t I'll ~Bi Given By f-t 4 D crc-u/n \je^CD ^-J U. S. DEPARTMENT OF AGRICULTURE ' DIVISION OF BIOLOGICAL SURVEY NORTH AMERICAN FAUNA No. 17 [Aetna! date of publication, June~6, l'.ioo] Mi 111- AMERICAN VOLES OF THE GENUS MICROTIS VERNON BAILEY C H I E h' [ IRLD NATURALIST Prepared under tin direction of Dr. C. HART MERRIAM CHIEF OF niVlSIONOF BIOLrvrrlCAL SURVEY \Y A SII IN(iTON &6 V I . i; N VI 1'. N T riMNTOi: < > F V L G E 1 '.- 0 0 U. S. DEPARTMENT OF AGRICULTURE . DIVISION OF BIOLOGICAL SURVEY NORTH AMERICAN FAUNA No. IT [Actual date of publication, June 6, 1900] Je£r REVISION OF AMERICAN VOLES OF THE GENUS MICROTIS VERNON BAILEY CHIEF FIELD NATURALIST . )'i:»'iia WASHINGTON GOVERNMENT PRINTING OFFICE 1900 LETTER OF TRANSMITTAL. U. S. Department of Agriculture, Washington, D. C, March 10, 1900. Sir: I have the honor to transmit for publication, as No. 17 of North American •Fauna, 'A Revision of the American Voles of the Genus Microtus,'' by Vernon Bailey, Chief Field Naturalist of the Biological Survey. Respectfully, Hon. James Wilson, Secretary of Agriculture. 2 .... : ; J C. Hart Merriam, Chief, Biological Survey. ■:Q CONTENTS. Page Introduction 5 Habits 6 Breeding 6 Food 6 Economic status , 7 Injury to trees aud crops 7 Protection of trees from voles 8 Destruction of voles 8 Determination of species 9 Material examined 9 Snip family Microtinw 10 Genus Microtus 10 Generic characters 10 Subgenera 10 Key to subgenera 11 List of species and subspecies, with type localities 11 Subgenus Microtus 13 Groups in the subgenus Microtus 13 Key to species and subspecies of the subgenus Microtus 11 Subgenus Ariicola 59 I'itymys 62 Lagurus 67 Chilotus 70 Pcdomys 72 Orthriomys 76 Herpetomys 77 Xeqfiber 78 3 ILLUSTRATIONS. PLATES. Page. Platk I. Microtus pennsylvanicus Frontispiece II. Skulls of representative species of the 9 subgenera (top view).. .. 80 III. Skulls of representative species of the 9 subgenera (bottom view). 82 IV. Skulls of representative species of 7 of the groups in tbo subgenus Microtus (top view) ... 84 V. Skulls of representative species of 7 of the groups in tbo subgenus Microtus (bottom view) 86 TEXT FKiURES. Fig. 1. Molar enamel pattern of Microtus peunsylvauicus 17 2. Molar enamel pattern of Microtus luoutanus 28 3. Molar enamel pattern of Microtus californicus 35 4. Molar enamel pattern of Microtus operarius 39 5. Molar enamel pattern of Microtus abbreviates 45 6. Molar enamel pattern of Microtus loiniscudi 17 7. Molar enamel pattern of Microtus mordax 19 8. Molar enamel pattern of Microtus phcuus 55 9. Molar enamel pattern of Microtus xanthognathus 57 10. Molar enamel pattern of Microtus macropux . 59 11. Molar enamel pattern of Microtus pinetorum 63 12. Molar enamel pattern of Microtus pallidas 67 13. Molar enamel pattern of Microtus oregoui 71 14. Molar enamel pattern of Microtus austerus 73 15. Molar enamel pattern of Microtus umbrosus 76 16. Molar enamel pattern of Microtus guatemalensis 78 17. Molar enamel pattern of Microtus alleni 79 4 No. 17. NORTH AMERICAN FAUNA. June, 1900. REVISION OF AMERICAN VOLES OF THE GENUS MICROTUS. By Vernon Bailey. INTRODUCTION. The following synopsis of American voles is based on a study of between 5,000 and 0,000 specimens from more than 800 localities, includ- ing types or topotypes of every recognized species with a known type locality, and also types or topotypes of most of the species placed in synonymy. Voles, or meadow mice, occur throughout the greater part of the northern hemisphere north of the Tropics. In North America both species and individuals reach their maximum abundance in the Canadian and Transition zones, and from this broad belt the number of species decreases on both sides. On the north a few species occur in the Hudsonian and Arctic zones, and individuals are abundant even in the -barren grounds, or 'tundras,'' north to the arctic coast. South of the Transition zone the decrease in species and individuals is rapid. In the Upper Austral zone they are scarce; in the Lower Austral rare and exceedingly local; while in the Tropical only a single species, of very limited distribution, is known. To the south, as individuals decrease in abundance and species become restricted to distinct areas, the degree of specific and superspeciiic differentiation becomes more and more marked. Of the nine American subgenera, one (Neojibcr) is confined to Florida, and two (Orthriomys and Rerpetomys) are restricted to two isolated mountains in southern Mexico. Another (Pitymys) is mainly Austral, and is confined to the southeastern United States and a small area in southeastern Mexico. Three others (Pedomys, Lagurus, and Chilotus) are found mainly in the Transition zone, and reach but little north of the United States. The subgenus Arvicola belongs to mountains in the Hudsonian and Canadian zones; and the polymor- phous subgenus Microtus is the only one that enters the arctic regions. Yoles adapt themselves to the mcst diverse conditions of environ- ment. Many of the species inhabit moist or wet ground and several are mainly aquatic; others inhabit areas of excessive humidity, while 5 6 NORTH AMERICAN FAUNA. [no. 17. a few live in dry and even arid regions. Some live in the perpetual shade of dense ibrests, others are exposed to the full effects of light on the open plains. Some of the most striking peculiarities of the differ- ent species result from these different conditions of environment. The development of oil and musk glands is most pronounced in the aquatic species of the subgenera Neofiber and Arvicola and least in the sub- genera Lagurus and Pedomys of the dry regions. The color is palest in species most exposed to light and dryness, as in curtatun and pallidus, and darkest and richest in species from shaded and humid areas, as in quasiater and umbrosus. The ranges of most of the species and subspecies conform to the limits of well-defined life zones, except in the subaquatic species, which follow water courses and often have the appearance of being out of their proper zones. HABITS. Certain peculiarities of habits are common to nearly all of the species. None are known to hibernate, but in the North they have snug winter homes under the snow, where they move about freely in numberless tun- nels. They burrow in the ground, and are famous for their little roads or smooth trails which run through the grass from burrow to burrow or away to their feeding grounds. Bulky nests of grass and soft plant fibers are placed in underground cavities, or on the surface of the ground under cover of snow, logs, or dense vegetation. The nest is depressed globular in form, with an open chamber in the center, which contains a soft bed, and has one or two round entrances at the sides. These nests are the sleeping places of the old and the nurseries of the young. They are kept surprisingly clean and fresh, and new ones are frefjuently made to take the place of those that are old or imperfect. Breeding. — Voles seem to have no definite breeding season. Four to eight young are usually produced at a birth, and as far north as Min- nesota I have found them in the nests at all seasons of the year. Their increase is accordingly very rapid, and is only partially counter- balanced by the host of enemies that prey upon them. They form the principal food of nearly all owls and some hawks, while weasels, minks, foxes, coyotes, cats, badgers, skunks, and many other animals, as well as certain snakes, feed extensively on them. But in spite of their enemies they seem to hold their own, and tend to increase faster as the country becomes more thickly settled and the larger mammals and birds are destroyed. Food. — Meadow mice choose a somewhat varied diet, but their food consists mainly of green vegetation, roots, and bark. Grass, especially the tender base of grass stems, forms the bulk of their food, but almost every plant with which they come in contact is eaten to some extent. Bark, both from roots and trunks of trees and shrubs, is a favorite winter food. Seeds and grain are eaten when found, but are not espe- cially sought; flesh in any form is never refused. As the animals are June, 1900.] ECONOMIC STATUS. < active all winter and food is always abundant, tbey do not ordinarily lay up stores, although Mr. E. W. Nelson found M. operarius, of Alaska, storing- roots." ECONOMIC STATUS. Injury to trees and crops. — Though small enough tobe commonly called mice (meadow inicfc, upland mice, field mice, pine mice, ground mice, bear mice, etc.), they make up in numbers what they lack in size, and over the whole breadth of the continent lay a heavy tribute on many products of the farm. Too small and too numerous to be suc- cessfnlly destroyed by traps, guns, or poison, they prove one of the most difficult enemies with which the farmer has to contend. If they would confine themselves to meadows, their mischief would be limited to the destruction of a comparatively small amount of grass; but they prefer growing grain to grass, and by running long tunnels under ground, or making little paths under cover of the vegetation, gain easy and safe access to the fields. With a stroke of their chisel-like teeth they fell the stalks of wheat and oats and eat the tender parts, together with some of the grain. It is so easy to cut down the stalks that they destroy many times as much as they need for food. The work of a few animals is insignificant, but the work of millions makes heavy inroads on growing crops. Later in the season, when the grain is cut and left standing in shocks or stacks, the field mice take possession, building their nests and establishing their homes under its cover. In shocks oi corn and wheat left for a long time the grain is often completely devoured, and that remaining all winter in stacks suffers in proportion to the number of the little animals that make their homes in it. Even stacks of hay are often found in spring with the lower parts cut to chaff and filled with the nests of meadow mice. When the snow comes these little rodents can safely leave their cover of weeds, grass, or bushes and plow their way under the snow on long exploring expeditions. The tunnels thus formed remain as open passages until the snow melts in spring, giving the animals free and safe conduct from the meadows to the uplands, into fields, orchards, gardens, and nurseries. There is no sign from above of what is happening below the surface; but later on, in spring, when the snow disappears, trees and shrubs are found stripped of their bark for a wide space near the ground. The marks of tiny teeth remain in the hard wood, and little piles of dry outer bark, mixed with character- istic pellets of excreta, show what animal has been at work. The uncovered roads may be seen leading from tree to tree, to winter nests on the surface of the ground, and back to the cover of brush or mead- ows. Shrubs and small trees are often stripped of their bark and killed, and sometimes even well-grown apple trees, 10 inches or a foot in diameter, are completely girdled. Usually, however, large trees are >Proc. Biol. Soc. Wash., VIII„140, 1893. 8 NORTH AMERICAN FAUNA. [no. 17. gnawed on only one side. In this case, although they are not killed at once, the wood, thus exposed usually decays iu a few years, the trees become hollow at the base, their productiveness is impaired, and they die prematurely. Protection of trees from voles. — Various means have been resorted to for protecting fruit trees and shrubs from these ravages, but with only partial success. Wire netting and tin cylinders placed around the bases of the trees in autumn are generally considered the surest pro- tection, but in most cases this is too expensive to be practicable. Wrapping the trunks with burlap or twisted ropes of straw, or coating them with whitewash, tar, or other unpalatable substances, are com- mon methods of protection used with varying degrees of success. But as some species of voles eat the bark from the roots below the surface of the ground, none of these resources insure perfect protection. Destruction of voles. — The importance of placing every possible check on the increase of these animals and of reducing their numbers when they become too numerous is obvious. No direct method of accom- plishing these ends has as yet been devised, but the desired result can be attained indirectly by avoiding or preventing the useless destruction of their natural enemies. Owls and some species of hawks live almost exclusively upon them, watching for them night and day in the grass, and are always ready to pounce on any that appear above the snow. Weasels run through their burrows and trails, and not only kill enough for food, but destroy great numbers for the mere pleasure of killing. In spite of these well known and often reiterated facts, boun- ties are still paid for the destruction of hawks and owls in counties where the annual loss in fruit trees and grain from the ravages of field mice if computed would amount to a startling sum. In the spring ot 1895 I examined a small apple orchard in Washtenaw County, Mich., in which several choice trees had been killed and many others injured during the preceding winter by the common vole (Microtus pennsylr ani- ens). The owner of the orchard considered $50 a low estimate of the damage done. At the same time the county of Washtenaw was taxing the farmers to pay a bounty of 25 cents each on all hawks and owls, while the several gun clubs of the county gave these birds a high count in their competitive hunting matches. Many similar instances could be cited. Who was ever known to miss an opportunity to destroy a hawk or weasel? The diminution of foxes, minks, coyotes, and such preda- tory mammals may be necessary, but if so, the protection and encour- agement of other less harmful species becomes doubly important, and in fact imperative, if we are to escape such devastating hordes of voles as have occasionally swept over certain parts of Europe, particularly in Scotland,1 Germany,2 Italy,2 Bussia,2 and Thessaly.1 1 Parliamentary Report of Plague of Field Voles iu Scotlaud, Loudon, 1893. 2 U. S. Consular Reports, L, No. 187, 539-543, 1896. June, 1900.] MATERIAL EXAMINED. V DETERMINATION OF SPECIES. It is not many years since certain prominent writers treated as mere varieties, or subspecies, animals that belong to widely different sub- genera, while others described and named with full specific rank every different condition of pelage in a single species. In some cases the original type was not preserved, or no type was designated by the describer, or still worse, the type locality was not given, so that sub- sequent writers renamed these same species or confounded them with others. The resulting confusion can now be cleared up by means of series of specimens collected within the past ten years at most of the known type localities, and in the general region of those not definitely known. The series of specimens available, and the number of localities represented, make it possible to define almost every North American species from typical specimens, and in most cases to give the various changes of pelage due to season and age. When possible, the original types have been compared with the new series of specimens from the. type localities, and in this way the names califomicus, trowbridgi, edax, occidentalism townsendi, longirostriSj and modest us have been sifted out with the following result: califoniieus stands for a widely distributed western species with trowbridgi as a synonym; edax as a well-marked species, but one in which the name has been persistently misapplied; occidentalis as a synonym of townsendi; longirostris as a synonym ot montanus; and modestus as a western form of pennsylvanicus. The type of montanus is lost, but a series of 57 specimens from the type locality agrees with Peale's description of the species. The types of modestus and edax are immature specimens made up with the skulls inside the skins. It was only by the removal of the skulls that even the group to which the species belonged could be determined.1 MATERIAL EXAMINED. The following synopsis of the genus Microtus is based mainly on a study of specimens in the collection of the Biological Survey and that of Dr. C. Hart Merriam, both of which are in the United States National Museum. For the use of much additional material, including types and topotypes, my thanks are heartily extended to Dr. F. W. True, executive curator, and Mr. Gen-it S. Miller, jr., assistant curator of mammals, United States National Museum; to Dr. J. A. Allen, curator of mammals and birds, and Mr. Frank M. Chapman, assistant curator, American Museum of Natural History; to Mr. D. G. Elliot, curator of the Department of Zoology, Field Columbian Museum; and to Mr. Outram Bangs. Most of all, I am indebted to Dr. C. Hart Merriam, 1 Through the kindness of Dr. True and Mr. Miller, skulls have also been removed from a large numher of specimens from Alaska and Arctic America, so that it has been possible for the first time to identify the species and make use of the localities in determining their rauges. 10 NORTH AMERICAN FAUNA. [no. 17. who, after doing muck work on tke genus, has placed kis manuscript, drawings, and large private collection of specimens at niy disposal, besides giving me constant criticism and advice. Among otkers wko kave contributed material or notes my thanks are especially due to Mr. E. W. Nelson, wko kas collected all the known Mexican species of Microtis and lias contributed the notes on their zonal distribution. Seventy species and subspecies are here recognized. Of these 54 actual types and series of topotypes of 13 additional forms have been examined, while of the three remaining forms, for which no type exists or is accessible and no definite type locality is known, specimens kave been examined from the type region, or as near to it as can be deter- mined. Tkree forms, Microtus californicus constrictus, M. ludoviciauus and M. scirpensis are described as new. Except for a relatively small number of alcoholics and a few skeletons, tke specimens are mostly well-prepared skins with cleaned skulls and are accompanied by col- lectors' measurements. All measurements are in millimeters, and external measurements, unless otherwise stated, are taken in tke flesh by collectors. Skull measurements are my own, made from perfect skulls unless otherwise stated. The skull drawings are by Dr. James C. McConnell. Most of the drawings of teeth have been used in previous publications of the Biological Survey. Subfamily MICROTINiE Cope.1 The subfamily Microtinw includes tlfe Voles of the genera Microtus, Evotomys, and Phenacomys; the Lemmings of the genera Lemmas, Discrostonyx, and Synaptomys; and the Muskrats of tke genus Fiber. As the genera and subgenera of the family have been recently treated in detail by Mr. Gerrit S. Miller, jr.,2 it is only necessary to give briefly the characters distinguishing the genus Microtus. Genus MICROTUS Schrank. Generic characters. — Lower incisors with roots extending far behind and on outer side of molar series; upper incisors not grooved; molars rootless, with outer and inner reentrant angles approximately equal. Palate with median ridge, distinct lateral pits, complete lateral bridges1 (not terminating in posterior shelf in any American species). Tail as long as or longer than hind foot, terete; claw of thumb pointed, not strap-shaped. SUBGENERA. Nine subgenera are here recognized among the living species of North America.4 Five of these (Chilotus, Pedomys, Eerpetomys, Orthri- omys, and Neofiber) are found only in North America. The remaining iMicrolida: Cope, Syllabus Lectures Geol. ami Paleont., p. 90^ 1891. Microtinas Rhoads, Am. Nat., XXIX, 940, Oct., 1895. - North American Fauna No. 12, Genera and Subgenera of Voles and Lemmings, 1896. :i Usually incomplete in Neofiber. ••The extinct species of Microtus are not included in the present paper. June, 1900.] SPECIES AND SUBSPECIES, WITH TYPE LOCALITIES. 1 1 four (Microtus, Pitymys, Arvicola, and Lagurus) include also Old World species. All of the nine subgenera, save Microtus, are sharply defined and easily distinguished by either cranial or external characters. The subgenus Microtus contains many more species than all of the other subgenera together, and species differing so widely that only the most general characterization can be applied to it. It is a composite group containing all forms that do not fit into the other more restricted subgenera and yet are not sufficiently differentiated to merit subgeneric rank. KEY TO SUBGENERA OF MICROTUS. m3 with 3 transverse loops and no closed triaugles. Plantar tubercles 5 or 6. m3 with 3 closed triangles, ' mammae 8.2 Plantar tubercles 6, side glands on hips in adult males (on flanks in xanthognathus) Microtus Plantar tubercles 5, side glands on flanks or else inconspicuous. Side glands conspicuous on flanks of adult males, size large Arvicola Side glands obscure or wanting, size small Chilotus m3 with 2 closed triangles, mammae 4 or 6. Skull wide and flat, tail very short, fur short and dense, mammae 4. . Pitymys Skull high and narrow, tail medium, fur coarse, mamma' G Pedomys m3 with 2 transverse loops and 2 median triangles, plantar tubercles .">. ml with 5 closed triaugles. Side glands conspicuous in both sexes, mamma* 6, size very large, tail long - - • Neofiber Side glands obscure, mamma* 8, size small, tail very short Lagurus ml with 3 closed triangles. m.3 with 3 closed triangles, mamma* 6, tail short Herpetomys m3 with 2 closed triangles, mamma; 4, tail long Orthriomys LIST OF SPECIES AND SUBSPECIES, WITH TYPE LOCALITIES. Microtus abbreviatus Miller. Hall Island, Bering Sea, Alaska. acadicus Bangs. Digby, Nova Scotia. aUcni (True). Georgiana, Brevard County, Florida. alticolus (Merriam). Little Spring, San Francisco Mountain, Arizona, 8,200 feet. anyusticeps Bailey. Crescent City, California. ari~onensis Bailey. Springerville, Arizona. arvicoloides (Rhoads). Lake Keechelus, Washington, 8,000 feet. auricularis Bailey. Washington, Mississippi. ansterus (LeConte). Racine, Wisconsin. azfecns (Allen). Aztec, Rio Arriba County, New Mexico, r>,900 feet. hairdi Merriam. Crater Lake (Glacier Peak), Oregon, 7,800 feet. hrcweri (Baird). Muskeget Island, Massachusetts. californicus (Peale). San Francisco Bay, California. canescais Bailey. Conconully, Washington. canicaudus Miller. McCoy, Oregon. chrotorrhinus (Miller). Mount Washington, head of Tuckerman Ravine, New Hampshire, r>,300 feet. 1 Except Microtus breweri, in which 2 are usually confluent, and chrotorrhinus, which has 5 closed triangles. 2 Except in the Microtus mexicanus group, in which the number is 4. 12 NORTH AMERICAN FAUNA. [no. 17. Microtus constrictus Bailey. Cape Mendocino, California. curtains (Cope). Pigeon Spring, Mount Magrnder, Nevada. drummondi (Aud. & Bach.). Rocky Mountains, vicinity of Jasper House, Alberta, Canada. dutcheri Bailey. Big Cottonwood Meadows, near Mount Whitney, Califor- nia, 10,000 feet. edax (Le Conte). California (south of San Francisco). enixns Bangs. Hamilton Inlet, Labrador. fishcri Merriam. St. Matthew Island, Bering Sea, Alaska. fontigenus Baugs. Lake Edward, Quebec. fulviventer Merriam. Cerro San Felipe, Oaxaca, Mexico. guatemalensis Merriam. Todos Santos, Huehuecheuaugo, Guatemala, 10,000 feet. haydeni (Baird). Fort Pierre, South Dakota. innuitU8 Merriam. St. Lawrence Island, Bering Sea, Alaska. kadiacensis Merriam. Kadiak Island, Alaska. labradorius Bailey. Fort Chimo, Ungava, Labrador. leucophceus (Allen). Graham Mountains, Arizona. longicaudus (Merriam). Custer, South Dakota. ludovlcianus Bailey. Iowa, Calcasieu Parish, Louisiana. macfarlani Merriam. Fort Anderson (north of Great Bear Lake), Arctic America. macropus (Merriam). Pahsimeroi Mountains, Idaho, 9,700 feet. macrurus Merriam. Lake Cushman, Olympic Mountains, Washington. mexicanus (De Saussure). Mount Orizaba,, Mexico. minor (Merriam). Bottineau, North Dakota. modestus (Baird). Sawatch Pass (Cochetopa Pass). Colorado. mogollonensis (Mearus). Baker Butte, Mogollon Mountains, Arizona. montanus (Peale). Headwaters of Sacramento River, near Mount Shasta, California. mordax (Merriam). Sawtooth (or Alturas) Lake, Idaho, 7,200 feet. nanus (Merriam). Pahsimeroi Mountains, Idaho. nemoralis Bailey. Stillwell (Boston Mountains), Indian Territory. nesophilus Bailey. Great (Jull Island, New York. nevadensis Bailey. Ash Meadows, Nye County, Nevada. nigrans Rhoads. Currituck, North Carolina. operarius (Nelson). St. Michael, Alaska. oregoni (Bachmau). Astoria, Oregon. pallidas (Merriam). Fort Buford, North Dakota. paupcrrimus (Cooper). Plains of Columbia, near Snake River, Washington. pennsylvanicus (Ord). Pennsylvania (near Philadelphia). pheeus Merriam. North slope Sierra Nevada de Colhna, Jalisco, Mexico, 10,000 feet. pinetorum (Le Conte). Pine forests of Georgia (probably near the old Le Conte plantation at Riceboro, Georgia.) l)opofensis Merriam. Popof Island, Shumagin Islands, Alaska. quas later (Cones). Jalapa, Vera Cruz, Mexico. rivularis Bailey. St. George, Utah. richardsoni (De Kay). Near foot of Rocky Mountains, vicinity of Jasper House, Alberta, Canada. scalopsoides (Aud. ifc Bach.). Long Island, New York. scirpensis Bailey. Ainargosa River, California, near California-Nevada line. serpens Merriam. Agassiz, British Columbia. sitkensis Merriam. Sitka, Alaska. Junk, 1900.] GROUPS IN THE SUBGENUS MICROTUS. 13 Microtua terrcenovcB (Bangs). Codroy, Newfoundland. ieirameru8 (Rhoads). Beacon Hill Park, Victoria, British Columbia. townsendi (Bacbman). On or near Wappatoo (Sauvie) Island, Willamette River, < )regon. unibrosus Merriam. Mount Zempoaltepec, Oaxaca, Mexico, 8,200 feet. unalascensis Merriam. Unalaska Island, Alaska. vallicola Bailey. Lone Pine, Inyo County, California. yakulatensis Merriam. Yakutat Bay, Alaska. xanthognathus (Leach). Hudson Bay. Subgenus MICROTUS Schrank. Type. — Microtus terrestris Schrank (=Mus arvalis Pallas). Microtus Schrank, Fauna Boica, I, lste Ahth., 72, 1798. Microtus Miller, N. Am. Fauna No. 12, 63, July 23, 1896 (subgenus). Geographic distribution (in North America). — From the Arctic Ocean southward to southern Mexico, and across the continent, niaiuly in Boreal, Transition, and Upper Austral zones. Subgeneric characters. — Plantar tubercles G; lateral glands on hips in adult males;1 inaninue normally 8, 4 inguinal and 4 pectoral;2 ears usually overtopping fur; ml normally with 5 closed triangles;3 m3 with 3 transverse loops and no triangles; m2with 4 closed sections, and in most eastern species an additional posterior inner loop; m3 with 3 closed triangles (except in chrotorrhinus and abbreviatus groups). GROUPS IN THE SUBGENUS MICROTUS. The subgenus Microtus is readily divided into 10 fairly well-marked groups of slightly superspecitic rank that may be conveniently desig- nated by the name of their best-known or most characteristic species. These groups are not of great importance or of equal rank, but for showing the relationship of species and for convenience in arrangement they serve a useful purpose. 1. Pennsylvanicus Group, characterized by a posterior fifth loop to middle upper molar, includes pennsylvanicus, nigrans, acadicus, modest us, fontigenus, labradorius, enixus, aztecus, drummondi, terrcenovce, nesophilus, andJjreweri. 2. Montanus Group, characterized by moderately short tail and constricted incisive foramina, includes montanus, arizonensis, nanus, canescens, canicaudus, nevadensis, rivularis, and dutcheri. 3. Townsendi Group, characterized by large size, long tail, and dark-brown color, includes townsendi and tetramerus. 4. Californicus Group, characterized by large size and wide-open incisive foramina, includes californicus, constrict us, vallicola, cdax, and scirpensis. 5. Longicaudus Group, characterized by long tail and gray color, includes lougi- caudus, mordax, macrurus, angusticeps, ulticolus, and leucophceus. 6. Mexicanus Group, characterized by short tail, brown color, and only! mammae, includes mexicanus, plurus, fulrirenter, and mogollonensis. 'In front of hips in xanthognathus and probably in chrotorrhinus. -Four in the mexicanus group, a pair of inguinal and a pair of pectoral. 3 With only four closed triangles in most of the Alaska species. 14 NORTH AMERICAN FAUNA. [no. 17. 7. Opararius Group, characterized by short tail and only 4 closed triangles in anterior lower molar, includes operarius, macfarlani, kadiacensis, itnalasccnsis, sitkensis, yakntatensis, popofensis, and innuitus. 8. Abbreviatus Group, characterized by robust form, very short tail, 5 closed tri- angles in anterior lower molar, and two closed and one open in posterior upper, includes abbreviat us and fisheri. 9. Ciirotorrhinus Group, characterized by yellow nose and five^closed triangles in posterior upper molar, includes chrotorrhinus and ravus. 10. Xanthognathus Group, characterized by yellow nose, large size, glands on Hanks, and 3 closed triangles in posterior upper molar, includes oue species, xanthognathus. In using the following key it will be necessary to have both skins and skulls in hand, and even then it will be impossible to identity some of the forms without actual comparison with their nearest allies. When- ever possible, several specimens should be examined, to avoid the danger of being led astray by abnormal molar patterns, for even the widest ranges of snbgeueric differences are sometimes covered by individual variation or abnormal tooth pattern. KEY TO SPECIES AND SUBSPECIES OF THE SUBGENUS MICROTUS. ml' with 4 closed angular sections aud a rounded posterior loop. m3 with two of the 3 triangles usually confluent. Interparietal about as wide as long, colors pale breweri 'Interparietal much wider than long, colors dark nesopliihis m3 with 3 closed triangles. ml with usually a sharp point or spur at base of posterior triangle; belly white with a median dusky line term no nr ml with normal truncate posterior triangle; belly without median dusky line. Interparietal more than half as long as wide, belly white. Skull long and narrow, braincase long; feet aud tail stout aztecus Interparietal about half as long as wide, belly usually dull colored. Skull wide, braincase short, molars small enixus Skull not wide, braincase medium, molars medium. Colors dusky or blackish. Size large, hind foot 23 nigrans Size small, hind foot 21 fontigemus Colors brownish or dark grayish. Size medium. Belly white or whitish acadicus Belly dull. Colors bright or dark brownish pennsylvanicus Colors paler, size less modestus Size small, feet and tail very slender. Skull low, incisors projecting, bulla' not large labradorius Skull high, incisors decurved, bulla' large drummondi m2 with 4 closed sections and no posterior loop (except irregularly in Californicus group. Mammae 4, inguinal, 1-1; pectoral, 1-1. Skull short and wide. Incisive foramina not constricted. Colors bright rich brown above and below fit 1 ri renter Colors dull brownish above and below, lielly but little lighter than back. Size medium mexicannx Size slightly larger phceua Belly much lighter than back ; size small mogollonensia June, 1900.] KEY TO SPECIES AND SUBSPECIES OF MICROTUS. 15 Mannme 8, inguinal, 2-2; pectoral, 2-2. ml with normally i closed triangles (sometimes 5 in sit ken sis) and rounded anterior loop. Bullae very small and narrow, molars very light. Skull narrow and slender opcrarius Skull wider and heavier kadiacensis Bulla: medium, molars moderately heavy. Incisors strongly projecting. Size large, hind foot 23 inmiitus Size small, hind foot 19 macfarlani Incisors not strongly projecting, size medium. Frontals heavily ridged in adult males. Prezygomatie notch deep, color dusky gray or ochraceous. Color dark ochraceous, helly dusky .*. sitkensis Color dusky gray, helly huffy gray yakutatensis Prezygomatie notch shallow, color ochraceous unalascensis Frontals not ridged in adults, color ochraceous popofensis ml with 5 or t> closed triangles. A pair of glands on flanks of males, nose yellowish. Size large, side glands conspicuous in adult males, m.3 with 3 closed tri- angles xanthognathus Size smaller, glauds ohscure or wanting, m3 with 5 closed triangles. Color hister chrotorrhinus Color grayish varus A pair of glands on hips of males, nose not yeljow. Incisive foramina not constricted posteriorly, m2 with or without posterior loop. Size large, colors dark, young hlackish. Nasals emarginate posteriorly erfa.r Nasals truncate posteriorly scirpensis Size smaller, colors grayish, young dusky califoruicus Colors clearer gray, hulhe smaller. Skull wide '. valHcola Skull narrow constrict it* Incisive foramina constricted posteriorly, m2 normally without posterior loop. Tail very short, size medium. Belly dusky, lips and tip of nose white dutcheri Belly, lips, and nose huffy. Rostrum and nasals slender abbrcriatus Rostrum and nasals heavy Jisheri Tail medium, size large or small. ml with 6 closed triangles aud deep-lohed trefoil. Size large, hind foot 24 nevadensis ml with 5 closed triangles and anterior trefoil. Size large, hind foot 23 , rivularis Size medium or small, hind foot 18-22, helly gray or whitish, ears large. Hind foot 20 or more. Color dark gray ahove mon la n us Color rusty gray ahove arizonensis Hind foot 20 or less. Lateral pits of palate deep, tail hicolor. Color grizzled gray nanus Color ashy gray canescens Lateral pits of palate shallow, tail mostly gray. ...... canicamlus. 16 NORTH AMERICAN FAUNA. - [no. 17. KEY TO SPECIES AND SUBSPECIES OF THE SUBGENUS MIC ROT US— Cont'd. Tail long, about J of total length. Hip glands conspicuous in males, colors dark brown. Hind foot averaging 25.4 townsendi Hind foot averaging 22 .- tetramerus Hip glands not conspicuous, colors grayisb, belly whitish. Size large, hind foot 24 macrurua Size medium, hind foot 22. Anterior arm of frontal acuminate. Skull narrow, bulhc small angusiiceps Skull normal, bullae large. Sides much grayer than back mordax Sides scarcely grayer than back longicaudus Anterior arm of frontal obliquely truncate. Size small, foot 20 alticolus Size larger, foot 22 leucophaus MICROTUS PENNSYLVANICUS (Ord). Meadow Vole. Mus pennsylranica Ord, Guthrie's Geography, 2d American edition, II, 292, 1815. (Rhoads' reprint.) Based on Wilson's description of the meadow mouse from meadows below Philadelphia and along the seashore. Ali/nomes pratensis Raiinosque, Am. Monthly Mag., II, 45, 1817. Based on Wilson's description of meadow mouse. Lemmus noveboraceitsis Rafinesque, Annals of Nature, 3, 1820. (New York and New Jersey.) Arricola riparius Ord, Journ. Acad. Nat. Sci. Philadelphia, IV, Pt. II, 305-306, 1825. (Type locality noc given.) Arricola paluslris Harlan, Fauna Americana, 136-138, 1825. (Swamp along the shores of the Delaware.) Arricola Itirsutus Emmons, Rept. Quad. Mass., 60, 1840. Arricola alborufescens Emmons, Rept. Quad. Mass., 60-61, 1840. (Williainstown, Mass.) Arricola fulra Aud. and Bach., Proc. Acad. Nat. Sci. Phila., I, 96, 1841. ("One of the Western States; wo believe Illinois.") Arricola nasula Aud. and Bach., Proc. Acad. Nat. Sci. Phila., I, 96-1)7, 1841. (Near Boston, Mass.) Arricola rvfescens DeKay, Zool. N. Y., Mammals, I, 85, 1842. (Oneida Lake, N. Y.) Arricola oncida DeKay, Zool. N. Y., Mammals, I, 88-89, 1842. (Oneida Lake, N. Y.) Arricola dekayi Aud. and Bach., Quad. N. Am., Ill, 287-288, 1854. (New York or Illinois.) Arricola riparia var. longipilis Baird, Mammals N. Am., 524, 1857. (West Northlirld, 111., and Racine, Wis.) Arricola rufidorsum Baird, Mammals N. Am., 526, 1857. ' (Holmes Hole, Marthas Vineyard, Mass.) Type locality. — Pennsylvania (meadows below Philadelphia). Geographic distribution. — Eastern United States and westward as far as Dakota aud Nebraska, shading into modestus of the western plains and Pocky Mountains. In a general way it occupies the Tran- sition zone from the Atlantic coast to the edge of the Great Plains. Habitat. — Meadows, iields, and especially grassy places near water. 1 Not having 6een the type of rufidorsum or any specimen from Marthas Vineyard, I hesitate to place this name in synonomy. June, 1900.] MICROTUS PENKSYLVANICUS. 17 General characters. — Size medium; tail at least twice as long as hind foot; fur long, overlaid with coarse hairs; ears moderate, conspicuous above fur in summer, almost concealed in winter pelage; colors dusky gray or brownish; skull long, well arched, and rather smooth; middle upper molar with four triangles and a posterior loop. Color. — Summer pelage: Upperparts dull chestnut brown, varying to bright yellowish chestnut, darkened along the back with coarse black hairs; belly dusky gray or tinged with cinnamon ; feet brownish; tail dusky above, slightly paler below. Winter pelage: Duller and grayer throughout ; tail indistinctly bicolor. Young : Blackish. Cranial characters, — Skull long, usually not angular or much ridged; incisors projecting well in front of nasals ; incisive foramina long, occu- pying two-thirds of the space between molars and incisors; bulla? mod- erately large and well rounded ; molar series long; m2 with 4 closed triangles and a posterior loop; m3 with an anterior crescent, 3 closed triangles, and a posterior loop with two inner lobes; ml with 5 closed triangles, anterior trefoil, 4 outer and 5 inner salient angles ; m3 with 3 long inner _ , _, . ... . ' Fig- 1.— Molar enamel pattern and 3 short outer salient angles. of mcrotus pennsyUanicus Measurements. — Average of 5 adults from (x5)- Washington, D. C: Total length, 171; tail vertebra?, 46; hind foot, 21.2. Skull (No. 30321, 9 ad., from Washington, D. C.) : Basal length, 27.4; nasals, 8.3; zygomatic breadth, 17.2; mastoid breadth, 12.7 ; alve- olar length of upper molar series, 7.2. Remarks. — The above description is based on a good series of speci- mens from the District of Columbia, showing the seasonal changes of pelage and agreeing perfectly with the Pennsylvania animal. From Pennsylvania south along the Atlantic coast, specimens show a notice- able increase in size and intensity of coloration, which reaches its maximum in the subspecies nigrans of North Carolina; while to the north they show a corresponding decrease in size and intensity of col- oration, which reaches its extreme in the subspecies acadicus of Nova Scotia. To the westward pennsylvanicus is fairly typical as far as southern Michigan and Iowa, but on the plains of Nebraska and South Dakota it grows paler as it grades into modestus. Northward in Min- nesota it becomes smaller until scarcely distinguishable from and perhaps grading into drummondi of northwest Canada. Three skulls in the U. S. National Museum, that seem to be typical pennsylvanicus^ are labeled as coming from Prairie Mer Eouge, La., but I am inclined to question the authenticity of the labeling. Specimens examined. — Total number, 716, from the following localities: Pennsylvania: Philadelphia, 1 ; Chester County, 1 ; Carlisle,6; Columbia, 2; Drurys Run (near Renovo), 6; Foxbury, 2; Pine Gleu, 1; Leasuresville, 2; Bear Lake (Warreu County), 2. 18392— No. 17 2 18 NORTH AMERICAN FAUNA. [no. 17. New York: Owego, 2; Nichols, 24; Lake George, 20 ; Alder Creek, 2 ; Locust Grove, 55; Peterboro, 9; Troy, 5; Geneva, 5; Brandon, 4; Catskill Moun- tains, 3; Highland Falls, 12; Mott Haven, 1; Oyster Bay, 2; Lake Grove, 2; Montauk Point, 45; Shelter Island, 6; Plum Island, 14; Roslyu, 4. Connecticut: East Hartford, 2. Massachusetts: Wilmington, 13; Middleboro, 19; Newtonville, 4; Holmes Hole, 1; Woods Holl, 1. Vermont: Burlington, 4. New Hampshire : Ossipee, 15. Maine : Addison, 1 ; Calais, 1. New Jersey: Tuckerton, 4; Mays Landing, 1; Sea Island City, 1. Maryland: Laurel, 23; Hyattsvillc, 7; liladensburg, 1 ; Mountain Lake Park, 2; Finzel, 1; Grautsville, 1. District of Columbia: Washington, 64. Virginia: Falls Church, 2; Dunn Loring, 2; Arlington, 2; Fort Mycr, 1; Bristoe, 1. "West Virginia: Travellers Repose, 2; White Sulphur Springs, 3. North Carolina: Roan Mountain, 45; Old Richmond, 3; Raleigh, 20. Ohio: Garrettsville, 10; Salem, 1. Michigan: Detroit River, 1; Manchester, 3; Ann Arbor, 2. Illinois: West Northliehl. 6. Wisconsin: Racine, 14; Busseyville, 1; Milwaukee, 4; Saxeville, 1; Fisher Lake (Iron County), 14. Iowa: Knoxville, 2. Missouri : St. Louis, 5. Louisiana : Prairie Mer Rouge, 3. Nebraska : Blair, 3 ; Valentine, 2. Minnesota: Elk River, 112; Fort Snelling, 3; Heron Lake, 1; Ortonville, 1; Tower, 6. South Dakota: Vermilion, 2; Pierre, 2; Travere, 1; Flandreau, 4; Fort Sis- seton, 18; Fort Wadswortb, 2. Ontario : Toronto, 1 ; Lome Park, 6. MICROTUS PENNSYLVAN1CUS NIGRANS Rkoads. Albemarle Meadow Vole. Microtus pennsylv aniens migrans Rhoads, Proc. Acad. Nat. Sci. Phila., 1897,307-308. Type locality. — Currituck, N. C. Geographic distribution. — (Typical form.) Coast region of northern North Carolina and southern Virginia, in the Austroripariau zone. Habitat. — Marshes and localities close to water. General characters. — Slightly larger than pennsylvanicus, with notice- ably larger hind feet and darker coloration. Color. — Summer pelage: Upperparts dull bister, much obscured by black hairs; belly smoky gray to dull cinnamon; tail black above, sooty below; feet blackish. Winter pelage (partly retained in April specimens): Darker, with dorsal area almost black. Young (to nearly half grown) : Sooty black all over. Cranial characters. — Skull averaging slightly larger than in typical Pennsylvania! s; rostrum slightly heavier, incisive foramina wider; den- tition the same. June, 1900.] MICEOTUS PENNSYLVANICUS ACADICUS. 19 Measurements. — Average of three not fully adult males from type locality: Total length, 165; tail vertebra?, 48; hind foot, 23. Skull (No. 72374, $ ad., from Eastville, Va.): Basal length, 29; nasals, 8.5; zygo- matic breadth, 17.5; mastoid breadth. 13.4: alveolar length of upper molar series, 7.3. Remarks. — Specimens of Microtias from a chain of localities along the Atlantic coast, from North Carolina to Nova Scotia, show a decrease in size and intensity of coloration from the south northward. Unfortu- nately the type of pennsylvanicus was taken from an intermediate locality, and it becomes necessary to recognize the extremes — acadieus and migrans — as slightly differentiated forms. Specimens examined. — Total number of typical specimens, 16, from the following localities: North Carolina : Currituck, 6. Virginia: Wallaceton, 7; Eastville, 1; Smiths Island, 2. MICEOTUS PENNSYLVANICUS ACADICUS Bangs. Acadian Vole. Microtus pennsylvanicus acadieus Bangs, Am. Nat., XXXI, 239-240, March, 1897. Type locality. — Digby, Nova Scotia. Geographic distribution. — Nova Scotia and Prince Edward Island. Habitat. — Fields and fresh-water marshes. General characters. — Slightly smaller than M. pennsylvanicus; color slightly paler, both in summer and winter. Color. — Summer pelage (July to October) : Upper parts yellowish bister, slightly lined with black hairs; belly washed with white or smoky gray; tail indistinctly bicolor, brownish black above, slightly paler below; feet dusky plumbeous. Winter pelage: Back buffy gray; sides paler; ears nearly concealed under bright ochraceous patch; belly washed with pure white; tail sharply bicolor, blackish above, white below; feet plumbeous. Young: Not so dark as those of pennsylvanicus. Cranial characters. — Skull usually distinguishable from that of penn- sylvanicus by projecting posterior point of palate ; posterior tip of nasals slightly emarginate or truncate, never rounded. Dentition as in pennsylvanicus. Measurements. — Type: Total length, 172; tail vertebra?, 49; hind foot, 20. Average of 5 topotypes: 170; 47; 21. Skull (No. 2145, $ — not fully adult): Basal length, 25.5; nasals, 7.7; zygomatic breadth, 14.8; mastoid breadth, 12; alveolar length of upper molar series, 6.5. Remarks. — None of the 19 topotypes before me are old, and the major- ity are not fully adult, but in a series of 40 specimens from Prince Edward Island, including fully adult specimens in both summer and winter pelages, the principal characters of M. acadieus are accentual d. The winter pelage is rather more distinctive than the summer. Specimens examined. — Total number, 67, from the following localities: Nova Scotia : Digby, 19; Halifax, 1. Prince Edward Island : 47. 20 NORTH AMERICAN FAUNA. [no. 17. MICROTUS PENNSYLVANICUS MODESTUS (Baird). Sawatcli Vole. ArvicoJa modesta Baird, Mamm. N. Am., 535-536, 1857. Arvicola insperatus Allen, Bui. Am. Mus. Nat. Hist., 347, 1894 (Custer, 8. Dak.). Type locality. — "Sawatcli Pass, Rocky Mountains'1 [same as Coche- topa Pass], Colorado. Geographic distribution. — Rocky Mountains and western Plains from New Mexico to Britisli Columbia, and from the Black Hills of South Dakota to central Idaho, and beyond, with slight variation, to the plains of the Columbia, mainly in Transition zone. Habitat. — Marshes and damp grassy places. General characters. — Size of M. pennsylvanicus, tail slightly shorter, color paler, more yellowish, never chestnut in summer pelage; skull heavier, becoming more ridged and angular with age. Color. — Summer pelage : Upperparts dull ochraceous, darkened with black-tipped hairs; belly washed with soiled whitish, smoky gray or pale cinnamon; feet plumbeous; tail indistinctly bicolor, blackish above, dull grayish below. Winter pelage : Much darkened above by long black hairs, especially early in the season, later becoming paler than in sum- mer as the under-fur grows longer; belly heavily washed with creamy white; feet paler; tail more sharply bicolor. Young: Slightly less blackish than in pennsylvanicus. Cranial characters. — Skull not positively distinguishable from that of 'pennsylvanicus, but in adults averaging heavier and more ridged. Measurements. — Average of 10 adults from Cochetopa Pass, Colorado : Total length, 170; tail vertebra', 44; hind foot, 20.0. Skull (No. 4S0r>3, 9 ad.): Basal length, 27 ; nasals, 7.0; zygomatic breadth, 10; mastoid breadth, 12.4; alveolar length of upper molar sSries, 0.7. Remarks. — Baird's type of modestus was collected at Sawatch or Cochetopa Pass in the Cochetopa Mountains. The type specimen in the United States National Museum is a half-grown young in the black pelage, and agrees perfectly in both external and cranial characters with specimens of the same age since collected at the type locality. The other specimen from Sawatch Pass (No. 593), which Professor Baird examined and believed to be distinct from modestus, but refrained from describing from a single immature specimen,1 is also in the United States National Museum, and proves to be Microtus nanus, a good series of which has since been collected at a point 3 miles east of Cochetopa Pass. Microtus mordax is the only other species known to occur in this part of Colorado. These three widely different species are readily dis- tinguishable at any age. Microtus modestus decreases in size to the northward until, in north- western Montana, it seems to merge into the little drummondi of the region farther north. Westward it becomes darker, specimens from Salt Lake Valley, Utah, being practically indistinguishable from typi- cal pennsylvanicus, while those from Cteur d'Alene, Idaho, and the 1 Mamm. N. Am., p. 535, 1857. Jdnk.1900.] MICROTUS PENNSYLVANICUS FONT1GENUS. 21 plains of the Columbia in eastern Washington are too small and dark to be typical modestus. Specimens examined. — Total number, 259, from the following localities : Colorado: Cochetopa Pass, 89; Fort Garland, 15; Loveland, 7; Twin Lakes, 1. Wyoming : Newcastle, 1 ; Bear Lodge Mountains, 2 ; Sundance, 1 ; Lower Geyser Basin, Yellowstone Park, 1. South Dakota: Custer, 2; Hill City, 1. North Dakota: Fort Buford, 1. Montana: Little Bighorn River, 2; Fort Custer, 3; Bozeman,2; Fort Ellis, 1; Big Snowy Mountains, 13; Pkilbrook, 1 ; Stanford, 1; Choteau, 1; Robare, 1; Blackfoot, 1; Fort Assinniboine, 1; Tobacco Plains, 3; Stillwater Lake, 8; Flathead Lake, 9; Little Bitterroot Creek, 2; Hot Springs Creek, 1; Horse Plains (8 miles east), 1. Idaho: Lemhi, 1 ; Salmon River, 3; Challis, 3; Birch Creek, 24; Cceur d'Alene^ 3; Fort Sherman, 1. Washington: Marshall, 15; Coulee City, 4; Conconully, 4; Colville, 20; Marcus, 1. Utah: Ogdcn, 7; Salt Lake, 1. MICROTUS PENNSYLVANICUS FONTIGENUS (Bangs). Forest Vole. Microtia fontigenus Bangs, Proc. Biol. Soc. Wash., X, 48-49, March 9, 1896. Microtus pennsylvanicus fontigenus Miller, Proc. Boston Soc. Nat. Hist., XXVIII, 14, April, 1897. Type locality. — Lake Edward, Quebec. Geographic distribution. Eastern Canada, in the Hudsonian zone. Habitat. — Marshes, fields, dry banks, and deep woods. General characters. — Smaller than pennsylvanicus, with short wide skull, large round bulla', and short incisive foramina. Color. — Autumn pelage (September specimens in long fur): Upper- parts dark bister mixed with black, slightly paler on sides and cheeks; belly washed with whitish or smoky gray; tail bicolor, blackish above, grayish below; feet plumbeous. Cranial characters. — Skull light and smooth, not ridged or angular; rostrum and incisive foramina short; braincase wide; audita! bulla' large and smoothly rounded; interpterygoid space narrow, ending squarely at palate; dentition as in pennsylvanicus. Measurements. — Type, 9 ad. : Total length, 151; tail vertebra1, 41 ; hind foot, 21. Topotype, S ad.: 150; 45; 21. Skull (No. 3839, 3): Basal length, 23, nasals, 6.2; zygomatic breadth, 14.3; mastoid breadth, 11.5; alveolar length of upper molar series, 6. Remarks. — The short rostrum, short, wide braincase, and short inci- sive foramina distinguish fontigenus from both pennsylvanicus and drum- mondi, with both of which species it seems to intergrade. In size it is intermediate, and in external characters not very dif- ferent from either. It is recorded by Mr. Miller from Nepigon and Peninsula Harbor, Ontario, but he considers the specimens obtained at those places not quite typical.1 A series of 10 specimens from Godbout, Quebec, are rather nearer fontigenus than acadicus. Specimens examined. — Total number, 0, topotypes (from the Bangs Coll,). 1 Proc. Boston Soc. Nat. Hist,, XXVIII, 14, April, 1897. 22 NORTH AMERICAN FAUNA. [no. 17. MICROTUS PENNSYLVANICUS LABRADORIUS Bailey. Little Labrador Vole. Microtus pennsylvanicus laoradorius Bailey, Proc. Biol. Soc. Wash., XII, 88, April 30, 1898. Type locality. — Fort Chimo, Ungava, Labrador. Geographic distribution. — Known only from the type locality. General characters. — Size of Microtus drummondi and of approxi- mately the same proportions. Skull flatter, with smaller audital bullae and more protruding upper incisors. Color. — (Much changed by alcohol.) Upperparts dark brownish; belly whitish; tail bicolor; feet pale. Cranial characters. — Skull low, not much ridged or angled ; postor- bital ridge prominent; nasals short, cuneate and scarcely reaching base of incisors; audital bulla? small; incisive foramina short; first upper molar usually with an inner posterior point, molar pattern other- wise as in pennsylvanicus. The skull is readily distinguishable from that of either drummondi or fontigenus by the protruding incisors and small audital bullae. Measurements. — Type, 2 ad. (in alcohol), measured by Dr. 0. Hart Merriam : Total length, 139 ; tail vertebra1,, 39 ; hind foot, 20. Average of 7 alcoholic specimens from type locality: 137; 37; 19. Skull (of type): Basal length, 24.3; nasals, 6.7; zygomatic breadth, 14.4; mastoid breadth, 11; alveolar length of upper molar series, G.2. Remarks. — M. labradorius shows closer affinity with drummondi than with fontigenus, though no doubt meeting and grading into the latter. It is widely different from enixus, and the two occur together at Fort Chimo. Specimens examined. — Total number, 9, from the type locality. MICROTUS DRUMMONDI (Ami. & Bach.). Druinmond Vole. Arvicola drummondi Aud. and Bach., Quad. N. Am., Ill, 166-167, 1854. Arvicola (Mynomes) microcephalia Rhoads, Proc. Acad. Nat. Sci. Phila., 1894, 286-287 (Lac La Hache, B. C). Microtus stonei Allen, Bui. Am. Mus. Nat. Hist., XII, 4, March, 1899 (Liard River, N.W.T.). Type locality. — Rocky Mountains, vicinity of Jasper House, Alberta. Geographic distribution. — From Hudson Bay to the west slope of the Rocky Mountains aud Alaska, and from the northern edge of the United States north to Fort Anderson, N. W. T., in Canadian and Hudsouian zones. Habitat. — Both marshes and dry upland. General characters. — Similar to Microtus pcnnsylv aniens, but much smaller, with slenderer feet aud tail, and paler color. Color. — Summer pelage: Upperparts yellowish bister with numerous dark-brown- or black-tipped hairs, sides of nose and hairs in front of ears more decidedly yellowish; belly white or rarely tinged with buffy, sometimes dusky during the molt; feet silvery gray; tail bicolor, black- ish above, whitish below. Winter Pelage: Paler than in summer; yel- low on ears and nose more conspicuous; Young: Paler and not so sooty as young pennsylvanicus. June, 1900.] MICROTUS DRUMMONDI. 23 Cranial characters. — Skull not much arched and rather flat topped, slender but sharply ridged in adults; audital bullae large and smoothly rounded; palate flattened in immature specimens, becoming higher with deep lateral pits in adults. Dentition as in pennsylvanicus. Except for the larger bullae and a few characters of minor weight, the skull of drummondi is a miniature of the skull of pennsylvanicus. Measurements. — Average of 6 adult males and females from Muskeg Creek, Alberta: Total length, 145; tail vertebra1, 39; hiud foot, 17.8. Largest specimen from Muskeg Creek : 160; 41; 18. Skull (No. 81487, 9 ad., same locality) : Basal length, 23 ; nasals, 6; zygomatic breadth, 14; mastoid breadth, 11; alveolar length of upper molar series, 6. Remarks. — The characters separating drummondi from pennsylvanicus and modestus are relative. There is no sharp distinction and the forms either merge into each other, or after approaching each other in size overlap in range and occur together at the same localities. Specimens from Blackfoot, Montana, are nearer modestus, while those from Summit and St. Marys Lake are almost typical drummondi. At Elk River, Minnesota, about half of the specimens are almost typical drummondi, while the others are a small form of pennsylvanicus ; but, as others fall between and cannot positively be placed with either form, it is impos- sible to decide whether the difference is due to individual variation in an intermediate form or whether two species meet and interbreed. To the eastward drummondi merges into fontiyenus, from which it differs in such slight degree that the two are not easily distinguishable. On the west slope in British Columbia drummondi becomes slightly darker, and in that respect less unlike fontigenus in appearance, but retains its cranial characters, Alaska and northwestern specimens (from Fort Wrangel, Nulato, and Fort Simpson) are larger and in gen- eral appearance less different from pennsylvanicus, but detailed cranial characters show them to be but a robust form of drummondi. Speci- mens from Liard River, including the type of M. stonei, are indistin- guishable from typical drummondi. Specimens examined. — Total number, 355, from the following localities: Alberta: Muskeg Creek (15 miles south of Smoky River), 13; Smoky Valley (50 miles north of Jasper House), 5; Fishing Lake (90 miles uorth of Jasper house), 2 ; Henry House ( 15 miles south), 4 ; South Edmonton, 110; St. Alberts, 26; Canmore, 2; Banff, 1; Red Deer, 1. Athabasca: Lesser Slave Lake, 1. Assiniboia: Indian Head, 30; Medicine Hat, 1. British Columbia: Shuswap, 13; Sicamous, 1; Cariboo Lake, 2 ; Okanagan, 11; Ducks, 2; Stuart Lake, 1. Saskatchewan: Cumberland House, 2; St. Louis, 1. Manitoba: Carberry, 2. Ontario: Rat Portage, 2; Coney Island (Lake of the Woods), 2. Northwest Territory : Fort Churchill, 2; Fort Simpson, 2; Fort Rae, 3; Fort Reliance, 1 ; Big Island (Great Slave Lake), 1 ; Fort Good Hope, 1; Fort MacPherson, 1; Fort Anderson, 2; head of Liard River, 1; Chandindu River, 1 ; Dawson, 1 ; Sixty-Mile Creek, 1 ; Fort Selkirk, 3 ; 50 miles below Fort Selkirk, 1 ; Rink Rapids, 6 ; Thirty-Mile River, 3 ; Lake Lebarge, 4; Fifty-Mile River, 6; Lake Marsh, 22; Caribou Crossing, 16. 24 NORTH AMERICAN FAUNA. [no. 17. Alaska: Fort Wrangell, 1; Nulato, 1; Charlie Creek (Yukon River), 2; Cana- dian Boundary and Yukon River, 4. North Dakota: Portland, 4; Lisbon, 1; Bottineau, 1. Montana : St. Marys Lake, 6 ; Summit, 8. Idaho : Priest Lake, 5. Washington : Loon Lake, 9. MICROTUS AZTECUS (Allen). Aztec Vole. Arricola (Mynomes) aztecus Allen, Bui. Am. Mus. Nat. Hist., V, 73-74, April 28, 1893. Type locality. — Aztec, New Mexico (altitude, 5,900 feet). Geographic distribution. — Valley of the San Juan River in north- western ISTew Mexico, in Transition zoue. Habitat. — Grassy places along the river banks and near irrigation ditches. General characters. — In size similar to M. pennsylvanicus, but with shorter tail and larger hind foot; skull long; braincase narrow; inter- parietal long. Color. — Winter pelage: Upperparts dull buffy, heavily mixed with black hairs; belly washed with creamy or pale buff; feet plumbeous; tail sharply bicolor, black above, soiled whitish below. Young (half- grown specimens in Dec.) : Scarcely different in color from adults. (Summer pelage not examined.) Cranial characters. — Skull long; braincase high and narrow; inter- parietal more than half as long as wide, extending back to plane of foranien magnum; audital bull* large; mandible short and heavy; angular process of mandible wide; dentition as in pennsylvanicus. Measurements. — Average of 7 adults from type locality: Total length, 171; tail vertebra?, 43; hind foot, 22. Slcull (No. 57432, & ad.): Basal length, 28.8; nasals, 8.4; zygomatic breadth, 17 ; mastoid breadth, 12.5; alveolar length of upper molar series, 7.2. Remarks. — Microtus aztecus belongs to the pennsylvanicus group. Externally it is not very different from modestus, but none of the specimens show any signs of intergradation ; and the skull characters are so well marked that there seems no doubt of its full specific rank. Specimens examined. — Total number, 45, from the following localities: New Mexico : Aztec, 42 ; La Plata, 3. MICROTUS ENIXUS Bangs. Large Labrador Vole. Microtus enixus Bangs, Am. Nat., XXX, 1051-1052, Dec. 5, 1896. Type locality. — Hamilton Inlet (north shore), Labrador. Geographic distribution. — Eastern coast of Labrador from Hamilton Inlet to Ungava Bay, in Hudsonian zone. General characters. — Size slightly larger than M. pennsylvanicus, with relatively longer tail and larger ears; coloration duller and darker; skull smaller, shorter, and wider, with lighter molars. Hip glands present in adult males. June, 1900.1 MICROTUS TERR^ENOV^. 25 Color. — Summer pelage (July and Aug. specimens) : Upperparts dark yellowish bister mixed with blackish; belly smoky gray or soiled whitish, sometimes tinged with buffy; feet dusky or blackish; tail black above, grayish brown below. Cranial characters. — Outline of skull shorter, wider, and less arched than in pennsylvanicus; prezygomatic notch deep; coronoid notch of mandible wide and rounded; molar series small and slender; m2 with posterior loop completely closed and circular in form; m3 with termi- nal loop shortened ; dentition otherwise as in pennsylvanicus. Measurements.1 — Type, 9 ad.: Total length, 210; tail vertebra?, 67; hind'foot, 22.5. Average of 10 adult topotypes : 189.4 ; 60.1 ; 22.1. SI nil (No. 1018, 9 ad.): Basal length, 27.3; nasals, 8.6; zygomatic breadth, 16.7; mastoid breadth, 12.3; alveolar length of upper molar series, 6.5. Remarks. — Microtus enixus appears to be a distinct and well marked species of the pennsylvanicus group, the long tail and small molars being the most convenient characters for recognition. From its geo- graphically nearest neighbors, Microtus p. fontigenus, of Lake Edward, Quebec, and Microtus p. labradorms, of Ungava, it shows a wider difference than from typical pennsylvanicus. Specimens examined. — Total number, 16, from the following localities: Labrador: Hamilton Inlet, 13; Fort Chimo, Ungava, 3. MICROTUS TERRiENOV^E (Bangs). Newfoundland Vole. Arvicola ierrcenom Bangs, Proc. Biol. Soc. Wash., IX, 129-132, July 27, 1894. Type locality. — Codroy, Newfoundland. Geographic distribution. — Newfoundland and Penguin Island. General characters. — Slightly larger thstn pennsylvanicus, with decid- edly larger hind foot, and more yellowish colors; belly with dusky median line; nose patch buffy; skull wide and angular. Color. — Summer pelage fin July and August specimens): Upper- parts dark russet, darkened by brown-tipped hairs, becoming paler on sides and across face; nose patch dark buffy or dull russet; belly whit- ish or smoke gray with a median streak of dusky cinnamon; tail dis- tinctly bicolor, blackish above, soiled whitish below; feet grayish brown. Winter pelage (retained in April specimens): Slightly paler russet above, whiter below, with sharper markings throughout. Young (nearly half-grown): Similar to adults. Cranial characters. — Skull short with wide-spreading zygomata; pre- zygomatic notch deep; nasals terminating even with arms of premax- illae; lateral pits of palate deep and wide; interpterygoid space narrow with median constriction; ml with an anterior spur or loop; m3 with posterior loop short, irregularly rounded, triangular or trifoliate; m2 and m3 normally with anterior point or spur, 1 From original description. 26 NORTH AMERICAN FAUNA. [no. 17. Measurements. — Average of 10 adults from type locality : Total length, 182; tail vertebrae, 52; hind foot, 23.4. Skull (No. 74029, $ ad.): Basal length, 28.5; nasals, 8.4; zygomatic breadth, 17; mastoid breadth, 13; alveolar length of upper molar series, 7. Remarks. — Microtus terrcenovcs shows very distinctive characters, and no close affinity with any neighboring species. Specimens showing very young and full winter pelage are still needed for a comprehensive description Specimens examined. — Total number, 43; from the following localities: Newfoundland : Coilroy, 35 ; Penguin Island, 8. MICROTUS BREWERI (Baird). Beach Vole. Arvicola Iretveri Baird, Mauira. N. Am., 525-526, 1857. Type locality. — Muskeget Island, Massachusetts. Geographic distribution. — Muskeget Island. Habitat. — Beach plum thickets on the sandy island. General characters. — Size a little larger than pennsylvanicus; colors pale grayish; pelage long and coarse; skull heavy with wide nasals and quadrate interparietal. Color. — Summer pelage (July 18) : Upperparts buffy gray with scat- tered brown- and black-tipped hairs, paler on sides; belly tinged with sulphur yellow; feet silvery gray; tail bicolor, rusty brown or blackish above, soiled whitish below. Young: Rather paler and duller than adults. Cranial characters. — Skull massive; nasals wide anteriorly; inter- parietal more than half as long as wide; inner edges of zygomata sharply notched close to lachrymals; rostrum heavy; upper incisors bent abruptly downward; molar pattern not very different from that of pennsylvanicus. Measurements. — Average of 10 adults from Muskeget Island : Total length, 182; tail vertebra?,, 54; hind foot, 22.3. Skull (No. 73141, $ ad.) : Basal length, 28.7; nasals, 8.3; zygomatic breadth, 17.3; mastoid breadth, 13; alveolar length of upper molar series, 7.2. Remarks,. — Nine of the 2G specimens have a small white spot on the forehead. This may be accidental or an only partially acquired character. Specimens examined. — Total number, 2G; all from the type locality. MICROTUS NESOPHILUS Bailey. Gull Island Vole. Microtus insularis Bailey, Proc. Biol. Soc. Wash., XII, 86, April 30, 1898. Name preoccupied by Lemmus insularis, Nillson (= Microtus agrestis L. ). Microtus nesophilns Bailey, Science, N. S., VIII, 782, Dec. 2, 1898. Type locality. — Great Gull Island, New York. Geographic distribution. — Great Gull Island (at entrance to Long Island Sound). June, 1900.] MICROTUS MONTANUS. 27 General characters. — Size of pennsylvanicus ; colors darker ; skull shorter and wider with more spreading zygomata and deeper prezygo- matic notches. Color. — Summer pelage: Upperparts, dark yellowish bister heavily mixed with black hairs, darkest on nose and face; belly dusky, washed with cinnamon; feet blackish; tail blackish above, dark brown below. Cranial characters. — The skull differs from that of pennsylvanicus in shorter, wider braincase, wider and more abruptly spreading zygomatic arches, more expanded jugal, and smaller audital bulla'; palate short, with a median point or spur and deep lateral pits; mo normally with anterior inner and outer triangles approximately opposite and conflu- ent; dentition otherwise similar to that of pennsylvanicus. Measurements. — Type (measured in dry skin): Tail, 29; hind foot, 20. No. 1943, Am. Mus., $ ad., 185: 41 : 21. SJcull (No. 539G9) : Basal length, 2G; zygomatic breadth, 1G.2; mastoid breadth, 12.3; alveolar length of upper molar series, G.8. (No. 1943) 26.6; 8.5; 16.2; 13; 6.6. Bemarlts. — Microtus nesophilus needs no comparison with breweri or ternvnovee, the other two insular forms from the Atlantic coast. In general appearance it more nearly resembles pennsylvanicus, but in cranial characters it is as distinctly different as either of the other island species. During the month of August, 1898, Mr. A. H. Howell visited Great Gull Island for the purpose of getting specimens of Microtus, but he found their old haunts covered by the earth moved in grading the island for fortifications, while no trace of the animals remained. He thinks they are completely exterminated. Specimens examined. — Total number, 15; all from the type locality. MICROTUS MONTANUS (Peale). Peale Vole. Arvicola i.iontanus Peale, U. S. Exploring Exp'd., Mammalogy, 41, 1848. Aroicola longiroslris Baird, Mamm. N. Am., 530-531, 1857. (From Tipper Pitt River, California.) Type locality. — Headwaters of Sacramento River, near Mount Shasta, California. Geographic distribution. — Northeastern California, eastern Oregon, northern Utah and Nevada, in the Upper Sonoran and Transition zones. Habitat. — Marshes, meadows, and tule swamps. General characters. — Size medium (about as in pennsylvanicus) ; tail about twice as long as hind foot; colors dark; hip glands conspicuous in adult males; incisors projecting well in front of nasals; incisive foramina narrow and constricted posteriorly. Color. — Summer pelage : Upperparts bister or ashy mixed with blackish; belly washed with soiled whitish, giving a smoky gray or dusky color; feet plumbeous; tail indistinctly bicolor, blackish above, plumbeous below; lips usually showing a trace of whitish. 28 NORTH AMERICAN FAUNA. [no.17. Cranial characters. — Skull generally slender and smooth, becoming angular and ridged in only a few very old individuals ; nasals narrow and short; interparietal wide and normally strap- shaped; incisive foramina narrow and constricted posteriorly; bulla? medium and well rounded; dentition rather light; m2 with 4 closed sections; m3 with 3 closed triangles; ml nor- mally with 5 closed triangles. Measurements. — Average of 10 specimens from Sisson, Cal. : Total length, 175; tail vertebra?., 52; hind foot, 21.5. Extremely large specimens from Fig. 2.-Mo1ar enamel pattern Sisson run ag bigll ag 192 . 54. 23. 