen the greatest length of skull is 15.8mm., in those from Dominica, 15.8-16 mm., while in the Mexican series it never exceeds 15 mm. Gerrit 8. Miller, Jr. Parus inornatus griseus renamed. Parus inornatus griseus (Ridgway, 1882), the name in current use for the Gray Titmouse, cannot, according to our present interpretation of the American Ornithologists' Union Code of Nomenclature, be continued, since there is a much older Parus griseus (Muller, Zool. Danicae, 1776, p. 34). It is therefore proposed to rename the Gray Titmouse Parus inor natus ridgwayi. Charles W. Richmond. 156 General Notes. The Hooded Warbler breeding near Washington, D. C. Although the Hooded Warbler ( Wilsonia mitrata) has been for a very few years known to be a regular but rare summer resident of the wooded slopes on the Virginia side of the Potomac near this city, there has been apparently no record of its nesting. On June 8, 1902, Mr. J. H. Riley and the writer took a young bird, evidently just out of the nest, on the laurel-covered slope of a run draining into the Potomac near Langley, Fairfax County, Virginia, i. e., nearly opposite Cabin John, Maryland. It was accompanied by both parents. Mr. Riley has observed the species resident near Falls Church, Virginia, in May, June and July; and one or two individuals have been taken in early summer near Great Falls. William R. Maxon, Washington, D. C. VOL. XV, PP. 157-158 JUNE 25, 1902 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON DESCRIPTIONS OF TWO NEW BIRDS FROM TRONG, LOWER SIAM.* BY CHARLES W. RICHMOND. Among the birds collected by Dr. W. L. Abbott in the province of Trong, Lower Siam, and forwarded to the U. S. National Museum, are two apparently new forms which are here described. The first, a species of Stachyris, was sent for deter mination to Dr. R. Bowdler Sharpe, who considered it new and nearly related to 8. chryscea. The latter species is not repre sented in the Museum collection, hence a direct comparison has not been made. Stachyris chrysops new species. Type. Adult male, No. 160,566, U. S. National Museum, summit of Khow Nom Plu, 3000 feet, Trong, Lower Siam, February 22", 1897, Dr. W. L. Abbott. Nape, back, scapulars, outer margins of wing and tail feathers, upper tail-coverts and rump, brownish olive green, paler and more yellowish on the latter; forehead, orange-ochraceous; crown and occiput ochre yellow, streaked with black; lores and a narrow rictal streak, black; throat, breast, abdomen, and bend of wing, ochre yellow, deeper on the throat; under tail-coverts, sides of body, sides of neck and ear-coverts, brownish wax yellow; under wing-coverts, axillaries, *Published by permission of the Secretary of the Smithsonian Institu tion. 29 BIOL. Soc. WASH. VOL. XV, 1902. (157) 158 Richmond New JSirds from Trong. and inner margins of wing feathers, pale Naples yellow; tail grayish olive green; wing feathers grayish black. Length (in the flesh), 121 mm.; wing (in skin), 50 mm.; tail, 47; tar sus, 18.5; culmen, 12 (bill, from gape, 15). "Iris red; feet yellowish brown; upper mandible black; lower mandi ble leaden, pink at base." Oreocincla horsfieldi af finis new subspecies. Type. Adult male, No. 169,982, U. S. National Museum, Know Nok Ram, 3000 feet, Trong, Lower Siam, January 13, 1899, Dr. W. L. Abbott. Similar to 0. horsfieldi, but has a smaller bill, longer wing, somewhat brighter colors, is more heavily mottled above, and has a different wing formula. According to Seebohm the "second primary is intermediate in length between the fifth and sixth, or sixth and seventh" in 0. hors- fieldi*, which statement agrees with a specimen in the National Museum collection; the fourth and fifth primaries are equal and longest. In the new form the second primary falls between the third and fourth, with the third longest, thus agreeing with 0. varia. Length (in flesh), 2C7 mm.; wing, 142; tail, 93; tarsus, 34; culmen, 23 (bill, from gape, 30.5); "Iris, dark brown; feet, pale brownish fleshy; bill, dark horny brown, pale fleshy beneath at base." The wing of 0. horsfieldi is given as 138 mm. (Sundevall, Journ. fur Orn., 1857, 161), and from 133.5 to 137 mm. (Seebohm, Catal. Birds Brit. Mus., V. 153). *Catal. Birds Brit. Mus., V, 153. VOL. XV, PP. 159-160 JUNE 25, 1902 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON DESCRIPTION OF A NEW SUBSPECIES OF STENOP- SIS CAYENNENSIS FROM CURAgAO.* BY CHARLES W. RICHMOND. The Goatsucker described below was presented to the U. S. National Museum by Capt. Wirt Robinson, U. S. A., who obtained it on the Island of Curacao, in June, 1900. Birds of the genus Stenopsis have been previously reported from Curacao by Messrs. Petersf and Hartert, but in each case the collector obtained only females, which Hartert was unable to separate from S. cayennensis. Capt. Robinson's specimen is a male, in fresh plumage, and may be described as follows: Stenopsis cayennensis insularis new subspecies. Type. Adult male, No. 175,245, U. S. National Museum, Curasao, June 30, 1900, Capt. Wirt Robinson, U. S. A. (collector's No. 1446). Upper parts, pale drab gray, somewhat lighter on head, all of the feathers finely vermiculated with buff, brown, or black; feathers of median part of head with broad black centers, edged with pale cinna mon; scapulars, black, the inner webs mottled with rufous, the outer *Published by permission of the Secretary of the Smithsonian Institu tion. fJourn. fiir Orn., 1892, 87. jlbis, 1893, 319. 30 BIOL. Soc. WASH. VOL. XV, 1902. (159) 160 Richmond New Subspecies of Steno2isis Cayennensis. webs with buff; feathers of rump and upper tail-coverts with black shaft streaks, the latter with indications of black cross-bars; nape, pale cin namon; wing-coverts and tertiaries similar in color to the back, the former spotted with buff and pale cinnamon; primary coverts, clove brown, with Mars brown markings on outer webs. Chin, throat, abdo men, and narrow line on forehead, white; lores, a narrow line over and under eye, ear-coverts, and sides of neck, pale cinnamon, with black and white markings; a patch of feathers on sides of throat, cinnamon, tipped with black; breast and sides of body, pale cinnamon, mixed with white and more or less barred and mottled with black; under wing-coverts, Mars brown, mottled with black, the innermost feathers, and axillaries, white. Under tail-coverts, deep brownish buff; middle tail feathers, drab gray, with fine dusky vermic illations and crossed by about nine irregular blackish bars; next three pairs of feathers mainly blackish on outer webs and tips (basal third mottled with pale cinnamon and black), inner webs white, with an oblique black bar near the middle of the feather; outer pair of feathers white, with an oblique black bar on the inner webs, and a small mottled patch at tip. Outer primaries, clove brown, with a grayish shade on outer webs and tips; a broad white bar across the four outer feathers; inner primaries, clove brown, with Mars brown markings on the outer webs; secondaries, blackish, with white tips, tinged with pale cinnamon on the innermost; outer webs mottled with Mars brown. Wing feathers at base, white, mottled with pale cin namon. Wing, 130 mm.; tail, 113mm.; tarsus, 18mm.; culmen, 12mm. This subspecies is closely related to 8. cayennensis, but is smaller and paler, of clearer color, with darker under tail-coverts. VOL. XV, PP. 161-165 AUGUST 6, 1902 PROCEEDINGS or THE BIOLOGICAL SOCIETY OF WASHINGTON SIX NEW SKUNKS OF THE GENUS CONEPATUS. BY C. HART MERRIAM. The seemingly inexhaustible collections of Mexican mammals made by E. W. Nelson and E. A. Goldman for the Biological Survey of the U. S. Department of Agriculture contain large series of skunks of all three genera Mephitis, Spilogale and Conepatus. Among the latter are several that appear to be new. One of these ( Conepatus tropicalis) has two dorsal stripes and is closely related to C. mapurito from Bogota. It belongs, apparently, to Gray's Marputim, which seems to be a valid sub- genus. The others are characterized by a single broad dorsal band and by cranial peculiarities that seem to be of sufficient weight to demand independent subgeneric recognition. The group may be characterized as follows: Submenus Oryctogale nob. Type. Conepatus leuconotus (Licht.) from Vera Cruz, Mexico. Characters. Snout very long, overhanging the incisors a full inch (25 mm.), with a large naked pad on upper side for rooting in the ground; soles of hind feet entirely naked to heel. Color black, with a single white median dorsal band, beginning on top of head between ears (or midway between eyes and ears) and reaching posteriorly to tail; tail 31-BiOL. Soc. WASH. VOL. XV, 1902. (161) 162 Merriam Six New Skunks, white except at base underneath, which is black.* Skull relatively high; anterior nares large and obliquely truncate (broadly open as seen from above); ascending arm of premaxillie reduced to a slender oblique splint; nasals and maxillae ending posteriorly on essentially same plane; anterior part of nasals (in profile) essentially in fronto-nasal plane (not upturned or pugged); postorbital constriction slight; antorbital foramen single throughout. The new forms are: Conepatus leuconotus texensis subsp. nov. Type locality. Brownsville, Lower Rio Grande, Texas. Type No. Iff II $ a( ^-> U. S. National Museum, Biological Survey Collection. July 20, 1892. F. B. Armstrong. Original No. 70. Range. Coast strip of Texas from Rockport, Aransas County, to mouth of Rio Grande; may extend southward into Tamaulipas. Characters. Similar to leuconotus but somewhat larger, with white dorsal area more restricted, usually contracting on rump to a narrow stripe and in some specimens disappearing altogether, leaving the rump largely or wholly black. Usually a narrow white line connects the white dorsal area with the white of the tail. Black of underside of tail more extensive than in leuconotus, usually spreading over sides of base, and reaching out on underside at least one-third and often one-half the distance from base to tip [in leuconotus ending opposite toes of out stretched hind feet]. Measurements. Average of 4 males from type locality: total length 825; tail vertebra? 365; hind foot 86. Average of 4 females from same place: total length 708; tail vertebrae 292; hind foot 76. Largest skull of $ : basal length 77; zygomatic breadth 55.5; palatal length 35; upper molar on outer edge 10. Largest skull of 9 '- basal length 66; zygomatic breadth 48; palatal length 31; upper molar on outer edge 9.5. Conepatus sonoriensis sp. nov. Type locality. Camoa, Rio Mayo, Sonora, Mexico. Type No. 95,914, 9 yg. ad., U. S. National Museum, Biological Survey Collection. October 29, 1898. E. A.Goldman. Original No. 13,213. Range. Northwestern Mexico from Valparaiso Mountains, Zacatecas, northwesterly over Sinaloa (specimens from Santa Cruz de Alaya) and into Sonora for an unknown distance (specimens from Camoa, Rio Mayo). Characters. Size large (hind foot in $ about 85; in 9 about 80); *Sometimes a few black hairs are inconspicuously mixed in the white of the tail. In some specimens of C. leuconotus texensis the white dorsal band fails on the rump; with this exception the color-pattern as above described is constant throughout the group. Merriam Six New Skunks. 163 white dorsal band broad throughout; tail long, the black at base be neath very short, falling short of outstretched feet; skull long and rela tively slender; f rentals even more elevated than in leuconotus; upper molar and sectorial large. Remarks. In cranial characters the relationship of this species to G, leuconotus, of eastern Mexico, is surprisingly close, particularly in view of the fact that their ranges are on opposite sides of the continent, sepa rated by a broad interval inhabited by a very different species. Un fortunately no skull of the male from the type region is available for comparison. Skulls of females resemble those of female leuconotus and texensis but are narrower throughout, with narrower anterior nares and shallower postorbital constriction. Measurements. $ yg. ad., from Valparaiso Mountains, Zacatecas: total length 715; tail vertebrae 30*0; hind foot 84. Average of 3 young adult 9 topotypes: total length 647; tail vertebrae 276; hind foot 79. Skull of type specimen ( 9 yg- ad.): basal length 66.5; zygomatic breadth 43; palatal length 33; upper molar on outer edge 8.5. Conepatus mesoleucus mearnsi subsp. nov. Type locality. Mason, Mason County, Texas. No. fJlf, 9 ad - Merriam Collection. February 20, 1886. Ira B. Henry. Range. Tableland of Mexico and adjacent southern part of the United States, from Jalisco (Guadalajara and Zapotlan) and southwestern San Luis Potosi northward to central Texas (Llano and Mason) and central Arizona (Phoenix and Fort Verde). Specimens from Colima appear to be nearer mesoleucus than mearnsi. Characters. Similar to mesoleucus but skull and molar teeth decidedly smaller. Measurements. $ yg. ad., from Boquillas, Texas: total length 670; tail vertebrae 290; hind foot 75. Adult 9 from same place: total length 580; tail vertebrae 220; hind foot 70. Average of 3 females from Hacien da La Parada, San Luis Potosi: total length 628; tai-1 vertebrae 243; hind foot 72. Conepatus filipensis sp. nov. Type locality. Cerro San Felipe, Oaxaca (altitude 10,000 feet). No. 68,172, $ ad., U. S. National Museum, Biological Survey Collection. August 24, 1894. E. W. Nelson and E. A. Goldman. Original No. 6,619. Range. Forests of pine and oak on summit of Cerro San Felipe from altitude of 9000 up to 10,500 feet. Characters. Size small (hind foot in $ only 74); pelage soft, full and woolly; dorsal white stripe rather narrow, especially on shoulders; skull small, depressed and flattened anteriorly, the profile sloping rather evenly 164 Merriam Six JVew Skunks. from middle of parietals to front of nasals; anterior nares small; postor- bital constriction strongly marked; zygomata broadest posteriorly; upper molar and sectorial smaller than in any known species except pedic a Ins. Remarks. Conepatus filipensis appears to be a high mountain species. Externally it differs from its relative of the surrounding low country, C. mesoleucus, in much smaller size, softer, fuller pelage, and the restriction of the white dorsal area. The skull and molariform teeth are so much smaller as not to require comparison. Measurements. Type specimen ( $ ad.): total length 630; tail vertebra 3 , 265; hind foot 74. Skull of type specimen: basal length 64; zygomatic breadth 47.5; palatal length 31; length of upper molar on outer edge 7.5. Conepatus pedicu^us sp. nov, Type locality. Sierra Guaclalupe, Coahuila, Mexico. No. 116,953, $ yg. ad., U. S. National Museum, Biological Survey Collection. April 25, 1902. E, W. Nelson and E. A. Goldman. Original No. 15,123. Characters. Size smallest of the known species (hind foot only 61); skull and teeth smallest (in male smaller than in female of C. mesoleucus mearnsi); skull short; nasals very broad and flat, ending posteriorly in a straight line; anterior nares smallest: braincase strongly elevated; fronto- nasal region flat, smooth, sloping strongly down ward, 'in profile a straight line. In the type and only specimen, there is a median circular bulge (probably abnormal) on the posterior half of the frontals. Remarks. Owing to its small size this species requires no direct com parison with others. The only one approaching it in the small size of the teeth is C.felipensis from Cerro San Felipe, Oaxaca. Measurements. Type specimen ( $ yg. ad.): total length 485; tail ver tebrae 178; hind foot 61. Skull: basal length 57; zygomatic breadth 43; palatal length 27; upper molar on outer side 7. Subgenus Marputius Gray. Conepatus tropicalis sp. nov. Type locality. Motzorongo, Vera Cruz, Mexico. Type No. 63,650, $ ad., U. S. National Museum, Biological Survey Collection. February 26, 1894. E. W. Nelson and E. A. Goldman. Original No. 5,903. Range. Tropical east coast of Vera Cruz and thence southeasterly to Yucatan. Limits of range unknown. Characters. Back with two rather narrow parallel white stripes (about 30 mm. in width, and separated by a still narrower internal stripe) reach- from nape (where they unite in a nuchal patch with convex front) to hips (where they taper off to a point); tail very short, white except on basal part which is black all round, the black covering about one-third of the vertebrae above and one-half below, and presenting a V-shaped indentation on median line above. Merriam Six New STcwriks. 165 Cranial character*. In addition to the cranial characters mentioned in the above diagnosis of the subgenus Marputiutt, it may be remarked that the skull has a "humpy" appearance, due chiefly to the bulging frontals and depth of the postorbital constriction. The profile is sinuous, pre senting three prominences and two depressions. The prominences are formed by the slightly upturned tips of the nasals, the interorbital bulge of the frontals, and a posterior prominence on the frontals behind the constriction. Remarks. So far as I am aware this is the only striped-backed Cone,- put-us inhabiting Mexico. All the other species of the genus have a single broad white band down the back and belong to the subgenus Oryctogale. Conepatus tropicalis differs strikingly in external characters and moder ately in cranial characters from both Conepatus proper ( C. humboldti from Patagonia) and the white-backed Mexican species here subgenerically distinguished under the name Oryctogale. In most respects it is inter mediate between the two. Externally it has the large nose pad and naked soles of Oryctogale, but the dorsal stripes and tail differ from either. In cranial characters it favors Conepatus. The fronto-nasal region is depressed less than in C. humloldti but much more than in Oryctogale; the anterior nares are intermediate in obliquity of truncation; the antorbital foramen opens anteriorly by two or more pores as in C. hum loldti, and the postorbital constriction is as deep as in that species. Measurements. Type specimen: total length 575; tail vertebrae 129; hind foot 74. Skutt: basal length 67.5; zygomatic breadth 50.5; palatal length 32.5; length of last upper molar on outer side 8. VOL. XV, PP. 167-172 AUGUST 6, 1902 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON FOUR NEW ARCTIC FOXES. BY C. HART MERRIAM. The White or Arctic fox of Lapland* and Siberia was de scribed by Linnaeus in 1758 under the name Canis lagopus (Syst. Nat., Ed. 10, p. 40, 1758). A related form from Iceland (the Sooty fox of Pennant), was named Canis fidiginosus by Shaw in the year 1800 (Gen. Zool., Vol. I, pt. II, p. 331, 1800). So far as I am aware no other names for members of the group were proposed until 1898, when Barrett-Hamilton and Bonhote in a joint paper described a small form from Spitzbergen as subspecies spitzbercjenensis (Ann. & Mag. Nat. Hist., 7th ser., I, p. 287, April, 1898). But they state that they are unable to distinguish the Spitzbergen form from the one from Iceland named fuliginosus by Shaw a century earlier. If the two are the same, spitzbergenensis of course becomes a synonym of fuliginosus. In 1900 I described a small Arctic fox from Hall Island, Ber ing Sea, under the name hallensis (Proc. Wash. Acad. Sci., II, p. 15, March 14, 1900). The material at present available for study is utterly insuf ficient to admit of a satisfactory revision of the group. Barrett- Hamilton and Bonhote had few if any skins with skulls from *The type locality may be restricted to Lapland. 32-BioL. Soc. WASH. VOL. XV, 1902. (167) 168 Merriam Four New Arctic Foxes. the old world, and none at all from America. Our National Museum is decidely better off, especially with reference to material from Alaska and Labrador, but we still lack reliable flesh measurements, and also lack series of skins and skulls put up according to modern standards. An examination of the specimens in the collections of the Biological Survey and the National Museum shows that several unrecognized forms exist. The animal inhabiting the mainland of Alaska and Canada from the mouth of the Yukon and Point Barrow easterly to Hudson Bay and Cumberland Island differs both from true lay opus of Scandinavia and from the animals inhabiting islands in Bering Sea.