8MU (No. of Microtus montanus (X5). nnnnt\ -, n r, . , , 9bb89, $ ad., from Sisson): Basal length, 28; na- sals, 8; zygomatic breadth, 17; mastoid breadth, 13.5; alveolar length of upper molar series, 7.3. General remarks. — The original description of M. montanus, though meager, agrees in all particulars with the animal from Sisson, at the west base of Mount Shasta. The measurements (total length, Clinches; tail, 1J inches=15G mm. and 38 mm.) give it too short a tail, which only serves to restrict it more closely to this form in distinction from either of the longer-tailed species (mordax or californicus) that occur at or near the type locality. Three mounted specimens in the United States National Museum, which Baird referred to montanus,1 and which came from Upper Klamath Lake and the Upper Des Chutes, are identical with those of the present series from Sisson, Fort Klamath, and Fort Crook. Specimens from the south end of Goose Lake, which is the source of Pitt Eiver, are the same as those from Sisson and from Fort Crook, lower down the river, and also the same as Baird's type of M. longirostris from 'Upper Pitt River.' M. montanus has a somewhat scattered and interrupted distribution and shows considerable geographic variation in widely separated local- ities. The extreme development of large size, large feet, and heavy angular skull is found in the big marshes of the Carson Sink, Nevada, while specimens from higher levels in the Transition zone are smaller, with slenderer feet and grayer coloration. The variation is mainly, but not entirely, zonal. To separate either extreme would tend to con- fusion rather than convenience, as the extremes point to nevadensis on the one hand and to nanus on the other. Specimens examined. — Total number, 2G3, from the following localities : California: Sisson, 57; Fort Crook, 23; Hayden Hill, 2; Fall Lake, 2; Cassel, 1; Tule Lake, 2; Goose Lake, 8; Greenville (8 miles NW.), 3; Bucks Ranch (Plumas Co.), 1; Quincy, 3; Summit, 1. Nevada: Washoe, 1; Deep Hole (south end of Granite Range), 1; Pine For- est Range, 1; Mountain City, 4; Wells, 13; Austin, 7; Carson, 16; Still- water, 10 ; Newark Valley, 5 ; Monitor Valley, 5 ; Ruby Lake, 5 ; Ruby Mountains, 3. i Mamm. N. Am., 530, 1857. June, 1900.] MICROTUS MONTANUS RIVULARIS. 29 Oregon: Klamath Basin (Lost River), 8; Klamath Falls, 1; Fort Klamath, 5; Swan Lake Valley, 5; Plush (west side Warner Lake), 9; Shirk, 2; Steen Mountains (east slope), 1; Summit NE. of Steen Mountains), 2; Burns, 2; Wapinitia, 4. Utah: Ogden, 10; Salt Lake City, 3; Provo, 1; Fairfield, 10; Manti, 20. MICROTUS MONTANUS ARIZONENSIS Bailey. Arizona Vole. Microtna montanus arizonensis Bailey, Proc. Biol. Soc. Wash., XII, 88, April 30, 1898. Type locality. — Springerville, Ariz. Geographic distribution. — Plateau country of eastern Arizona, at head of Little Colorado, in the Transition Zone. Habitat. — Creek banks and damp meadows. General characters. — Similar to M. montanus, but brighter and more ferruginous in color; lateral pits of palate shallower. Color. — Early winter pelage (October and November specimens): Upperparts yellowish or rusty brown; belly washed with white; feet dark grayish; tail bicolor, blackish above, grayish below; lips whitish. Slightly immature specimens are a little duller colored than adults. Cranial characters. — Skull very similar to that of montanus, but easily distinguished by the flatter palate with shallower lateral pits and by thicker pterygoids; condyloid process of mandible slightly shorter. Dentition not different. Measurements. — Type: Total length, 184; tail vertebra*, 55; hind foot, 20. Average of 7 specimens from type locality: 158; 41; 20.0. Skull (of type): Basal length, 27.3; nasals, 8; zygomatic breadth, 10; mastoid breadth, 12.2; alveolar length of upper molar series, 0.5. General remarks. — Although widely separated geographically from M. montanus by desert country through which continuity of range is improbable, this form is so closely related to that, species that its posi- tion is best indicated by subspecific rank. Specimens examined. — Total number, 12, from the following localities: Arizona: Springerville, 11. New Mexico : Nutria, 1. MICROTUS MONTANUS RIVULARIS Bailey. Utah Vole. Microtus nevadensis rivitlaris Bailey, Proc. Biol. Soc. Wash., XII, 87, April 30, 1898. Type locality. — St. George, Utah. Geographic distribution. — Known only from type locality, probably restricted to Lower Sonoran zone. Habitat. — Tule marshes along the banks of the Virgin Eiver. The runways were always found in wet places among sedges and rushes. General characters. — Larger and lighter colored than typical mon- tanus; skull more arched; nasals narrower. Color. — Winter pelage: Upperparts dull bister, darkened with black ish-tipped hairs; sides scarcely paler; belly washed with white j feet dull grayish; tail bicolor, blackish above, grayish below. Young: Darker than adult, but not black backed as in nevadensis. 30 NORTH AMERICAN FAUNA. [no. 17. Cranial characters. — Skull well arched, not much ridged; nasals con- spicuously narrower than in montanus; frontals narrower posteriorly ; basioccipital more constricted anteriorly; dentition essentially the same. Measurements. — Type: Total length, 179; tail vertebrae, 48; bind foot, 23. A nearly adult female topotype: 1G3; 43; 21. Skull (of type) : Basal length, 28.2; nasals, 8.3; zygomatic breadth, 17; mastoid breadth, 13.3; alveolar length of upper molar series, 7.3. General remarks. — Since rivularis was described in 1898 as a sub- species of nevadensis, a series of 50 specimens has been collected at the type locality of montanus, showing for the first time the real characters and the range of variation in that species, and, moreover, as Dr. Merriam had previously suggested to me, that rivularis comes nearer to typical montanus than to nevadensis. So far as at present known it has an isolated range in a Lower Sonoran valley, but it may readily extend northward to meet and merge into montanus in central Utah. Specimens examined. —Total number, 4, from the type locality. MICROTUS NANUS (Merriam). Dwarf Vole. Arvicola nanus Merriam, North American Fauna No. 5, 62-63, pi. II, figs. 5 and 6, July 30, 1891. Type locality. — Pahsimeroi Mountains, Idaho (altitude 9,350 feet). Geographic distribution. — Rocky Mountains and outlying ranges, from central Idaho southward to central Nevada and southern Colo- rado, in Canadian zone. Habitat. — Dry, grassy parks on mountain slopes. General characters. — Size small; tail short; ears short and rounded; color dark grayish ; skull slender. Color. — Summer pelage: Upperparts uniformly grizzled gray mixed with sepia and blackish hairs; belly washed with white; feet grayish or plumbeous; tailbicolor, dusky gray above, whitish below. ( Winter pelage unknown.) Young: Similar to adult, but slightly duller throughout. Cranial characters. — Skull small, slender and well arched, with slen- der zygomata and large well-rounded bullae; superciliary ridges prom- inent, sometimes confluent in old age; incisors projecting well beyond nasals; molars light, with short, wide triangles; enamel pattern scarcely distinguishable from that of mordax and montanus. Measurements. — Type, 6 ad.: Total length, 151; tail vertebrae, 41; hind foot, 18. Average of five adults from type locality: 143; 37; 18.4. Skull (of type) : Basal length, 23.7; nasals, G.7; zygomatic breadth, 14; mastoid breadth, 10.8; alveolar length of upper molar series, 5.8. Remarks. — Microtus nanus belongs to the montanus group but occu- pies a higher zone and has more of the habits of Pedomys or Lagurus. It is rarely found in wet places or near water. Specimens from certain isolated localities are not entirely typical, but do not differ enough to warrant separation. June, 1000.] MICROTUS NANUS CANESCENS. 31 Specimens exa m ined. — Total number, 114, from the following localities : Idaho: Pahsimeroi Mountains, 13; Lost River Mountains, 1; Ckallis, 7; Saw- tooth Lake, 5; Tkree Creek, 3; Montpelier Creek, 3; Seven Devils Moun- tains, 4. Utah: Uinta Mountains, head of Smitk Fork, 1. Wyoming: Fort Bridger, 9; Kinney Ranck, 6; Beaver, 1 (no skull) ; La Barge Creek (near kead), 1; Ckeyenne, 3; Sherman, 2; Laramie, 2; South Pass City, 20; Bighorn Mountains, head of Powder River, 9. Montana: Beartooth Mountains, 3; Big Snowy Mountains, 1. Colorado: Estes Park, 1; Cocketopa Pass, 17 ; Twin River, 1; Twin Lakes, 1. MICROTUS NANUS CANESCENS Bailey. Gray Vole. Microtus nanus ranescens Bailey, Proc. Biol. Soc. Wash., XII, 87, April 30, 1898. Type locality. — Conconully, Washington. Geographic distribution. — Northern Washington and southern British Columbia, east of the Cascades. Apparently confined to the Transi- tion zone. Habitat. — Dry grassy ground. General characters. — Like nanus but lighter, clearer gray; skull with larger bullae and greater mastoid breadth; zygomatic arches less widely spreading; upper incisors bent more abruptly downward. Hip glands conspicuous in adult males. Color. — Summer pelage: Upperparts clear, dark grayish, formed by pale buffy and black- tipped hairs; sides shading to lighter gray and belly to white; feet dark gray; tail bicolor, blackish above, grayish below. ( Young and winter pelage not shown in present material.) Cranial characters. — Skull slightly narrower and more elongate than in nanus; interparietal averaging longer; bullae decidedly larger and fuller; mastoid breadth relatively greater; incisors scarcely reaching beyond nasals; molar pattern as in nanus. Measurements. — Type: Total length, 149; tail vertebra?, 42; hind foot, 20. STcull (of type): Occipital condyle to anterior base of molars, 17.4; posterior tip of nasals to foramen magnum, 19.2; zygomatic breadth, 15; mastoid breadth, 12.3; alveolar length of upper molar series, G.3. General remarks. — In its extreme development this northern form is readily distinguishable from typical nanus. From intermediate locali- ties, Flathead Lake and the Plains of the Columbia, specimens are not typical of either but show slight peculiarities of local development interesting in themselves but not sufficiently marked for even sub- specific distinction. To a certain extent they are intermediate between nanus and canescens. Specimens examined. — Total number, 47, from the following localities: British Columbia: Okanagan, 11; Ducks, 2; Vernon, 7. Washington: Conconully, 1 ; Wenatchee, 1; Fort Walla Walla, 1; Oakesdale, 2; Wawawai, 4; Ckeney, 1. Oregon: Elgin, 2; Wallowa Mountains (near Joseph), <>. Montana: West arm of Flatkead Lake, 5; Hot Spring Creek (a branch of the Little Bittcrroot), 4. 32 NORTH AMERICAN FAUNA. [no. 17. MICROTUS CANICAUDUS Miller. Gray-tailed Vole. Microtus canicaudus Miller, Proc. Biol. Soc. Wash., XI, 67-68, April 21, 1897. Type locality. — McCoy, Oregon. Geographic distribution. — Willamette Valley, Oregon, and the east base of the Cascades in southern Washington, in Transition zone. General characters. — Size and proportions about as in nanus; ears larger, skull heavier, more arched, with fuller, rounder bullae, and shal- lower lateral pits of palate, coloration more yellowish, tail grayer. Color. — Winter pelage: Upperparts bright yellowish bister, darkened with blackish-tipped hairs, slightly paler on sides; belly and whole lower parts whitish-gray; feet grayish or pale plumbeous; tail in winter adults uniformly grayish, with a half-concealed dusky dorsal line. In the only summer specimen (from North Yakima, Wash., and perhaps not typical) the tail is sharply bicolor with a blackish dorsal line. Young (half-grown November specimens): Sooty gray above and scarcely lighter below; feet dusky; tail gray, with a blackish dorsal stripe. Cranial characters. — Skull high, smooth, and well arched, with scarcely a trace of superciliary ridges; interparietal lozenge-shaped; bill las larger and more rounded than in nanus; interpterygoid fossa narrower and more acuminate; lateral pits of palate shallower; incisors less protruding; enamel pattern of molars the same as in nanus. Measurements. — Type: Total length, 135; tail vertebras, 33; hind foot, 20. Average of 8 adults from type locality: 141; 35.7; 20. Skull (of type): Basal length, 21.2; nasals, 7.3; zygomatic breadth, 15.3; mastoid breadth, 12.8; alveolar length of upper molar series, 6. Remarlis. — A single specimen with a badly broken skull from North Yakima, Wash., seems to be true canicaudus in summer pelage, and indicates that the range of the species is much more extensive than is at present known. Specimens examined.— -Total number, 14, from the following localities: Oregon: McCoy, 9; Beaverton, 2; Sheridan, 2 (im). Washington: North Yakima, 1. MICROTUS DUTCHERI Bailey, Dutcher Vole. Microtus dutcheri Bailey, Proc. Biol. Soc. Wash., XII, 85, April 30, 1898. Type locality. — Big Cottonwood Meadows, near Mount Whitney, Cal- ifornia (altitude, 10,000 feet). Geographic distribution. — Iludsonian zone of the southern Sierra Nevada. Habitat. — Wet alpine meadows. General characters. — Size rather small; tail short; ears small, nearly concealed by fur; colors dark above and below; lips and usually nose white; hip glands present in adult males. Color. — Summer pelage : Upperparts dark bister with brown tips to the long hairs; below, dull cinnamon or buffy-brown; feet whitish or June, 1900.] MICROTUS NEVADENSIS. 33 plumbeous-gray; tail bicolor, brown or blackish, above, whitish below; lips and usually tip of nose white. ( Winter pelage unknown.) Young: Dull brown above and scarcely lighter below; feet and tail blackish; lips and nose usually white. Cranial characters. — Skull similar to that of montanus but differing in many details; rostrum slightly longer; bulla; smaller and less globu- lar; lateral pits of palate shallower; dentition the same. Measurements. — Type, $ ad.: Total length, 1G7; tail vertebrae, 35; hind foot, 20. Average of 10 adults, 5 males and 5 females, from type locality: 163; 37; 20.6. SJcull (of type): Basal length, 27.4; nasals, 8; zygomatic breadth, 16.7; mastoid breadth, 12.2; alveolar length of upper molar series, 6.5. General remarks. — The nearest relative of M. dutcheri is montanus, but the two species occupy widely separated zones, and show no evi- dence of inter gradation. Specimens examined. — Total number, 65, from the following localities in the Sierra Nevada: California: Big Cottonwood Meadows, 28; Whitney Meadows, 11; Menaclie Meadows, 2; Olancha Peak, 3; Head of Kern River, 1; Mammoth, 12; Pine City, 3; Head of San Joaquin River, 5. MICROTUS NEVADENSIS Bailey. Nevada Vole. Microtus nevadenais Bailey, Proc. Biol. Soc. Wash., XII, 86, April 30, 1898. Type locality. — Ash Meadows,1 Nye County, Nevada. Geographic distribution. — Known only from the type locality and Pahranagat Valley, about 100 miles to the northeast. Both localities are in the Lower Sonoran zone. Habitat. — Salt grass and tule marshes in alkaline valleys. Runways always found in wet, muddy places, and often extending through shal- low water. General characters. — Size large; ears small; tail rather short; fur coarse and lax; colors dark; hip glands conspicuous in adult males. Skull massive and angular; incisive foramina narrow and closing to a point posteriorly. Color. — Whiter pelage (March specimens): Upperparts dark sepia or bister, much obscured by blackish hairs; sides lighter; belly smoky gray; feet dark gray; tail indistinctly bicolor, blackish above, gray or brownish below ; lips usually white ; tip of nose usually whitish. Young: With a blackish dorsal stripe and dusky feet and tail. Cranial characters. — Skull heavy, angular, and much ridged; fron- talshigh; rostrum bent downward; nasals truncate or rounded pos- teriorly, terminating even with arms of premaxillse; incisive foramina short, rather narrow and constricted to a point posteriorly; dentition heavy; upper incisors curved abruptly downward; molar pattern vari- 'Ash Meadows is on the Nevada side of the Nevada-California line near wly Vernon Bailey. Original number, 2520. Geographic distribution. — Known only from the type locality. Habitat. — Wet ground under tall tales (Scirpus olneyi), where the runways extend through mud and water in a little marsh around a warm spring. General characters. — Size and proportions about as in edax, colors not so dark, tail long, skull heavy and angular, middle upper molar with rounded open or elosed posterior loop. Colors. — Winter pelage: Upperparts dark buffy gray, slightly darker than in califomicus, but not so black as in edax; belly smoky gray, tail indistinctly bicolor, brown above, grayish below; feet brownish gray, not dusky. Young: Upperparts black, belly grayish, a black dor- sal stripe retained until the animals are half grown. Cranial characters. — Skull of adult angular and heavily ridged ; in general characters resembling that of edax, but with more truncate posterior tip of nasals, heavier dentition, and well-developed inner pos- terior loop of middle upper molar. The same characters and larger size distinguish it from those of vallicola and calif ornicus, and the wide incisive foramina with many other characters distinguish it from that of its nearest neighbor — nevadensis. Measurements. — Type: Total length, 210; tail vertebra1, G7 ; hind foot, 25. Average of 6 adults: 203; 65; 25.1. Skull (of type) : Basal length, 31; nasals, 10; zygomatic breadth, 19; mastoid breadth, 13.0; alveolar length of upper molar series, 8.7. June, 1900.] MICROTIA OPERARIUS. 39 General remarks. — Microtus scirpensis stands nearest to M. edax, and, except for the more completely developed posterior loop of middle upper molar, fits into the californicus group. Among 14 specimens the loop is closed in 7 and open in 7, while among 43 specimens of edax it is closed in 2, open in 32, and absent in 9, and among 100 specimens of californicus it is closed in 3, open in 20, and absent in 77. Although resembling pennsylvanicus in the fifth loop, in other characters it does not approach that group or any of its forms. In range it comes nearer to vallicola than to edax or californicus, but from rallicola it differs in the same way as from californicus. Specimens examined. — Total number, 14, from the type locality. MICROTUfl OPERARIUS (Nelson). Tundra Vole. Arvicola operarius Nelson, Proc. Biol. Soe. Wash., VIII, 139, Dec. 28, 1893. Type locality. — St. Michael, Alaska. Geographic distribution. — Barren grounds from Bristol Bay, St. Michael, and Kowak River, Alaska, east to Anderson River. Habitat. — Mossy tundras. General characters. — Size small; tail short, densely haired; ears small and wholly concealed in long winter fur: colors yellowish; skull slender and narrow ; dentition light. Color. — Winter pelage: Upperparts dark rich buff, slightly tinged along back with black-tipped hairs; sides paler; belly pale buffy or creamy white; tail soiled whitish below and on sides, a partly con- cealed blackish dorsal line; feet gray; heels tinged with dusky. Sum- mer pelage: Darker yellowish above, more buffy below. Cranial characters. — Skull rather slender and narrow, angular and well ridged in adults; nasals slender, ending even with arm of pre- inaxilke; bullae small and narrow; palate low; incisive foramina short, constricted posteriorly ; incisors projecting well in front of nasals ; molars very light ; m2 with 4 closed sections; m3 with 3 closed triangles, 3 outer and 3 inner salient angles and terminal loop; ml with 4 closed triangles, 3 outer and 5 inner salient angles, and fifth triangle open and confluent with short fig. 4.— Molar enamel pat- terminal loop, as in M. ratticeps of Europe. tern of Mierotu* "J'era- ™ . . riw*(X5). Measurements. — Type (immature, measured in dry skin): Total length, 110; tail vertebras, 28; hind foot, 18. Aver- age of 10 adult topotypes, 1G8; 40; 10.7. Skull (of type) : Basal length, 22.4; nasals, 5.8; zygomatic breadth, 12.4; mastoid breadth, 10.7; alve- olar length of upper molar series, 5.5. Skull (of adult $ , No. 9899) : 27; 8; 16; 12.3; 6.3. General remarks. — Mr. E.W. Nelson found these voles abundant along the coast tundras of Bering Sea from Cape Vancouver north to Bering 40 NORTH AMERICAN FAUNA. [no. 17. Strait, and on Nelson, St. Michael, and Stewart islands.1 Mr. W. H. Osgood found them extending southward into the timbered region as far as the point where the Yukon crosses the Alaska boundary. Specimens examined. — Total number, 81, from the following localities: Alaska: St. Michael, 65; Kowak River, 1; Kagiktowik, 1; Bristol Bay, 1; Fort Yukon, 8; Circle, 1; 40 miles above Circle, 2; International boundary on Yukon, 1; Yukon River (200 miles southwest of Fort Yukon), 1. MICROTUS MACFARLANI Merriam. Maclarlane Vole Microtus macfarJani Merriam, Proc. Wash. Acad. Sci., II, 24, March 14, 1000. Type locality. — Fort Anderson, Anderson Eiver, Northwest Terri- tory. Geographic distribution. — Tundra region of Arctic America, east of the Mackenzie Eiver. General characters. — Similar to operarius in external characters, but with shorter tail, shorter, wider skull and more projecting incisors. Fur very long and soft in winter specimens. Color. — Winter pelage: Upperparts light buffy gray; belly whitish; feet silvery gray; tail sharply bicolor, black above, white below. Sum- mer pelage: Darker and brighter buff or ochraceous. Young: More grayish. Cranial characters. — Compared with operarius: Skull short and wide; nasals shorter; incisors more projecting; bullae wider; incisive fora- mina shorter; molars slightly heavier; enamel pattern the same. With yakutatensis: Size smaller; coloration brighter; skull Hatter; nasals shorter; incisors more projecting; interparietal smaller. Measurements. — Type (in dry skin): Tail vertebra1, 29; hind foot, 18.5; topotype (No. 9144) : tail, 27 ; hind foot, 19. Skull (of type) : Basal length, 26; nasals, 7; zygomatic breadth, 15.5; mastoid breadth, 12.5; alveolar length of upper molar series, 6. General remarks. — Material is so scanty from the Arctic regions that little is known of the range of this form, whether it meets and grades into operarius or yakutatensis, or whether it has a restricted and isolated range. Specimens examined. — Total number, 18, from the following local- ities : Northwest Territory: Fort Anderson, 4; Mackenzie River, 11; 'Arctic Coast,' 3. MICROTUS YAKUTATENSIS Merriam. Yakutat Vole. Microtus yakutatensis Merriam, Proc. Wash. Acad. Sci., II, 22. March 14, 1900. Type locality. — Yakutat Bay (north shore), Alaska. Geographic distribution. — Mainland of Alaska from Glacier Bay to Prince William Sound. » Proc. Biol. Soc. Wash., VIII, 140, 1893. June, 1900.] MICROTUS KADIACENSIS. 41 General characters. — Size medium, about equal to operarius, less than that of any of the island forms of the group. Color dusky, as in sitkensis, but belly whitish. Skull of adult male heavily ridged; bullae medium, rounded; interparietal large, shield-shaped. Color. — Summer pelage: Upperparts dusky gray, with a trace of brownish, darker dorsally; belly washed with soiled white or pale buffy; tail sharply bicolor, sooty or black above, whitish below; feet silvery gray, soles black. Young: Quarter-grown specimen (June 19), darker gray than adult, with black nape, whitish belly, sharply bicolor, black and white tail. Cranial characters. — Skull heavy, ridged and angular in adult male; interparietal large, shield-shaped; nasals long, with median constric- tion; dentition heavy. From the skull of operarius it differs in greater width, larger bulla1, heavier dentition; from that of unalascensis in smaller size, larger interparietal, slenderer nasals, smaller bulhe, shorter pterygoids. Molar pattern as in operarius. Skulls of adult females conspicuously smoother and less ridged than in males. Measurements. — Average of 10 adults (5 males and 5 females) from type locality: Total length, 161; tail vertebrae, 37; hind foot, 20.G. Skull (of type, $ ad., jSTo. 98005) : Basal length, 28; nasals, 8; zygomatic breadth, 16; mastoid breadth, 13; alveolar length of upper molar series, 7. General remarks. — This mainland form is readily distinguished from any of the island species of the group by either cranial or external characters, although it shows closer relationship with some of them than with the neighboring mainland species, operarius. If it has an uninterrupted range to the north it may grade into operarius, but at present there is no intermediate material to show whether it does or not. Specimens examined. — Total number, 17, from the following localities: Alaska: Yukutat, 29; Glacier Bay, 17 ; Prince William Sound, 1. MICROTUS KADIACENSIS Merriam. Kadiak Vole. Microtus ladiaccHsis Merriam, Proc. Biol. Soc. Wash., XI, 222, July 15, 1897. Type locality. — Kadiak Island, Alaska. Geographic distribution. — Known only from the type locality. General characters. — Size about that of sitkensis; belly white; ears very small; bulla' small and narrow; basioccipital short and wide. Color. — Summer pelage: Yellowish brown above, with scattered black hairs; sides paler; belly washed with pure white; feet silvery gray, heels dusky, soles blackish; tail not sharply bicolor, black above, whitish below. Young (in June) : Dull buffy gray above, maltese below. Cranial characters. — Skull flat, long, and narrow; audital bulhe small and laterally compressed; basioccipital short and wide; palate low, with sloping median ridge, lateral pits deeper, and incisive foramina wider than in sitkensis; nasals short, not reaching posterior tips of pre- 42 NORTH AMERICAN FAUNA. [no. 17. maxillae; incisors projecting; molars small; m3 with 3 closed triangles and elliptical terminal loop, making 3 inner and 3 outer salient angles, as in operarius. Measurements. — Average of five adult males : Total length, 188; tail vertebra^, 50 ; hind foot, 2 1 . SMll (of topotype, $ ad. No. 97969,) : Basal length, 28; nasals, 8; zygomatic breadth, 16.5; mastoid breadth, 13.2; alveolar length of upper molar series, 6.4. Remarks. — This insular species belongs to the operarius group, but differs from operarius in larger size, smaller bulla1, and transversely longer and narrower interparietal. Specimens examined. — Total number, 12, from the type locality. MICROTUS UNALASCENSIS Merriam. Unalaska Vole. Microtus unalascensis Merriam, Proc. Biol. Soc. Wash., XI, 222, July 15, 1897. Type locality. — Unalaska, Alaska. Geographic distribution — Island of Unalaska. General characters. — Larger and more robust than operarius; belly white; feet light gray; skull well arched and heavily ridged; bulla' large and well rounded. Color. — Upperparts dull yellowish brown, darkest on head and rump; end of nose whitish; belly white or slightly soiled whitish; feet light gray with dusky soles; tail bicolor, a narrow line of blackish above, soiled white below. Young: Similar to adult. Cranial characters. — Skull considerably arched, deep, heavy, and angular; frontals heavily ridged in old age; bulla' medium, much larger than in operarius or hadiaccnsis; basioccipital narrowly constricted between bulla' ; dentition actually and relatively heavier than in opera- rius; molar pattern the same; m3 with 3 closed triangles and an inner salient angle confluent with rounded posterior loop; ml with 4 closed triangles, the fifth triangle confluent with shortened anterior loop. Measurements. — Type, $ im.: Total length, 12^; tail vertebra?, 28; hind foot, 19. Adult $ topotype, No. 97963: 181; 38; 22. Skull (of topotype): Basal length, 30; nasals, 8; zygomatic breadth, 17.7; mas- toid breadth, 14; alveolar length of upper molars, 7. Specimens examined. — Total number, 9, from the type locality. MICROTUS UNALASCENSIS POPOFENSIS Merriam. Popof Island Vole. Microtus unalascensis jmpofensis Merriam, Proc. Wash. Acad. Sci., II, 22, March 14, 1900. Type locality. — Popof Island, Shumagin group, Alaska. Geographic distribution. — Known only from Popof Island. General characters. — Similar to Jcadiacensis, but slightly larger, with larger feet, relatively shorter and sharply bicolor tail. Skull less ridged, with larger bulla' and heavier molars. In size and proportions more nearly agreeing with unalascensis, from which it differs in wholly dusky nose, less ridged skull, smaller audital bullae and deeper prezygomatic notch. ■Tune, 1900.] MIOROTUS SITKENSIS. 43 Color. — General coloration not readily distinguishable from that of kadiacensis; upperparts dark yellowish brown; nose dusky to tip; belly soiled white or pale buffy; tail sharply bicolor, whitish below, dusky or black above; feet silvery gray, with black soles and dusky heels. Cranial characters. — Skull rather long and narrow; frontals not ridged in adults; prezygoinatic notch deep; audital bullae medium, not narrowly constricted as in kadiacensis, nor large and rounded as in unalascensis; palate with posterior point projecting into pterygoid fossa; incisive foramina short and wide; molar pattern as in operarius except in m.3, which has normally 4 inner and 3 or 4 outer salient angles; ml has 4 closed and 1 open triangle as in operarius. Measurements. — Average of 3 topotypes: Total length, 165; tail ver- tebrae, 38; hind foot, 22.4. An adult 9 , No. 97959: 188; 43; 22. Skull (of type, No. 97950, $ ad.): Basal length, 29.4; nasals, 8f)-7 zygomatic breadth, 17.5; mastoid breadth, 13.5; alveolar length of upper molar series, 7.2. General remarks. — This species needs comparison only with unalas- censis and kadiacensis, from both of which it differs in slight external and well-marked cranial characters. The three are evidently from the same original stock that from long insular separation has beeu modified by somewhat varied conditions. Specimens examined. — Total number, 7, from the type locality. MICROTUS SITKENSIS Mcrriam. Sitka Vole. Microti!* sitkensis Mcrriam, Proc. Biol. Soc. Wash., XI, 221, July 15, 1897. Type locality. — Sitka, Alaska. Geographic distribution. — Known only from Baranof Island, Alaska. General characters. — Size medium, about that of unalascensis; color yellowish brown above and below; skull rather tiat, wide interorbit- ally; interparietal triangular; molars small; ml with 4 or 5 closed triangles. Color. — August pelage : Upperparts rusty brown, brightest on rump and nose, besprinkled with blackish hairs; sides paler; belly washed with dark buff; nose blackish; feet silvery plumbeous; heels and soles black; tail sharply bicolor, black above, pale buff below. Cranial characters. — Skull long and flat, with no trace of superciliary ridges, wide iuterorbitally; tip of nasals reaching back of premaxilla?; interparietal narrow, subtriaugular; bullae medium and globose; pal. ate long and flattened, lateral bridges low, lateral pits shallow; incisive foramina short and narrow; incisors projecting well beyond nasals; dentition slightly more intricate than in operarius; ml has 4 or 5 closed triangles and a rounded terminal loop with a sharp inner salient angle; m.3 has 3 closed triangles and 4 inner and 4 outer salient angles. 44 NORTH AMERICAN FAUNA. [no. 17. Measurements. — Type, & ad.: Total length, 155; tail vertebrae, 42; hind foot, 23. Adult 5 topotype: 190; 45; 22. Skull: Basal length, 25.5; nasals, 7; zygomatic breadth, 14.2; mastoid breadth, 11.3; alve- olar length of upper molar series, 6. Skull (of topotype, V ad.): 30; 8; 17.7; 14; 7. Remarks. — Microtus sitkensis belongs to a well-marked group of tlie subgenus Microtus with the molar pattern of M. ratticeps of Europe, although in sitkensis ml is usually closed up, making 5- triangles instead of 4. Specimens examined. — Total number, 2, from the type locality. MICROTUS INNUITUS Merriara. Innuit Vole. Microtus innuitus Merriam, Proc. Wash. Acad. Sci., II, 21, March 14, 1900. Type locality. — St. Lawrence Island, Bering Sea, Alaska. Geographic distribution. — Known only from St. Lawrence Island. - General characters. — Size large; tail of medium length, sharply bicolor; skull wide and low, with projecting incisors; dentition mainly as in the operarius group. Cranial characters. — Skull ridged and angular, not much arched; braincase short and wide; nasals short and cuueate, falling consider- ably back of base of incisors; interparietal sin all, semicircular; bulla? large, somewhat flattened and angular; pterygoids short; interptery- goid fossa very narrow; dentition heavy, incisors conspicuously pro- jecting; molars with sharply constricted enamel folds; ml with only 4 closed triangles, m3 with three closed triangles, a short posterior loop, and long posterior inner salient angle. Measurements. — Tail vertebrae, 44; hind foot, 23. Skull (of type) : Basal length, 32.5; nasals, 9; zygomatic breadth, 19.5; mastoid breadth, 15.3; alveolar length of upper molar series, 7.2. General remarks. — The specimens from St. Lawrence Island were taken from regurgitated pellets of owls and jaegers, and consist of skulls, feet, tails, and imperfect skeletons. The animals are abundant. Many were seen running in the grass by members of the Harrimau party who landed for a short time on the island. The species of Microtus coming geographically nearest to St. Lawrence Island is tshuktshorum Miller, from Plover Bay, on the Siberian coast, a tiny spe- cies bearing little resemblance to the present one. Specimens examined. — Ten more or less imperfect skulls, besides feet, tails, and parts of skeletons, from the type locality. MICROTUS ABBREVIATUS Miller. Hall Island Vole. iJicrvtus abhreviatus Miller, Proc. Biol. Soc. Wash., XIII, 13, Jan. 31, 1899. Type locality. — Hall Island, Bering Sea, Alaska. Geographic distribution. — Known only from Hall Island. General characters. — Size rather large; tail very short, and densely haired; ears concealed in long fur; feet large and stout, measuring about 23. Junk, 1000.] MICROTUS ABBREVIATUS FISHERI. 45 Color. — July pelage: Upperparts dark buff or yellowish brown, bright- est over ears, face, and ramp; belly creamy white or pale buff; tail sharply bicolor, a narrow line of dark brownish above, creamy below; feet soiled white. Young (half-grown specimens): Duller and darker. Cranial characters. — Skull similar to that of unalascensis in size and general appearance, but more heavily ridged, with deeper prezygomatic notches, larger, more quadrangular interparietal, shallower lateral pits of palate; slightly smaller and especially narrower audital bulke; rather lighter dentition, with different molar pattern, nil and m2 with base of posterior triangles broadly open; m3 shortened, with but 2 closed triangles, 3 outer, and 4 iuner salient - „ r\ f\ f\/\/\ t^(%/\ -\ augles, the third triangle opening into short pos- ■ j-V \ _ vA_ j fry terior loop; nil with 5 completely closed triangles, and a well developed anterior trefoil, 4 outer, and 5 inner sharp, salient angles back of terminal QwM^vw loop. From the St. Matthew Island subspecies . . Fig. 5.— Molar enamel pat- fisheri, it diners in molar pattern as trom unettas- tern of Microtia abbren- ccnsis, and also in slenderer skull and rostrum, atu*( 5>- slenderer zygomata, and wider interpterygoid fossa. Measurements. — Average of 4 adult topotypes: Total length, 100; tail vertebra', 25.7; hind foot, 23.3. Skull (of No. 97981, $ ad.): Basal length, 30.3; nasals, 8.8; zygomatic breadth, 18; mastoid breadth, 14; alveolar length of upper molar series, 7: General remarks. — Microtus abbreviates was described from an alco- holic specimen retaining none of the original colors. On the Harriman expedition 7 specimens Mere collected July 14 on Hall Island, and for the first time the natural appearance of the animal was made known. In external characters it strongly resembles a lemming on account of the short tail, long fur, and stout form, but the skull is that of a robust Microtus. It belongs to the subgenus Microtus, and in general char- acter comes nearest to the operarius group, from which it is excluded, however, by its unique molar pattern — m3 having but 2 closed tri- angles, while ml has 5. Specimens examined. — Total number, .4. Skull (No. 50907, from Oregon City, Oreg.): Basal length, 29.2; nasals, 8.4; zygo- matic breadth, 17.1; mastoid breadth, 13.5; alveolar length of upper molar series, 7.5. General remarks. — Microtus townsendi has no close affinities with any species except tetramerus of Vancouver Island. Except for the COllSpicilOUS hip glands, it COineS nearest Fie. 6 —Molar enamel pattern of ,■17. 7 ' -1 7 Microtus townsendi (■ 5). to the longicaudus and moraax group in com- bination of general characters. Almost no variation is shown through- out its rather limited range, and there are no characters by which to recognize occideutalis even as a subspecies. Specimens examined. — Total number, 74, from the following localities: Oregon: Oregon City, 1; Aumsville, 2; Yaquina Bay, 5; Newport, 6; Shel- burn, 1 (iin); Salem, 1. Washington: Teniuo, 2; Steilacoom, 3; Roy, 2; Kent, 1; Lake Washington, Seattle (south end), 3; Avon, 25; Mount Vernon, 4; Sank, 1. British Columbia: Port Moody, 16; Chilliwack, 1. MICROTUS TETRAMERUS (Rhoads). Vancouver Vole. Arr.icola ( Tetramerodon) tetramerus Rhoads, Proc. Acad. Nat. Sci. Phila., Oct., 1894, 283. Type locality. — Beacon Hill Park, Victoria, British Columbia. Geographic distribution. — Southern end of Vancouver Island. General characters. — Like townsendi, but slightly smaller, with slen- derer feet and tail and narrower, slenderer skull. Color. — Indistinguishable from that of toicnscndi, in either winter or summer pelage. Cranial characters. — Skull smaller, slenderer, and less arched than in townsendi, with superciliary ridges never quite meeting; nasals rela- tively shorter and more spreading anteriorly; incisors slenderer; molars smaller but with the same enamel pattern. Measurements. — Average of 0 adult males from Coldstream (near Victoria), Vancouver Island: Total length, 177; tail vertebra', 54.3; hind foot, 22. The largest of a series of 14: 190; 60; 23. Skull (No. 91901, £ ad.): Basal length, 27.3; nasals, 7.8; zygomatic breadth, 16; mastoid breadth, 12.4; alveolar length of upper molar series, 6.7. General remarks. — The marked difference in size, together with slight cranial characters, separates this insular form as an easily recognizable species. Specimens of townsendi from the nearest localities on the mainland, Port Moody, British Columbia, and Avon, Wash., show no tendency toward tetramerus. The small series of specimens includes both summer and winter pelage. Specimens examined. — Total number, 17, from near the type locality: Vancouver Island, B. C: Goldstreaiu, 16; near Victoria, 1. 48 NORTH AMERICAN FAUNA. [no. 17. MICROTUS LONGICAUDUS (Merriam). Long-tailed Vole. Arvicola (Mynomes) longicaudus Merriam, Am. Nat., XXII, 934-935, Oct., 1888. Type locality. — Custer, S. Dak. (in the Black Hills at an altitude of about 5,500 feet). Geographic distribution. — Boreal cap of the Black Hills and down some of the cold streams well into the Transition zone. Habitat. — Banks of cold streams and in mountain meadows. General characters. — Size of body about equal to tbat of Mierotus pennsylvanicus ; tail much longer; ears larger; colors grayer; skull flatter ; braincase wider. Color. — Summer pelage: Upperparts dull bister, darkened with nu- merous black-tipped hairs, becoming grayish on the sides and shading into dull, buffy gray on belly ; feet plumbeous; tail dimly bicolor, black- ish above, soiled whitish below. Winter pelage (old and faded in a June specimen from Sundance, Wyo.): Upperparts grayish bister, mixed with blackish-tipped hairs, shading gradually into slightly paler sides and dull whitish belly; tail distinctly bicolor; feet soiled whitish. Cranial characters. — Skull long and not much arched; rostrum long; nasals reaching to anterior plane of incisors; bulla1 large and rounded; molar pattern similar to that of pennsylvanicus, except for absence of posterior loop in middle upper molar; m3 with 3 closed triangles, 3 outer and 4 inner salient angles; ml with anterior loop, 5 closed tri- angles, 4 outer and 5 inner salient angles. From mordax it differs in slightly shorter, heavier rostrum and wider nasals; narrower interptery- goid fossa; wider expansion of jugal; shorter and wider condyloid ramus of mandible. Measurements. — Type, 9 ad.: Total length, 185; tail vertebra?., 05; hind foot, 21. Topotype, 9 ad.: 181; 61 j 22. Skull (of type): Basal length, 25; nasals, 7.8; zygomatic breadth, 15.2; mastoid breadth, 11.6; alveolar length of upper molar series, 6.3. General remarks. — Mierotus longicaudus stands as one of the few out- lying and isolated forms, though the first-described species of its widely distributed group. Its nearest neighbor is M. mordax of the Bighorn Mountains, Wyoming, between which range and the Black Hills neither species is known to occur. Specimens examined. — Total number, 6, from the following localities: South Dakota: Custer, 2. Wyoming: Sundance (in the western edge of the Black Hills), 4. MICROTUS MORDAX (Merriam). Cantankerous Vole. Arvicola (Mynomes) mordax Merriam, North American Fauna No. 5, 61, July 30, 1891. Mierotus vellerosus Allen,1 Bui. Am. Mus. Nat. Hist., XII, 7, March, 1899. (Liard River, Northwest Territory.) Mierotus eautus Allen,1 Bui. Am. Mus. Nat. Hist., XII, 7, March, 1899. (Hell Gate, Liard River, Northwest Territory.) 1 The types of Mierotus rcllerosus and H. eautus, kindly loaned me for comparison by Dr. Allen, agree in every character with specimens in corresponding pelage of M. mordax from its type locality. The type of vellerosus, collected May 4, shows the dark brownish-gray back of imperfect summer pelage, while the type of eautus, col- lected November 29, shows the light gray pelage of early winter. Junb, 1900.] MICROTUS MORDAX. 49 Type locality. — Sawtooth (or Alturas) Lake, east foot of the Sawtooth Mountains, Idaho. Geographic distribution. — Rocky Mountains and outlying ranges from latitude 60° to northern New Mexico, and south in the Cascades and Sierra Nevada as far as Kaweah and Kern rivers, California. In the Cascades mainly confined to the east slope, but extending west to the Siskiyous, in southern Oregon, and Salmon and Trinity mountains, in northern California. Found in most of the isolated ranges of east- ern Oregon and northern and central Nevada. Common in Canadian and Hudsonian zones. Habitat. — Marshes and wet woods, but more especially the banks of cold mountain streams, down which it often extends into the Transition zone. General characters. — Size medium; tail long; ears large; feet small; no conspicuous side or hip glands in males. Very similar to longicaudus. Color. — Summer pelage : Back grayish bister; sides olive gray; belly washed with whitish; nose dusky; feet plumbeous; tail dimly bicolor, dusky above, soiled whitish below. Winter pelage: Lighter colored than in summer; dorsal stripe of yellowish bister more sharply con- trasted with the deeper gray of sides and face; belly heavily washed with pure white; tail sharply bicolor; feet whitish. Young: Darker, less sharply marked than the adults; feet and tail dusky. Cranial characters. — Skull light and slender, similar to that of longi- caudus, but with slightly longer, slenderer rostrum and nasals; slen- derer zygomata, and longer condylar ramus of mandible; dentition essentially the same; m2 with 4 closed sections, the posterior open ; m3 with anterior crescent, 3 closed triangles, and pos- terior loop with 2 inner salient angles; ml with 5 closed triangles, 5 inner and 4 outer salient angles F'G' 7_„M"1,ar enamel Pat- ° ' & tern of Microtus mordax back of anterior loop; second and third lower (x5). molars each with 3 outer and 3 inner salient angles. Measurements. — Type: Total length, 200; tail vertebra?, 77; hind foot, 22. Average of five adults from type locality, 182; 00; 22. Skull (of type): Basal length, 20.5; nasals, 8.0; zygomatic breadth, 10.2; mas- toid breadth, 12.8; alveolar length of upper molar series, CO. General remarks. — The species has a wide and frequently interrupted range, but shows remarkably slight variation of characters. Even from the southern extremities of the Rocky Mountains and Sierra Nevada the variation is too slight for subspecific recognition. Speci- mens from isolated ranges in Nevada are practically typical. Specimens examined. — Total number, 708, from the following localities : Idaho: Sawtooth Lake, 35; Lemhi, 4; Lost River Mountains, 2; Sahnou River Mountains, 1; Three Creek, 1; Preuss Mountains, 1; Montpelier Creek, 3; Kingston, 1; Osborn, 1; Mullau, 10; Cceur d'Alcne, 7; Craig Mountains, 1; Seven Devils Mountains, 3; Priest Lake (east side), 2. Utah: Laketown, 6; Park City, 1; Barclay, 3; near Barclay, 4. 18392— No. 17 1 50 NORTH AMERICAN FAUNA. [no. 17. Colorado: Estes Park, 2; Ward, 4; Gold Hill, 2; Longs Peak, 10; Canyon City, 1 ; Lake City, 2 ; Silverton, 5 ; Fort Garland, 12. New Mexico: Chama, 1; Martinez, 1; Agna de Lobo, 1 (no skull). Wyoming: Bridger Pass, 14; Bighorn Mountains, 1; Lake Fork, near Bull Lake, 4; Clark Fork, mouth of Crandle Creek, 3; Tower Falls, Yellow- stone Park, 2. Montana: Red Lodge, 2; Pryor Mountains, 8; Fort Custer, 2; Big Snowy Mountains, 2; Jefferson River, 1; Blackfoot, 1 ; St. Marys Lake, 3 ; Java, 4; Summit, 5; Flathead Lake, 8; Tobacco Plains, 2; Horse Plains, 3; Upper Stillwater Lake, 6; Prospect Creek, 11 ; Thompson Falls, 2 ; Silver, 4. Nevada : Reese River, 18 ; Arc Dome, 15 ; Indian Creek, 1 ; Shoshone Mountains, north of Cloverdale, 5; Pine Forest Range, 3; Granite Creek, 8; Mountain City, 15; Bull Run Mountains, 1; White Mountains, 6; Ruby Mountains, 14; Monitor Mountains, 2. California : Near Mount Whitney, 31 ; Olancha Peak, 1 ; Mulkey Meadows (near Olancha Peak), 1 ; Soda Springs (on North Fork of Kern River), 1 ; Mineral King (on East Fork Kaweah River), 19; Upper San Joaquin River, 2; Lone Pine, 2; Bishop Creek, 2; Queen Station, 1; Sequoia National Park, 24; Yosemite Valley, 3; Pine City, 1; Mammoth, 7; White Mountains, 4; Dou- ner, 3; Sierra Valley, 3; Hope Valley, 2; Carberry Ranch, 10; Emerald Bay, 5 ; Goose Lake, 2 ; Warner Creek, 1 ; Lassen Creek, 2 ; Lassen Peak, 19 ; Etna, 1; Trinity Mountains, 4; Canyon Creek, 16; Plumas County (20 miles southwest of Quincy), 2; Mount Shasta, 59; Sisson, 15; Goose Nest Moun- tain, 1; Bear Creek (Shasta County), 1. Oregon: Siskiyou. 6; Fort Klamath, 18; Crater Lake, 5; Diamond Lake, 5; Sink Creek (east of Mount Thielson), 1 ; Upper Des Chutes River, Little Meadows (near head of Des Chutes River), 2; Farewell Bend (Des Chutes River, 30 miles southwest Prineville), 2; Swan Lake Valley, 2; head of Drews Creek, 1 ; Warner Mountains, 3; Steen Mountains, 6; Summit (east of Malheur Lake), 1; 10 miles north of Harney, 9; Maury Mountains, 5; Wallowa Mountains, 3; Lone Rock, 1. Washington: Cleveland, 2; Wenatchee, 5; head of Lake Chelan, 12 ; Easton, 2; Conconully, 3. British Columbia : Mount Richtor, 2 ; Nelson, 1 ; Sicamous, 5 ; Hope, 2 ; Shus- wap, 1 ; Glacier, 3; Okanagan, 2; Bennett City, 6. Alberta: Henry House, 19 ; 15 miles south of Henry House, 4; south of Smoky River, 3; Banff, 1. Northwest Territory : Liard River, 2 ; Rink Rapids (upper Yukon), 1; Lake Lebarge, 2; Lake Marsh, 1. Alaska : Charlie Creek (Upper Yukon), 4 ; Circle, 1 ; White Pass, 5 ; Skagway, 1. MICROTUS MACRURUS Merriam. Olympic Vole. Microtus macrurus Merriam, Proc. Acad. Nat. Sci. Phila., Aug., 1898, 353. Type locality. — Lake Cushtnan, Olympic Mountains, Washington. Geographic distribution. — (The typical form) Olympic Mountains. (With slight variation) along the coast strip of British Columbia ami Alaska north to Yakutat. Habitat. — Marshes and borders of cold streams. General characters. — Like mordax, but considerably larger, with con spicuously larger hind foot and darker coloration. Color. — Summer pelage: Upperparts dark bister, shaded with numer- ous black hairs, becoming sooty gray in some specimens; sides slightly paler; belly washed with dull buffy or whitish; feet plumbeous; tail distinctly bicolor, blackish or brownish above, soiled whitish below, June, 1900.] MICROTUS ANGUSTICEPS. 51 usually white-tipped. ( Winter pelage unknown.) Young: Darker than adult, with blackish feet and tail. Cranial characters. — Skull averaging much larger than mordax, with wider interorbital region, heavier rostrum, smaller audital bullae, aud heavier dentition; molars especially wider; lower jaw conspicuously more massive, with wide, heavy molars. Measurements. — Type: Total length, 220; tail vertebrae, 88; hind foot, 24. Average of five specimens from three localities in the Olympic Mountains: 204; 80; 24.3. Skull (of type, No. GG151, $ ad.): Basal length, 27.7; nasals, 8.2; zygomatic breadth, 16; mastoid breadth, 12.5; alveolar length of upper molar series, 7. General remarks. — Microtis macrurus is the most conspicuously marked and easily recognizable form of the longicaudus group, though with less deeply seated characters than some forms that arc externally scarcely distinguishable from each other. In the Olympic Mountains its range is completely isolated, being separated from that of mordax by the intervening low country, the habitat of the larger toicnsendi, and by the high Cascades, in which neither form occurs. To the northward it again occurs, in nearly typical form, on the coast at Lund, British Columbia (lat. 50°), and extends northward along the coast to Yakntat, Alaska, becoming slightly smaller and less markedly different from true mordax. Specimens examined. — Total number, 84, from the following localities: Washington: Lake Cushman, 7; head of Skokoiuish River, I; head of Soleduc River, 1; Quineault Lake, 4; Granville, 1. British Columbia : Lund, on Malaspina Inlet, 3; River Inlet (head), 14; Fort Simpson, 6. Alaska: Loring, 3; Wrangell, 4; Juneau, 12; Yakutat, 7; Yakutat Bay (north shore), 10; Glacier Bay, 11. MICROTUS ANGUSTICEPS Bailey. Coast Vole. Microtus angusiiceps Bailey, Proc. Biol. Soc. Wash., XII, 86, April 30, 1899. Type locality. — Crescent City, California. Geographic distribution. — Coast region of northwestern California and southwestern Oregon. Habitat. — Damp pastures in the Sitka spruce belt. General characters. — Smaller aud darker colored than typical mordax, with very narrow, slender skull and small audital bullae. Color. — Summer pelage: Upperparts dark bister, lined with black hairs, darkest on face and nose; sides paler; belly washed with creamy white; feet plumbeous gray; tail distinctly bicolor, blackish above, soiled white below. Cranial characters. — Skull small and very narrow, distinctly ridged in adidts; nasals projecting in front of incisors; incisive foramina short; audital bullae very small and constricted; coronoid notch of mandible narrow; incisors slender; molars small, with narrow, sharp angles; enamel pattern as in mordax. 52 NORTH AMERICAN FAUNA. [no. 17. Measurements. — Type, S ad.: Total length, 170; tail vertebra1, 56; hind foot, 22. An adult 9 topotype: 170; 55; 22. Skull (of type): Basal length, 23.4; nasals, 7.6; zygomatic breadth, 13.5; mastoid breadth, 10.8; alveolar length of upper molar series, 6. General remarks. — Externally this species is not very different from true mordax, but the skull shows such marked characters as to warrant full specific rank. Specimens examined. — Total number, 45, from the following localities: California: Crescent City, 31; Areata (Humboldt Bay), 13. Oregon: Gold Beach, 1. MICROTUS ALTICOLUS (Merriam). Mountain Vole. Arvicola (Hynomes) alticolus Merriam, North American Fauna No. 3, 67-69, PI. V, figs. 1 and 2; PI. VI, figs. 1, 2, 3, and 4; Sept. 11, 1890. Type locality. — San Francisco Mountain, Arizona (Little Spring, on northwest side of mountain, altitude S,200 feet). Geographic range. — Boreal zone of San Francisco Mountain, from 8,200 feet altitude up to timberline at 11,000 feet. Habitat. — Vicinity of springs and cold streams on the slopes of the mountain. General characters. — Similar to longicaudus, but tail shorter, hind foot and ear smaller, and skull with truncate, instead of pointed, anterior end of frontal and deeper lateral pit of palate. Color. — Summer pelage: Upperparts uniform sepia or dull bister, darkened with blackish-tipped hairs; sides scarcely lighter; belly pale huffy or whitish ; feet dull grayish or dirty whitish; tail not sharply bicolor, blackish above, grayish below. Young: Similar to adults, but with woolly fur aud long, scattered, black-tipped hairs. ( Winter pelage unknown.) Cranial characters. — Skull similar to that of mordax, but readily dis- tinguished from it and those of all other forms of the group by truncate end of anterior arm of frontal. Other characters are, deeper lateral pits of the palate; wider interpterygoid fossa; slightly longer, more open, incisive foramina. Dentition similar to that of longicaudus and mordax. Measurements. — Type, 9 ad.: Total length, 170; tail vertebra1, 56; hind foot, 20. Average of 5 adults from type locality: 178; 56; 20. Skull: Basal length, 25; nasals, 7.5; zygomatic breadth, 14.8; mastoid breadth, 12.3; alveolar length of upper molar series, 6.5. General remarks. — Microtus alticolus, with its subspecies leucophceus, is the most isolated form in the longicaudus group. Its geographieally nearest neighbor and probably nearest relative is mordax, in the moun- tains of Colorado aud northern New Mexico. Specimens examined. — Total number, 13, from the type locality. One immature specimen from Springerville, in the White Mountains, may be either alticolus or leucophwus. June, 1900.] MICROTUS MEXICANUS. 53 MICROTUS ALTICOLUS LEUCOPH.EUS (Allen). Graham Mountain Vole. Jrricola leucophceus Allen, Bui. Am. Mus. Nat. Hist., VI, 320-321, Nov. 7, 1894. Type locality. — Graham Mountains, Arizona. Geographic distribution. — Known only from the type locality. General characters. — Similar to alticolus, and of the same proportions, hut slightly larger, color the same, skull wider interorbitally and with other slight differences. Color. — Summer pelage : Upperparts sepia or dull bister, but little paler on sides and faintly lined with blackish hairs; belly washed with soiled whitish; feet dull grayish; tail distinctly bicolur, brown above, grayish below. (Young and winter pelages not represented.) Cranial characters. — Skull wider interorbitally than in mordax or alti- colus; anterior arm of frontal with triangular instead of truncate point; incisive foramina wider than in alticolus, slightly constricted posteriorly; lateral pits of palate wide and shallow; coronoid notch narrow and sharp. Dentition as in alticolus and mordax. Measurements. — Type, 9 ad.: Total length, 173; tail vertebrae, 50; hind foot, 22.5. Topotype, 9 ad.: 183; 50;" 23. Skull (of type) : Basal length, 26.5; nasals, 8.3; zygomatic breadth, 15.2; mastoid breadth, 12.3; interorbital width, 4.2; alveolar length of upper molar series, 6.3. General remarks. — Microtus leucophceus belongs to the longicaudus group. It is closely related to alticolus, from which size and slight cranial characters separate it as a fairly well-marked subspecies. Through the kindness of Dr. J. A. Allen, of the American Museum of Natural History, I have the type and a topotype of M. leucophceus for comparison with the Biological Survey series of alticolus, mordax, and longicaudus. Specimens examined. — Total number, 2, from the type locality. MICROTUS MEXICANUS (De Saussure). Mexican Vole. Arvicola (Hemiotoviys) mexicauus De Saussure, Revue et Mag. tie Zool., 2e s6r., XIII, 3, Jan., 1861. Type locality. — Mount Orizaba, Puebla, Mexico. Geographic distribution.— Eastern Puebla and to the north and west, grading into its subspecies phceus. Habitat. — Grassy places in open forests, in upper Austral and Tran- sition zones. General characters. — Size rather small; tail short; ears conspicuous; pelage coarse and lax; colors brownish; skull wide, with short, wide incisive foramina; ml normally with 6 inner salient angles. Color. — Winter pelage: Upperparts grizzled brown, from a mixture of dull cinnamon and black; sides paler; belly washed with cinnamon or huffy, or rarely with whitish ; sides of nose and ear coverts usually a brighter shade of cinnamon; feet clear gray; tail dusky above, gray below. Summer pelage (imperfectly represented): Evidently darker and less ferruginous. Young: Duller and darker than adult. 54 NORTH AMERICAN FAUNA. [no. 17. Cranial characters. — Skull rather angular, with wide- spreading zygo- matic arches, narrow interorbital constriction, and large, well-rounded audital bullae; incisive foramina short and wide, truncate posteriorly; zygomata broadly flattened; palate with a median groove between the lateral pits where a spur or ridge appears in most species of Microtus; upper incisors abruptly decurved; molar pattern differing from that of nanus and montanus — mainly in extra angle of anterior trefoil of ml; most of the salient angles acute; m2 has 4 closed sections; m3 has 3 closed triangles, 3 outer and 4 inner salient angles; ml has 5 closed triangles, 5 outer and G inner salient angles.1 Measurements. — Average of 10 adults from the type locality (5559): Total length, 138; tail vertebrae, 29; hind foot, 19.35; maximum: 148; 30; 20. SJcull (of topotype, No. 51496, $ ad.): Basal length, 21.5; nasals, 7.4; zygomatic breadth, 15.3; mastoid breadth, 11.6; alveolar length of upper molar series, 6.6. General remarks. — Microtus mexicanus, phceus, fulviventer, and mogol- lonensis form a well-marked and closely united group of small, short- tailed, brownish voles, distinguished by the arrangement of mamma? in two pairs, a pair of inguinal, and a pair of pectoral ; by wide-spreading zygomatic arches and narrow interorbital constriction; by wide inci- sive foramina and grooved posterior ridge of palate; and by similar habits and habitat. They need comparison only among themselves. While mogollonensis and fulviv enter are well-marked forms, occupying widely separated and probably disconnected areas, mexicanus and phceus merely show the extremes of differentiation found in one wide- ranging and somewhat variable form. As only those from the type localities are really typical, any line separating mexicanus and phceus is purely arbitrary. Specimens examined. — Total number, 194, from the following localities in Mexico : Puebla : Mount Orizaba, 27 ; Chalchicomula, 22. Vera Cruz : Cofre de Perote, 29; Las Vegas, 11. Tlaxcala: Mount Malincbe, 1 ; Huamantla, 2. Hidalgo: Sierra de Pachucii, 7; Tulaucingo, 7; Real del Monte, 10. Morelos : Huitzilac, 4. • Mexico: Ajusco, 6; Toluca Valley, 20; North slope Volcan de Toluca, 9; Mount Popocatepetl; 19 ; Ainecameca, 1 ; Salazar, 19. MICROTUS MEXICANUS PH^US (Merriam). Colima Vole. Microtus phceus Merriam, Proc. Biol. Soc. Wash., VII, 171-172, Sept. 29. 1892. Type locality. — North slope of Sierra Nevada de Colima, Jalisco, Mexico (altitude 10,000 feet). Geographic distribution. — Southern Jalisco and northward to north- western Chihuahua (to the eastward, grading into mexicamis), occu- pying Boreal and Transition zones. 'This extra number of angles is a weak character depending on tbo slightly un- usual development of tbo anterior trefoil of ml, so tbat a pair of short points or angles may in most cases be counted on its inner and outer sides. June, 1900.] MICROTUS FULVIVENTER. 55 Habitat. — Grassy parks in open timber. General characters. — Similar to mexicanus, but slightly larger and a shade darker, and with slight cranial differences. Color. — Winter pelage: Upperparts uniform dark cinnamon brown mixed with blackish; belly a lighter shade of cinnamon or buffy, or sometimes whitish; feet brownish gray; tail brownish gray, paler below. Summer pelage: Not shown in specimens from near the type locality, but June specimens from El Salto, Durango, are brighter and darker ferruginous than topotypes in winter pelage. Young: Dull brownish. Cranial characters. — Skull similar to that of mexicanus, but with less constricted interorbital region, slightly shorter incisive foramina, and shallower prezygomatic notches. Dentition essentially the same. Measurements. — Average of 10 adult topo- types (5 $ and 5 9): Total length, 151; tail vertebra?, 35; hind foot, 20.5. Skull (topo- type, Sad., No. 45(345): Basal length, 25.2; nasals, 7.3; zygomatic breadth, 15.5; mastoid breadth, 12; alveolar length of upper molar Series 6 08 ^la' 8-— Molar enamel pattern of ,Y ' , , __.. , . , Microtus 2>hcnus (X 5). General remarks. — Microtus m. plmus is not a strongly or sharply marked form of mexicanus, although apparently the more widespread form. Ajusco and Salazar specimens might as well be referred to it as to mexicanus. A large series from near Guad- alupe, in southwestern Chihuahua, are indistinguishable from typical pJuvus, and those from Miquihuana, western Tamaulipas, are nearer to plueus than to mexicanus. Specimens examined. — Total number, 13G, from the following localities in Mexico: Jalisco: Sierra Nevada de Colinia, 17. Michoacan: Nabuatzin, 23. Queretaro: Pinal de Amoles, 9. Tamaulipas: Miquihuana, 20. Durango : El Salto, 25. Chihuahua: Sierra Madre uear Guadalupe y Calvo, 19; Colouia Garcia, 23. MICROTUS FULVIVENTER Merriam. Oaxaca Vole. Microiitsfidriventer Merriam, Proc. Biol. Soc. Wash., XII, 106, April 30, 1898. Type locality. — Cerro San Felipe, Oaxaca, Mexico. Geographic distribution. — Central part of the State of Oaxaca. Habitat. — Open grassy places and along edges of fields in the Boreal zone. General characters. — Slightly larger than mexicanus and of nearly the same proportions; darker and. richer in coloration; ears conspicuous above fur; tail a little more than one and a half times the length of hind foot. Color. — Summer pelage: Upperparts umber brown, darkened by black hairs; under parts fulvous or dull chestnut brown ; feet grayish brown ; 56 NORTH AMERICAN FAUNA. [no. 17. tail dusky brown above, fulvous below, darker toward the end. Winter pelage (in October and March specimens) : Less deeply colored. Young: Dull sooty, with scarcely a trace of fulvous. Cranial characters. — Skull similar to that of mexicanus, but with smaller bulla?, longer incisive foramina; sharper posterior point of frontals; molars slightly heavier, enamel pattern almost the same; ml has a more rounded anterior loop. Measurements. — Average of 10 adult topotypes (5 $ and 5 9): Total length, 152 ; tail vertebra?, 35 ; hind foot, 20.5. Type ( $ ad.) : 151 ; 38 ; 20. Skull (of type) : Basal length, 25.4; nasals, 7.4; zygomatic breadth, 15.5; mastoid breadth, 12.4; alveolar length of upper molar series, 6.5. General remarks. — M. fulviv enter belongs to the mexicanus group, but is sharply separated in its distinguishing characters as well as in geo. graphic range. But little variation is shown throughout its range and although specimens from the mountains near Ozolotepec show differences, these are slight and unimportant. Specimens examined. — Total number, 120, from the following localities in Mexico : Oaxaca: Cerro San Felipe, 32; Reyes, 23; 15 miles west of Oaxaca, 20; Mount Zempoaltepec, 28 ; nearCajones, 5; Guajainaloya, 1 (im.); mountains near Ozolotepec, 9; Totontepec, 8. MICROTUS MOGOLLONENS1S (Mearns). Mogollon Mountain Vole. Arvicola mogollonensis Mearns, Bui. Am. Mus. Nat. Hist., II, No. 4, 283-284, Feb., 1890. Type locality. — Baker Buttes, Mogollon Mountains, Arizona. Geographic distribution. — Plateau country of central Arizona. Habitat. — Dry grassy parks among the yellow pines of the Transition zone. General characters. — Size small; tail and feet short; color dull rusty brown; fur long and soft; ears not concealed; skull short, wide, and angular; lateral pits of palate very deep; an inner projecting point at base of posterior triangle of middle upper molar. Color. — Upperparts dull rusty brown, brightest on tips of ears; sides slightly paler; belly cinnamon or buffy gray; feet grayish brown; tail brownish gray above, grayish below. Cranial characters. — Skull short and well arched, with wide-spreading zygomata and sharply constricted interorbital region ; zygomatic shield broad and flat; interparietal small and narrow; nasals notched pos- teriorly, falling considerably short of terminus of premaxillse; bulla1 full and rounded; incisive foramina short, wide, and open; incisors wider than m nanus and bent more abruptly downward; molar pattern as in mexicanus, except in m2, which has an inner point at base of pos- terior triangle, and in ml, which has 5 closed triangles and only 5 inner and 4 outer salient angles and an abbreviated terminal loop. June, 1900.] M1CROTUS XANTHOGNATHUS. 57 Measurements. — Average of 10 adults from San Francisco Mountain, Arizona: Total length, 131; tail vertebrae, 28.5; hind foot, 18. Skull (adult $ , No. 21503): Basal length, 23.6; nasals, 7; zyg matic breadth, 15.2; mastoid breadth, 12; alveolar length of upper molar series, 0.3. Remarks. — M. mogollonensis is widely separated, both geographically and specifically, from the other members of its group. Its nearest ally is phceus from Mexico. Specimens examined. — Total number, 51 ; from the following localities : Arizona: San Francisco Mountain (Little Spring on north side of mountain), 15; Springerville, 35. New Mexico: Fort Wingate, 1. MICROTUS XANTHOGNATHUS (Leach). Yellow-cheeked Vole. Arricola xanthognalha Leach, Zool. Miscel., 1,60, 1814. Type locality. — Hudson Bay. Geographic distribution. — Northwestern Canada and Alaska, from central Alberta north to the Arctic coast and west to central Alaska. General characters. — Size large, almost equaling that of richardsoni, but tail shorter and ears larger; colors dull ; nose and ear patch yellow ; skull heavy, ridged, and angular. Side glands as in richardsoni, or a little farther back on flanks. Color (March and May specimens). — Upperparts dark sepia to bister, heavily lined with coarse black hairs over the back; sides of nose and ear patch bright rusty yellowish, a tinge of the same around eyes and on cheeks; belly dusky gray; breast sooty; tail indistinctly bicolor, blackish above, dusky gray below ; feet sooty. Cranial characters. — Skull smaller than that of richardsoni and relatively longer and narrower, with less projecting incisors; nasals long and narrow; bulla1 large; incisive foramina long and narrow; den- tition heavy; molar pattern scarcely dif- ferent from that of townsendi; anterior loop of m.l small and triangular; middle sec- tion of m.3 frequently divided into two nearly closed triangles. Measurements. — Two dry skins from Fort Resolution, Great Slave Lake, Canada, adult females, in U. S. Nat. Mus. No. 4501 : F,G;f 9-Molar enamel patten, of Total length, 210; tad vertebra',, 50; hind foot, 27. No. 4502: 218; 45; 25. Skull (No. 4504): Basal length (approximately), 34.5; nasals, 10.3; zygomatic breadth, 20; mastoid breadth, 15.7; alveolar length of upper molar series, 8.5. Remarks. — Microtias xanthognathus shows no close relationship to any other American species. In the position of side glands it resembles richardsoni, but in no other characters. It shows a strong superficial resemblance to chrotorrhinus in color, but the great difference in size prevents the possibility of one ever being mistaken for the other. 58 NORTH AMERICAN FAUNA. [No. 17. Specimens examined. — Total number, 44, from the following localities: Canada: Nelson River, N. W. T., 1; Cache Apocotte (40 miles east of Henry House, Alberta), 1; Fort Resolution, 22; Great Slave Lake, 1 ; FortRae, 1; Liard River, 1; La Pierre House, 1; Anderson River, 1; Arctic coast (east of Fort Anderson), 2. Alaska: Mouth of Porcupine River, 1; Yukon (200 miles southwest of mouth of Porcupine), 3; Charlie Creek (Upper Yukon), 9. M1CROTUS CHROTORRHINUS (Miller). Rock Vole. ArvicoU clirotorrhinns Miller, Proc. Boston Soc. Nat. Hist., XXVI, 189-193, pi. 3, 1894. Type locality. — Mount Washington, New Hampshire, head of Tuck- erman Kavine (altitude, 5,300 feet). Geographic distribution. — Mount Washington, the Catskills, central Quebec, and northern New Brunswick, in the Hudsonian zone. Habitat. — Rocky places near water on the mountains, and in deep spruce forests farther north. General characters. — Size and proportions of pennsylvanicus except slightly smaller hind foot; ears larger; fur lax; conspicuously yellow- ish about nose, ears and rump; skull comparatively thin- walled and smooth; dentition unique. - Color. — Summer pelage (July 14): Upperparts bright glossy bister, lined with black hairs; nose to eyes dull orange rufus; hairs around ears and on rump yellowish; belly plumbeous; feet dark gray; tail grayish brown, slightly paler below. Worn, left-over winter pelage: Darker and more rusty above. Cranial characters. — Skull light and smooth, somewhat flattened on top, superficially reseinblingthat of Evotomys; bulke large and rounded ; incisive foramina short and wide; dentition rather light; incisors bent down at right angles to axis of skull, extending scarcely beyond tip of nasals; m2 with 4 closed sections; ni3 normally with 5 closed triangles, 5 inner and 5 outer salient angles and a double-lobed posterior loop; m3 with outer salient angles prominent and reentrant angles deep. Measurements. — Type: Total length, 165; tail vertebrae, 45; hind foot, 19.4. Average of 4 adult topotypes : 170 ; 48 ; 19.6. Skull (of ad. 6 , No. 2523, Bangs Coll.): Basal length, 25.4; nasals, 7.2; zygomatic breadth, 15; mastoid breadth, 12; alveolar length of upper molar series, 6.4. Remarks. — Microtus chrotorrhinus shows a marked superficial resem- blance to M. xanthognathus, but in cranial characters it differs widely from this and all other American species and is quite unique in the subgenus Microtus. In the specimens examined there is no trace of hip glands, but in two adult males (3845 and 3849) there appear to be rudiments of side glands on the flanks. Specimens examined.1 — Total number, 8, from the following localities :- New Hampshire: Mount Washington, 3. New York: Catskill Mountains, 1. Quebec: Lake Edward, 4. 1 Type in collection of G. S. Miller, jr. ; other specimens in the collection of E. A. and O. Bangs. 2Mr. Miller records a specimen in the Am. Mus. Nat. Hist, from Trousers Lake, New Brunswick— Proc. Bost. Soc. Nat. Hist., XXVI, 193, 1894. June, 1900.] SUBGENUS ARVICOLA. 59 MICROTUS CHROTORRHINUS RAVUS Bangs. Gray Rock Vole. Microtns ehrotorrhinus ravus Bangs, Proc. Biol. Soc. Wash., XII, 187, Nov. 16, 1898. Type locality. — Black Bay (north shore of Strait of Belle Tsle), Labrador. Geographic distribution. — Known only from the type locality. General characters. — Similar to ehrotorrhinus, but slightly grayer, and with noticeably more yellowish on nose and face. Skull slenderer, with lighter dentition. Color. — Summer pelage (July specimens): Upperparts grayish bister, becoming yellowish on rump; whole face from behind ears suffused with yellowish, brighter on nose; belly thinly washed with white over the plumbeous; feet buffy gray; tail brownish above, slightly paler below. Cranial characters. — Skull, compared with that of ehrotorrhinus, slightly thinner, lighter, and slenderer throughout; interorbital con- striction narrower; rostrum longer and narrower; incisive foramina longer; molar series shorter and narrower; tooth pattern as in ehrotor- rhinus. Measurements. — Type,1 )' angles ; m2 with the anterior pair of triangles confluent; m3 with 4 closed sections including 2 median triangles. .Tune, iooo.] SUBGENUS HERPETOMYS. 77 MICROTUS UMBROSUS iMerriam. Zempoaltepec Vole. Microtus umbrosns Merriam, Proc. Biol. Soc. Wash., XII, 108, Apr. 30, 1898. Type locality. — Mount Zempoaltepec, Oaxaca, Mexico (altitude 8,200 feet). Geographic distribution.— Known only from the east slope of Mount Zempoaltepec, in the humid Upper Austral zone. Habitat. — Dense oak forests, living in burrows and long underground tunnels. General characters. — Size rather large: tail long; ears large; fur long and lax; colors dark; skull long and flat, with small bullae and peculiar dentition. Color. — Upperparts uniform dusky, with brown-tipped hairs; below dark plumbeous thinly washed with fulvous; feet and tail thinly haired, concolor, dark brown. Cranial characters. — Skull long, narrow, and but little arched, with smooth outlines, and slender zygomatic arches; bullae very small; palate low, with slender or incomplete lateral bridges, shallow lateral pits and grooved posterior ridge; interpterygoid fossa wide and quad- rate; incisive foramina short and widest in the middle. Dentition heavy; incisors abruptly decurved; inner salient 'angles' of upper and posterior lower molars rounded instead of acute; ra3 with a small outer and a large inner closed triangle and a posterior trefoil with large inner and small outer lobe; m3 with 2 median closed triangles, an outer and inner, and broad terminal loops. Measurements. — Average of 7 specimens from type locality: Total length, 184; tail vertebra, 65; hind foot, 23. Type: 177; 61; 23.5. Skull (of type): Basal length, 26.5; nasals, 7.3; zygomatic breadth, 16; mastoid breadth, 12; alveolar leugth of upper molar series, 7. Specimens examined: Total number, 15; from the following localities in Mexico. Oaxaca: Mount Zempoaltepec (above Totontepec), 8; Totontepec, 7. Subgenus HERPETOMYS Merriam. Herpetomys Merriam, Proc. Biol. Soc. Wash., XII, 107, April 30, 1898. Type, Micro- tus guatemalensis Merriam, from Todos Santos, Guatemala. Geographic distribution. — That of the type species. Subgencric characters. — Plantar tubercles, 5; side glands' on thinks of males small and obscure or sometimes wanting; mammae, 6, pecto- ral, 2-2, inguinal, 1-1 (the latter apparently rudimentary and func- tionless); ears large; pelage long and soft; colors dark brownish. Skull with smooth outlines and large globose audital bullae; m3 with 3 closed triangles; mT with 3 closed triangles and an interior confluent 1 In some specimens no side glands can be discovered, and in others they are marked by a pencil of white hairs. There is some doubt as to whether the white hairs are a product of the glands or occur there accidentally or from injury, as they sometimes do over other parts of the body. 78 NORTH AMERICAN FAUNA. [no. 17. pair of triangles opening into terminal loop, and with 5 inner and 4 outer salient angles; ni3 with 4 closed sections including a pair of subequal median triangles. MICROTUS GUATEMALENSIS Merriani. Guatemalan Vole. Microtus guatemalensix Merriam, Proc. Biol. Soc. Wash., XII, 107, April 30, 1898. Type locality. — Todos Santos, Huehuetenango, Guatemala (altitude 10,000 feet). Geographic distribution. — Known only from type locality. Habitat. — Open ground on damp north slopes under rank growth of brush and weeds in the Boreal zone (altitude 9,800-11,000 feet). General characters. — Size medium; tail less than twice the length of hind foot; ears large but nearly concealed in the long fur; colors dark. Color. — Winter pelage: Upperparts dark umber brown; nose blackish; lips white; belly clear plumbeous or lightly washed with dull ochraceous; feet and tail dusky brown, con-_ fig. i6.-Moiar enamel pattern of color, and thinly haired. ( Summer pelage not Microtus (Herpetomys) guatemai- seen.) Young (half grown individuals in De- cember): Slightly duller than adult. Cranial characters. — Skull rather long and but little arched, without conspicuous ridges and angles; interorbital space wide; braincaselong; bulhc large and globose (larger than those of any other Mexican species) ; palate with steep and lightly grooved posterior median ridge; incisive foramina wide and short. Dentition heavy; most of the salient angles of molars acute; prisms deltoid; m3 with anterior crescent, 3 closed triangles, and a posterior crescent with two inner horns; m3 with pos- terior and anterior transverse crescents and a pair of subequal median triangles. Measurements. — Average of 20 specimens from the type locality : Total length, 150; tail vertebne, 37; hind foot, 21. Type ( 8 ad.): 155; 40; 21. Skull (of type) : Basal length, 25.6 ; nasals, 7.G ; zygomatic breadth, 15; mastoid breadth, 12.3; alveolar length of upper molar series, 7. General remarks. — So far as at present known, this is the southern- most species of Microtus in America. Its nearest relatives are umbrosus and mcxicanus, with both of which it has some characters in common, but from which it differs so widely as to require subgeneric separation. Specimens examined: Total number, 34, from the type locality. Subgenus NEOFIBER True. Neofiber True, Science, IV, 34, July 11, 1884 (genus). Type, Neofiber alleni True Neofiber Merriam, North American Fauna No. 5, 59, July, 1891 (subgenus). Geographic distribution. — That of the type species. Jun», 1900.] MIC ROT US ALLENI. 79 Subgeneric characters. — Plantar tubercles, .5; side glands conspicuous in both sexes and in young, situated half-way between hips and shoulders, the glandular area marked by brownish base of fur and half-en- circled above by a semilunar area of fur with white base; mamma', 0, ingui- nal, 2-2, pectoral, 1-1: feet and far modified for aquatic life; soles naked; a dorsal keel of long hair on rump. Skull maSSive ; palate long With illCOm- FlG- ".-Molar enamel pattern of Microtus , , , , , , . , . , (Xeofiber) alleni (X 5). plete lateral bridges; pterygoids wing- like; m3 with 2 closed triangles; ml with 5 closed triangles; ra3 with 2 median triangles and 2 transverse terminal loops. MICROTUS ALLENI (True). Florida Water-Eat. Neofiber alleni True, Science, IV, 34, July 11, 1884. Microtus (Neofiber) alleni Miller, North American Fauna No. 12, 70, .July 23, 1896. Type locality. — Georgiana, Brevard County, Florida. Geographic distribution. — Eastern and central Florida. Habitat. — Marshes, shallow lakes, and banks of streams. General characters. — In appearance very similar to a small muskrat, but with a round tail, a tuft of long hair above the tail, hind feet less modified for aquatic life; fur dense, with color and texture of muskrat fur; skull resembliug that of the muskrat, but with the rootless molars of Microtus. Color. — Upperparts dark brown, darkened on head and along back by coarse blackish hairs; nose black; chin dusky; belly pale buff or soiled silvery whitish; tail dark brown or blackish, darker toward the tip; feet dark brown. Young: Dark inaltese, with sooty backs. Cranial characters. — Skull high and short, with heavy ridges aud sharp angles; prezygomatic notches deep; postorbital shelf projecting; palate bone longer than in any other Microtus, shorter than in Fiber; lateral bridges of palate interrupted ; pterygoids wing-like (as in Fiber); dentition heavy; upper incisors bent abruptly downward. Measurements. — Average of 3 adult specimens from Canaveral, Fla.: Total length, 320; tail vertebra), 121; hind foot, 11. Largest adult, i : 330; 130; 41. Skull (No. 23150, 9 ad.): Basal length, 11.6; nasals, 12.5; zygomatic breadth, 26; mastoid breadth, 20.5; alveolar length of upper molar series, 12. General remarks. — The striking resemblance between M. alleni of Florida and M. amphibius of England proves on comparison of cranial characters to be only superficial ; the differences are subgeneric. Specimens examined: Total number, 17, from the following localities: Florida: Georgiana, 3; Titirsville, 1 ; Eden, 3; Canaveral, 5; Geneva, 3; Lake Harney, 1; Oaklodge (on peninsula opposite Micco), 1. PLATK II. Skulls of 9 subgenera, upper view. [Enlarged one and one-half times.] Fi<;. 1. Microtus (Microtus) pennsylvanicus. - Hyattsville, Md. (No. 87163, 9 ad., U.S.Nat. Mus.) 2. Microtus (Arvicola) macropus. Sawtooth Lake, Idaho. (No. 31451, 9 ad., U. S. Nat. Mus. ) 3. Microtus (Neofiber) alleni. Eden, Fla. (No. 24112, 9 ad., U.S. Nat. Mus.) 4. Microtus (Pedomys) austerus. Racine, Wis. (No. 92851, £ ad., U. S. Nat. Mus. ) 5. Microtus (Pitymys) pin et or am. Frbgmore, S. C. (No. 1523, £ ad., Merriam collection.) 6. Microtus (Lay ur us) curtatus. Mouui Magruder, Nov. (No. 41017, £ ad., U. S. Nat. Mus. ) 7. Microtus (Chilotus) oregoni. Astoria, Greg. (No. 24255, £ ad., U. S. Nat. Mus. ) 8. Microtus (Orthriomys) umbrosus. Mount Zempoaltepec, Oaxaca, Mexico. (No. (58469, 9 ad., U. S. Nat. Mus.) 9. Microtus (Ihrpetomys) guatemalensis. Todos Santos, Guatemala. (No. 76776, £ ad., U. S. Nat. Mus.) 80 North American Fauna, No. 17. Plate II. Skulls of Representative Species of the Nine Subgenera of Microtus (top view). 1. Microtus (Microtus) pennsylvanicus. 2. Microtus (Arvicola) macropus. '■'>. Microtus (Neofiber) altcni. 4. Microti/* i Pedomys) austerus. 5. Mir rut us (Pitymys) pinetorum. 6. Microtus (Lagurus) curtatus. 7. Microtus (Chilotus) oregoni. 8. Microtus (Ortfiriomys) umbrosus. 9. Microtus (Herpetomys) guatemalensis. 18392— No. 17- PLATE I IT. Skulls of 9 subgenera, lower view. [Enlarged imo and one-half times.] Fig. 1. Microtus (Microtus) pennsylvanicus. Hyattsville, Md. (No. 87163, 9 ad., U. S. Nat. Mus. ) ' 2. Microtus (Arvicola) macropus. Sawtooth Lake, Idaho. (No. 31451, 9 ad., U.S. Nat. Mus.) 3. Microtus {Neofiber) alleni. Eden, Fla. (No. 24 112, 9 ad., U.S. Nat. Mus.) 4. Microtus (Pedomys) uusti :rus. Racine, Wis. (No. 92851, $ ad., U. S. Nat. Mus.) 5. Microtus (Pitymys) pinetorum. Froginore, S. C. (No. 1523, $ ad., Merriam Collection.) 6. Microtus (Lagurus) curtatus. Mount Magruder, Nev. (No. 41017, $ ad., U. S. Nat. Mus.) 7. Microtus (Chilotus) orcgoni. Astoria, Oreg. (No. 24255, $ ad., U. S. Nat. Mus.) 8. Microtus ( Orthriomys) umhrosus. Mt. Zempoaltepee, Oaxaca, Mexico. (No. 68469, 9 ad., U.S. Nat. Mus.) 9. Microtus (Herpetomys) guatemalensis. Todos Santos, Guatemala. (No. 76776, $ ad., U. S. Nat. Mus.) 82 North American Fauna No. 17. Plate III. .Skulls of Representative Species of the Nine Subgenera of Microtus (bottom view). 1. Microtus {Microtus) pennsylvanicus. 2. Microtus (Arvicola) macropus. 3. Microtus (Neofiber) alleni. 4. Microtus (Pedomys) ouster us. 5. Microtus (Pitymys) pinetorum. 6. Microtus (Lagurus) curtatus. 7. Microtus {ChUotus) oregoni. 8. Microtus (Orthriomys) wmbrosus. 9. Microtus (Herpetomys) guatemalensis. PLATE IV. Skulls of 7 groups in subgenus Microtus, upper view. [Enlarged one and one-half times.] Fig. 1. Microtus mordax. Red Lodge, Mont. (No. 67305, 9 ad., U. S. Nat. Mus.) 2. Microtus nevadensis. Ash Meadows, Nev. (No. 39663, 3 ad., LI. S.Nat, Mus.) 3. Microtus nanus. Sawtooth Lake, Idaho. (No. 75181, $ ad., U. S. Nat. Mus.) 4. Microtus operarius. St. Michael, Alaska. (No. 22214, $ ad., U. S.Nat. Mus.) 5. Microtus chrotorrhiuus. Mount Washington, N. H. (No. 1501, $ ad., Langs Collection.) 6. Microtus townsendi. Sieilacoom, Wash. (No. 42SI21, $ ad., U. S. Nat. Mus.) 7. Microtus californicus. Walnut Creek, Cal. (No. 44678, $ ad., U. S. Nat. Mus.) 8. Microtus mexicanus. Orizaba, Puebla, Mexico. (No. 53406, 9 ad., U. S. Nat. Mus.) 84 .North American Fauna, No. 17. Plate IV. Skulls of Representative Species of Seven of the Groups in the Subgenus Microtis (top view). 1. Microtus mordax. 2. Microtus in null n.U. S. Nat. Mus. , Biological Survey- Coll. Collected June 26, 1892, by J. Alden Loring. Orig. No., 272. Distribution. — Known ordy from the type locality. General characters. —Size large; tail long, well haired, slightly peni- cillate; hind sole naked medially to posterior fifth, which is hairy; pelage soft, full, and long; antitragus not lobed; mastoids greatly developed. Color. — Above, pinkish buff delicately lined with black; basal fifth of hairs plumbeous; underparts white; lateral line buff, rather wide, extending on forelegs nearly to wrist; orbital area pale; white spot present at base of ear above and below; tail buff, mixed with black above. Skull. — Size large; mastoids excessively developed, bulging in all directions and reaching the maximum shown in the genus; audital bulla? relatively small, about as large as in P. apache, weakly apposed sept., 1900.] PEROGNATHUS LONGIMEMBRIS. 33 anteriorly; interparietal relatively very small, pentagonal, about as long as broad; rostrum long and slender, nasals more slender than in apache, nasal branches of premaxillse wider; zygomata narrowing anteriorly; interorbital width moderate; lower premolar about equal to or very slightly larger than last molar. Measurements.— Type: Total length, 155; tail vertebrae, 80; hind foot, 20. Sfadl: (See table, p. 62.) Remark*. — Both externally and cranially P. amplus is very peculiar and evident!}' has no close relation with any previously known species. In proportions (not in size) and character of pelage it is not very unlike P. panamintinus, and from some of the forms of this species it is but slightly dissimilar in color, but its remarkable skull and slightly haired hind foot are unique. The great development of mastoids which it shows is not at all correlated with an equal enlargement of the audital bulla?, as these are no larger than in P. apache. It has no important characters in common with apache and can not be closely related to it. Specimens examined. — One, the type. PEROGNATHUS LONGIMEMBRIS (Coues). San Joaquin Pocket Mouse. Otognosis longimembris Coues, Proc. Acad. Nat. Sci. Phila., 1875, 305, under Cricetodipus parvus. (Type from Fort Tejon. ) Cricetodipus parvus True, Proc. U. S. Nat. Mus., IV, 474, 1882. Perognathus inornatus Merriam, N. Am. Fauna No. 1, 15, 1889. (Type from Fresno. ) Type locality. — Fort Tejon, Canada de las Uvas, Kern County, Cal. Distribution. — Sonoran zone of the San Joaquin Valley, California, and its immediate extensions. General characters. — Size large, equaling P. apache; color uniform, all markings reduced; antitragus not lobed; pelage rather harsh; proximal third of hind sole hairy. Color. — Above, buff mixed with more or less black; below, white; bases of hairs on rump slightly or not plumbeous; lateral line poorly defined, concolor with upper sides; tail buff, paler on lower surface, faintly dusky above; upper side of forelegs generally buff to wrist; ears buffy outside, dusky within, a slight stripe of white on inflexed portion and the usual white spot at base. Young adults darker than adults, and showing fig.7.— skuii of a Slight tinge of olivaceOUS. Perognathuslong- Skull. — Size large, mastoids and audital bulla? moder- ate, not bulging as in brevinasus; interparietal subquadrate, relatively smaller than in brevinasus; interorbital space very narrow, often dor- sally concave in old individuals; nasals long; lower premolar larger than last molar. 3794— No. 18 3 34 NORTH AMERICAN FAUNA. [no.18. Measurements. — Average of 4 adult males from Fresno, Cal. : Total length, 145.2; tail vertebrae, 74.5; hind foot, 18.7. Of 4 adult females: Total length, 136; tail vertebrae, 71.5; hind foot, 18.3. Skull: (See table, p. 62.) Remarks. — The above description is based mainly on specimens from Fresno, the type locality of ' inornatus.^ The type of longi- membris is immature, but its skull shows the narrow interorbital space peculiar to the San Joaquin Valley form. The only available topotype is fortunately a young adult which agrees perfectly with specimens from Fresno and other points in the San Joaquin Valley. Two young specimens from San Emigdio and Rose Station, both very near Fort Tejon, are also clearly the same as those from Fresno, having the harsher pelage and slight olivaceous effect so different from the soft hairs and delicate pearly color of the young of panamintinus and sub- species. Thus it seems that the name longimembris should be applied to the animal recently called inornatus rather than to the San Bernar- dino form. The species is very distinct, though its range is limited. It seems to be exclusively confined to the San Joaquin Valley, where it is the only representative of the genus. Young adults may be distinguished from old by their smaller size and darker color. Females are con- stantly smaller than males. Among adults two phases of color are apparent, one in which the hairs are grayish from the roots and another in which they are buffy. Specimens examined. — Total number, 111, from localities as follows: California: Alila, 2; Bakersfield, 5; Delano, 2; Fort Tejon, 2; Fresno, 54; Huron, 3; Livingston, 11; Lodi, 3; Oakdale, 2; Ripon, 2; Rose Station, Kern County, 1; San Emigdio, Kern County, 1; Three Rivers, 2; Tipton, 7; Walker Basin, Kern County, 14. PEROGNATHUS PARVUS (Peale). Oregon Pocket Mouse. Cricetodipus parvus Peale, U. S. Expl. Exp'd., VIII, Mamm. and Ornith., 52-54, 1848. Perognathus parvus Cassin, U. S. Expl. Exp'd., Mamm. and Ornith., 48-49, 1858; Merriam, N. Am. Fauna No. 1, 28, 1889 — Peale' s description copied. Perognathus monticola Baird, Mamm. N. Am., 422, 1857; Merriam, N. Am. Fauna No. 1, 17, 1889. Type locality. — Oregon. Assumed to be The Dalles, Oreg. Distribution. — Valley of the Yakima River, Washington, and thence southward to central and southeastern Oregon. Upper Sonoran zone. Geyieral characters. — Size large: tail slightly penicillate, its verte- bra? longer than head and body; ears moderate, well haired, antitragus prominently lobed; proximal fourth of hind sole hairy; color vari- able, presenting two extremes, a gray and a buff.1 1 This species is certainly to some degree dichromatic, for the color variation is evidently not due to age, sex, or season. In one phase the huff is reduced to grayish bjspt., 1900.1 PEROGKNATHUS PARVUS. 35 Color. — Gray phase: Above, pale slaty buff mixed with black, dark- est in center of back; below, white, except belly, the hairs of which are normally plumbeous, with pale tips; sides like back, but paler; black-tipped hairs of back running forward across sides and reaching or nearly reaching forearm: lateral line buff; tail tricolor, dusk}7 above, becoming black terminally, buff on sides, generally white below, but sometimes suffused with buff}7; ears dusky, lighter on mar- gins; subauricular spot moderate; feet white; inner side of hind legs dusky to heel. Buffphast : Everywhere as in gray phase, but general color buff or ochraceous buff instead of slaty. Young: Above, clear, light plumbeous, tips of hairs very pale buff, gradually intensifying with increasing age; below, as in adult. In late fall the high pelage which succeeds the breeding pelage becomes much paler as the black tips of the hairs wear off and expose the undercolor. Skull. — Size large: cranium slightly arched; rostrum somewhat attenuate; audita] bulla? and mastoids moderately developed; audital bulla?, meeting anteriorly in a well-defined symphysis; interparietal wide, pentagonal, anterior angle strong; lower premolar smaller than last molar. Measurements. — Average of five adults from Mabton, Wash. :x Total length, 171.8; tail vertebras, 91.8; hind foot, 22.4. Skull: (See table, p. 62.) Remarks. — The group for which parvus stands contains seven closel}7 related forms. All are of relatively large size and have the antitragus distinctly lobed, thus requiring but slight comparison with the other members of the subgenus. P. p. olivaceus is the most centralized form. It occupies the main part of the Great Basin proper and the others, which are found in the various Great Basin extensions, have evidently been derived from it. The name parvus, though one of the earliest proposed for a pocket mouse, has been usually incorrectly applied. Peale assigns the species to Oregon, and his original description and measurements indicate one of the larger members of the genus.2 Since but one species is found in the part of Oregon traversed by the Wilkes expedition, and since this agrees in general with Peale's description, there seems to be no reason why the name parvus should not now be applied to it. The drab, and in another it is developed into cinnamon, or even bright ochraceous. Between these extremes occur various intermediate stages. As might be expected, one phase is often much more numerous at a given locality than the other, though both are found together. The two are perfectly distinct in both adults and young. 1 Although numerous specimens from The Dalles have been examined none are sufficiently adult to afford satisfactory measurements, so that it has been necessary to use the Mabton series for this purpose. 2 The measurements alone are sufficient to prove that the name should never be used for alive-toed kangaroo rat. Cf. Rhoads, Proc. Acad. Nat. Sci. Phila., 1893, 407-410. 36 NORTH AMERICAN FAUNA. [no. 18. form found at The Dalles is here considered typical. The chances that the type was taken there are considerable since the species is very abundant there and members of the Wilkes expedition camped at or near that place on several different occasions.1 It is also not improbable that the type of Baird's ' monticola'' was also taken at The Dalles. Baird's queried statement that it came from St. Mary's Mission, Mont., is rendered much more doubtful by the unsuccessful efforts of recent collectors to obtain additional specimens from that locality. Dr. Suckley, who collected this type, stopped for some time at The Dalles and may have obtained it there, as pocket mice are probably more abundant there than at any other point at which he stopped. Its skull agrees more nearly with that of parvus than with that of any other form. Specimens examined. — Total number, 103, from localities as follows: Oregon: Antelope, 1; Burns, 5; Crown Rock, John Day River, 3; Harney, 1; Heppner, 2; Lost River, Klamath Basin, 5; Narrows, Malheur Lake, 6; North Dalles, 11; Prineville, 1; Rock Creek Sink, 2; Shirk, 5; The Dalles, 13; Tule Lake, 5; Tumtum Lake, 7; Twelve-mile Creek, 1; Umatilla, 2; Willows Junction, 2. "Washington: Mahton, 25; North Yakima, 6. PEROGNATHUS PARVUS MOLLIPILOSUS (Coues). Coues Pocket Mouse. Pewgnathus mollipilosus Coues, Proc. Acad. Nat. Sci. Phila., 1875, 296 (under P. monticola). Perognathus monticola Townsend, Proc. U. S. Nat. Mus., X, 177, 1888. Type locality. — Fort Crook, Shasta County, Cal.2 Distribution. — Great Basin extension of northeastern California, north to Klamath Basin, Oregon. Upper Sonoran zone, except on Mount Shasta, where it ascends to the Boreal. General characters. — Size somewhat smaller than parvus; ears much larger, antitragal lobe prominent; coloration dark; markings intense. Color.- — Above, rich ochraceous buff, black-tipped hairs very abun- dant; lateral line prominent; white subauricular spot very faint or not evident; below, white, varying to tawny ochraceous on belly. /Skull. — Size relatively rather small; very similar to P. olwact us, but with the ascending branches of the premaxillse abruptly truncated, not exceeding the nasals. Measurements. — Average of three adults from the type locality: Total length, 168.3; tail vertebra?, 88; hind foot, 22.3; ear from meatus (dry), 8.2. Skull: (See table, p. 62.) Remarks. — The specimens from Fort Crook and Fall River Valley are the only ones that may be considered strictly typical. They are 1 Wilkes, Narrative U. S. Expl. Exp'd, IV, 403-432, 1845. 2 Fort Crook, now abandoned, was located about 2 miles northeast of the present site of Burgettville, or Swasey. bept.,1900.] PEROGNATHUS PARVUS OLIVACEUS. 37 well characterized by large ears, rich color, obsolescent subauricular spots, and truncated premaxilla?. Nearly all the others here referred to mollvpilosus show greater or less tendency toward olivaceus. The form seems to be one like raagruderensis, which is rather ill defined, but of a type too strongly characterized to be left unrecognized. Specimens from the Boreal zone on Mount Shasta do not seem to be separable, notwithstanding their very anomalous distribution.1 Speci/mens r.iut 111 nied. — Total number, 44, from localities as follows: California: Alturas, 1; Cassel, 6; Edgewood, 3; Fall Lake, Fall River Val- ley, 1; Fort Crook, 5; Likely, 1; Madeline Plains, 2; Mount Shasta (head of Panther Creek, altitude 7,800 feet, 8; pine helt, south base 4), 12; Sisson, 2; Susan ville, 2. Oregon: Summer Lake, 2; Swan Lake Valley, 4: Williamson River, 3. PEROGNATHUS PARVUS OLIVACEUS (Merriam). Great Basin Pocket Mouse. Perognalhus olivaceus Merriam, N. Am. Fauna No. 1, 15, 1889; ibid., No. 5,71,1891; Elliott, Field Columbian Mus., Zool. Ser., I, No. 10, 211, 1898. Perognafhus olivaceus amanus Merriam, N. Am. Fauna No. 1, 16, 1889. Type locality. — Kelton, Utah. Distribution. — Upper Sonoran zone throughout the Great Basin, from northern Utah and southern Idaho southwest to Owens Valley, California, and west to southern Oregon and northeastern California. General characters. — Similar to P. parvus; differing in softer pelage, lighter color, and slight cranial characters. Color. — Similar to the buff phase of P. parvus, but with clearer, softer colors; above, bright cinnamon buff finely mixed with black; lateral line distinct; subauricular spot conspicuous: hairs of belly pure white or with plumbeous bases and buff tips; inner side of foreleg white or buff. Late fall pelage paler. Skull. — Similar to that of parvus but slightly larger; mastoids more inflated; interparietal slightly smaller (ratio of interparietal width to basilar length of Hen- sel, 27. 8); audital bullae meeting anteriorly in a very weak sj^mphysis or not meeting; ascending branches of premaxilla? generally exceeding nasals. Measurements.- Tyj><: Total length, 184; tail verte- bra1, 101; hind foot, 23. Average of three males from fig. 8.— skuii of Salt Lake City, Utah: Total length, 175.6; tail verte- ^gnathus "'"'"" brae, 95.6; hind foot, 22. Average of three females from Ogden, Utah: Total length, 167.7; tail vertebrae, 88; hind foot, 21.7. Skull: (See table, p. 62.) Remarks. — In the wide range of this form are found numerous 1 See N. Am. Fauna No. 16, 98, 1899. 38 NORTH AMERICAN FAUNA. [no. 18. more or less trivial deviations from the type. Most of these are of size only and probably represent nothing more than individual vari- ation, which in this respect is often considerable. A difference in size between the sexes is also quite noticeable. The dark undercolor shown by the type of iamcmus'> has been observed in many speci- mens from various localities, and in the series now available from Nephi are individuals with pure white belly hairs, as in the type of olivaceus. /Specimens examined. — Total number, 126, from localities as follows: California: Benton, 1; Bishop (Veek, 1 ; Long Valley, 4; Lower Alkali Lake, 1; Moran, 4. Idaho: Bear Lake (east side), 10; Big Butte, 1; Birch Creek, 3; Blackfoot, 2; Lemhi, 1 ; Pahsimeroi Valley, 3. Nevada: Anderson, 1; Bull Run Mountains, 3; Carson Valley, 1; Cottonwood Range, 5; Elko, 6; Golconda, 1; Granite Creek, 5; Halleck, 5; Monitor Valley, 5; Mountain City, 3; Pyramid Lake, 3; Reese River, 6; Ruby Valley, 9; Winnemucca, 1. Utah: Blacksmith Fork, Cache County, 2; Kelton, 2; Laketown, 2; Nephi, 9; Ogden, 17; Otter Creek, 2; Salt Lake City, 4. Wyoming: Fort Bridger, 1. PEROGNATHUS PARVUS MAGRUDERENSIS subsp. nov. Mount Magruder Pocket Mouse. Type from Mount Magruder, Nev. (altitude 8,000 feet). $ ad., No. |j^, U. S. Nat. Mus., Biological Survey Coll. Collected June 6, 1891, by Vernon Bailey. Orig. No., 2899. Distribution. — Upper Sonoran and Transition zones of the desert ranges of southern Nevada and adjoining portion of California. General characters. — Similar to P.p. olivaceus, but very much larger, being the largest member of the parvus group. Color. — As in P. p. olivaceus. Skull. — Very much as in olivaceus, but considerably larger and heav- ier; interparietal relatively narrower (ratio of interparietal width to basilar length of Hensel, 25.1). Measurements. — Type: Total length, 198; tail vertebrae, 107; hind foot, 26. Average of five adult topotypes: Total length, 191; tail vertebrae, 102.2; hind foot, 24.2. Skull: (See table, p. 62.) Remarks. — P. p. magruderensis is a large incompletely differentiated mountain form closely related to olivaceus which is found near it at a lower altitude. The form found on the Panamint Mountains shows trifling differences from typical magruderensis, but is here considered the same. Specimens examined. — Total number, 27, from localities1 as follows: California: Coso, 8; Inyo Mountains, 2; Panamint Mountains, 7; White Mountains, 2. Nevada: Mount Magruder, 7; Grapevine Mountains, 1. 'See N. Am. Fauna, No. 7, 361-384, 1893. seft.,1900.] PEROGNATHUS LORDI. 3i> PEROGNATHUS ALTICOLA Rhoads. White-earkd Pocket Mouse. Perognathus aMcolus Rhoads, Proc. Acad. Nat. Sci., Phila., December, 1893,412. Type locality. — Squirrel Inn, San Bernardino Mountains, California. Distribution. — Known only from the type locality. General characters. — Similar to P.p. olivaceus, from which it differs in somewhat smaller size, in color of ears and tail, and in slight cra- nial characters. ( blor. — Above, as in P.p. oH/uaceus; sides like back, lateral line not prominent; below, white; ears clothed within and without with clear white hairs; tail faint buff above, terminal fourth slightly dusky, white below. Skull. — Essentially as in P. p. olivaceus; ascending branches of supraoccipital very broad and heavy; interparietal rather narrow. Measurements. — -Average of two adult topotypes: Total length, 165; tail vertebrae, 83.5; hind foot, 22.2. Skull: (See table, p. 62.) Remarks. — This isolated species may be immediately distinguished from the other members of the parvus group by its light ears and tail. The type agrees perfectly with the topotypes upon which the descrip- tion is based. Specimens examined. — Total number, 4, all from the type locality. PEROGNATHUS LORDI (Gray). Northwest Pocket Mouse. Abrom>/s lordi Gray, Proc. Zool. Soc. Loudon, 18H8, 202. Perognathus lordi Rhoads, Proc. Acad. Nat. Sci. Phila., 1893, 405. Type locality. — British Columbia. Distribution . — Upper Sonoran and Transition zones of the plains of the Columbia River, Washington, and suitable adjacent territory in southern British Columbia. General characters. — Similar to P. parvus; size large (nearly equal- ing magruderensis)) tail long; feet and ears moderate; antitraguslobed; color dark; interparietal narrow. Color. — Above, pale slaty buff, strongly mixed with black; general color as in the gray phase of P. parvus/ hairs of belly generally with plumbeous bases and buffy tips, leaving a small inguinal and a large pectoral patch pure white; subauricular spot small but distinct; tail tricolor, as in parvus. Skull. — Size large; audital bulla? and mastoids inflated; audita! bullae alwaj s connected anteriorly; interparietal squarish pentagonal, deeply notched by occipital. Measurements — Average of seven adults from Oroville, Wash.: Total length, 183 , tail vertebrae, 97.7; hind foot, 23.2. Skull: (See table, p. 62.) Remarks. — The numerous specimens examined from various parts of the country in which John Keast Lord collected leave little doubt 40 NORTH AMERICAN FAUNA. [no.18. that this was the pocket mouse to which his name was given by Gray in 1868; but in order to remove all uncertainty, specimens were sent to Mr. Oldfield Thomas, curator of mammals in the British Museum, who kindly compared them with the type and found that they agreed in every essential particular. In color lordi is almost identical with the gray phase of P. monticola, but its large size and small interpari- etal show it to be a very different species. Apparently it does not occur on the west side of the Columbia at Wenatchee or south of that point. Specimens from Coulee City, Douglas, and vicinity are grading toward columbianus. Specimens examined.— -Total number, 131, from localities as follows: British Columbia: Ashcroft,14; Kamloops, 6; Okanagan, 12; Vernon, 2. Idaho: Lewiston, 1. "Washington: Almota, 16; Asotin, 11; Chelan, 2; Cheney, 3; Conconully, 3; Coulee City, 6; Douglas, 11; Fort Spokane, 7; Marcus, 1; Orondo, 7; Oro- ville, 9; Spokane Bridge, 11; Wenatchee (east bank of Columbia), 9. PEROGANATHUS LORDI COLUMBIANUS (Merriam). Columbian Pocket Mouse. Perognathns columbianus Merriam, Proc.Acad. Nat. Sci. Phila., September 27, 1894, 236. Type locality. — Pasco, Wash. Distribution. — Vicinity of type locality. General characters. — Similar to P. lordi, from which it differs in slight cranial characters. Color. — As in P. lordi. Skull. — Audita! bulla? and mastoids highly developed: interparietal width much reduced; otherwise as in P. lordi. Measurements. — Average of five adults from Pasco, Wash. : Total length, 179.8; tail vertebra?, 92; hind foot, 22.8. Skull: (See table, p. 62.) Remarks. — This form is found only on the hot plains about the Great Bend of the Columbia. The great development of audita! bullae and mastoids and consequent reduction of interparietal width exhibited by it is the extreme shown in the parvus group. Specimens examined. — Total number, 26, from localities as follows: Washington: Pasco, 12; Touchet, 14. PEROGNATHUS FORMOSUS Merriam. Long-tailed Pocket Mouse. Perognathus formosun Merriam, N. Am. Fauna No. 1, 17, October 25, 1889. Type locality. — St. George, Utah. Distribution. — Southwestern Utah, southern Nevada, and the adjoin- ing portion of California. Lower Sonoran zone. General characters. — Size large (about equal to P.p. yg. ad., No. 91877, U. S. Nat. Mus., Bio- logical Survey Coll. Collected December 16, 1897, by E. A. Goldman. Orig. No., 11968. Distribution. — Upper Sonoran zone from Valparaiso, Zacatecas, to Celaya, Guanajuato, Mexico. General characters. — Somewhat larger and darker-colored than his- pidus; otherwise similar. <'<>l<>r. — Much darker and more olivaceous than in hispid us; general color of upperparts between the hair-brown and olive of Ridgway ; bases of hairs very dark plumbeous; lateral line pure ochraceous, well defined, slightly paler than in hispidus; spots at base of whiskers intensely black and very conspicuous; tail sharply black above; under- parts white. Skull. — As in hispidus, but somewhat larger. Measurements. — Type: Total length, 211; tail vertebra?, 105; hind foot, 27.5. Skull: (See table, p. 62.) Remarks. — This form seems to be related most nearl}' to paradoxus and, like it, inhabits the Upper Sonoran zone. Its dark olivaceous color makes it easily recognizable. Specimens examined. — Total number, 10, from localities in Mexico, as follows: Guanajuato : Celaya, 1. Zacatecas : Valparaiso, 9. PEROGNATHUS PENICILLATUS Woodhouse. Desert Pocket Mouse. Perognathus penecillalus Woodhouse, Proc. Acad. Nat. Sci. Phila., 1852, 200. Perognathus penicillatus Woodhouse, Sitgreaves Exp'd. Zufii and Colorado River, 49, pi. 3, 1854; Merriam, N. Am. Fauna No. 1, 22, 1889. Type locality. — San Francisco Mountain, Arizona.1 1 Woodhouse does not specify exactly where the type was taken. It seems to have been between his camps 15 and 18, which were on the northeast side of the mountain. It is not unlikely that the type came from the Little Colorado Desert, a few miles farther to the northeast. 46 NORTH AMERICAN FAUNA. [no. 18. Distribution. — Vicinity of Colorado River, from Bunkerville, Nev., to Yuma, Ariz., where it meets the range of its subspecies angusti- rostris. The type is the only specimen known from the type locality. Lower Sonoran zone. General characters. — Size rather large, about equal to formosus; tail long, heavily crested, penicillate; sole of hind foot naked to heel; ears scantily haired, shorter and rounder than in formosus, antitragus lobed; pelage rather soft; no spines on rump; color very uniform, markings almost obsolete. Color. — Above, vinaceous buff very finely sprinkled with black; sides exactly like back; lateral line obsolete; subauricular spot present; face and cheeks like back except for a slight darkening under ears; no black spots at base of whiskers; ears outside like back, inside slightly dusky; tail white below to pencil, upper surface and pencil dusky brownish. In the 'left-over' winter pelage the general color is ecru drab instead of vinaceous buff. Skull. — Size medium or rather large; rostrum heavy and high; parietals somewhat flattened; mastoid side of parietal about equaling squamosal side, much exceeded by others; interparietal moderate, all angles rounded, especially posterior ones, anterior angle rounded but distinctly evident; ascending branches of supraoccipital quite heavy; audital bullae widely separated anteriorly; lower premolar larger than last molar. Measurements. — Average of four adults from bend of Colorado River, Nevada: Total length, 205; tail vertebras, 109; hind foot, 25.5. Skull: (See table, p. 62.) Remarks. — The members of the penicillatus group are true Chmtodi- pus, but none of them have rump spines. Characters marking the typical form are large size, uniform color, subdued markings, and heavy skull. The skull of the type which is now available for exam- ination does not agree perfectly with any of the series from the bend of the Colorado River. It is larger and heavier than these, the anterior part is much elevated, and the rostrum broad. These characters, how- ever, are quite pronounced in the Colorado River specimens, and it seems safe to consider them penicillatus, even though no exact dupli- cates of the type are among them. Even the most northern of the Colorado River specimens is somewhat intermediate between true penicillatus and angustirostris. Specimens examined. — Total number, 55, from localities as follows: Arizona: Ehrenberg, 5; Harper Ferry, 3; Fort Mohave, 9; Norton, 4; Wan Francisco Mountain, 1 (type). California: Mohave Mountains, 1. Nevada: Bunkerville, 3; Colorado River, Lincoln County, 8; Colorado River, near Callville, 8; Vegas Valley, 13. bept.,1900.] PEROGNATHUS PENICILLATUS PRICEI. 47 PEROGNATHUS PENICILLATUS ANGUSTIROSTRIS subsp. nov. California Desert Pocket Mouse. Type from Carriso Creek, Colorado Desert, Cal. $ ad., No. 73881, U. S. Nat. Mus., Biological Survey Coll. Collected March 31, 1895, by A. W. Anthony. Orig. No., 22. Distribution. — Colorado Desert; south to northern Lower Califor- nia and east to the Colorado River and southwestern Arizona, where it meets the range of penicillatus and pricei. Lower Sonoran zone. General characters. — Similar to P. penicillatus , but smaller; color about the same; skull lighter and with longer and more slender rostrum. Color. — As in P. penicillatus. SkuU. — Similar in general to P. penicillatus; nasals and ascending premaxilla? long and narrow, much more slender than in penicillatus; interparietal averaging larger and more angular. Measurements. — Type: Total length, 191; tail vertebrae, 105; hind foot (measured dry), 24.4. Average of five topotypes: Total length, 181; tail vertebra?, 103; hind foot, 24. Skull: (See table, p. 62.) Remarks. — The numerous specimens of this subspecies which have been examined include many which are not strictly typical. This is true of the large series from the Colorado River at monument No. 204 and the several localities in the vicinity of Yuma, all of which tend in differing degrees toward tvue penicillatus. From Yuma eastward the tendency is toward pricei. The characters of small size and slender rostrum are very constant in the many specimens from the Colorado Desert, California. Specimens examined. — Total number, 253, from localities as follows: Arizona: Bradshaw City, 1; Gila City, 3; Yuma, 9. California: l Agua Caliente, 3; Baregas Springs, 4; Carriso Creek, 15; Colo- rado Desert, 7; Coyote Wells, 3; Indian Wells, 1; Laguna, 5; Mexican Boundary monument No. 204, near Colorado River, 78; Palm Springs, 55; Salt Creek, 1; San Felipe Canyon, 6; Unlucky Lagoon, 9; Vallecitas, 10; Walters, 7; Whitewater, 2; Fort Yuma, 15. Lower California: Gardner Lagoon, 5; Hardee River (head, near mouth of Colorado River), 2; Poso Vicente, 2; Seven Wells, 10. PEROGNATHUS PENICILLATUS PRICEI (Allen). Price Pocket Mouse. Perogncdhus pricei Allen, Bull. Am. Mus. Nat, Hist., N. Y., VI, 318, November, 1894. Perognathm obscurw Allen, Bull. Am. Mus. Nat. Hist., N. Y., VII, 216, June, 1895. Type locality. — Oposura, Sonora, Mexico. Distribution. — South central Arizona and Northwestern Mexico, west of the Sierra Madre. General characters. — Similar to penicillatus^ but smaller; pelage 1 Nearlv all these localities are in the Colorado Desert, 48 NORTH AMERICAN FAUNA [no.18; harsher, no spines on rump; upperparts more strongly mixed with black; skull short and heavy. Color. — Above, general effect drab or broccoli brown, produced by vinaceous buff strongly lined with black; sides like back, lateral line faintly evident; ears very scantily haired, same color as back; under- pays white; tail bicolor, white below, dusky above. Skull. — Size medium, much smaller than in 'penicillatus; rostrum short and heavy; nasal branches of premaxillre barely exceeding nasals; interparietal moderately wide, anterior angle often obliterated; lower premolar larger than last molar. Contrasted with ■penicillatus the skull of pricei is much smaller, smoother, or less angular, and has very much shorter nasals. In comparison with intermedius it is heavier and less arched, the rostrum is broader, and the nasals are shorter, the mastoids are smaller, and the interparietal is narrower. Measurements. — Average of seventeen adults from Hermosillo, Sonora, Mexico: Total length, 172.5; tail vertebrae, 92.8; hind foot, 22.3. Skull: (See table, p. 62.) Remarks. — The type of P. pricei is very immature, but its skull shows characters amply sufficient to prove that it belongs to the peni- cillatus rather than the intermedius group. Although these groups inhabit the same general region and resemble each other so closely in superficial characters, the skulls are so markedly different as to indi- cate that they bear no close relation to one another. The only exter- nal difference is found in the rump spines. This is not to be relied upon absolutely, however, for though never present in penicillatus and its forms, they are sometimes, though very rarely, absent or unde- veloped in intermedius. In local habitat the two also differ in an interesting way, pricei being found in sandy places, while intermedins prefers the rocks. The extreme form of pricei is found in southern Sonora, where it is so different from typical penicillatus as to suggest full specific rank. Specimens examined. — Total number, 187, from localities as follows: Arizona: Calabasas, 6; Dos Cabezos, 1; Fairbank, 28; Fort Bowie, 2; Fort Huachuca, 1; Fort Lowell, 39; La Osa, 2; Mammoth, 12; New River, 5; Phoenix, 5; Santa Cruz River (west of Patayone Mountain), 3; Sentinel, 2; Tubac, 3; Tucson (twenty miles south), 3; Willcox, 6. Sonora: Batomotal, 13; Hermosillo, 17; Magdalena, 6; Oposura, 8; Ortiz, 10; Quitobaquita, 10; Sonora, 1; Sonoyta, 4. PEROGNATHUS PENICILLATUS EREMICUS (Mearns). Eastern Desert Pocket Mouse. Perognatfms (Chsetodipus) eremicus Mearns, Bull. Am. Mus. Nat. Hist., N. Y., X, 300, August 31, 1898. Type locality. — Fort Hancock, El Paso County, Tex. Distribution. — Extreme western Texas, thence south into north sept.. 1900.] PEROGNATHUS STEPHENSI. 49 central Mexico east of the Sierra Madre at least to La Ventura, Coahuila. General characters. — Size about equal to pricei; color slightly paler; pelage softer; nasals longer and more slender; skull otherwise peculiar. Color. — Essentially as in pricei, but paler; general effect fawn lightly mixed with black; dark area below ears quite prominent; spot at base of whiskers faint. Skull.— Similar to pricei; cranium somewhat arched; nasals long and slender; nasal branches of premaxillas widened at extremities, extending much beyond nasals; supraoccipital slightly bulging behind. Measurements. — Type: Total length, 163; tail vertebras, 83; hind foot, 22.1. Skull: (See table, p. 62.) Remarks. — The average difference between this eastern form of the penicillatus group and its western relative pricei is considerable. The long slender nasals and high arched skull of this form are never found in specimens from west of the Sierra Madre. Specimens from Chi- huahua and Coahuila appear to be quite typical. P. eremicus differs from pricei much as angustirostris does from true penicillatus. In fact, its skull is not very unlike that of angustirostris, but the two are not likely to be confused, on account of the difference in size and color. Specimens from San Bernardino ranch, Arizona, are not typ- ical, being dark-colored and otherwise intermediate. Specimens examined. — Total number, 93, from localities as follows: Arizona: San Bernardino Ranch, Cochise County, on Mexican boundary, 27. Texas: El Paso, 5; Fort Hancock, 3. Chihuahua: Ciudad Juarez, 2; Escalon, 1; Samalayuca, 3; Santa Rosalia, 24. Coahuila: Jimulco, 1; La Ventura, 12; Torreon, 14. Durang-o: Mapimi, 1. PEROGNATHUS STEPHENSI Merriam. Stephens Pocket Mouse. Perognathus Stephens! Merriam, Proc. Acad. Nat. Sci. Phila., September 27, 1894, 267. Type locality. — Mesquite Valley, northwest arm of Death Valley, In} o County, Cal. Distribution. — Known only from the type locality. General characters. — Similar to penicillatus; size very much smaller; tail long, well crested; hind feet naked below; very little or no black in color. Color. — ' Left -or, :■/•' /rinter prhoj, : Above, between pinkish buff and vinaceous buff; effect perfectly uniform, no traces of black anywhere; ears sparsely haired, same color as back; lateral line entirely obliter- ated; face slightly lighter than back and sides; below, white; tail below white, above like back. The post-breeding pelage is doubtless darker and ma}r have more or less black in it. Skull. — Size small; general form much like that of penicillatus; 37(,>4— No. 18 4 50 NORTH AMERICAN FAUNA. [no. is. cranium slightly arched; mastoids rather small; interparietal corre- spondingly large; ascending branches of supraoccipital relatively heavy; lower premolar very large, nearly twice as large as last molar. Measurements. — Type: Total length, 177; tail vertebras, 96; hind foot, 21. Skull: (See table, p. 62.) Remarks. — P. stephensi is a miniature of penicillatus and but slightly larger than arenarius. It is at once separated from the former by its small size and from the latter b}7 its cranial characters. Further col- lections from the desert region of California will doubtless yield more of this interesting species, but at present it is known only from the two specimens which Mr. Stephens caught in the extension of Death Valley known as Mesquite Valley. Specimens examined. — Total number, 2, the type and one topotype. PEROGNATHUS ARENARIUS Mernam. Little Desert Pocket Mouse. Perognathus arenarius Merriam, Proc. Cal. Acad. Sci.,2d ser., IV, 461, September 25, 1894. Type locality. — San Jorge, near Comondu, Lower California. Distribution.— -Known only from the type locality. General characters. — Size very small; tail short, slightly exceeding head and body; pelage rather soft, no bristles anywhere; color plain and uniform, lateral line obsolete; skull short and broad. Color. — Very similar to penicillatus; dorsum buffy drab, finely mixed with black; sides somewhat paler, lateral line not evident; ears dusky, a minute white spot on lower margins; underparts white; tail bicolor. Skull. — Size very small; cranium slightly arched; interorbital and mastoid width relatively great; mastoids moderate, relatively larger than in penicillatus; interparietal broadly pentagonal; nasals rather slender, slightly emarginate at frontal endings; zygomata extremely frail and light; lower premolar larger than last molar. Measurements. — Type (from dry skin): Total length, 136; tail ver- tebras, 70; hind foot, 20. Skull: (See table, p. 62.) Remarks. — P. arenarius is a very aberrant member of the penicil- latus series. It is about the same color as stephensi, but differs from it in size and cranial details, such as more slender nasals, wider inter- orbital space, larger mastoids, and shorter premaxillas. As far as known it is the smallest member of the subgenus Chwtodipus. Specimen examined. — The type. PEROGNATHUS PERNIX Allen. Sinaloa Pocket Mouse. Perognathus perms Allen, Bull. Am. Mus. Nat. Hist., N. Y., X, 149, April, 1898. Type locality. — Rosario, Sinaloa, Mexico. Distribution. — Coast of western Mexico in the States of Sinaloa and Jalisco. sept., woo.] PEKOGNATHUS PERNIX ROSTRATUS. 51 Gene ml characters. — Size small; tail rather long, thinly haired, slightly crested; colors dark; pelage slightly hispid, no spines or bristles anywhere; ears medium; feet naked below. Color. — General color above, hair-brown, uniform over all parts above the lateral line; lateral line distinct, between pinkish buff and oehraceous buff; underparts soiled white; ears dusky, a minute white spot on inferior margins; tail brownish black above, whitish below. Skull. — Size rather small; form narrow and elongate; mastoids quite small; interorbital space much constricted; nasals rather broad and flattened, of medium length; naso-frontal suture not emarginate; interparietal wide, somewhat produced anteriorly; posterior angles much rounded; molar teeth small and weak; lower premolar larger than last molar. Measurements. — Average of four adult topotypes: Total length, 175; tail vertebrae, 97; hind foot, 22.3. Skull: (See table, p. 63.) Remarks. — Perognathus pemix differs from other Mexican species in much smaller size. Its dark color, narrow interorbital space and long nasals distinguish it from all other Chcetodipus not having rump spines. Sp>ecimens examined. — Total number, 48, from localities in Mexico, as follows: Sinaloa: Altata, 2; Culiacan, 17; Mazatlan, 11 (not typical); Rosario, 10. Tepic: Acaponeta, 8. PEROGNATHUS PERNIX ROSTRATUS subsp. now Broad-nosed Pocket Mouse. Type from Camoa, Rio Mayo, Sonora, Mexico. $ yg. ad., No. 95818, U. S. Nat. Mus., Biological Survey Coll. Collected October 28, 1898, by E. A. Goldman. Orig. No., 13167. Distribution. — Coast plains of southern Sonora and northern Sin- aloa, Mexico. General characters. — Size, proportions, and general color about as in P. pemix; skull quite different. Color. — Above, slightly lighter and grayer than pemix; general color oftener broccoli brown than hair-brown; facial area distinctly paler than back and sides; lateral line pinkish buff; lower parts soiled white. Skull. — Similar to pemix, but shorter and broader; rostrum very heavy; nasals, premaxillae, and premaxillaiy branches of zygomata all heavier than in pemix; nasals shorter; interorbital space wider; inter- parietal, mastoids, and audital bullae not tangibly different. Measurements. — Type: Total length, 162; tail vertebrae, 94; hind foot, 23.5. Average of four topotypes: Total length, 161; tail verte- brae, 88; hind foot, 22.5. Skull: (See table, p. 63.) Remarks. — This form is quite a departure from pemix, but inter- gradation with that species is evidenced by a single specimen from 52 NORTH AMEBIC AN FAUNA. [no. 18. Sinaloa. The series of topotypes from Carnoa are constant in their cranial differences f rom pernix, and though no external characters are evident the form seems well worth recognition. Specimens examined. — Total number, 10, from localities in Mexico, as follows: Sinaloa: Sinaloa, 1. Sonora: Carnoa, 9. PEROGNATHUS INTERMEDIUS Merriam. Intermediate Pocket Mouse. Perogncdhus intermedins Merriam, N. Am. Fauna No. 1, 18-19, 1889; ibid., No. 3, 74, 1890. Perognathus obscurus Merriam, ibid., No. 1, 20-21, 1889. Type locality. — Mud Spring, Mohave County, Ariz. Distribution. — Known from several scattered localities in the Sono- ran zone of Arizona, New Mexico, and northern Mexico. General characters. — Size medium, smaller than jjenicillatus / color much darker, with well-defined markings; rump spines rather weak; skull rather small and light. Color. — Winter p>elage: Above, general effect drab, with a strong mixture of black on back and rump; sides paler than back; lateral line pale fawn, quite narrow; ears dusky; tail dusky above, becoming black toward pencil, whitish below, faintly buffy on sides; underparts white, with sugges- tions of bull'. Skull. — Size medium; cranium well arched; rost- rum slender, somewhat depressed; interparietal very wide and strap-shaped, anterior angle normally oblit- erated, others but slightly rounded; lower premolar larger than last molar. Compared with penicillatus it is smaller and less angular; rostrum and nasals much fig. 12.— skuii of more slender; zygomata more sloping; mastoids rela- Pcmgnathus inter- tivelv larger and fuller; ascending- branches of supra- medius. occipital much lighter; interorbital space wider. -Measurements. — Average of four adults from the Grand Canyon of the Colorado, Arizona: Total length, 179.5; tail vertebra?, 102.7; hind foot, 22.7; ear from anterior base, 7. Skull: (See table, p. 62.) Remarks. — Specimens of typical Intermedius are not numerous at present, and the few that are available are in the winter pelage. This makes the determination of \P. obscurus'' a little difficult. The latter is identical with -Intermedium in cranial characters, but slightly more ruddy in color. P. intermedins is much rarer than penicillatus, some form of which is often found near it. In the vicinity of El Paso, Tex., Mr. Vernon Bailey collected both mtermedius and eremicus, the one being found sept., 1900.] PEROGNATHUS NELSONI. 53 in the rocks and the other in the sandy places. At other localities where both occur the same conditions seem to obtain. Specimens examined. — Total number, 46, from localities as follows: Arizona: Grand Canyon, 4; Harper Ferry, 1 ; Fort Bowie, 1; Fort Huachuca, 1; Little Colorado River, Painted Desert, 2; Mud Spring, 2; Willow Spring, 1. New Mexico: Camp Apache, Grant County, 14. Texas: Alpine, 1; El Paso, 2. Chihuahua: Casas Grandes, 4; Chihuahua, 13. PEROGNATHUS NELSONI Merriam. Nelson Pocket Mouse. Perognathus nelsoni Merriam, Proc. Acad. Nat. Sci. Phila., September 27, 1894, 266. Type locality. — Hacienda La Parada, San Luis Potosi, Mexico. Distribution. — Upper and Lower Sonoran zone of central Mexico, covering the table-land from Inde, Durango, south to Lagos, Jalisco, and east to Jaumave, Tamaulipas. General characters. — Similar to intermedins, but larger, darker, and harsher pelaged; tail heavily crested; rump bristled. Color. — Above, general effect hair-brown; hairs dark plumbeous, basally followed by a narrow grayish fawn zone and a heavy black tip; sides like back, orbital region scarcely lighter; lateral line fawn, well denned; underparts dirty whitish; ears dusky, slightly hoary on margins; tail bicolor, black above, whitish below. Worn pelage much paler, becoming drab or ecru drab. Skidl. — Similar to intermedins, but larger and heavier, rostrum and nasals particularly so; interparietal smaller; nasal branches of pre- maxilke exceeding nasals; ascending branches of supraoccipital heavy. Measurement*. — Average of ten adults from the type locality: Total length, 182; tail vertebra?, 104; hind foot, 23; ear from anterior base, 8. Shdl: (See table, p. 62.) Remarks. — This is the commonest pocket mouse of Mexico. It is found in suitable localities over the entire table-land. It is closely related to intermedins and possibly intergrades with it. There are some slight variations in the species, but none are marked enough to warrant separation. Specimens examined. — Total number, 65, from localities in Mexico, as follows: Aguas Calientes: Chicalote, 5. Coahuila: Jimulco, 1; La Ventura, 1; Sierra Encarnacion, 1. Durango: Durango City, 10; Inde, 3; Mapimi, 1. Jalisco: Lagos, 9. San Luis Potosi: Hacienda La Parada, 19; Jesus Maria, 3. Zacatecas: Berriozabal, 9; Canitas, 1; Hacienda San Juan Capistrano, 1; Valparaiso Mountains, 1. 54 NORTH AMERICAN FAUNA. [no. 18. PEROGNATHUS NELSONI CANESCENS (Merriam). Gray Pocket Mouse. Perognathus intermedins canescens Merriam, Proc. Acad. Nat. Sci. Phila., September 27, 1894, 267. Type locality. — Jaral, Coahuila, Mexico. Distribution. — Known only from the type locality. General characters. — Size larger than intermedium; color much paler and more grayish; skull similar to that of P. nelson i. Color. — General color of upperparts drab gray; lateral line pinkish buff, rather narrow; underparts pure white; tail bicolor, mouse gray above, white below. Skull. — Similar to that of nelsonl; differs in more slender nasals, constricted interorbital space, and slightly smaller mastoids. Measurements. — Type: Total length, 193; tail vertebras, 117; hind foot, 22. One topotype: Total length, 184; tail vertebrae, 105; hind foot, 22. Skull: (See table, p. 63.) Remarks. — This form seems to be quite localized. Its habitat is similar to that of the other members of the group. The type and cotypes were caught in the cliffs of a rocky canyon. Specimens examined. — Total number, 3, from the type locality. PEROGNATHUS GOLDMANI sp. nov. Goldman Pocket Mouse. Type from Sinaloa, Sinaloa, Mexico. 9 a(l, No. 96673, U. S. Nat. Mus., Biological Survey Coll. Collected February 15, 1899, /by E. A. Goldman. Orig. No., 13428. Distribution. — Coast plains of northern Sinaloa and southern Sonora, Mexico. General characters.- — Size large; tail moderately long and heavily crested; pelage somewhat hispid, rump with a few short bristles; ears relatively large, much larger than those of nelson i; antitragal lobe prominent, wider at base than at apex; in color and markings similar to nelsoni; skull relatively large and heavy. t^e^A— Similar in general to nelsoni,' general color across shoulders and anterior portion of upperparts, broccoli brown; posterior half of dorsum much darkened by admixture of black; lateral line pinkish buff; ears blackish with hoary margins, externally whitish for distal half; subauricular spot present; tail sharply bicolor, blackish above, white below. Skull. — Size large, much heavier than in nelsoni; mastoids some- what smaller and more ridged; nasals much larger and heavier; skull noticeably higher and not so wide posteriorly, thus making the zygo- mata more nearly parallel. Measurements. — Type: Total length, 202; tail vertebrae, 108; hind foot, 28. Average of five topotypes: Total length, 202; tail vertebra?, 112; hind foot, 28; ear from anterior base, 11. Shall: (See table, p. 63.) Remarks. — The large orbicular ears of this species easily distinguish bept.,1900.] PEROGNATHUS FALLAX. 55 it from nelsoni, its nearest relative. It is one of the several forms peculiar to western Mexico, and, like the others, its known range is quite limited. Specimens from Camoa and Alamos are slightly smaller than those from Sinaloa. Specimens examined. — Total number, 36, from localities in Mexico, as follows: Sinaloa: Sinaloa, 7. Sonora: Alamos, 18; Camoa, 11. PEROGNATHUS ARTUS ap. now Batopilas Pocket Mouse. Type from Batopilas, Chihuahua, Mexico. J ad., No. 96298, U. S. Nat. Mus., Biolog- ical Survey Coll. Collected October 6, 1898, by E. A. Goldman. Orig. No. , 13090. Distribution. — Known only from a few scattered localities in west- ern Mexico. General characters. — Externally similar to goldmani; rump bristles weak or undeveloped; skull distinctive. Color. — As in goldmani. Skull. — Similar to that of goldmani, but smaller and narrower; mastoids much smaller with more strongly marked transverse ridges; audital bullae smaller; nasals moderate, exceeded by ascending pre- nmxilla?; interparietal nearly elliptical, slightly produced anteriorly; zygomata nearly parallel. Measurements. — Average of live adult topotypes: Total length, 191; tail vertebrge, 106; hind foot, 21.6. Skull: (See table, p. 63.) Remarks. — The large size of this species at once distinguishes it from pernix and rostratus, and its very small mastoids separate it from other Mexican species. Externally it is very similar to goldmani, but it has less prominent rump bristles; in fact, they are not at all evident in the majority of specimens. P. pernix was generally found by Mr. Goldman at the same localities as P. goldmani, but at Culiacan he found it in company with P. artus. Specimen*- r.mmitied. — Total number, 15, from localities in Mexico, as follows: Chihuahua: Batopilas, 8. Durang-o: Chacala, 3. Sinaloa: Culiacan, 4. PEROGNATHUS FALLAX Merriam. Short-eared California Pocket Mouse. Perognathus fallax Merriam, N. Am. Fauna No. 1, 19, 1889; Allen, Bull. Am. Mus. Nat. Hist., N. Y., V, 184, 1893. Type locality. — Reche Canyon, 3 miles southeast of Colton, San Bernardino County, Cal. Distribution. — Extreme southwestern California, occupying the region west of the San Bernardino and San Jacinto ranges and extend- ing south into northern Lower California. 56 NORTH AMERICAN FAUNA. [no. 18. General characters. — Size medium, somewhat larger than interme- dins; general color similar but darker; wider and brighter lateral line; rump bristles heavier; tail long and crested; ears moderate. Color. — Above, general effect bister, middle of back and rump with a strong element of black; lateral line and subterminal zone of hairs of upperparts pinkish buff; underparts creamy white; ears dusky on inflexed portions, hoary on inner sides; tail bicolor. Skull. — Similar to intermedlus; cranium arched; nasals slender; mastoids rather large and full; interparietal wide, anterior angle slightly developed; naso-frontal suture slightly or not emarginate. Measurements. — Average of six adults from the type locality: Total length, 192; tail vertebra?, 11; hind foot, 23; ear from anterior base, 9. Skull: (See table, p. 63.) Remarks. — This species falls readily into the small group typified by intermedins. It differs from the other members in size, color, and shape of interparietal. It has been much confused with femoralis on account of its similar color, but its much smaller ear is a convenient external character for distinguishing it. Two specimens from Turtle Bay, Lower California, are similar in color to anthonyi, but cranially the same as fallax, to which they are here referred. Specimens examined. — Total number. 120, from localities as follows: California:1 Ballenas, 1; Bergmann, Riverside County, 1; Carlsbad, 1; Chi- huahua Mountains, 1; Dulzura, 24; El Nido, 3; Encinitas, 1; Herron, San Bernardino County, 5; Jacumba, 8; Lajolla, 1; Mountain Spring, 11; Radec, 5; Reche Canyon, Riverside County, 10; Riverside, Riverside County, 1; Rose Canyon, 10; San Felipe Valley, 4; San Pasqual Valley, 4; Santa Ysabel, 10; San Ygnacio Valley, 1; Summit (Coast Range), San Bernardino County, 4; Temescal, Riverside County, 1. liower California: Cape Colnett, 2; Ensenada, 1; Gato Creek, 1; Jamul Creek, 1; San Isidro Ranch, 2; Sanos Cedros, 1; San Quintin Bay, 1; Tia Juana, 2; Turtle or San Bartolome Bay, 2. PEROGNATHUS ANTHONYI sp. nov. Cerros Island Pocket Mouse. Type from South Bay, Cerros Island, Lower California. 9 ad., No. 81058, U. S. Nat. Mus., Biological Survey Coll. Collected July 29, 1896, by A. W. Anthony. V Orig. No., 71. Distribution. — Known only from the type locality. General characters. — Similar in general to P. fallax; differing in slightly smaller size, more ruddy color, and cranial characters. Color. — Above, grayish fawn mixed with black; lateral line brown- ish fawn, poorly defined; ears dusky; white subauricular spot present; tail dusky above, whitish below. Skull. — Similar to P. fallax; cranium less arched; rostrum heavier; mastoids smaller; interparietal smaller and shorter; zygomatic breadth greater anteriorly. 1 Most of these localities, unless otherwise stated, are in San Diego County. SEPT., 1900.] PEROGNATHUS FEMORALJS. 57 Measurements. — Type: Total length, 168; tail vertebras, 92; hind foot, 23.5. Skull: (See table, p. 63.) Specimens examined. — One, the type. PEROGNATHUS FEMORALIS Allen. Great California Pocket Mouse. Perognalhus femoralis Allen, Bull. Am. Mus. Nat. Hist., N. Y., Ill, 281, June 30, 1891; Rhoads, Proc. Acad. Nat. Sci. Phila., 1893, 407. Type locality. — Dulzura, San Diego County, Cal. Distribution. — Known from a few localities in San Diego County, in extreme southern California, and the adjoining part of Lower California. General characters. — Size very large; tail long, heavily crested peni- cillate; color dark; ears large and elongate; pelage harsh; rump and flanks furnished with strong bristles or spines; skull large and heavy. Color. — Similar to fallow, but quite intensified; above, general color bister, hairs heavily tipped with intense black; lateral line rich pinkish buff; under- parts dirty whitish, sometimes washed or flecked with buffy; tail bicolor. Skull. — Large and heavy; less arched than in fallax; rostrum and nasals much heavier; mastoids relatively smaller; molar teeth relatively weaker: inter- parietal subquadrate, rarely developing a fifth angle; naso-frontal suture slightly emarginate. Measurements. — Average of six adults Fig. 13.