* The most perplexing Arctic foxes I have seen are from Pribi- lof Islands in Bering Sea.f The skulls from St. Paul Island are of three sizes; a very large and long skull with long and rather slender rostrum, like a red fox; a very small skull with small teeth (smaller even than our specimens of haUen-sis)\ and one which is intermediate in size and characters between the others and which agrees essentially with the common Arctic fox of the Alaska mainland. The collections contain a number of skulls of each of these three forms, though the number of the smallest size is much less than of the others. What do these facts signify? Three theories occur to me: (1) that the large skulls represent a large resident species while the two others are stragglers from St. Matthew [or some other] Island and the mainland respectively, reaching the Pribilofs by means of the pack ice; (2) that the large skulls represent a large resident species; the small ones stragglers from St. Matthew [or some other] Island, while the middle sized ones are hybrids between these two; (3) that all three belong to a single species which presents extraordinary and unprecedented variations in size. Personally, I believe that the large animal is a well marked *A specimen in the National Museum from Cumberland Island is es sentially identical with specimens from Alaska except that the lower premolars are somewhat larger, in this respect resembling the Labrador form, which is here described as subspecies ungava. fThe National Museum has a series of skulls and several skins from St. Paul Island, collected by Dr. F. W. True, Dr. D. W. Prentiss, and Mr. Wm. Palmer; and during my visit to the islands in 1891 I secured several specimens on St. George Island. Merriam Four New Arctic Foxes. 169 insular species; the small one a straggler from Hall or St. Matthew [or some other] Island, the others hybrids between these two, or in some cases perhaps, stragglers from the main land. On this assumption I have named the large form pribi- lofensis. In winter the pack ice from the north often reaches St. Matthew Island and sometimes pushes south to the Pribilof s. The distance between St. Matthew and St. Paul is about 225 statute miles; that between Nunivak (from which we have no specimens) and St. Paul is even less. But wherever the animals come from, the fact is well-known that when the pack ice reaches the Pribilofs, white foxes come ashore on the Islands. They have been seen to do this repeatedly. When observed, they are pursued and killed, if possible, in order to prevent in terbreeding with the native blue foxes. On Bering Island, on the Siberian side of Bering Sea, there appear also to be two forms: the mainland species (lagopus) and a large insular species here named beringensis. Pelages. The characteristic markings of the Arctic foxes are: In winter pelage, white throughout, the only marking be ing the small black pad on end of nose; in summer pelage, head, back, a cross-bar over shoulders, outer side of fore and hind legs, upper surfaces of fore and hind feet, and stripe on upper- side of tail, brownish dusky, usually darkest on top of head and rump; face strongly, feet moderately mixed with white hairs; ears strongly edged with white; chin grayish dusky; underparts soiled whitish or buffy, becoming strongly buffy on flanks. The Blue fox of the Pribilofs is a sooty-blue all over at all seasons. tiexual differences. As usual among foxes the males are somewhat larger than the females. Among the Arctic foxes the difference in size of skull is usually not great, but the large upper molar, and the carnassial above and below are noticeably larger in the males. The Arctic foxes here recognized in North America (including Bering Island but excluding Greenland) are as follows: Vulpes lagopus (Linn.). Type locality Lapland. Believed to occur on Bering Island. Vulpes lagopus innuitus nob. Type locality Point Barrow, Alaska. Ranges from Alaska to Baffin Bay. 170 Merriam Four New Arctic Foxes. Vulpes lagopus ungava nob. Type locality Ungava, Labra dor. Vulpes hallensis (Merriam). Type locality Hall Island, Ber ing Sea. Occurs on St. Matthew Island also. Vulpes pribilofensis nob. Type locality St. George Island, Pribilof Islands, Bering Sea. Occurs on St. Paul Island, also. Vulpes beringensis nob. Type locality Bering Island, Bering Sea. Vulpes lagopus innuitus subsp. nov. Type from Point Barrow (Karogar River), Arctic Alaska. No. 107,020, 9 ad., U. S. National Museum, Biological Survey Collection. June 27, 1898. E. A. Mcllhenny. Original No. 831. Characters. Similar to lagopus in size and general characters, but braincase broader and more pyriform, and tapering much more abruptly behind broadest part; nasals much broader. Contrasted with pribilofen sis it is so much smaller and shorter as not to require close comparison. Contrasted with hallensis it maybe distinguished by the following char acters: nasals broader; median frontal sulcus deeper; rostral constriction less pronounced; lower premolars larger (most noticeable in third and fourth) ; first upper molar larger. Contrasted with ungava the frontal shield is flatter and the postorbital processes are heavier and more deeply excavated posteriorly. Cranial measurements. Skull of type specimen 9 a d.: basal length 117; zygomatic breadth 09; palatal length 60; postpalatal length 50; breadth of rostrum at second premolar 23.5; upper carnassial (on cin- gulum) 12.5; first upper molar (tranverse diameter from notch on outer side) 10. Vulpes lagopus ungava subsp. nov. Type from Fort Chimo, Ungava, Labrador. No, 23,195, [ $ ad.] U. S. National Museum. Collected by L. M. Turner. Original No. 2,302. Characters Similar to innuitus but slightly larger; rostrum slightly longer; frontal shield less flat, swollen anteriorly on each side of median sulcus; postorbital processes more pointed, less massive, and less deeply excavated posteriorly; under jaw much broader vertically and deeply bellied under sectorial. Compared with Vulpes lagopus from Lapland and Sweden, ungava differs as follows: postorbital processes more decurved and pointed and less excavated posteriorly; braincase broader and more pyriform (broadly inflated on plane of posterior edge of posterior root of zygoma, behind which it tapers much more abruptly); frontal shield less flattened; nasal broader; crown of fourth lower premolar larger and longer: base of Mcrriam Four New Arctic Foxes. 17 1 skull decidedly shorter, as shown in shorter basioccipital and basisphenoid and in the distance from foramen magnum to plane of upper sectorial; similarly, the lower sectorial is nearer the condyle. Under jaw much broader vertically and more bellied under sectorial. Measurements. Skull of type specimen: basal length 118; zygomatic breadth 70; palatal length 63; postpalatal length 55; breadth of rostrum at second premolar 26; upper carnassial (on cingulum) 12.5; first upper molar (transverse diameter from notch on outer side) 9.5. Vulpes pribilofensis sp. nov. Type from St. George Island, Pribilof Islands, Bering Sea. No. 42,624, $ ad., U. S. National Museum, Biological Survey Collection. August, 1891. C. Hart Merriam. Characters. Largest of the lagopus group. Skull much elongated, resembling that of a Red fox more than that of the Arctic foxes; rostrum long and set far forward, the postpalatal length exceeding that of any other known form. In the type specimen the frontals are narrow anteriorly and the postorbital processes are only slightly developed. In skulls from St. Paul Island the frontal shield is broader and the postor bital processes are much more prominent. Contrasted with skulls of Vulpes from the mainland (St. Michaels and Lower Yukon region) the differences are very marked. F. pribilofensis* is larger throughout but the difference is most marked in the total length and zygomatic breadth, and in the greater production of the rostrum. The rostrum is not only longer but is set much farther forward so that the postpalatal length is very much greater. In most specimens the palatal and ppstpalatal lengths are subequal, but in one or two skulls of old males from St. Paul Island the palatal length considerably exceeds the postpalatal. [In F. innuitus the postpalatal length is always very much less than the palatal.] The brain- case is longer and flatter; the sagittal crest more strongly developed poste riorly; the frontal hump (at base of nasals) more marked. The teeth are essentially the same as in innuitus. The species requires no comparison with the small F. Jiallensis from Hall and St. Matthew islands. Cranial measurements. Skull of type specimen ($ ad.): basal length 127; zygomatic breadth 72; palatal length 66; postpalatal length 61 ; breadth of rostrum at second premolar 25; upper carnassial (on cingulum) 12.5; first upper molar (transversed diameter from notch on outer side) 9.5. An old $ from St. Paul Island: basal length 130; zygomatic breadth 78; palatal length 67; postpalatal length 62; breadth of rostrum at second premolar 26. Vulpes beringensis sp. nov. * Type from Bering Island, Bering Sea. No. 47,109, [ 9 yg. ad.], U. S. National Museum, Biological Survey Collection. June 3, 1892. B. W. Evermann. 172 Merriam Four New Arctic Foxes. Characters. Size large, nearly equalling pribilof ens-is; rostrum broad and massive; frontals high anteriorly and swollen on each side of median sulcus; premolars large and swollen. Compared with lagopus the skull is decidely larger; the rostrum and nasals broader; frontals more elevated anteriorly; molars and premolars larger. Compared with pribilofensis, with which it nearly agrees in size, the rostrum is shorter and broader; rostral constriction or 'step' much more pronounced; frontals anteriorly more elevated; base of cranium shorter; molars and premolars decidedly larger and more swollen. Cranial measurements. Skull of type specimen [ 9 yg- ad.] : basal length 123; zygomatic breadth 71; palatal length 66; postpalatal length 56; breadth of rostrum opposite second premolar 27. An old 9 f rom type locality: basal length 128; zygomatic breadth 77; palatal length 68; postpalatal length 60; breadth of rostrum at second premolar 26. Upper carnassial (on cingulum) in young $ 13.5; first upper molar (transverse diameter from notch on outer side) 11. VOL. XV, PP. 173-175 AUGUST C, 1902 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON TWO NEW MALAYAN MOUSE DEER.* BY GERRIT S. MILLER, JR. The series of Mouse Deer in the United States National Museum contains two hitherto undescribed forms, one from the Malay Peninsula, the other from Borneo. They may be named and defined as follows: Tragulus ravus sp. nov. Type. Immature! female (skin and skull) No. 83,506 United States National Museum. Collected in Trong, Lower Siam, September 6, 1896, by Dr. W. L. Abbott. Characters. A member of the kancMQ group resembling Tragulus pallidus from Pulo Laut, North Natuna Islands, but larger, the general color less pallid, legs more red, and dark nape stripe less well defined. Color. Back and sides ochraceous buff heavily clouded by the black ish hair tips, so that the general effect is a yellowish gray, darker on back and shoulders, where black is distinctly in excess, noticeably lighter on sides. Legs, except for the usual white markings, ochraceous, ^Published here by permission of the Secretary of the Smithsonian Institution. fUterus contained a full grown fetus. Milk premolars still in place, though very much worn. JFor use of this name see Stone and Rehn, Proc. Acad. Nat. Sci. Philadelphia, 1902, pp. 128-131. June 4, 1902. 33-BiOL. Soc. WASH. VOL. XV, 1902. (173) 174 Miller Ttco New Malayan Manse Deer. slightly paler than that of Ridgway. A narrow, very indistinct line of this color divides color of sides from the pure white of underparts. In axillary region this line widens to cover sides of chest and send back an irregular and variable streak along middle of belly. Neck ochraceous, less bright than that of legs and much sprinkled with black. Nape- band ill-defined, black, considerably mixed with ochraceous. Throat markings normal, the dark bands like sides of neck and well defined collar slightly paler. Head and face like sides of neck, but ochraceous less bright and black hairs more numerous. Tail white beneath and at tip, dull ochraceous above. Skutt. The skull shows no characters worthy of note, except that in size it distinctly exceeds that of Tragulus pallidus. Measurements. External measurements of the type: total length, 535; head and body, 470; tail vertebrte, 65; hind foot, 120; hind foot without hoofs, 108. Weight, 1.9 kg. An adult male from the type locality: total length, 520; head and body, 445; tail vertebrae, 75; hind foot, 112; hind foot without hoofs, 102. Skull of type: greatest length, 94; basal length, 86; basilar length, 82; occipito-nasal length, 89. Skull of adult male: greatest length, 93; basal length, 85; basilar length, 80; occi pito-nasal length, 86. Specimens examined. Eight, all from the type locality. Remarks. This species is readily distinguishable from Tragulus kan- chil by its pallid coloration, and from T. pallidus by its greater size. Tragulus borneanus sp. nov. Type. Aclult male (skull only) No. 34,924 United States National Mu seum. Collected on the Suanlamba River, British North Borneo, by C. F. Adams. Characters. In general similar to Tragulus napu from Sumatra, but smaller and darker; toothrow longer than in the Sumatran animal. Color. Through preparation by means of an acid preservative fluid the skins examined have been altered in color by the removal of most of the red and yellow tints. The color pattern, however, remains un changed. Underparts as in Tragulus napu, the middle of belly with a dusky wash (probably rusty before action of acid) on tips of hairs. Up- perparts with decidedly more black than in T. napu, the crown and nape stripe nearly clear black, and the black clouding of the back much more noticeable, in this respect suggesting T. pretiosus. Skull and teeth. The skull is distinctly smaller than that of Tragulus napu, so much so that those of immature individuals suggest those of T. kanchil. In general form there appear to be no tangible differences. The ascending portion of the premaxillary is broader than in the Suma tran animal, and the horizontal portion less produced forward. Teeth actually as well as relatively larger than in T. napu, the difference par ticularly noticeable in the lower premolars. Miller Two New Malayan Mouse Deer. 175 Measurements. External measurements* of adult female from Sapa- gaya River, British North Borneo (No. 19,213 U. S. Nat. Mus.): total length, 600; head and body, 520; tail vertebrae, 80; hind foot, 132 (118). Cranial measurements of type: greatest length, 107 (115)f; basal length, 102 (108); basilar length, 96.5 (100); occipito-nasal length, 101 (104); zygomatic breadth, 49 (49); least interorbital breadth, 29 (32); mandible 84 (90); maxillary toothrow, 42 (37.4); mandibular toothrow, 48 (42.4). Specimens examined. Seven skins and skulls, all from British North Borneo. Remarks. That the Bornean Napu is distinct from that of Sumatra might readily have been suspected in view of the known tendency of the group to vary. Actual comparison shows that the animals are read ily separable. *From well made skin. fMeasurements in parenthesis are those of an adult male Tragulus napu from Tapanuli Bay, northwestern Sumatra. VOL. XV, PP. 177-179 AUGUST 6, 1902 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON FOUR NEW PLANTS FROM NEW MEXICO.* BY CHARLES LOUIS POLLARD AND T. D. A. COCKERELL. The mountain ranges of New Mexico, numerous and largely isolated as they are, offer unusually favorable conditions for the development of local types of plants and animals. Only a few of these ranges have been explored for plants with any degree of thoroughness, and even these are still yielding novel ties whenever visited. The Sandia Mountains, within sight of Albuquerque, were visited long ago by Bigelow, but have since then been strangely neglected. Miss C. Ellis recently obtained a small series of plants in these mountains, and we find among them such conspicuous novelties as the Primula and Achillea herewith described. The Las Vegas Range, being really con tinuous to the north with the mountains of Colorado, would not be expected to have a peculiar flora; but as a matter of fact many of the plants are quite different from their congeners in Colorado. This statement is made with some degree of con fidence, because the difference is seen in many conspicuous mem bers of the flora, and is not easily overlooked; moreover, the junior writer of this paper spent three years at the foot of the Sangre de Cristo mountains in Colorado, and became familiar with the plants of that region. ^Published here by permission of the Secretary of the Smithsonian In stitution. 34-BiOL. Soc. WASH. VOL. XV, 1802. (177) 178 Pollard and CocJterett New Plants from New Mexico. A Primula and a violet are here described from the Las Vegas Range, the former from the Hudsonian zone, the latter from the Canadian zone. Viola WiJmattae. Acaulescent, low (5-7 cm. high at flowering), tufted, the scape barely exceeding the foliage; leaves practically glabrous, cordate in general outline, all palmately cleft and lobed with numerous divisions; stipules lanceolate, scarious; scapes bibracteale at about the middle; flowers deep violet, 2 cm. broad; sepals lanceolate, slightly scarious-margined; petals narrowly oblong, well bearded : fruit not observed. Type No. 404,924, in the United States National Herbarium, collected in Sapello Canon, Beulah, New Mexico (altitude about 8000 feet), by Mrs. Wilmatte P. Cockerell, for whom it is named. The species is of interest as being the only representative of the pal- matae occurring in the Southwest. It is related to V. cognata Greene much as V. palmata is related to F. papilionacea of the Eastern States. From the compestrine F. pedatifida and F. Bernardi it may be dis tinguished by the color of the flowers, the much smaller, narrowly oblong leaves, and the remarkably uniform cut of the latter. In this connection it is worthy of note that Mrs. Cockerell has collected F. pedatifida further up the same canon. Primula Ellisiae Plant about 1 dm. high, from a stout vertical caudex; leaves minutely scabrous on both surfaces, oblong-spatulale in outline, tapering to scar ious-margined petioles, the upper half of the blade irregularly and sharply serrulate with salient teeth, the apex obtuse or acutish; scape barely surpassing the leaves, bearing a dense umbel of rather large flowers; calyx-teeth lanceolate, exceeding the tube, the whole calyx densely farinose, but tending to become glabrate with age; corolla tube twice the length of the calyx; limb of the corolla f cm. in diameter, lavender-purple with a yellow eye, the lobes truncate and retuse. Type No. 404,914 in the United States National Herbarium, collected by Miss C. Ellis in the Sandia Mountains of New Mexico. This beautiful species is allied to P. Rusbyi Greene, the type of which was collected by Dr. Rusby in the Mogollon Mountains of New Mexico in 1881. It differs in having much larger flowers and a conspicuously farinose calyx; the leaves are also shorter-petioled, more spatulate in outline, and more distinctly serrulate on the margins. In the size of its flowers it even approaches P. Parryi Gray, of the Rocky Mountain region, while the white, farinose calyx resembles that of P. farinosa. ' Pollard and Cockerel! New Plants from New Mexico. 179 Primula angustifolia Helenae. Leaves narrowly linear and, remotely denticulate; corolla lobes longer and narrower than in the type; flowers white with a yellow eye. Type No. 404,913 in the United States National Herbarium, collected on the summit of the Las Vegas Mountains of New Mexico (altitude about 11,000 feet), June 29, 1901, by Miss Helen Blake, for whom it is named. The typical P. angustifolia is common in the same region. Achillea laxiflora. Perennial, glabrous; leaves coarsely pinnatifid, the upper ones 5-6 cm. long; divisions irregularly lobed or cleft, the ultimate segments acute; inflorescence loosely corymbose, the peduncles long and inclined to droop; heads turbinate, about 8 mm. high; bracts closely imbricated, scarious, with a greenish keel; rays orbicular, usually large (5 mm. broad in the dried specimen), pure white; achene linear, the apex prominently scarious-margined, the sides scarcely so. Type No. 404,884 in the United States National Herbarium, collected by Miss C. Ellis in the Sandia Mountains of New Mexico, at an altitude of 8000 feet. The plant attracts attention by its loose and comparatively few-flowered corymb and by the large heads with conspicuous orbicular rays. The segmentation of the foliage is also of an entirely different type from that observable in other western yarrows; and the apically margined achenes afford a conspicuous character. VOL. XV, PP. 181-182 AUGUST 6, 1902 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON DESCRIPTIONS OF NEW SPECIES OF SYNAPTOMYS AND PHENACOMYS FROM MACKENZIE, CANADA. BY EDWARD A. PREBLE. In studying a collection of mammals from the region of Athabasca and Great Slave lakes, I find the following forms of Synaptomys and Phenacomys which appear to be undescribed. Neither of these genera has been previously recorded from this region. Synaptomys (Mictomys) bullatus sp. nov. Type from Trout Rock, near Fort Rae, Great Slave Lake, Macken zie, Canada. Skin and skull No. 110,632, $ ad., U. S. National Museum, Biological Survey Collection. Collected August 17, 1901, by Edward A. Preble. Original No., 4511. General characters. Similar to Synaptomys (Mictomys) mnuitus, but differing in cranial characters; closely resembling 8. datti cranially, but differing in color; hind foot small. Color. Fur at base dark plumbeous, tipped with black and reddish- brown, the black usually predominating on the back; sides lighter than back; belly ashy-plumbeous, rather abruptly separated from the color of sides; feet dusky brown; tail bicolor, dusky above, ashy beneath. Skutt. Compared with skulls of Synaptomys (Mictomys} innuitus from Labrador, skulls of bullatus differ as follows: skulls less flat, thezygomatic 35-BioL. Soc. WASH. VOL. XV, 1902. (181) 182 Preble New Species of Syrictptomys and Phenacomys. arch descending farther below level of top of skull; bullae more inflated. Measurements. Of type: total length, 130; tail vertebrae, 24; hind foot, 18. The skull of the type measures: occipito-nasal length (tip of nasals to foramen magnum), 26; zygomatic breadth, 16; interorbital con striction, 3; length of upper molar series (crowns), 7. General remarks. Cranially Synaptomys bullata resembles closely S. dalli of Alaska, but in color it shows much less reddish-brown than either wrangeli or dalli. The hind foot of bullata is smaller than that of wrangeli, dalli, or innuitus. Phenacomys mackenzii sp. nov. Type from Fort Smith, Slave River (near the Athabasca-Mackenzie boundary line), Canada* Skin and skull No. 110,625, $ ad., U. S. Na tional Museum, Biological Survey Collection. Collected June 29, 1901, by Edward A. Preble and Alfred E. Preble. Original No. 4271 . General characters. Similar to Phenacomys celatus and P. c. crassus, but differing in cranial characters. Color. Fur everywhere plumbeous at base, on back tipped with yel lowish-brown, black, and gray in varying proportions; beneath grayish- white, passing gradually on sides into color of back; tail bicolor, nar rowly brownish above, grayish-white beneath. Cranial characters. Compared with skulls of P. celatus, those of mackenzii have the bullaa more roundly inflated, palate shorter, rostrum more slender, postorbital process of squamosal more slender, and in terorbital constriction narrower. Interorbital beads prominent, even in youth, converging early in life and nearly uniting in old age. Measurements The type measures: total length, 142; tail vertebras, 34; hind foot, 17. Ten adult specimens of both sexes from type locality average: total length, 140.7; tail vertebrae, 32.7; hind foot, 17. General remarks. In color the type series resembles quite closely a portion of the type series of Phenacomys c. crassus, kindly lent me for comparison by Mr. Outram Bangs. I have not examined specimens of Phenacomys constablei, described by Allen from Telegraph Creek, British Columbia,* which seems to belong to this group, but its measurements show that it has a larger hind foot than P. mackenzii. *Bull. Am. Mus. Nat. Hist., Vol. XII, p. 4, 1899. VOL. XV, PP. 183-186 AUGUST 6, 1902 PROCEEDINGS / OF THE BIOLOGICAL SOCIETY OF WASHINGTON THE GENUS HEDYSARUM IN THE ROCKY MOUNTAINS. / BY AVEN NELSON. Most of the material that has been collected in the genus lledysarum in the Rocky Mountain Region has found its way into the herbaria either as H. boreale or II. Mackenzii. More recently some of it is appearing under other names. Of the numerous collections made by the writer some have been distributed under the foregoing names, but always with the feeling that a careful study of ample material would show that material so distributed usually represented an aggregate. For the sake of light on this point careful study of the literature and the specimens at hand has been made with the following conclusions respecting all the species heretofore attributed to this region. Hedysarum Americanum (Michx.) Britt. lledysarum alpinum Americanum Michx. Fl. 2:74; Pursh, Fl. 2:484. //. boreale Nutt. in D. C. Prodr. 2:343, in part; Hook. Fl. Bor. Am. 1:155, in part; and later authors in part. Not //. boreale Nutt. Gen. 2:110. It does not seem to me that there is sufficient evidence to show that the plants of Michaux "In borealibus Canadae et in cataractis montium Al- le^hanis" and those of Nuttall "In arid and denudated soils around Fort Mandan on the banks of the Missouri" are the same species. In 36-BIOL. SOC. WASH. VOL. XV, 1902. (183) 184 Nelson Genus Hedysarum in Rocky Mountains. fact the descriptions do not tally very closely and Nuttall merely suggests the possibility. The two habitats are so wholly different as to suggest in themselves the distinctness of the plants. Hedysarum boreale Nutt. Hedysarum boreale Nutt. Gen. 2:110; and of many later authors in part. See citations and the note on the preceding. When the proper locality shall have been visited, Nuttall 's subvillous species with its obovate leaflets and round articulation in the loment will probably be found again. Hedysarum cinerascens Rydb. Hedysarum cinerascens Rydb. Mem. N. Y. Bot. Gard. 1:257. 77. can- escens Nutt. T. & G. Fl. N. Am. 1:358. This species is not of frequent occurrence, but is occasionally found in typical form from central Wyoming, northward and westward. The cinereous pubescence throughout, the small stipules and the short corolla- wings are distinctive. Hedysarum lancifolium Rydb. Hedysarum lancifolium Rydb. Mem. N. Y. Bot. Gard. 1 :256. No specimen of this is at hand, but judging by the description it is an excellent species. Known only from the original locality, "Head water of Jocko River, Mont., 1883, Canby, 93." Hedysarum marginatum Greene. Hedysarum marginatum Greene, Pitt. 138. 1900. Of this species the description is not just now at hand but judging by a specimen of the type number (Pagosa Springs, Colo., Baker, 1899) it is strongly marked. Presumably that is still the only collection of it. Hedysarum Mackenzii Rich. Hedysarum Mackenzii Rich. Frankl. Jour. 745. . It is usual to call the commonest form of Hedysarum in the Rocky Mountains by this name in spite of the fact that it does not at all closely Nelson Genus Hedysarum in Rocky Mountains. 185 accord with the original description. It seems highly probable that no true H. Mackenzii occurs within the United States, at least no specimens have come under my observation with leaves that are "canescently hairy on both sides," or "utrinque eanescenti-pilosis, " and with the joints of the legume "rugose and pilose." The original specimens from "Barren grounds, north of Lat. 04, as far as the shores of the Arc tic Sea" must have been very different. The Rocky Mountain plant may be named as follows: Hedysarum pabulare n. sp. Stems several from the crown of a woody root, decumbent at base and moderately slender, 4-7 dm. long, subcinereous and obscurely striate: leaves petioled. 5-10 cm. long; leaflets 9-15, from narrowly to broadly oblong or rarely on some of the lower leaves obovate and emarginate, 1-2 cm. long, subcinereous with a minute pubescence below, green, glab- rate and minutely puncticulose above; stipules rather small, more or less united, with linear acuminations as long as the body; racemes long, on axillary peduncles surpassing the leaves, its flowers tardily or not at all reflexed; calyx-tube short (about 2 mm.), its4eeth linear-acuminate and 3-4 mm. long; corolla lilac or light purple; its standard 12-13 mm. long, broadly obovate, emarginate; the keel obovate-cuneate, equaling the standard and about one-fourth longer than the narrowly oblong wings; joints of the loment, 3-5, suborbicular, 5-7 mm. broad, usually closely united or confluent, more rarely with a narrow canescent con nective, the transverse nerves (scarcely reticulate) ending in a thickened margin, light green and appearing glabrous, but sparsely puberulent under a lens. This species is reputed an excellent forage plant. As previously stated it has often figured as H. Mackenzii. The following collections seem to me* to be representative of it. Wyoming: 752, Wind River, 1894; 1087, Snake River, 1894; 201, Bates Creek, 1901, by L. N. Goodding. Colorado: Durango, 1898, by C. S. Crandall; 83, Mancos, 1898. by Baker, Earle and Tracy. Utah: 5592, Soldier Summit, 1894, by M. K. Jones. Hedysarum philoscia n. sp. Bright green and nearly glabrous throughout; stems slender, 4-7 dm. high, few to several from a freely branched semi-fleshy root; leaves crowded, short petioled, with lanceolate membranous semiconnate stip ules: leaflets 15-23, mostly oblong (15-30 mm. long and 7-12 mm. broad), obtuse with a minute mucro, some minute straggling hairs below, especial ly on the midrib: peduncles axillary, slender, overtopping the leaves; race me slender, puberulent, acuminate; flowers white, sharply reflexed in bud, rising somewhat in anthesis; bracts filiform; calyx-tube campanulate, with triangular-lanceolate teeth shorter than the tube; corolla narrow, 186 Nelson Genus Hedysarum in Rocky Mountains. the standard shorter than the narrow keel; the wings linear, with a nar row basal lobe equaling the claw; legume glabrate, mostly three jointed; the joints oval to orbicular, 6-8 mm. long, lightly reticulate from a some what larger marginal nerve. Only two collections of this species are at hand, both of them from rich moist copses on stream banks in the Laramie Mountains of Albany Co., Wyoming. Nos. 2034, Crow Creek, 1896; 3367, Willow Creek, 1897. Hedysarum sulphurescens Rydb. Hedysarum mlpJinrescens Rydb. Bull. Torrey Club 24:253: //. flaves- cms Coult. & Fish. Bot. Gaz. 18:300; not Regel & Schm. This fine species is of frequent occurence in northwestern Wyoming, arid through Montana. Hedysarum uintahense n. sp. Green and apparently glabrous throughout, but under a lens sparsely and minutely pubescent: stems stout ish, noticeably striate, mostly erect, the short lower nodes sheathed by the large brown connate stipules; leaves nearly sessile, large, 10-15 cm. long; leaflets 11-23, variable in shape (oval, ovate or even nearly lanceolate), mostly obtuse, 1-3 cm. long; stipules large, semiconnate, brownish, membranous; racemes axil lary, in the uppermost approximated nodes appearing clustered, corym bose, or even umbellate, surpassing the leaves; the purple or lavender flow ers strongly reflexed from the earliest anthesis; calyx-tube nearly glabrous, campanulate, with oblique margin and short unequal teeth (tube 4 mm. long and teeth usually much less than half as long); corolla large, 15 mm. or more long; the keel surpassing thesubequal standard and wings; the wings broadly linear, with a slender claw and a free lobe as long as the claw: loment stipitate, its 2-5 large joints often with a canescent connective; the j oints obscurely puberulent, lightly reticulated, narrowly margined, oval to obovate or oblong, 10-15 mm. long. I cite as type my No. 7198 from the moist draws in the Uinta foot-hills, Evanston, Wyo. This species has been variously treated heretofore but most specimens are probably labelled H. boreale. The following I think belong here. Wyoming: My Nos. 877 and 3839; B. C. Buff urn's from South Fork, Crazy Woman Creek, 1892; 977, Merrill and Wilcox, Teton Pass, 1901. Washington: 1850, L. H. Henderson, Olympic Mountains. Colorado: 464, Baker, Earle and Tracy, near La Plata. The last two are given with some reserve. VOL. XV, PP. 187-190 AUGUST 6, 1902 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON DESCRIPTIONS OF EIGHT NEW BIRDS FROM ISLANDS OFF THE WEST COAST OF SUMATRA.* BY CHARLES W. RICHMOND. A collection of several hundred bird skins made by Dr. W. L. Abbott during the winter of 1901-2 on various islands off the west coast of Sumatra has been lately received at the United States National Museum. This material was gathered chiefly on Simalur Island; but birds from the small islands of Pulo Babi and Pulo Lasia, southeast of Simalur, and from the Ban- jak Islands are also well represented. The collection proves to be of unusual interest, and contains several apparently unnamed species, eight of which are de scribed below. Macropygia simalurensis sp. nov. Type. Adult male, No. 179,067, U. S. National Museum, Simalur Island (west coast of Sumatra), December 9, 1901, Dr. W. L. Abbott. Head (except ear-coverts), underparts (except middle of breast and abdomen), under wing-coverts and axillaries, reddish chestnut; feathers of chest with dusky blackish bases which are more or less visi ble; middle of breast and abdomen cinnamon; ear-coverts, sides of neck, and nape cinnamon, with dull black bars; feathers of sides of neck with ^"Published here by permission of the Secretary of the Smithsonian Institution. 37-BIOL. S6C. WASH. VOL. XV, 1902. (187) 188 Richmond Ely Jit New Birds from Sumatra. green and amethyst reflections; mantle and scapulars sepia, the feathers of the latter with reddish chestnut tips; wings and wing-coverts of the same color, the feathers (mainly of the coverts) edged with reddish chestnut, the lesser coverts being entirely of this color, and the middle coverts nearly so; back, rump, and upper tail-coverts burnt umber, the feathers with rather paler edges; three central pairs of rectrices Prout's brown, outer feathers cinnamon rufous, each with a dusky black (gray on under surface) oblique patch near the tip, mainly on the inner web. Length (in flesh), 324 mm.; wing, 157; tail, 167; tarsus, 18; culmen, 14. The female differs in being duller below, with more prominent black mottling on the foreneck and chest; the tertiaries, secondaries, wing-cov erts, back, rump and upper tail coverts are more heavily edged with red dish chestnut (paler than in the male) ; sides of neck, nape, and mantle dull blackish, with pale cinnamon and whitish bars (no glossy reflections), and the forehead is heavily streaked with black. "Iris bluish gray; feet dark purple brown; bill dark brown; black at tip." The nearest relative of this species appears to be M. ruficeps. Palaeornis major sp. nov. Type. Adult male, No. 179,111, U. S. National Museum, Pulo Babi (west coast of Sumatra), January 14, 1902, Dr. W. L. Abbott. Identical in color with P.fasciatiis, but much larger. Length (in the flesh), 438 mm.; wing, 192; tail, 230; tarsus, 18; cul men, 18. A male P. fasciatus from the Andaman islands measures: length, 387 mm. ; wing, 168; tail, 200. A specimen of the same sex from Tenasserim, has a total length of 324 mm.; wing, 150; tail, 140. Seven specimens of P. major from Pulo Babi and Pulo Lasia are very large and agree in dimensions with the type. Psittinus abbotti sp. nov. Type. Adult male, No. 179,106, U. S. National Museum, Simalur Island (west coast of Sumatra), December 11, 1901, Dr. W. L. Abbott. Mantle, scapulars, tertiaries, and inner middle and greater wing-cov erts, parrot green, the feathers with paler edges, those of the tertiaries and wing-coverts with yellowish green edges; back cobalt blue; rump, upper tail-coverts, middle pair of tail feathers, outer wing-coverts, and upper surf ace of primaries and secondaries, grass green, each feather with a paler green border; innermost lesser wing-coverts maroon purple; edge of wing and primary coverts greenish blue; head, including nape and hind neck, campanula blue, duller on the sides, becoming green (be tween emerald and apple green) on the forehead, lores, anterior part of crown and round the eyes, the green feathers being tipped with campa nula blue; a narrow band of bluish black feathers separates the blue of Richmond Eight New Birds from Sumatra. 189 the head from the green mantle. Underparts, including under tail-cov erts, apple green; under wing-coverts and axillaries carmine; under sur face of wing and under primary coverts sooty black, a few of the outer primaries with narrow yellowish edges; tail (except middle pair of feathers), pale lemon yellow, with a greenish wash on outer webs and at tips. "Iris pale yellow; cere dull green; feet greenish; upper mandible red, lower pale fleshy brown." Length (in flesh), 216 mm.; wing, 141; tail, 59; tarsus, 1C; culmen, 22. The black band on the hind neck is much broader on another male. The female is duller in color, and has a green head. Thriponax parvus sp. nov. Type. Adult male, No. 179,160, U. S. National Museum, Simalur Island (west coast of Sumatra), December 3, 1901, Dr. W. L. Abbott. Similar to T. javensis, but of very much smaller dimensions. Length, 362 mm.; wing, 1G9; tail, 124; tarsus, 28; culmen, 39 (bill from gape, 42). A male of T.javensfa, from the Malay Peninsula, measures: length (in flesh), 457 mm.; wing, 227; tail, 160; tarsus, 36; culmen, 58. Twelve specimens, from Simalur Island, all agreeing with the meas urements given above. Hypothymis abbotti sp. nov. Type. Adult male, No. 179,426, U. S. National Museum, Pulo Babi (west coast of Sumatra), January 11, 1902, 'Dr. W. L. Abbott. Wholly blue, without the black occipital spot and band across fore- neck of H. azurea. The color is bright light cyanine blue (of Ridgway's ' Nomenclature of Colors'), with a tinge of deep campanula blue on breast, abdomen, and sides. Wings and tail black, with a bluish shade above; under wing-coverts and axillaries, dusky gray, with bluish tips; wing feathers, from below, with dusky gray edges. Length (in flesh), 181 mm.; wing, 76; tail, 77; tarsus, 19; culmen, 13 (bill from gape, 20). This species seems to have no near relative; H. puella from Celebes being entirely different. It is strictly congeneric with H. azurea, and has the ruff of feathers on the fore-neck common to species of this genus. Known from Pulo Babi and Pulo Lasia, on the west coast of Sumatra. Hypothymis consobrina sp. nov. Type. Adult male, No. 179,433, U. S. National Museum, Simalur Island (west coast of Sumatra), December 24, 1901, Dr. W. L. Abbott. This species is similar to //. tytleri, from the Andaman islands, but has 190 Richmond Eight New Birds from Sumatra. a shorter wing and tail, and a narrower black band across the fore-neck; the abdcmen appears to be of a rather deeper blue. Length (in the flesh), 165 mm.; wing, 68; tail, 68; tarsus, 16; culmen, 12, (bill, from gape, 18). Malacopteron no tat urn sp. nov. Type. Adult male, No. 179,346, U. S. National Museum, Pulo Bang Karu, Banjak Islands (west coast of Sumatra), January 17, 1902, Dr. W. L. Abbott. Top of head, including nape, sooty black; lores, a narrow line round the eye, cheeks, entire underparts, under wing-coverts, and axillaries, white, the feathers of cheeks, throat and chest obscurely streaked with dusky; ear-coverts, smoke gray; sides of neck and mantle brownish drab-gray (each feather white at base), shading into wood brown on back, rump and tertiaries; wing-coverts and edges of primaries and secondaries rather grayer in color; upper tail-coverts deep cinnamon- rufous, the feathers white at their bases; tail Prout's brown, the inner webs of the feathers edged with pale rufous; wing feathers dusky, with white edges on lower surface of inner webs. Length, 178 mm.; wing, 77; tail, 69; tarsus, 23.5; culmen, 17 (bill, from gape, 21.5). This species is related to M. affine (Blyth.), but has a sooty blackcap instead of a brownish black one, and there are other minor differences of color. It differs from M. cantori (Moore) in having a distinct cap, in lacking a decided malar stripe, etc. Stachyris banjakensis sp. nov. Type. Adult male, No. 179,344, U. S. National Museum, PuloTuanku, Banjak Islands (west coast of Sumatra), January 24, 1902, Dr. W. L. Abbott. Similar to S. maculata, from Borneo, but larger, and color of nape and back grayer. Length, 200 mm.; wing, 87; tail, 71; tarsus, 27; culmen, 22 (bill, from gape, 27). "Iris brownish yellow; feet, leaden; bill, black, leaden be neath; naked skin on sides of neck pale blue." VOL. XV, PP. 191-193 OCTOBER 10, 1902 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON NOTES ON LITTLE-KNOWN NEW MEXICAN MAMMALS AND SPECIES APPARENTLY NOT RECORDED FROM THE TERRITORY. BY C. M. BARBER. The present paper is intended to record certain species of mammals not previously known to occur within New Mexico and to give such notes on their habits as may be of interest. The notes are taken from a field diary made between the years 1897 and 1902. Six species and two genera new to the terri tory are given. Ovis mexicanus has not been previously re corded from the United States. The known range of several species has been greatly extended. Ovis mexicanus Merriam. The skins and skulls of a young ram, and old ewe, and a very young lamb, recently came into the author's possession. They were killed in March, 1901, on the Gaudalupe Mountains, near the boundary line between New Mexico and Texas, by Mr. W. E. Schreffler. He informs me that a small band live on this range of mountains, and that they pass up and down the range from New Mexico to Texas. Comparison of these specimens with several specimens of the Mexican big-horn from near the type locality leaves no doubt that the Mexican sheep is found in certain favorable localities in Texas and Southern New Mexico. 