— Ear of (a) Perognathus fallax; (b) Perognathus femoralis. tebras, 126; hind ible, p. 63. from the t}^pe locality: Total length, 223; tail verte foot, 27.5; ear from anterior base, 12. Skull: (See U Remarks. — This species has the longest tail and largest hind foot found in the genus, but its body is light in comparison with that of •paradoxus. In color it has a remarkable resemblance to fallax, which is found within its range, but its large size, long ears, and heavy skull are amply sufficient to distinguish it. Specimens examined. — Total number, 60, from localities as follows: California (San Diego County): Dulzura, :!2; Santa Ysabel, 9; Twin Oaks, 16. Lower California: Nachoguero Valley, 3. 58 NORTH AMERICAN FAUNA. [no. 18. PEROGNATHUS CALIFORNICUS Merriam. California Pocket Mousse. Perognathus califomicus Merriam, N. Am. Fauna No. 1, 26, 1889; Allen, Bull. Am. Mus. Nat. Hist., N. Y., 263, 1896; Elliott, Field Columbian Mus., Zool. Ser., I, No. 10, 211, 1898. Perognathus armatus Merriam, 1. c, 27. Type locality. — Berkeley, Cal. Distribution. — Vicinity of San Francisco Bay and south to Bear Valley, San Benito County, where it meets the range of its subspecies dispar. General characters. — Similar to P. femoralis, but smaller; about equal in size to fallax; ears quite elongate; rump and flanks well supplied with bristles; skull very peculiar. Color. — Nearly the same as femoralis, much darker than fallax,' general effect of upperparts bister; hairs pale plumbeous basalty, darkening distally; subterminal zone pinkish buff fol- lowed by heavy black tips; tail bicolor; underparts and feet yellowish white. Skull. — Size medium; cranium considerably arched; mastoids exceedingly small; mastoid width greatly reduced; occiput bulging greatly; interparietal about twice as broad as long, anterior angle very slightly developed; naso-frontal suture deeply emarginate or V-shaped; lower premolar slightly larger than last molar. Measurements. — Average of five adults from the fig. i4.-skuii of type locality: Total length, 192; tail vertebra, 103; /amicus. hind foot, 21; ear from anterior base, 10.5. Skull: (See table, p. 63.) Remarks. — P. califomicus 'is remarkable for its very small mastoids. It has no close relation to fallax, with which it has sometimes been confused. Its long ears and its cranial characters indicate that its closest affinities are with femoralis. Even within its very limited range it is quite a rare animal, and but few specimens are in collections. Specimens examined. — Total number, 18, from localities as follows: California: Berkeley, 7; Bear Valley, San Benito County, 2; Gilroy, 3; Por- tola, San Mateo County, 2; Stanford University, 2. PEROGNATHUS CALIFORNICUS DISPAR subsp. nov. Allen Pocket Mouse. Type from Carpenteria, Santa Barbara County, Cal. $ ad., No. jogggi U. S. Nat. Mus., Biological Survey Coll. Collected December 19, 1891, by E. W. Nelson. Orig. No., 1655. Distribution. — Coast valleys of California from San Bernardino to San Benito County and north along the foothills of the west slope of the Sierras to Placer County. bept.,1900.] PEROGNATHUS SPIKATUS. 59 General elm ruder*. — Larger and paler colored than califomicus^ pelage somewhat softer; skull quite different. Color. — Similar to fallax, paler than califomicus or femoralis,' above, general color bister; facial area slightly lightened; lateral line pinkish buff', sometimes approaching ochraceous buff; underparts buffy white; tail bicolor. Skull. — Similar to that of califomicus, but larger and heavier; in general form resembling that of femoralis,' mastoids quite small; nasals heavy, somewhat elongate; interorbital space narrow. Measurements. — Type: Total length, 218; tail vertebra?, 120; hind foot, 27. Average of six typical adults: Total length, 210; tail verte- bra?, 117; hind foot, 26; ear from anterior base, 12. Skull: (See table, p. 63.) Remarks. — Although this subspecies is somewhat intermediate in character between califomicus and femoralis there seems to be no good evidence of any connection with the latter. It intergrades with cali- fomicus in the vicinity of Bear Valle}^, San Benito County. In typi- cal form, its skull presents the characters of small mastoids and narrow interorbital space found in calif brnicus at the same time almost attain- ing the large size of the skull oi femoralis. Specimens examined. — Total number, 56, from localities as follows: California: Auburn, 1; Bitter Water, 3; Carpenteria, 4; Fort Tejon, 2; Hue- neme (10 miles west), 1; Kern River (25 miles above Kernville), 1; Las Virgines Creek, Los Angeles County, 1; Milo, 1; Nordhoff, 4; Raymond, 1; San Bernardino Peak, 3; San Emigdio, 4; San Fernando, 3; San Luis Obispo, 8; San Simeon, 1; Santa Monica, 1; Santa Paula, 1; Three Rivers, 9; Ventura River, 7. PEROGNATHUS SP1NATUS Merriam. Spiny Pocket Mouse. Perognathus spinaius Merriam, N. Am. Fauna No. 1, 21, October 25, 1889. Type locality. — Twenty-five miles below the Needles, Colorado River, California. Distribution. — Desert region of southern California and northern Lower California. General characters. — Size medium, tail moderately long and crested; ears small and orbicular; pelage hispid, spines large and prominent on rump, scattered on flanks and sides and often extending to shoul- ders; lateral line very faint or wanting. Color. — Above, general effect drab brown; hairs plumbeous basalty, ecru drab subterminally and black-tipped; sides and orbital region slightl}T paler than back; underparts buffy white; lateral line gen- erally faint, in very bright pelage showing as a slender line of ecru drab; ears dusky, subauricular spot small; tail hair-brown above, whitish below; spines white with dusky tips, except on sides where the tips are also white. Many of the hairs of the back often end with a 60 NORTH AMERICAN FAUNA. [no. 18. broad zone of ecru drab without the usual black tip. These, when combined with those having black tips, cause a peculiar mottled appear- ance. The ' left-over1 winter pelage is much paler and grayer, the gen- eral effect being pale drab. Skull. — Size medium; cranium rather slender and much flattened; parietals on nearly same plane as interparietal; mastoids small, not so full as xnfallax and intermedium; interparietal broad but normally with slight evidence of an anterior or fifth angle; supraorbital ridge slightly trenchant; lower premolar about equal to last molar. Measurements. — Average of five adults from Palm Springs, Cal.: Total length, 181; tail vertebrae, 101; hind foot (measured dry), 21.5. Skull: (See table, p. (VS.) Remarks. — Perognathus spinatus has a limited range, rig." is.— skuii of and aside from the excellent series from Palm Springs Peroonathusspin- \n the Bangs collection but few specimens have found (ttUS. , . 11- T • 1 their wa}T into collections. It is the type of the sub- genus Chcetodipus and the representative of a small group quite dis- tinct from all others. Young adults of this group differ from old in being of a grayish color and in having weaker and less numerous spines. In juveniles the first pelage is soft and without spines, which do not appear until an entire new pelage is acquired. In this species, as in most others, the males average slightly larger than the females. Specimens examined. — Total number. 16, from localities as follows: California: Colorado River (twenty-five miles below Needles), 1 (type); Coast Range, San Diego County, 3; LaPuerta, San Diego County, 6; Palm Springs, 21; San Felipe Canyon, 12; Vallecitas, San Diego County, 2. Lower California: Cocopah Mountains, 1. PEROGNATHUS SPINATUS PENINSULA Merriam. Cape St. Lucas Pocket Mouse. Perognathus spinatus •peninsulas, Merriam, Proc. Cal. Acad. Sci., 2d ser., IV, 460, Sep- tember 25, 1894. Type locality. — San Jose del Cabo, Lower California. Distribution. — Cape region of Lower California. General characters. — Similar to P. spinatt/x, but much larger; pelage a trifle more hispid; tail more scantily haired and relatively shorter; ears large and rounded. Colo?'. — As in P. spinatus. Skull. — Similar to that of P. spinatus, but averaging much larger; supraorbital ridges flattened and shelf-like, with very trenchant edges, forming a nearly straight line from mastoids to lachrymals. Measurements. — Average of five adults from San Jose del Cabo and Cape St. Lucas: Total length, 188; tail vertebrae, 101; hind foot, 24. Skull: (See table, p. 63.) sept., 1900.] PEROGNATHUS MARGARITA. 61 Specimens examined. Total number, 23, from localities as follows: Lower California: Cape St. Lucas, 7; Comondu, 2 (intermediate); San Jose del Cabo, 5; Santa Anita, 9. PEROGNATHUS BRYANTI Merriara. Bryant Pocket Mouse. Perognathus bryanti Merriam, Proc. Cal. Acad. Sci., 2dser., IV, 458, September 25, 1894. Type locality. — San Jose Island, Lower California. Distribution. — Known only from the type locality. General characters. — Larger and longer-tailed than peninsulce; skull slightly characterized; otherwise similar. Color. — Apparently as in peninsulce. Skull. — Slightly larger and heavier than in peninsulce; somewhat more elongate; nasals longer and more slender; interparietal wide and subquadrate; lower premolar equal to or slightly larger than last molar. Measurements. — Type: Total length, 216; tail vertebrae, 12T; hind foot, 25. One adult topotype: Total length, 225; tail vertebra?, 128; hind foot, 25. Skull: (See table, p. 63.) Remarks. — This insular form is well characterized by its large size and long tail. In color it is probably the same as peninsulcB, though the material examined is not sufficient to determine with certainty. Specimens examined. — Total number, 8, all from the type locality. PEROGNATHUS MARGARITA Merriam. Margarita Pocket Mouse. Perognathus margarilse Merriam, Proc. Cal. Acad. Sci., 2d ser., IV, 459, September 25, 1894. Type locality. — Santa Margarita Island, Lower California. Distribution. — Known only from the type locality. General characters. — Size medium; tail longer than head and bod}T; ears moderate; pelage rather harsh, rump and flanks with a few bris- tles; skull peculiar. Color. — Above, much as in spinatus, pale fawn mottled and lined with hair-brown and black; lateral line scarcely evident; subauricular spot present, but very small; underparts and feet dirty white. Skull. — Size rather small; cranium somewhat arched; mastoids exceedingly small, fully as small as in calif ornicus/ nasals moderate, nasofrontal suture emarginate; occiput not projecting posteriorly; interparietal wide, anterior angle evident, others very much rounded; interorbital space moderate, about as in calif ornicus ; lower premolar larger than last molar. Measurements. — Type (from diy skin): Total length, 180; tail verte- bra?, 102; hind foot, 22.5. Skull: (See table, p. 63.) Remarks. — So far as known, this species has no near relative on the mainland adjacent to its habitat. In cranial characters it seems to be 62 NORTH AMERICAN FAUNA. [NO. 18. somewhat similar to calif ornicus, while externally it is a combination of fallax and spmatus. Specimen examined. — The type. Cranial measurements of Perognathus. [All measurements are in millimeters.] Species. Perognath u s fascial u s P.J. infraluteus P.flavescens P. merriami P. m. gilvus P.flavus P.flavus P.f. bimaculatus P. apache P. apache melanotis 2. P. caUistus P.panamintinus P.p. brevinasus P. nevadensis P. pacificus " P. longimcmbris P. amplus - P. parvus P.p.olivaceus P.p magruderensis .. P.p. mollipilosus P. alticola P. lordi P. I. columbianus P. formo&us P. bailcyi P. hispidus P. h. paradoxus P. h. sacatecse - P. penicillatus - P. penicillatus P. p. angustirostris. . . P. p. pricei P. p. eremicus - P. stephcnsi - P. arenarius - P. intermedins - P. nelsoni Locality. Tilyou ranch, Mont Loveland, Colo Kennedy, Nebr Brownsville, Tex Eddy.N.Mex El Paso, Tex Fort Huachuca, Ariz Fort Whipple, Ariz Keams Canyon, Ariz CasasGrandes, Chihuahua, Mexico. Kinney ranch, Wyo Panamint Mountains, Cali- fornia. San Bernardino, Cal Halleck, Nev Edge of Pacific Ocean, Mexican boundary. Fresno, Cal Fort Verde, Ariz Mabton, Wash Salt Lake City, Utah Mount Magruder, Nev Fort Crook, Cal San Bernardino Mts., Cal ... Oroville, Wash Pasco, Wash St. George, Utah Magdalena, Mexico Mier, Mexico Kansas and Nebraska Valparaiso Mountains, Mex- ico. Little Colorado Desert, Arizona. Colorado River, near Call- ville, Nev. Carriso Creek, California... Oposura, Sonora, Mexico... Fort Hancock, Tex Mesquite Valley, California San Jorge, Lower California Mud Spring, Ariz La Parada, Mexico See note, p. 14. "o 'en CO oS CO a ■ . .a Pr-1 a o O 'c CO S3 -a CO "£ ^ o! 0J u O 16.5 23 11.8 15.4 21.8 11.7 15.1 22.1 11.6 14.8 20.4 11.2 15 20.7 11.7 14.6 21 12 14.5 20.3 11.7 15.9 21.6 13 16.3 23 13 15.3 21.5 11.8 16 22.9 13 14.9 21.4 11.9 14.2 20.1 11.8 14.8 20.6 12 13 19 11 17.5 23.6 13 16.8 23.6 13.9 17.6 25.1 13 17.6 12.6 19.1 27 13.6 17.8 25.3 12.8 18 25.4 12.6 18.7 26.7 13.6 18.6 25.9 14.1 18.6 26.7 14 21.5 30 15.6 20.2 28 13.8 24 32 15 22.5 30.2 15 20 28.3 14.5 19.5 27.8 13.4 18 26 13 18.4 26 13 17.5 12.6 16 22.7 12 15.3 23 12 17 24. 5 13.5 18 26 13.8 ">"u 2.3 2.2 2.6 2.4 2.5 2.9 2.8 2.8 2.5 3 2.2 2.2 3 3 3.1 3.7 3.1 3.1 3.2 3.3 3.1 3.7 4.2 4.1 4.7 4 3.5 3.1 3.4 3 3 3.5 3 3. 5 4.3 4.8 4.8 3.6 3.3 2.9 2.8 3.1 4 4.3 4.6 3.5 3.8 3.7 3.4 3.8 3.3 5.4 4.9 4.8 5.1 4.6 4.7 4.1 5.8 6.8 7.2 6.7 7 6.7 6.4 8 I 7.6 ■d'o '££ II S o p h 4.5 5 5 4.S 5.2 5 5.2 4.9 4.8 5 6.6 6.8 7.5 6.4 6.2 6.4 6 6.2 6.3 6.7 7.6 8.6 8.3 7.1 8.2 7 9.2 9.2 10.4 12.2 10. 6 13.2 12 10 10.8 10.3 9.4 9.8 9 8.8 9.4 10 - Type. CRANIAL MEASUREMENTS OF PEROGNATHUS. 63 ( ranial measurements of Perognathus — Continued. [All measurements are in millimeters.] Species. P. n. canescens ' P. goldmani P. artus P.fallax P. anthonyi 1 ... P.femoralis P. calif ornicus . . P.c.dispar P.sjrinatus P. s. peninsulas . P.bryanti P. margarilx ' . . P.pemix P.p. roslraius .. Local itv. ~. Jaral, Coahuila, Mexico 17.5 Sinaloa, Sinaloa, Mexico 20.6 Batopilas, Chihuahua, 18.8 Mexico. Reche Canyon, California.. 18 Cerros Island, Lower Cali- 17.4 fornia. Dulzura, Cal 20. 3 Berkeley, Cal 18.9 Carpenteria, Cal 19.8 Palm Springs, Cal 17 San Jose del Cabo, Lower 18 California. San Jose Island, Lower Cal- 18.9 i fornia. Margarita Island, Lower 18 California. Rosario, Sinaloa, Mexico 17. 4 Camoa, Sonora, Mexico 16. 5 1 Type. ^ -c o CI . 03 oja V ee V O O 25 13.5 27.7 14.3 25. 4 12.4 26 14 25.4 12.9 29.6 14.3 27.4 13 28 13.1 24.3 12.3 26.5 13 27.3 13.1 25.9 12 24. 4 12.2 22.7 11.7 «_• 3.7 3.8 3.3 3.8 2.6 4 4.4 4 3.4 3.7 3.5 3.7 3.3 3.4 ■3a ■So .Q O ° t 6.1 6.5 6.1 6.6 6 7.1 6.4 6.5 6.2 6.6 6.8 6.5 5.4 5.5 INDEX. [Names of new species in black-face type, synonyms in italics.] Abromys, 13. Abromys lordi, 39. Chaetodipus, subgenus, 14,41-62. Cricetodipus, 13. ( 'ricetodipus flavus, 21. parvus, 33,34. Otognosis, 13. Otognosis longimembriq, 33. Perognathus, characters of subgenera classification, 12-13. color and pelages, 11-12. cranial measurements. 02-63. distribution, 11. genus, 13. habits, 12. history and material, 9-11. key to species, 14— 1H. Chsetodipus, 15-16. Perognathus, 14-15. list of species, 17-18. Ch*todipus, 17-18. Perognathus, 17. new species, 13. subgenus. 14,18-41. Perognathus alticola, 14.39. amoznus, ::?. unpins, 15,32-33. anarustirostris, 10,47. anthonyl, 16,56-57 apache, 15,26-27. apache, 24. arenarius, 16,50. arenicola,29. armatus, 58. art us, 16,55. baileyi. 16.41-42. bangsi, 15,29-30. bimaculatus, 15, 24-25. brevhiasus. 15,30-31. bryanti, 16,61. californicus, 15,58. eallistus, 15,28. canescens, it'., 54. columbianus. 14.40. conditi, 44. copei, 20. dispar, 15,58-59. 679J— No. 18 5 Perognathus— Continued. eremicus, 16,48-49. fallax, 16,55-56. fasciatus, 15, 18-1'.'. fasciatuif, 42. femoralis, 15, 57. flavescens, 15.20-21. flavus, 15, 23-24. flavus, 21,26. formosus, 14,40— 11. fuliginosus, 15, 25. gilvus, i5,22-23. goldmani, 16,54-55. hispidus. 16,42-44. infraluteus, 15, 19-20. inornatus, 33. intermedins, 16. 52-53. latirostris, 44. longimembris. 15,33-34. lordi, 14,39-40. magruderensis, l i.;s. margarit*, 16, 61-62. mearnsi, 21. luelanotis. 15,27. merriami, 15, 21-22. mexicanus, 15,25-26. mollipilosus, 14.36-37. monticola, 34, 36. nelsoni. 16,53. nevadensis, 15.31. obscurus, 47. 52. olivaceous. 14.37-38. pacificus, 15,31-32. panamintinus, 15,28-29. paradoxus, 16, 44-45. paradoxus, 42.44. parvus. 14,34-36. peneciUaius, i">. penicillatus, 16,45-46. peninsula-. 16,60-61. pernix, 16,50-51. pricei, 16. 47- 18. rostratus. 16,51-52. spilotus, 42. spinatus, 16,59-60. stephensi, 16. 49-50. zacatecse, 16. 45. 65 PLATE I. [one and one-half times natural size.] Fig. 1. Perognathus fiavus Baird. Topotype. El Paso, Tex. (-No. 25029, U. S. Nat. Mus.) 2. Perognathus ampins Osgood. Type. Fort Verde, Ariz. (Type No. 46711, U. S. Nat, Mus.) 3. Perognathus merriami Allen. Topotype. Brownsville, Tex. (No. 41764, U.S. Nat. Mus.) 4. Perognathus flavus bimaculatus (Merriam). Topotype. Fort Whipple, Ariz. (No. 46478, IT. S. Nat. Mus. ) 5. Perognathus ( Chsetodipus) californicus Merriam. Topotype. Berkeley, Gal. (No. 55560, U. S. Nat. Mus. ) 6. Perognathus ( Chsetodipus) pernix Allen. Topotype. Rosario, Sinaloa, Mexico. (No. 91324, U.S. Nat, Mus.) 7. Perognathus (Chsetodipus) penicillatus Woodhouse. Type. Near San Fran- cisco Mountain, Ariz. (Type No. 2676, U. S. Nat, Mus. ) 8. Perognaihus (Chsetodipus) pernix rostratus Osgood. Type. Camoa, Sonora, Mexico. (Type No. 95818, U. S. Nat. Mus. ) 66 North American Fauna, No. If Plate I. Skulls of Perognathus. 1. Perognath us flavus. 2. P. ampins. 3. P. merriami. 4. P. flavus bimaculatus. 5. P. (Chsetodvpus) caUfornicus. 6. P. (Chsstodipusj pernix. 7. P. (Chsetodipus) penicillatus. 8. P. (Chxtodipus) pernix rostraiu8. PLATE II. [one and one-half times natural size' Fig. 1. Perognathus panamintinus (Merriam). Type. Panamint Mountains, Cal. (Type No. 39866, U. S. Nat. Mus.) 2. Perognathus lordi columbianus (Merriam). Type. Pasco, Wash. (Type No. 39450, U. S. Nat. Mus. ) 3. Perognathus nevadewis Merriam. Topotype. Halleck, Nev. (No. 54565, V. S. Nat. Mus.) 4. Perognathus (Chsetodipus) bryanti Merriam. Type. San Jose Island, Lower California, Mexico. (No. 550, Coll. Calif. Acad. Sci.) 5. Perognathus (Chtetodipus) margaritie Merriam. Type. Santa Margarita Island, Lower California, Mexico. (No. 90, Coll. Calif. Acad. Sci.) 6. Perognathus (Chsetodipus) spinatus peninsula: Merriam. Type. San Jose del Cabo, Lower California, Mexico. (No. 274, Coll. Calif. Acad. Sci.) 7. Perognathus (Chsetodipus) arenarms Merriam. Type. San Jorge, near Co- mondu, Lower California, Mexico. (No. 99, Coll. Calif. Acad. Sci.) S. Perognathus (Chsetodipus) stephensi Merriam. Topotype. Mesquite Valley, Cal. (No. 39874, U. S. Nat. Mus.) 9. Perognathus (Chsetodipus) nelsoui Merriam. Type. Hacienda La Parada, Sail Luis Potosi, Mexico. (Type No. 50214, U. S. Nat. Mus. ) North American Fauna, No. 1 Skulls of Perognathus. 1. Periifii)iillni< jut na in intinus. 2. P. lordi columbianus. 3. P. nevadensis. 4. P. (Chsetodipus) bryanti. 5. P. (Chsetodtpus) margaritse. ti. P. (Chaetodipus) spinqtus penimvtse. 7. P. (Chaetodipus) arenarius. 8. P. iChmtodipui) stepkensi. 9. P. I Chxtodipus) nelsoni. PLATE III. Map showing distribution of the subgenus Perognathus. 70 PLATE IV. Map showing distribution of the subgenus Chsetodipus. o North American Fauna, No. 19. Frontispiece ,78' %■ 17*' .TO' .66 162' IM* I5»' .50' 146" l«' 136' 13*" 130" 126" I22~ m" i» nf i 106" / \ / ">^ / / / 1 1 I 1 6 A / ^O^r^ih^ ■ ,cV5--''\ ^^ \ > \ \ \ / / \ ^ /\ \ > \ p \ \ \ %v \ z-V^l ~^CMV A */v / / / 1 iTiv s*^^ \ l\ \ i \ *^1 \ >> \ vs///>'' / 1^-3sLJP / ""^ s K 'Af> / / '^ / <,/f / •MJr.McKinl.jy/ ST] ^VT ^T?*(vg S^ / &' > AV Prinze o1 \7 A'J, / c A \ 0'^ /^ A / SIBERIA CANADA AND WASHINGTON '> / / ■ \ ^^^/.^^ / SHOWING 1 /^~^^ RoTite of the Biological Survery Expedition 1899 -— / Scole \ / TT||] «, o ». ~o 3o=Mil„s I l5*' "•"" "*' '"' ,3S" l34" .so- 126- U. S. DEPARTMENT OF AGRICULTURE DIVISION OF BIOLOGICAL SURVEY NORTH AMERICAN FAUNA No. 19 [Actual date of publication, October <;, 1900] RESULTS OF A BIOLOGICAL RECONNOISSANCE OF THE YUKON RIVER REGION General Account of the Region Annotated List of Mammals By Wilfred II. Osgood Annotated List of Birds By Louis B. Bishop, M. D. Prepared under the direction of Dr. C. HART MERRIAM CHIEF OF DIVISION OF BIOLOGICAL SURVEY WASHINGTON GOVERNMENT PRINTING OFFICE 1 9 0 0 LETTER OK TRANSMITTAL. U. S. Department of Agriculture, Washington, D. C, July 28, 1900. Sik: I have the honor to transmit for publication, as No. 19 of North American Fauna, a report entitled ' Results of a Biologieal Reconnois- sance of the Yukon River Region,' by Wilfred H. Osgood and Louis B. Bishop. Under instructions dated May 11, 1899, Wilfred.' H. Osgood, an assistant in the Biological Survey, proceeded-, to. Skagway, Alaska, and thence over White Pass to the headwaters of the Yukon and down the entire length of the Yukon River to St. Michael. He was accom- panied by Dr. Louis B. Bishop, of New Haven, as volunteer assistant; Dr. Bishop has prepared the report on the birds observed during the trip. These are the first investigations of the kind undertaken on the Upper Yukon, and the results herewith presented will be found to contain many important facts concerning the distribution of mammals, birds, and trees in this region. Respectfully, Hon. James Wilson, Secretary of ^ Vgricultwe. C. Hart Merriam, Chief, Biological Survey. CONTENTS. Page. General account of the region, by Wilfred II. Osgood 7 Itinerary 7 Faunal districts 8 Lynn Canal district 8 White Pass district 9 Canadian Yukon district — Lake subdivision 10 River subdivision 12 Hudsonian Yukon district 13 Alaska Tundra district 15 Summary of faunal districts 16 Previous work 18 New species 19 Mammals of the Yukon region, by Wilfred II. Osgood 21 Introduction 21 List of species and subspecies 22 Annotated list of species 22 Birds of the Yukon region, with notes on other species, by Louis B. Bishop.. 47 Introduction 47 Classified lists i >f species 51 Annotated list of species 57 5 ILLUSTRATIONS. Facing page. Plate I. Map of Alaska Frontispiece IT. Fig. 1. — Summit of White Pass. Fig. 2. — Canadian police station at Caribou Crossing 10 III. Fig. 1. — Cliffs on east side of Lake Lebarge. Fig. 2. — Yukon River 50 miles below Fort Selkirkc 12 IV. Fig. 1. — Nests of red squirrels in spruce thicket. Fig. 2. — Burrows made by red squirrels in loose scales stripped from spruce cones 26 V. Skulls of Sciurus vancouvereiixix, Sriurus hudsonicus pelulans, Neo- toma cinerea drummondi, and Ncoloma saxamans 34 VI. Skulls of Lutreola vison energumenos, Lutreola v. ingem, Fiber zibeth- icus, and Fiber sjxdulatus 42 VII. Skulls of Mustela americana brumalis, Mustela americava actuosa, and Mustela americana 44 6 No. 19. NORTH AMERICAN FAUNA. October, 1900. RESULTS OF A BIOLOGICAL RECONNOISSANCE OF THE YUKON RIVER REGION. GENERAL ACCOUNT OF THE REGION. By Wilfred H. Osgood. Nowhere else in North America is such a vast extent of boreal coun- try so easily accessible as along- the Yukon. The navigable waters of the river begin at Lake Bennett, only 35 miles from the port of Skagway. on the coast of southeast Alaska, and with but one short interruption, extend northward as far as the Arctic Circle and then westward to Bering Sea; in all, a distance of more than 1,800 miles. The recent developments resulting from the discovery of gold in this region include a modern railroad from Skagway to Bennett and a tram- car service around the dangerous White Horse Rapids. The chief obstacles to ready access to the territory have thus been removed, and an opportunity is afforded for obtaining specimens and information from a region much of which was previously unknown to naturalists. Accordingly, with Dr. Louis B. Bishop as voluntary companion and A. G. Maddren as assistant, I was detailed to make a hasty biological reconnoissance of this region during the summer of 1899. ITINERARY. After outfitting at Seattle, Wash., we sailed via the Inside Passage direct to Skagway, Alaska, where we arrived on May 30. From Skagway we worked slowly over White Pass and down to Lake Ben- nett, at the head of navigation on the Yukon. Here we embarked in a small flat-bottomed boat suited to our needs and sailed down the series of lakes that follow one another for nearly 200 miles. From the lakes we passed into Thirty-Mile River, thence into Lewes River, and finally into the Yukon proper, stopping frequently and making collections at favorable points. With the aid of the swift, even cur- 7 8 NORTH AMERICAN FAUNA. [no. 19. rent we were able to make eas}r and rapid progress. Thus we con- tinued until an unfortunate capsize between Fort Yukon and Fort Hamlin prevented further detailed work on the river, and we were obliged to proceed direct to St. Michael, where a month was spent in collecting on the coast and tundra. Finally, late in September, our work was brought to a close by the approach of the long arctic winter. We returned to Seattle on the U. S. revenue cutter Corwin, which stopped on her way for a few hours at St. George Island and at Unalaska, at each of which places we collected a few birds. A relatively large part of our time was spent in the White Pass region and about the headwaters of the Yukon, as this was an abso- lutely virgin field, whereas part of the lower river had been previously visited by naturalists. We were unable to do any collecting in the mountains which lie back from the river, owing to the great distance to be covered and the shortness of the season. Legions of mosquitoes were attendant upon us almost constantly. At first they seemed posi- tively unbearable and wrere a real hindrance to the work, but we grad- ually became accustomed to them, and by the use of gloves, head nets, and canopies to sleep under, managed to exist in comparative comfort. Aside from insect pests, however, outdoor life on the Yukon in June and July is very enjoyable; good camping places are abundant, and the weather is mild and beautiful. During the latter part of August and in September strong winds sweep up the river and frequent rains occur. FAUNAL TOSTRICTS. The country traversed may be divided for convenience into five dis- tricts: (1) The Lynn Canal district, (2) the White Pass district, (3) the Canadian Yukon district, (±) the Hudsonian Yukon district, and (5) the Alaska Tundra district. These districts arc limited in a general way by their respective life zones, but the}7 are not of equal extent or importance, and the names applied to them are used not to specially designate restricted parts of zones alread}7 recognized, but purely as a matter of convenience. They are longitudinal districts — that is, they are very much longer than wide, and each is merely a nar- row tract covered by our route through some larger faunal region. Lynn Canal district. — Skagway and the country bordering L}rnn Canal are in the northern part of that faunal area which Nelson has called the 'Sitkan district' and which has often been included in the Northwest Coast district. The trees and shrubs are much the same as those at Juneau, Wrangell, and other points farther south, but the vege- tation is not quite so dense and luxuriant. The shores of Lynn Canal are steep, rocky, and comparatively sparsely timbered, but in some places, as at Haines, low, swampy ground and heavy saturated forests are found. At Skagway, poplars {Popukis tntnuloides and Popultis OCT.,1900.] FAUNAL DISTRICTS. 9 balsamifera) are very common; they share the river bottom with wil- lows and extend well up the steep canyon sides, where they occupy large areas adjacent to the pines, firs, and spruces. Skagway is surrounded by high mountains, and its fauna is limited chiefly b}r altitude. Glacier Station, 14 miles distant, and about 1,900 feet higher, is near the boundary between the Lynn Canal and White Pass districts. The station is situated on the side of a wooded gulch through which a fork of Skagway River Hows. The immediate vicinity is similar to the country about Skagway, but shows the influence, of the Hudsonian /one of the White Pass district, which begins only a short distance beyond. On either side of the gulch are glaciated granite dirt's supporting an irregularly distributed vegetation, chiefly groves of poplars and dense thickets of alders, while in the bottom of the gulch conifers are the prevailing trees. The most common trees and shrubs are lodgepole pines (Pi/ius murrayana), alpine firs {Abies lasiocarpa), tidewater spruces {Picea sitchensis), poplars or aspens (Poj'uhis tremuloides and Pqpidus balsamifera) , alders {Akius sinuata), dwarf birches {Betula (/landulom), currants {Ribes laxijloruiit), and huckleberries ( Vaccmium oral! folium). The black crowberry {Envpe- t I'n in nigrum) and several other heather-like plants occur in the gulch but are more common higher up. Along the trickling streams are many ferns and mosses, as well as occasional patches of the lichen known as 'reindeer moss.' Among the mammals of this region are the Streator shrew {Sorex p. streatori), the Bangs white-footed mouse {Peromyscus oreas), the Dawson red-backed mouse (Evotomys dawsoni), the long-tailed vole {Microtvs inordax), and the red squirrel {Sciurus h. petulans). Characteristic birds are the sooty song sparrow {MeJo- sjriza m. rufina), the Townsend fox sparrow {Passerella i. townsendi), the Oregon snowbird {Junco h. oregonus), and the varied thrush (JPes- 'perociclilii 1 1 nvid). White Pass district. — The summits of the mountains that rise directly east of Skagwa}' are covered with glaciers and perpetual snow, which feed numerous streams that flow down between mass- ive walls of granite. The sides of the wider canyons have been smoothed and scored by glaciation, and the smaller and more recent ones are but jagged rock-bound chasms. These unfavorable conditions cause a rapid change in the character of the plant and animal life, and from Glacier to the summit of White Pass the zones are Hudsonian and Arctic-alpine. A few hundred feet above Glacier the trees become smaller and more scattered, and at Summit only the alpine juniper (Juniperus nana), the bearberry (Arctostajjhylos uvaursi), and depau- perate alpine hemlocks (Tsuya merteiisiana) occur. Heathers and mosses prevail and large areas of reindeer moss are conspicuous. For some distance on the summit of White Pass (Plate II, fig. 1) the elevation and physiography are much the same; the country is slightly 10 NORTH AMERICAN FAUNA. [no. 19. rolling and consists entirely of granite rock, about which cling many mosses and heathers, while small alpine junipers and hemlocks struggle for existence in favorable places. The breeding birds found with these Hudsonian plants were ptarmigan (Lagopus rujpestrvs and L. leucurus), pipits (Anthus pensil/vanicus), rosy finches {Leucosticte t. littoralis), and golden -crowned sparrows (Zonotrichiacoronata). Char- acteristic mammals noted were pikas (Ochotona collaris), hoary mar- mots (Arctomys caligatus), and mountain goats (Oreamnos montanus). Canadian Yukon district. — Lake subdivision: On the north side of the divide the hemlocks are soon replaced by pines and spruces, and in the vicinity of Shallow Lake the boundary of the interior fauna and flora is reached. The change is complete at Log Cabin, British Colum- bia, nine miles from the head of Lake Bennett, where the characteristic features of the Canadian zone are again established and the general aspect of the country is very different. The most abundant tree is the white spruce (Picea canadensis), and among shrubs seen for the first time the buffalo berry (Lepargyrcea canadensis) is very common. Birds marking a change of fauna are the slate-colored junco (Junco hyemalis), the Alaska jay (Perisoreus canadensis fumifrons), the inter- mediate sparrow (Zonotrichia 1. gambeli), and the black-poll warbler (Dendroica striata). A new chipmunk (Eutaniias caniceps) is very conspicuous. At the head of Lake Bennett another change occurs; the country becomes more arid and rocky and there is a tinge of Hudsonian. Lake Bennett is a long, narrow sheet of water inclosed by high granite cliffs, the sides of which are often so steep as to be unfavorable for plant and animal life, and whose summits are doubtless similar to White Pass in fauna and flora. Cold winds sweep down the lake much of the time, and cool shadows envelop the east side most of the fore- noon and the west side most of the afternoon, so that opportunity for warmth by direct sunlight is limited. Hence there is quite a strong Hudsonian element about the lake. Among the plants1 collected here are the pale dwarf laurel (Kalmia glauca), the Greek valerian (Polemonium humile), the forget-me-not (Myosotis sylvatica aJpestris), the alpine juniper (Juniperus nana), the bush cranberry {Viburnum puuciflorum), the dwarf birch {Betula glanchtlosa), the bearberry (Arctostaphylos uvaursi), the buffalo berry (Lepargyrcea canadensis), the shadbush (Amdanehicr aini folia), the Labrador tea (Ledum grm?i- landicum), and the black crowberry (Empetruin nigrum). Where trickling streams come down to the lake alder thickets abound, and along terraces of rock clumps of pines and spruces as well as poplars tind support. Among Hudsonian mammals were found pikas (■Ocho- tona collaris), hoary marmots (Arctomys caligai/m), and Pall sheep 1 Identified by F. V. Coville, chief botanist, U. S. Department of Agriculture. North American Fauna, No. 19. Plate II. Fig. 1.— Summit of White Pass. A'. Irf-j* Fig. 2.— Canadian Police Station at Caribou Crossing. oct., 1900.] FAUNAL DISTRICTS. 11 (Ovis daffli). Although the lake widens slightly at its lower end, its outlet is a narrow stream about '2 miles long, called Caribou Cross- ing (Plate II, fig. 2), on the north side of which is an open, grassy .swamp bordered by willow thickets. This low country, though very limited and not extending to the next lake, affords a breeding place for a few mammals and birds not found about Bennett. Lake Tagish, which receives the waters of Bennett through Caribou Crossing, is like Bennett in character, though not so closely walled, and is characterized by practically the same plants and animals. The surrounding mountains are covered with dense forests, which in many places are almost impenetrable. Connected with Lake Tagish by a short, narrow stream, known as Six-Mile River, is Lake Marsh, a long shallow lake on each side of which extends low country, with rolling hills farther back. The valley widens here quite appreciably, and the open country is like that at Caribou Crossing. On the east side are sedgy bogs surrounded by willow thickets, and in many places a wide margin of beautiful green sedge meets the edge of the water. Rocky shores are found at some points on the northwest side, but in general the country is low and moist, in marked contrast to that about Bennett and Tagish. The mountain animals of those lakes are of course absent, and the bird life is also somewhat different. Fifty -Mile River, into which the Yukon waters proceed from Lake Marsh, is rather narrow, and for a short distance at White Horse Rap- ids very swift. Its banks are chiefly abrupt bluffs of sandy clay (from 50 to 100 feet high) but at Miles Canyon it is confined between walls of basalt. Below the rapids the stream widens somewhat and the high banks become less frequent, often being replaced by low ones thickly grown with willows. The timber is somewhat scattered, and on the rolling hills back from the river bare granite spaces may be frequently seen. At the head of Fifty-Mile River, we first met with birch trees (Bctula papyrifera ?), and from that time on they were seen dail}r. They do not grow to large size — trees more than 8 inches in diame- ter were seldom seen. Several small streams flow into Fiftj^-Mile River, which favor the growth of thickets of alders along their banks and large clumps of willows about their mouths. The little boreal sagebrush {Artemisia frigida) grows abundantly on the warm exposed slopes that occasionally alternate with the sandy bluffs. Lodgepole pines are also abundant and frequently occupy large areas to the exclu- sion of all other trees. Spruce and poplar, however, are still the strong- est elements in the forest. From Fifty-Mile River we enter Lake Lebarge, the last and largest of the lakes. All about its clear, cold waters are low granite moun- tains (Plate III, fig. 1). Occasionally patches of heavy spruce forest are found near the water, but in many places cliffs rise abruptly from 12 NORTH AMERICAN FAUNA. [no.19. the water's edge, and the timber is very sparsely sprinkled over them. The rocks found here and a few in Thirty-Mile River are the last we saw showing signs of glaciation. Lake Lebarge is quite different from Lake Marsh, and is more similar to Lakes Tagish and Bennett, though all the Hudsonian elements of these are not present. River subdivision: This area includes the section from the foot of Lake Lebarge to the mouth of the Pell y River at Fort Selkirk. There is very little variety in the character of the country between these points. Thirty -Mile River, which proceeds from Lake Lebarge, is a swift, narrow stream, and at low water is barely navigable for small steamers. A conspicuous feature of its banks, which are cut abruptly like those of Fiftv-Mile River, is a narrow ribbon-like stratum of vol- canic ash about 6 inches below the surface that maj' be seen wherever the bank is exposed. On the mountains a short distance from the river the forest of spruce is heavier and purer than any previously noted. The poplars and willows are more confined to the brink of the river, and the birches are scattered. Thirty -Mile River is simply that por- tion of the Yukon between Lake Lebarge and the mouth of the Hoota- linqua or Teslin River. The stream is greatly augmented b}r the waters of the Hootalinqua, and from this point on to Fort Selkirk is known as the Lewes River. Below the Hootalinqua it cuts through the Semenow Hills, for the most part abrupt, rocky, and rather barren mountains from 2,000 to 3,000 feet high. Near their bases and at the water's edge are forested areas, but the exposed hillsides are covered with boreal sagebrush (Artemisia frig idd), with here and there a pros- trate juniper or a small clump of spruces. The river now widens rapidly, receiving in succession the waters of the Big Salmon, the Little Salmon, and the Nordenskiold. The rolling hills are sometimes a mile or several miles from the river bank, with low willow swamps intervening. Islands varying from 1 to 100 acres in extent and covered with luxuriant vegetation are abundant. The distribution of trees on the small, regular-shaped islands is very uniform, the different kinds being grouped in concentric belts. Alders generally form the outer margin; next come the willows; next the poplars, rising somewhat higher; and finally the dark-green spruces, which occupy the central area. The whole effect is quite picturesque. On the larger islands the spruces are larger, and usual l}r predominate to such an extent that almost everything else should be classed as under- growth (including trees and shrubs belonging to the genera Alnus, Salix, Populus, Zepargyrwa, Cormis, Viburnum, Rosa, Ledum, Yac- cinium, Ribes, and others). Lodgepole pines still occur, though unlike the spruces they nowhere form continuous forest and disappear entirely a short distance beyond Fort Selkirk. The Canadian Yukon district as a whole is very well marked. Char- acteristic mammals are the gray-headed chipmunk [Eutdm ins com iceps), North Amencan Fauna, No. 19 Plate III. Fig. 1.— Cliffs on East Side of Lake Lebarge. Fig. 2.— Yukon River, 50 Miles below Fort Selkirk. oct., 1900.] FAUNAL DISTRICTS. 13 the Bennett ground squirrel (/Sjpt rmojphihts < ///j>< tra plesius), the North* era bushy-tailed rat (Neotoma saxa/ma/ns), white-footed mice {Peromys- cus areas and Perom/yscus mandculatw cvrcticus), and the varying hare {Lepus salii ns). All of these species and three of the genera, Eutwm ias, Neotoma, land Perom/yscus, find their northern limits in this district. Among birds that are known from the Yukon only in this district may be mentioned the sparrow hawk (JFaZco sparverms), the screech owl {Megascops asio h rmicoUif), the night hawk (Chordeiles virgwdan/us), the tree swallow (Tachycmeta fticolor), the Tennessee warbler (JTehnm- thophUajp( regrina), the pileolated warbler ( Wilsonia pimllajpileolata), and the mountain bluebird (Sialia arctica). Of these, GhoroZeUes is perhaps the most noteworthy, as it is decidedly a southern genus. It is very common, and was seen nightly from Caribou Crossing to Rink Rapids, but after we had passed that point it disappeared. Its range in this region as observed by us is probably accurate and corresponds with the limits of the district. Among trees, the lodgepole pine [Pinus m/urrayana) is common throughout the district, but does not extend beyond it. Hudson ian Yukon district. — This district, as here considered, includes all of the Yukon region from Fort Selkirk to the limit of trees. The Lewes River is joined at Fort Selkirk by the Felly, after which the increased volume of water flows on between heavily forested slopes and jutting cliffs (Flate III, fig. 2), which replace the sandy banks of the upper river. From the mouth of the Selwyn northward the topography of the river banks is but slightly different. The num- ber of poplars in the forest is much increased; the spruces are corre- spondingly decreased not only in number but also in size; Avhile the birches about hold their own, and the pines are not present at all, having disappeared between Fort Selkirk and the mouth of the Selwyn River. As we approach Dawson spruces become dwarfed and entirely subordinate to the poplars, which crowd their bushy tops together for miles and miles. The spruces are in the gulches and in small clumps elsewhere, and a few are scattered about, their dark-green spike tops showing off well against the billowy mass of the lighter foliage of poplar and birch. The undergrowth remains much the same, and deep moss covers the ground and rocks. In damp sandy places along shore and on islands occasionally overflowed a bright-green scouring rush. {Equi 'fsetum) grows so abundantly as to be a characteristic plant. The alpine juniper (Juniperus nana) is found occasionally on hill- sides not too thickly grown with poplars, and on the more open hill- sides the landscape is brightened by masses of fiieweed (Chammneriori arujusti folium), for even here forest fires are not a novelty. Two more large rivers, the AVhite and the Stewart, empty into the Yukon in this vicinity. About the mouths of these and other tributaries is more or less low country covered with willows. Islands become 14 NORTH AMERICAN FAUNA. [no. 19. .still more numerous and larger, and have a forest growth that is more uniform in character than that of the river banks. High cliffs over- hanging the river are of frequent occurrence. From Dawson to the Alaska boundary and thence to Circle the country is about the same. For a long distance in the vicinity of the boundary a range of high mountains is visible to the northward on the right bank of the river. The low, rolling hills which border the upper river do not quite reach Circle, but are replaced by a broad, flat country known as the 'Yukon Flats,' which extends from near Circle to Fort Hamlin, a distance of about 200 miles. Through the ' Flats ' the course of the river breaks up into a great many channels, and the islands still further increase in size and number. These are composed of sand and silt, in which poplars thrive better than spruces, though the latter are by no means eliminated. A wild rose [Rosa cinnamomea?) is the most abundant shrubby plant, and on the ground below it the Equisetum is rampant. The larger islands are identical in character with the mainland, and on them the spruces form quite a heavy forest, with deep moss beneath. At Fort Hamlin the river narrows again and flows between rolling wooded hills, which are similar to many farther up the river. Small streams enter the main river frequently, and the timber is much the same; poplars, alders, and birches cover the hills in dense thickets, through which spruces are sprinkled. Alders are more numerous than before. The hills vary in height from 500 to 3,000 feet, and the highest have a distinct timberline at about 2,000 feet. At the mouth of the Tanana the hills become smaller and the river very much wider. Here, at Fort Gibbon, Dr. Bishop found the larch {Larix americana) quite abundant. This was the only point at which it was seen by any of our party. The Lower Yukon beyond the Tanana is very uniform in character. The banks are low and rolling and overgrown with willows and alders; farther back are higher hills covered with poplars and birch; occasion- ally the summits of a few hills higher than the rest are devoid of trees. On the sandy islands the willow thickets are impenetrable, and where a cut bank exposes a section of them their slender perpendicu- lar trunks stand so closely as to present a solid front like a thick hedge or canebrake. Thus it continues until the limit of timber is reached at Andraefski, 90 miles above the mouth of the Yukon. This district as a whole is characterized by absence of southern plants and animals. Among migratory birds a few have their center of abundance farther south, but all the mammals are northern forms, and nearly all belong to genera of circumpolar distribution.1 Plant life, though quite luxuriant, is made up of only a small number of hardy species. Characteristic mammals are the Fort Yukon ground 'The only exceptions are Synaptomys, Fiber, aud Erethizon. oct., 1900.] FAUNAL DISTRICTS. 15 squirrel (SpemwphUvs osgoodi), Dawson red-backed mouse (Evotomys dawsoni), yellow-cheeked vole (Mtcrohts xanthognathus), Yukon lem- ming (Lemm usyuhom mis), Dall varying hare (Lepus american us dalli), and tundra weasel (Patori/us arcticus). Of the birds, the most char- acteristic are the duck hawk (Falco peregrvn/us anatum), pigeon hawk [Falco columbarius), Alaska longspur ( ( 'alcariu* /ajponicus alascensis), hoary redpoll (Acanthis hornemanni exilipes), fox sparrow (Passerella iliaca), Bohemian waxwing (Ampelis garrukts), and wheatear (Saxlcola (VnantJu). Alaska Tundra district. — The Yukon from Andraefski to the coast of Norton Sound is bounded on both sides by typical tundra. The country is low and gently undulating, and its surface a short distance away appears to be thickly carpeted with grass. That this is not the real condition a short walk ashore soon demonstrates; but the delusion is so complete that were it not for the presence of the great river one might fancy himself looking out over the undulating plains at the eastern base of the Rocky Mountains in the western United States. The flora of the tundra, though devoid of trees deserving of the name, is Jound on careful examination to be quite varied. Besides the numerous mosses and heathers and many small berry-bearing plants are dwarf willows, birches, and alders. The alders attain the greatest size, but are usually found in isolated clumps in favorable spots, where they often grow from 6 to 8 feet high. The ground is frozen a few inches below the surface, and the hea\w, spongelike cover- ing of vegetation is kept constantly saturated. Occasional high bluffs on the coast in exposed situations are bleak and bare, but besides these there is scarcely a spot not covered with lowr, matted vegetation. Numerous small ponds are irregularly distributed over the tundra, and around them the vegetation is ranker than elsewhere. Broken lava borders the shores of St. Michael Island, and small moss-covered heaps of it, which form practically the only solid footing on the island, are scattered about over the tundra. Characteristic mammals at St. Michael are the Hall Island fox (Vidpes hallensis), Nelson vole (Microtus operarius), tundra red-backed mouse (Evotomys dawsoni alascensis), Nelson pied lemming (Dicro- stonyx nelsoni), Alaska lemming (Lemm us alascensis), and Alaska Arctic hare (Lepus othus). Land birds known to breed are the hoary redpoll (Acanthis hornemanni exilipes), common redpoll (Acanthis linaria), Alaska longspur (Calcarius lapponicus alascensis-), western tree spar- row (Spisella monticola ochracea), golden-crowned sparrow (Zonotrichia coronata), and Siberian yellow wagtail (Biulijt. es fiavus leucostriatus). Common tundra plants1 are Cassiopea tetragona, Andromeda polifolia, Vacdnium vitisideea, Mairania alpina, Ledum pdl/ustre, Artemisia 1 Nelson, Report upon Natural History Collections in Alaska, 30, 1887. 16 NORTH AMERICAN FAUNA. [no. 19, arctica, Bubus chamcemorus, Rulnis arctieus, Betula nana, Alnus sinuata, ChamcBcisttcs procitmbms, and Tmsilago frig Ida. SUMMARY OF FAUNAL DISTRICTS. All the country here considered is in the boreal zones, the Tundra district and a small part of the White Pass district belonging to the Arctic subdivison, and the Yukon Valley principally to the Hud- sonian, though it has also a well-marked Canadian section. Birds are comparatively rare in all the interior region, and it is difficult to determine the exact range of many species. Some were seen but once or twice; others appeared sporadically at rather long intervals; while still others that are known from the region were not seen at all; so it is hardly safe, in making generalizations, to rely too much on the ranges observed by us. The distribution of trees and shrubby plants and of many of the mammals, however, could be determined with much greater accuracy and constitute reliable guides in fixing the limits of the districts. These districts are in general the same as those recognized by Nelson, but with more definite and somewhat modified limits. Names slightly different from those he used are adopted here in order to agree with the commonly accepted names of the primary zones of North America. Thus the part of his 'Alas- kan-Canadian' district here considered is called the 'Hudsonian Yukon' district, since it lies entirely within the transcontinental Hud- sonian zone. Owing to iluviatile conditions, the boundaries of the Yukon districts doubtless do not agree in latitude with those which might be made away from the rivers. The zones which we successively traversed in going from Skagway to St. Michael via White Pass and the Yukon are: (1) Canadian; (2) Hudsonian; (3) Arctic- Alpine; (4) Canadian; (5) Hudsonian, and (6) Arctic. The Lynn Canal district is in the Canadian zone, but it has some slight peculiarities such as are to be expected in a coast district. Though it does not have the Hudsonian animals of the northern coast, it lacks several of the typical Canadian forms of the coast farther south.1 It is really near the northern limit of the Canadian zone on JThe coast of Alaska south of the peninsula, or what has been known as the 'Sit- kan district,' may be easily divided into two districts corresponding to the Canadian and Hudsonian zones. Lynn Canal is situated near the boundary between these districts. Among Canadian mammals which have their center of abundance in the restricted Sitkan district on the coast south of Lynn Canal are: Odocoileus sitkensis, Sciurus vancouverensis, Peromyscus macrorhinus, Evotomys wrangdi, Microius macrurus, Synaptomys ivrangeli, Zapus saltator, and Myalls alascensis. Among the Hudsonian forms found on the coast only north of Lynn Canal are: Rangifer sp., Ovis dalli, Sciurus hudsonicus, Spermophilus e. plesius, Zapus h. alascensis, Ochotona collaris, and Myotis lucifugus. Among trees which find their northern limit in the vicinity of Lynn Canal are: Thuja pllcata, Abies lasiocarpa, and Firms murrayana. The northern district from Lynn Canal to Kadiak is so similar to the great interior Hudsonian region that it hardly merits recognition as a distinct district, but it certainly should not be included in the Sitkan district. oct.,1900.] SUMMARY OF FAUNAL DISTRICTS. 17 the Pacific coast. The occurrence at Skagway of mammals of the interior, such as Microtus mordax,1 Eootomys dawsoni, and Perom/yscus areas, seems to show an approach to the condition farther north where the coast and interior forms are practically the same. The Canadian zone of the Lynn Canal district gives way to the Hudsonian and Arctic- Alpine in the White Pass district. The character of this dis- trict is essentially the same as that of other mountain regions in western North America. This is well indicated by the fact that its mammals include the hoary marmot (Arctomys caligatus), the Alaska pika (Ochotona coUaris), and the mountain goat (Oreamnos montanus), and its birds the ptarmigan (Lagopus I ucurus and L. rwpestris), the pipit {Anthuspt nsilvcmicus), and the rosy finch (Leucostictet. littoralis). The Canadian Yukon district from Bennett to Fort Selkirk merely represents the extent to which our route entered the extreme northern part of the Canadian zone; that is, its limits are those of the section that our route made across the end of a narrow tongue which extends northward from the great areas occupied by the Canadian zone farther south. Owing to its being so near the border of the Hudsonian zone, its character is not purely Canadian, but the presence of such forms as chipmunks (JEhdamias) and white-footed mice (Peromyscus) among- mammals, night hawks (Ghordeiles) among birds, and lodgepole pines (Pin us murrayana) among trees, makes it evident that the Canadian element is very strong. The Hudsonian Yukon district represents the complete section which the Yukon River makes through the great northern forest belt of the Hudsonian zone. This belt corresponds to the Alaskan-Canadian district outlined by Nelson. It is bounded on the south by the Canadian zone and the extreme northern limit of southern forms, and on the north by the treeless tundra. On the west it probably reaches and includes the coast from Kadiak to Lynn Canal; on the east its limits are unknown. The Alaska Tundra district defines itself. Its character is the combined result of latitude and rigorous coast climate. Our experience in this treeless district was limited to St. Michael Island and the ninety miles between Andraefski and the mouth of the Yukon. The animals of this region are not all abso- lutely confined to it, many of them ranging some distance into the forest belt. Small mammals, such as the Nelson vole (Microtus ojpe- rarius), occupy so-called 'islands' — local spots offering what are prac- tical ly tundra conditions — as far within the forest belt as Circle. JThe case of Microtus mordax at the head of Lynn Canal is particularly interesting, since a closely related form, M. macrurux, has been found at Glacier Bay on the north side of the mouth of the canal and also at Juneau on the south side, and would therefore be expected at Skagway, which is halfway between. Assuming that macrurua has been dispersed from the south northward on the coast, it seems that it did not reach Glacier Bay by way of the present mainland, otherwise it would be found at Skagway. M. mordax doubtless reached Skagway from the interior. 4494— No. 19 2 18 NORTH AMERICAN FAUNA. [no. 19. The fauna of the Yukon basin as a whole is composed of two prin- cipal elements, one containing absolutely circumpolar forms, evidently derived from the north, the other containing forms which have their center of abundance farther south. This is particularly true of the mammals. Among the genera belonging to the north may be men- tioned Rangijer, Evotomys, Lemmus, and Dicrostonyx, all of which are circumpolar in distribution; those from farther south are Alces, Schiropterus, Eutamias, Peromyscm, JVeotoma, Fiber, and Synaptomys. With the exception of alpine species and a few wide-ranging forms, chiefly carnivores, the variations of which are not sufficiently known to be of use in defining faunal regions, no species of mammals are common to the Yukon region and the Sitkan coast district. From this it seems that all the southern forms which reach the Yukon region have been derived from the interior rather than from the coast. This is also true of the trees and to a great degree of the birds. But, on the other hand, some species of land birds are common to the lower Yukon and the Sitkan district while a large intervening area in the interior is uninhabited by them.1 Selasphorus rufus, Dendroica townsendi, and Ilylocichla aonalaschkcB were found on both sides of White Pass, but only rarely and for a very short distance on the interior side. PREVIOUS WORK. Our knowledge of the natural history of the Yukon region has been derived chiefly from two sources — the members of the Russo-Ameri- can Telegraph Expedition and the Signal Service officers formerly stationed at St. Michael. The first information was gathered by the scientific corps of the Telegraph Expedition of 1865 to 1868. Promi- nent among the members of this corps were Robert Kennicott, Wil- liam H. Dall, and J. T. Rothrock, all of whom secured valuable speci- mens and information. The notes of Kennicott were not published, owing to his untimely death at Nulato, May 13, 1866, but numerous specimens, particularly from the vicinity of Fort Yukon, are now in the National Museum, a monument to his faithful pioneer work. Among the numerous papers on various subjects relating to Alaska published by Dail are lists of birds and mammals.2 A list of plants including some records from Fort Yukon was published by Rothrock.3 1One of these species is the varied thrush (Hesperotichla nievia) which was found in the Lynn Canal district, but not in any numbers in the Yukon Valley above Daw- son. Below Dawson it is quite common along the Yukon and undoubtedly breeds there. It is well known to range along the Pacific coast to Kadiak, and thence to the shores of Kotzebue Sound and up the Kowak River. Its absence on the Upper Yukon and its occurrence all along the coast make it extremely probable that in reaching the Yukon its course of migration is up the river from its mouth. 2 List of Birds of Alaska, by W. H. Dall and II. M. Bannister < Trans. Chicago Acad- emy of Sciences, I, pt, II, 267-310, 1869; also Alaska and its Resources, by W. H. Dall, Boston, Lee & Shepard, 1870. 3 Sketch of the Flora of Alaska, by J. T. Rothrock < Annual Report Smithsonian Institution, 1867, 433-463. OCT.jim] NEW SPECIES. 19 Iii 1874, with the establishment of a meteorological station at St. Michael, work was begun by Lucien M. Turner. He collected about St. Michael and secured specimens from the fur traders and natives on the Yukon as far up as Fort Yukon. The results of his work were published in the Arctic Series of the Signal Service.1 Turner was relieved in 1877 by E. W. Nelson, who continued to collect specimens until 1881. His work was more extensive than that of Turner, hut was carried out along the same lines. He made several trips up and down the coast from St. Michael, and also worked about the Yukon delta and up the river as far as Anvik. L. N. McQuesten, who conducted a trading post at Fort Reliance, furnished him with numerous specimens and much valuable information. The results of Nelson's work were also published in the Arctic Series of the Signal Service a and form by far the most valuable contribution to the natural history of Alaska yet made. In 1889 an important report :! was pub- lished by Dr. George M. Dawson, covering the region of the sources of the Yukon and down as far as Fort Selkirk. This report contains detailed descriptions of the physical features of the upper river, notes on natural history, particularly on the distribution of trees, and a list of plants by John Macoun. NEW SPECIES. Nine new species and subspecies of mammals are described in the present report. They are as follows: Seiuwpterus yukonensis. Sciurus hudsonicus p>etidans. Eutamias caniceps. SpermophUus empetra plesvus. Neotoma saxamans. Fiber spalulatus. Lejjus saliens. Lutreola vison ingens. Mustela americana actuosa. In the collection of birds, three new forms were found. These have been described by Dr. Bishop4 as follows: ( 'anachites canadensis osgoodi. Contopus richardsoni satunttHx. Sayornis saya yukonensis. 1 Contributions to the Natural History of Alaska, by L. M. Turner, Arctic Series, Signal Service, No. II, Washington, 1886. 2 Eeport upon Natural History Collections made in Alaska, by E. W. Nelson, Arctic Series, Signal Service, No. Ill, Washington, 1887. 2 Geological and Natural History Survey of Canada, Annual Report, III (1887-88), Pt. I, 6 B-277 B, 1889. 4 Auk, XVII, 113-120, April, 1900. 20 NORTH AMERICAN FAUNA. Oo.l9. Besides the above, several new mammals which come within the scope of the present report have been recently described by Dr. Mer- riam.1 These are as follows: SpermophUus osgoodi. Lemmus yukonensis. Lemmus alascensis. Dicrostonyx nelsoni. Erethizon epixanthus myops. Lepus americanus dalli. Lepus othus. Vulpes hallensis. Sorex personatus arcticus. Sorex tundrentis. In addition to these, three new forms recently described by Witmer Stone 2 should be mentioned: Dicrostonyx hudsonius alascensis (equals I). nelsoni Merriam). Putorius rixosus eskimo. Lynx canadensis mollipilosus. 1 Proc. Wash. Acad. Sci., II, 13-30, March 14, 1900. * Proc. Acad. Nat. Sci. Phila., March 24, 1900, 33-49. MAMMALS OF THE YUKON REGION. By Wilfred II. Osgood. INTRODUCTION. The following list, primarily based on collections made during the past 37ear, includes all the known mammals of the Yukon region. Besides the species which belong' strictly to the Yukon, are included those found in the Lynn Canal and White Pass districts and those known to occur at St. Michael. This makes a list containing the majority of the mammals known from Alaska, which is not, however, intended to be comprehensive, but should be considered as supplementary to the list published by Nelson in 188T. As may be seen from the itin- erary, our collections were made during a hasty trip from the coast of southeastern Alaska to the head waters of the Yukon and thence down the river to St. Michael. Good series of all the common small mam- mals were secured, but the larger and rarer species were not often obtained. It was not only difficult to secure specimens of the larger mammals, but it was hard to gain much accurate information in regard to them. Most of the miners we met had been in the country but a short time and their knowledge of animals was limited; natives were seldom met on the upper river and the few that were interviewed seemed disinclined to talk. The fur trade on the Yukon has dwindled to comparatively meager proportions. The Indians still bring a few furs to the traders every }^ear and receive pittances of flour and tea in return; but the trade is apparently very small and were it not for the transportation business which has recently become so important, the large companies would doubtless find it difficult to maintain themselves. In identifying the recently collected specimens and studying their geographical distribution, it has been necessary to refer constantly to the specimens collected by Kennicott, Dall, Nelson, and Turner. Many of these, which are in the National Museum, were found to be in poor condition and required considerable renovating to make them compar- able with modern specimens. For the free use of these specimens I am indebted to Gerrit S. Miller, jr., assistant curator of mammals in the National Museum. I am also indebted to Outram Bangs for the use of specimens, and E. W. Nelson for much valuable informa- tion. The identifications of some of the mammals have been verified 21 22 NORTH AMERICAN FAUNA. [NO. 19. by specialists as follows: The species of Sorex by Dr. C. Hart Merriam; of Microtus by Vernon Bailey, and of Zapus by Edward A. Preble. All measurements are in millimeters. LIST OF SPECIES AND SUBSPECIES. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. Rangifer montanus Seton-Thornpson. Rangifer arcticus (Richardson). Rangifer tarandus (Linnaeus). Alces gigas Miller. Ovis dalli Nelson. Oreamnos montanus (Ord). Sciuropterus yukonensis sp. nov. Sciurus hudsonicus Erxleben. Sciurus hudsonicus petulans subsp. nov. Eutamias caniceps sp. nov. Spermophilus empetra plesius subsp. nov. Spermophilus osgoodi Merriam. Arctomys caligatus Eschscholtz. Castor canadensis Kuhl. Mus decumanus Pallas. Peromyscus oreas Bangs. Peromyscus maniculaius arcticus (Mearns). Neotoma saxamans sp. nov. Evotomys dawsoni Merriam. Evotomys dawsoni alascensis (Miller). Microtus mordax ( Merriam ) . Microtus drummondi (Aud.and Bach. ). Microtus xanthognathus (Leach). Microtus opcrarius (Nelson ) . Fiber spatulatus sp. nov. Synaptomys dalli Merriam. Lemmus yukonensis Merriam. 28. Lemmus alascensis Merriam. 29. Dicrostonyx nelsoni Merriam. 30. Zapus hudsonius alascensis Merriam. 31. Erethizon epixanthus myojis Merriam. 32. Ochotona collaris (Nelson) 33. Lepus saliens sp. nov: 34. Lepus americanus dalli Merriam. 35. Lepus othus Merriam. 36. Lynx canadensis moUipilosus Stone. 37. Canis occidentalis Richardson. 38. f Vulpes fulvus (Desmarest). 39. Vulpes hallensis Merriam. 40. Ursus americanus Pallas. 41. Ursus horribilis alascensis Merriam. 42. Lutra canadensis (Schreber) . 43. Lutreola vison ingens subsp. nov. 44. Putorius arcticus Merriam. 45. Putorius cieognani alascensis (Mer- riam.) 46. Putorius rixosus eskimo Stone. 47. Mustela, americana actuosa subsp. nov. 48. Mustela pennanti Erxleben. 49. Qulo luscus (Linnaeus). 50. Sorex personatus streatori Merriam. 51. Sorex ])ersonatus arcticus Merriam. 52. Sorex obscurus Merriam. 53. Sorex tundrensis Merriam. 54. Myotis lucifugus (Le Conte). ANNOTATED LIST OF SPECIES. Rangifer montanus Seton-Thompson. Mountain Caribou. Rangifer montanus Seton-Thompson, Ottawa Naturalist, XIII, No. 5, 6, Aug. 11, 1899; Allen, Bull. Am. Mus. Nat. Hist., N. Y., XIII, 1-18, April 3, 1900. This large woodland caribou is reported as quite common in northern British Columbia about the head waters of the Yukon and for an indefinite distance northward. It does not occur on the coast south of Cook Inlet, but is reported from many points immediately beyond the summit of the coast mountains. It prefers the higher ground in summer and is not found along river bottoms like the moose, for which reason few are killed by parties descending the river. Its flesh is smoked and dried by the Indians for winter food, and when so cured is preferred to all other meat of the country. The hides, like those of the moose, serve the natives for various articles of clothing and a"e utilized especially for sleeping robes. oct.. 1900.] MAMMALS OF THE YUKON REGION. 23 Rangifer arcticus (Richardson). Barren Ground Caribou. The barren ground caribou ranges over nearly all of extreme north- ern North America from northwestern Labrador to the Aleutian Islands. It was formerly abundant over this great territory, but is now quite rare. Even at the time of Nelson's work in 1877 it had become comparatively uncommon, though it was once common all about Norton Sound and for some distance up the river. The south- ern and interior limits of its range are uncertain. During our stay in St. Michael, I saw half a dozen skins which had been secured near Andraefski, 90 miles above the mouth of the Yukon. There are specimens in the National Museum from Nushagak and llnalakleet, Alaska; and from Rampart House and La Pierre House, Northwest Territory. Rangifer tarandus (Linnaeus). Domesticated Reindeer. During the past few years, as is well known, an effort has been made to introduce domesticated reindeer from Siberia into Alaska. The animals as a rule have been carefully herded, but in a few cases they have had opportunities to stray awajT and run wild. The herd that had perhaps the best chance to stray was one which was brought from Lapland to Haines in 1898, and driven inland over the Dalton trail. A short time after it started several of the animals were seen in the forest near Haines, and one of them was killed. This was the only instance of the kind brought to my attention, but I have no doubt that reindeer have occasionally wandered from the care of the herders at other times and in other places. Alces gigas Miller. Alaska Moose. Alces gigas Miller, Proc. Biol. Soc. Wash., XIII, 57-59, May 29, 1S99. The Alaska moose, as has frequently been stated, is the largest of the deer family in North America. Its distribution along the Yukon extends from Lakes Atlin and Tagish at least to the mouth of the Tanana and probably somewhat farther. Whymper 1 says that it was 'never known as low as Nulato^ even in the time of its greatest abundance. It is evident, however, from the record of Nelson 2 at the Yukon delta, and that of Richardson 3 at the mouth of the Mackenzie, that it does occasionally leave its favorite woodlands of the interior and wander as far as the Arctic coast. At present it is still quite numerous, but is chiefly confined to the small streams tributary to the Yukon. According to reports which came to me it is abundant in the region about the upper waters of the Stewart, Pellv. and MacMillan rivers. Along the great river itself numbers have been killed during 1 Travels in Alaska and on the Yukon, 245, 1869. 