38 BIOL. Soc. WASH. VOL. XV, 1902. (191) 192 JJarber Little-known New Mexican Mammals. Castor canadensis f rondator M earns. In September of 1898, I located a large colony of beavers about six miles below Ruidoso Post Office, on the Ruidoso Creek, Lincoln County, New Mexico. After a period of probation spent in learning their habits and how to trap them, I succeeded in securing a series of eleven speci mens, representing all ages. This species lives here in the creek bank, and little attempt was made to deepen the stream by damming it. I found old cuttings up the stream to an altitude of about 8,000 feet. At the time I visited tha colony they were living among the Mexican ranches at about 6,000 feet elevation. At that season (September to November) they were feeding on corn alone, although a few scarred willows were to be seen along the bank. The Mexicans had planted their corn 'right down to the water's edge to utilize all the available ground in the narrow valley. The beavers were cutting and dragging the corn to the stream, then floating it to their dens. In places there was a wagon load of stalks in the water. Near the dens they had cleaned the ground for a hundred feet on either side and made great trails in dragging stalks to the stream. The Mexicans could not trap them, and as they never appeared in the day time and seldom in the twilight, very few were killed. Felis hernandesii (Gray). Mr. Nat. Straw, hunter and trapper, informed me that he trapped a jaguar near Graf ton, on Taylor Creek, Socorro County, New Mexico, in May, 1900. He gave its length as 8 feet and 3 inches (2439 mm.) I saw the skin made up into a rug. I have heard of several others being seen or killed. It is probable that they find their way into the Mogollon Mountains by ascending the Gila River. Conepatus sp. inc. On August 12, 1901, I trapped an immature individual of this genus in the Domingo Baca canyon on the west side of the Sandia Mountains, 18 miles east of Albuquerque, New Mexico. The trap was . baited with carrion and set at the base of a large pine (Pinus ponderosa). The alti tude was about 7,000 feet. I trapped for several weeks in the same locality but could secure no further specimens. The one secured is not over half grown and does not show specific characters. I can find no record of specimens from New Mexico nor from any point as far north. The species does not appear to be abundant. It seems very strange to find this young individual several hundred miles north of the known range of the genus, and at a high altitude at that. The specimen is now in the Museum of Comparative Zoology, Cambridge, Massachusetts. JSarber Little-knoicn New Mexican Mammals. 193 Sperm ophilus interpres (Merriam). On January 5th, 1898, I secured three specimens apparently referable to this species at the base of the Organ Mountains, near the San Augus tine Pass. That they were mating was evident by their actions and the enlarged sexual organs in both sexes. They were on the east side of the range, among the rocks in a sunny canyon in the foothills. They do not hibernate at all or perhaps only during occasional storms. Efforts to trace their range further north in the San Andreas range were unsuccess ful, nor have I found the species in any of the other ranges in southern New Mexico. There is a small gap between the Organ Mountains and the Mt. Franklin group (the type locality). These specimens extend th known range about fifty miles northward. The following measure ments were taken in the flesh: Male. Length, 223; tail, 65; hind foot, 38 mm. Male. Length, 240; tail, 86; hind foot, 37 mm. Female. Length, 205; tail, 72; hind foot, 36 mm. Putorius frenatus neomexicanus Barber and Cockerell. A Mexican brought me a nearly typical specimen of this weasle at Albuquerque, New Mexico, on December 6, 1900. It was a fine male and measured as follows: Total length, 444; tail vertebrae, 176; hind foot, 46 mm. This extends the known range of the species about 200 miles north in the Rio Grande valley and is, so far as I know, the sixth speci men to be taken. It is now in the Biological Survey Collection at Washington, D. C. Putorius arizonensis Mearns. Professor G. E. Coghill shot three weasles on the Pecos Forest Reserve above Willis, New Mexico, on July 23d, 1898. The late F. J. Birtwell secured possession of them and I was permitted to examine them. They answer Dr. Mearns' description of the Arizona weasel and are apparently that species. The only measurements available are those of a female as follows: Total length, 332; tail vertebrae, 126; hind foot, 38 mm. These seem to be the first specimens of this species taken in New Mexico. Two of these specimen are now in the Museum of Comparative Zoology at Cambridge, and the third is in the A. & M. College collection at Mesilla Park, New Mexico. VOL. XV, PP. 195-199 OCTOBER 10, 1902 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON GENERAL NOTES. The occurrence of certain tropical plants in Mississippi. The flora of the coastal region of Mississippi is made more interesting by the naturalization of some tropical shrubs, introduced for ornamental purposes. These are of three species: Lagerstrcemia indica Linn., the Crepe Myrtle; Zizyphus vulgaris Lam., the Jujube Tree, and Vitex agnus-castus Linn., the Chaste-Tree. The first of these species is still very common in cultivation; in the wild state it occurs as a shrub of rather bushy habit, from four to ten feet in height. The pink flowers are borne very profusely all summer, and through September. Zizyplms vulgaris is cultivated to some extent as a lawn shrub, but is also spontaneous. Owing to the slight edibility of the fruit, and the large size of the seed, the plants are infrequently found outside of the limits of the yard where it is grown. The usual habit is bushy, and the aver age height about twelve feet, but in some cases the plant is a tree of eight inches in diameter, and twenty feet in height. The drupes are somewhat eaten by birds Mimus polyglottos and Melanerpes erythroce- phalus. The fruit ripens in August. Vitex agnus-castus is now little cultivated, it is fairly common on the beach, growing in the sand, often below extreme high-water mark. It is usually a shrub less than ten feet high, but in one case, at Bay St. Louis, reaches a height of about twenty feet. The flowers appear in May. Andrew Allison, Bay St. Louis, Mississippi. 39 BIOL. Soc. WASH. VOL. XV, 1902. (195) 196 Getieral Notes. Note on the generic name of the horseshoe crab. Mr. Pocock in his article ''The Taxonomy of Recent Species of Limulus" (Ann. Mag. Nat Hist. 7th ser., IX, p. 257, 1902), considers that the validity of the generic name Xiphosura rests solely on the Gronovian foundation. On the contrary, before 1785, the date of Mailer's Limulus, Xipho8ura was adopted by at least three binomial writers, viz: by Brun- nich (Zoologiae Fimdamenta, p. 208, 1772), Scopoli (Introd. Hist. Nat., p. 405, 1777) and Meuschen (Mus. Gronov., p. 83, lllS).Mary J. Rathbun. A further note on the name of the Argentine Viscacha. As has been fully shown,* there is no doubt of the pertinence of the generic name Viscaccia Schinz, 1825, to "la Vizcache," of Azara, the Argentine Viscacha. But it turns out that Oken (Lehrb. d. Naturg., Theil III, Abth. 2, p. 835, 1826) used the same' term in 1816, in nearly the same sense. Oken included in his group or subgenus Viscaccia only two species: (1) Lepus chilensis and (2) Mus laniger. The first, notwith standing the name chilensis, is based, as far as the description is con cerned, wholly on "la Vizcache" of Azara, while in his diagnosis of the group Viscaccia he says "Zehen vorn 4, hinten 3," which would exclude his second species, the Mus laniger of Molina, and hence the Chinchilla of Peru. As Bennett in 1829, made the Chinchilla the type of his genus Chinchilla, the Argentine Vischacha also becomes by restriction the type and only species of Oken's Viscaccia, the authority for which name is thus Oken (1816) instead of Schinz (1825). It also unfortunately happens that Oken's name cliilensis has one year's priority over maximus of Desmarestf (Dipus maximus Desm. ex Blainville M. S.). Hence, apparently, the Argentine Viscacha must be called Viscaccia chilensis (Oken). From his account of the animal, he appears to have believed, as did Azara, that it was found in Peru, and also in Chili, as shown by his reference to the use made of its fur in those countries. He gives its distribution correctly, however, as follows: "In Paraguay ist es nur westlich des Flusses Uruguay, vom 30 B. gegen Siiden, siidiich von Buenos Ayres sehr gemein." Oken was almost erratic and irregular in nomenclatorial matters, viewed from the standpoint of present day usuages, as was Zimmermann in his "Specimen Zoologiae Geographicae" 1777. His use of the generic names Lepus and Mus for the species he placed under Viscaccia will not sur prise systematists who are familiar with the character of Oken's "Lehr- buch, ' ' although the name Lepus chilensis is apparently Oken's own name. It appears to have heretofore escaped citation. J. A. Allen. *Cf. Palmer, Science, N. S., VI, p. 21, July 2, 1897; Thomas, Proc. Biol. Soc. Wash., XIV, p. 25, April 2, 1901; Allen, ibid., p. 181, Dec. 2, 1901. fNouv. Diet. d'Hist. Nat., XIII, p. 117, 1817. General Notes. 197 On the generic names Notophorus, Alces, Dama, and Cephalotes, with remarks on the "one-letter rule" in Nomenclature. Since, in a matter so essentially important as nomenclature, I look upon any obstinacy in upholding one's own opinions in doubtful cases as both childish and criminal, I gladly adopt the view advocated by Dr. Allen* that Notophorus Fisch. should be considered as a pure synonym of Tayassu Fisch., and therefore Dr. Merriam's Olidosus should stand. I do this mainly on the ground that Fischer's placing of Tayassu as equiva lent to Notophorus, may be considered as synonymizing them ab initio, without discussing the reasons given by Dr. Allen against my accept ance of Sus tayassu Linn. Gmel. as a type species. Dr. Allen's renaming of the Elk (Moose) on the ground that Alces is antedated and invalidated by Alee is of course in direct conflict with the principles advocated in my own remarks on CallorJiinus versus Cal- lirhinus and Stenorhinchus v. Stenorhynchu8.\ But far from thinking that the Alces case is one to cause hesitation in accepting what has been called the "one-letter rule," I look upon the resulting retention of so well and long known a name as Alces for the Elks as an instance in its favor. As Dr. Allen shows, this name has been used for the Elks for a long period by naturalists of all nations, and the fact that Alee would also be valid, for the fossil Irish Deer, is but a small drawback compared to the advantage of retaining Alces. When we look at the retention of other well known names affected by this rule, such as Macroglossus, He- UophoUus, Callorhinus, &c., I think the balance of advantage will be oh its side. Nor is it evident where a line is to be drawn between the ex treme one-letter principle, and the confusion of such differently spelt words as Prionodon and Priodontes, or even Odocoileus and Cododon.\ There appear to me to be only two alternatives in any given case either (1) the two names are the same and should be spelt the same (old fashioned emendation combined with non-compatibility of words of like origin, or (2) the names being differently spelt, are to be treated as dif ferent, however nearly like (modern non-emendation, and its logical con sequence the one letter rule). The position adopted by some writers of stoutly opposing emendation and at the same time objecting to the one- letter rule seems to me supported by neither logic nor classicality. But with specific names the case is different. There, far from being barred, emendation, when demanded by classic correctness, is univer sally practiced, and no doubt rightly so. Its consequence therefore, the one letter rule, does not follow, and, while admitting both Picus and Pica as valid generic names, we should not dream of admitting both picatus and picata as valid specific ones in the same genus, for the reason that either would be emended to agree with the gender of the generic name, *Bull. Amer. Mus. Nat. Hist., XVI, pp. 162 and 168. July 1, 1902. fSupra p. 154. JThe earlier of these has actually been said to invalidate the use of the later. 198 General Notes. and would thus clash into the other. Similarly grcecus and greens could not be admitted side by side, nor borneoensis and borneemis, canadiensis and canadensis. Besides the emendation reason for this distinction between the treat ment of genera and species, it must be remembered that the chances are greatly against two closely similar generic names coming very near each other in the system, or being frequently used by the same person. On the other hand, to have two species of the same genus, and even from the same country, with closely similar names would produce the maxi mum of inconvenience, with no corresponding advantage in the direc tion of stability, \fnif ormity or logicalness. Dr. Allen quotes me as agreeing with him in the use of Dama for the American Deer, but I happened to be correcting the proofs of the de scription of "Dama rothseh&di" when his paper came in, and I used the name without very close enquiry. On further consideration, however, I think I was wrong, and should now be disposed to adopt on this subject both Mr. Miller's conclusions* and his reasons for them, calling the Coiba Deer Odocoileus rothschildi. In connection with disputed points in nomenclature I may take this opportunity of pointing out that Nyctimene, Bechsteinf, with the type " Vespertilio cephalotes" Pallas, antedates Cephalotes Geoffrey (1810), and that the type of the latter should also be F. cepJialotes, following the rule recently published in Science^. The name Cephalotes will become a synonym of Nyctimene, and Dobsonia Palmer, will stand for the bat called Cephalotes by Dobson and Matschie. Oldfield Thomas. P. S. Sept. 1, 1902. It has been objected that in the above note the undeniable difference between a mere adjectival declension such as picat-us, -a -um, and a more essential modification, such as from Abrothrix to Habrothrix, has been too much ignored, and that my arguments depend on there being no real difference between them. But this is not so, for quite apart from any such argument, I hold that since on the one hand such double forms as grcecus and grecus, canadensis and canadiensis should not be both admitted side by side, and on the other that essentially similar forms of generic names such as Pridontes and Prionodon should be so admitted, a line has to be drawn somewhere between the two extremes, and that this line may most naturally, certainly and conveniently be placed between genus and species. No other suggestions, so far as I know, have been made for the draw ing of the line required, authors (like Dr. Allen in the case of Alee and Alces] who cavil at the one-letter rule not saying at what point they would cease to consider two similar words synonymous. Nor do I know what is the opinion of the extreme one-letter men about such specific names as those above quoted. 0. T. *Supra, p. 39. fSyst. Uebers. Yierf. Thiere. II, p. 615. 1800. {Science, N. S. XVI, p. 114. July 18, 1902. General Notes. 199 On the distribution of Hyla evittata Miller. In my recent paper on the Batrachians and Reptiles of the District of Columbia I gave the known range of Hi/la evittata as extending along the Potomac River from Four Mile Run to Little Hunting Creek, a dis tance of about fifteen miles. I have since learned from Mr. A. H. Howell that he has several times heard its unmistakable note in the neighborhood of Analostan Island; and Mr. G. S. Miller, Jr., has taken a specimen (No. 29,021, U. S. National Museum) at Quantico, some fifteen miles below Mount Vernon. In early August of the present summer I secured one specimen and saw another in a small fresh water pool along the York River, about half a mile above Yorktown, Va., Mr. Howell 's record probably represents the northern limit of the range of this species as I have repeatedly searched in favorable localities in the neighborhood of Plummer's Island and Great Falls without seeing or hearing it. I might also add that a careful search in a Nymphcea swamp in Accomack County, Va., failed to reveal it. W. P. Hay. VOL. XV, PP. 201-203 OCTOBER 10, 1902 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON TWO NEW VIOLETS FROM THE EASTERN UNITED STATES.* BY CHARLES LOUIS POLLARD. Viola tenuipes. Stem simple, erect, 8-15 cm. high, from a short horizontal or ascend ing knotted rootstock emitting numerous long, thick fibrous roots; leaves, 3 to 5, usually 4, borne at the summit of the stem; blades glabrate, ovate-lanceolate or hastate-lanceolate, with a truncate base and acute apex, the margins repand-denticulate; stipules ovate, membranaceous; petiole 1-1.5 cm. long, appreciably shorter than the blade; flowers on filiform peduncles exceeding the leaves; sepals irregular, from linear- ovate to linear; corolla yellow, 1.5 cm. broad, the petals narrow, beard less, and quite free from markings; capsules smooth, ovoid. In dry soil, northern Georgia and Alabama to Florida. Type from, Chattahoochee, Florida, March, 1897 in the Chapman herbarium, now the property of the Biltmore herbarium. The type sheet is a remark ably full one, bearing twelve individuals of uniform size and characters except that the rootstock of one of these produces two stems in place of a solitary one. A photograph of the type sheet is deposited in the U. S. National Herbarium. The species is also well represented by collections made at Auburn, Alabama, April 11, 1899, by Mrs. F. S. Earle; at Cull- man, Alabama, May 1, 1901, by representatives of the Biltmore Herbarium; at Tuscaloosa, Alabama, by Professor Lester F. Ward; and at Rome, Georgia, by Dr. Chapman. *Published here by permission of the Secretary of the Smithsonian In stitution. 40 BIOL. Soc. WASH. VOL. XV, 1902. (201) 202 Pollard Two New Violets. The publication of, this species is the result of a critical study of the violets belonging to the hastata group in the South, wherein my own field observations have been supplemented by information kindly given me by other collectors. For many years Elliott's V. tripartita was re garded as a variety of V. hastata, although the most superficial examina tion of the rootstock in the two species ought to have dispelled such an opinion. Yet even after Dr. Small had demonstrated this character it was noticeable to southern botanists at least that certain forms remained which could not well be disposed of in connection with either of these species. More recently Mr. Roland M. Harper* attempted to solve the difficulty by the reinstatement of Gingins' V. hastata glaberrima^ , which he treated as an entire leaved variety of tripartita, remarking in this connection: "It differs from typical tripartita in having leaves all un divided and glabrous, but is in other respects very similar. It seems to extend farther south than the type, and is the plant which was taken for V. hastata in Florida." But Mr. Harper has here fallen into the very natural error of placing in a single category all the plants with un divided leaves, regardless of range, and his statement in the concluding portion of above quoted paragraph is also a trifle too broad. While much of the material referred by Dr. Chapman to V. hastata glaberrima has nothing in common with hastata, the latter species does, notwithstanding, occur in Florida, as is abundantly proven by excellent specimens in the Biltmore Herbarium, collected in Liberty County. A mere glance at the whitish, succulent, tuberous rootstock is sufficient to establish the identity of this species. Let us now examine the range of the dubious forms under considera tion. The type of V. tripartita came from Athens, Georgia, and the species has been collected there by many different persons since Elliott's day, so that we have no difficulty in establishing a positive diagnosis of its characters. North of this point it ranges through the mountains of North and South Carolina and eastern Tennessee, being common near Biltmore, N. C., where I have carefully studied it. Throughout this range, and nearly always associated with the typical form, as shown by notes on collector's labels, is a plant of almost identical appearance, ex cept that the leaves are undivided and have a tendency to become glabrous with age. This is the true glaberrima, which may well stand as a variety of tripartita, where it has been placed by Mr. Harper. If now we examine the collections from south of Athens, we find an entire-leaved yellow-flowered violet represented in abundance, but no tripartita. It is this plant which was familiar to Dr. Chapman in Florida* and which Mr. Harper has confused with the more northerly form. V. tripartita and its variety are relatively tall, coarse plants, even at flower ing time, having flowers with the corolla distinctly veined, and with rather broad sepals. This violet is very slender, of remarkably uniform size, with smaller flowers, quite immaculate petals, and leaves strongly sug- *Bull. Torr. Club. 27: 337. 1900. J-Gingins in D. C. Prodr. 1: 300. 1824. Pollard Two New Violets. 203 gesting those of Jiastata, while the leaves of tripartita glaberrima resemble rather those of V. scabriuscula in shape. This species, which I have described above under the name tenuipes, is of southern range ex clusively. From the true kastata, which also occurs in Florida, it may be distinguished by the rootstock, which is of the same type as that of tripartita. I wish to express my appreciation of the courtesy of Mr. C. D. Beadle, Curator of the Biltmore Herbarium, for the loan of many sheets of speci mens, and to Mr. Frank Boynton, of the same institution, for assistance in field work. Viola Mulfordae. Acaulescent, tufted, from a short, thick and nearly vertical caudex: scapes and foliage finely puberulent; leaf -blades oblong or ovate-oblong outline, in the earliest 1.5-2 cm., the latter 3-4 cm. long; the margins coarsely crenate, frequently incised at base with one or more lobes, the apex very obtuse; petioles about twice the length of the blades; scapes 12-15 cm. high, surpassing the leaves; flowers deep violet-purple, nearly 2 cm. broad; sepals linear-lanceolate, finely ciliate; petals obovate, very obtuse, copiously bearded at the base with glistening white hairs; scapes of the cleistogenes evidently erect. Type, No. 404,998 in the United States National Herbarium, collected by Miss F. A. Mulford at Hempstead Plains, Long Island, N. Y., May 13, 1902. The species belongs to the coastal plain region, its affinities being with V. Brittoniana , with which it is found growing. Miss Mul ford was the first to detect the obvious differences in both flowers and foliage, and after a full season's observation of both plants in the field concluded that they should not be referred to the same species. In recognition of her courtesy in contributing material and the result of her investigations, I take pleasure in naming the plant as above. It will be remembered that the leaves of V. Brittoniana are distinctly ovate in outline and pinnately lobed, while the flowers are of another shade, and do not exhibit the white pubescence. VOL. XV, PP. 205-206 OCTOBER 10, 1902 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON AN ADDITION TO THE CORAL FAUNA OF THE AQUIA EOCENE FORMATION OF MARYLAND. BY T. WAYLAND VAUGHAN. " Since my report on the Eocene corals of Virginia and Mary land, published by the Maryland Geological Survey*, was writ ten, Mr. Chas. Schuchert and Mr. R. S. Bassler have collected some interesting specimens at upper Marlboro, Maryland, mak ing one important addition to the fauna. The additional species for this locality, collected by these gentlemen are Paracyathus marylandicus Vaughan (collected by Mr. Schuchert), originally described from Piscataway, Maryland, and Haimesiastrcea con- ferta Vaughan (collected by Mr. Bassler), the types coming from Gregg's Landing, Alabama. The first species has no especial geologic significance, although it is interesting to be able to note it from another locality. The second species is represented by a single small and imperfectly developed specimen, but pos sesses much importance, as it had hitherto been found only in Alabama, where it ranges in the Eocene from the Midway to the Wood's Bluff, its best development being in the Gregg's or Bell's Landing horizon. The occurence of this species is ad ditional evidence for correlating the Aquia formation with the Gregg's Land ing horizon, but what is much more important, it is *Eocene, 1901, pp. 222-232, pi. LXI, figs. 3-15. 41 BIOL. Soc. WASH. VOL. XV, 1802. (205) 206 Vaughan Addition to Maryland Coral Fauna. the first compound coral that has been found in the Eocene of Virginia and Maryland, and shows a northward range of an Ala baman species. The physical conditions, however, apparently were not favorable for luxuriant growth. Two factors were probably active. Apparently the Aquia deposits were laid down in water somewhat deeper than those of the same age in Ala bama. More probably the most important factor was the colder temperature in the Maryland-Virginia region. Both deep water and low temperature are detrimental to a development of colonial corals, especially those that tend to be somewhat or pronouncedly massive. The species previously known from the Aquia formation are: JFlabellumsp., Turbinolia acuticosta ta Vaughan, Trochocyathus clarkeanus Vaughan, Paracyathus marylandicus Vaughan, Balanophyllia desmophyllum Milne Edwards and Haime, and Eupsammia elaborata (Conrad). Only the last two species were known to occur also in Alabama. The Flabellum may be drop ped from the list, as it could not be specifically determined. Now of the six determined species (including the addition of Haimesiastraia confer to), we have three common to the Aquia formation of Virginia-Maryland and the Gregg's Landing horizon of Alabama. VOL. XV, PP. 207-209 OCTOBER 10, 1902 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A REDESCRIPTION OF THE CORAL PL A TYTR CHUS 8PECIOSUS* BY T. WAYLAND VAUGHAN. I was obliged, because of the imperfection of the original description and the inadequacy of the figures, to place Platy- trochus speciosus Gabb and Horn in the category of < 'Doubtful Species" in my memoir on the Eocene and Lower Oligocene Coral Faunas of the United States, f Through the kindness of Professor L. C. Glenn, of Vanderbilt University, who has re cently sent me the types, I am now able to present a new de scription of the species and to give figures, from drawings by Dr. J. C. McConnell. Trochocyathus speciosus (Gabb and Horn). Figs. 1, la, 15, 2, 2a. 1860. Platytrochus speciosus, Gabb and Horn, Journ. Acad. Nat. Sci. Phila., 2nd ser., Vol. IV, p. 399, pi. LXIX. Figs. 15, 16, 17 1900. Platytrochus speciosus, Vaughan, U. S. Geological Sur vey, Mon. XXXIX, p. 196. ^Published here by permission of the Director of the U. S. Geological Survey. fU. S. Geological Survey, Mon. XXXIX, 1900. 42 BIOL. Soc. WASH. VOL. XV, 1902. (207) 208 Vaughan Redes cription of a Coral. Not Platytrochus speciosus, C. W. Johnson, Geological Survey, New Jersey, Ann. Rep. for 1897, p. 265, 1898; Platytrochus speciosus, C. W. John son, Proc. Acad. Nat. Sci. Phila., 1898, p. 462, = Trochocyathus woolmani, Vaughan, Proc, Acad. Nat. Sci. Phila., 1900, pp. 436, 437, figs. 1,2,3. Corallum inversely conical, slightly curved, living attached by a rather small base, transverse outline elliptical. DIMENSIONS: Greater transverse Lesser transverse Height of diameter of calice diameter of calice corallum Specimen 1. 13.5mm. 11.5mm. 12.25mm. Specimen 2. 11 10 11 The transverse measurements are made to the outer edge of the costae. Wall thin around upper edge, thicker below. Forty-eight costae, which are rather prominent, regularly alternately larger and smaller, gradually decreasing in size and prominence as the base is approached. Those corresponding to the fourth cycle of septa are continued very near or actually to the base. Their edges are acute, may be transversely undulated, serrately or crenately dentate, the serra tions not very tall; small granulations on the sides. No vestige of epitheca discernible. Septa rather thin, thicker at the wall, in four complete cycles, grouped into six distinct systems. Members of the first and second cycles free; those of the fourth joining by their inner margins to the sides of the third. The upper margins of the larger septa may project as much as 1.25 mm. above the edge of the wall. Septal margins entire or very faintly crenate. Lateral ornamentation, of granulations arranged along- definite lines, with the line of divergence interior to the wall. Pali in two distinct crowns. Those before the septa of the first and second cycles are situated far down in the otilice, but still distinctly above the upper surface of the columella and are narrow. Those before the third cycle extend high up into the calice, and are quite wide, usually about twice as wide as those first described. All of the pali are thin and transversely undulated. The calice is deep, I would say that its depth is about half the height of the corallum. The upper surface of the columella is suddenly sunken. Columella well developed, fascicular, composed of numerous twisted, coalescing irregular laths. Its upper termination is not papillose, being formed by the upper ends of the irregular laths. Vaughan RedescripHon of a Coral. 209 la. Cotypes of PLATYTROCHUS SPECIOSUS. Fig. 1. General view of specimen, height, 11.25mm.; Fig. la, costse of the same specimen considerably enlarged; Fig. Ib. two systems of septa, showing pali, of the same specimen, slightly diagrammatic; Fig. 2. general view of another specimen, height, 11 mm.; Fig. 2a, calice of the same, greater diameter, 11 mm. [Figures reproduced through the courtesy of the Director of the U. S. Geological Survey.] Locality* "M. & C. R. R. or Walker's Bank, at Tennessee State Line, Hardeman County, Tennessee." Geological Horizon. "Midway Eocene." Cotypes. No. 87. Department of Geology, Vanderbilt University. Two specimens. Remarks. This species presents an individuality so striking that com parisons with other species can scarcely be made. I pointed out in de scribing TrocJiocyathus icoolmani two important differences between the two species, basing my comparison on Gabb and Horn's brief original de scription, viz: that T. speciosus is three times as large as T. woolmani and that its calice is much deeper. A third difference is that the former possesses one more cycle of septa than the latter, even when the speci mens of the two species are of the same size. T. speciosus bears considerable external resemblance to occasional large specimens of ParacyatJms alternatus Vaughan. The pali and columella are quite different, and even in form there is discernible difference. The ratio of the diameters of the calice to the height of corallum is greater in T. speciosus, i. e. the diameters of the calice in T. speciosus are relatively greater than in P. alternatus. *Taken from the label accompanying the specimens. VOL. XV, p. 211 OCTOBER 10, 1902 PROCEEDINGS OF THB BIOLOGICAL SOCIETY OF WASHINGTON A NEW SUBSPECIES OF THE CUBAN CLIFF SWALLOW. / BY E. W. NELSON. The discovery of a subspecies of Petrochelidon fulva occupy ing a limited area on the Mexican tableland, remote from the seacoast, between the ranges of P. melanogaster and P. luni- frons, and apparently not mtergradmg with either, is one of those curious cases of distribution so difficult to understand. A similar and even more striking case is the presence in the mountains of Sinaloa of a marten (Progne sinaloce) closely re lated to P. dominicensis but with its habitat surrounded by the territory occupied by P. subis and P. chalybea. Petrochelidon fulva pallida new subspecies. Coahuila Cliff Swallow. Type^o. 183,703 $ ad., U. S.Nat. Mus., Biological Survey Collection. Saltillo, Coahuila, Mexico, April 17, 1902, E. W. Nelson and E. A. Gold man. Distribution. Arid border of table-land in northeastern Mexico in por tions of Coahuila and Nuevo Leon. Subspecific Characters. Similar to P. fulva from which it differs in the slightly larger size and paler colors. The rufous frontlet is only a little paler but the nuchal collar and rump are decidedly paler and more rusty rufous. The wash of reddish on sides and underparts of head and neck is paler and is usually absent along sides of breast and body. The dark centres of under tail coverts average paler. Dimensions of type. Wing, 108; tail, 50; culmen, 7; tarsus, 12. Remarks. The amount of differentiation of this form from the Cuban bird is very slight considering their isolation from one another and the great differences of physical surroundings in the homes of the two forms. 43 BIOL. Soc. WASH. VOL. XV, 1902. (211) VOL. XV, pp. 213-234 NOVEMBER 20, 1902 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON OBSERVATIONS ON THE CONDITIONS OF LIGHT PRODUCTION IN LUMINOUS BACTERIA. BY RANDOLPH E. B. McKENNEY. Introduction. Although the phenomenon of the production of light by organic or organized matter was known in the days of Aristotle, comparatively little is known concerning the actual conditions of luminescence. The number of organisms or organic bodies said to possess the so-called phosphorescent* properties is a large and varied one. Both animal and plant forms, simple and complex types of life, are to be found in the category of light- producing organisms. In view of the comparative ease of culture and simplicity of form, the photogenic bacteria constitute perhaps the best material for the study of luminescence in living forms a phenomenon, the essential conditions of which are probably the same in all forms of life. The study of the conditions of luminescence was undertaken at the suggestion of Prof. Dr. W. Pfeffer and was carried out in the botanical laboratories of Leipzig and Basel. It is a pleasant duty to here give expres sion to my appreciation of the friendly counsels of Professors W. Pfeffer and A. F. W. Schimper during the course of my experiments. *The term phosphorescence as applied to the light produced by lumi nous organisms is inappropriate. It should only be applied to light emitted in the dark by bodies which have been previously illuminated. The light emitted by all known luminescent forms of life is entirely in dependent of previous illumination. 44 BIOL. Soo. WASH. VOL. XV, 1902. (213) 214 McKenney On Luminous Bacteria. Literature. It is only a quarter of a century since Pfltiger discovered the first luminous bacterium Micrococcus phosphoreus, Cohn.* Previous to this discovery, Pfliiger had expressed the belief that the (so-called) phos phorescence phenomena would prove to be intimately connected with the respiratory processes. He obtained confirmation of this belief when by experiment he showed the need of free oxygen for the production of light by the newly discovered bacterium. Since light production then seemed bound up with respiration, Pfliiger asserted that luminescence was a vital phenomenon that it was inseparably bound up with life. Those who had studied luminosity in animals before Pfliiger's discovery of luminous bacteria had arrived at the conclusion that it was the pro toplasm which was luminous that the luminous matter was "lebendiges Eiweiss." Pfliiger's experiments apparently confirmed this position. In 1880, Radziszewski attacked the question from the chemical stand point. It had previously been known that certain substances when raised to a certain temperature could be made to emit light. Radzis- zewski discovered a large number of additional non-living organic com pounds which could be made to give out light. He also found that in many cases a high temperature was not required and that with lophin no higher temperature than 10 was needed. Further he ascertained that the conditions of luminescence were a markedly alkaline reaction and a slow oxidation. Organic and inorganic bases might be employed to produce the alkaline reaction. The light produced by these substances has a spectrum very closely resembling that of the photogenic forms giving a continuous band between D and G, with the brightest part be tween E and F. In view of the resemblance of the spectra of the light produced by these chemicals and by luminous organisms, Radziszewski is of the opinion that the light of luminous organisms is due to the presence of certain of the photogenic substances which he discovered. While Rad ziszewski differs from Pfliiger in holding luminescence not to be a vital phenomenon, yet both agree that oxidation plays a very important role in luminescence. Dubois records a symbiotic relation between a marine mollusk and a species of bacterium. In this case Dubois asserts that the animal excretes a substance "Luciferin" which through action of the bacterium is caused to emit light. Beijerinck, who has studied a number of forms, finds that light ceases with the death of the bacteria and that the light intensity may be diminished or increased by varying the nature of the nutrient media. His conclusion concerning luminescence is, that it is a vital process that it is due to the liberation of radiant energy by the oxidation of peptone at the moment of its conversion into living protoplasm. The observations of B. Fischer, Forster, Lehmann and Tolhausen of the production of light at and below seem hardly to agree with the *Since 1875 there has been as many as twenty-five species of luminous bacteria recorded from widely separated parts of the world. In all probability, many of these species names will prove to be synonyms, or at least varieties. Mclfenney On Luminous Bacteria. 