2 Report upon Natural History Collections in Alaska, 287, 1887. :! Fauna Boreali-Anicrirana, 233, 1829. 24 NORTH AMERICAN FAUNA. [no. 19. the recent influx of prospectors. At the beginning of the Klondike rush, it was not uncommon for a party to secure one or two moose while descending the river, but such is rarely the case at present. Our party failed to see any, though we spent nearly three months in the region; during this time we heard of but two animals being killed, one near the foot of Lake Lebarge and another on upper Charlie Creek, a short distance above Circle; both were secured by Indians. We saw comparative^ few fresh tracks. In winter, moose meat is the staple diet of both Indians and whites and has readily sold in the mining camps at $1 to $2 per pound. Such a price, even in this country of high wages, has been a great incentive to hunting, and many a miner has left his claim to pursue the moose. The hides also are a source of profit, particularly to the Indians, who tan them and make them into mittens and moccasins. What the Indians do not need they sell readily to miners and prospectors. It is dif- ficult to estimate the number of animals that have been killed, but it must be very large, for the demand has been steady and a comparatively large population has been supplied with meat. On one hunt, an ac- count of which has been given by Tappan Adney,1 -i-t moose were killed in about one month, and a single party of Indians was credited with a total of 80 moose and 65 caribou in one winter. Ovis dalli (Nelson). Dall Mountain Sheep. Most of the specimens of the Dall sheep which have reached our museums were secured in the vicinity of Cook Inlet, but the animal occurs in nearly all the high mountains of Alaska, and in the north ranges to the Arctic coast. Since we were at a distance from the mountains during the greater part of our trip, I was unable to secure much information in regard to the distribution of the species. Sheep are said to occur about the West Arm of Lake Bennett, and Windy Arm of Lake Tagish. A prospector with whom I talked at Lake Tagish said he had seen and killed them at both of these places. Lake Bennett is not far from the type locality of 0. stone!, and it is possible that this species occurs there with 0. dalli. Both white and gray sheep are reported, though all are said to be white in winter. I was told that white sheep were killed some years ago on the cliffs about Lake Lebarge, but I failed to find signs of them there. Prospectors at Fort Selkirk say that sheep are always to be found in the mountains along Pelly River, particularly in the MacMillan Mountains '' near the mouth 1 Harper's Magazine, C, 495-507, March, 1900. 2 The sheep from the MacMillan Mountains are said to he the ' hlack sheep,' which name could hardly apply to 0. dalli, hut is the name commonly given to 0. stoneu If stonei really does occur in these mountains the record is a very interesting one, and the locality much farther north than any from which thespecies has been previously recorded. Oct., 1900.] MAMMALS OF THE YUKON REGION. 25 of MacMillan River, and they were also reported from the head waters of the Stewart and from the Tanana Hills. Oreamnos montanus (Orel). Mountain Goat. Goats occur on the high granite cliffs which inclose the upper part of Lynn Canal; they arc also common on the mountains near White Pass and about the rocky walls of Lake Bennett. I was told that they had been killed recently at the upper end of Little Windy Arm on Lake Tagish, but 1 could obtain no reliable report of their occurrence in the interior beyond this point. At Lake Lc) targe they were very doubtfully reported. Their range is known to extend north to White Pass in the coast mountains at least to Copper River,1 but does not reach far into the interior. Hunters from the mountains about the upper filters of the Pelly and Stewart rivers asserted positively that none had been heard of in that region. The station agent at Glacier, near White Pass summit, told me that goats frequently appear on the cliffs within easy view of his house. He also showed me the hide of one- that had been killed near there a short time before our arrival. I made one short trip into these mountains, but failed to see any goats. The character of the cliffs is ideal for them, but they had evidently gone farther back to their summer feeding grounds, as the abundant tracks and dung were a few weeks old. Sciuropterus yukonensis sp. now Yukon Flying Squirrel. Tijpr from Camp Davidson, Yukon River, near Alaska-Canada boundary. No. yff°§, U. S. Nat Mus. Collected December 8, 1890, by R. E. Carson. Characters. — Size largest of North American flying squirrels; tail exceedingly long; color rather dark, underparts suffused with fulvous; skull slightly characterized. ('<>}<>/■. — Top of head, neck, and upperparts to base of tail pale cin- namon or between the wood brown and cinnamon of Ridgway; under- fur bluish black, partially exposed on legs and membranes; underparts dull whitish, irregularly suffused with cinnamon fawn; feet dusky above, lightly edged with creamy white, buffy white below; cheeks and sides of head ashy, lightly mixed with cinnamon; end of nose slightly paler than top of head, not light ashy as in S. sabrmvA/ black eye-ring prominent; tail light fawn below, with a light edging of dusky, becoming broader toward tip; tail above fawn heavily mixed with black, which predominates for terminal fifth. Skull. — Size large, slightly larger than in S. alpmm; audita! bullae larger; width at postorbital constriction greater; molars heavier, particularly the mandibular series. 'H. T. Allen, Science, VII, 57, 1886. 26 NORTH AMERICAN FAUNA. [no. 19. Measurements. — Total length 365; tail 180; hind foot (measured dry) 41. Skull: Occipito nasal length 40; zygomatic breadth 25; postorbital constriction 10. Remarks. — This species is distinguished from both S. sahrinus and 8. alpinus by its large size and very long tail, but it is also very dif- ferent from either in color. It is evidently a very rare squirrel, as the type and one topotype are the only specimens known. A speci- men from Chilkoot Inlet which may possibly be this species has been recorded by Dr. George M. Dawson.1 Camp Davidson is the north- ernmost point at which the genus Sciurojpterus is known to occur. The type and one other specimen were secured by R. E. Carson, who was a member of the boundary survey party of the IT. S. Coast and Geo- detic Survey under J. E. McGrath, in 1890. Dr. W. W. Kingsbury, also a member of the party, writes me as follows in regard to these specimens: I send you the following notes taken from my journal regarding two Flying Squir- rels which were captured by a member of our party while in Alaska, in 1890; their skins were sent to the National Museum at Washington. The female was caught Dec. 8th, 1890, and the male Dec. 9th, 1890. Both squirrels were caught in a trap known as the 'dead fall,' which was set by R. E. Carson for marten. The traps were set in the bed of a frozen stream, where it ran through a clump of spruce trees about one-fourth of a mile back from the Yukon river. This clump of trees is about 2\ miles east of the International boundary line, and on the east bank of the Yukon river. We showed these skins to both McQuesten and Mayo, two traders who had been in that country over twenty years, and who said that they had seen Flying Squirrels along the Yukon river quite a number of times before, and had also seen them at Ft. Reliance and Ft. Yukon; but had not seen any of them for a number of years before this date. We also showed the skins to an Indian, who said these squirrels would attack a man by flying in his face, and the Indians would not eat them because the squirrels ate dirt. During the winter and spring of that year, I hunted very carefully in the vicinity where these squirrels were captured, but failed to find further trace of them. The stomachs of both of these squirrels were empty. The traps in which they were caught were set for martens, and two or three had been caught, but none were caught in these traps after the squirrels were captured. Scrums hudsonicus Erxleben. Hudson Bay Red Squirrel. All the red squirrels from the Yukon basin and northern Alaska, as far as can be determined at present, are referable to Scinrus hudsoni- cus ' proper,' although those from the Upper Yukon show considerable tendency toward S. h. streatori. Most of the Yukon specimens are in summer pelage, while the few available specimens from eastern Canada and the vicinity of Hudson Ray are in winter pelage, so that close comparison is not possible. Specimens from various points along the Yukon from Bennett to Nulato have been examined. The 1 Geol. and Nat, Hist, Survey of Canada, Annual Report, III (1887-88), pt. I, 50 A, 1889. North American Fauna, No. 19. Plate IV. Fig. 1.— Nests of Red Squirrels in Spruce Thicket. Fig. 2.— Burrows made by Red Squirrels in loose Scales stripped from Spruce Cones. on-., 1M0.] MAMMALS OF THE YUKON REGION. 27 animal is exceedingly abundant in all the spruce forest, and doubtless ranges northward to the limit of trees. Evidences of its activity are to be found all through the spruce forest. Its globular nests of grass, moss, bark, and refuse are com- mon (Plate IV, fig. 1), and are usually situated near the trunk of some slender spruce, 10 or 20 feet from the ground. Sometimes several will be found in the same tree, and a half dozen or more are very often to be seen at the same time. Little excavations in the moss show where the chickarees have been digging- for roots; and spruce cones tucked away in these and other out-of-the-way places are fur- ther evidence of their sagacit}'. The ground is often strewn for some distance with the scales of spruce cones which they have stripped (Plate IV, fig. 2). Near Lake Marsh I found one such place 20 feet square which was covered 6 inches deep with scales. Sciurus hudsonicus petulans subsp. now Tyjh from Glacier, White Pas?, Alaska (altitude 1,870 feet). No. 97457, U. S. Nat. Mus., Biological Survey Collection, 9 ad. Collected June 4, 1899, by W. H. Osgood. Original No., 370. ( 'luii ■elage: Above, mottled as in S. empetra, but general colorless fulvous; upperparts, mixed black, white and yellow- ish gray extending forward to top of head, becoming narrower and slightly grayer between shoulders; top of head chestnut mixed with black; nose and forehead clear hazel; under side of bod}' cinnamon rufous paling to nearly white around chin and extending to sides of body, neck and cheeks, and both sides of legs; under side of tail some- what deeper cinnamon rufous margined by }'ellowish white; subterminal black in tail less extensive than in empetra and hidiacensis; median part of upper side of tail grizzled black and yellowish, narrow sub- margin and subterminal zone black, the whole edged and overlaid with yellowish white. The hairs of the back in 8. plesius are of two kinds, some being of several colors arranged in zones and some pure black for their entire length. The former, which are most abundant, are dark sooty plumbeous at the base followed by a zone of light gray, 1 See Allen, Bull. Am. Mus. Nat. Hist., N. Y., Ill, 109, 1890. 30 NORTH AMERICAN FAUNA. [no. 19. then one of black, then yellowish white, and finally a black tip. In jS. empetra, the arrangement is practically the same, but the upper part of the light gray zone blends into fulvous. As this is the widest zone, it gives a fulvous suffusion to the entire upperparts of the ani- mal. In plesius the black submargin of the tail never shows through on the under side. Worn, spring pelage: Upperparts yellowish gray; top of head, forehead, and nose cinnamon rufous; thighs with faint suggestions of rufous; shoulders and neck hoary; sides and under- parts grayish white washed with yellowish and flecked with ochrace- ous; feet pale buffy ochraceous; tail paler than in postbreeding pelage. Skull. — Similar to that of /Sr. empetra from Unalaska, but smaller and lighter; nasals shorter and wider in proportion to their length; postpalatal notch extending farther forward, being almost on a plane with the last molar; molar teeth decidedly larger than those of kadiacensls. Measurement*. — Type (from dry skin): Total length 315; tail verte- brae 93; hind foot 50. Skull of type: Basal length 45; zygomatic breadth 35; postorbital constriction 13; length of nasals 18; least width of nasals 6; alveolar length of molar series 13. Remarhs. — The material representing Spermopkilus empetra is still very scanty and imperfect. Specimens from the Arctic coast are few in number and poor in quality, while from Hudson Bay one flat skin, unaccompanied by a skull, is all I am able to find. I have considered this (No. 13932, U.S.N.M.) to be typical of empetra and have used it in making skin comparisons. Since it agrees fairly well with speci- mens of the ground squirrel which has been introduced on Unalaska, I have used the skulls of these for skull comparisons. Specimens from Bristol Bay and the Alaska peninsula are apparently intermediate between empetra and plesius. S. hadiacensis is apparently confined to Kadiak Island, as specimens from the mainland immediately opposite the island are cranially and dentally distinct. The southern members of the group, columbianus and erythroglutwus, also need not be con- sidered, as they are very different from empetra and plesius. $. plesius was first met with on the south side of White Pass near Glacier, where a small colony was found on a steep rocky slope above the canyon. They were active here in early June while patches of snow still lay on the ground. On the summit of White Pass another small colony was found, and at Lake Bennett they were very abundant. Here their burrows are to be found wherever the conformation of the rocks affords lodgment of sufficient soil. From Bennett on to Fort Selkirk they are exceedingly abundant. We saw them daily about all the lakes, and as we floated down Fift}^-Mile and Thirty -Mile rivers, we often saw them bobbing in and out of their burrows or scurrying along their little trails which score the banks. From sunrise till late in the afternoon, their sharp clicking cries oci.,1900.] MAMMALS OF THE YUKON REGION. 31 rang out across the water, so that, if not to be seen, they at least reminded us of their presence nearly all the time. When alarmed, they .stand erect on their haunches near their burrows and violently utter their .sharp, high-pitched clickety ctick as long as the exciting cause is in .sight, always emphasizing each cry by vigorously slapping their short tails against the ground behind them. As a rule they were quite wary, and it was not possible to get within gunshot without some concealment and careful stalking. The limit of the range of the spe- cies along the river is near Fort Selkirk. The last specimen secured was caught near Rink Rapids, but I learned that quite a colony of ground squirrels exists on the west bank of the river just below Fort Selkirk. Spermophihis osgoodi Merriam. Fort Yukon Ground Squirrel. Spermophilus osgoodi Merriam, Proc. Wash. Acad. Sri., II, is, March 14, 1900. From Fort Selkirk, near the limit of Spennophihis plesius in the inte- rior, nearly to Circle, we saw no signs of ground squirrels of any kind. Just before reaching Circle, however, we began to sec unmistakable signs of them and were soon attracted to a small colony by their click- ing calls which reached our ears as we floated down in midstream. The call is executed in about the same time as that of S. plesnts, but its pitch is much lower and its effect on the ear is utterly different. It suggests the click of castanets. On going ashore we found their burrows and connecting paths scattered over quite an area on the hill- side. The colony occupied the open hillside and a few ledges of loose rock, and even extended down into a thicket of alder and willow at the foot of the hill. The animals were very shy and became much excited at our approach. Their long tails were very noticeable in marked contrast to the short ones of 8. plesms, which we had been accustomed to seeing. Fifteen specimens were secured.1 At this time (Aug. 14) they were all very fat and in splendid postbreeding pelage; the entire underparts were rich ferruginous without a trace of any other color; the back was very dark, and the long tail was full and bushy. One specimen was pure glossy black with faint shadowy indications of vermiculation on the back. Among the specimens in the National Museum from Fort Yukon are several in this melanistic condition, showing that it is not uncommon. The range of this spe- cies on the Yukon begins about 20 miles above Circle and extends at least to Fort YTukon and probably to the mouth of the Tanana. Arctomys caligatus Eschscholtz. Hoary Marmot. Six specimens of the hoary marmot were secured in the White Pass region and about Lake Bennett, where it wfas common. It is confiued 1 This valuable series was unfortunately destroyed. 32 NORTH AMERICAN FAUNA. [no. 19. to rocky, mountainous parts of the Hudsonian zone, and consequently we did not meet with it during the latter part of our trip, and only heard of it through reports from the mountains at the headwaters of the White and Tanana rivers. As elsewhere, it is familiarly known as the 'whistler,' although occasionally rather inappropriately called 'ground hog.' Its long drawn whistle is peculiarly mournful, par- ticularly when it breaks the deathly silence of some rocky canyon. It loves to stretch at full length on top of a large rock and bask in the sun. I frequently found it quietty enjoying itself in this manner. Castor canadensis Kuhl. American Beaver. It hardly seems possible that half a million or more beaver skins have been secured in the Territory of Alaska. The animal is now almost as rare there as it is in the United States, the inevitable result of continued pursuit by both whites and natives, which has so many parallels that it is useless to emphasize it here. At Fort Selkirk I saw several beaver skins taken on a small tributary of Stewart River, and at St. Michael I found a very few in the warehouses of the trad- ing companies. Beyond this I saw or heard nothing of them. Mus decumanus Pallas. Norway Rat. Large rats are exceedingly abundant at St. Michael. Their intro- duction must have been effected very recently, as they were unknown there at the time of Nelson's work. Unalaska has long been their northern limit on the Pacific coast. They find shelter about the wharves and lumber piles at St. Michael and also infest the buildings, particularly food warehouses. Their distribution will undoubtedly soon be extended all along the Yukon by means of the many steamers now plying between St. Michael and Dawson. Peromyscus oreas Bangs. Bangs White-footed Mouse. Peromyscus oreas Bangs, Proc. Biol. Soc. Wash., XII, 84, March 24, 1898. Long-tailed mice were taken at Skagway, Glacier, Summit, Bennett, Caribou Crossing, Fifty-Mile River, and Rink Rapids. In general they seemed to be more woodland loving than the short-tailed species, though at Bennett a number were taken among bare rocks at the very water's edge. I first noticed them here while walking along the shore at night. They were darting in and out among the rocks, chasing each other as if pla}'ing* a game of tag, and often four or five were in sight at once. P. oreas from the type locality is somewhat intermedi- ate between my specimens and those which come from the coast of Puget Sound and southern British Columbia. Northern specimens are slightly larger, paler, and less ruddy brown than typical oreas. They are very similar in color to canadensis and increase the prob- ora.,1900.] MAMMALS OF THE YUKON REGION. 38 ability that the latter has a transcontinental range. Their skulls are larger and have fuller braincases than those of either oreas or cana- densis. Peromyscus maniculatus arcticus (Mearns). Arctic White-footed Mouse. Hesperomys /< ucopus arcticus Mearns, Bull. Am. Mus. Nat, Hist., N. Y., II, 285, Feb., 1890. A short-tailed white-footed mouse was found to be very common from Lake Marsh to Lake Lebarge. Thirt}T specimens were secured, most of them about the crevices of low ledges of rock along the lake shores. The name arcticus is only tentatively used for these speci- mens, as its applicability can not be positively known until a series of Labrador specimens is obtained. My specimens do not differ from topotvpes of arcticus, and these in turn, as stated by Bangs,1 do not differ in color and size from typical maniculatus. The description of the skull of the Great Whale River specimen examined by Bangs, how- ever, does not agree well with the characters of the skulls of arcticus, so it seems advisable to recognize arcticus as a subspecies of manicu- latus. It is probable that more material will amply justify this treat- ment of the western form. Neotoma saxamans sp. nov. Northern Bushy -tailed Rat. Type from Bennett City, head of Lake Bennett, British Columbia. No. 98923, U. S. Nat. Mus., Biological Survey Collection, $ ad. Collected June 19, 1899, by W. H. Osgood. Original No., 462. (See Plate V, fig. 4.) Characters. — Similar to Neotoma cinerea drummondi, but somewhat darker; underparts pure white; skull strongly characterized. Color. — (Type:) Above, grayish fawn mixed with black, becoming brighter on sides, where the quantity of black is much diminished; underparts and feet pure white; eyelids intense black with a limited sooty area about them; nose and anterior cheeks ashy; tail slaty above, white below. Skull. — Similar to that of N drummondi (Plate V, fig. 3s) but with interorbital space narrower; nasals narrower and more attenuate pos- teriorly; maxillary arm of zygoma lighter; sphenopalatine vacuities open; ventral surface of occipital with a high trenchant median ridge; front of incisors very pale. Measurements. — Type (from dry skin): Total length 452; tail verte- brae 192; hind foot 46. Skull of type: Basal length 52; zygomatic breadth 29; interorbital width 5; length of nasals 23. Remarks. — Neotoma saxamans differs from N. cinerea? N. occiden- 1Am. Naturalist, XXXII, 496, July, 1898. 2 Neotoma c. Columbiana Elliot does not differ cranially from N. tinerca, and there- fore need not be considered in this connection. 3No. 75907, U. S. Nat. Mus., from Jasper House, Alberta. 4494— No. 19 3 34 NORTH AMERICAN FAUNA. [no. 19. iaHs, and N. drunnnondi chiefly in its long attenuate nasals, open sphenopalatine vacuities, and pale incisors. The only specimens secured were caught in a slide of large granite bowlders at the head of Lake Bennett. It was ascertained to occur, however, from White Pass to the Semenow Hills. In the cliffs above Glacier on the coast side of White Pass I found signs of JVeotoma, and once one peeped out of a crevice at me while I was busily engaged stalking a hoary marmot. It also occurs sparingly in the cliffs about Lake Lebarge and in the Semenow Hills, where the last evidences of its presence were seen. This distribution makes it the northernmost species of the genus. One night about 10.30, as I was returning to camp at Bennett, I saw one of these rats frisking about in the rocks. It was still quite light, and I immediately stopped and stood motionless while he darted in and out of the rocks. His movements were utterly noiseless and so quick that my eye could scarcely follow them. For some time his little whiskered nose appeared and disappeared at various openings in the rocks about ten feet away. Each time he would look steadily at me for a moment or two and then silently vanish. Gradually his curiosity overcame his caution, and in decreasing circles he came nearer and nearer until he bobbed out right before me and then cautiously approached until he could sniff at the toe of my shoe. A slight grat- ing of my gun barrel against a rock caused him to vanish like a flash, and this time he did not reappear. Evotomys dawsoni Merriam. Dawson Red-backed Mouse. Red-backed mice are by far the most abundant mammals in the Yukon region. Although but one specimen was taken at Bennett, and none between there and Fifty -Mile River, in spite of considerable trapping, aside from this they were found all along our route from Skagway to Fort Yukon. The following are the most important localities at which specimens were secured: Skagway, Glacier, Ben- nett, White Horse Rapids, Lake Lebarge, Rink Rapids, Fort Selkirk, Dawson, Char-lie Creek, and Circle. From a study of this series, which numbers over 100 specimens, it appears that all belong to one species, E. dawsoni. Its range probably reaches northward almost if not quite to the limit of trees. Specimens were trapped in all sorts of localities; along cold streams, under logs, in heavy moss, in Microtus runways, and among rocks. They abound on the large islands, where they were generally caught in dry, brush}^ places, in the dead leaves which cover the ground. We occasionally saw them during the day, and often heard them rust- ling the dead leaves on the ground about us as we lay in our blankets at night. They are the vermin of the miner's larder, and are always to be found about log cabins. North American Fauna. No. 19. Plate V. Skulls of Sciurus and Neotoma. (x m.) 1. Sciurus vancouverensis. 2. Sciurus hudsonicus petulans. 3. Neotoma cinerea drummondi. 4. Neotoma saxamans. ocr.,1900.] MAMMALS OF THE YUKON REGION. 35 Evotomys dawsoni alascensis (Miller). Tundra Rod-backed Mouse. Eootomys alascenm Miller, Proc. A.cad. Nat. Sci. Phila., L898, 364-367. The fflvotom/ys found at St. Michael has heretofore been compared only with the Asiatic E. rut /I us. Its closest relationship is really with K dawsoni, with which its range is doubtless continuous. On com- paring the series secured at St. Michael with those in the same condi- tion of pelage from Rink Rapids,1 Northwest Territory, I am unable to tind even the slightest difference in color or size. The skull of alascensis is slightly characterized by small, narrow molar series, and by nasals which have their posterior end truncate. The palate and audital bullae are not peculiar. The tail is often thick and bristly in winter pelage and in immature specimens of both dawsoni and alascen- sis. From this it appears that alascensis may be only a slightly marked subspecies. The favorite habitat of these mice about St. Michael is in the heaps of broken lava scattered about over the tundra. They are very rarely taken in the Microtus runways. They are common in the warehouses, which they seem to enter more readily than other mice of the tundra. Microtus mordax (Mcrriam). Long-tailed Vole. Specimens of this vole were taken at Skagway, Glacier, Bennett, Lake Marsh, Lake Lebarge, Rink Rapids, and near Charlie Village. Specimens from near the coast are almost exactly like those of the inte- rior and all are quite typical of the species. They were found in various environments, but the general habitat of the species was dry places rather than moist. At Glacier and Bennett they were secured on dry, rocky hillsides; at Lake Marsh two specimens were taken in the crev- ises of some granite rocks; at Lake Lebarge they were taken in the kitchen of a log cabin; at Rink Rapids, in an open, sandy place; and near Charlie Village, on the side of a cut bank, where they had made burrows and runways among the exposed roots of trees. Charlie Village is by far the northernmost locality from which the species has been recorded. Microtus drummondi (Aud. and Bach.). Drummond Vole. This is the most common meadow vole of the Yukon region. At Caribou Crossing and Lake Marsh its runways form interminable labyrinths in the level, open stretches of sedge at the margin of the water. It occurs in nearly all moist, grassy places from Caribou Crossing to Fort Yukon. From there it undoubtedly ranges farther on, at least to Nulato, where Dali took several specimens. It is most active during the day, as I easily learned by visiting traps night and 1 No good series of specimens is available from any point nearer Finlayson River, the type locality of E. dawsoni, than Rink Rapids. This series is therefore used to represent the species. 36 NOKTH AMERICAN FAUNA. [no. 19. morning. Near Fort Yukon I found its runways on recently depos- ited silt sparsely grown up to Equisetum. Its burrows in this soft material were very numerous, and at the entrance to each a little heap of earth in small globular lumps, as if carried in the mouth, was always to be seen. Microtus xanthognathus (Leach). Yellow-cheeked Vole. This tine species was met with only once. A small colony was found on a little stream near Charlie Village, occupying an old log ]am, part of which had become embedded in a matrix of sand and mud and overgrown with weeds. Burrows perforated this structure in numer- ous places, and well-beaten, open runways connected various openings about the protruding logs. The little animals were quite active during the daytime, and as I walked over the logs I occasionally saw one flash from one opening under a log to another and heard sharp little squeaks sounding all about beneath me. A liberal number of traps placed about yielded nine specimens, chiefly immature. The colony was apparently confined to the log jam, as traps set in suitable places but a few yards away secured only M. drummondi. Four specimens of this vole collected by Robert Kennicott are in the National Museum, one from the mouth of the Porcupine and three from the Yukon, 200 miles southwest of that point. Microtus operarius (Nelson). Nelson Vole. This vole was taken on a small stream about 40 miles above Circle, and a few more were secured between that point and Fort Yukon. It doubtless ranges from there to the coast. Forty-nine specimens were taken at St. Michael. These represent all stages of growth and several phases of color and seem to offer pretty conclusive proof that but one species of Microtus occurs at St. Michael. It was found in all moist parts of the tundra, being particularly numerous along the banks of the small ponds in the tall grass and rank, weedy growths. Fiber spatulatus1 sp. nov. Northwest Muskrat. Type from Lake Marsh, Northwest Territory, Canada. No. 98567, U. S. Nat. Mus., Biological Survey Collection, $ yg. ad. Collected July 3, 1899, by W. H. Osgood. Original No., 552. (See Plate VI, fig. 4.) Characters. — Similar in general to Fiber zibetliicus; size small; color rather dark; skull small; molar teeth very small; nasals short and much expanded anteriorly. Color. — Similar to Fiber zibetliicus, but apparently less suffused with fulvous. Skull. — Similar to that of Fiber zibetliicus (Plate VI, fig. 32), but smaller; jugals more slender, and but slightty produced dorsally ; audital bullae smaller; molar teeth decidedl}7 smaller; nasals much shortened and 1 Spatulatus, spatulate, in allusion to the shape of the nasals. 2No. 76259, U. S. Nat. Mus., from Wilmington, Mass. oct, 1900.] MAMMALS OF THE YUKON REGION. 37 widely expanded anteriorly, rapidly becoming compressed posteriorly; angular process of mandible short, blunt, and upturned; condyle narrow and somewhat rounded. Measurements. — Type (from dry skin): Total length 495; tail verte- brae 170; hind foot 73. Skull of type: Basal length 57; zygomatic breadth 38; length of nasals 21; alveolar length of molar series 14. Remarks. — Specimens of this species from Ugashik, Fort Kenai, Nushagak, and Nulato, in Alaska, have been examined. Besides these, 1 tind two specimens from Alberta which seem to be referable to it, one from South Edmonton and one from Henry House. These all agree in having very small molar teeth and short, spatulate nasals, characters which are amply sufficient to distinguish the species from all other forms in the genus. The specimens secured by Nelson at St. Michael can not now be found, but they doubtless show the same characters. Fiber osoyoosensis has Larger teeth and a much longer rostrum than spatulatus, so need not be further compared with it. From these facts it appears that Fiber spafodatus is the form occupying all of northwest America, and is derived from a form east of the Rocky Mountains rather than from a western one. Muskrats occur sparingly all along the Yukon, where the}r find par- ticularly favorable environment about the many small swamp-invested ponds a short distance from the river banks. At St. Michael a few are still found about the open ponds on the tundra. Synaptomys dalli Merriam. Dall Lemming Mouse. Lemming mice were taken at the foot of Lake Lebarge, at Rink Rapids, and near the mouth of the Chandindu River. At Lake Lebarge they were found in the long grass at the edge of a small pond; at other localities in cold, boggy places near small streams. The external characters of S. dalli have been unknown up to the present time, but, as was to be expected, they are in accordance with the gen- eral type so uniform throughout this genus. The color of the upper- parts is chiefly raw umber mixed with black; the lower parts are uni- form bluish white, and the feet and tail are dusky. The ears are of medium size and partially hidden by long hairs growing from the anterior base; a conspicuous bluish- white side gland is present in the males. The skull of the type of dalli is not fully mature and does not agree in all particulars with my specimens from the Upper Yukon. In these the skull is somewhat larger and heavier and the nasals are a trifle longer and more noticeably constricted posterior^. Lemmus yukonensis Merriam. Yukon Lemming. Lemmus yukonensis Merriam, Proc. Wash. Acad. Sci., II, 27, March 14, 1900. This lemming was found at only two localities — Rink Rapids, where five specimens were secured, and Charlie Creek, where five more were 38 NORTH AMERICAN FAUNA. [no. 19. taken. Considerable careful trapping was done at various points between these two places, but no other specimens were secured. At Rink Rapids they were caught about old logs and among dry leaves in places usually frequented by red-backed mice. At Charlie Creek one was caught in a Microtus runway and several were secured on the side of a cut bank. On one occasion one was seen running about under a brush heap in midday. Lemmus alascensis Merriam. Alaska Lemming. Lemmus alascensis Merriam, Proc. Wash. Acad. Sci., II, 26-27, March 14,1900. All efforts to secure this species at St. Michael proved fruitless. I kept large numbers of traps out for more than two weeks and set them in all conceivable locations about the tundra, but failed to catch any lemmings. Dicrostonyx nelsoni Merriam. Nelson Pied Lemming. Dicroslonyx nelsoni Merriam, Proc. Wash. Acad. Sci., II, 25-26, March 14, 1900. Dicrostonyx hudsonius alascensis Stone, Proc. Acad. Nat. Sci. Phila., March 24, 1900, 37-38. No specimens of this species were taken. Nelson says of it: Specimens were brought me by the fur traders from above Fort Yukon and from Nulato, Anvik, and Kotlik, along the course of the Yukon, and also from the Kaviak Peninsula and about Kotzebue Sound. A few were taken near St. Michael, but they were not numerous there. They are more plentiful about Bering Straits than any other district visited by me, if the number of their skins among the native children can be taken as a guide. Zapus hudsonius alascensis Merriam. Alaska Jumping Mouse. Three typical specimens of this jumping mouse were taken in a sedgy swamp near the foot of Lake Lebarge. Similar swamps exist near the Yukon, at least as far as Fort Yukon, but I was unable to do any trapping in them. No specimens were taken elsewhere. Erethizon epixanthus myops Merriam. Alaska Porcupine. Erethizon epixanthus myops Merriam, Proc. Wash. Acad. Sci., II, 27-28, March 14, 1900. Porcupines are quite common in all the forest region of Alaska. I noticed signs of them at man}7 places along the Yukon. They were abundant about Glacier, in the White Pass region, and 1 shot one there one evening as it swayed back and forth in the tap of a slender alder. It was eating the leaf buds which were just bursting. Ochotona collaris (Nelson). Alaska Pika. Two specimens of an ashy gray Ochotona were taken, one at the summit of White Pass, another at the head of Lake Bennett. The species was apparently quite rare at these localities and it was only with considerable difficulty that these individuals were secured. Both are very pale, ashy gray, with pure white underparts, no traces of OCT..1W0.1 MAMMALS OF THE YUKON REGION. 89 fulvous, and very indistinct collars. They are in the early spring or left-over winter pelage, and agree quite well with specimens in the same pelage collected in the Chigmit Mountains, near Bristol Bay, by C. L. McKay. The type and topotypes of 0. collaris are in the summer or post-breeding pelage and present quite a different appearance. The species apparently occurs in the high mountains throughout Alaska. It was reported to me from the MacMillan Mountains, the Upper Stewart River, the Upper White, and the Upper Tanana. Fragments of a skull were found in an owd pellet picked up by Dr. Bishop near Windy Arm, Lake Tagish. The present record from White Pass is the most southern one. There is suitable country for it farther south, and it will be interesting to trace its range in this direction. Lepus saliens sp. nov. Type from Caribou Crossing, between Lake Bennett and Lake Tagish, Northwest Territory. Canada. No. 98956, U. S. Nat. Mus., Biological Survey Collection, $ ad. Collected June 26, 1899, by W. II. Osgood. Original No., 504. Characters. — Similar to Lepus bairdi, but more yellowish and less ruddy; dorsal hairs with plumbeous roots; feet nearly white in sum- mer; similar to L. cohmibiensisj but with greater amount of black in dorsal region; feet much lighter; skull similar in general to that of Lepus a. da Hi/ audital bulla; very large. Colo?\ — Type in worn spring pelage: Upperparts mixed black and yellowish buff, writh patches of plumbeous under -fur exposed in places; black hairs predominating on rump and middle of back, forming an ill-defined dorsal stripe; outer edge of thighs, outer side of forelegs and pectoral band buff; ears and head, except sides of nose, buff with black hairs sprinkled through; sides of nose gray; ears margined with white; hairs of fore and hind feet plumbeous at base, rufous in cen- tral part, and broadly white at tips; general appearance of feet white, lightly mixed with rufous; underparts, except pectoral band, white. Skull. — Similar to that of dalli but somewhat larger; teeth heavier; nasals long, heavy, and very broad anteriorly; audital bulla? very large; palate short; malars rather wide, deeply channeled anteriorly; postorbital and antorbital processes of frontais well developed. Measurements. — Type (measured from dry skin): Total length 395; hind foot 134; ear from crown 74. Skull of type: Occipitonasal •length 77; greatest zygomatic breadth 38; length of nasals 33; great- est width of nasals 17; alveolar length of molar series 15. Remarks. — The exact relation in which this species stands to ameri- canus, bairdi, and columbiensis is difficult to determine at present. Its light feet point to relationship with bairdi, while its dark under color 1 Rhoads, Proc. Acad. Nat. Sci. Phila., 1895, 242-243. 40 NORTH AMERICAN FAUNA. [no. 19. and general buffy appearance are more like columhiensk. Its skull is quite distinctive, the large audital bulla? and broad nasals being- unequaled in the group. It seems probable that it i.s a northern form of hairdi not related to colwmfa'ensis, which is nearer to wctshmg- tmii. There are no specimens available to show whether or not it has any connection with dalli, which is the form found on the Lower Yukon. But two specimens were secured — the type, which I shot in a Lepargyrcea thicket at Caribou Crossing, and one very young female which Dr. Bishop took in a willow bog near Bennett City. It seems to have been a decidedly 'off year' for rabbits, for these two were the only ones we saw on our entire trip, though numerous signs of their former abundance were seen daily. Lepus americanus dalli Merriam. Dall Varying Hare. Lepus americanus dalli Merriain, Proc. Wash. Acad. Sci., II, 29-30, March 14, 1900. This rabbit is doubtless abundant at certain times all along the Lower Yukon, but we heard very little of it. It is subject to epidem- ics and frequently becomes locally extinct, which probably accounts for its scarcity last year. Lepus othus Merriam. Alaska Arctic Hare. Lepus othus Merriam, Proc. Wash. Acad. Sci., II, 28, March 14, 1900. Signs of Arctic hares were occasionally noticed about St. Michael, but we did not see any of the animals. The Eskimos were hunting continually, and brought numbers of ducks and geese to the village to sell, but they brought no rabbits during our stay. Lynx canadensis mollipilosus Stone. Arctic Lynx. Lynx canadensis mollipilosus Stone, Proc. Acad. Nat. Sci. Phila., March 24, 1900, 48-49. The Canada lynx is not as common in the interior of Alaska as might be expected. I saw no signs of it and could obtain only very scanty information as to its occurrence. The police sergeant in charge of the station at the foot of Lake Lebarge told me that the tracks of but one had been seen in that vicinity during the previous winter. Lynx- skin robes are in common use in the country, but the majority of them are imported. This I learned from a trader at Circle, who had several for sale that came from eastern and southern Canada. Lynx skulls from the following localities are in the National Museum: Tanana River, Russian Mission, Nulato, Andraefski, and mountains near Unalakleet. Canis occidentalis Richardson. Wolf. The country along the Yukon is not well suited for wolves, and they are seldom seen there. A prospector showed me the skin of a large gray one from the upper waters of the MacMillan river — the only one I saw on the trip. oct 1900] MAMMALS OF THE YUKON REGION. 41 Vulpes fulvus (Desmarest) \ Red Fox. Occasional reports of foxes were received all along- our route, but no specimens were secured. Owing- to their natural sagacity, foxes are doubtless able to hold their own against trappers better than most other fur-bearing animals. Their skins are quite common among traders and natives. Vulpes hallensis Merriam. Hall Island Fox. Vulpes hallensis Merriam, Proc. Wash. Acad. Sci., II, 15-16, March 14, 1900. White fox skins are common among the natives and traders at St. Michael, and could be bought at from $1 to $4 each, according to quality. During our stay there one of the animals was seen on the island, which indicates that they are still far from extermination. Ursus americanus Pallas. Black Bear. Black and brown bears are common all along the Yukon. We found them common on the upper river, and Nelson records them as far down as Anvik. We saw tracks very frequently, but owing to the thick forest and underbrush, and the fact that we made no special hunts for them, the animals themselves were rarely observed. A young adult female in glossy black pelage was killed at Glacier by A. G. Maddren, and several others were seen during our stay there. I was told at Lake Lebarge and at White Horse Rapids that brown bears were seen very frequently. At Fort Selkirk I saw skins brought from the Pelly River. Near Charlie Village I saw the skin of a large brown bear that had been killed there shortly before our arrival. One afternoon while sitting in the boat preparing specimens, about 20 miles above Circle, I saw a good-sized black bear walking deliberately across an open space on a hillside a short distance away. We gave chase, but did not see it again. At the mouth of the Tatondu River I saw numerous tracks, and on the border of a stagnant pool found evidences that bruin had been enjoying a mud bath. Moss uprooted by bears in digging for roots was noticed at several places. Ursus horribilis alascensis Merriam. Alaska Grizzly Bear. Very little accurate information is obtainable in regard to the grizzly in the Yukon region. It doubtless occurs sparingly all along the river, but miners and prospectors report any large bear as a grizzly, and without doubt often mistake the brown bear for it. There are a num- ber of its skulls from Norton Sound in the Biological Survey collection. Lutra canadensis (Schreber). American Otter. The fate of the otter in Alaska is much the same as that of the bea- ver. There are doubtless a few on some of the smaller streams of the interior and about the Yukon delta, but they are now quite rare in comparison wTith their former abundance. 42 NORTH AMERICAN FAUNA. [no. 19. Lutreola vison ingens subsp. nov. Alaska Mink. Type (skull) from Fort Yukon, Alaska. No. 6530, U. S. Nat. Mus., $ ad., old. Collected by Robert Kennicott. (See Plate VI, fig. 2.) Characters. — Size largest of North American mink; similar to L. v. energumenos, but lighter in color and very much larger; skull and teeth very large and heavy. Color. — Similar in general to Lutreola v. ensrgumenos, but paler. Skull.- — Very large, angular, and ridged; rostrum very wide; brain- case relatively shallow and very wide; zygomata heavy; audital bulla? large and relatively wide; dentition heavy. (Compare with skull of Lutreola v. energumenos, Plate VI, tig. I.1) Measurements. — No. 13880, U. S. National Museum, St. Michael, Alaska (from dry skin): Total length 720; tail vertebra? 180; hind foot 75. Skull of type: Occipitonasal length 69; zygomatic breadth 47; mastoid breadth 41; breadth across postorbital processes 23; length of audital bulla 17. Average of five adults: Occipitonasal length 44.5; mastoid breadth 39.5; breadth across postorbital processes 21; length of audital bulla 17.5. Remarks. — The large size of the Alaska mink has been noted by various authors,2 but each has dismissed the subject by concluding that it is the natural result of the animal's northern range, and the form has remained unnamed, while less marked forms from other local- ities have been recognized. The largest mink previously described is L. v. energumenos, which is very much smaller than ingens and also averages much darker. The minks of the Yukon region are caught mostly on the tributary streams, and, as stated by Nelson, are very abundant in the area between the deltas of the Yukon and the Kuskokwim. Along the Yukon itself our party did not see any, and very few signs of them were observed. Their skins were seldom seen among the Indians and Eskimos. They were reported, however, from the Porcupine, Koyu- kuk, Tauana, and various other streams tributary to the Yukon, and without doubt occur in suitable places all over Alaska. Putorius arcticus Merriam. Tundra Weasel. Putorius arcticus Merriam, N. Am. Fauna No. 11, 15, June, 1896. Putorius cicognani richardsoni Merriam, /. c, 11-12 (part). Three immature specimens of this weasel were taken at St. Michael. They were caught in traps baited with sandpipers and set among the lava rocks along the shore. Several specimens which were also secured at St. Michael by Nelson and Turner are in the National Museum. Besides these I find specimens from Nulato, Fort Yukon, and Fort Reliance, which gives the species a more extensive range in the interior than it has been supposed to have. Most of these specimens are 1 No. 5537, Bangs collection, from Sumas, B. C. 2 See Allen, Bull. U. S. Geol. and Geog. Survey Terr., II, 327-328, 1876. North American Fauna, No. 19. Plate VI. Skulls of Lutreola and Fiber. (Natural size.) 1. Lutreola vison energumcno& 2. Lutreola vison ingens. ?,. Fiber zibethicus. i. Fiber spatulatux. oct.,1900.] MAMMALS OF THE YUKON REGION. 43 imperfect, but enough skulls are now at hand to show conclusively that all the Yukon specimens heretofore identified as richardsoni arc really practically identical with P. arciicus from Point Barrow. Putorius cicognani alascensis (Merriam). Juneau Weasel. A single immature specimen taken 20 miles below Fort Selkirk is referred to this form. Its skull is rather large and indicates a possible intergradation with /'. arcticus; otherwise it agrees with alascensis. Putorius rixosus eskimo Stone. Alaska Least Weasel. Putorius rixosus eskimo Stone, Proc. Acad. Nat. Sci. Phila., March 24, L900, 44-45. No specimens of this rare weasel were obtained. There are three imperfect specimens in the National Museum, two from St. Michael and one from Fort Reliance. Besides these the only ones recorded are the type and four topotypes from Point Barrow, Alaska, and the specimen mentioned by Stone (loc. cit.) from Bethel, Kuskokwim River, Alaska. Mustela americana actuosa subsp. nov. Alaska Marten. Typt (skull) from Fort Yukon, Alaska. No. 6043, U. S. Nat. Mus., $ ad., old. Collected by Robert Kennicott. (See Plate VII, fig. 2. ) ( (hcwacters. — Similar to M. orumalis, but larger; cranial and dental characters distinctive. Color.— (Topotype, No. 6116, U.S.N.M., $ ad.): Posterior half of upperparts pale ochraceous buff, shoulders and anterior part of upper- parts gradually becoming grayish; entire upperparts, except head, overlaid with coarse brown hairs; head, including cheeks and throat, pale grayish-white lightly mixed with brown, especially on nose and chin; inside and edges of ears whitish, outside and bases of ears brown; underparts similar to upperparts, but darker and more brown- ish on chest; an irregular patch of creamy buff mixed with white on chest; legs and feet dark brown, front of legs with mixture of gray hairs; tail brown, somewhat darker at tip, and with a slight mixture of gray hairs. Skull. — Similar to that of M. brtmaMs (Plate VII, fig. I1), but some- what larger; relatively longer and narrower; interorbital space slightly narrower; audital bulhv very much larger and longer; dentition rela- tively much weaker; last upper molar decidedly smaller. Mt asurements. — Average of four adult male topotypes measured in the flesh by the collector: Total length 26.22 inches (665 mm.); tail vertebrae 8.08 inches (223 mm.); hind foot 4.36 inches (109 mm.). Skull of type: Occipitonasal length 85; greatest zygomatic breadth 55; breadth across postorbital processes 21; palatal length H; length of audital bulla? lit. 1 Type No. 7417, Bangs collection, from Okak, Labrador. 44 NOKTH AMERICAN FAUNA. [no. 19. Remarks. — This form is the largest of the subspecies of Mustda americana. 3f. hrumalis is also large, but does not equal actuosa, and notwithstanding its smaller size has heavier dentition. The enor- mous audital bulla1 of actvom are not equaled hy those of any other member of the group. The skulls of am.ericana (Plate VII, fig. 31) and caurina are so very much smaller than those of hrumalis and actu- osa that they do not need to be closely compared. In a good series of actuosa from Fort Yukon and Fort McPherson the characters are very constant. A large number of skins from these localities present very little variation, and nearly all are quite light colored like the one described above. The marten is still the commonest fur-bearing animal of Alaska, notwithstanding the hundreds of thousands that have already been taken. Trappers are always confident of a harvest of martens whether other animals are abundant or not. Mustela pennanti Frxleben. Fisher. Dr. Elliott Coues2 states that he has examined specimens of the fisher from Alaska, but does not give the exact localit}\ At present no specimens are at hand to corroborate this record, but there is little doubt that the animal occurs along the Upper Yukon, as it is known from similar latitudes to the eastward. It was not met with by our party, and I received no reliable information in regard to it. Gulo luscus (Linmeus). Wolverine. Wolverines seem to be quite common in the Yukon region. They were often reported, and I saw a number of skins among the natives on the lower river. One was said to have been trapped at Tagish in the winter of 1898, and others were seen in the vicinity. They are seen frequently about Lake Lebarge in winter, and trappers from the MacMillan River say they are abundant in that region. Sorex personatus streatori Merriam. Streator Shrew. Specimens of this shrew were secured as follows: Haines 1, Skag- way 6, Glacier 1, Bennett 3, Caribou Crossing 2, Lake Lebarge 1, 50 miles below Fort Selkirk 1, mouth of Chandindu River 1, and 40 miles above Circle 1. Although the conditions along the Yukon seem to be ideal for shrews, I was unable to secure many specimens, and could only conclude that they were not common there, for the same methods of trapping were much more successful in the coast regions. Sorex personatus arcticus Merriam. Arctic Shrew. Sorex personatus arcticus Merriam, Proc. Wash. Acad. Sci., II, 17, Mar. 14,1900. Twenty specimens were taken at St. Michael. They occur through- out the tundra in much the same situations as S. tundrensis, but were also found in the lava heaps and along high banks near the coast. xNo. 4934, Merriam collection, from the Adirondack^, New York. 2Fnr-bearinensilvanicus at Circle, but the others were not seen again. To one accustomed to the orchards, fields, and forests of Connect!- ocr.,1900.] INTRODUCTION. 49 cut, the duck marshes of North Dakota, or even the balsam thickets of northern Now England, the Yukon Valley seems wanting in bird life — not the center of abundance of its avifauna, but rather a deposit for the overflow from more favored regions. There are exceptions to this rule, notably wandering flocks of crossbills, the colonies of bank swallows of Fifty-Mile and Thirty-Mile rivers and the Yukon proper, the spotted sandpipers that continually flitted across our bow, the intermediate sparrows and juncos that seldom failed to greet us as we stepped ashore, and the Alma thrushes, whose songs sounded all night, wherever we happened to camp. Bird life is fairly abundant, too, in certain favored places such as Log Cabin, Caribou Crossing, the swampy shores of Lake Marsh, and the ponds and level country at the lower end of Lake Lebarge. Near Miles Canyon I noticed 23 species on July 11, but individuals of each, with the exception of bank swallows, were few. In the entire Upper Yukon Valley breeding colonies of shore and water birds were conspicuously absent. The precipitous shores of the lakes, the comparative absence of islands, the swift current of the Y^ukon, and its high banks cut by narrow, wooded valleys, are a sufficient explanation of this; and I can not believe that either geese, ducks, or shore birds ever bred abundantly in most of the region visited, though their number has doubtless been reduced in recent years. In the Yukon flats the condition changes, and no doubt many of these birds find a summer home in the ponds a few miles back from the river as they do at the foot of Lake Lebarge; but these we had no opportunity of visiting. Our study of the bird life of the Yukon was chiefly confined to what could be seen or heard from our boat or on the banks in the immediate vicinity of camping places. From the lakes to the Alaska boundary snow-capped peaks were absent, and no species were found that did not also occur upon the banks of the river, although we climbed hills, visited deep woods, and ascended small streams for some distance. As we proceeded north, however, several birds were found at lower altitudes than those at which they had been already noted. Away from the river, birds were rarer than immediately upon its banks. We learned little regarding the Upper Yukon as a migratory high- way for species breeding farther north, though we heard that thou- sands of geese and ducks passed Lower Lebarge in the spring. It was too late for the spring migration, and the southward movement of ducks and geese had hardly begun on August 20, when we left Circle. The fall migration of the Limicola? should have been well under way at this date, but very few of these birds were observed. If they do pass in large numbers they must frequent the ponds back from the river. Several times at Circle, I walked a long distance over the sand flats left bare by the falling Yukon without seeing any 1491— No. 19 1 50 NORTH AMERICAN FAUNA. [no. 19. shore birds, or anything on which they could feed. This was very different from the constantly passing flocks I saw on the Yukon Delta August 27-28, and the abundance of Limicolee at St. Michael in Sep- tember. The smaller land birds we often saw late in July and in August. They were usually in family parties, and most of them seemed to be traveling up the river. At Circle the intermediate sparrow, western tree sparrow, and western savanna sparrow were abundant, and were evidently migrating August 19-20. Forty -two species of migratory birds, exclusive of those possessing a continental range, certainly occur as summer residents in the Yukon Basin above Fort Yukon. Of these, 13 (31 per cent) have their center of distribution in eastern North America, 14 (33 per cent) near the Pacific coast, and 15 (36 per cent) in western North America not far from the Rocky Mountains. The eastern birds reach the Yukon through the Rocky Mountains. Some of these, such as Chordeiles virginianus, were found only above the Tatchun River; others, as Empidonax t. alnorum, were absent above the Pelly and common from there to Fort Yukon; others, as Wilsonia pusilla, were not found above the Chandindu River; others, as Helmmtlwphila pere- grina were each found at a single place, while still others, as Junco hyemalis and Merula migratoria, were regularly distributed along the river. The Pacific coast forms probably all reach the Yukon over the Alaska coast range. These disappear as one goes north, Hylo- cichla aonalaschkce extending through Lake Bennett, Wilsonia p. pileo- lata to Lake Marsh, Dendroica townsendi to Lake Lebarge, Myadestes townsendi to Dawson, and Tachycineta thalassina to Circle. Last and most important in number of species, abundance of individuals, and regularity of distribution are birds which breed in the Yukon Valley and spend the winter in the western United States, as Zono- trichia I. gambeli, Spizella s. arizonce, and the small Ammodramus s. alaudinus of the Yukon lakes, and those which probably enter by the mouth of the Yukon, as the large Ammodramus s. alaudinus, found below Alaska boundary, and Seiurus n. notabilis, first met near Dawson. In coloring, Yukon birds, especially in juvenile plumage, show a strong tendency to replace the buff-ochraceous markings of Eastern forms by white, cream color, and gray. Canacliites c. osgoodi, Parus h. evura, and Ilylocichla u. alma are good examples of this characteristic. I take this opportunity to express my hearty thanks to Dr. Merriam for the privilege of visiting Alaska as a member of the Biological Sur- vey party, of writing this report, and of using the collection of the Biological Survey in its preparation; also to Mr. Osgood and Mr. Oberholser of the Biological Survey for aid in determining species. I am also greatly indebted to Mr. Robert Ridgway and Dr. Charles W. Richmond for the opportunity of studying the collection of the OCT., 1900.] CLASSIFIED LISTS OF SPECIES. 51 United States National Museum and for much valuable assistance; to Dr. J. A. Allen and Mr. F. M. Chapman for the hours which I spent with the birds in the American Museum of Natural History; to Mr. William Brewster for the courtesy of allowing me to compare my specimens with those in his valuable collection, and to Mr. Walter Deane for much help in this study. CLASSIFIED LISTS OP SPECIES. NEW SUBSPECIES. Canachites canadensis osgoodi. Contopus richardsoni saturatus. Sayornis saya vukonensis. SPECIES NOT HITHERTO RECORDED FROM WESTERN NORTH AMERICA. Halireetus albicilla. SPECIES NOT HITHERTO RECORDED FROM SOUTHEASTERN ALASKA. iEchmophorus occidentalis. Xema sabinii. Lagopus leucurus. Picoides americanus alascensis. Coutopus richardsoni saturatus. Empidonax hammondi. Junco hyemalis connectens. Sitta canadensis. Merula migratoria. SPECIES NOT HITHERTO RECORDED FROM UNALASKA. Larus Philadelphia. Tringa maculata. Tringa acuminata. Loxia curvirostra minor. SPECIES NOT HITHERTO RECORDED FROM THE PRIBILOF ISLANDS. ? Arenaria melanocephala. Larus Philadelphia Tringa acuminata. SPECIES NOT HITHERTO RECORDED FROM ST. MICHAEL. Calidris arenaria. SPECIES NOT HITHERTO RECORDED FROM THE YUKON ABOVE FORT YUKON. Tringa hairdi. Symphemia semipalmata inornata. Buteo borealis calurus. Falco sparverius. ? Megascops asio kennicotti. ? Dryobates villosus hyloscopus. Contopus borealis. Contopus richardsoni saturatus. Empidonax trailli alnorum. Empidonax hammondi. Spinus pinus. Spizella socialis arizonae. Passerella iliaca. Helminthophila peregrina. Dendroica townsendi. Wilsonia pusilla pileolata. Sitta canadensis. Hylocichla aonalaschkaj. Hylocichla aonalaschkaj pallasi. Saxicola cenanthe. LIST OF SPECIES KNOWN FROM THE YUKON BASIN. Colymbus holbcelli. Colymbus auritus. Gavia imber. Gavia arctica. Gavia lumme. Stercorarius pomarinus.1 Stercorarius parasiticus.1 Stercorarius longicaudus.1 1 Known only from Fort Yukon or below. 52 NORTH AMERICAN FAUNA. [NO. 19. Rissa tridactyla pollicaris.1 Larus barrovianus.1 Larus argentatus smithsonianus. Larus vegae.1 Larus brachyrhynchus. Larus Philadelphia. Xema sabinii.1 Sterna caspia.1 Sterna paradisaea. Sterna aleutica. 1 Hydrochelidon nigra surinamensis.1 Phalacrocorax pelagicus robustus.1 Merganser americanus. Merganser serrator.1 Anas boschas. Mareca americana. Nettion carol inensis. Querquedula disco rs.1 Spatula clypeata. Dafila acuta. Aythya vallisneria.1 Aythya marila. Aythya affinis. Clangula clangula americana. Clangula islandica. Charitonetta albeola. Harelda hyemalis. Histrionicus histrionicus. Arctonetta fischeri.1 Somateria v-nigra.1 Somateria spectabilis.1 Oidemia americana.1 Oidemia deglandi. Oidemia perspicillata. Chen hyperborea.1 Anser albifrons gambeli.1 Branta canadensis hutchinsi. Branta canadensis minima. Branta nigricans. * Philacte canagica.1 Olor Columbian us. 1 Olor buccinator.1 Grus canadensis. Fulica americana.1 Crymophilus fulicarius.1 Phalaropus lobatus. Gallinago delicata. Macrorhamphus scolopaceus. l Tringa canutus. l Tringa couesi.1 Tringa maculata. Tringa bairdi. Tringa minutilla. Tringa alpina pacifica. Ereunetes occidental is.1 Calidris arenaria.1 Limosa lapponica baueri.1 Limosa haemastica.1 Totanus flavipes. Helodromas solitarius cinnamomeus. Heteractitis incanus.1 Bartramia longicauda. ' Symphemia semipalmata inornata.2 Tryngites subruficollis ] Actitis macular ia. Numenius hudsonicus. Numenius borealis.1 Squatarola squatarola. Charadrius dominicus. Charadrius dominicus fulvus.1 ^Egialitis semipalmata. Arenaria interpres.1 Arenaria melanocephala. ' Canachites canadensis osgoodi. Bonasa umbellus umbelloides. Lagopus lagopus. Lagopus rupestris. Pedicecetes phasianellus columbianus. Circus hudsonius. Accipiter velox. Aceipiter atricapillus. Buteo borealis calurus.2 Buteo swainsoni.1 Archibuteo lagopus.1 Haliaeetus leucocephalus alascanus. Falco rusticolus gyrfalco. Falco peregrinus anatum. Falco columbarius. Falco columbarius richardsoni.2 Falco sparverius.2 Pandion haliaetus carolinensis.1 Asio accipitrinus.1 Scotiaptex cinerea. Scotiaptex cinerea lapponica.1 Nyctala tengmalmi richardsoni. ?Megascops asio kennicotti.2 Bul)o virginianus pallescens. Nyctea nyctea.1 Surnia ulula caparoch. Ceryle alcyon. Dryobates villosus leucomelas. ?Dryobates villosus hyloscopus.2 Known only from Fort Yukon or below. 1 Known only above Fort Yukon. OCT., 190(1. J CLASSIFIED LISTS OF SPECIES. 53 Dryobatea pubescens nelsoni. Pieoides arcticus. Picoides americanua alascenais. Colaptes auratus luteus. Chordeilea virginianus.2 Selasphorus rufua.2 Sayomia saya yukonensis.' Con t opus borea lis. Contopus richardsoni saturatus.' Empidonax trail]].1 Empidonax trailli alnorum.2 Empidonax hammondi.'-' Otocoris alpeatria leucolamia. Pica pica hudsonica. Perisoreua canadensis fumifrona. Corvua corax principalis. Scolecophagus carolinns. Pinicola enucleator alascenais. Pyrrhnla cassini.1 Loxia leucoptera. Acanthia hornemanni exilipes. Acanthis linaria. Spinus pinus.2 Paaaerina nivalia. Calcarius lapponicua alascenais. Calcariua pictns.1 Ammodramua sandwichensis alaudinns. Zonotrichia leucophrys gambeli. Zonotrichia coronata. Spizella monticola ochracea. Spizella socialia arizonse.2 Junco hyemalia. Melospiza lincolni. Pasaerella iliaca. Petrochelidon lunifrona. Hirundo erythrogastra unalaachkensia. Tachycineta bicolor. Tachycineta thalassina.2 Clivicola riparia. Ampelia garrulus. Lanius borealis. Helminthophila celata. Helminthophila peregrina.2 Dendroica a?stiva rubiginosa. Dendroica coronata hooveri. Dendroica striata. Dendroica townsendi.2 Seiurua aurocapillus.1 Seiurus noveboracensis notabilis. Wilsonia pusilla. Wilsonia pusilla pileolata.2 Bndytes flavns leucoatriatus.1 Anthus pensilvanicus. Cinclua mexicanus. Sitta canadenais.2 Parus atricapillus septentrionalis. Parus cinctua alascensis.1 Parus hudsonicua evura. Phyllopseuates borealis.1 Regulus calendula. Myadeatea townsendi.2 Hylocichla alicise. Hylocichla uatulata almse. Hylocichla aonalaschkse.2 Hylocichla aonalaschkse pallasi.2 Merula migratoria. Hesperocichla naevia. Saxicola cenanthe. Sialia arctica.2 8PECIES WHOSE OCCURRENCE ON THE YUKON IS DOUBTFUL. Chaulelasmua atreperua. Eniconetta stelleri. Branta canadenaia. Macrorhamphua griaeua. Aquila chrysaetoa. Nucifraga colunibiana. Loxia curvirostra minor. Motacilla ocularis. 1 Known only from Fort Yukon or below. 2 Known only above Fort Yukon. 54 NORTH AMERICAN FAUNA. [NO. 19. SPECIES AND SUBSPECIES OCCURRING IN THE YUKON BASIN AND HAVING THEIR CENTER OP ABUNDANCE DURING THE BREEDING SEASON IN ALASKA AND BERING SEA. Rissa tridactyla pollicaris.1 Larus barrovianus.1 Larus vegfe.1 Larus brachyrhynchus. Sterna aleutica.1 Phalacrocorax pelagicus robustus.1 Arctonetta fischeri.1 Somateria v-nigra.1 Chen hyperborea.2 Branta canadensis minima. Branta nigricans. 2 Philacte canagica.1 Grus canadensis.2 Macrorhampus scolopaceus. Tringa couesi.2 Ereunetes occidentalis. Heteractitis incanus. Arenaria melanocephala. Of these 35 forms, 1 is a subspecies of an Asiatic bird, 5 are chiefly confined to Bering Sea, 2 range in winter to the western Pacific, 7 are resident subspecies of northern North American birds, and the remaining 20 pass in winter to the western United States or beyond. EASTERN NORTH AMERICAN SPECIES FOUND IN THE YUKON BASIN. Canachites canadensis osgoodi. Halieeetus leucocephalus alascanus. ? Megascops asio kennicotti. Picoides americanus alascensis. Sayornis saya yukonensis. Contopus richardsoni saturatus. Perisoreus canadensis fumifrons. Pinicola enucleator alascensis. Leucosticte tephrocotis littoralis. Calcarius lapponicus alascensis. Hirundo erythrogastra unalaschkensis. Dendroica jestiva rubiginosa. Dendroica coronata hooveri. Paras cinctus alascensis. Parus hudsonicus evura. Hylocichla ustulata almee. Hesperocichla nsevia. Limosa hsemastica. Numenius borealis. Accipiter atrieapillus. Falco sparverius. Colaptes auratus luteus. Chordeiles virginianus. Empidonax trailli alnorum. Junco liyemalis. Passerella iliaca. Helminthophila celata. Helminthophila peregrina. Dendroica striata. Seiurus aurocapillus. Wilsonia pusilla. Hylocichla alicise. Hylocichla aonalaschkpe pallasi. Merula migratoria. WESTERN NORTH AMERICAN SPECIES FOUND IN THE YUKON BASIN. Anser albifrons gambeli. Branta canadensis hutchinsi. Olor buccinator. Symphemia semipalmata inornata. Bonasa umbellus urnbelloides. Pedicecetes phasianellus columbianus. Buteo borealis calurus. Buteo swainsoni. Falco columbarius richardsoni. Bubo virginianus pallescens. ? Dryobates villosus hyloscopus. Picoides americanus alascensis. Selasphorus rufus. Empidonax trailli. Empidonax hammondi. Otocoris alpestris leucolfema. Pica pica Hudsonica. Calcarius pictus. Ammodramus sandwichensis alaudinus. Zonotrichia leucophrys gambeli. Spizella monticola ochracea. Spizella socialis arizonse. Seiurus noveboracensis notabilis. Cinclus mexicanus. Parus atrieapillus septentrionalis. Myadestes townsendi. Sialia arctica. 1 Reported only from the Yukon Delta. 2 Known only as migrants. OCT., 1900. CLASSIFIED LISTS OF SPECIES. 55 AMERICAN PACIFIC COAST SPECIES FOUND IN THK YUKON BASIN. Helodromas solitarius cinnamomeus. Tachycineta thalassina. Zoiiotrk'hia coronata. Dendroica townsendi. Wilsonia pusilla pileolata. Hylocichla aonalaschkse. ASIATIC AND PACIFIC SPECIES FOUND IN THE YUKON BASIN. Limosa lapponica baueriJ Gharadrins dominions fulvus.1 Aichibutet i lag* >pus. ScotiapteN cinerea lapponica.1 Pvrrhula cassini. Budytee fiavus leucostriatus.1 Phyllopseustes borealis.1 MIGRATORY SPECIES NOT COMMON TO NORTHERN NORTH AMERICA FOUND DURING BREEDING SEASON IN THE YUKON BASIN ABOVE FORT YUKON. Eastern species. Accipiter atricapillus. Falco sparverius. Colaptes auratus luteus. Chordeiles virginianus. Empidonax trailli alnorum. Junco hyemalis. ? Passerella iliaca. Helminthophila celata. Helminthophila peregrina. Dendroica striata. Wilsonia pusilla. Hylocichla aliciaj. Hylocichla aonalaschkse pallasi. Merula migratoria. Western species. Branta canadensis hutchinsi. ? Grus canadensis. Sympheniia semipahnata inornata. Buteo borealis calurus. ? Otocoris alpestris leucolaema. Pica pica hudsonica. ? Calcarius lapponicus alascensis. Ammodramus sandwichensis alaudinus. Zonotrichia leucophrys gambeli. Falco colurnbarius ricbardsoni. Selasphorus rufus. Sayornis saya yukonensis. Empidonax hammondi. Spizella monticola ochracea. Spizella socialis arizonse. Seiurus noveboracensis notabilis. Myadestes townsendi. Hylocichla ustulata alinse. Sialia arctica. Pacific coast species. Larus brachyrhynchus. Helodromas solitarius cinnamomeus. Leucosticte tephrocotis littoralis. Zonotrichia coronata. Hirundo erythrogastra unalaschkensis. Tachycineta thalassina. Contopus richardsoni saturatus. Dendroica sestiva rubiginosa. Dendroica coronata hooveri. Dendroica townsendi. Wilsonia pusilla pileolata. Hylocichla aonalaschkse. Hesperocichla nsevia. SPECIES OCCURRING ON THE COAST OF SOUTHEAST ALASKA AND IN THE YUKON VALLEY. Gavia imber. Larus Philadelphia. Anas boschas. Histrionicus histrionicus. Oidemia deglandi. Oidemia perspicillata. Phalaropus lobatus. Actitis macularia. Haliteetus leucocephalus alascanus. Picoides americanus alascensis. 1 Known only from the Yukon Delta. 56 NORTH AMERICAN FAUNA. [NO. 19. Colaptes auratus luteus. Selasphorus rufus. Contopus richardsoni saturatus. Empidonax hamiiiondi. Sterna paradissea. Corvus corax principalis. Ammodramus sandwichensis alaudinus. Spizella monticola ochracea. Hirundo erythrogastra unalaschkensis. Tachycineta bicolor. Dendroica coronata hooveri. Dendroica townsendi. Wilsonia pusilla pileolata. Anthus pensilvanicus. Sitta canadensis. Hylocichla aonalaschkse. Merula migratoria. Hesperocichla nsevia. SPECIES OCCURRING ON WHITE PASS SUMMIT AND IN THE YUKON VALLEY. Sayornis saya yukonensis. Zonotrichia coronata. ? Wilsonia pusilla pileolata. Anthus pensilvanicus. Hirundo erythrogastra unalaschkensis. SPECIES FOUND BY US ONLY ON WHITE PASS SUMMIT. Lagopus rupestris. Lagopus leucurus. Zonotrichia coronata. Leucosticte tephrocotis littoralis. SPECIES FOUND BY US ONLY ON THE COAST OP SOUTHEAST ALASKA. ? Dendragapus obscurus fuliginosus. Sphyrapicus ruber. Cyanocitta stelleri. Corvus caurinus. Junco hyemalis oregonus. Junco hyemalis connectens. Melospiza melodia rufina. Melospiza lincolni striata. Passerella iliaca townsendi. Helminthophila celata lutescens. Anorthura hiemalis pacifica. Parus rufescens. Regulus satrapa olivaceus. Regulus calendula grinnelli. LAND BIRDS FOUND IN LYNN CANAL DISTRICT ONLY NEAR TIDE WATER. Contopus richardsoni saturatus. Sphyrapicus ruber. Cyanocitta stelleri. Corvus caurinus. Spizella monticola ochracea. Junco hyemalis oregonus. Melospiza melodia rufina. Melospiza lincolni striata. Tachycineta bicolor. Helminthophila celata lutescens. Anthus pensilvanicus. Sitta canadensis. Merula miarratoria. LAND BIRDS FOUND IN LYNN CANAL DISTRICT ONLY NEAR THE LEVEL OF GLACIER. Colaptes auratus luteus. ?Dendragapus obscurus fuliginosus. Junco hyemalis connectens. Wilsonia pusilla pileolata. Cinclus mexicanus. Anorthura hiemalis pacifica. Regulus satrapa olivaceus. SPECIES RECORDED FROM THE UPPER YUKON ONLY ABOVE THE PELLY RIVER. Tringa minutilla. Symphemia semipalmata inornata. ? Megascops asio kennicotti. ? Dryobates villosus hyloscopus. Chordeiles virginianus. Selasphorus rufus. Contopus borealis. Hirundo erythrogastra unalaschkensis. Tachycineta bicolor. Helminthophila peregrina. Dendroica townsendi. Wilsonia pusilla pileolata. Hylocichla aonalaschkse. Hylocichla aonalaschkpe pallasi. Sialia arctica. ocr.,1900.] BIRDS OF THE YUKON REGION. 57 SPECIES RECORDED FROM THE YUKON VALLEY BETWEEN POET YUKON AND TOE PELLY R1VEK. Fako peregrinus auatum. Acanthis hornemanni exilipes. ?Falco columbarius. Passerella iliaca. ?Falco columbarius richardsoni. Seiurua noveboracensis notabilis. Empidon&x trailli alnorum. Wilsonia pusilla. < >tocoris alpestris leucoleema. Hylocichla alicise. Calcarius lapponicua alascensis. Saxicola oenanthe. ANNOTATED LIST OF SPECIES. 1. iEchmophorus occidentalis. Western Grebe. Several seen at Bocadequadra, near Dixon Entrance, May 28. 2. Colymbus holboelli. Holboell Grebe. A young male was taken on the 'Canal' at St. Michael September 22. The irides were primrose yellow; basal two-thirds of the culmen, outside tarsi, and lobes, seal brown; rest of bill, ocher 3^ellow; inside of the tarsi and lobes, maize yellow; nails, yellowish olive buff. 3. Gavia imber. Loon. Seen at Bocadequadra May 28 and in the Inside Passage May L".». Several seen on Lake Bennett and a pair at Caribou Crossing between June 17 and 28. On Lake Marsh they were common and were fre- quently heard, especially at night. The last loon certainly referable to this species was seen there July 6. 4. Gavia arctica. Black-throated Loon. A loon that flew over our boat on Thirty-Mile River Jul}- 18, and another seen near Big Salmon River July 20, I believe were Garni a nt lea. I saw several loons at the Aphoon mouth of the Yukon August 27 and one at St. Michael on September 5 and 16. We obtained none of them, but the experience of others makes it probable that all were the black-throated. Dr. Romig, of the Moravian Mission on the Kuskokwim River, told me that his party killed two on August 27 on the portage from Bethel on the Kuskokwim to Hendricks Station on the Yukon Delta. 5. Lunda cirrhata. Tufted Puffin. Osgood saw one at Whale Island, near St. Michael, September 8. 6. Fratercula corniculata. Horned Puffin. We took two and saw about a dozen puffins near Whale Island Sep- tember 8. Irides, drab gray; ring on eyelid and lip of bill, flame scarlet; rest of bill dull straw yellow; bare skin at gape and line along base of maxilla, cadmium yellow; line below lower eyelid and horns, black; palmations, cadmium orange; tarsi and toes, cadmium orange above, chrome yellow below; nails varying from drab gray to slate color. 58 NOETH AMEKICAN FAUNA. [no. 19. 7. Simorhynchus pusillus. Least Auklet. Anklets were seen several times while we were crossing Bering Sea in the Corwin October 1-2 and increased in numbers as we approached the Pribilofs. They were common with various other (unidentified) species of water birds off Unalaska October 1 and abundant in Akutan Pass October 6. I refer them to this species, as Nelson found it the most abundant in these waters. 8. Brachyramphus marmoratus. Marbled Murrelet. This bird was fairly common in the Inside Passage May 28-29, and one was killed at Bocadequadra. We saw a few on Lynn Canal May 30, and I shot one near Skagway May 31. Doubtless some of the many murrelets seen with auklets near the Pribilof and Aleutian islands in October were this species. 9. Cepphus columba. Pigeon Guillemot. Seen at Bocadequadra and along the Inside Passage May 28-29. Guillemots which I saw near Unalaska October 1 were probably this species. 10. TJria lomvia arra. Pallas Murre. The murres seen near St. Michael August 29 and about St. George Island and Unalaska in October were probably chiefly this species, though some may have been TJria troile calif ornica. 11. Stercorarius parasiticus. Parasitic Jaeger. Common at the Aphoon mouth of the Yukon August 27-28, and alx>ut St. Michael until September 10. About this time their num- bers decreased, and the last one was seen September 16. All appeared to be adults (as were the four collected), and only one was in the black plumage. 12. Stercorarius longicaudus. Long-tailed Jaeger. I saw one at the Aphoon mouth August 28, and both Osgood and I occasionally saw the species at St. Michael until September 12. 13. Rissa tridactyla pollicaris. Pacific Kittiwake. Adult kittiwakes were tolerably common at St. Michael from Sep- tember 19 to the end of our stay, but no young were seen. As we crossed Bering Sea October 1-5, and at Unalaska October 5-6, young kittiwakes were common, and we saw no adults except at St. George and Unalaska. The irides of the adult are vandyke brown; ring on eyelid orange rufous; bill sulphur yellow, whitish at tip; gape rufous; tarsi, toes, palmations, and nails slate black. 11. Rissa brevirostris. Red-legged Kittiwake. One was seen by Osgood at Unalaska (Dutch Harbor) October 5. oct.,1900.] BIRDS OF THE YUKON REGION. 59 15. Larus barrovianus. Point Burrow Gull. Abundant on the Lower Yukon, at the Aphoon mouth, and during September at St. Michael, though most of the adults had gone by the middle of the month. While crossing Bering Sea we saw several young October 2 and others near Unalaska October 1. A young bird shot near St. Michael September 19 had the head of a recently killed ptarmigan in its throat. The irides of the young are Prout's brown: tip of bill and sides of nails black; rest of bill, toes, and pal- mations vinaceous buff; rest of nails drab gra}^. 16. Larus glaucescens. Glaucous-winged Gull. Large gulls, which doubtless were chiefly this species, were common from Dixon Entrance to Lynn Canal May 28-30, and we saw a few near Skagway June 1-2. At Unalaska, where I collected two, they were abundant October 1—6. A few gulls that followed the Oorwim in the North Pacific I think also belonged to this species. IT. Larus argentatus smithsonianus. American Herring Gull. The only large gulls I took on the Yukon — a female which had fin- ished laying, collected at Lake Tagish June 30, and another taken near Charlie Creek August 8 — were this species, and no others came close enough to make identification positive; hence I must refer all the large gulls seen to Larus a. smithsonicmus, although on several we could see no black on the primaries. I saw one flying over White Pass Summit June 12 and another at Bennett City June 19. We saw eight or ten at Caribou Crossing and a few on Lake Tagish. No more were observed until we reached Lake Lebarge July 13; but from this point to the mouth of the river large gulls slowly became more numerous, one or two being noted every few days. Three fully grown young, with their parents, were seen on a sand bar about 15 miles above Circle August 12. 18. Larus brachyrhynchus. Short-billed Gull. Our acquaintance with this bird dates from our arrival at Lake Marsh, July 1, where we found it common, and took downy young the next day. From this time, until we reached the Tatchun River, July 23, hardly a day passed that we did not see several; on July 20 we counted fourteen on a sand bar near Little Salmon River. After July 23 we saw no more until September 6, when young of the year appeared at St. Michael, and were common there until the 23d. The only adult seen at St. Michael was noted on September 11. The adult has the irides Prout's brown; ring on eyelids and skin at commissural angle reddish orange; gape orange; bill, tarsi, and toes olive j'ellow; nails black, french gray at base. Natal plumage: Creamy white, becoming pale cream color on fore- head, chin, and anterior breast, mottled with different shades of brown, 60 NORTH AMERICAN FAUNA. [no. 19. except the center of chest and abdomen. Head markings slate-black, distinctly defined and numerous, the most characteristic being one that covers the entire nasal region, a V on the pileum, a W on the occi- put, and a somewhat interrupted U on each side of the throat. On the upperparts the markings become pale seal brown, and with lighter tips render the lower neck, sides of breast, flanks, and anal region grayish. Bill brownish black; tip of bill, tarsi, toes, and palmations whitish; nails and edges of scutelloe of tarsi and toes hair brown. 19. Lams Philadelphia. Bonaparte Gull I saw several small black-headed gulls, probably this species, in the Inside Passage May 29. I took a Bonaparte gull at Caribou Crossing on June 24 and saw several others. We saw one on Lake Marsh July 1, a few young at St. George Island October 3, and found them common at Unalaska October 4-5. 20. Xema sabinii. Sabine Gull. Osgood found a dead bird of this species on the shores of Chilkat Inlet June 1. The specimen, unfortunately, was not in a condition to permit its preservation, but it was carefully identified at the time and showed no apparent variance from the description and figure in Ridg way's Manual. 21. Sterna paradissea. Arctic Tern. We saw a large flock of terns in the Inside Passage May 29, and two days later at Skagway saw a few more, securing two, which proved to be of this species. At Bennett, between June 15 and 20, we fre- quently saw two or three, and I was informed that arctic terns bred on a small lake near Log Cabin, British Columbia. We found a breed- ing colony of about twenty on a small rocky island lying in the entrance to Windy Arm, Lake Tagish, July 1. I found four single eggs (three fresh and one well advanced in incubation), one set of two (one fresh and the other at point of hatching), and also a young bird which had just left the shell. There were no nests; the young bird and eggs were in the short grass on the top of the island. Except a single bird, seen at Lake Marsh and probably belonging to this colony, we did not meet with terns again until August 27, when I found this species common at the Aphoon mouth. A single tern with injured primaries was seen frequently at St. Michael up to September 21. The downy young differs from the description given in Baird, Brewer and Ridgway's 'Water Birds,' in having the forehead plain dusky, the chin whitish, the basal half of bill, tarsi, and toes salmon pink, and the rest of bill and nails black. 22. Diomedea albatrus. Short-tailed Albatross. A dark-brown albatross, probably the young of D. albatrus, joined the Corwin October 1, about 150 miles from St. Michael. It was soon OCT., 1900. BIRDS OF THE YUKON REGION. 01 accompanied by others, and until we reached Cape Scott, October 12, a glance astern would seldom fail to show two or three following the vessel. 23. Fulmarus glacialis glupischa. Pacific Fulmar. A single dark-colored fulmar, possibly this form, was seen ( October 1, between St. George and Unalaska. 21. Oceanodroma furcata. Forked-tailed Petrel. To this species 1 refer a few light-colored petrels seen October 3, on Bering Sea north of the Pribilof Islands. 25. Phalacrocorax pelagicus. Pelagic Cormorant. Cormorants were seen October 1 near Unalaska, where this species is reported as common. 26. Phalacrocorax pelagicus robustus. Violet-green Cormorant. We saw a single cormorant at Whale Island September 8; and one — possibly the same bird — was seen by Osgood several times at St. Michael. 27. Phalacrocorax urile. Red-faced Cormorant. This is the only cormorant reported by William Palmer from St. George, where we saw several October 3. 28. Merganser americanus. American Merganser. A pair of mergansers was breeding on a small, rocky island in Lake Tagish, at the entrance to Windy Arm, June 30-July 1. The nest was found by Osgood in a crevice in the cliffs about 15 feet above the water. It was made of down, and contained seven eggs about one week advanced in incubation. Retrieving would have been impossible had we shot the bird, but as I succeeded in watching the female on the nest from a distance of less than 6 feet I feel positive of the species. A few other mergansers, usually in pairs, were seen on Lake Tagish July 1, on Lake Marsh July 8, at Fifty-Mile River July 9 and 12 (a flock of a dozen males flying up the river in the evening of the latter date), near Little Salmon River July 20, and about 25 miles above Circle August 12. Near Charlie Creek we found the dried wing of an adult male of this species August 8. 29. Anas boschas. Mallard. On the flats of Chilkat Inlet I saw seven June 2. In no part of the Yukon Valley above Circle did we find ducks abundant, except surf scoters, but the mallard undoubtedly occurs at all suitable places throughout the region. It must breed very early, as on June 21, only three weeks after the lakes were open to steamer navigation, I found a female with two young at Caribou Crossing, and on June 28 I shot 62 NORTH AMERICAN FAUNA. [no. 19. another female there and caught two of her half dozen downy young. Two ducks, probably mallards, were seen on Lake Marsh July 6, and at Miles Canyon Maddren was informed they had been common there earlier in the season. We saw several females with young in the marshy ponds at the foot of Lake Lebarge July 17, a few adults near the Little Salmon July 20, and a good-sized flock near Charlie Village August 10. Osgood shot one near Fort Yukon August 21. In the large flocks of geese and ducks disturbed by the steamer on the Lower Yukon were two young mallards, secured at Hendricks Station August 25. Mallards were common at the Aphoon mouth August 27, and we saw a few at St. Michael September 2. 30. Mareca americana. Baldpate. Five ducks that I took to be baldpates were seen a short distance above Fort Selkirk July 25. 31. Nettion carolinensis. Green- winged Teal. Three teal that 1 saw in the creek at Circle, August 19, were prob- ably this species. Green-winged teal were common in the tundra ponds about St. Michael during the first half of September, but apparently did not occur after September 16. All that were taken were young birds. 32. Dafila acuta. Pintail. Maddren was told at Miles Canyon, July 11, that pintails were com- mon, but we saw none near enough for identification until August 27, when I found them abundant at the Aphoon mouth. Seven were here killed by a passenger on the steamer. During September young, pintails far outnumbered all other ducks on the marshes and tundra ponds about St. Michael. Large numbers were killed by the Eskimos, but no adults were seen. Their numbers had greatly decreased by September 20. 33. Aythya marila. Scaup Duck. We saw a flock of about a dozen adult males at Caribou Crossing June 24, and another of about twenty on the Yukon, a short distance above Fort Selkirk, July 25. 31. Aythya affinis. Lesser Scaup Duck. We found a pair with young on a small pond at Lower Lebarge July 17. Osgood secured the female. 35. Clangula clangula americana. American Golden-eye. I am confident that a flock of ducks seen about 25 miles above Circle August 12 were males of this species or of C. islandica. oor.,1900,] BIRDS OF THE YUKON REGION. (53 36. Charitonetta albeola. Buffle-head. I shot a female on a small pond near Lake Marsh July 8, and saw a male near Little Salmon River July 20. Maddren was informed that bume-heads were common near Miles Canyon, and a boy at Lower Lebarge said they bred commonly on the ponds near there, and that he took two young- July 16. 37. Harelda hyemalis. Old Squaw. Single young birds were found frequently during- September in the small ponds about St. Miehael, and a flock of about a dozen was seen in the harbor September 11. No adults were observed. One young bird, taken early in September, still retained natal down on the hind- neck. 38. Histrionicus histrionicus. Harlequin Duck. We saw a male and two females in Wrangell Narrows May 2!>. A Hock of twelve males came eloso to the shore at Bennett June 18; and on June 23 a single male swam so near that men sitting on the beach threw stones at it. One other harlequin was seen a few miles above Fort Selkirk July 25. Dr Romig told me he saw a number on the portage from the Kuskokwim to the Yukon August 24-25. 39. Somateria v-nigra. Pacific Eider. We saw the head of a male of this species lying in the window of the hotel at St. Michael, and the soldiers at the barracks had a mounted bird, shot near St. Michael in the spring, but we saw no living eiders of any species during our trip. 40. Oidemia americana. American Scoter. We noticed a few in Wrangell Narrows May 29, and I saw a number off Unalaska October 5. 41. Oidemia deglandi. White- winged Scoter. This species was fairly numerous at Bocadequadra, Wrangell Nar- rows, and Lynn Canal May 28-30. We saw two on Lake Marsh July 6, two on Lake Lebarge July 14, and a flock of about twenty-five Hying up Fifty-Mile River from Lake Lebarge on the evening of July 12. 42. Oidemia perspicillata. Surf Scoter. In Lynn Canal, near Haines, June 1 we noted a large flock of surf scoters, most of which had disappeared the next day. They were abundant on all the Yukon lakes except Bennett, which was almost destitute of bird life. On Lake Tagish we saw fourteen June 30, and at Lake Marsh thirty to forty males almost every day between July 1 and 8. We saw no more, except a pair on July 11 on Fifty-Mile River, which connects Lake Marsh with Lake Lebarge, until we 64 NORTH AMERICAN FAUNA. [no. 19 entered Lake Lebarge on the evening of Jul}7 12, when a flock of at least a hundred flew high overhead from the direction of the lake. About 8 p. m. and at 10 p. m. of the same evening, and on the next morning, we saw what we took to be the same flock. The birds were probably taking a morning and evening flight, such as E. 8. Bryant has noticed in the case of the white- winged scoters breeding at Devils Lake, North Dakota; and I believe that with both species these flights are taken chiefly to exercise the wing muscles. We saw no females on any of the lakes, nor could we find them on the shore, though they were undoubtedly nesting in the vioinity. We observed several on Thirty-Mile River July 18 and two near the Little Salmon July 20. Near Whale Island, at St. Michael, we saw a number September 8, and two scoters, probably young of this species, September 21. I think there were a few with the American scoters I saw at Unalaska October 5. 43. Chen hyperborea. Lesser Snow Goose. 1 saw live snow geese at the Aphoon mouth August 28, and a large flock at St. Michael September 11. 44. Anser albifrons gambeli. American White-fronted Goose. A single white-fronted goose was seen by Osgood among a number of other birds killed by natives about the Yukon Delta August 29. 45. Branta canadensis hutchinsi. Hutchins Goose. Although Maddren was informed that a goose with four young was seen near White Horse Rapids about July 11, and although the sergeant in charge of the police station of Lower Lebarge told us that thousands of geese and ducks passed there in the spring, and that he had counted twenty -four distinct species, and had killed both Hutchins and cackling geese, we did not see a goose of any species until we were in the neighborhood of Charlie Village, August 10. There we saw a flock of about twenty of the Branta canadensis group, and Osgood shot two hutchinsi and saw many more near Fort Yukon August 21. Brown geese, apparently chiefly this subspecies, were common on the Yukon flats and on the lower river, especially the Yukon Delta. A Hutchins goose was brought to the steamer Robert Kerr by an Eskimo August 26, and I found the bird common at the Aphoon mouth August 27-28. Prospectors on the Kerr told me that geese bred abundantly at the head waters of the Porcupine and the marshes at the source of Birch Creek. During September this species was common about St. Michael in small flocks, but very shy; Osgood took one September 23. [Philacte canagica. Emperor Goose. Dr. Romig told me they were common on the tundra along the Kuskokwim.] OCT.,1900.] BIRDS OF THE YUKON REGION. 65 | Olor colwmbiamis. Whistling Swan. We were told that a swan — probably this species- was killed at Circle during the spring-. J lo. Grus canadensis. Little Brown Crane. Along the Yukon we did not see any cranes, although I thought I heard one near the Little Salmon July 21, and a man who had spent, the summer at Circle told me he had heard and seen the 'sand- hill crane' there frequently during the past two months. I was also informed by prospectors that these cranes were found in small num- bers at the head waters of Birch Creek and Porcupine River. Near St. Michael we saw flocks of from two to six individuals each almost daily during the first half of September, but none later than Sep- tember L5. On the night of September 13 and all the following day there was a hard southwest gale. On the 11th we saw large num- bers— Osgood counted ninety-six — flying south, high in the air. IT. Crymophilus fulicarius. Red Phalarope. We saw a small flock near Skagway in Lynn Canal June 2, and others I believed to be this species near Wrangell Narrows and in Prince Frederick Sound May 29. Osgood took one at St. Michael September 17 during a heavy storm. IS. Phalaropus lobatus. Northern Phalarope. Large flocks were seen near Dixon Entrance May 28, and smaller ones on the Inside Passage May 29. From a flock of about twenty on Lake Lebarge July 13 I shot a female that was changing to winter plumage, and on a small pond at Lower Lebarge July 17 I took a male that was in worn breeding plumage. At St. Michael September 2 I caught a young bird that had but one wing, and on St. George Island October 3 I shot one that was swimming alone in a pool. Phalaropes, probably this species, were seen on Bering Sea October 1 and 1. 19. Gallinago delicata. Wilson Snipe. At Haines May 31 I was told that several Wilson snipe had been seen that dajT, but was unable to find them. We saw one on Fifty- Mile River not far below Lake Marsh Juty 10, and another in the marsh at Lower Lebarge July 17. Osgood saw one at Circle August 18, and I killed two from a small flock at Hendricks Station August 25. At St. Michael we saw eight or ten single birds between Sep- tember 12 and 22. 50. Tringa couesi. Aleutian Sandpiper. Common about the lava rocks that line the shore at St. Michael, where flocks of five to fifty were observed, but only small flocks after September 15. A few were occasionally seen on the tidal mud flats, H:91r— No. 19 5 66 NORTH AMERICAN FAUNA. [no. 19. but none about the ponds in the interior of the island or on the salt meadow behind the town. Out of eighty specimens taken only eight were adults, and five of these were taken before September 9. On the rocky shores of a point opposite Dutch Harbor, Unalaska, I found them common October 5. Those taken at St. Michael were molting into first winter plumage, which is practically complete in the Una- laska birds. In this plumage there is considerable individual differ- ence in the width and shade of the pale edgings of the feathers of the upperparts. The irides were Vandyke brown; bill, black changing to olive buff in basal half; tarsi and toes, yellowish olive buff washed with black; nails black. I find great sexual variations in size in this species, the females, as in many other species of Limicolse, averaging considerably larger, especially in length of bill. Measurements of twenty-nine males: Length 8.06 to 8.94 (average 8.57) inches; wing 4.37 to 5.12 (average 4.89) inches; exposed culmen 0.96 to 1.13 (average 1.06) inches; tar- sus 0.91 to 1.03 (average 0.96) inches. Measurements of thirty-four females: Length 8.56 to 9.56 (average 9.03) inches; wing 4.47 to 5.31 (average 4.98) inches; exposed culmen 1.16 to 1.42 (average 1.24) inches; tarsus 0.96 to 1.05 (average 0.99) inches. 51. Tringa ptilocnemis. Pribilof Sandpiper. We saw a number on St. George October 3, but too close to the rookery of fur seals to be obtained. 52. Tringa acuminata. Sharp-tailed Sandpiper. First found September 18, when six were seen with a large flock of T. a. pacijica at St Michael. We did not see more than a dozen of this species during the rest of September. Although the species has not been hitherto recorded from St. George Island, we took three and saw about a dozen during the short time we were there October 3. At Unalaska, October 5, I secured one which was with T. couesi on the rocky beach. The irides were vandyke brown; maxilla and distal half of mandible, dark seal brown, mandible changing to dull olive buff at base; gape ecru drab; tarsi and toes, greenish maize yellow; nails black. 53. Tringa maculata. Pectoral Sandpiper. This species was present throughout our stay at St. Michael, usually associating with flocks of T. a. paciflca, but in very small numbers, not more than twenty being seen. All the specimens taken were young birds. Osgood took one at St. George October 3, and 1 one at Unalaska October 5. ocr.,1900.1 BIRDS OF THE YUKON REGION. <*>7 54. Tringa bairdi. Baird Sandpiper. Two .sandpipers, probably of this species. Hew by us <>n Lake Marsh, and we saw tour more in the marshes of Lower Lebarge, but failed to secure an}' of them. I shot one young bird near the Tahkandik River August 7. Osgood shot one from a flock of four at Circle August 1 5, and another near Fort Yukon August 21. 55. Tringa minutilla. Least Sandpiper. At the southern end of Lake Marsh, not far from where Six-Mile River enters, the surrounding country is level, and at high water the lake stretches far back through a dense growth of willows. At the time of our visit the retreating water of the lake had left a bell of grass between these willows and its margin. Here on the evening of July 2 I found three pairs of least sandpipers, and after a long search, somewhat interrupted by hordes of mosquitoes, I came upon a female surrounded by four downy young*. Both parents tried time and again the well-known 'wounded-bird' tactics to lure me from the spot where the young were hidden in the bunches of grass, and, finding this a failure, would circle around me only a few yards off, uttering a plaintive twitter. I saw two other least sandpipers on the west shore of Lake Marsh July 7. Natal plumage: Lower parts, forehead and orbital region, brownish white. Upperparts bright cinnamon rufous mottled with black; many feathers, especially on head, rump, and tail, tipped with white. Post- orbital line and loral line blackish, and spot of bright cinnamon rufous on sides of chest. Irides dark brown; bill and nails, slate black; tarsi and toes, pale slate. 56. Tringa alpina pacifica. Red-backed Sandpiper. Young red-backed sandpipers were very abundant at St. Michael during our stay, many times outnumbering all other Limicohe. Early in September they frequented chiefly the mud flats on the coast, but after the middle of the month large numbers were found only about the pools of the salt marsh. September 24, when the tundra was quite thoroughly frozen, with snow in every hollow and a skimming of ice on the pools, I saw at least one hundred in this latter place. In several taken early in September the back of the neck was still covered with down, but the majority were in full juvenile plumage. Sonic -till retained this plumage at the time we left St. Michael, but the larger portion had molted into winter plumage. Only two adults weir taken, September 1 and 5. A few were seen at St. George Island October 3. 57. Calidris arenaria. Sanderling. 1 saw three at St. Michael September 11 and collected one. which proved to lie a young female. 68 NORTH AMERICAN FAUNA [no. 19. 58. Totanus flavipes. Yellow-legs. On July 1, while floating down Six-Mile River close to its entrance into Lake Marsh, we were attracted by the anxious cries of a pair of yellow-legs. Osgood shot both birds, and we found two downy young in the grass on the shore of the river. Entering Lake Marsh we heard a yellow-legs' whistle, and on July 2 I saw a yellow-legs near where I found the least sandpiper. I collected a female on the west shore of Lake Marsh July 8, and a male, the last bird of this species seen, near a small pond at Lower Lebarge July 17. Both these birds undoubtedly had eggs or young close by, for they alighted exclusively in trees, scolded vociferously, tilting the body with each cry, and refused to leave. Bare spaces on the breast show that both sexes assist in incubation. Natal plumage: Upperparts and thighs, dark seal brown, many of the feathers tipped with cream buff and whitish; longitudinal lines on rump, cream color, inclosing central, seal-brown space. Forehead, buffy white, extending in narrow lines on sides of crown to occiput, and in broader lines above eye to nape, the latter crossed by trans- verse dark lines extending from eye to occiput. Line beginning at base of culmen enlarged to dark space on crown and occiput, extend- ing down neck to back, seal brown; other dark lines extending from crown above eye to occiput, and from nostrils through eye to nape. Throat and center of abdomen silvery white; rest of lower parts and sides of neck, buffy white; each feather of lower parts becoming brownish black at base. Irides, Vandyke brown; bill, black at tip, changing to greenish olive at base; tarsi and toes, yellow, paler than in adult, and mottled with brown; nails, brown. The juvenile plumage is appearing, in this specimen, on wings, wing coverts, chest, and sides. 59. Helodromas solitarius cinnamomeus. Western Solitary Sandpiper. At Log Cabin, British Columbia, on the evening of June 14, we noticed a sandpiper wheeling through the air, like the woodcock at its breeding place, occasionally uttering a rather musical whistle. The next morning I found it feeding in a small swamp. It proved to be a solitary sandpiper, as I had suspected on the previous evening. Osgood saw another near Lake Marsh July 5, and I saw two near Little Salmon River July 21. On July 8, after rowing a few miles down Lake Marsh, we stopped for lunch on the west shore, where a forest tire had killed most of the trees, and fallen trunks piled in end- less confusion, brush, small pools, and hordes of mosquitoes rendered the place anything but a paradise. Here 1 startled a solitary sand- piper and a yellow-legs at the same instant. They lighted on the half- fallen trees and scolded me, tilting their bodies at each cry. The solitary sandpiper, which doubtless had a nest there, differed chiefly from eastern specimens of solita/rius in having dark, wavy markings oct., 1900.] BIRDS OF THE YUKON REGION. 69 on inner webs of outer primaries. Osgood took a typical young of dnnamomeus and saw another on an island near Sixty-Mile Creek August 1. 60. Symphemia semipalmata inornata. Western Willet. While in a meadow a short distance back from the southeast end of Lake Marsh July 2 I heard a willet whistle several times its unmis- takable 'pill-willet,' but failed to see the bird. 61. Heteractitis incanus. Wandering Tattler. Osgood took an adult at Skagway May 31. I shot a young bird from a flock of three at St. Michael September 1, saw one on Whale Island September 8, and secured two at Unalaska October 5. The irides of the adult were vandyke brown; bill, black, base of mandible brownish; tarsi and toes, brownish ocher; nails, black. In the }Toung, the bill changed from black to sage green in basal third of mandible, and to greenish olive at base of maxilla; tarsi and toes, dull gallstone yellow, greenish at joints. 62. Actitis macularia. Spotted Sandpiper. 1 saw one at Skagwa}' June 3, and Osgood one at Glacier June 8. This is preeminently the shore bird of the Yukon Basin; we saw two at Bennett June 18, and until we reached Circle, August 15, hardly a day passed without our seeing many running along the shore, or skimming over the river. They were especially abundant between White Horse Rapids and Lake Lebarge. After the 1st of August most of the spotted sandpipers seen seemed to be traveling upstream in small flocks. We saw no adults after August 4. The first set of eggs wras found at Caribou Crossing June 27; the last at the Tatchun River July 23. The first young noticed were in a nest containing three young and one pipped egir found on Lake Marsh July 7. Both sexes were incubating. Nests were close to the shore, and also on small rocky islands in the lakes. 63. Numenius hudsonicus. Hudsonian Curlew. I secured one from a flock of four curlews on the marshes of Chilkat Inlet, and Osgood found a dead bird in the woods at Haines, June 1. Three young were brought to the steamer by an Eskimo at the Aphoon mouth August 28. I saw one at St. Michael September 2, and, I think, another September 14. Adult: Irides, vandyke brown; bill and nails; black; tarsi and toes cinereous. Young (Massachusetts specimen): Irides, raw umber; maxilla, black; mandible, clove brown, blackish at tip, vinaceous toward base; tarsi and toes, olive gray; nails, black. 70 NORTH AMERICAN FAUNA. [no. id. 61. Squatarola squatarola. Black-bellied Plover. At the Aphoon mouth of the Yukon I saw a flock August 28. Osgood saw three young which had been shot on the mainland near St. Michael September 10, and from this date to the end of our stay we saw occasionally one or two birds of the year, one of which was taken September 16. 65. Charadrius dominieus fulvus. Pacific Golden Plover. None were seen until September 16, after which young birds became fairly common on the boggy tundra about St. Michael and the mud flats along the shore. The only adult seen was taken by Osgood Sep- tember 25. We saw a number of young birds on St. George Island October 3, and Osgood secured one. Crossing Bering Sea we saw some near Unalaska October 4, and 1 saw one on October 8, when we were several hundred miles south of the Aleutian Islands. This bird flew several times around the Corwin, answering my every whistle, and seemed anxious to alight. The specimens collected differ greatly in the amount of the golden coloring, but all are far more golden than Massa- chusetts skins of dominicus, and all have the shorter wings of fulvus. Irides, Vandyke brown; bill and nails, black; tarsi and toes, slate gray. 66. iEgialitis semipalmata. Semipalmated Plover. Osgood collected a male at Caribou Crossing June 21. and a pair of adults and one pipped egg at the southern end of Lake Marsh July 2. I removed the young bird from the shell, and within half an hour the down was almost dry, the eyes were open, and it could hop about on its 'knees.'1 Maddren took another adult at this place July 6, and I a female and four eggs nearly hatched, on the west shore of Lake Marsh on the same day- The nest was a hollow, lined with a few grasses and dead leaves, and was situated about 8 feet from the water in the drift debris among the stones of the beach. We saw three or four on a sand flat near Charlie Village August 10; a few about 15 miles above Circle August 12, and the last at Circle August 15. Bare pectoral spaces showed that both sexes assist in incubation. Natal plumage: Lower parts, white, separated by broad bare space on neck, changing to cream color on lower tail coverts. Above, cream -color, mottled with black, changing to buff on wings and tail. Fore- head and infraorbital patches, cream color; broad band on neck encircling head, white, bordered above by narrow band of black extending from bill around occiput, and connecting in malar region with black line leading to inner canthus of eye. Spot on forehead, on sides of chest at lower border of bare space, on sides and on flanks, black. Irides, dark; bill and nails, black; tarsi and toes, slate color, whitish posteriorly. OCT., 1900.1 BIRDS OF THE YUKON REGION. 71 67. Arenaria melanocephala. Black Turnstone. We found a small flock on the rocky .shore at St. Michael August 31; I took three young- there the next day, and on September ."• 1 saw a single turnstone flying across the marsh. On St. George I slant I. October 5, we saw a number of birds that we had no doubt were black turnstones, but I do not tind this species recorded from the Pribilofs, and we were unable to obtain specimens. Irides, Vandyke brown; bill, olive black; tarsi and toes varying- from clay color to vinaceous cinnamon, and washed with black; nails, black. 68. Dendragapus obscurus fuliginosus. Sooty Grouse. We were told that grouse were common on the heights above Skag- way. but although we often found droppings we saw no birds, and the spring "calling' of the male had ceased. Maddren and J heard a bird that must have been this species "booming-1 far up on the hillside from the ravine above Glacier June 8. 69. Canachites canadensis osgoodi. Alaska Grouse. Canachites canadensis osgoodi Bishop, Auk, XVII, 114, April, 1900. We first met the Alaska grouse at Bennett City, where Osgood shot a laying female June 22. At Caribou Crossing he found feathers of this grouse in a magpie's nest and in one of his mammal traps. At Lake Marsh he shot four females and four young July 4-5, at Lake Lebarge a female July 11, at Lower Lebarge a female and one young July IT, and on Thirty -Mile River an adult male July 19. He found the birds frequenting the thickets of poplars and young spruces and remarkably easy to approach. I saw a male at Lake Lebarge July 16, and shot a well-grown young near the Tatchun River July 16, but did not meet with the species elsewhere. This bird was reported as common at Lower Lebarge by the police sergeant stationed there: at Rampart City by Mr. Burkman, and along the Kuskokwim by Dr. Romig. 70. Bonasa umbellus umbelloides. Gray Ruffed Grouse. I secured a female and one young bird on the west shore of Lake Lebarge July 14, and another female that had a brood of young, two- thirds grown, at Lower Lebarge July 17. Osgood took a .young bird from a covey near Rink Rapids Juty 22. The sergeant at Lower Lebarge called this species rare, but 1 was told it was common near Rampart City. 71. Lagopus lagopus. Willow Ptarmigan. Two flocks were found on the tundra at the Aphoon mouth August 28, one alighting close to the steamer. Not seen at St. Michael until September 11, when about one hundred appeared. These were seen frequently after this date, but were exceedingly shy. Most of those 72 NORTH AMERICAN FAUNA. [no. 19. taken were young- birds, and all were in full molt. The irides of a young male taken September 19 were vandyke brown; skin above eye, rufous; bill, slate black, whitish at tip and salmon buff at base of mandible; nails, white. We were told that ptarmigan were very abundant near Atlin, British Columbia, at the head waters of the Porcupine River and Birch Creek, near Rampart City, along the Kuskokwim, and in winter at Glacier and Lower Lebarge. Doubtless some of these statements refer to the following species. 72. Lagopus rupestris. Rock Ptarmigan. At White Pass Summit, June 11 and 13, we took three males still in white plumage (excepting a few dark feathers on head and lower neck), and saw a few others. Osgood found two eggs there, probably of the previous year, lying on the moss under an alpine hemlock. Dr. Romig told me that this sDecies was more common than L. lagopus along the Kuskokwim. [Lagopus rupestris nelsoni. Nelson Ptarmigan. We were told at Unalaska that this species had been abundant during the summer on Unalaska Island, but that the birds had been almost exterminated by the officers of an English man-of-war. We saw none during the day and a half we were there.] 73. Lagopus leucurus. White-tailed Ptarmigan. Osgood took a white-tailed ptarmigan June 8 on the summit of the cliffs above Glacier, and saw several other ptarmigan, probably of this species. On June 8 he found at the same place, on the moss under an alpine hemlock, fragments of two ptarmigan eggs, sparingly dotted with brown as in leucurus. 71. Circus hudsonius. Marsh Hawk. We saw one at Lake Marsh July 8, one at Lake Lebarge July 12, a young bird on which duck hawks were feeding near the Tahkandik River August 7, one about 20 miles above Circle August 12, and two at Circle August 15 and 20. At the Aphoon mouth 1 saw several August 28. At St. Michael we secured a young bird September 2, and saw single marsh hawks on September 6, 7, and 11. The young bird taken is noticeably darker than young from Dakota and New England. 75. Accipiter velox. Sharp-shinned Hawk. I saw one at Lower Lebarge Jul}" 17, and two near White River July 30; Osgood found one feeding on a thrush near Charlie Creek August 9; at Circle I saw one August 17 and shot an adult female August 19. Osgood found a nest of this species, about 15 feet oct.,1900.] BIRDS OF THE YUKON REGION. 73 from the ground, in a small .spruce in the center of an island near the Nordenskiold River July 22, and 1 secured the female, whose crop held the tibia, tarsus, and toes of a nicker. The nest contained three downy, but very pugnacious young, one infertile egg, and the remains of a young intermediate sparrow. I kept two of the young alive until July 31, when both were well feathered and trying to fly and were as irascible as ever. The last survivor succeeded in getting out of his box w'lile we were moored at Dawson, flew into the Yukon, and was carried rapidly along by the current, though struggling valiantly to reach the shore. 1 suspect that it succeeded, as I heard a man who hurried after it say later that he would have ' fricasseed chicken for dinner.1 76. Accipiter atricapillus. American Goshawk. I saw an adult flying high above the shore of Lake Marsh July 8 with a mammal, probably a ground squirrel, in its talons. 77. Buteo borealis calurus. Western Red -tail. This is presumably the common hawk of the Upper Yukon; for the two large hawks taken are this species, and the numerous others seen resembled these in appearauce, flight, and cry. About half were in the melanistic plumage. Passing down Six-Mile River July 1 we saw three large buteos circling, and we noticed others frequently, usually in pairs, until we left Circle. Osgood and Maddren found a nest near Lake Marsh July 5 regarding which a pair of these birds were very solicitous. It was high in a spruce, and was empty except for a dead ground squirrel. On Fifty-Mile River July 10 I found a nest that was about 55 feet up in a spruce and contained two downy 3Toung. Osgood shot the female, which was in light plumage; the male, a melanistic bird, escaped. Osgood shot a melanistic female at Lower Lebarge July 17, and I found a pair — one light, the other dark — near Fort Selkirk July 25. These had a nest that was 60 feet up in a spruce and contained two young able to fly. I saw another nest with the birds about it on an island near the White River July 31. 78. Archibuteo lagopus. Rough-legged Hawk. On September 1, 6, 7, and 9 we saw at St. Michael large hawks which from their proportions and flight were either buteos or archi- buteos. Osgood shot one on Whale Island September 8, but could not retrieve it. Mr. Nelson's experience with the hawk family at St. Michael leads me to refer these birds to this species. 79. Haligeetus albicilla. Gray Sea Eagle. Lieutenant Satterlee, of the Corwm, found a dead binl of this species at Unalaska October 5, which proved to be a young female. 74 NORTH AMERICAN FAUNA. [no. 19. The wings had been removed at the carpal joint, but the unruffled plumage — the down yet persisting on the ends of the secondaries — removes all probability that it had been a caged bird. This is the first record of the occurrence of this species in western North America, although it is common in Japan and occurs in Kamchatka and occa- sionally on the Commander Islands. 80. Haliaeetus leucocephalus alascanus. Northern Bald Eagle. We found this bird common along the Inside Passage, especially near Wrangell Narrows, and from the steamer I noticed three occupied nests. We visited one which was high in a gigantic dead cedar on a small island near Bocadequadra. Broken shells at the foot of the tree made it probable that the nest contained young. The female parent was secured by Maddren. On the flats of Chilkat Inlet June 1 I saw 28 eagles feeding. Here I found another occupied nest at least 100 feet up in a living spruce (it was so high that heavy charges of No. 4 shot did the bird no harm). A man passing by shot the male with a rifle. The next day I saw the female again on the nest. In the interior this bird is much rarer, though I saw one at Log Cabin June 20, and another at Bennett June 19. We saw the birds occasionally about the lakes (I found a deserted nest on Lake Marsh), and once or twice along the river, the last being observed near the White River July 31. 81. Falco rusticolus. Gray Gyrfalcon. A female was caught in a steel trap set on a post at St. Michael September 21. Its stomach contained feathers. The irides were Vandyke brown; tip of bill and nails, black; tarsi, toes, cere, gape, and rest of bill, pearl gray, the bill changing to pearl blue on maxilla near commissure. 82. Falco peregriims anatum. Duck Hawk. At Fort Selkirk the character of the Yukon Valley changes, and the high, sandy bluffs which have been constantly visible on one bank or the other are frequently replaced by rocky cliffs of varying height. Flying about one of these cliffs near Stewart River Jul}7 31 was the first duck hawk we noted. From that point to the Yukon Flats, a few miles above Circle, a da}T seldom passed without our seeing or hearing them, and from Camp Davidson to Circle I think there was at least one breeding pair every 10 miles. We saw a number of their nests on shelves on the cliffs, but at this time, the first half of August, the young had flown. Osgood secured a young female August 5 on the cliff known as 'Old Woman,' and an adult female August 7 near the Tahkandik River, and shot several others which he failed to bag. I took a young male from a family on ' Castle Rock' August 5. We found that those taken OCT.,1900.] BIRDS OF THE YUKON REGION. 75 had been feeding on marsh hawks. Alaska jays, white-winged cross- bills, intermediate sparrows, and varied thrushes. 1 saw two duck hawks near Nulato August 24, and a tame young bird spent part of the rainy evening- of August 30 perched on the back of a chair in the hotel at St. Michael. The cere and hill of the young- male were french gray, changing to black on tip of bill and along culmen and cere above; tarsi and toes, pale, grayish green; soles, tarsi behind, and edges of scutellse in front, yellow; nails, black. 83. Falco peregrinus pealei. Peale Falcon. One flew around the Corwvn when we were some distance south of the Aleutian Islands and out of sight of land October 7. 84. Falco columbarius. Pigeon Hawk. We saw a pigeon hawk feeding on a large vole near Charlie (reek August 8. Osgood took a young male at a point 12 miles above Circle August 13. and I saw one at the Aphoon mouth August 28. 85. Falco columbarius richardsoni. Richardson Merlin. At Circle August 18 I shot a young female merlin which is inter- mediate between cohimhariw and richardsoni. In general coloring both above and below, it is between typical examples of the two forms and approaches very closely a specimen of richardsoni taken by Captain Bendire at Walla Walla, Washington, December 3, 1880, and now in the American Museum of Natural History. My bird has light spots on outer webs of primaries and six light bars on tail similar to those of richardsoni, but the bars are narrower and more interrupted. The crop and stomach contained the remains of a red-backed mouse. The irides were vandyke brown; cere, greenish-yellow; maxilla, slate black at tip. changing to greenish-white toward cere and pale french gray at commissure; mandible, pale dull greenish, changing to pale french gray toward tip and commissure; tarsi and toes, straw yellow, the latter inclining toward sulphur yellowT; nails, black. Mr. Cant- well writes in the L Osprey ' 1 of having seen Richardson's merlin, but does not state that he took specimens. These are the only records for this bird in the Yukon Valley. 86. Falco sparverius. American Sparrow Hawk. We saw this species at Log- Cabin June 14, Semenow Hills July 19 and 20, near the Tatchun River July 23, near Fort Selkirk July 26, about 30 miles below Fort Selkirk July 28, and, 1 think, at Circle August 15. We took specimens on July 19 and 28. This species has not previously been reported from along the Yukon. 1 Osprey, III, 25, Oct., 1898. 76 NORTH AMERICAN FAUNA. [no. 19. [Pandion haliaetus carolinemis. American Osprey. While the steamer was anchored near Holy Cross Mission August 25, one of the passengers, Mr. J. F. Burkman, fired at, but failed to get, a large hawk which he was positive was this species. As Mr. Nelson records it from the Lower Yukon, I see no reason to doubt Mr. Burkman's identification.] 87. Asio accipitrinus. Short-eared Owl. We saw a short-eared owl flying overhead at St. Michael on the evening of September 7, and I flushed one from some bushes on Whale Island the next day. September 9 I set three steel traps near St. Michael on poles in the tundra. One of them failed to catch anything, but before September 25 the others yielded 6 short-eared owls and the nails of another. These birds had been feeding on mice and shrews. Osgood shot a short-eared owl at Unalaska October 5. These speci- mens average slightly darker, with the white of the face purer, than fall birds from New England. 88. Scotiaptex cinerea. Great Gra}T Owl. From some low growth on a steep hillside at Miles Canyon July 11 we flushed a large gray owl that I am confident was this species. We saw a mounted specimen in Dawson August 2 and I was told at Circle that an owl answering the description of this species had been killed there recently. 89. Nyctala tengmalmi richardsoni. Richardson Owl. While lying awake under my mosquito netting in a clearing at the base of the Semenow Hills on the night of July 19 I saw a small, round-headed owl alight on the limb of a dead tree only a few feet away. It flow before I could bring my gun to bear, but I have no doubt it was this species. Osgood took a young bird near Rink Rap- ids July 22. I was told at Circle that a small owl was common there, and that one had been caught recently. 90. ? Megascops asio kennicotti. Kennicott Screech Owl. A reddish-brown owl, of the size and appearance of a screech owl, was seen by Maddren and myself at Caribou Crossing on the afternoon of June 27. We were drawn from camp by its peculiar notes, and saw it fly from a poplar across an opening to a spruce thicket. Later that day Osgood caught a glimpse of another, or perhaps the same bird, as it flew from the top of a small poplar. 91. Bubo virginianus pallescens. Western Horned Owl. Owl pellets, some of them remarkably large, containing chiefly bones of rabbits, ground squirrels, and red squirrels, were found in great abundance, especially at Caribou Crossing and on Windy Island, Lake Tagish, but the most careful hunting failed to disclose the owls. ocr.,1900.] BIRDS OF THE YUKON REGION. 77 On Fifty-Mile River, near Lake Marsh July 8, we heard the hoot- ing of a horned owl; and at our camp at Lower Lebargeone flow over, about midnight July 16, and lit in the top of a spruce just out of gun- shot. I hurried after it but merely succeeded in seeing the bird swoop into the surrounding gloom. At our camp near the Tatchun River July 22 one flew by and set- tled for an instant not far off; and the next day Osgood saw three extremely light-colored horned owls near by. We also heard the hoot- ing of this species near the Yukon at the following places: Near Little Salmon River July 21, 20 miles below Fort Selkirk July 27, 20 miles below the Selwyn River July 29, near the Tatondu River August 6, about 15 miles above Circle August 12, and opposite Circle August 14. In the last case the identification is not without doubt, but the notes of the others were unmistakable. [Surnia vfojUa caparoch. American Hawk Owl. At Bennett, June 18-22, a bird with a peculiarly weird cry flew about the cliffs above our camp every night. By a process of elimination 1 have attributed the serenade to this species.] 92. Ceryle alcyon. Belted Kingfisher. This bird occurs about the Yukon lakes, but in small numbers. Osgood saw one at Bennett June 20; I heard one at Caribou Crossing- June 29, and saw another on Lake Lebarge July 13. We found kingfishers fairhT common on Fifty-Mile River, and still more common on Thirty -Mile River. As the cliff's replaced the high banks below Fort Selkirk kingfishers became fewer, and none were seen after August -1, when we were about -10 miles below Dawson. Young able to fly were seen near Five-Finger Rapids July 22. 93. Dryobates villosus leucomelas. Northern Hairy Woodpecker. Osgood took a single hairy woodpecker on Fifty -Mile River a few miles above Miles Canyon July 10. 91. ? Dryobates villosus hyloscopus. Cabanis Woodpecker. Near the Little Salmon River July 21 I took a young female that corresponded in size and plumage with some young of this subspecies. It was seated in the entrance to a cavity in a burnt spruce. This is the first record of the occurrence of this bird in the Yukon Basin. 95. Picoides arcticus. Arctic Three-toed Woodpecker. On July 1 I was attracted by the loud cries of a young bird, and traced the sound about 100 yards through a spruce grove on the bank of Six-Mile River. The noise proceeded from a full-fledged young woodpecker of this species that had thrust its head out of the opening to its house and kept up a continual screaming. With Osgood's assist- ance, the nest was opened but only this young bird was found. The 78 NORTH AMERICAN FAUNA. [no. 19. entrance was 5 feet 8 inches from the ground, on the lower side of a living, slightly leaning spruce, and the cavity was 10 inches deep. Osgood shot what we both supposed was one of the parents, for it cer- tainly came in answer to the cries of the young; yet this bird proved to be a typical adult male of P. americd-nus alascensls. We saw no other woodpeckers there, except flickers. 96. Picoides americanus alascensis. Alaska Three-toed Woodpecker. Osgood found the remains of an Alaska three-toed woodpecker at Haines June 1, and I shot a laying female near Glacier June 10. In the Yukon Valley we secured one on Six-Mile River; three on Fifty- Mile River above Miles Canyon July 10-11, two of them young adults; two on the Lewes River between Big Salmon and Little Salmon rivers July 20-21, and two at Circle, August 19-20. The young have whiter backs than the adults. 97. Sphyrapicus ruber. Red-breasted Sapsucker. I took an adult male at Skagway May 31, and heard what 1 suppose was its mate. 98. Colaptes auratus luteus. Northern Flicker. We saw and heard flickers several times at Glacier. One, which Osgood flushed from a hole high in a dead pine June 8, had yellow quills. In the Yukon Valley this is by far the most common woodpecker. We found it quite regularly from Log Cabin to Circle, but, like most Yukon birds, it was shy. At Caribou Crossing June 27 Osgood secured a female and found her nest, containing 8 young and 3 eggs, in a cavity 3 feet from the ground in a partly dead poplar. At Six-Mile River we found a nest about 6 feet from the ground, and at Lower Lebarge July 17 I found 7 well-Hedged young in a cavity about 5 feet from the ground in a small dead tree in a burnt tract. July 25 1 took a full- grown young near Selkirk. Adult flickers from Alaska average slightly darker than luteus from Canada and farther south, the wings, tail, and bars of upperparts being somewhat blacker, and the light parts more olive and less buffy. Three young — one from near Fort Selkirk, the others nestlings from Lower Lebarge — show this difference in a marked degree, having the wings, tail, and bars of upperparts deep black, and the ground color above smoky olive, instead of buffy olive as in luteus; they are even darker than the young of auratus from Florida. But the slightness of the difference shown by the adults, the small number of specimens from Alaska, and the possibility that the plumage of the three young may have been discolored by the burnt trees where they were found- though microscopic examination shows no sign of this — make their separation as a subspecies inadvisable at present. oct.,1900.] BIRDS OF THE YUKON REGION. 79 99. Chordeiles virginianus. Nighthawk. From Caribou Crossing, where 1 shot two females June 27, until after passing the mouth of the Tatchun River July 24, we met with nighthawks on numerous occasions. I took an adult male at White Horse Rapids July 11. These birds were very fat, as might beexpected from the abundance of flying insects. They are slightly darker than virginianus from the East. 100. Selasphorus rufus. Rufous Hummingbird. We saw a rufous hummingbird on 'Eagle Island' at Bocadequadra May 28. At Glacier Osgood saw one June 6, and on June lo I found a nest with two slightly incubated eggs 3| feet from the ground on the branch of a small conifer near the falls of the river. I secured the female, and also one of two males which I saw the same day in the open country below Glacier. On Lake Bennett we saw one opposite West Arm June 24. Mr. George G. Cantwell has already added both this species and Cli<>r just after a frosty night, and among other species I saw a single fox sparrow. The bird was too close to leave identification doubtful. 135. Passerella iliaca townsendi.1 Townsend Fox Sparrow. Osgood saw one at Skagway, and we noticed several at Glacier which were exceedingly shy. Osgood collected two at Glacier June 8-9, one of which was too badly shot to preserve; the other Mr. Ridgway pronounces somewhat nearer this form than annectms. 136. Petrochelidon lunifrons. Cliff Swallow. This species was common at Log Cabin June 15 and 20. At Cari- bou Crossing we saw a few June 29, probably members of the small colony breeding on the cliffs of an island in Lake Tagish July 1. We next saw cliff swallows near the Hootalinqua River July 19, and from this point to a few miles above Dawson, August 1, we frequently met with colonies of varying size, the largest being near White River. Their nests were attached to cliffs bordering the river, except at Fort Selkirk, where they were breeding under the eaves of houses. Full- fledged young were taken July 25, and I think the absence of this spe- cies below Dawson was due to their having already migrated. I was told that both cliff and bank swallows were exceedingly abundant along the Porcupine. 137. Hirundo erythrogastra unalaschkensis.3 Alaska Swallow. A few were flying over the marshes of Chilkat Inlet June 1; 1 heard that they were common at White Pass City June 9, and we saw two about the buildings of White Pass Summit June 10. At Log Cabin they were common on June 11, 15, and 20, and on the last date I took a male. A few were noticed at Bennett June 19-21. I refer all seen to this subspecies, for all had remarkably long tails. The single specimen taken had a length of 7.96 inches, wing 4.68 inches, tail 4.10 inches, fork of tail 2.33 inches. The forehead, lower wing-coverts, and abdomen are more highly colored than eastern skins of II. < rythro- gmtra, and the shafts of the long tail feathers are whitish. 138. Tachycineta bicolor. Tree Swallow. I saw several at SkagwTay May 31 and June 3, and over the Chilkat marshes June 1. We saw others near Caribou Crossing June 29; one July 6 and a pair July 7 at Lake Marsh; and several at Miles Canyon July 11. A few miles above Fort Selkirk July 25 I saw several enter- ing and leaving an old flicker hole in a dead spruce. JAuk, XVII, 30, Jan., 1900. 2 Kept. Fur-Seal Investigations, 1896-97, pt. 3, 422, 1899. 88 NORTH AMERICAN FAUNA. [no. 19. 139. Tachycineta thalassina. Violet-green Swallow. Mr. Cantwell1 has already added this species to the list of birds known to inhabit the Yukon Valley. We saw a single male among flocks of bank swallows flying over Fifty-Mile River above Miles Canyon July 11 and another between White Horse Rapids and Lake Lebarge. On July 18 I took a male from several that we saw near Hootalinqua, and at the Semenow Hills July 20 Osgood secured a female, finding her nest with four young in a crevice in the cliffs. Maddren shot a young July 28. After this we frequently saw colonies of from six to ten birds of this species, and one near White River that must have contained over fifty. They were nesting about the cliffs as a rule, but several times we saw them enter holes in banks similar to those of Clivicola riparia, while at Fort Selkirk they were nesting in the interstices between the logs of the cabins. We often met with small colonies until within 15 miles of Circle, but after August 5 they kept so high about the cliffs that identification was possible only by their characteristic twitter. The two adult males have green rumps. 140. Clivicola riparia. Bank swallow. We found a small colony nesting at the northern end of Lake Tagish July 1, and a larger one on the west shore of Lake Marsh July 7, but we were entirely unprepared for the great abundance of this species on Fifty-Mile River above Miles Canyon. There almost every bank was honeycombed with their holes. Along the rest of the Yukon as far as Circle bank swallows were common and often abundant, but after August 1 their former presence was generally manifested only by the deserted holes. At Circle 1 saw about thirty August 17, and a single bird on the following day. Eggs advanced in incubation were found July 7, and by the 22d the young were flying, and all acting as if preparing to migrate. As it grew dusk on the evening of August 5 we watched a large Hock which circled over the Yukon, rising higher with each revolution, and at last disappeared toward some mountains due south. 141. Ampelis garrulus. Bohemian Waxwing. We saw several on Six -Mile River Jul}' 1, two at Lake Marsh July 7, one on Fifty-Mile River July 10, two pairs (one of which was secured) at Miles Cai^on July 11, one at Lower Lebarge July 16, two about a mile apart on Thirty -Mile River July 18, and four near the Selwyn River July 28. We took two adults and three young from a flock of twenty about 15 miles below the Selwyn July 29, and four from a similiar flock near Sixty-Mile Creek July 31. We saw them again at the Chandindu River August 4, Camp Davidson August 5, • ^sprey, III, 25, Oct., 1898. oct., 1900.] RIRDS OF THE YUKON REGION. 