215 theory of the vital nature of luminescence. However, these observers hold that inasmuch as no luminous substance has ever been isolated from photobacteria, luminescence must be inseparable from life. Lud- wig, and with him Dubois, is of the opinion that the light is produced by some specific substance, similar to those discovered by Radziszewski. As the matter rests now, there is almost as much evidence for the "luminous substance" theory as for the intracellular vital theory; the latter having, however, slightly the better of the argument. In order to come a little nearer to the cause of light production, I decided to ex amine more minutely into the nutrition of the photobacteria and the effect of various external agents on the light production. Material. Most of the succeeding observations and experiments were made with Bacillus phosphor escens, B. Fischer (Photobacterium indicum, Beij.) and Microspira luminosa, (Beij.) Mig. (Ph. luminosum Beij.)- Some experi ments were also made with Bacterium phosphor excens, B. Fischer, (Ph. phosphor escens, Beij.). Cultures of these species were obtained from Krai's laboratory in Prague. The culture of Microspira luminosa thus obtained emitted a weak light. Strongly luminous cultures of this species were obtained, however, through the kindness of Prof. Dr. Beijerinck, of Delft. The morphologic characters of the above mentioned species are quite fully set forth in the papers of Beijerinck and in Migula's "System der Bakterien." It may simply be noted here that the Bacillus and Micro spira are motile and liquefy gelatine, while the Bacterium is non-motile and does not liquefy gelatine. When not otherwise indicated the results recorded will refer to Bacillus phosphor escens. General Methods of Culture. For most of the experimental work a liquid culture medium was found best, but control experiments were frequently made with solid culture media. About 500 grams of fresh fish were extracted over a water bath with two litres of water. Herring, pike and carp yielded good extracts, but that obtained from a couple of species of flounder was decidedly less favorable to both growth and light production. To the filtered fish ex tract the following ingredients were added: Peptone 1.0$ Asparagin .5$ Glycerol 2.0$ Na Cl 2.0$ Mg 01* 1.0$ The liquid thus obtained was made weakly alkaline with Na OH, and constitutes what will later be designated as normal fish bouillon. Ap propriate solid media were obtained by adding to this bouillon either \% of good agar or 6 to 8$ of best grade gelatine. As containers for the bouillon, Erlenmeyer flasks of ca. 100 c. c. capacity were employed. From 10 to 20 c. c. of the bouillon was intro duced into each flask. The broad base of the flask at once insured 216 McKenney On Luminous Bacteria. stability of the cultures and permitted access of free oxygen to all parts of the media. Relations to Acids and Bases. In his study of Bacterium phosphor escens, Beijerinck came to the con clusion that certain acids, e. g. lactic, malic, glyceric and aspartic, ac celerated light production, while others, such as formic, acetic, propionic and butyric decreased light emission. Still others, e. g. citric, mucic. oxalic and glycolic, appeared to be without effect on the bacterium. Beijerinck further observed that certain of the salts of these acids reacted toward the bacteria much as did the free acid. That free acids should in all cases prove injurious, or even that they should all be beneficial, would not be particularly surprising, This variation in the action of the acids was, however, difficult to understand. A quantity of normal fish bouillon, agar, and gelatine were made weakly acid with HC1, HNO f , HjPO ? and CH^COOH respectively. These media were then inoculated with Bacillus phosphorescens and Bacterium phosphorescens; but no growth ever appeared. Since the acid might perhaps inhibit initial, but not later growth, and not the light production, luminous cultures were obtained in normal media and the acids then added. To a stab gelatine culture of Bacterium phospJiorescens, 4 drops of decinormal HC1 were added with a pipette. The acid was dropped directly on the bacterial growth. The light was instantly extinguished. During the 6 hours following treatment no light was emitted, but after 24 hours a faint light was visible in the culture. This experiment was repeated a number of times and each time with the same result. Evi dently the acid was injurious to light production, but not for the life of the organism. In a second series of experiments, cultures in normal fish bouillon were employed. To such cultures, which were strongly luminous, 4 drops of decinormal HC1 were added. No effect on light emission was ob served. Examination showed that the amount of HC1 added had not been sufficient to give the medium an acid reaction. Decinormal HC1 was then added to another lightning bouillion culture until the light emission ceased. The culture fluid was then found to be slightly acid to litmus paper. Normal, double normal, and fairly concentrated HC1 were in turn added to a series of light-emitting bouillon cultures until the light disap peared. In each case the media at the end of the experiment were slightly acid to litmus. Naturally the more concentrated the acid the less was required to cause cessation of light production. In no case, however, did the light disappear until the media became slightly acid. The experiments were repeated in bouillon cultures containing litmus. In all cases light disappeared as soon as the medium turned faint red, i. e., was acid, and not before. One may interpret the difference between the results obtained with solid and fluid cultures by the fact, that with the solid media the acid acted at once on all of the bacteria, while in fluid culture only some of McKenney On Luminous Bacteria. 217 the bacteria were subject to the action of the acid before its neutraliza tion. Immediately the media became acid, i. e., the moment all the bacteria were subjected to the action of the acid, the light instantly dis appeared. Numerous experiments with nitric, sulphuric, orthophosphoric, formic, acetic, lactic, succinic, malic, tartaric, oxalic and citric acids, gave re sults essentially the same as those obtained with hydrochloric acid. At the moment the media turned just weakly acid, the light emission at once ceased. Naturally, in proportion as the normal acid was weak, or the acid dilute, so was the actual quantity of acid solution required to give an acid reaction to the medium and destroy light, the larger. The end result of a dark culture and slightly acid reaction of the medium was the same in all cases. A few experiments were made to learn the effect of the acid salts. The dihydric phosphates of sodium and potassium NaH^PO ? and KH;jPO T were employed for this purpose. Quite large quantities of the solutions of these salts were needed to render the culture media acid. In each case, however, as soon as the medium became slightly acid, the culture became at once dark. In cultures thus treated with acids the light never returned. In most cases, even when the culture was made weakly alkaline within five minutes of the acid treatment, light did not again appear in the culture. In cultures which had been made alkaline after acidification with the acid phosphates, light was again emitted within 12 hours of the addition of the alkali. A few experiments to learn the effect of excess of Na OH and KOH in the media were also tried. Growth only occurs in media which turns red litmus light blue. If 2 to 4 drops of decinormal KOH or NaOH be added to a good luminous bouillon culture light production ceases in stantly, and subsequent reduction of excessive alkalinity never permits any return of light. Inoculations made from such cultures do not take, showing the bacteria to have been killed and not simply rendered in active, as is the case when light is destroyed by acids. The experiments here recorded for Bacterium phosphorescens were re peated with Bacillus phosphorescens and Microspira luminosa. Like re sults were obtained with both. It may be well to briefly note here the methods employed for the in troduction of reagents into the cultures. In the preliminary experi ments the cotton plug was removed, the quantity of sterile reagent quickly introduced with a pipette and the cotton plug at once replaced. Although experience showed that there was rarely any bacterial con tamination by this method, still there was the danger. Since the cul tures were kept under observation for some days after treatment, a method of experimentation was devised which entirely precludes bacter ial contamination during the course of the experiment. Small glass tubes were taken, drawn out to form small capillary tubes and on one end of such a tube a very thin-walled bulb was blown. Care was taken to have the walls of the tube heavier than the bulb wall. A measured quantity of the desired reagent was introduced into the bulb 218 McKenney On Luminous Bacteria. and the open end of the tube sealed. The tube was then shoved through the cotton plug so that the bulb was just a little distance above the culture fluid in the bottom of the flask. The tube and bulb with the contained reagent were found light enough to be held in place by the cotton plug. The arrangement is shown in Fig. 1. The whole, containing culture media and reagent was sterilized in the usual way, the cotton plug slightly raised to permit the in sertion of the needle and the media inocu lated. When the culture is luminous and the reagent is to be applied, one presses on the end of the tube (a) and the bulb (b) is pressed against the bottom of the flask and shattered, bringing the reagent in direct contact with the bacteria. In this method, the only danger of contamination is that which is usually incidental to inoculation, and this experience shows to be extremely small. The acids used in these experiments in clude mona-, di-, and tribasic members of both the inorganic and organic series. Since these representative acids all destroyed light emission and often the life of the organism, it seems probable that all acids would react in the same way. My results, then, are op posed to the conclusion of Beijerinck. I find all acids to be injurious to light production, lactic and malic (which Beijerinck distinctly labels photogenic) fully as much as citric and acetic (which he classes as indifferent and injurious respectively). A consideration of Beijerinck's methods may explain his results. In the auxanogram method which he used, the reagent was dropped on a nutrient gelatine plate containing a rich bacterial growth. The acid diffused in radiating fashion from the point of contact, and as the dif fusion circles widened, the reagent came in contact with the bacteria. Now the effect noted could not have been that of the free acid, since the moment the acid came in contact with the gelatine it would react with the contained alkali and form a salt. Consequently the effect noted must have been that of a probably neutral salt and not that of the free acid. The beneficial effect of the salts of certain acids noted by Beijerinck, I have been able to confirm. It is clear then that the error in Beijerinck's account, as far as acids are concerned, is not one of result but of interpretation of these results, since his experiments did not show the effect of the free acid, but rather of its salts. The few experiments conducted with the hydrates of sodium and potassium show that while the photobacteria thrive in and, in fact, need a slightly alkaline nutrient medium, still the maximum point is rather sharply defined, and but a slight excess over that is even more fatal than an excess in the other direction. FIG. 1. Diagram of culture flask and bulb for introduc tion of sterile reagents. McKenney On Luminous Bacteria. 219 Relations to Temperature. Aside from some observations as to the maximum, minimum and optimum temperatures for light production, nothing has been recorded as to the relations of photobacteria to temperature. If the phenomenon of luminescence be primarily an irritabile function, we would expect to find evidence of this in the relations of the organisms to temperature. It must be borne in mind, however, that the term irritability rather ex presses our ignorance than our knowledge of the phenomena usually classed under that designation. And this is especially true for the lower forms of life. In the subjoined tables I give the earlier as well as my own records for the temperature minima, maxima and optima for both luminescence and growth in the three species I have examined.* 44 t> d e h d d M bo | m o> JbO 1 S8 1 f 5 M *) of media and these too in 24 hours exhibited not only luxuriant 224 McKenney On luminous Bacteria. growth but a strong luminescence. It is to be noted that at every trans fer every 24 hours the bell-glass was removed a few minutes. When the bell-glass was replaced, a dish of freshly prepared 5$ ether water was placed under it instead of the old dish of ether water. In this way the supply of free oxygen was maintained as well as the action of the ether; the amount of ether which could evaporate from a 5$ water solution being a limited amount and not enough to exclude the oxygen from the bell-glass and the organisms. This experiment was twice repeated and essentially the same results were obtained. In one case the B cultures when about 4 days old also emitted light, i. e., about 2 days after their daughter (C) cultures were luminous. From these experiments it is clear that ether, when not too concen trated, exerts a partial narcosis on the bacteria. While it inhibits light production, it does not inhibit growth and multiplication and hence not all of the metabolic activities. In the case of ether we find a second adaptation of the organisms to environment. Nutrition. Naturally the first culture medium used for the culture of photobac- teria was fish the substratum from which they had been first isolated. Later the organisms were grown on agar and gelatine contain ing sea salt, peptone, asparagin, etc. The culture media best suited to cultivation of photobacteria are those first used by Beijerinck. These are described in a previous section of this paper as normal fish bouillon, fish agar and fish gelatine. That nutrient conditions exert some effect on light production, we know from Beijerinck's researches. His results indicate that certain substances which are plastic are not photogenic and vice versa. Further it seems for the six species studied by Beijerinck which include those I have examined part of the nitrogen must be fur nished as peptone and in some cases all of the nitrogen may be given in this form. In some species (peptone forms) peptone wi'll alone cover all the carbon and nitrogen requirements, while in other species (peptone- carbon forms), although peptone will suffice for the nitrogen needs, an additional source is needed for the carbon. Bacillus phosphor escens and Microspira luminosa are peptone forms, while Bacterium pJwsphorescens is a peptone-carbon form. In all cases, however, peptone seems to be a necessary part of the nutrient media. ORGANIC NEEDS. In order to test the conclusion of Beijerinck that peptone or a related protein was absolutely essential, a large series protein-free media were made up. These were inoculated with Bacillus phos- phorescem and were kept under observation for from 4 days to one McKenney On Luminous Bacteria. 225 week. Aside from the various protein-free media of Cohn, Fraenkel, Gamaleia, Niigeli, Pasteur, Proskauer and Beck, and Uschinski, a large number of original synthetic media, free from protein, were devised. In almost all instances negative results were obtained, and hence it will be needless to detail all of the synthetic protein-free media employed. In the following two media, growth was at times, although not always, obtained: I. II. Protogen, \%. Protogen, 1$. NaNO f , \%. Glycerol, \%. in distilled water. Glucose, 1$. in distilled water. In both of these media growth was slight and after 3 or 4 days entirely ceased. In no case, however, was any luminescence evident. " The growth in medium I. was better than that in II. Protogen which formed the basis of these media is, however, a complex substance, the compo sition of which is hardly understood, and it may prove to be a protein compound. While media containing peptone and needed inorganic salts will per mit growth and luminescence, still the addition of certain amides causes a more luxuriant growth. These amides include asparagin, lactamid, isobutylamin, isovaleramid, and glycocoll. Asparagin is considered by Beijerinck to be specially stimulating to luminescence. In my exper ience, while it very much promoted growth, it did not cause the least increase in the intensity of the light produced. Leucin, tyrosin, and sodium asparaginate were apparently without effect. On the other hand, methylamin, hexamethylamin, hexamethyltetramin, uric acid, hippu- ric acid and alanin, all nitrogen containing compounds, were injurious since they either retarded growth and light production or entirely pre vented growth. The inorganic nitrogen compounds as a rule did not prove plastic or photogenic. Among ammonia compounds, only the valerianate accel erated growth; it did not, however, affect the light intensity. The fol lowing ammonium compounds proved either injurious or at least indif ferent: tartrate, bimalate, chloride, carbonate, nitrate, sulphate, phos phates, and aldehyde-ammonia. Nitrate of sodium proved not alone plastic but particularly photo genic. The nitrates of potassium, lithium and calcium proved neither plastic nor photogenic. In none of the synthetic media containing peptone, amides and inor ganic salts, in which distilled water was used as the solvent, was the growth nearly as good as when, in place of the distilled water, fish ex tract was used as the solvent for the peptone, amides and inorganic salts. Evidently, while peptone may be an essential organic constituent of the nutrient media, it alone -or with any of the plastic nitrogen com- 226 McKenney On Luminous Bacteria. pounds mentioned is not sufficient to produce the best growth of the organisms. In order to determine whether perhaps a second form of carbon supply was needed, a series of sugars and related compounds were added to the media as secondary sources of carbon. When, in addition to peptone, \% of either dextrose, lactose, cane sugar or dulcite, was added, the growth was a little more than that in the control without the sugar or the alcohol. Maltose (1%) at first accelerated growth, then retarded it, and later again caused acceleration. Arabinose and levulose retarded both growth and luminescence, while the presence of \% of inulin was sufficient to entirely prevent growth. The glucosides arbutin, aesculin and agaracin retarded growth or were at least indifferent in their action. A number of additional organic compounds were experimented with. Among these protogen, lecithin, glycerol, sodium lactate, sodium phos- pholactate and sodium oleinate produced increased growth, but seemed without effect on light production. The following retarded growth: ethyl alcohol, butter, palmitin, stearin, cholestrin, camphor, turpentine, xylol, and citrus, olive, and bone oils. Peptone and sea-salts dissolved in distilled water constituted a medium sufficient to enable the photobacteria to produce fully as intense a light as when fish or fish extract, peptone and sea salt, etc., were employed. The growth was, however, never as luxuriant in the purely synthetic media as in the media containing fish extract. MINERAL NEEDS. All observers have emphasized the fact that in order to insure the best growth a certain amount of sea salt must be added to the culture media. Beyond this nothing is known concerning the inorganic needs of the photobacteria. One great difficulty in the way of investigating mineral needs of luminous bacteria is the fact that peptone must form part of the nutrient medium. All preparations of peptone contain a consider able amount of ash. Griibler's purified peptone", which was the best at my disposal, contained about \% of ash, while the Witte peptone, which was rarely employed, contains rather more than \%. The ash of the Griibler peptone (that used in the following experiments) contains iron, barium, sodium and potassium. Since peptone is essential, it at first seemed very improbable that anything could be learned about the mineral needs of the bacteria. The various culture media employed for preceding experiments have all contained a varying number of inorganic salts. The question then arose, might there not be enough mineral matter in the peptone to pro vide for its inorganic needs? Distilled water containing 1 or 2$ of pep tone, however, remained free from growth even a week after inocula tion. When, however, 2$ of sea salts was added to the \% peptone water fairly good growth and a very strong luminescence were obtained. In order to learn whether the complete mixture of salts contained in sea water was necessary, or only certain of these, media were made up con- McKenney On Luminous Bacteria. 227 taining, in addition to 1$ peptone, varying quantities of each of the salts in sea water. Since NaCl constitutes the bulk of the sea salt, it was first experi mented with. To a series of flasks containing 1$ peptone in double distilled water, NaCl was added in amounts of .25, .5, 1, 2, 3, 5, 10 and 15 per cent, respectively. Each of these flasks of peptone and NaCl was inoculated with Bacillus phospJiorescens. After 18 hours, growth was present in all of the media except the one containing .25$ of NaCl. At no time within the course of the next ten days was any growth what ever to be found in this flask. The growth in the flasks containing .5, 10 and 15$ of NaCl was very slight, that in the latter two being less than in the first. The growth continued in the cultures with 10$ and 15$ of NaCl. for only a few days and then ceased entirely. Cultures containing 1$, 2$ and 3$ NaCl showed a luxuriant growth, which con tinued for nearly a week. In the cultures containing .5$ and 5$ NaCl, growth continued for 9 or 10 days. While growth occurred in all except one of these peptone NaCl media, in only three was any luminescence to be observed. The cul tures containing 1$, 2$ and 3$ NaCl all emitted a strong light. In none of the other cultures was light discernable at any time. The light in these three cultures was fully as bright as when the bacteria were grown in normal fish bouillon. The growth was not, however, quite as luxu riant. These experiments were repeated five times and yielded the same results. Evidently, then, a single one of the ingredients of sea salt (NaCl) is sufficient for the needs of luminescence. Would any one of the other salts contained in sea water or belonging to the groups of alkali or alka line earth metals do just as well as NaCl? To answer this question, to 1$ peptone in distilled water I added the following amounts of MgCl^: .1$, .25$, .5$, 1$, 2$, and 5$, respectively. In 18 hours, growth was evident in all of these except the one containing but .1$ MgCW. Only those cultures containing 1$ and 2$ MgCl.j, however, became luminous. The light in these cultures appeared from 24 to 48 hours later than in the corresponding NaCl cultures and was rather weak. Media in which KC1, CaCl^, NH C1, and BaCl- 2 - were employed in the place of NaCl remained perfectly clear for a week after inoculation. In addition to the above salts, the following were also used in the place of NaCl; KNO^, K^SO T , LiNog, RbSO T , Ca(NO T )v,, and Sr(NO )*. None of these were, however, even sufficient for growth of the organism the media remaining perfectly clear for the week during which they were observed. Two of the salts of sea water, NaCl and MgCl^, are evidently of prime importance for the growth and light production of these bacteria, and are interchangeable. Further, the optimum amount Of MgCl^ approximates the optimum amount of NaCl. The remaining salts of sea-water are insufficient for the needs of the photobacteria. Not only is this so, but the addition of potassium or calcium salts to a peptone-NaCl medium appears to retard the growth and also light production. 228 McKenney On luminous Hacteria. The question then presented itself as to whether the metal or the haloid was the important element, or whether both were required. If other salts of sodium could replace the chloride, then the metal would be the important element. If this were not the case, then the chlorine ion or the entire molecule would be required. To test this I used NaNO y in the place of NaCl. Not only did I get growth in such media, but the light obtained with 1$, 2$, and 3$ of the salt was even more in tense than when the chloride had been used. Further, the minimum amount of NaNOg, like that of NaCl, was .5$. Less than this was not sufficient for growth. The sulphate of sodium, Na^SO ? , was likewise found capable of replac ing NaCl, and while the growth was fully as good as when NaCl was used, still the cultures did not become luminous as soon, nor were they as bright as when NaCl was used. In addition to the chloride, the nitrate and the sulphate of sodium, ten other sodium salts were experimented with, namely: monobasic phos phate, dibasic phosphate, sulphite, phospho-lactate, citrate, carbonate, acid carbonate, nitrite, tartrate, and bitartrate. Of these the first five when added to a 1$ peptone solution were sufficient for growth, and, except the sulphite, were sufficient for luminescence. The remaining five salts were found not to be able to replace NaCl. This insufficiency is, however, in all probability, due to the character of the ion linked with the sodium rather than to the sodium ion itself, since eight of the thirteen salts of sodium investigated, when added in sufficient quantity to \% peptone solution gave good growth. Since MgCl^ was capable of replacing NaCl, it seemed strange that the closely related KC1 could not replace NaCl. In order to make sure that the insufficiency of KC1 was not an osmotic one, media was made in which KC1 and KNO were added to peptone in quantities isoosmotic with 1, 2 and 3$ of NaCl and NaNo T , respectively. However, in no case was any growth to be observed even a week after inoculation. Of the salts of sodium, the nitrate, chloride and sulphate are the best forms in which to furnish sodium to the bacteria. When the nitrate is used, a far brighter light is obtained than when any of the other salts are employed. The chloride is also used to better advantage than is the sulphate. The fact that the nitrate is more advantageous than the chloride of sodium indicates that the sodium need can hardly be a ques tion simply of adaptation to its primitive invironment. The Theory of Luminescence. At present, those who have studied the luminous bacteria may fairly be said to be divided into two camps, one holding that luminescence is intracellular and that it is inseparably bound up with life, while the other considers it to be extracellular, and not inseparable from life that it is capable of reproduction in the laboratory. Among those holding the intracellular view may be mentioned Pfliiger, Beijerinck and Lehmann. McKenney On Luminous Bacteria. 229 Radziszewski, Ludwig and Dubois believe in the extracellular theory. The observations at hand give almost equal support to both sides of the question, although the intracellular theory seems to have a little the better of the argument. That light is an oxidation phenomenon, is pretty largely accepted by all students of the luminous bacteria. The questions to be settled are, however, numerous, e. g., what it is that is oxidized, the conditions of oxidation, how the light is produced by oxidation, and is the oxidation internal or external? We know that even in an abundant supply of fresh oxygen the photobacteria may be non-luminous. Further, con tinued growth and light production do not necessarily go hand in hand. We have seen that photogenic bacteria may grow at a high temperature without producing light, e. g. Bacillus phospkorescens will grow at 38, but remains perfectly dark. A culture of Bacillus phosphorescens does not emit light as soon as the first growth takes place. Usually it is not luminous until from 18 to 24 hours after inoculation. During this period the culture medium is seen to become more and more clouded with a white growth, and finally a white skin of bacterial growth covers the surface of the culture liquid. Then the culture becomes luminous. This is not due to contact with the air, because when the culture is luminous, it is luminous to a depth of 2 to 3 centimeters. Again we have seen how a certain amount of ether may prevent light production and yet not growth. During the 18 hours immediately following inoculation, and before luminescence begins, the bacteria are actively swimming about the cul ture liquid. After light production begins not only are the bacteria of the surface skin motionless, but also those in the depths of the liquid. In no case have I observed light while the bacteria were motile, and conversely I have not been able to find the bacteria in motile condition while they were in a luminous condition. Indeed it would seem at least for Bacillus phosphorescens that light and motion are opposing functions, since they are not performed at the same time, but one follows the other. As long as the bacteria are in motion, the culture has but little odor (unless fish extract has been used) and is of a light yellow color. Shortly after the culture becomes luminous, the color changes. First it becomes dark yellow, then it is light brown, then more and more reddish. By this time the odor is very marked and reagents are hardly needed to demonstrate the presence of the lower fatty acids and of skatole. And reagents^confirm the olfactory evidence. It has been shown that the elements sodium and magnesium are of importance for both light and growth of luminous bacteria. Just what the connection may ..be between the oxidation which causes light emis sion and sodium or magnesium is still a question needing further experi mentation. That there is some connection is clear. A comparison of luminous bacteria with the electric ray is rather suggestive. The elec tric organ of the torpedo is known to be rich in NaCl, usually having as much as 3$. Further, it is commonly accepted that the electric organ 230 McKenney On Luminous Bacteria. is modified muscular or contractile tissue. In any case there is analogy between the bacterial contractile flagellse and contractile animal muscle. Today physicists are pointing out the close connection between light and electricity. The large. NaCl content in the electric (or modified con tractile) organ of the torpedo on the one hand and the large sodium need of the photobacteria taken with the relations between light and motion on the other hand are full of significance. With the facts at hand one may reasonably- draw a few conclusions concerning the nature of light production. The fact that no luminous substance has ever been certainly isolated rather inclines one to disbe lieve the extracellular theory. The fact that the temperature limits for life are without the limits for luminescence points to the intracellular theory. The fact that a slight amount of ether may cause a cessation of light emission and yet not stop growth points in the same direction. While these facts lead us to strongly believe that luminescence is an internal (oxidation) process, yet there are not facts enough at hand to warrant the assumption that this process is inseparable from life and incapable of exact reproduction in the laboratory. The beautiful re searches of Radziszewski show us the possibility of such a thing. Still it is yet to be proven that these same processes occur in the bacterium and are responsible for its luminesence. I see no warrant for Beijerinck's assumption that light is produced by sudden union of oxygen and pep tone at the moment of conversion into living protoplasm. To begin with, it is still to be demonstrated that peptone is capable of direct con version into protoplasm. Sny thesis is not always a recapitulation of analysis. To me it seems that luminescence is connected with metabolism, and since its appearance is closely followed by the presence in the culture liquid of the products of portein decomposition, that it is a phase of destructive metabolism. It also seems highly probable that the phe nomenon of contractility (motility) and luminescence are closely related to one another, since the one appears when the other disappears. Fur ther, it seems possible that the sodium ion may serve as a strongly reducing agent, possibly rendering oxygen atomic and so providing for a very active oxidation with consequent liberation of energy as light. In the near future I expect to be able to test the hypothesis suggested in the latter part of this paper. Summary. In conclusion I may summarize the chief results of the experiments as follows: 1. All acids are injurious to light production. A slight excess of alkali is even more injurious than a slight excess of an acid. 2. The temperature limits for light emission are within those necessary for growth. 3. Change of temperature, either sudden or gradual, is without effect on luminescence, i. e., does not stimulate. McKenney On Luminous Bacteria. 231 4. There is no luminescence at or below 0. 5. Exposures to temperatures above the growth maximum are highly injurious to the power of light production, while exposure to very low temperatures seems to serve as a stimulus to light production. 6. Bacillus phosphorescens is capable of adapting itself to high tem peratures, producing a race capable of light production at 35, which is 5 above the normal maximum for luminescence. 7. A certain degree of continued illumination is without effect, and it is possible for the bacteria to live their entire lives in the dark and yet emit a brilliant light. 8. Ether acts as a narcotic, preventing luminescence, but not growth and multiplication. 9. It is possible to develop a race of bacteria so immune to the action of small amounts of ether as to be still luminous in its presence. 10. Peptone or related protein is required for the nutrition of luminous bacteria. 11. Dextrose, and certain of the higher sugars may be utilized ad vantageously by Bacillus phosphoresccns. 12. Either sodium or magnesium is required for growth, and especially for light production. Minimum, maximum and optimum amounts of sodium are observed for growth and luminescence. 13. Potassium, ammonium, lithium, rubidium, calcium, barium and strontium cannot replace sodium (or magnesium). 232 McKenney On Luminous Bacteria. Bibliography. Bancel, C. et Husson, C. : Sur la phosphorescence de la viande de hommard; Compt. rend., 88:191. 1879. Beijerinck, M. W.: Le photobacterium luminosum, bacterie lumi- neuse de la Mer du Nord; Archives Neerlandaises, 23:104. 1889. L'auxanographie ou la M6thode de 1'hydrodiffusion dans la gelatine appliquee aux recherches micro- biologiques; Archives N6erlandaises, 23:367. 1889. Les bacteries lumineuses dans leurs rapparts avec 1'oxygene. Archives Neerlandaises, 23:416. 1889. Over lichtvoedsal en plastisch voedsal van licht bac- terien; Verslagen en Mededeeling der Koninklijke Akadamie van Wetenschappen. Afdeeling Natuur- kunde. Ill Reeks. Deel. VII :239. Amsterdam. 1890. Sur 1'aliment photogene et 1'aliment plastique des bacteries lumineuses; Archives Neerlandaises, 24:369. 1891. (A French translation of the preceding paper). Conn, F: Briefliche Mitteilungen an J. Penn, abgedruckt in Vezameling van stukken betreffende het geneeskundig staatsoetzicht in Neederland. Jaarg. 1878, p. 126. Cited by Schroter : Pilze in Krypt. Flora von Schlesien, p. 146. Dubois, R: Sur le r61e de la symbiose chez certaines animaux marins lumineux; Compt. rend., 107:502. 1888. Nouvelles recherches sur la production de la luminie're par les animaux et lesvegetaux; Compt. rend., 111:363. 1890. Lecons de Physiologie. Paris. 1898. Eijkman, G: Liclvtgevende Bakterien; Jaarverslag van het Labara- torium voor pathologische Anatomie en Bakteriologie te Weltfreden over het Jaar 1891. (Geneeskundig Tijdschrift voor Neederl. Ind., 32:169. 1692). Re viewed in Centralbl. f. Bakt., 12:656. McKenney On Luminous Bacteria. 233 Fischer, B. : Ueber einen lichtentwickelnden im Meerswasser gefundenen Spaltpilz; Zeitsch. f. Hygiene, 2:54. 1887. Ueber einen neuen lichtentwickelnden Spaltpilz; Cen- tralb. f. Bakt., 3:105 and 137. 1888. Bakterien Wachstum by 0. Centralbl. f. Bakt., 4:89. 1888. Die Bakterien des Meeres Nach den Untersuceungen der Plankton expedition. Kiel und Leipzig. 1894. Forster, J.: Ueber einige Eigenschaften leuchtender Bakterien; Centralbl. f. Bakt. 2:337. 1887. Ueber die Entwickelung von Bakterien bei niederer Temperatur; Centralbl. f. Bakt., 12:431. 1892. Fliigge, C. : Mikroorganisnlen, 3te Auflage, Bd 1 :65. 1896. Gadeau de Kerville, H.: Les -v^getaux et les animaux lumineux, 1890. Reviewed in Bot. centralbl. Beiheft 4:228. 1894. Hermes, : Der leuchtende Nordsee Bacillus. (Review in Baum- garten's Jahreshericht der Pathogenen Mikroorgan- ismen. 3:344. 1888. Katz, O.: Preliminary remarks on phosphorescent Bacteria; Pro ceed. Linn. Soc. New South Wales. Ser. 2, Vol. 2: 331 and 628. 1887. Zur Kenntnis der Leuchtbakterien; Centralbl. f. Bakt., 9;157. 1891. Kutscher, : Zur Phosphorescenz der Elbvibrionen; Centralbl. f. Bakt., 8:424, 1890. Ein Beitrag zur Kenntnis der den Choleravibrionen ahnliche Wasserbakterien; Deutsche Med. Wochen- schrift. No. 49. 1893. Lassar, O: Die Mikrokokken der Phosphorescenz; Archiv. f. die gesamte Physiologie, 21:104. 1880. Lehmann, K. B. : Studien iiber Bacterium phosphorescens; Centralbl. f. Bakt., 5:785. 1889. 234 McKenney On Isutninoud Hacteria. Ludwig, F.: MicroeoccusPflugeri; Hedwigia, 23:33. 1884. Die bisherigen Untersuchungen iiber photogene Bak- terien; Centrtlbl. f. Bakt., 2: 372 and 401. 1887. Lehrbuch der niederen Kryptogamen. p. 68. 1892. Migula, W. : System der Bakterien. 1:335, 2: 434, 865 and 1015. Pfeffe*, W.: Pflsnzenphysiologie, Ite Auflage. 2:418. 1881. Pfluger, E. : Die Phosphorescenz lebendigen Organismen und ihre Bedeutung fiir die Principien der Respiration; Archiv. f. die gesamte Physiologie, 10:275. 1875. Ueber die Phosphorescenz verwesender Organismen; Archiv f. die gesamte Physiologic, 11:222. 1875. Radzisze wski, B. : Ueber die Pho&phorescenz der orgariischen und organisierten Korper. Liebig's Annalen der Chemie, 203:305. 1880. Tollhausen, P.: Untersuchunger iiber Bacterium phosphorescens, Fischer; Inaug. Dissert. Wurzburg. 1889. VOL. XV, PP. 235-238 DECEMBER 16, 1902 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON SOME GENERIC NAMES OF TURTLES. By LEONHARD STEJNEGER. I. Merrem, in 1820, (Tentamen, p. 27) was the first author to give a generic name, viz: Terrapene, to the Emydine turtles with a movable plastron, which some earlier writers, such as Oppel (1811) and Cuvier (1817), had indicated as a section of the genus Emys. In the genus he included six valid species (his T. boscii being only a synomyne of T. ornatcr) as follows: (1) T. odorata (-f- boscii), (2) T. Pennsylvania, (3) T. amboinensis, (4) T. tricar inata, (5) T. nigricans, (6) T. clausa. Two years later Fleming (Philos. Zool. II, p. 270) appar ently without knowing Merrem's work, gave the name Cistuda to the same group of turtles, without mentioning any species whatsoever. This makes it an unconditional synonym of Ter- /apene, a conclusion quite in consonance with Say's use of Fleming's name in 1825 (Journ. Phila. Acad., IV, ii, p. 205) for the species C. clausa, C. pennsylvanica, and C. odorata. In 1824, Spix (Testud. Brasil., p. 17) instituted the genus Kinosternon, thus taking out of Merrem's Terrapene his T. tri- carinata. J. E. Gray in a paper entitled "A Synopsis of the Genera of 45 BIOL. Soc. WASH. VOL. XV, 1902. (235) 236 Stejneger Some Generic Names of Turtles. Reptiles and Amphibia" (Aun. Philos. (n. s.) X, pp. 193 seqv.) published in September, 1825, subdivided Merrem's genus and fixed T. clausa definitely as the type of Terraphene, as he spelled it. Kinosternon he also adopted as proposed by Spix, and in addition he instituted for T. odorata and T. pcnx- sylvanica, a new genus for which he used a manuscript name by Bell, viz, SternotJierus. Both of the species which constitute his new genus being congeneric with the type of Kinosternon^ SternotJierus becomes an unconditional synonym of the latter. Thomas Bell, whose MS8. name Gray had thus been appro priating, scarcely more than a month later (Zool. Journ., II, No. 7, Oct. 1825, p. 305) used the same name (though spelt slightly different, Sternothaerus) in a somewhat different sense, including in it, besides T. odorata and the conspecific T. boscii, two new species S. trifasciatus and 8. leachiamis. This action ties the name SternotJierus down to T. odorata, if the latter be considered generically distinct from Kinostemon, thus antedat ing AromocJielys. The genus which is now usually known as Sternothcerus, therefore, must have another name, and Pelusios of Wag let- (Nat. Syst. Amph., 1830, p. 137) becomes available with P. nigricans for type.