89 50 miles above Circle August LI, and 15 miles lower August L2. The female taken July 11, which Lacked the wax tips on the second- aries, contained an egg ready for the shell. The young- resemble those of A. cedrorvm, but are grayer, have less white on the abdomen, no pale streaking above, and have the wings, tail, and lower tail- coverts like adult ga/rrvlvs. They lack the cinnamon suffusion of the head of the adult, have only a few black feathers on the throat, a much shorter crest, the wax-like tips of the secondaries peach-blossom pink instead of scarlet, and the lower tail-coverts paler. A still younger bird than the two described is obscurely streaked with whitish both on back and lower parts. On one of the young' the wax tips are very small. In habits and notes the Bohemian waxwing closely resembles the common cedar wax wing. Two males that we noticed while descending Thirty-Mile River were perched on the topmost sprays of tall spruces, uttering a lisping whistle at frequent intervals. One of them Hew after a passing insect in the manner of a flycatcher. Flocks were easily approached, and when one bird was shot the rest would scatter, and each would alight on the top branch of some spruce and utter a characteristic call note. This note, which we often heard from pass- ing flocks, wTas similar to the whistle just mentioned. The birds that we collected had been feeding on the purple berries of some uniden- tified plant. 142. Helminthophila celata. Orange-crowned Warbler. Osgood took an adult male at Caribou Crossing June 20; I a female and two young 20 miles below Fort Selkirk July 27, and a young near Dawson August 2. Osgood secured an adult and one young at Camp Davidson August 5 and 6, and I saw one young 15 miles above Fort Yukon August 21. All taken were in alders or willows close to the water. 113. Helminthophila celata lutescens. Lutescent Warbler. Common at Haines, where we took live June 1 and 2. 114. Helminthophila peregrina. Tennessee Warbler. Found only at Caribou Crossing, where I heard four males singing and secured three of them June 25 and 27. They were in compara- tively open swamps of willows and low spruces. 145. Dendroica sestiva rubiginosa.1 Alaska Yellow Warbler. I am positive I often heard the song of this ■species at Bennett June 17-22. I took an adult male at Caribou Crossing June 27, and think I heard the song about Lake Marsh. An adult female was taken l>y xAuk, XIV, 76, 123, 1897. 90 NORTH AMERICAN FAUNA. [no. 19. Osgood near the Nordenskiold River July 22, and family parties were often found in the alders and willow thickets between the Felly River and Circle. I took a young from a small flock 15 miles above Fort Yukon August 21, saw one at the Aphoon mouth August 28, and a few I thought this species at Hendricks Station August 25. Birds from the Yukon Valley do not differ from those of the Alaska coast. A young female is duller above and more buffy below than the young female of D. cestiva. 116. Dendroica coronata hooveri.1 Hoover Warbler. We found Hoover warblers common at Skagway, Glacier, Log Cabin, and Caribou Crossing, and also noted them at Haines, Bennett, Lake Tagish, Miles Canyon, White River, Sixty-Mile Creek, and 12 miles above Circle. At Skagway May 31 they were still in flocks, but at Glacier June 1-10 they seemed to be mated and settled for the summer. At Log Cabin we found a flock June 15, but five days later those still remaining there were beginning to nest. A small flock seen on an island near Sixty-Mile Creek August 1 showed that the return migration had begun. I took a young in striped plumage August 1. Adult males average paler below than typical I). coronata, the black markings being narrower, thus giving an effect of broad longitudinal markings rather than black clouding on the chest. Eight specimens of both sexes average slightly larger in length of wing and tail than the corresponding sexes from eastern and central United States. In six males, the exposed culmen averages 0.02 inch longer than in males from Connecticut, but the bill from nostril averages the same, as do both measurements in females. In juvenile plumage hoover/' is darker than coronata^ the black markings are broader and blacker, both above and below, and the brownish edgings to the feathers greatly restricted— entirely wanting on the lower parts and middle back. 117. Dendroica striata. Black-poll Warbler. At Log Cabin June 15 this species was common, but on my return June 20 I saw only one pair — which I secured — and one other male. July 5 I took a male at Lake Marsh. Two birds taken at Caribou Crossing are somewhat smaller than average specimens from Dakota and Connecticut. 118. Dendroica townsendi. Townsend Warbler. Osgood took a male at Skagway May 31. At Glacier it was tolera- bly common in the dense woods of spruce and fir, and unquestionably nesting; altogether we noticed about twenty individuals during our stay. Osgood took an adult at the southern end of Lake Marsh July 1, and I an adult female and young female on the west shore of Lake 1 Bull. Cooper Omith. Club, I, 32, 1899. oct.,1900.] BIRDS OF THE YUKON REGION. 91 Lebarge July 14. The juvenile plumage differs from that of D. virens only in being slightly less brown on crown and back. This is a new species for the Yukon Valley. 149. Seiurus noveboracensis notabilis. Grinnell Water Thrush. The first sound that I heard on the morning of August 1. when we were on a small island about 1<» miles below Sixty-Mile Creek, was the unmistakable alarm note of the water thrush. This was the first time we had met with this species, and before starting that morning- on our daily Yukon drift, Osgood and I each secured a young bird. Near Forty-Mile Creek. Tatondu River, and Charlie Creek water thrushes were again met with. At Circle I saw several August 16-20, took one 15 miles above Fort Yukon August 21, and saw two in a thicket at the Aphoon mouth August 28. The young in fall plumage taken on the Yukon are clove-brown above, including wings and tail — far darker than is usual in notabilis — and have darker streaks below. 150. Wilsonia pusilla. Wilson Warbler. Osgood took an adult female near the Chandindu River August -1, and I a young female near Charlie Village August 11 and a young- male 25 miles above Circle August 12. 1 also saw one 3Q miles below Circle August 20. These birds, while not typical pusilla, are, like those of the Lower Yukon, nearer it than pileolata. 151. Wilsonia pusilla pileolata. Pileolated Warbler. We found this the most abundant bird at Glacier June 5-10, fre- quenting the alder thickets from the valley as far as they extended up the hills. 1 saw a yellow warbler I thought this species on White Pass Summit June 12. Pileolated warblers were common at Log Cabin, Bennett, and Caribou Crossing, and I am confident I heard them singing at Lake Marsh. Adult males from Glacier resemble normal pileolata closely, but have the back rather more green; those from the Yukon Valley, while having the orange forehead and lower parts of this form, have the duller green back of pusilla. [Mbtaeilla ocularis. Swinhoe Wagtail. On the morning of August 28 the Robert Kerr, on which I was a passenger, wajs hindered from proceeding by a gale and low water on the bar, and was made fast to the bank at the Aphoon mouth of the Yukon. As 1 came on deck I saw half a dozen wrhite wagtails fly about the vessel and settle in the grass close by. While I returned for my gun they left, but a thor- ough acquaintance with MotaciUa alba in Egypt, where it is abundant during the winter, leaves me no doubt that these birds were wagtails.] 152. Anthus pensilvanicus. American Pipit. A male taken at Skagway June 3 was probably a belated migrant. On the heights above Glacier Osgood saw several June 5, and we 92 NOETH AMEEICAN FAUNA. [no. 19. found them common at Summit June 11-13. A female taken June 13 was laying, and a fresh but empty nest I found the same day I attrib- uted to this species, no other being near. This nest was loosely formed of fine dry grass in a hollow in the deep moss which covered the almost perpendicular side of a bowlder lying on a hill high above Summit, only a small hole for entrance showing in the moss. We often saw the song-flight at Summit. Launching himself with a sharp preliminary ' chip ' from one of the granite bowlders that abound there, the male would rise rapidly to a height of a hundred feet or more, uttering a sweet, clear song. After poising high in air and repeating this song for several minutes the singer would slowly float toward earth and alight 100 yards from where he started, soon to repeat the same performance. We found a pair on the heights above Ben- nett June 17, and a few, possibly members of one family, at Circle August 15-20. 153. Cinclus mexicanus. American Dipper. We collected a female and set of four fresh eggs at the falls at Glacier June 8. A single ouzel seen farther down the river June 10 was probably the mate of the ore taken. Osgood also took one at Unalaska October 5. 151. Anorthura hiemalis pacifica. Western Winter Wren. We noticed a few at Glacier June 4-10, and I took a male there June 6. 155. Anorthura alascensis. Alaska Wren. I saw one at St. George October 3, and we collected five at Unalaska October 5. The young were then molting. 156. Sitta canadensis. Red-breasted Nuthatch. I took a male at Skagway May 31 and another near Log Cabin June 20, and heard one on an island at the junction of the Lewes and Pelly rivers July 26. This species has not heretofore been noted in the Yukon Valley. 157. Parus atricapillus septentrionalis. Long-tailed Chickadee. We took this species at Bennett June 19, west shore of Lake Ben- nett June 24, Caribou Crossing June 26, Lake Marsh July 7, and Lake Lebarge July 15, but did not notice it again until we reached the Lower Yukon, although chickadees were heard several times whose specific identity was not determined. Thirty miles below Holy Cross Mission I took two August 25, and at the Aphoon mouth I saw a small flock August 28. Young able to fly were taken July 7. One taken August 25 had completed the molt into first winter plumage, while an adult taken the same day was in fresh plumage. oct.,1900.] BIRDS OF THE YUKON REGION. 93 158. Parus hudsonicus evura.' Yukon Chickadee1. We took the Yukon chickadee at Caribou Crossing June 27, Lake Tagish Juno 30, Lake Marsh July 5, and Lake Lebarge July 14, and after reaching Thirty-Mile River July 19, found it regularly distrib- uted in families or large flocks all the way to Fort Yukon, 15 miles above which I saw a flock August 21. At St. Michael 1 took a young- female in first winter plumage September 20. Young able to fly were first taken July 5 and molting birds August 13. We took adults in full molt June 27, and one in which the molt was almost completed July 21. 159. Parus rufescens. Chestnut-backed Chickadee. We found a few at Haines and Skagway, and I took one and heard another at Glacier June 5. A female taken at Skagway June 3 had finished laying. 160. Regulus satrapa olivaceus. Western Golden-crowned Kinglet. Tolerably common at Glacier; often heard but seldom seen, and difficult to procure. A female that I took June 10 had the last egg ready for the shell. 161. Regulus calendula. Ruby-crowned Kinglet. I took a male at Log Cabin, and on June 20, between that point and Bennett, heard another singing. Osgood took two young speci- mens, one at Lower Lebarge July 17, and the other 20 miles below Dawson August 1. 162. Regulus calendula grinnelli.2 Sitka Kinglet. At Skagway I heard a Sitka kinglet singing May 31, and at Haines took a male and heard another singing June 1. At Glacier 1 took a male June 6, and during our stay heard two or three others singing. While the Log Cabin bird is normal calendula, the Haines and Glacier birds have the more olive back and darker sides of crown of grinnelli. 163. Myadestes townsendi. Townsend Solitaire. On the heights above Bennett 1 took an adult male June 17. On the hot noon of June 26, while seated on the summit of a hill some 1,500 feet above Caribou Crossing, I heard the most beautiful bird song that has ever delighted my ear. It seemed to combine the strength of the robin, the joyousness and soaring quality of the bobo- link, and the sweetness and purity of the wood thrush. Starting low and apparently far awa}r, it gained in intensity and volume until it filled the air, and I looked for the singer just above my head. I finally traced the song to a Townsend solitaire that was seated on 1 Auk, XVII, 118, April, 1900. 2 Auk, XIV, 399, 1897. 94 NORTH AMERICAN FAUNA. [no. 19. a dead tree about 150 yards away, pouring' forth this volume of melody without leaving its perch. The singer came close enough later to make identification certain. Osgood and Maddren saw one at Lake Lebarge July 14. Osgood took an adult at Miles Canyon July 11, another at the Semenow Hills July 20, a young in the spotted plumage 20 miles below the Selwyn River July 29, and another young 30 miles above the White River July 30. I saw an adult near the Selwyn River July 29, and took a molting adult near Sixty-Mile Creek August 1. Mr. Cantwell found this species in the Yukon Valley. 161. Hylocichla alicise. Gray-cheeked Thrush. Several thrushes which we heard singing on the west shore of Lake Marsh July 8 were, I think, this species, as their song differed from that of the dwarf, hermit, and Alma thrushes. 1 saw two, but they were so shy that 1 could not secure either. Near Sixty-Mile Creek, July 31, 1 took a young in spotted plumage, which was with the young of alrnoe which Osgood shot. At Circle 1 took a young in first winter plumage, also with almce. 165. Hylocichla ustulata almge.1 Alma Thrush. This is the common thrush of the Yukon basin, occurring every- where from Log- Cabin to Circle, perhaps in largest numbers at Caribou Crossing and Lake Marsh. Fifteen miles above Fort Yukon 1 took one, and saw others August 21. We found many nests, usually 6 to 10 feet from the ground in thick growths of young- spruces, but none contained eggs. A nest containing four young just hatched, which 1 found at Caribou Crossing June 25, was about 8 feet from the ground in a thicket of small spruces. The nest resem- bled that of II. a. swainsoni. At Miles Canyon July 11 we saw young able to fly. Osgood took young in spotted plumage July 31, but those taken August 20 had assumed first winter plumage. They were usually silent by day, but sang frequently during the short nights. At Caribou Crossing, the last of June, their song could be heard constantly from 8 p. m. to 8 a. m., one taking up the strain as another stopped. The song is much superior to that of Hylocichla aonalaxchkm and almost equal to that of II. fmcescens. It has whispered notes like that of II. mustelinus. By the middle of July the song season was practically over, though we heard one of the birds singing July 23. When the nights became really dark in August, I often heard the call-note of this bird near our camp between 2 and 3 a. m. 1 Auk, XVII, 119, April, 1900. ocr.,1900.] BIRDS OF THE YUKON REGION. 95 166. Hylocichla aonalaschkse. Dwarf Hermit Thrush. We hoard several singing at Skagway, and Osgood took one at Haines June 2. At Glacier they were tolerably common, and we secured several, but they were very shy, keeping in the thickets dur- ing the day and singing for several hours in the evening from the topmost spray of some spruce well up the mountainside. Several thrushes' nests in small spruces 6 to 8 feet from the ground were empty, for which condition the abundant red squirrels were probably responsible. At Log- Cabin and Bennett we heard a few singing, and at Caribou Crossing Osgood took one June 27. 167. Hylocichla aonalaschkae pallasi. Hermit Thrush. About 15 miles below Little Salmon River July 22 we secured a pair, whose nest, containing four well-grown young. Osgood had found the evening before. Far from selecting the secluded nesting site usual with this species, this pair had placed their nest between two small bunches of flowers on an open southern hillside, just above a small piece of burnt poplar woodland, and exposed to the full glare of the sun. 168. Merula migratoria. American Robin. Tolerably common at Haines and Skagway, but not found at Gla- cier. At Haines I took a female and four well-incubated eggs June 2. Robins were common at Log Cabin June 15, and were found regu- larly, but in gradually decreasing numbers, until August 1, when the last was noted near Sixty-Mile Creek. A flock seen Jul}7 29 showed that the southern migration had commenced. We found an empty nest 30 miles below Dawson and heard that the birds bred near Fort Yukon. Although robins were b}- no means common at Caribou Crossing, I found, on June 25, 13 empty nests, most of them evidently built that year, and 4 empty nests of the Alma thrush, in a small patch of spruces. The red squirrels which lived in a hollow tree near by probably knew the location of most of these nests. Osgood took a well-grown young robin here on June 26. 169. Hesperocichla naevia. Varied Thrush. At Haines I saw several June 1, and Osgood took one June 2. At Glacier varied thrushes were rather common, but exceedingly shy. About an hour before sunset they would fly to the top of some tall tree and repeatedly utter a long-drawn, plaintive whistle until dark- ness fell. Sometimes on cloudy days we would hear their song, but it was infrequent and had about stopped when we left Glacier, June 11. We next saw this species near the Tatchun River, where I took a young bird July 23. Thirty miles below Dawson we took young, 96 NORTH AMERICAN FAUNA. [i and met with the birds several times until August 21, when large flocks were seen near Fort Yukon. At Glacier I found on June 7 a nest containing four eggs, varying from fresh to several days incubated. It was very large, built of sticks and moss and lined with dry grass, and was situated 15 feet from the ground, near the top of a small spruce growing in dense woods a short distance from the river. When I put my hand on the tree, the female flew from the nest with a hoarse, cackling cry and settled a few feet away; the male did not appear. The eggs average 1.25 by 0.81 inches and are nile blue sparingly spotted with ecru drab and seal brown. 170. Saxicola cenanthe. Wheatear. Osgood saw two young wheatears at Circle August 19, and secured one. At the Aphoon mouth I shot one on August 27, which fell into the river and was carried away by the rapid current, but I saw the white rump plainly. 171. Sialia arctica. Mountain Bluebird. We found a pair on Fifty-Mile River a short distance above Miles Canyon July 10. The next day I secured the female and found the nest with four well-incubated eggs in a hole about 8 feet from the ground in a dead spruce in the midst of a burnt tract. July 22 I shot a male near the point where Fifty-Mile River empties into Lake Lebarge. Mr. Cantwell also found this species on Fifty-Mile River. INDEX [Names Of new species iii lilack-faci' type.] Abies lasiocarpa, 9, 16. Acanthis exilipes,82. Hnaria.vJ. Accipitei atricapillus, 73. velox.72. Actitis macularia, 69. /Echmophorus occidentalis, 57. .Egialitis semipalmata, 70. Alas1' i tundra district, 15-16. Albatross, short-tailed, 00-61. Alces gigas. 23. Alnus sinuata, 9, 16. Amelanehier alnifolia. 10. Ammodiamus alaudinus, 83-8 1. sandwichensis, S3. Ampelis garrulus, 88-89. Anas boschas. 61-62. Andromeda polifolia, 15. Anorthura alascensis, 92. pacifica, 92. Anser gambeli, 64. Anthus pensilvanicus, 91-92. Archibuteo lagopus. 73. Arctomys caligatus, 31-32. Arctostaphylos uvauxsi, 9, 10. Arenaria melanoeephala, 71. Artemisia arctica, 15-16. frigida, 11, 12. Asio accipitrinus, 76. Anklet, least, 58. Aythya affinis,62. marila, 62. Baldpate,62. Bat. little brown, 45. Bear, Alaska grizzly, 41. black, 41. Beaver. American, 32. Betula glandulosa, 9, 10. nana, 16. papyrif era, 11. Birds, classified lists of, 51-57. Blackbird, rusty, 81. Bluebird, mountain, 96. Bonasa umbelloides, 71. Brachyramphus marmoratus, 58. Branta hutchinsi, 64. Bubo pallescens, 70-77. Buffalo berry, 10. Buffle-head, 63. Buteo calurus. 73. Calcarius alascensis, S3. Calidris arenaria, 67. Canachites osgoodi, 19, 71. Canadian Yukon district, 10-13. Canis occidentalis, 40. Caribou, barren ground, 23. mountain, 22. Cassiopea tetragona, 15. Castor canadensis, 32. 4494— No. 19 — Cepphus columba, 58. Ceryle alcyon,77. Chamaecistus procumbens, 16. Chameeneriou angnstifolium, 13. Charadrius fulvus, 70. Charitonetta albeola, 68. Chen hyperborea, 64. Chickadee, chestnut-backed, 93. Icmg-tailed, 92. Yukon, 93. Chipmunk, gray-headed, 28-29. Chordeiles virginianus, 79. Ciuclua mexicanus, 92. Circus hudsonius, 72. Clangula americana, 62. Clivicola riparia, 88. Coccothraustes montanus, 81-82. Colaptes luteus, 78. Colymbus holboelli,57. Contopus borealis, 79. saturatus, 19, 79-80. Cormorant, pelagic, 61. red-faced, 61. violet-green, 61. I '< irvus americanus, 81. caurinus, 81. principalis, 81. Crane, little brown, 65. Crossbill, red, 82, white- winged, 82. Crow, American, 81. northwest, 81. Crymophilus fulicarius, 65. Curlew, Hudsonian, 69. Cyanocitta stelleri, 80. ■ Dafila acuta, 62. Dendiagapus fuliginosus, 71. Dendroica hooveri, 90. rubiginosa, 89. striata, 90. townsendi, 90-91. Dicrostonyx alascensis, 20, 38. nelsoni,20, 38. Diomedea albatrus, 60-61. Dipper, American, 92. Dryobates hyloscopus, 77 leucomelas, 77. Duck, harlequin, 63. lesser scaup, 62. scaup, 62. Eagle, gray sea, 73. northern bald, 74. Eider, Pacific, 63. Empetrum nigrum, 9, 10. Empidonax alnorum,80. hammondi, 80. Equisetum, 13, 14. Erethizon myops, 20, 38. Eutaiuias caniceps, 19,28-29. 98 INDEX. Evotomys alascensis, 35. dawsoni, 34. Falco anatum, 74. columbarius, 75. pealei, 75. richardsoni, 75. rusticolus,74. sparverius, 75. Falcon, Peale, 75. Fauna! districts, 8-18. Fiber spatulatus, 19, 36-37. Fisher, 44. Flicker, northern, 78. Flycatcher, alder, 80. Hammond, 80. olive-sided, 79. Fox, Hall Island, 41. red, 41. Fratercula corniculata, 57. Fulmar, Pacific, 61. Fulmarus glupischa, 61. Gallinago delicata, 65. Gavia arctica, 57. imber, 57. Goat, mountain, 25. Golden-eye, American, 62. Goose, American white-fronted, 04. emperor. 64. Hutchins, 64. lesser snow, 64. Goshawk, American, 73. Grebe, Holbcell, 57. western, 57. Grosbeak, Alaska pine, 82. western evening, 81-82 Grouse, Alaska, 71. gray ruffed, 71. sooty, 71. Grus canadensis, 65. Guillemot, pigeon, 58. Gull. American herring, 59. Bonaparte, 60. glaucous-winged, 59. Point Barrow, 59. Sabine, 60. short-billed, 59-60. Gulo luscus, 44. Gyrfaleon, gray, 74. Hare, Alaska arctic, 40. Dall varying, 40. Harelda hyemalis, 63. Haliseetus alascanus, 74. albicilla, 73. Hawk, duck, 74. American sparrow, 75 marsh, 72. pigeon, 75. rough-legged, 73. sharp-shinned, 72. western red-tailed, 73. Helminthophila celata, 89. lutescens, 89. peregrina, 89. Helodromas cinnamomeus, 68. Hesperocichla narvia, 95-96. Hesperomys arcticus, 33. Heteractitis ineanus, 69. Hirundo unalaschkensis, 87. Histrionicus histrionicus, 63. Hudsonian Yukon district, 13-15. Hummingbird, rufous, 79. Hylocichla alicire, 94. alma-, 94. aonalaschkse, 95. pallasi, 95. Jfeger, long-tailed, 58. parasitic, 58. Jay, Alaska, 80-81. Steller, 80. Junco, Oregon, 86. Shufeldt, 86. slate-colored, 85-86. Junco connectens, 86. hyemalis, 85-86. oregonus, 86. Juniperus nana, 9, 10, 13. Kalniia glauca, 10. Kingfisher, belted, 77. Kinglet, ruby-crowned, 93. Sitka, 93. western golden-crowned. 93. Kittiwake, Pacific, 5S. red-legged, 58. Lagopus lagopus, 71. leucurus, 72. nelsoni, 72. rupestris, 72. Larix americana, 14. Larus barrovianus, 59. brachyrhynchus, 59-60. glaucescens, 59. Philadelphia, 60. smithsonianus, 59. Ledum grcenlandicum, 10. palustre, 15. Lemming, Alaska, 38. Nelson pied, 38. Yukon, 37-38. Lemmus alascensis, 20, 38. yukonensis, 20, 37-38. Lepargyrsea canadensis, 10. Lepus dalli, 20. 40. othus, 20, 40. saliens, 19, 39-40. Leucosticte, Aleutian, 82. Hepburn, 82. Leucosticte griseonucha, 82. littoralis, 82. Longspur, Alaska, 83. Loon, 57. black-throated, 57. Loxia leucoptera, 82. minor, 82. Lunda cirrhata, 57. Lutra canadensis, 41. Lutreola ingens, 19, 42. Lynn canal district, 8-9. Lynx, arctic, 40. Lynx mollipilosus,20,40. Magpie, American, 80. Mairania alpina, 15. Mallard. 01-62. Mammals, list, 22. Mareca americana, 62. Marmot, hoary, 31-32. Marten, Alaska, 43-44. INDEX. 99 Megascops kennicotti, 76. Melospiza cinerea, 86. lincolni. 86. rufina, 86. striata, 86. Merganser, American, 61. Merganser americanus, 61. Merlin. Richardson, 75. Merula migratoria, 95. Mierotus drummondi, 35-36. mordax, IT. 35. operarius, 36. xanthognathus, 36. Mink. Alaska. 42. Moose. Alaska, 23. Motacilla ocularis, 91. Mouse, Alaska jumping, 38. Arctic white-footed, 33. Bangs white-footed, 32-33. Dall lemming. 37. Dawson red-backed, 34. tundra red-backed, 35. Murre, 1 'alias. 58. Murrelet, marbled, 58. Mus decumanus, 32. Muskrat, northwest, 30.-37. Hnstela actuosa, 19,43-44. pennanti. 44. Myadestea townsendi, 93-94. Myosotifi alpestris. 10. Myotis lucifugus, 45. Neotoma saxamans, 19,33-34. Nettion carolinensis, 02. New speeies, 19. Nighthawk, 79. Nurnenius hudsonicus, 09. Nuthatch, red-breasted, 92. Nyctala richardsoni, 70. Oceanodronia furcata, 01. Ochotona collaris, 38-39. Oidemia amerieana, 0:'.. deglandi, 03. perspicillata, 03-04. 01d-scpuaw,63. Olor Columbian us, 65. Oreamnos montanus, 25. Osprey, American, 70. Otter. American. 41. Oris dalli, 24-25. stonei,24. Owl, great gray, to. American hawk, 77. Kennicott screech, 76. Richardson, 76. short-eared, 76. western horned, 70-77. Pandion carolinensis, 76. Parus evura. 93. rufescens, 93. septentrionalis, 92. Passerella iliaca,87. townsendi, 87. Passerina nivalis, 83. Perisoreus fumifrons, 80-81. Peromyscus arcticus, 33. oreas, 32-33. Petrel, forked-tailed, 61. Petroehelidon lunifrons, 87. Pewee, Alaska wood, 79-80. Phalacrocorax pelagicus, 61. robust us, 61. urile, o.i. Phalarope, northern 65. red, 05. Phalaropua lobatus, 65. Philacte canagica, 64. Phoebe, Yukon, 79. Pica hudsonica, 80. Pieea canadensis, 10. sitchcnsis, 9. Picoides alascensis, 78. arcticus, 77-78. Pika, Alaska, 38-39. Pinicola alascensis, 82. Pintail, 02. Pinus murrayana. 9, 13, 10, 17. sitchensis, 9. Pipit, American, 91-92. Plover, black-bellied, 70. Pacific golden, 70. semipalmated, 70. Polemonitun humile, 10. Populus balsamifera, 9. tremuloides, 8, 9. Porcupine, Alaska, 38. Ptarmigan, Nelson, 72. n irk, 72. white-tailed, 72. willow, 71. Puffin, horned, 57. tufted, 57. Putorius alascensis, 43. arcticus, 42-43. eskimo, 20, 43. richardsoni, 42. Rangifer arcticus, 23 montanus, 22. tarandus, 23. Rat, northern bushy-tailed, 33-34. Norway, 32. Haven, northern, 81. Redpoll, 82. hoary, 82. Regulus calendula, 93. grinnelli, 93. olivaceus, 93. Reindeer, domesticated, 23. Ribes laxifloruni, 9. Rissa brevirostris. 58. pollicaris, 58. Robin, American, 95. Rosa cinnamomea ,11. Rubus arcticus. 10. chamsemorus, 16. Sanderling, 67. Sandpiper. Aleutian, 05. Baird, 07. least, 07. pectoral, 60. Pribilof, 00. red-backed, 67. sharp-tailed, 66. spotted, 69. western solitary. 6S. Sapsucker, red-breasted. 78. Sa xicola cenanthe, 90. 100 INDEX. Sayornis yukonensis, 19, 79. Sciuropterus yukonensis, 19,25-2(3. Sciurus hudsonicus, 26-27. petnlans, 19, 27-28. vancouverensis, 27. Scolecophagus carolinus, 81. Scoter, American, 63. surf, 63-64. white-winged, 63. Scotiaptex cinerea, 76. Seiurus notabilis, 91. Selasphorus rufus, 79. Sheep, Dall mountain, 24-25. Shrew, arctic, 44. mountain, 45. Streator, 44. tundra, 45. Sialia arctica, 96. Simorhynchus pusillus, 58. Siskin, pine, 83. Sitta canadensis, 92. Snipe, Wilson, 65. Snowflake, 83. Solitaire, Townsend, 93-94. Somateria v-nigra, 63. Sorex arcticus, 20, 44. obscurus, 45. streatori,44. tundrensis, 20, 45. Sparrow, Aleutian song, S6. Forbush,86. Fox, 87. golden-crowned, 84. intermediate, 84. Lincoln, 86. sandwich, 83. sooty song, 86. Townsend fox, 87. western chipping, 85. western savanna, 83-84. western tree, 84-85. Spermophilus osgoodi, 20, 31. plesius, 19,29-31. Sphyrapicus ruber, 78. Spinus pinus,83. Spizella arizonse, 85. ochracea, 84-85. Squatarola squatarola, 70. Squirrel, Bennett ground, 29-31. Fort Yukon ground, 31. Hudson Bay red, 26-27. Yukon flying, 25-26. Stercorarius longicaudus, 58. parasiticus, 58. Sterna paradiseea, 60. Surnia caparoch,77. Swallow, Alaska, 87. bank, 88. cliff, 87. tree, 87. violet-green, 88. Swan, whistling, 65. Symphemia inornata, 69. Synaptomys dalli, 37. Tachycineta bicolor, 87. thalassina, 88. Tattler, wandering, 69. Teal, green-winged, 62. Tern, arctic, 60. Thrush, Alma, 94. dwarf hermit, 95. gray-cheeked, 94. hermit, 95. varied, 95-96. Thuja plicata, 16. Totanus flavipes, 68. Tringa acuminata, 66. bairdi, 67. couesi, 65. maculata, 66. minutilla, 67. paciflca, 67. ptilocnemis, 66. Tsuga mertensiana, 9. Turnstone, black, 71. Tussilago frigida, 16. Uria arra, 58. Ursus alascensis,41. americanus, 41. Vaccinium ovalifolium,9. vitisidsea, 15. Viburnum pauciflorum, 10. Vole, Drummond, 35-36. long-tailed, 35. Nelson, 36. yellow-cheeked, 36. Vulpes fulvus,41. hallensis,20,41. Wagtail, Swinhoe, 91. Warbler, Alaska yellow, 89-90. black-poll, 90. Hoover, 90. lutescent, 89. orange-crowned , 89. pileolated, 91. Tennessee, 89. Townsend, 90-91. Wilson, 91. Water thrush, Grinnell, 91. Wax wing, Bohemian, 88-89. Weasel, Alaska least, 43. Juneau, 43. tundra, 42-43. Wheatear, 96. White Pass district, 9-10. Willet, western, 69. Wilsonia pileolata, 91. pusilla, 91. Wolf, 40. Wolverine, 44. Woodpecker, Alaska three-toed, 78. arctic three-toed, 77-78. Cabanis, 77. northern hairy, 77. Wren, Alaska, 92. western winter, 92. Xema sabinii,60. Yellow-legs, 68. Y'ukon flats, 14. Zapus alascensis, 38. Zones, 16. Zonotrichia coronata,,si. gambeli, 84. o U. S. DEPARTMENT OF AGRICULTURE DIVISION OF BIOLOGICAL SURVEY NORTH AMERICAN FAUNA ]sr<). do [Actual date of publication, August 31, 1901] REVISION OF THE SKUNKS OF THE GENUS CHINCHA ARTHUR H. HOWELL ASSISTANT, BIOLOGICAL SURVEY Prepared under the direction of Dr. C. HART MERRIAM CHIEF OF DIVISION OF BIOLOGICAL SURVEY WASHINGTON GOVERNMENT PRINTING OFFICE 1901 ir LETTER OF TRANSMITTAL. U. S. Department of Agriculture, Division of Biological Survey, Washington, D. C, July 5, 1901. Sir: I have the honor to transmit herewith for publication as No. '20 of North American Fauna a report entitled "Revision of the Skunks of the Genus Chincha" by Arthur H. Howell, assistant in the Biological Survey. Respectfully, T. S. Palmer, Acting Chief, Biological Survey. Hon. James Wilson, Seen tary of . Vgrieultun . CONTENTS. Page. Introduction 9 History 9 Distribution 10 Habits 11 External characters 12 Material 13 Nomenclature 13 < Jeneric names 14 Specific and subspecific nanus 15 Genua ( 'hincha 20 Generic characters 20 Key to species and subspecies 20 List of species and subspecies, with type localities 21 Subgenus Ch India 22 Subgeneric characters 22 I descriptions of species and subspecies 22 Subgenus Leucomiira 39 Subgeneric- characters 39 Descriptions of species and subspecies 41 Tal >le of cranial measurements 44 5 ILLUSTRATIONS. Page. Plate I. Skins of Chincha putida and C. elongata 46 II. Skins of Chincha mephitis, C. mesomekis, and C. estor 48 III. Skins < if Chincha holzneri, C. notata, and C. spissigrada 50 IV. Skins of < nincha macroura and ( '. milleri 52 V Skulls of Chincha mephitis and C. putida 54 VI. Skulls of Chincha hudsonica and C. mesomelas 56 VII. Skulls i if ( 'hincha occidentalis and C. estor 58 VIII. Skulls of Chincha milleri and C. vittala 60 7 No. 20. NORTH AMERICAN FAUNA. August, 1901. REVISION OF THE SKUNKS OF THE GENUS CHINCHA.' By Aktiu i; II. Howell. Assistant, Biological Survey. INTRODUCTION. HISTORY. Skunks have figured in literature since the early part of the seven- teenth century. Their peculiar means of defense has served to make them conspicuous, but the repugnance in which they are commonly held has doubtless prevented as thorough study of their habits and characteristics as has been accorded to other common mammals. Apparently tin1 hist account of them is that given by Gabriel Sagard- Theodat in his history of Canada, published in 1636, in which he refers to the ill-smelling qualities of these ' enfcms chi (UableS* In 1651 Hernandez gave an account of the Mexican skunks, which he confused somewhat with the nasua.3 It was on this description that Lin- nseus primarily based his Vwerra memphitis. Buffon, in 1765, gave a description and figure of a skunk, which he called 'leckincke'* — a name taken from Feuillee, who in 1711, in an account of lis travels in South America, had recorded this as the native name of the skunk occurring near Buenos Ayres. 5 Buffon's account has served as the basis for many of the later names applied to the skunks. He ascribes l"lechiiiche'> to South America, but his figure evidently represents one of the skunks of the genus Chincha, which is strictly North Amer- ican. The confusion of the skunks of the two continents, for which this error is partly responsible, has continued, though in a lessening degree, to the present day. 1 Chincha of this paper is the equivalent of Mephitis of recent authors. For a dis- cussion of the reasons for this change, see p. 14. 2Histoire du Canada, p. 748, Paris, 1636. 'Thesaurus Rer. Med. Novae His].., p. 332, Rome, 1651. 4Histoire Naturelle, XIII, p. 294, pi. 39, Paris, 1765. 5Journal du P. Feuillee, p. 21'1\ Paris, 1714. 10 NORTH AMERICAN FAUNA. [no. 20. Schreber, in 1776, gave a good description of the large North American skunk, and his specific name mephitis is the first one avail- able for any member of the genus. Cuvier, in 1800, applied the same name in a generic sense to the ' mouffettesj his group including both the large and small skunks of North America, since separated into different genera.1 During the early years of the nineteenth century the large skunks, which form the basis of this paper, were treated by many authors, and some new names were proposed; but for the most part the descriptions are hopelessly confused and the names are unidentifiable. In 1829 Richardson described hudsonica from central Canada; in 1832 Lichtenstein named mesomelas from Louisiana and macroura and mttata from Mexico; and in 1837 Gray described varians from Texas and mexicana (the same as macroura) from Mexico. Up to 1838 the North American skunks had all been included in a single undivided genus, Mephitis, but in this year Lichtenstein pub- lished an extended revision of the group, in which he classed the white-backed skunks as a subgenus under the name Thiosmus, but did not separate the little spotted skunks from their larger relatives.2 In 1842 Lesson published a brief but important synopsis of the genus, in which he arranged the species in three groups, restricting Mephitis in a subgeneric sense to the little spotted skunks, naming the large striped skunks Chincha and retaining Thiosrrms for the white-backed skunks.3 Lesson's Mephitis and Chincha, proposed as subgenera, are perfectly tenable as genera, and are here used in that sense, Mephitis for the little spotted skunks, commonly known as Spilogale, and Chincha for the large striped skunks, commonly known as Mephitis. Boitard, in 1812, proposed three new names for the large skunks, all of which probably refer to the same species, though only one, putida, is certainly identifiable. This is available for the eastern skunk. In 1857 Baird named the California skunk occidentalis. For several decades after this the group received very little attention from systematists, but during the last few years there has been considerable activity in the naming of species. In 1890 estor was proposed by Merriam; in 1895, elongata by Bangs; in 1896, scrutator by Bangs, and fossidens and orthostichus (fossils) by Cope; in 1897, holzneri and miller i by Mearns; in 1898, avia and spissigrada by Bangs, and in 1899, Jwhdenta by Elliot, and leptops and obtusaPm (fossils) by Cope. DISTRIBUTION. The skunks of the genus Chincha range over the' greater part of North America, from the Hudsonian zone in Canada to Guatemala. The northern limits of their distribution are not definitely known. 'Legons d'Anat. Comp., I, tabl. 1, 1800. 2Abh. Akad. Wiss. Berlin for 1836, pp. 249-312, 1838. sNouv. Tabl. de Regne Anim., p. 67, 1842. aug., 1901.] HABITS. 11 In tht' interior they occur as far north as Great Slave Lake, as shown by a specimen from that locality in the V. S. National Museum, and by the statement of B. R. Ross, who says that while he has never seen a living specimen on the Mackenzie River, he has found the bones and pari of the skin a short distance from the shores of Great Slave Lake.1 On the Atlantic coast they have not been recorded north of Nova Scotia. On the Pacific slope the most northerly record is Stuart Lake, British Columbia, though on the immediate coast of the Pacific they are not reported farther north than Howe Sound, in southern British Columbia, and probably do not occur much beyond this point. In the United States they are gener- ally distributed, except in the higher mountains, and in many places are extremely numerous. They are absent or very rare in eastern North Carolina, according to Mr. Outran) Bangs,2 and perhaps the same is true of other small areas within their range. The subgenus Ghincha ranges but a short distance into Mexico; 0. estor reaches southern Chihuahua in the Sierra Madre, holzneri enters Lower California, and variatis occurs in eastern Mexico in the Rio Grande Valley. The subgenus Zeucomitra3 occupies nearly the whole of Mexico, and its range overlaps that of the subgenus Chincha in southern Arizona. It is chiefly confined to the Sonoran and Tran- sition zones, not occurring in the tropical lowlands except on the coast of Oaxaca. The southern limit of its range is not known, but it probably does not extend beyond the highlands of Guatemala. HABITS. The large skunks are wholly terrestrial, living in caves, in the de- serted burrows of other animals, or in barrows of their own excavation. They do not avoid the habitations of man, but seem rather to prefer the clearings and pastures about the farm, and frequently make their abode under a house or other building. They are sluggish in move- ment, and usually show little fear of human beings, depending for safety more on the efficacy of their malodorous discharge than on attempts to escape. Although chiefly nocturnal, they are often seen moving about in the daytime, especially in the morning and evening. They hibernate only during the severest part of the winter, and prob- ably only in northern latitudes. Their food consists largely of small mammals, reptiles, batrachians, insects, and birds' eggs; and they apparently have no difficulty in secur- ing an abundance, for the}T frequently become excessively fat. They are particularly fond of insects, and during the seasons when grasshop- pers are abundant feed extensively on these pests. In many parts of 1 Canadian Nat. and Geol., VII, p. 139, 1862. 2Proc. Biol. Soc. Wash., X, p. 139, 1896. 3 See p. 39. 12 NORTH AMERICAN FAUNA. [no. 20. the United States they have been found useful in destroying- the 'white grub,' a great pest in lawns and meadows. Fish and crusta- ceans, when available, and even carrion or other refuse, are readily eaten. They occasionally rob the poultry yard, but such depredations as they commit are more than offset by their destruction of noxious mammals and insects. Vegetable matter is sometimes eaten. Mr. E. A. Goldman reports that he has found roots and wild fruits in the stomach of a hooded skunk. Skunks have been extensively trapped for furs ever since the settle- ment of the country by white men,1 and within the last few years attempts have been made to breed them in confinement; but although w skunk farms' have been started in several States, the industry seems not yet well established. EXTERNAL CHARACTERS. Skunks are too well known to need more than a passing reference to their external appearance. They are stocky animals, particularly heavy behind, with slender nose, small ears, short legs, long fore claws, and long bushy tail. The pelage is long, loose, and silky; the under-fur is quite dense, especially in northern latitudes; and the tail hairs are coarse and flaccid. The colors are black and white, the white usually arranged in a narrow frontal stripe and two na row dorsal bands or one broad band covering the whole back, but the pattern of coloration is subject to much variation. During spring and summer the pelage often becomes much worn and faded, and the glossy black of the winter pelage is frequently replaced by a dull brown. There is apparently but one molt in a }rear, and this usually occurs in late summer or in autumn. The females are always smaller than the males and less specialized in their cranial characters, which are often of so little prominence that they do not serve to readily distinguish closely related species. The difference in size is most marked in the larger species. Thus in Chincha notata the average length of the skull is 10 percent greater in males than in females, and the average zygomatic breadth 11 per- cent greater. The teats usually number 10 to 11. The young are born in litters of from 1 to 10, and reach maturity at a very early age, the bones of the skull completely coalescing before the teeth show any appreciable signs of wear. That which particularly distinguishes skunks from other animals is their means of defense, consisting of a characteristic malodorous fluid, which, when ejected, speedily disarms the boldest aggressor. The 1 During the forty years from 1850 to 1890 the Hudson Bay Company alone shipped over 250,000 skunk skins to England. An;.. 1901.] NOMENCLATURE. 13 fumes from the fresh Liquid are overpowering in their pungency, and are possessed of remarkably penetrating and Lasting properties.3 The liuid is secreted by two anal glands, similar in character to those possessed by other members of the Mustelidse, but Larger and more muscular. Dr. C. Hart Merriam has given a concise description of these glands, as follows: The glands lie on either >i«ic of the rectum, and are imbedded in a dense, gizzard- like mass of muscle which serves to compress them so forcibly that the contained fluid may be ejected to the distance of four or five metres (approximately 13 to 16$ feet). Each sac is furnished with a single duct that leads into a prominent, nipple- like papilla that is capable pf being protruded from the anus, and by means of which the direction of the. jet is governed.2 MATERIAL*. The present revision is based chiefly on a study of the specimens in the collections of the IT. S. Biological Survey and the IT. S. National Museum, supplemented by much additional material from the collec- tion of the American Museum of Natural History, New York, and the private collections of Dr. C. Hart Merriam and Mr. Outram Bangs. The total number of specimens examined exceeds 950, among which are all the existing- types. The Biological Survey collection contains good series of skins and skulls of most of the species, and is particu- larly rich in large series of extra skulls, which have proved of great value in showing the range of individual variation. My thanks are cordially extended to Dr. C. Hart Merriam for the privilege of using the collections under his charge. Thanks are also due Dr. J. A. Allen and Mr. Frank M. Chapman, of the American Museum of Natural History. Mr. Gerrit S. Miller, jr., of the U. S. National Museum. Mr. D. G. Elliot, of the Field Columbian Museum, Mr. Witmer Stone, of the Academy of Natural Sciences, Philadelphia, and Mr. William Brewster, of the Museum of Comparative Zoology, Cambridge, for the loan of material under their charge. I am espe- cially indebted to Mr. Outram Hangs for the use of a tine series of skunks from his collection, including the types of all his species. The illustrations in the present paper are from photographs taken by Dr. A. K. Fisher, of the Biological Survey. One new subgenus (Leucomit/ra) and three new species and subspe- cies (Chincha Occident nil* nuijor, O. occidentalis notata, and C. platy- rliinti) are here described. NOMENCLATURE. Three generic and 35 specific names have been applied to the skunks of this group. The general similarity in external characters among the members of the genus has led to the inclusion of widely different • 1 For a valuable account of the chemical properties of this fluid see T. B. Aldrich, Journ. Exp. Med., I, pp. 323-340, 1896; II, pp. 439-452, 1897. 2 Mammals of the Adirondack Region, Trans. Linn. Soc., X. Y., I, p. 76, 1882, 14 • NORTH AMERICAN FAUNA. [no. 20. forms under one name, while the extreme variability in some species has multiplied synonyms. The confused state of the nomenclature makes necessary a brief statement of the status of each name. GENERIC NAMES. Viverra Linnaeus, 1758. Syst. Nat., ed. X, pp. 43-44, The genus comprehended five species: V. ichneumon, V. memphitis, V. 2>utorius,V. zibetha (usually considered the type), and V genetta. The American skunks were referred to it by authors prior to Cuvier, 1800, but it is now restricted to the Old World civets. Mephitis Cuvier, 1800. Lecons d'Anat. Comp., I, tabl. 1. Cuvier proposed this genus to include the ' rfwuffettesj but as no species are mentioned, it is necessary, in order to determine its specific constituents, to refer to his 'Tableau Elementalfe,' published two years previously.1 In this work he places the mouffettes as a subgroup of Mustela, and mentions two species, Viverra putorius Linn, and V. mephitis Linn.2 The latter was removed in 1842 to become the type of the genus Chincha Lesson, which leaves V. putorius, one of the little spotted skunks, as the type of Mephitis. The name Spilogale^ pro- posed in 1865 by (Tray for the little spotted skunks, will therefore have to be abandoned, becoming a synonym of Mephitis, which thus unfortunately proves to be the name of this group of skunks instead of the group for which it has so long been used. Chincha Lesson, 1842. Nouv. Tableau Regne Anim., Marnm., p. 67. Lesson proposed this as the name of a subgenus of Mephitis, with Chincha americcwia&s the type species; hudsonica Richardson is given as a variety, but no other species are placed in the group. The ref- erences show that his type species is based on Viverra mephitis Erx- leben, which in turn is based on T" mephitis Schreber — a plainly recognizable species. If we assume (as we can with all propriety) that Cuvier, in placing ' Viverra mephitis L.' as one of the types of his genus Mephitis referred to V. mephitis of Gmelin's edition, we then have for the type of Chincha a species which is one of the two orig- inally composing the Cuvierian genus Mephitis, and one that is like- wise identifiable, for V. mephitis Gmelin is based on V. mephitis Schreber. 'Tabl'. Element, de l'Hist. Nat. des Anim., p. 116, 1798. 2 This name does not appear in either the 10th or 12th editions of Linnaeus, so we must assume that the reference is to Gmelin, 1788; memphitis of the 10th edition is a different name (see p. 18) and Cuvier's description shows that the animal he had in mind was a skunk, and not the composite species which Linmeus described under the name memphitis. atjg., 1901.] SPECIFIC AND SUBSPECIFIC NAMES. 15 It is perfectly clear, therefore, that Lesson intended to apply the name < fhincha to the large two-striped North American skunks, and it is used for those in a generic sense in the present paper. [Note. — In a list of names of Mexican animals by A. L. Herrera,1 ■ M/i!t!sii.s macrura: is used for the lsorillo: (the vernacular name of the hooded skunk in Mexico). All of Herrera's names are modified forms of accepted generic names, and until such time as his system shall have been adopted, his names require no consideration.] SPECIFIC AND SUBSPECIFIC NAMES. americana {Mephitis) Desmarest, 1818. Nbuv. Diet. < 1' Hist. Nat., Paris, XXI, p. 514. This nana1 was applied to a composite species, including- as varieties all the skunks of North and South America. The first author to use it in a restricted sense was Sabine, who in 1823 applied it to the skunks of Canada.'- He says that the animals "under examination are the particular sort designated as the Vwerra mephitis of Gmelin," which is variety No. 7 of Desmarest. The name americana therefore becomes a synonym of mephitis. avia (Mephitis) Bangs, 1898. Proc. Biol. Soc. Wash., XII, p. 32. Under this name Bangs described a form from San Jcse, Illinois. It proves to be closely related to mesomelas, of which it is considered a subspecies. bivirg-ata (Mephitis americana) Hani. Smith, 1839. Jardine's Nat. Libr., Manmi., I, p. 196. The description under this name probably refers to the eastern skunk (Chinch" j.>>/f/'r/a), but no t}Tpe locality is assigned, and the species can not be identified with certainty. cinche i Viverra) Miiller, 1776. Natursyst. Suppl., p. 32. The species thus named is based on the chinehe of Buffon, which probably belongs to this genus, but is not certainly identifiable. Muller follows Buffon in ascribing the animal to South America, where Chincha does not occur. concolor ( Mephitis) Gray (Verreaux MS.?), 1865. Proc. Zool. Soc. London, 1865, p. 149. Published by Gray under his * var. <■. vittata\ with 'Verreaux MS. V as authority. The form thus named is evident!}7 only one of the many variations to which vittata is subject, and although Gray after- wards raised it to subspecitic rank 3, the name must be regarded as a synonym. 1 Sinonimia Vulgar y Cientifica de los Principalis Vertebrados Mexicanos, p. 30, 1899. 2 Franklin's Narrative of a Journey to the Polar Sea (1819-1822), App., p. 653, 1823. 3 Cat. Oarnivora, Brit. Mus., p. 138, 1869. 16 NORTH AMERICAN FAUNA. [no. 20. edulis (Memphitis) Coues (Berlandier MRS.), 1877. Fur-Bearing Animals, p. 236. Berlandier (as quoted by Coues) gives a brief description of a skunk from Mexico which is probably macroura. The name will stand as ;t synonym of the latter. elongata (Mephitis mephitica) Bangs, 1895. Proc. Boston Soc. Nat. Hist., XXVI, p. 531. Under this name Bangs described the Florida skunk. It proves to be a distinct species, most nearly related to put Ida. estor (Mephitis) Merriam, 1890. N. Am. Fauna No. 3, p. 81. The species described under this name by Merriam is a well-marked form inhabiting- Arizona and northern Mexico. fetidissima (Mephitis) Boitard, 1842. Jardin des Plantes, Mamm., p. 147. Boitard proposed this name on the same page with olida and putida. The form thus named is probably one of the many variations of the common eastern skunk, but as no more definite t}^pe locality is assigned than ' United States,' the name can not be specifically applied. fceda (Yivcrra) Boddsert, 1785. Elenchus Animalium, I, p. 84. This is one of the many early names which can not be specificalty identified. Boddsert quotes Buffon, Schreber, and other early authors, but assigns his species to Mexico. fcetulenta (Mephitis) Elliot, 1899. Field Columbian Museum, Pub. 32, Zool. Ker., I, no. 13, p. 269. The form to which this name was applied proves to be practically identical with that to which Bangs had previously given the name spissigrada, of which fwtulenta is, therefore, a synonym. fossidens (Mephitis) Cope, 1896. Proc. Acad. Nat. Sci. Phila., p. 386. This name was applied by Cope to a well-marked fossil species from the Pleistocene bone caves at Port Kennedy, Pennsylvania. frontata (Mephitis) Coues, 1875. Bull. U. S. Geol. it (ieog. Surv. Terr., 2d series, no. 1, p. 7. The form described by Coues under this name was based on a skull found in the Post-pliocene deposits of Pennsylvania. While the char- acters assigned to it by him are of slight weight, it differs from the living species in dental characters, and seems worthy of specific recognition. holzneri (Mephitis occideritalis) Mearnn, 1897. Proc. U. S. Nat. Mus., XX, p. 461. Mearns proposed this name for the form of occidentaUs occupying southern California and northern Lower California (type locality, San Isidro Ranch). It is a rather poorly marked subspecies that ranges north to the vicinit}' of Monterey Bay. ait... 1901.] SPECIFIC AND SUBSPECIFIC NAMES. 17 kudsonica (Mephitis americana) Richardson, 1829. Fauna Boreali-Americana, I, Mamm., p. 55. Richardson's aame, applied to the skunk of the Northern plains, proves available for the species ranging from Colorado north to the interior of Canada. intermedia (Mephitis vittata, var. b. ) ( rray, 1869. Cat. Carnivora Brit. Mns., p. 138. This name is applied to one of the numerous varieties of the hooded skunks due to individual variation. It is a synonym of vittata. laticaudata i Mephitis) E. I teoffroy, 1803. Cat. Mamm. Mus. National d'Hist. Nat., Paris, p. L09. This name can not be satisfactorily identified. The references indi- cate that the type specimen belonged to the genus Chi/ncka, but the description applies more nearly to one of the South American skunks. leptops (Mephitis) Cope. 1899. Journ. Acad. Nat. Sci. Phila.,2d scries, XI, pt. 2, p. 235. This name was applied to a fossil species from the Port Kennedy hone caves. Montgomery County, Pennsylvania. I have not had an opportunity to compare it with specimens of recent species. long-icaudata (Mephitis) Tomes, 1861. Proc. Zool. Soc. London, 1861, p. 280. Under this name Tomes described the form from Duefias, Guatemala. Specimens from near the type locality have been compared with those of iDaaroura and are found to be practically identical. macroura (Mephitis) Lichtenstein, 1832. 1 Darst. Saugeth., pi. 4H, with accom- panying text. Lichtenstein's plate and description are sufficient to identify the species thus named by him, which came from the 'mountains north- west of the City of Mexico.' This is the first name applied to any of the hooded skunks. memphitis ( Vwerra) Linnaeus, 1758. Syst. Nat., p. 44. A name applied by Linnaeus to an unrecognizable species, evidently part skunk and part nasua There is nothing in the description to indicate even the genus to which it refers, and it seems best to reject the name as indeterminable. It is quite distinct from mephitis, as shown below. mephitica ( Viverra) Shaw, 1792. Museum Leverianum, p. 171. This name has been generally adopted by recent authors for the eastern skunk, and was restricted by Bangs in 1895 to the northeastern JThe date of publication of that portion of the ' Darstellung ' in which the skunks are described is fixed by Lichtenstein himself in a later paper (Abh. Akad. Wiss. Berlin for 1836, p. 303, 1838). 2909— No. 20—01 2 18 NORTH AMERICAN FAUNA. [no. 20. species. Schreber's name mephitis, however, is perfectly tenable for the same species, and being of earlier date is adopted instead of mephitica. mephitis ( Viverra) Schreber, 1776. Saugth., Ill, p. 44J, tab. 121. Recent authors have rejected this name on the ground that it is preoccupied by Viverra memphitis Linn., 1758, supposed to be a mis- print for mephitis. Through the kindness of Mr. J. E. Harting, of London, who has examined for me a cop}T of the tenth edition of Linnaaus's 'Systema Naturae' in the Linnsean Society Library, which contains numerous corrections and annotations in the author's own handwriting, sufficient evidence has been brought to light to show that Linnaeus intended to write 'memphitis, which must therefore be considered entirely distinct from mephitis. In this copy of Linnaeus Mr. Harting finds that although certain alterations1 are made in the diagnosis, the spelling of memphitis is not corrected. Still other evidence of the validity of the name is adduced by Mr. Harting, who writes: "That he [Linnaeus] meant memphitis to stand is clear, not only from his leaving the spelling uncorrected in his annotated tenth edition, but by his rewriting it in a marginal note to his copj^ of Ra}r,s 'Synopsis Animalium,' 1693, wherein, on p. 181, he identifies it with Ray's ' Yzquiepatl seu Vulpecula.7 Opposite these words he has written ' Viverra memphitis. ,' distinctly." Schreber's name mephitis, then, is not preoccupied, and being accompanied by a recognizable description, is adopted as the first name for the Canada skunk. mesomelas (Mephitis) Liechtenstein, 1832. Darst. Saugeth., pi. 45, with accom- panying text. Lichtenstein's specimen, on which the species thus named is based, was secured from a natural history dealer, and was said to have come from Louisiana. The measurements show it to be the small species inhabiting the southern Mississippi Valley, subsequently named scrutator by Bangs. mexicana (Mephitis) Gray, 1837. Charlesworth's Mag. Nat. Hist., I, p. 581. Gray's brief description under this name probably refers to one of the forms of macroura, of which mexicana is therefore a synonym. 1 The alterations referred, to consist in the . substitution for the original diagnosis, of the following: "V. alba subtus nigro maculata;" and the entry in the margin of the query, "an nasim?" The diagnosis as altered agrees with the later portion of the description, and the nasua element is largely removed, although the references point principally to that animal (of. Bangs, Proc. Boston Soc. Nat. Hist., XXVI, p. 529, 1895). That Linnaeus himself was in doubt as to the validity of the species is shown both by the annotations above mentioned, and by the omission of the name from his twelfth edition. AUG., 1901.] SPECIFIC AND SUBSPECIFIC NAMES. 19 milleri (Mephitis) Mourns, 1897. Proc U. 8. Nat. Mus., NX, p. 467. The form thus named was described from Fort Lowell, Arizona. It proves to be a northern race of the hooded skunk, for which this Is the only name. obtusatus (Mephitis) Cope, 1899. Journ. A. ad. Nat. Sci. Phila., 2d scries, XI, pt. 2, p. !_•:>(>. Cope gave this name to an extinct species based on a single jaw discovered in the Port Kennedy bone deposit (Pennsylvania), but his type can not now be found. It is described as a very small species, the size of a weasel. occidentalis (Mephitis) Baird, 1857. Mamm. N. Am., p. 194. Under this name Baird described the western skunk. It is a wide- ranging species for which there is no other name. olida (Mephitis) Boitard, 1842. Jardin des Plantes, Mamm. p. 147. As in the case of fetidissima, described on the same page, this name was probably intended to apply to one of the forms of the eastern skunk, but in the absence of a definite type locality, no specific appli- cation can be made. orthostichus (Mephitis) Cope, 1896. Proc. Acad. Nat. Sci. Phila., p. 389. Cope described the species thus named from remains found in the Port Kenned}' bone caves, Pennsylvania. It appears to be quite dis- tinct from any living- species. putida (Mephitis) Boitard, 1842. Jardin de.s Plantes, Mamm. p. 147. This is the first name applicable to the eastern skunk in connection with which the type locality is definitely fixed. Boitard refers to it as " La Moufebte de New Jersey ■' and gives a brief description of the animal. The name is adopted for the species long known as m&phitica. scrutator (Mephitis mephitica) Bangs, 1896. Proc. Biol. Soc. Wash., X, p. 141. Under this name Bangs described the form from Louisiana as a subspecies of the Canada skunk. It proves to be quite distinct from the latter, and is a well-marked form, for which, however, Lichten- stein's name mesomelas, of much earlier date, must be used. spissigrada (Mephitis) Bangs, 1898. Proc. Biol. Soc. Wash., XII, p. 31. The form thus named was described by Bangs from Sumas, British Columbia. It proves to be a northern race of occidentalis. varians (Mephitis) Gray, 1837. Charlesworth's Mag. Nat, Hist., I, p. 581. The description of the form thus named is inadequate, and the type locality (; Texas') indefinite, but Gray's statement in a later paper that the tail is as long as the body fixes the name to the large skunk of southern and western Texas. The type specimen can not be found. 20 NORTH AMERICAN FAUNA. [no, 20. vittata (Mephitis) Lichtenstein, 1832. Darst. Siiugeth. plate 47, with accompany- ing text. Lichtenstein described under this name the species found at San Mateo del Mar, Oaxaca, Mexico. It proves to be a well-marked form of the hooded skunk group. Genus CHINCHA Lesson. Generic characters. — Skull highly arched, highest in frontal region; rostrum truncated with slight obliquity; posterior margin of palate nearly on a line with posterior border of last molars; periotic region not inflated; mastoid and paroccipital processes prominent; post-orbital processes not prominent; coronoid process of mandible conical, erect. Dental formula: i. ^%, c. jij, pm. 3^3, m. 2i2 = 34. Snout not greatly produced; nostrils lateral; tail long and bushy. The genus Chincha is a member of the subfamily Mephitina?, which also includes three other genera. Mephitis, TMosmus,1 and Canepatus, of which only the first two occur in North America. Mephitis may be readily distinguished by the color pattern, which consists of numerous white stripes (always more than two) broken into many patches or spots. Thiosimis differs from the other two North America genera in having the snout produced, and bare for a considerable distance from the tip, with the nostrils inferior; tail short and sparsely haired; and usual color pattern black, with a solid white band covering the entire back and part or all of the tail. Key to species and subspecies. [Based on adult males.] Audital bulke greatly inflated; back usually either wholly black or wholly white (Subgenus Leucomitra, p. 39) Smaller; bullfe much inflated vittata (p. 43) Larger; bullae less inflated. Tail much longer than body mitteri (p. 42) Tail not longer than body macroura (p. 41) Audital bullae not greatly inflated; back usually with a white stripe, divided posteriorly (Subgenus Chincha, p. 22) Palate with prominent spine. Tail longer than body elongata ( p. 27 ) Tail shorter than body putida ( p. 25 ) 1 Thiosmus Lichtenstein, Abh. Akad. Wiss. Berlin for 1836, p. 270, 1838. Material in the collection of the Biological Survey indicates that Conepatus Gray, described from Patagonia, will have to be separated from the white-backed skunks usually known under this name. The white-backed skunks will, in such event, require another name; and since Thiosmus Lichtenstein seems to be the earliest one that is available, it is here provisionally adopted for this group. aug.. 1901.] LIST OF SPECIES AND SUBSPECIES. 21 Palate without prominent spine. Tail less than halt the length of body mephitis ( p. 22) Tail more than half the length of body. Tail usually more than 350 mm. (Texas) varians (p. 31) Tail usually less than 350 mm. Skull small (basilar length1 less than 66 mm.). Body stripes very broad estor ( p. 32) Body stripes narrower. Tail more than 250 mm holzneri (p. 38) Tail less than 250 mm. skull smaller (basilar length usually under 60 mm.). mesomelas (p. 29) skull larger | basilar length usually over 60 mm. ) avia ( p. .">(>) Skull large (basilar length more than 66 mm.). Body stripes narrow: frequently not continuous notata (p. 36) Body stripes broad; always continuous. Tail usually more than 285 mm. Skull very broad (mastoid breadth more than 45 mm.) ..major (p. 37) Skull narrower (mastoid breadth less than 4.5 mm.). Rostrum very broad (breadth across post-orbital processes more than 24 mm. ) platyrhina (p. 39) Rostrum narrower (breadth across post-orbital processes less than 24 mm. ) .oceidentalis (p. 34) Tail usually less than 285 mm. Palate extending back of last mi >lars hudsonica ( p. 24) Palate ending on a line with last molars spisxigrada ( p. 35) List of Species and Subspecies, ■with. Type Localities. Subgenus ( Jhincha. Species and subspecies. Type localities. Chincha mephitis (Schreber) America (restricted to eastern Canada). hudsonica \ Richardson) Plains of the Saskatchewan. putida ( Boitard) New Jersey. elongaia ( Bangs) Micco, Brevard ( Jounty, Florida. mesorru las | Lichtenstein ) Louisiana. mesomelas avia (Bangs) San Jose, Illinois. mesomelas varians (Gray ) Texas. > star i Merriam) San Francisco Mountain. Arizona. oceidentalis (Baird ) Petaluma, ( Jalifo >rnia. oceidentalis spissigrada (Bangs) Sumas, British Columbia. nceidi ntalis notata nobis Trout Lake, Mount Adams, Washington. oceidentalis major nobis Fort Klamath, Oregon. oceidentalis holzneri (Mearns) San Isidro Ranch, Lower California. platyrh ina nobis South Fork Kern River, California. Subgenus Leitomitka. Chincha maeroura (Lichtenstein) Mountains northwest, of City of Mexico. iintcroura milleri (Mearns) Fort Lowell, Arizona. maeroura vittata ( Lichtenstein ) San Mateo del Mar, Oaxaca, Mexico. 1Basilar length of Hensel, measured from inferior lip of foramen magnum to pos- terior rim of alveolus of middle incisor. 22 NORTH AMERICAN FAUNA. [no. 20. Subgenus CHINCH A. Large Striped Skunks. Subgeneric characters. — Skull (PLs. V, VI, and VII) long and relatively narrow interorbitally; zygomata usually spreading broadly; inter- pterygoid fossa broad; palate ending either squarely or with a median notch or spine; audital bullae not greatly inflated; anterior palatine foramina usually small and narrow; mastoids, and sagittal and supra- orbital crests well developed. Size medium to large; build heavy; soles broad; ears not prominent; fur dense. The usual color pattern is as follows: A narrow median white stripe extends from nose to nape; a white dorsal band beginning with a broad nuchal patch, and narrowing between the shoulders, divides into two lateral stripes, which continue to the tail and sometimes down its sides; the rest of the body is black; the tail is of mixed black and white hairs, the white ones much the longer; all the tail hairs are white at the base. (See PI. II, iig. 1.) Great variability is exhibited by the species inhabiting the eastern United States — -putida, elongata, and mesomelas — which grade in color pattern from specimens wholly white above, including the tail, to specimens in which the only white areas are the frontal stripe and a dash on the nape. In the western species and in O. mephitis from Canada the variability is slight, and consists chiefly in the breadth of the body stripes and the amount of white on the tail. (See Pis. I, II, and III.) The white areas are often of a creamy hue, and are never mixed with black hairs, as they are in the hooded skunks (subgenus Leucomitra), and specimens in the black phase never show any trace of the white on the sides that is usual in the black phase of Leucomitra. Descriptions of Species and Subspecies. CHINCH A MEPHITIS (Sihreber). Canada Skunk. Viverra mephitis Schreber, Saugth., Ill, p. -444, tab. 121, 1776. Viverra mephitica Shaw, Mus. Leverianum, p. 171, 1792 (part). Viverra mephitis Oken, Lehrbuch der Naturg., p. 994, 1816. Chincha americana Lesson, Nouv. Tabl. Regne Anini., Mamm., p. 67, 1842. Mephitis mephitica Bangs, Proc. Boston Soc. Nat. Hist., XXVI, p. 533, 1895; Elliot, Synop. Mamm. N. A., Field Columbian Museum, Zool. Ser., II, p. 322, 1901 (part). Type locality. — America. Geographic distribution. — Eastern Canada — Nova Scotia, Quebec, and northern Ontario; west and north at least to Oxford House, Keewatin. General characters. — Size large; tail short and slender, mixed black and white; markings constant: skull large and massive; palate ending in an even curve, without notch, or spine. mto., 1901.] CHINCHA MEPHITIS. 23 Color. — The color pattern, which is quite constant, agrees with that described in the subgeneric diagnosis (p. 22). The white stripes are never very broad and are sometimes separated for their entire length, as in ( '. notata. They extend down the sides of the tail and project beyond its tip. (See PI. II, tig. 1.) ('/■ e< rra m< mphitis, Linn. ; but as shown in the remarks on nomenclature (p. 18), the two names are quite distinct. The Canada skunk is a large, stocky animal with a heavy skull and a very short tail that narrows gradually to the tip. It is most nearly related to hudsonica of the western plains, from which it differs in the slender tail, as well as in skull characters. From putida of the Atlantic States it differs in larger size, shorter tail, and constancy of markings. Skulls of females are considerably smaller than those of males, and are equaled in size by skulls of males in putida, but in other respects they maintain the characters assigned to the species. Specimens examined. — Total number, 18, from the following localities: Nova Scotia: Annapolis, 1; Digby, 1: Halifax, 1. Quebec: Lake Edward, 4. Ontario: Moose Factory, 1; North Bay, Lake Nipissing, 2; Little Pic River, Lake Superior, 1. Keewatin: Oxford House, 1; Pine Lake, 1. 