* The synonymies of these genera, as here mentioned, would then stand as follows: Terrapene Merrem. 1820. Terrapene Merrem, Tent. Syst. Amph., p. 27 (type, as restricted by Gray, Sept., 1825, T. dausa). 1822. TJierapene Schinz, Citvier's Thierr. Uebers., II, p. 13 (emend.). 1822. Cistuda Fleming, Philos. Zool., II, 270 (no species men tioned). 1825. TerrapJiene Gray, Ann. Philos. (n. s.) X, Sep., p. 211 (err. typogr.). *If SternotJierus had not been disposed of by Gray in 1825, it would have stood for the genus now known as Cyclemys, since Wagler in 1880 clearly restricted it to C. trifasciata a whole year before Gray applied it exclusively to Pelusios nigricavK. Stejncijer Some Generic Nama* of Turtle*, 237 Kinosternon Spix. 1824. Kinosternon Spix, Testud. Brasil., p. 17 (type K. Ion- gicaudatum -f- J\. brevicaudatum = Testudo scorpi- oides = Terrapene tricaritiata MerreraJ. 1825. Sternotherus Gray, Ann. Philos. (n. s.) X, Sep., p. 211 (type 6*. odoratus). 1825. jSteruotlMerus Bell, Zool. Journ., II, p. 305 (type S. odoratus). 1829. Stemoteirits Gravenhorst, Delic. Mus. Vratislav., p. 17 (emend.). 1855. Aromochelys Gray, Cat. Shield Rept. Brit. Mus., I, p. 46 (A. odor at um). Pelusios Wagler. 1830. Pelusios Wagler, Nat. Syst. Amph., p. 137 (type IZmys castanea -j_ E. subniyra = Terrapene nigricans Merrera). 1831. Sternotherus Gray, Synops. Rept., p. 37 (same type) (not of 1825). Cyclemys Bell, 1830. Stemothatrus Wagler, Nat. Syst. Ampb., p. 137 (type S. trifasciatus) (not of Bell 1825). 1834. Cyclemys Bell, Proc. Zooi. Soc. London, 1834, p. 17 (type C. orbwulata), II. Gray's generic name Nicoria is plainly untenable being ante dated by the same author's Geoemyda. The type of the latter has been believed to be G. spinosa, but as I am going to show, this assumption is entirely erroneous. The type of Qeo&nyda, on the contrary, is un doubtedly G. spengleri which is so characterized in the original communication establishing the name in the following words (Proc. Zool. Soc. London, 1834, p. 99): "Specimens were exhibited of several Reptiles, which were accompanied by notes by Mr. Gray. These notes were read: 238 Stejneger Some Generic Names <>f 44 Mr. Gray regards the Testudo Spenglcri, Walb., as the type of a new genus of EmydvdcB 4 * From the beautiful figure of the animal of Em. spit'tomr given by Mr. Bell in his 'Monograph of the Testudinata' Mr. Gray is inclined to believe that this species belongs to the same genus with Em. Spengleri." As will be seen nothing could be more explicit. The genus was also so accepted by contemporary authors as shown by Bonaparte's use of the emended name Geoemys in 1837 with G. spengleri as the only species. That Gray himself afterwards (1855) shifted the name to G. spinosa has nothing to do with the case, except that it necessi tates the coining of a new name for the genus for which the latter species stands as type. As a substitute I would propose Jleosemys* with the three species JJeosemys spinosa, grandi* and depressa. The synonymy of the two genera would then stand as follows: Heosemys Stejneger. 1855. Geovmyda Gray, Cat. Shield Kept. Brit. Mug., I, p. 1(5 (type G. spinosa) (not of 1834). 1902. Jleosemys Stejneger, Proc. Biol. Soc. Washington, XV, Dec., 1902, p. 216 (same type). Geoemyda Gray. 1834. Geoemyda Gray, Proc. Zool. Soc. London, 1834, p. 100 (type Testudo spengleri). 1837. Geoetnys Bonaparte, Mag. Zool. Botan., II, No. vii, p. 60 (emend.; same type). 1855. Nicoria Gray, Cat. Shield Kept. Brit. Mus., I, p. 17 (same type). 1869. Melanochelys Gray, Proc. Zool. Soc. London, 1869, p. 187 (type M. trijuga). 1876. Chaibassia Theobald, Cat. Kept. Brit. Ind. (p. 6) (type Ch. tricar inata). *From >, East and sjuvS, turtle, formed in analogy to VOL. XV, PP. 239-240 DECEMBER 16, 1902 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A SALAMANDER NEW TO THE DISTRICT OF COLUMBIA. BY LEONHARD STEJNEGER. I wish to place on record the capture of a species of Amby stoma additional to those enumerated in Prof. W. P. Hay's 4 List of the Batrachia and Reptiles of the District of Colum bia" (Proc. Biol. Soc., Washington, XV, 1902, pp. 121-145), viz : Ambystoma maculatum (Shaw). A large specimen was caught at Twining City, D. C. on Oct. 19, 1902, by Mr. A. E. Thorn, and is now in the National Museum (No. 30,906). Ambystoma maculatum is usually known as A. punctatum (Linnaeus), being the Lacerta punctata of the twelfth edition of the Sy sterna Naturae. This name was preoccupied by Linnaeus himself in the tenth edition for a true lizard, now known as Riopa punctata. Its use is therefore precluded. The next name in time is Bechstein's Salamandra palustris, but this name was also preoccupied the year previously by Schneider for the female yellow-spotted Salamander of Europe. Shaw's Lacerta maculata is the next name in time, and is apparently not preoccupied. These names are absolutely equivalent, being all based on Catesby's "Car. 3, p. 10, t. 10, f. 10." The synonymy of this species, consequently, will stand as follows : 46 BIOL. Soc. WASH. VOL. XV, 1902. (239) 240 Stejneger A Salamander New to D. C. Ambystoma maculatum (Shaw). 1766. Lacerta punctata Linnaeus, Syst. Nat., 12 ed., I, p. 370 (type loc. " Carolina ") (not of 1758). 1800. Salamandra palustris Bechstein in Lacepede's Naturg. Amphib., II, p. 544 (substitute name) (not of Schneider 1799). 1802. Lacerta maculaa Shaw, Gen. Zool., Ill, i, p. 304 (sub stitute name). 1803. Salamandra venenosa Daudin, Hist. Nat. Kept., VIII, p. 229 (type loc. "near Philadelphia"). 1804. Lacerta subviolacea Barton, Amer. Philos. Trans., VI, i, p. 109, pi. iv, fig. 6 (type loc. "a few miles from the city of Philadelphia) ". 1850. Ambystoma carolince Gray, Cat. Batr, Grad. Brit. Mus., p. 35 (substitute name). 1854. Ambystome argus Dumeril and Bibron, Krpet. Gen., IX, p. 103 (substitute name). 1854. Salamandra margaritifera MS. Mus. Paris, fide Dumeril and Bibron, op. cit. p. 105. This salamander has costal grooves as indicated for Amby stoma opacum, by Professor Hay in his * l Key to the species of Caudata" but the two species may be distinguished easily by their color alone, Ambystoma opacitm being black with bluish gray cross-bars above, while A. maculatum is black with one series of large rounded yellow spots on each side of the back. These notes are published here with the permission of the Secretary of the Smithonian Institution. VOL. XV, PP. 241-242 DECEMBER 16, 1902 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON THREE NEW SKUNKS OF THE GENUS SPILOGALE. BY ARTHUR H. HOWBLL. A study of the skunks of this genus reveals three apparently undescribed forms, which, in advance of a more formal paper on the genus, are here briefly characterized. Spilogale tenuis sp. nov. Type from Arkins, Colorado, $ adult, No. 99,365, U. S. Nat. Mus., Biological Survey Collection. Collected Nov. 13, 1899, by R. 8. Weldon. Original number, 2198 X. General characters. Similar in markings to Spilogale leucoparia, but lateral stripes narrower; frontal patch long and narrow (in type speci men, 32x16 mm.); terminal third of tail white. Skull. Similar in general shape to that of indianola, but longer, and relatively narrower; braincase broad, and very flat; rostrum, postorbital region, and anterior portion of braincase very narrow; palate long. Measurements. Type: total length, 450; tail vertebrae, 165; hind foot, 51. Skull (of type): basilar length of Hensel, 52; zygomatic breadth, 34.7; mastoid breadth, 32.5; interorbital breadth, 14.3. *If the rule for fixing generic types recently promulgated by a num ber of working zoologists (see Science, N. S., Vol. XVI, p. 114, July 18, 1902) be generally adopted, as seems probable, the name Spilogale will have to be used for the little spotted skunks, instead of Mephitis, the name which I advocated in my previous paper in the present volume (pp. 2-6). 47 BIOL. SOC. WASH. VOL. XV, 1902. (241) 242 Ho well Three New Skunks. Remarks. This species differs strikingly in external characters from Spilogale interrupta of the plains, while from leucoparia and gracilis it may be readily distinguished by its skull characters. It is apparently a mountain animal, but is at present known from only two localities Arkins and Estes Park, Colorado. Spilogale angustifrons sp. nov. Type from Tlalpam, Valley of Mexico, $ adult, No. 50,825, U. S. Nat. Mus., Biological Survey Collection. Collected Dec. 15, 1892, by E. W. Nelson and E. A. Goldman. Original number, 4035. General characters. Size small; coloration as in S. ambigua, but usu ally without the white bands on thighs. Skull slender, and without prominent ridges. Skull. Similar to that of ambigua, but smaller and narrower; cranium highly arched; rostrum narrow: audital bullae relatively large and high; molars smaller than in ambigua, the carnassial teeth decidedly so. Measurements. Average of 3 adult males from type locality: total length, 345; tail vertebrae, 130; hind foot, 40. Skull (of type): basilar length, of Hensel, 43; zygomatic breadth, 31.3; mastoid breadth, 27.5: interorbital breadth, 13. Remarks. This form belongs with the group of narrow-skulled species inhabiting the eastern United States, in which group ambigua also be longs. Intergradation with the latter species is probable, thc-ugh no in termediate specimens have been examined. The present form occupies the southern portion of the Mexican table-land, from Guanajuato to Chiapas. Spilogale angustifrons tropicalis subsp. nov. Type from San Mateo del Mar, Oaxaca, $ adult, No. 73,523, U. S. Nat. Mus., Biological Survey Collection. Collected May 16, 1895, by E. W. Nelson and E. A. Goldman. Original number, 7958. General characters. Similar to angustifrons, but larger; skull with smaller teeth and larger mastoid capsules. Skull. Larger and more angular than that of angustifrons; mastoid capsules more inflated; audital bullae flatter; rostrum broader: upper molars and last lower molars relatively smaller. Measurements. Average of 3 adult males from type locality: total length, 378; tail vertebrae, 144; hind foot, 44.7. Skull (of type): basilar length of Hensel, 47; zygomatic breadth, 33; mastoid breadth, 30.5; inter- orbital breadth, 14.5. Remarks. This form occupies the tropical lowlands of Oaxaca, inter- grading apparently with angustifrons in the neighborhood of Tehuan- tepec. Its skull resembles that of ambigua quite closely, but has decidedly smaller teeth and larger mastoid capsules. VOL. XV, PP. 243-244 DECEMBER 16, 1902 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A NEW BAT FROM THE ISLAND OF DOMINICA.* BY GERRIT S. MILLER, JK. Among some West Indian bats collected for the United States National Museum by Mr. H. Selwyn Branch are twenty-five specimens of a Myotis from Dominica which represent a species related to M. niyricans and M. nesopolus, but distinct from either. It may be known as: Myotis dominicensis sp. nov. Type. Adult male (in alcohol), No. 113,564, United States National Museum. Collected on the Island of Dominica, West Indies, July 20, 1901, by H. Selwyn Branch. Characters. Similar to Myotis nigricans from Brazil and Paraguay, but smaller; skull with faceline more abruptly elevated above level of rostrum. Color. After immersion in alcohol for sixteen months the fur is uni form prouts brown above, and slightly tinged with burnt umber below, the hairs of the back indistinctly blackish at extreme base, those of the underparts slaty black through proximal half. Everywhere a wash of broccoli brown is visible in certain lights. Ears dull brown; membranes blackish. Skull and teeth. The skull is distinctly smaller than that of Myoti* nigricans, but in form the only tangible peculiarity appears to be the ^Published here by permission of the Secretary of the Smithsonian Institution. 48 BIOL. Soc. WASH. VOL. XV, 1902. (243) 244 Miller ^1 N"ew Bat From Dominica. more abrupt angle at which the forehead rises above the level of the rostrum. The skulls are so small that this angle is difficult to measure, but it is approximately 18 in M. nigricans and 26 in M. dominicensis. The teeth of the Dominican animal are smaller than those of Myotis ni(jricans. The crowns of the upper molars are slightly broader rela tively to their length, and the hypocone of the first and second is less developed. Small premolars both above and below less crowded than in the related species. Measurements. External measurements of type: total length, 63 (78)*; head and body, 36 (45); tail vertebrae, 27 (33); tibia, 12.4 (14); foot, 7 (7); forearm, 32.4 (33); first digit, 6.4(6); second digit, 28 (30); third digit, 48 (56); fourth digit, 41 (45); fifth digit, 37 (40); ear from meatus, 11 ( ); ear from crown, 8 ( ); width of ear, 6 ( ). Cranial measurements of type: greatest length, 12.4 (13)*; basal length, 11.2 (12); basilar length, 9.4 (10); zygomatic breadth, 7.2 (7.4); inter- orbital constriction 3 (3.6): mandible, 8.8 (9.4); maxillary toothrow (exclusive of incisors), 5(5.4); mandibular toothrow (exclusive of incisors); 5 (5.4). Specimens examined. Twenty-five, all from the Island of Dominica. Remarks. In all external features except size Myotis dominicensis ap pears to agree perfectly with M. nigric.ans. The cranial and dental characters which separate the two species are very constant in an ex cellent series of specimens of each form. From the Myotis nesopolus of Curasao the Dominican bat may be distinguished by its much more slender skull and smaller audital bullae. *Measurements in parenthesis are those of an adult male Myotis nig- rie.ans (No. 105,621) taken near Sapucay, Paraguay. VOL. XV, pp. 245-246 DECEMBER 16, 1902 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON TWO NEW TROPICAL OLD WORLD BATS.* BY GERRIT S. MILLER, JR. Among the Old World bats in the United States National Museum are representatives of two forms of Molosnidae that have not hitherto been described. Both are insular represent atives of well known continental species. Nyctinomus pusillus sp. nov. Type. Adult female (in alcohol), No. f^fff, United States National Museum. Collected on Aldabra Island, Indian Ocean, by Dr. W. L. Abbott. Characters. Externally similar to Nyctinomua pumilus. Skull and teeth noticeably smaller than in the related species. External features. In size as well as in details of external form Nyctino mus pusillus agrees so closely with N. pumilus as to need no description. Skull and teeth. The skull is similar to that of N. pumilus except that it is smaller and more lightly built, the interorbital profile is more con cave, the basisphenoid pits are better defined, and the interpterygoid space is relatively wider. Teeth much smaller than in N. pumilus and inner segment of upper molars relatively broader, so that the protocone and hypocone of the first and second are actually further apart than in the larger species. Measurements. External measurements of type : total tength, 84 ; head and body, 53 ; tail, 31 ; tibia, 11 ; foot, 7.8 (7) : forearm, 46.6 ; first digit, 6.4 : second digit, 35 ; third digit, 70 ; fourth digit, 58 : fifth digit, 34; ear from meat us, 15.4 ; ear from crown, 10 ; width of ear, 14. *Published here by permission of the Secretary of the Smithsonian Institution. 49-BiOL. Soc. WASH. VOL. XV, 190%', (5M5) 248 General Notes. in recent years, yet Bow Island is a small atoll in the South Pacific, one of a large number of coral islets composing the Paumotu chain, and withal a most unlikely place for any member of either the Dicmida or Fringillidce. Darwin (Zool. Voy. 'Beagle,' Birds, 1841, p. 105), in concluding his account of the Galapagos finches refers to the present bird, as follows: "I may here mention that a third and well characterized species of Cactornis has lately been sent by Captain Belcher, R. N. to the Zoological Society; as Capt. Belcher visited Cocos Island, which is the nearest land to the Galapagos Archipelago, being less than 400 miles distant, it is very probable that the species came thence. " As we now know some thing of the ornis of Cocos Island, Darwin's suggestion would lead us to suspect the lately described Cocornis agassizi Townsend (Bull. Mus. Comp. Zool., XXVII, 1895, p. 123, pi.), to be identical with Pinaroloxw* inornata, and a comparison between a female of the former and Sharpe's description (Catal. Birds Brit. Mus., X, p. 52) of the latter, actually proves them to be one and the same species. That there was some un certainty about the habitat of Cactornis inornata at the time Darwin wrote is evident from his surmise that Cocos Island was its true home. From the foregoing it will be seen that the ornis of Polynesia can no longer claim Pinaroloxias inornata, which name should in future be applied to the Cocos Island bird. Charles W. Richmond. The common Nyctinomus of the Greater Antilles. In the original description of Nyctinomus antiUularum, the common free-tailed bat of the Lesser Antilles (Proc. Acad. Nat. Sci., Philadelphia. 1902, p. 398, September 12, 1902), I compared the species with the re lated form occurring in the Greater Antilles, but neglected to mention the technical name of the latter. This is Nyctinomus musculus Gund- lach (Monatsber. k. Preuss. Akad. Wissensch, Berlin, 1861, p. 149), based on Cuban specimens. The animal is readily distinguishable from all of the known continental members of the Nyctinomus brasiliensis group by its smaller size, shorter ear, and rudimentary, peg-like anterior lower premolar. Gerrit 8. Miller, Jr. Lophostoma Venezuelan changed to Tonatia Venezuela?. In publishing a paper in conjuction with Capt. Wirt Robinson, on a collection of mammals made in the vicinity La Guaira, Venezuela, Proc. U. S. Nat. Mus., XXIV p. 154, Oct. 3, 1901, I overlooked Dr. Palmer's "Random notes on the nomenclature of the Chiroptera," Proc. Biol. Soc. Wash., XII p. Ill, April 30, 1898, where he shows that Lop?iostomcri8ta 19-21 Two new violets from the Eastern U. S 201-203 Pollard, C. L. and Cockerell, T. D. A.: Four new plants from New Mex ico 177-179 Porites clavaria 57 polymorpha 57 porites 56 Preble, Edward A. : Descriptions of new species of Synaptomys and Phenacomys from Mackenzie, Canada 181-182 Primula ellisiae 1W Jhelenae 179 rusbyi 178 Priodontes 197 Prionodon 197 Progne sinalose 211 Promops orthotis 250 Pseudemys rubiventris 143 Psittinus a"b"botti 188 Pterygistes azoreum 250 Ptychozoon homalocephalum 37 horsfieldii 37 kulili 37 Putorius arizpnensis 19 neomexicanus 193 pcrdus 67 striatus 2, 8, 60, 62, 67 Rana catesbeiana 131 clamata 131 palustris 13 pipiens 130 sylvatica 131 Rathbun, M. J.: Note on the generic name of the horse-shoe crab 19< Regina leberis 139 Rhogeessa minutilla 95 parvula P Rhopocichla 85 Rhopornis 35 255 Rlius floriilana littoralls 148 radicans 14T toxicodendron 147 Ribes aureum 23 beatonii ,. 2({ billiardii 26 cereum 100 chrysococcum 21 cinernscens 26 coccineum 26 ebracteatum 26 flabellaris 20 flavum 26 fontainesii 26 fragrans 26 glabratum 20 gordonianum 26 inodorum 26 intermedium 26 irveguum 100 jasmiriiflorum 26 lacustre 28 lentum 26, 100 leiobotrys 26 lindleyanum 26 longiflorum 26,100 loudoni 26 mlssouriense 26 mogollonicum 100 - moile 26 montigeuum 26 nubigenum 26 odoratum 26 Oregon i 26 oxyacauthoides 27 palmatum 26 parvuluin ; 27 praecox 26 proximum 26 revolutum 26 sanguineum.. 26 serotinuui ..... 26 setosum 27 tenuiflorum 24 tubiflorum 26 vc