1 All measurements are in millimeters. - Fide Bangs. 3Proc. Boston 8oc. Nat. Hist, XXVI, p. 533, 1895. 4 Examination of a large number of specimens from northern New York and New England shows that putida is the form occupying the greater part of the Canadian zone. 24 NORTH AMERICAN FAUNA. [no. 20. CHINCHA HUDSONICA (Richardson). Northern Plains Skunk. Mephitis americana, var. hudsonica Richardson, Fauna Boreali-Americana, I, Mamm., p. 55, 1829. Mephitis mephitica, var. occidentalis Merriam, Ann. Rep. IT. S. Geol. Surv. for 1872, p. 662, 1873. Mephitis mephitica hudsoyiio* Elliot, Synop. Mamm. N. A., Field Columbian Museum, Zool. Ser., II, p. 322, 1901 (part). Type locality. — Plains of the Saskatchewan. Geographic distribution.— Western Canada, from Manitoba to Brit- ish Columbia (east of the Cascades); south in the United States to Colorado, Nebraska, and Minnesota. General characters. — Size very large; tail heavy and of medium length, ending in a blunt brush wit/tout a white pencil; skull heavy, with a long palate; zygomata' broadly spreading. Color. — This species exhibits the usual color pattern of skunks of this group. The white stripes are of medium width, bifurcate just behind the shoulders, and extend nearly to the tip of the tail. The tail, which is very full and bushy, usually ends in a blunt, black brush, and has an indistinct band of white about two-thirds of the distance from the root to the tip. Cranial characters. — Compared with mephitis: Skull both longer and broader; zygomata very widely expanded; mastoid processes promi- nent; pal at i' long, ending- behind plane of posterior molars, without notch or spine; interorbital constriction marked; nasals and ascending branches of premaxilhe short; dentition heavy, the lower carnassial relatively large. (See PI. VI, figs. 1 and 2.) Measurements. — Average of three adult males from Saskatchewan, Montana, and Wyoming: Total length, 726; tail vertebrae, 268; hind foot, 82. Average of 3 adult females from Montana and Idaho, 602; 250; 71. Skull: (See table, p. U). General remarks. — This skunk may be readily distinguished from the skunks of the eastern States by its large size and bushy tail with- out a distinct white pencil. Compared with mephitis, its nearest ally, it has a longer and heavier tail, broad, heavy soles, and a skull of quite different proportions. The most noticeable difference between the skulls of the two species is in the widely spreading zygomata of hvd- sonica and the contracted mastoids of mephitis. The skull of // udson lea. viewed from above, appears narrower *than that of mephitis in the frontal and parietal regions. It resembles that of occidental!* in gen- eral shape, but has more widely expanded zygomata. Typical hud- sonica may be distinguished from both varians and occidental/is by the short ascending branches of the premaxillse. The species has an extensive range on the northern plains, and spreads over most of the northern portions of the Rocky Mountains from Colorado to British Columbia. Its range meets that of mephitis in Manitoba, and the two species may possibly intergrade, though aug., 1901.] CHINOHA PUTIDA. 25 present material docs not fully show thai such is the case. Specimens from British Columbia have slightly shorter tails, but in other respects arc typical. A large series from Arkins, Colorado, consisting of skulls and a few skins, show that two species arc present in i hat region. Of 31 skulls of adult males. It; are fairly typical hudsonica, L2 are just as typical varians, and 3are indeterminate. It is not difficult to separate them, the hudsonica scries having broad, heavy skulls, with spreading zygomata and lone- palates, the varians series much slenderer and nar- rower skulls, with shorter palates. No skins of the hudsonica form arc available, but those of the varians form are typical, except that they have the shorter tails usual in specimens of varicms from the northern part of its range. Under these rather unusual conditions it seems hardly possible to consider that the two species intergrade, but rather that their ranges overlap at this point, each remaining distinct, save for an occasional hybrid. Specimens examined. — Total number, 80, from the following localities: Mackenzie: Great Slave Lake, 1. Saskatchewan: Wingard, 1. Alberta: Jasper House, 1. British Columbia: Shuswap, 6; Sicamous, 4; Asheroft, 1; Okanagan, 4; Kamloops, 1; Stuart Lake, 2; Ducks, 1. "Washington: Fort Spokane, 1. Idaho: Bear Lake (east side), 1; Cceur d'Alene, 1. Montana: Stanford, 1; Great Falls, 1; St. Marys Lake, 1; Nyack, Teton Mountains, 1; Bear Paw Mountains, 3; Prospect Creek, near Thompson, 1; Tobacco Plains, 2; Yellowstone River, (2(5 miles from mouth), 1. Wyoming1: Rona, Sheridan County, I; Lower Geyser Basin, 1; Shoshone Lake, 1; Fort Bridger, 1; Big Horn Mountains, 1; Bull Lake Creek, Fre- mont County, 1; Fort Laramie, 1. Colorado: Arkins, Larimer County, 19 (skulls). Nebraska: Johnstown, 1. South Dakota: Custer, 1; Fort Pierre, 1; Fort Randall, 1; Fort Sisseton, Marshall County, 1; Rapid City, 1. Minnesota: Elk River, 9; Fort . Snelling, 1; Roseau River, at Point d'Orme, 1; Bois de Sioux, 1. CHINCH A PUTIDA (Boitard). Eastern Skunk. Mephitis putida Boitard, Jardin des Plantes, Mamm., p. 147, 1842. Mephitis olida Boitard, Jardin des Plantes, Mamm., p. 147, 1842. Mephitis fetidissima Boitard, Jardin des Plantes, Mamm., p. 147, 1842. Mephitis mephitica Baird, Mamm. N. Am., p. 195, 1857 (part); Elliot, Synop. Mamm. X. A., Field Columbian Museum, Zool. Ser., 11, ]>. 322, 1901 (part). Mephitis varians, var. I>. mephitica, Cray, Cat. Carnivora Brit. Mis., p. 137, 1869. Mephitis varians, var. e. chinga, Gray, Cat. Carnivora Brit. Mus., ]>. 137, 1869. Type locality. — New Jersey. G< ographic distribution. — New England and Middle Atlantic States; south to Virginia; west to Indiana. Gt neral characters. — Size medium; tail longer than that of mephitis, 26 NORTH AMERICAN FAUNA. [no. 20. black, with a distinct white pencil,' skull short and relatively broad zygomatically, with a prominent spine on the palate. Color. — This species exhibits the usual color pattern of the skunks of this group, with considerable variation in the breadth and extent of the white stripes. The tail is usually wholly black, excepting- a white pencil, which extends from 100 to 150 mm. beyond the end of the vertebrae. The white stripes are usually broader than in mephitis, and specimens frequently occur in which the back is almost wholly white. In many individuals, however, the stripes are much reduced both in length and breadth, and occasionally are entirely wanting, the white being confined to a small patch on the nape and the usual frontal stripe. Only two individuals in the large series examined have the white stripes continued down the sides of the tail. (See PI. I, fig. 1.) Cranial characters. — Skull of medium size and. relatively broad across zygomata/ interorbital constriction marked; posterior border of palate with a prominent spine; palate usually ending about on a line with last molars; mastoid processes prominent. (See PI. V, figs. 3 and 4). Measurements. — Average of 6 adult males from Hastings, New York: Total length, 575; tail vertebras 229, hind foot 60. Average of 6" adult females from same locality: 603; 223; 62. Skull: (See table, p. 11). General remarks. — This species, the common skunk of the Eastern States, is generally distributed from Maine to Virginia, but reaches the Mississippi Valley only, so far as known, in Ohio and eastern Indiana. It has long borne the specific name mephitica, which until recently it shared with all its congeners in eastern North America; but since this is a synon3rm of C. mephitis (Schreber), which is here adopted for the Canada skunk, it becomes necessary to select the next available name. While a great many names were proposed in the early years of the present century for the American skunks, the earliest used in a sufficiently restricted sense to be available for the present species is putida of Boitard, proposed in 18-12 for the skunk of New Jersey. There is but one species of skunk in New Jerse}7: hence, although Boitard's description is not in itself sufficient for identification, no doubt can exist as to the applicability of the name. The two other names (olida and fetidissima) proposed by him on the same page probably apply also to this form, but these are also inade- quately described and lack the specific mention of a type locality. In 1896 Bangs described a new form from Louisiana under the name Mephitis mephitica, scrutator (=mesomelas), and expressed the opinion that the skunks inhabiting the New England and Middle4 Atlantic States are intermediates between it and impltit i< ia(= mephitis).1 A careful study of a large number of specimens from eastern North America, including Bangs's types, shows that the skunks of the Atlantic coast States are very distinct from those inhabiting the JProc. Biol. Soc. Wash., X, pp. 139-144, 1896. ato., 1901.] OHINCHA ELONGATA. 27 Mississippi Valley, and that the form from New England and the Middle Atlantic States is also quite distinct from mephitis. Hence it is entitled to separation as a full species. It may be at once distinguished from mephitis by its longer tail and very different skull. The skull is not only much smaller and weaker than that of mephitis, hut tin1 relatively great breadth across the zygo- mata and the marked interorbital constriction give it a very different appearance. The skulls of some of the largest males are almost as broad zygomatically as those of mephitis, though very much shorter. The spine on the posterior border of the palate, which is a fairly con- stant character (absent in only a very few individuals in the large series examined), distihguishesy;////^ alike from mep/tif/N and from mrxomdas and its subspecies a/via. The presence of a palatal spine, the great mastoid breadth, and other differences in the skulls of males of this species, as compared with those of ama, whose range meets that of put iilii in Indiana, show that the two forms are quite distinct. Skulls of females, however, have the spine on the palate less pronounced, the zygomata less abruptly spreading, and the mastoid processes reduced, and thus resemble rather closely the females of avia. As in all the skunks, the skulls of the females of putida are very much smaller than those of the males, although occasionally the skull of a very large female will equal that of a small male. The skulls show a large amount of individual variation, particularly in size but also in other respects. Specimens from New Hampshire and northern New York average larger than the typical form, which might be regarded as due to the influence of mephitis but for the complete absence of other signs of intei gradation with that species. Sp( c'iik us ,-,(■( (in! n id. — Total number, 182, from the following locali- ties: New York: Adirondack .Mountains, 51 (skulls) ; Lake George, 1<>; Tomhan- nock, 1; Locust Grove, 11; Severance, 13 (skulls); Catskill Mountains, 1; Hastings-on-Hudson, 10; Highland Falls, :-!; Big Moose Lake, Hamilton County, 1; Mayville, 2; Westchester County, 1; Sing Sing, 3; Montauk Point, 2; Shelter Island, 19 (skulls); Miller Place, Suffolk County, 2. New Hampshire: Ossipee, 8. Maine: Bucksport, 1; BrookUn, 2. Massachusetts: Wilmington, 15; Burlington, 2; Taunton, 1; Ipswich, 1; Woods Hole, 2. Connecticut: East Hartford, 3. Pennsylvania: Carlisle, 1. Ohio: Garrettsville, 1. Indiana: Marion County, 3; B< >one County, 1; Denver, 1. Maryland: Jefferson, 1. District of Columbia: Washington, 3. CHLNCHA ELONGATA | Bangs). Florida Skcxk. Mephitis mephiiica dongata Banirs. Proc. Boston Soc. Nat. Hist., XXVI, p. 531, 1895. Mephitis elongate Bangs, Proc. Biol. Soc. Wash., X, p. 142, 1896. Type locality. — Micco, Brevard Count}', Florida. 28 NORTH AMERICAN FAUNA. [no. 20., Geographic distribution. — Florida (from vicinity of Lake Worth) to North Carolina, and in the mountains to West Virginia; west on the Gulf coast to the Mississippi River. General characters. — Size medium; tail very long, marked with white on the sides, and with a long' white pencil; marking's variable, but white stripes usually very broad; skull peculiar — larger than that of putida. Color. — Markings similar to those of putida, but white stripes aver- aging broader (about 45 mm. in width) and bifurcating about the mid- dle of the back; tail mostly white above and on the sides, this color reaching around and almost meeting beneath the root of the tail. Great variability in markings is exhibited, some specimens being wholly black, save for a few irregular white patches on the shoulders, others nearly all white above, including the tail, and mixed with white below. (See PI. I, figs. 2 and 3.) ('ran ial characters. — Skull larger and heavier than that of putida,' highly archied in frontal region; rostrum, very hroad; anterior palatine foramina large; zygomata spreading less abruptly; dentition heavy, lower carnassial especially large; audita] bulla? rather large; spine on palate prominent; interpterygoid fossa broad. M< tisnr, in, nis. Average of 2 aduH males from type locality: Total length, 703; tail vertebrae, 317; hind foot, 74. Average of 2 adult females from St. Marys, Georgia: 710; 315; 73. A series from Lake Harney and Mullett Lake are somewhat smaller, 3 adult males aver- aging 689; 292; 64. Skull: (See table, p. 44.) General remarks. — The Florida skunk is closely related to putida, but is veiy distinct from rnesomelas of the lower Mississippi Valley, from which it may be distinguished by its long tail and heavy skull. It was described by Bangs as a subspecies of the eastern skunk, but in a later paper he accorded it specific rank, and stated that its range does not meet that of mephitica (=putida) on the Atlantic coast, since the coastal region of North Carolina is practically uninhabited by skunks. More recently Mr. Bangs has received a specimen of elon- gata from Raleigh, North Carolina, the only one, so far as known, ever taken in the eastern part of the State. Two specimens from West Virginia are typical except that the tail is not quite the usual length — a character found occasionally even in specimens from Florida. Specimens exam ined. — Total number, 39, from the following localities: Florida: Micco, 2; New Berlin, 2; Blitches Ferry, Citrus County, 3; Lake Harney, 5; Mullett Lake, 5; Gainesville, 1; .Sebastian, 1 ; Fort Kissimmee, 3; Lake Worth, 1; Hernando County, 1. Alabama: Baldwin County, 2. Mississippi: Bay St. Louis, 2. Georgia: St. Marys, 2; Pinetueky, 2; Mcintosh County, 1; Nashville, 1. North. Carolina: Raleigh, 1; Weaverville, 2. West Virginia: Green Bank, 1; Travellers Repose, 1. atjg., 1901.] CHINCHA MESOMELAS. 29 CHINCHA MESOMELAS (Lichtenstein). Louisiana Skunk. Mephitis mesomdas Licht., Darst. Saugeth., pi. 45, fig. 2, with accompanying text, 1832. Mephitis mephitica scrutator Bangs, Proc. Biol. Sue. Wash., X, p. 141, 1896; Elliot, Synop. Mamm. X. A., Field Columbian Museum, Zool. Ser., II, p. 324, 1901. '/)//» locality. Louisiana. Geographic distribution. -West side of Mississippi Valley from southern Louisiana to Missouri; westward along the eoastof Texas to Matagorda Island; and up the Red River Valley as far at least as Wichita Falls. General characters. Siet very small; tail short, usually wholly Mack; skull small and relatively narrow. Color. More variable than varians, but apparently less so than putida. In specimens from the type locality the white stripes are nar- row, and usually terminate about the middle of the hack, though they occasionally extend to the root of the tail;1 tail usually wholly black; the white pencil generally absent, but if present, shorter than in putida. (See PI. II, tig. 2.) In specimens from Texas the stripes usually reach to the tail, and the coloration is more constant. Cranial characters. Skull very small mid relatively narrow; m; teeth small: audital bullae usually more inflated than in putida. (See PL VI, figs. •"> and -t.) Measurements. — Average of 4 adult males from Louisiana: Total length. 576; tail vertebras, 223; hind foot, 63. Average of 3 adult females from same localities: 566; 224; 62. Average of 4 adults (both sexes) from Marble Cave, Missouri: 628; 232; 67. Skull: (See table, p. 44.) (i, a, rnl remarks. — In Lichtenstein's original description of this species he remarked that his type was secured from a natural history dealer, and was said to have come from Louisiana. In a later paper2 he gave its range as ' Ludoviciana et ad Missouri jVu/vium.^ By rea- son of his assignment of a definite type locality, and by the aid of the measurements of his specimen,3 we are able to apply the name with certainty to the small species inhabiting the lower Mississippi 'Lichtenstein figures a specimen in which the white stripes reach to the tail and down its Bides. None of those examined from Louisiana have as much white as this specimen, but it is stated by Mr. Levi Spalding, of Iowa, Louisiana, that all gradations of color occur in the skunks of that section. When more specimens are obtained from the State, many individuals will probably be found that agree perfectly with the figure of the type. -Ahh. Akad. Wiss. Berlin for 1836, p. 277, 1838. 3His measurements are: Head and body, 1 ft. 7 in.; tail, 9 in.; hind foot, 2 in. Reduced to millimeters (assuming that he used the Rhineland foot) these measure- ments are as follows: Total length, 7o.'i; tail vertebra1, 235; hind foot, 52; which do not differ radically, except in length of body, from measurements of specimens recently taken in Louisiana. The great length of body was undoubtedly due to stretching of the skin. 30 NORTH AMERICAN FAUNA. [no. 20. Valley. The hairy soles ascribed by him to this species have not been observed in any species of the genus. The Louisiana skunk was later described by Bangs as a subspecies of ' 'mephitica? (= mephitis and putida). With mephitis it apparently has no connection; from putida it may be distinguished by its small size, short tail, and narrow skull with square palate. These charac- ters also distinguish it from elongata. In most of the specimens examined the tail ends in a blunt black brush, as in varlans and hud- sonica; but a few from both extremes of its range have a slender white pencil, shorter than in putida. One specimen from Marble Cave, Missouri, is nearly all white above, including the tail. The shape and size of the anterior palatine foramina are variable in this species, some individuals having the large rounded foramina which appear in the subgenus Leucomitra. Specimens examined. — Total number, 39, from the following localities: Louisiana: Cartville, Acadia Parish, 2; Point aux Loups Springs, 5; Calca- sieu Parish, 11; Calcasieu Pass, 1. Texas: Matagorda Peninsula, 3; Virginia Point, 1; San Antonio, 3; Aransas County, 1; Gainesville, 2; Henrietta, 1; Wichita Falls, 1. Oklahoma: Fort Cobb, Washita River, 1. Missouri: Marble Cave, Stone County, 7. CHINCH A MESOMELAS AVI A (Bangs). Illinois Skunk. Mephitis m Bangs, Proc. Biol. Soc. Wash., XII, p. 32, 1898. Type locality. — San Jose, Illinois. Geographic distribution. — Prairie region of Illinois, western Indiana, and eastern Iowa; boundaries of range imperfectly known. General characters. — Resembles mesomelas very closely, but skull slightly larger. Color. — The series from the type locality is variable; in some the white stripes terminate about the middle of the back, in others they reach to the root of the tail. Tail wholly black, with or without a white pencil. Cranial characters. — Skull slightly larger than that of mesomelas; z}'gomata more widely expanded; upper carnassial large; palate vari- able in length, ending sometimes in front of and sometimes behind plane of last molars. Measurements. — Average of two adult males from type locality:1 Total length, 641; tail vertebrae, 181; hind foot, 65; one adult female from Freeport, Illinois: 610; 220; 68.2 Shall: (See table, p. 44). General remarks. — This form is very closely related to mesomelas from which it differs chiefly in greater size and perhaps shorter tail. 1 Fide Bangs. 2 These measurements, taken in the flesh from a specimen which died in the National Zoological'Park, show that the species may have a s nnewhat longer tail than Bangs's measurements indicate, and not appreciably shorter than that of mesomelas. aug., 1901.] CHINCHA MESOMELAS VARIANS. 31 While its range meets that oiputida in Indiana intergradation seems not to take place. The differences between avi; Freeport, 1; 'Illinois,' 1. Indiana: Fowler, Benton County, 1. Iowa: Delaware County, 1. CHINCHA MESOMELAS VARIANS (Gray). Long-tailed Texas Skunk. Mephitis varians Cray, Charlesworth's Mag. Nat. Hist., I, p. 581, 1837. Mephitis macroura Ami. & Bach., Quad. N. Am., Ill, p. 11, L854 [not M. macroura Licht]. Mephitis mesomelas Allen, Bull. Am. Mus. Nat, Hist., VI, p. 188, L894 [not .1/. meso- melas Licht.]; Elliot, Synop. Mamm. N. A., Field Columbian Museum, Zool. Ser., II. p. 325, L901 [not M. mesomelas Licht,]. Type locality. Texas (specimens from lower Rio Grande Valley considered typical). Geographic distribution. — Southern and western Texas, eastern New Mexico, and adjacent parts of Mexico; north into Oklahoma, Colorado, Kansas, and Nebraska. (t, n, i;il characters. — Size large; tail very long; markings similar to those of hudsonica, constant: skull longer than that of mesomelas. Color. — Similar to that of hudsonica/ white stripes narrower than in estor/ tail ending in a black brush without a pencil; white hairs intermixed in the tail, usually showing prominently in upper surface to about the middle of the tail, where they form an indistinct band. Cranial characters. — Skull of medium size, smaller and narrower than that of hudsonica/ longer than that of mesomelas/ zygomata spreading less abruptly, and palate averaging shorter than in hud- sonica/ ascending branches of premaxillaB very long. Measurt im nts. — Average of 4 adult males from vicinity of Browns- ville. Texas: Total length, 758; tail vertebra", 393; hind foot, 71. Average of 4 adult females from lower Rio Grande Valley (Laredo and vicinity): 681; 376; 69. Skull: (See table, p. 44.) General remarks. — In Gray's original description of this species he remarks that it inhabits Texas, and in a later paper mentions that the tail is as long as the body.1 Two forms are found in Texas, either of which might be the subject of the original description, but only one of these, the larger, has a tail as long as the bod}'. To this form, therefore, the name is restricted.2 JCat. Carnivore P.rit. Mus., p. 186, 1869. 2Dr. J. A. Allen has endeavored to fix Lichtenstein's name, mesomelas, to this form (Bull. Am. Mus. Nat. Hist., VI, p. 188, 1894), but that this can not be done is evident upon comparing the measurements of this species with those of memmelas as given by Lichtenstein i see footnote p. 29). The specimens on which Dr. Allen based his views came from ( >klahoma and belong to the large form — not the small one, of whose presence in Louisiana he was at the time unaware. The body and tail measurements made by him from dry skins, do not correctly represent the aver- age measurements of varians; but the size and characters of the skull leave no doubt as to the identity of the specimens. 32 NORTH AMERICAN FAUNA. [no. 20. The species varies greatly in size: thus specimens from the lower Rio Grande Valley have much longer tails and rather smaller skulls than those inhabiting central Texas, Oklahoma, and Colorado. In southeastern Texas, in the vicinity of Matagorda Bay, it intergrades with mesomelas, whose range extends westward along the coast from Louisiana. In a series of 1 specimens from O'Connorport (opposite Matagorda Island), which are evidently intermediate, the markings are like those of varians while the skulls are small like those of mesomelas, and the tails average 312 mm. — much longer than the average tail measurement of mesomelas. Seven skins from Mason, Texas, show much more white than typical specimens, in this respect approaching estor, from which some of them can be distinguished only by greater size and longer tail. More than half of these Mason skins have the white hairs of the tail extending beyond the tip, and some have a dis- tinct white pencil. Specimens examined. — Total number, 139, from the following localities: Tamaulipas: Matamoras, 2; Mier, 1. Texas: Brownsville, 7; Corpus Ohristi, 1; Nueces Bay, 2; Santa Tomas, 1; Hidalgo, 2; Rio Grande City, 1; Padre Island, 1; Laredo, 2; Eagle Pass, 3; Mouth of Pecos River, 1; Presidio County, 2; El Paso, 2; East Painted Cave, Valverde County, 1; Indianola, 1; O'Connorport, 7; Rockport, .">; Fort Richardson, Jack County, 1; Berne, 1; Mason 7; Gail, 1; Colorado, 5; Fort Clark, Kinney County, 1; Sherwood, 2; Langtry, 2; Pecos High Bridge (Southern Pacific R. R. ), 2; Chisos Mountains, 1; Davis Moun- tains, 1. New Mexico: Hall Peak, 2; Eddy, 1. Oklahoma: Beaver River, Beaver County, 9. Colorado: Arkins, Larimer County, 20; Chivington, 1; Canon City, 1; Boulder County, 1; Costilla County, 2; Conrow, 1; Loveland, 1. Kansas: Cedarvale, 1; Neosho Falls, 1; Trego County, 5; Long Island, 2; Onaga, 3. Nebraska: Johnstown, 21 (skulls); Valentine, 1; Cherry County, 1; Loup Fork River, 1. CHINCHA ESTOR (Merriam) Arizona Skunk. Mephitis estor .Merriam, N. Am. Fauna No. 3, p. 81, 1890. Type local lit/. — San Francisco Mountain, Arizona. Geographic distribution. — Arizona, western New Mexico, Sonora, Chihuahua, and northern Lower California; south in the Sierra Madre to southern Chihuahua; limits of range unknown. General characters. — Size rather small; tail shorter than that of varians; much white on body and tall; skull resembling that of vari- ans, but smaller. Color. — White stripes on back very broad — almost confluent; pos- terior back wholly white in some specimens; tail of black and white hairs, the white longer and chiefly on the upper surface, where they AUG., L901.] < IIINCHA ESTOR. 33 extend beyond and nearly conceal the black; white pencil at tip. (Sec PL II. fig. 3.) Cranial characters. — Skull resembling1 that of various in general shape but smaller and slenderer/ palate ending about on a line with posterior molars, either square or with a very small notch; molars smaller than in either varians or occidentalism anterior palatine fora- mina small and narrow. (See PI. VII, tigs. 3 and 4.) Measurements. A.verage of 7 adult males from Arizona and adja- cent parts of Mexico: Total length, 639; tail vertebrae, 285; hind foot, 69. Average of 4 adult females from same localities: 580; 273; 63. Shall: (See table, p. 4-4.) General remarks. -The Arizona skunk is a very distinct species inhabiting Arizona. New Mexico, ami adjacent parts of Mexico. In southern Arizona and Chihuahua its range overlaps that of milleri, the two species being often found at the same place. By reason of this fact, and on account of the extreme variability of milleri, the two have been frequently confused by authors, and many references to estor really apply to miUeri. There need be no difficulty, however, in distinguishing them, either by skin or skull. The hooded skunks (to which group milleri belongs), while extremely variable, are usually either wholly black or wholly white on the back, and never have the two white stripes of Ghimcha; the tail is longer than the head and body (about 52 per cent of the total length). Estor is rather constant in markings, and has the white stripes of the other United States spe- cie- inclosing a small patch of black, while the tail is shorter than the head and body (about 44 per cent of the total length). In the few cases in which estor has the whole back white, the purity of the white will serve to distinguish it from milleri, in which the white is of a grayish hue. due to the intermixture of black hairs. The very pro- nounced skull characters distinguishing the two groups are pointed out under the description of the subgenus Leucomitra (p. 39). Estor differs from both varians and occidentalis in smaller size and shorter tail, and in the much greater extent of white on its bod}' and tail. Specimens from the Mexican boundary line at the west base of the San Luis Mountains are somewhat larger than the typical form of estor, and one of them is plainly referable to varians, so it is pos- sible that intergradation takes place between the two. Specimens examined. — Total number, 55, from the following locali- ties: Arizona: Sun Francisco Mountain, 3; Springerville, 3; Eiolbrook, 1; Cala- basas, 2; Yuma, 1; Fort Mohave, 1; Fort Verde, 12; Fori Huachuca, 1; Huachuca Mountains, 2; Huachuca Station, 1; Prescott, 1: Pinal County, 4; Whipple Barracks, 1. New Mexico: Fort Wingate, 1: Cloverdale, « riant County, 1, Hachita, 1. 2909— No. 20—01 8* 34 NORTH AMERICAN FAUNA [no. 20. Sonora (near Mexican boundary line): Santa Cruz River, 1 ; San Pedro River, J; Patagonia Mountains, 2; west side San Luis Mountains, 1; San Luis Springs, 1; Animas Valley, 1; San Bernardino Ranch (monument 77, Mex- ican boundary line), 1; La Noria (monument 111, Mexican boundary line), 1. Chihuahua: White Water, 1; Oajon Bonita Creek (near Mexican boundary), 1; Colonia Garcia, 6; Sierra Madre (near Guadalupe y Calvo), 1. Lower California: Poso Vicente, 1. CHINCHA OCCIDENTALIS (Baird). California Skunk. Mephitis occidentalis Baird, Mamm. N. A., p. 194, 1857. Type locality. — Petaluma, California. Geographic distribution. — Northern and central California, from vicinity of Monterey Bay northward, west of the Sierra and Cascades, to the Willamette Valley, Oregon. General characters. — Size rather large; resembling- hudsonica quite closely, but tail longer; skull relatively narrow across zygomata; palate rather long, sometimes with a distinct median notch. Color. — Closely resembling hudsonica in pattern of coloration; white stripes of medium width, and frequently extending down sides of tail, though the white hairs never reach beyond the tip. The mark- ings show little variation. Cranial characters. — Compared with that of hudsonica, the skull of occidentalis is much narrower across tht zygomata, which spread less abruptly and are more nearly parallel to the axis of the skull. Pal- ate rather long, usually with a distinct median notch, though this is not always present, even in the typical form. The largest skulls of occidentalis equal those of h udson lea in length and in mastoid breadth, but the majority are somewhat shorter and narrower. (See PI. VII, figs. 1 and 2.) Measurements. — Type:1 Total length, 800; tail vertebrae, 312; hind foot, 77. Average of 5 adult males from vicinity of San Francisco Bay: 693; 303; 78. One adult female from Auburn, California: 700; 330: 75. Shall: (See table, p. 44.) General remarks. — The California skunk was recognized as distinct by Baird in 1857. With its four subspecies it forms a well-marked group, quite distinct from any of the eastern members of the genus, and has an extensive range. It equals hudsonica in size, but has a longer tail, and differs materially in cranial characters. The colors are very constant, but the skulls show great individual variation. The length of the tail is also variable, ranging from 265 to 370 mm. A series of skulls from Cassel, Shasta County, California, average larger than the typical form, and two specimens in the series arc clearly intermediate between occidentalis and major. A specimen from Lake Tahoe likewise shows characters intermediate between 1 Fide Baird. Aug., tan.] CHINCHA OOOIDENTALIS SPISSIGRADA. 35 these two forms. On the north intergradation with spissigrada is shown by a large series from Lake Gushman, Washington; while to the southward the typical form merges gradually into kolzneri. Its relationships with notata and platyrhina arc not dear, but intergrada- tion with these is probable. Specimens examined. — Total number, L<>7, from the following locali- ties: California: Petaluma, 1: Glen Ellen, 1; Novate, 1; Nicasio, 5; Point Reyes, 2; Mi. Tamalpais, 1: Fairfield, 2; Walnut Creek. 1; Santa Clara, 1; Salt Springs, Fresno River, 1; Wawona, L2 (skulls); Pine City, 7 (skulls); Yosemite Valley, t: Mariposa County, .) i ranial characters. Skull similar to that of occidental is, hut shorter, and relatively broader across zygomata; rostrum averaging broader; bullae slightly larger: />. Average of 5 adult females from The Dalles, Oregon: 659; 286; 69. Skull: (See table, p. +4.) General remarks. —This form shows greater variability in markings than any other of the western skunks and is the only one in which the body stripes arc ever interrupted, In skull characters it resem- bles occidentalis quite closely, hut lacks tin1 notch in the palate and has much larger molars. It probably intergrades with both occiden- talis and major. Specimens examined.- -Total number, 41, from the following locali- ties: Washington: Trout Lake, Mount Adams, :;i (skins with skulls, 6; skulls only, 25); Rockland, Klickitat County, 1; Goldendale, 1. Oregon: The Dalles, 8. CHINCHA OCCIDENTALIS MAJOE subsp. nov. Great Basin Skunk. Typt from Kurt Klamath, I >regon. $ adult. No. 92238, I". S. Nat. Mus., Biological Sur- vey Coll. Collected Jan. 5, 1898, by B. L. Cunningham. Original number, 80. Geographic distriJmtion. — Eastern Oregon, northern California, and Nevada; east to the Wasatch Mountains in Utah. General char oxters. — Similar to occidental^ but much larger; hind font longer; skull larger and more heavily built. ('<>h>r. — Much as in occidentalis: white stripes broad, bifurcating near the middle of the back, and extending- only a short distance on the tail, which is nearly all black exteriorly. ( rcmial choyractt rs. — Skull larger and more heavily built than that of occidentalism rostrum broader and much flattened; hraincase broader ,nnl not xo J, .-, ji, thus giving a flattened appearance to the upper sur- face of the skull: dentition heavier; palate long, usually ending in a concave line, sometimes irregularly notched; ascending branches of premaxillae short and broad. Measurements.- -Average of 5 adult nudes from type locality: Total length, 705; tail vertebra1. 306; hind foot. 84. Skull: (See table, p. 44.) General remarks. — This subspecies seems to he the largest and heavi- est skunk in the genus; the hind foot is both longer and broader than in any other member of the genus, and the front foot is correspondingly large. The large skull with its broad hraincase readily distinguishes the form from its congeners. In the series from the type locality, the characters are constant, but the subspecies undoubtedly intergrades with occidentalis in northern California, and possibly with notata in northern Oregon. Specimens from western Nevada are provisionally included with majo?; although by reason of the fact that no males 38 NORTH AMERICAN FAUNA. [no. 20. have been examined — female skunks are less readily separable — their exact relationships are uncertain. Specimens from Ogden, Utah, are clearly referable to this form. Immature skulls of major may be dis- tinguished from those of both occidentalis and hudsonica by the broad premaxillse. Specimens examined. — Total number. i;T, from the following locali- ties: Oregon: Fort Klamath, 6; Tule Lake, 2; Plush, Lake County, 1; Shirk, Har- ney County, 1; Harney, 2; Elgin, 1. Washing-ton: Touchet, 1. California: Lassen Creek, Shasta County, 1; Honey Lake, 1; Sierra Valley, 1. Nevada: Carson, 1; Reno, 1; Quinn River Crossing, Humboldt County, 1. Utah: Ogden, 6; Provo, 1. CI 1 1 X( 1 1 A ( )CCIDENTALIS HOLZNERI (Mearns) . Southern California Skunk. Mephitis occidentalis holzneri Mearns, Proc. U. S. Nat. Mus., XX, p. 461, 1897. Type locality. — San Isidro Ranch, Lower California. Geographic distribution. — Southern California, from vicinity of Monterey Bay south into Lower California: east to the Sierra Nevada and San Bernardino Range; limits of southward range unknown. General characters. — Similar to occidentalis but smaller. < '<>/ or. — There are no appreciable color differences to distinguish this form from occidentalis. (See PI. Ill, tig. 1.) Cranial characters. — Skull smaller than that of occidentalis and relatively narrow in mastoid region ; audita! bullae more circumscribed, but rather prominent; palatal notch usually absent. Measurements. —Type (adult male): x Total length, 665; tail vertebra?, 273; hind foot, 72. Average of 3 adult males from Twin Oaks, San Diego County: 637; 292; 71. Average of 3 females (barely adult) from San Diego County: 605; 291; 61. Skull: (See table, p. 44.) General remarks. — This form does not differ greatly from occiden- talis, but averages considerably smaller in cranial measurements. The body and tail measurements of holzneri are slightly smaller than those of occidentalis, and the hind foot is decidedly shorter. Specimens from Ventura and adjacent counties are intermediate between the two forms, having the long tail of occidentalis, but very small skulls. Specimens examined. — Total number, 11, from the following locali- ties: Lower California: San Isidro Ranch, 3. California: Dulzura, 3; Twin Oaks, 5; Witch Creek, 2; Unlucky Lake, San Diego County, 1; Pacific coast at Mexican boundary, 1; Santa Ysabel, San Diego County, 1; Santa Paula, 2; San Fernando, 1 ; Ventura River, 1; Santa Ynez Mission, 3; Gaviota Pass, Santa Barbara County, 2; San Emigdio, Kern County, 1; Morro, 1; San Luis Obispo, 3; San Simeon, 2; Monterey, 3; South Fork Kern River ( 25 miles east of Kernville), 1; Three Rivers, 5. /•'/'(/, Mearns. A.06., 1901.] SUBGENUS LEUOOM1TR A. 39 CHINCHA PLATYRHINA sp. nov. Broad-nosed Skunk. Type from Smith Fork of Kern River (25 miles east of Kernville), California, J1 adult, No. ],'-;:;, r. S. Nat. Mus., Biological Survey Coll. Collected July."), L891, by Vernon Bailey. Original number, 2998. Geographic distribution. Known only from the type locality and from Owens Valley. General characters. Externally much like occidentalis,' skull with peculiarly shaped zygomata ami very broad rostrwn. Color.- Resembling occidentalis: white stripes of medium width, produced only a short distance on the sides of the tail; tail Mack externally, except for an indistinct hand of white on the upper surface about | the distance from base to tip. Specimens from Owens Valley have most of the tail hairs chestnut instead of black, exteriorly; and in one case most of the body is chestnut. This is probably due to fading. < '/■fluid! characters. — Skull resembling that of occidentalism but shorter and relatively broader: much flattened in frontal region; ros- t rn in very broad — actually and relatively broader than in largest speci- mens of occidentalis y zygomata spreading less abruptly and in an even curve nearly parallel to the axis of the skull; palate nearly square, with only a very slight notch; audital bullae small, and slightly inflated: tube of auditory meatus short; nasals short and broad. Measurements. — Type (adult male): Total length, 750; tail vertebras, 320; hind foot. 90. Average of 3 adult females from Owens Valley: (Hit; 332; 79. Skull: (See table, p. M). G< io ml r< marks. — This species shows no marked charactersof pelage to distinguish it from occidentalis, which it equals or exceeds in size, hut its peculiar skull at once serves to separate it. As typical holzneri occurs at the type locality of platyrhiha, it is evident that intergrada- tion does not take place between the two species at this point. It is quite likely, however, that platyrhina intergrades with major; but until there is material available to show such intergradation, a binomial designation seems preferable. The skull resembles that of major in general shape, but is very much smaller, and relatively broader across the postorbital processes. Specinu nsexamined. — Total number, 9, from the following localities: California: South Fork Kern River, 5; Owens- Valley, 2; < >wens Lake, 2. Subgenus LEUCOMITRA. LEUCOM1TKA, subgen. nor. Hooded Skunks. Type. — Chincha macroura (Lichtenstein). From mountains northwest of the City of Mexico. Subgeneric characters. — Skull (PI. VIII) short and broad; inter- pterygoid fossa narrowly U-shaped: palate without notch or spine: 40 NORTH AMERICAN FAUNA. [no. 20. anterior palatine foramina large 7. Average of 4 adult females from same local- ities: 594; 297; 60.5. Skull: (See table, p. 44). General remarks. — This species was described by Lichtenstein in L832, and his name has been very generally adopted by subsequent authors. His type came from the mountains northwest of the city of Mexico, and specimens from Salazar are' considered typical. The Guatemala form has been described by Tomes under the name longi- caudata, but specimens from the vicinity of Dueiias show that it is not separable from the typical form. The southern limit of range is unknown, but it will probably be found little south of the highlands of Guatemala. To the northward macroura grades imperceptibly into mill, ri. Specimens em/mined. — Total number, 60, from the following locali- ties: Mexico: Tlalpam, 4; Amecameca, 1; Salazar, '1. Hidalgo: Marques, 1; El Chico, 1; Irolo, 1; Encarnacion, 1; Zimapan, 1; heal del Monte, 1. Michoacan: Querendaro, 4; Patzcuaro, 1; Nahuatzin, 2. Colima: Colima, 4; Hacienda Ma: < 'asas ( rrandes, 2; Guadalupe y ( lalvo (mountains near), .'?. Sinaloa: Sierra de Choix, 1. Coahuila: Jimulco, 2; La Ventura, 1. CHINCHA MACROURA VITTATA (Lichtenstein) . Least Bo an Skunk. Mephitis vittata Lichtenstein, Darst. Saugeth.,pl. 47. with accompanying text, L832. Mephitis concolor Gray (Verreaus MS.?), Proc. Zool. Sue. London, L865, p. L49. Mephitis vittata var. b. intermedia Gray, Cat. Carnivora Brit. Mus., p. L38, L869. Mephitis vittata var c. concofor Gray, Cat. Carnivora Brit. Mus., p. L38, L869. ////'< locality. — ' San Matteo el Mar,' Oaxaca, Mexico. Geographic distribution. — Known only from the type locality; prob- ably ranges over the coast region of Oaxaca and Chiapas. General characters. Smaller than macroura, with very small skull, and slightly developed mastoids and sagittal crest. Color. As in macroura; pencil not distinct. In the scries of 18 topotypes examined. ;» are in the black phase. Cranial character's. — Skull decidedly smaller than that of macroura/ relatively narrow across zygomata, and mastoids much red/uced; sagit- tal crest very slightly developed; bulla disproportionately large. (See PL VIII, figs. 3 and 4.) Measurements. -Average of 6 adult males from the type locality: Total length. 558; tail vertebrae, 275; hind foot, 60.4. Average of L2 adult females from the type locality: 585; 300; 59.5. Skull: (See table, p. 44.) General remarks. — This form was described by Lichtenstein at the same time as macroura, and his description was accompanied by a good figure of a specimen in the black phase. The description is too meager in details, in the absence of material from the type locality, to clearly establish the validity of the subspecies, but all uncertainty has been removed by the fine series of specimens collected at San Mateo del Mar in 1895 by E. W. Nelson and E. A. Goldman. The form is well marked, differing more from macroura than does milleri. It is the smallest of the genus, and -may be recognized by this fact as well as by its peculiar skull characters. The pelage is rather thin and coarse. The tail is relatively longer than that of macroura. Although, as usual, the males have the greater average skull meas- urements, yet in total length the average of the females is greater. A very large proportion of the skulls examined had been infested with parasites, and the distortion of the cranium through this cause is greater than in any other species examined. One specimen in par- ticular lias the swelling produced fully 7 mm. above the normal top of the cranium and spread to a width of 28 mm., although the mastoid breadth of the same specimen is but 31 mm. Specimens examined. — Eighteen, all from the type locality. 44 NORTH AMERICAN FAUNA. [no. 20. ■paStuaAB suaunoads jo laqtaim i- 10 co co m co c< 1" ■sjiqoui %sv.\ jo auBjd o} mnuSiuu nara'BJOj r-l rH O CO CO CO CO CO CO CO CO — 00 CO O ID cO'tbio'ioncN'ioido) COCOCCCOCOCO-PCOCO-fCO ■|[)iill.l[ 1K|K[UcI-1S(i,I CO^J1^COCOCOCOCOCOCOCOCOCOCOCOCOCO"*COCO-PCO CO CO CO CO •I[)^U.)[ lU}TqH,I •qjptojq [lijiqjojajtii ^suai in in CM -i CM ia -t CO CR CM CO t> CO to Tj< lO H CM CM lO rH lO en CO o en o> X 1- r. oo o - / t> I- 1 - r. Cft CO 00 O CO CO o o CI CM rH rH Ihiid ssoiob qipeajg •qipuaaq in CO m CO m m CO CO lO en CM H in CO L~ t> CO CO m in in -cp CM CM -tf CM CO :1 CM CM o CM CM Cl = o : i r. CM CM O CI M CO CM H C-l CO CI CI Cl Cl in -f rH CO CO 1- en m rH to CO lO -# CI CO t* X iC -r o lO :: Oi CO o CO CO l~ -* -r to in 1^ SO to "dl C-l 1~ — — = ■f 00 CO -r :: CCMMCOMCOCOMCO •qjptfajq t* H ■* « 'C i-H t> t^ M t-' -f ci rH -r Cl C* CO Cl CO © © 00 jo i{?3ii9[ iBiis^g; iO 00 © ci co © © © r*- o © io iO iO CM CO 1^ 00 >C t- I> © © iC H M B Ot CO O0 'H Cl OJ W O O t> O Cl © CC © iO © iO »o »o © »o © lO I> © © © © ■q^Straj lusug in lO H tO in -- CO lO j: rH CO to CO CM CO CO CO 00 co in rH m ■O Cl to CC -- S CO m o o i^ — o •* CM -1 en CM m CO ~ i- to I> i^ to to in to to to in to to to to ~ i* to to to to 3 e £ <5 o o u ^ ^ t- tc* 'fz '* -g =; ■? e -5 £ &H S - c A ■d •-§ £ 3 s -s | o 2 a * a 'bo'bVjO'b'bot'DO^o'oO'bcu^o'bot'ij'bo AUG., 1901.J AVKKAiiK CRANIAL MEASUREMENTS. 45 ?l • .: 1- — - x '- ~ / I- X ' '~ X :'. i- :: M h - 5i X. — — X ^-< O CO i— ' —*:: — :: :: :: — '■'■ '■'■ '-'■ '■'■ ■" — to co /:' — :: — :: :: :: c> in e >o « - :t / to A o "if i-i o a> :: :: :: :: :: :: :: -i Tl > ,- ,- CO 1- iK o» /" r. >: C^l 1 - s 1 - >n Oi /; :i ?i :: ?1 ci t- IS OC r-i CO 00 CO ri 00 Q t-i 0> a ' r ,-^ — ri ?i — — — — — :: * oo «, g^ r- - _ -, t. i^ iC N !C co - H •-: -- -i - — " _ -• "1 - O o ri o — -r. i~ — t- — 1~ CC :: CO .- »- s - to -1 -1 — :: — :: -r CO CO - ■-: T .- t~ s. 01 oi - te CO CO !- -1 ( -l Ol X L.0 — — CO CO ?) L" iC iC ,- t- '- CO •*>« ■* to IC If I- ~ r 0-1 ■^ I - SB m ■-Z iC .- IK p. - K - c~ tN CO 1- CO -r •-T CO i- -- "I" Ol 1~ t~ - .n >o SiS'fi'So'Sii.S o — " — ^- -"- c — — x x s O « s , mesomelas, 29, 31. mexicana, 41. milleri, 12. obtusatus, 19. occidentalis, 24, 34. olida, 25. orthostichus, 19. putida, 25. scrutator, 29. spissigrada, 35. varians, 31. vittata, 43. Skunk: Arizona, 32-34. Broad-nosed, 39. California, 34-35. Canada. 22-23. Cascade, 36-37. Eastern. 25-27. Florida, 27-28. Great Basin, 37-38. Hooded. II 12. Illinois, 30-31. Least Hooded, 13. Long-tailed Texas, 31-32. Louisiana, 29-30. Northern II led, 12-13. Northern Plains, 21-2-">. Puget Sound. 35-36. Southern California, 38. Skunks: Hooded, 39-43. Large striped, 22-39. Little spotted. 10, 14. White-backer 1. 10. Spilogale, 10, 14. Thiosmus, 10, 20. Viri rra, 14. cinche, 15. fosda, 16. memphitis, 9, 17. mephitica, 22. mephitis. 18, 22. putorius, 14. 47 PLATE T. [Greatly reduced, and relative sizes not accurately shown, owing to differences in preparation of skins.] Fig. 1. Ckincha jmtida (Boitard). Burlington, Massachusetts. (No. 77878, U. S. Nat. Mus.) 2. Chincha elongata (Bangs). Fort Kissimmee, Florida. (No. 64017, U. S. Nat. Mus. ) 3. Ohincha elongata (Bangs). Fort Kissimmee, Florida. (No. 64016, U. S. Nat. Mus.) 48 North American Fauna, No. 20. Plate I. Skins of Chincha. 1. Chincha putida. 2,3. Chincha elongata. 2909— No. 20—01- 49 PLATE II. [Greatly reduced; and relative sizea not accurately shown, owing to differences in preparation of skins.] Fig. 1. Chincha mephitis (Schreber). Pine Lake, Keewatin. (No. 107226, U. S. Nat. Mns. ) 2. Chincha mesomelas (Licht. ) . Calcasieu Parish, Louisiana. (No. 99831, U. S. Nat. Mus. ) 3. Chv wha estor (Merriam). Holbrook, Arizona. (No. 53209, V. S. Nat. 31 us. ) 50 North American Fauna, No. 20. Plate II. 1. Chincha mephitis Skins of Chincha. 2. Chincha rm ■-■mm las. 3. Chincha estor. PLATE III. [( rreatly reduced, and relative sizes not accurately shown, owing to differences in preparation of skins.] Fig. 1. Chincha occidentalis holzneri (Mearns). Three Rivers, California. (No. 31244, U. S. Nat. Mus.) 2. Chincha occidentalis notata Howell. Trout Lake, Mt. Adams, Washington. (No. 87042, U. S. Nat. Mus.) 3. Chincha occidentalis spissigrada (Bangs). Neah Lay, Washington. (No. 88650, U. S. Nat. Mus. ) 52 North American Fauna, No. 20. Plate III. Skins of Chincha. 1. Cliincha occidentalis hokneri. '-!. Chincha occidentalis notata. 3. Chincha occidentalis spissigrada. PLATE IV. [Greatly reduced, and relative sizes not accurately shown, owing todifferenees in preparation of skins.] Fig. 1. Chincha (Leucomitra) macroura (Licht. ). Perote, Vera Cruz. (No. 54225, U. S. Nat. Mus. ) 2. Chincha (Leucomitra) macroura. milleri (Mearns). Oamoa, Sonora. (No. 95927, U. S. Nat. Mus. ) 3. Chincha (Leucomitra) macroura milleri (Mearns). Oamoa, Sonora. (No. 95923, U. S. Nat. Mus. ) 4. Chincha (Leucomitra) macroura milleri (Mearns). Camoa, Sonora, (No, 95931, U. S, Nat. Mus.) 54 North American Fauna, No. 20. Plate IV. Skins of Chincha < Leucomitra^. 1. Chincha macroura. 2,3, I. Chinclia nacroura mttleri. PLATE V. [Natural size.] Figs. 1 and 2. Chincha mephitis (Schreber). $, Lake Edward, Quebec. (No. 3805, Coll. E. A. & 0. Bangs.) 3 and 4. Chincha putida (Boitard). $, Highland Falls, New York. (No. 2020, Am. Mus. Nat, Hist.) North American Fauna, No. 20. Plate V. Skulls of Chincha. 1,2. Chincha mephitis. 3,4. Chincha pulida. PLATE VI. [Natural Mze.] Figs. 1 and 2. Chincka hudsonica (Richardson). A LETTER OF TRANSMITTAL. U. S. Department of Agriculture, Washington, D. C, July 5, 1901. Siu: 1 have the honor to transmit herewith for publication, as No. 21 <>f North American Fauna, two special reports on the natural history of little-known parts of the northwest coast of North America, the Queen Charlotte Islands, British Columbia, and Cook Inlet, Alaska, both by my assistant. Wilfred H. Osgood. Owing to the absence of definite information concerning- the faunas of these areas, Mr. Osgood was sent there to conduct biological explorations during the field season of 1900. The results of his trip form an important contribution to the natural history of the northwest coast region. Respectfully. C. Hart Merriam, Chief, Biological Surrey. Hon. James Wilson, Srrretary of Agriculture. 3 CONTENTS. Page. Natural history of the Queen Charlotte Islands 7 Introduction and itinerary 7 Acknowledgments 9 Physiography 9 Fl< »ra 11 Fauna 16 Life zones 20 Bibliography 22 Annotated list of mammals 25 Annotated list of birds 38 Natural history of the Cook Inlet Region 51 Intri xluctii >n and itinerary 51 Physiography 51 Flora 53 Fauna 56 Life zones 59 Previt his work 60 Annotated list of mammals 61 Annotated list of birds 72 5 ILLUSTRATIONS. PLATES. Plate I. Map of the Queen Charlotte Islands Frontispiece. II. Fig. 1 . Moss-grown spruce, Cumshewa Inlet 12 Fig. 2. Salmon-herry thicket, Cumshewa Inlet. III. Fig. 1. Mouth of stream, Cumshewa Inlet 20 Fig. 2. Shore of Bare Island, Skidegate Inlet. IV. Fig. 1. Skull of Ursiis carlottae 30 Fig. 2. Skull of Ursus americanus. V. Figs. 1 and 2. Skull of Mustela cavrina 34 Figs. 3 and 4. Skull of Mustela nesophila. VI. Fig. 1. Peat hog and mixed woods near Tyonek 52 Fig. 2. Looking toward Turnagain Arm from head of Bear Creek. VII. Fig. 1. Peak at head of Bear Creek 60 Fig. 2. Canyon of Upper Bear Creek. TEXT FIGURE. Figure 1. Outline' map of the Queen Charlotte Islands, showing extent of the Hudsonian zone 21 6 No, 21. NORTH AMERICAN FAUNA. September, 1901. NATURAL HISTORY OF THE QUEEN CHARLOTTE ISLANDS. By Wilfred H. Osgood, Assistant Biologist, Biological Survey. INTRODUCTION AND ITINERARY. The Queen Charlotte Islands lie off the coast of British Columbia, just south of the Alaskan boundary, between latitude 51 55' N. and .~>4 l.V X. They are slightly farther from the mainland than any of the islands of the Alexander Archipelago, to the north of them, and are not in the track of regular coasting steamers. The3T were visited by several of the early navigators of the northwest coast, but until 1787 no name was given them and no account of them had been pub- lished. In this }• ear Capt. George Dixon cruised about the islands from July 1 to August 3, trading with the natives and roughly chart- ing the roast. He named the group after Queen Charlotte, the consort of George III of England, and in the report of his voyage which appeared later included a very interesting account of his visit, together with maps and illustrations. In the early part of the nineteenth cen- tury various fur-trading vessels stopped frequently at the Queen Char- lottes, and later the discovery of gold and coal in small quantities has caused sporadic invasions by prospectors. No important attempt has been made, however, to develop the resources of the islands. The interior has not been explored to any great extent, and proba- bly will not be for some time to come, since the difficulties of travel are insurmountable to ordinary expeditions. The principal harbors and most of the east coast have been surveyed from time to time by officers of the royal navy, and the late Dr. George M. Dawson spent the summer of 1878 in studying the geology and littoral topography of the group. The report1 that he published is exceedingly interest- ing and important. It contains accounts of the history, geology, and ethnology of the islands, with some notes on the natural history, more especially of invertebrates. The vertebrate fauna as a whole had never been studied until the present year, however, and the little that was known of it was entirely due to the zeal of Rev. J. H. Keen, who for lGeol. Survey of Canada, Report of Progress for 1878-79, Pt. Ill, pp. 1-239, Mont- real, 1880. 7 8 NORTH AMERICAN FAUNA. [no. 21. eight years was engaged in missionary work at Massett. The present white population consists of several missionaries and three or four other men, who are engaged in trading and fishing. To supply necessities to these and to carry mail to them and the educated natives, small steamers make irregular trips to the islands. From one of these steamers I was landed June 13, 1900, with ray assistant, Mr. Edmund Heller, at the fishing station in Cumshewa Inlet, known as Clew (also spelled Klue), on the north side of the inlet about 10 miles from its mouth. We were very kindly received by Mr. W. H. Dempster, who conducts a plant here for obtaining oil from the dogfish which abound in the waters about the islands. We made small collections at Clew, and then devoted a number of days to cruising about the inlet in a canoe, collecting and making observations at various points. The first and most important trip was to the head of West Arm of Cumshewa Inlet, where we camped from June 17 to 25. After the coast at this point had been worked an expedition was made to the top of the highest mountain near the head of West Arm.1 This trip was exceedingly trying, as we were obliged to carry food, bedding, and traps on our backs and beat our way through the deep forest and heavy underbrush. We reached an altitude estimated at 4,500 to 5,000 feet. After working here in the Hudsonian zone as much as possible in the brief time available we returned to Clew, and from there made short trips to Louise Island and the mouth of the inlet. The next move was to Prevost Island, at the south end of the group, which was reached by means of a small fishing schooner. We dropped anchor July 3 in Houston Stewart channel, near Ellen Island, from which point both Prevost Island and the south end of Moresby Island were worked. In a few days we sailed north again and landed at Skidegate July 9. Here work was done about Skidegate Inlet and on the south end of Graham Island until July 18, when the trip was drawn to a close. Our entire time on the islands was thus about five weeks, and we were able to visit the three largest and most important of the group, namely Graham, Moresby, and Prevost. The weather during practically all this time was extremely disagreeable, rain being not only continuous for long periods, but often so severe as to interfere seriously with work. Most of the natives, unfortunately, were away at the time. They find employment in summer at the salmon canneries on the coast, the men being skilled in fishing and the women in packing. They are much reduced in numbers, and the entire population is divided between the two small villages of Massett and Skidegate, though ruins of former villages are abundant on other parts of the islands. lrThis mountain is indicated, but not named, on the map published by Dawson. The only Indian I was able to interview said it was sometimes called Haida Moun- tain; but the white men had no name for it, though it is a very conspicuous peak. sept.. 1901.] PHYSIOGRAPHY. V ACKNOWLEDGMENTS. The preparation of this report and the following one1 has been greatly facilitated by the free access which has been accorded me to the collections of tin* V . S. National Museum. Mr. Robert Kidgway, curator, and Dr. Charles \Y. Richmond, assistant curator of birds, have been exceedingly kind, not only in placing at my disposal the collections under their charge, but in numerous other ways as well. 1 am likewise indebted to Mr. (ierrit S. Miller, jr., assistant curator of mammals, for similar favors from his department. Mr. Joseph Grinnell, of California, has kindly forwarded me specimens for examination from his private collection. Through the kindness of Rev. J. H. Keen it has been possible to include in the list of birds many migratory and winter resident specie-. Mr. Keen lived at Massett for eight years, and during1 that time collected and identified a large number of birds and other ani- mals. He has very generously furnished a list of Massett birds, with notes for use in the present report, giving all the species positively identified by him. Special acknowledgment has been made elsewhere to Mr. Frederick Y. Coville and other botanists who have so kindly assisted in the determination of plants (see p. 13). PHYSIOGRAPHY. The Queen Charlottes are part of a submerged mountain chain like most of the other large islands, of the same coast, and were regarded by Dawson as a continuation northwestward of the ranges of Vancouver Island and the Olympic Peninsula. Their general trend is northwest and southeast, practically parallel with the mainland. The greatest length of the entire group is 156 miles,1 and the greatest width 52 miles; the area is unknown. The main islands of the group are. consecutively from north to south: North, Graham, Moresby, and Prevost (set1 frontispiece). All are very closely connected, the width of each intervening channel being reduced, at least at some points, to less than a mile. The shortest distance between the Queen Charlottes and the islands very closely connected with the mainland is i;7 miles, from Kose Point, Graham Island, to Stephens Island. The wide channel known as Hecate Strait, which lies between the Queen Charlottes and the mainland, is rather shallow; that part between Graham Island and the mainland seldom exceeds 2<> fathoms in depth. Graham Island is the largest of the group; its greatest length is 67 miles, and its width 52 miles. The coast on the north end is very deeply indented by Massett Inlet, and to a lesser degree by Naden Harbor; on the east side it is comparatively regular, and the west is character- ized by deep, unsurveyed sounds. The east side of the island is low distances are stated in nautical miles, and on the authority of Dawson 10 NORTH AMERICAN FAUNA. [no. 21 and comparatively level; the northwest part is slightly higher and somewhat rolling; and the southwestern corner is quite mountainous. This mountainous district may be roughly indicated as extending from Cartwright Sound south to Skidegate Inlet and east to Bearskin Baj^. Some of the mountains are high enough to maintain perennial banks of snow, which feed numerous streams; these, however, are not very large, and are exceeded by the lowland streams, which drain greater areas, particularly those in the northern part of the island, emptying into Massett Inlet and Naden Harbor. Heavy forest covers almost the entire island and fringes the coast to the very water's edge. In the northern part, not far from Massett, a few open, swampy meadows are known, and near Rose Point there are grassy sand hills, but else- where all is dense forest. Moresby Island is next in size to Graham; it is, in fact, about 5 miles longer, but is so dissected by sounds and inlets that its area is very indefinitely known, though certainly less than that of Graham Island. Its east coast bounds a succession of inlets, which make deep, transverse cuts into it and frequently unite with each other to detach small islets. The island is thus a mere skeleton; or perhaps it might be more properly called a mere backbone, since it is a nearly continu- ous mountain chain. Apparently the only part of it which is not mountainous is the peninsula lying between Skidegate and Cumshewa inlets. Cumshewa Inlet, the first deep indentation in the east coast south of Skidegate Inlet, is one of the largest of the island. Its south side, formed by Louise Island, and the region about its upper end are very mountainous. Among the peaks is one which rises near the head of West Arm to a height of -1,000 feet or more. From the inlet, its sharp cliffs and heavy snow banks present a rugged, imposing appearance, much heightened by contrast with the low, rounded, and somewhat undulating, forest-covered hills near the shore line. One very deep canyon cuts down its east side and through it a stream of moderate size runs into West Arm. Numerous smaller streams enter the inlet in the same vicinity. To the east, west, or south from the summit of this mountain one looks out over many other snow-laden peaks not so high, but of the same character and crowded together in tremendous masses. These mountains are practically continuous from the north end of Moresby Island south at least as far as Skincuttle Inlet. From a boat about 10 miles offshore in Hecate Strait one can look across Darwin and Juan Perez sounds and obtain an excellent view of the San Cristoval mountains, the best part of the chain. Prevost Island is the southernmost of the Queen Charlottes. It is quite small, being but 11£ miles in length and about 8 miles in extreme width. It is low and rolling, and not so densely forested as Graham and Moresby. The hills on the north end are perhaps 600 feet in alti- tude and in other parts of the island they probably do not exceed 1,000 sept., 1901.] FLORA. 11 feet. A few small streams take their rise in the interior, several of which empty into Houston Stewart Channel. FLORA. The vegetation of the northwest coast region, which is so well known for its almost tropical luxuriance, is probably nowhere more highly developed than in fcheQueen Charlotte Islands. The magnificent dark forests are comparatively endless, the underbrush is omnipresent and well-nigh impenetrable, and mosses and lichens everywhere festoon the trees and shrubs and carpet the rocks and soil (see Plate II, fig-. 1). The Indian, on the rare occasions when he can not travel by canoe, discreetly follows the beach; hence the interior wilderness remains almost as trackless as it* human beings had never set foot on the islands. Relatively open forest is found on the higher slopes of the mountains, but can be reached only by hand to hand conflict with the tangle lying between it and tide water. From the tops of the mountains in the northern part of Moresby Island one can look out over vast stretches of forest to the northeast on Moresby and Graham islands as far as the eye can perceive. Of coniferous trees at least seven species are found, namely, the Sitka spruce (Pipea sitchensis), the western hemlock (Tsuga heterophylla), the alpine hemlock (Tsuga mertensiana), the giant cedar {TJmja plicata), the yellow cedar (ChamcBeyparis nootkatensis), the northwest coast pine (Pinus con- torta), and the Pacific yew (Taxus hremfolid). The Sitka spruce is the most important. It is well distributed and generally becomes large, being second in size only to the giant cedar. It was found to be the predominating tree about Cumshewa and Skide- gate inlets, and, though smaller, equally abundant on Prevost Island and the south end of Moresby Island. Owing to exposed position and rocky soil, the trees on Prevost Island are rather small, and dead tops are so mixed writh the live ones that from a little distance the dark green hillsides appear to be uniformly overcast with a light hoariness. Dawson reports spruce1 as abundant in Skincuttle Inlet and about Darwin Sound on Moresby Island; he also found large forests of it in the eastern and northern parts of Graham Island, particularly about Naden River. Much of this timber is merchantable, though as yet no serious attempt to exploit it has been made. About Cumshewa Inlet the spruces stand in magnificent groves, the grandeur of which is appreciated only when one gets above the tangle of undergrowth and obtains an unobstructed view of the tall, straight, reddish-barked trunks, column after column extending far into the forest, until the dim light is finally entirely obscured and individual trees can no longer be distinguished. 'The spruce spoken of by Daw sun was called in his report Abies menziexii, but from the context it is evident that Pveea sitchensig was meant. 12 NORTH AMERICAN FAUNA. [no. 21. The giant cedar {Thuja plicata) is sparingly scattered through the forest in all parts of the islands except the higher mountains. Large individuals were rarely seen near the shore, owing, doubtless, to the fact that for many years the Indians have used the most accessible ones for making canoes. The yellow cedar {Chamcecyparis) is rather rare, except at high elevations. A few individuals of small size were found at the head of West Arm, Cumshewa Inlet, and at the head of Rose Harbor. Dawson found the species in cold places about other inlets of Moresby Island, as well as on the more exposed west coast. And Rev. J. H. Keen reports it in small quantities near Massett. On the mountains of the northern part of Moresby Island it is very abundant, and ranges from an altitude of about 2,000 feet to the upper limit of timber. The western hemlock {Tsuga heterophylla) is probably second in abundance to the spruce, and its distribution, below an altitude of 2,000 feet, is also general. It does not, however, occur in great num- bers on the actual shores, like the other conifers, but becomes more abundant inland. The very deepest, darkest forests are largely composed of this hemlock. The alpine hemlock {Txuya mertensiana) was found only in the moun- tains near the head of Cumshewa Inlet. It appears with the yellow cedar at an elevation of about 2,000 feet, and soon becoming well established, persists to the highest limit of trees. It is slightly more abundant than the 3rellow cedar, and with it straggles in fantastic shapes up the ridges or flattens in thick mats on sunny slopes. Now and then it attains fair size and regularit}' of branching in cold canyons or about small seepage pools in little heather meadows. The northwest coast pine (Pinus contorta) is rather rare. A few small individuals were noticed on rocky detached islets in Skidegate and Cumshewa Inlets and in Houston Stewart Channel; and, curiously, a few very depauperate individuals were found well toward the sum- mit of the mountain near the head of Cumshewa Inlet. The Pacific yew {Taxus brevifolia) was found on Cumshewa Inlet from Clew to the head of West Arm, being most abundant about West Arm. It is quite common around the shores of Prevost Island and the south end of Moresby Island. It is said to occur toward the west coast in Skidegate Inlet, and Rev. J. H. Keen reports that it is not uncommon in some places near Massett. It does not grow to large size and is always found near the shore or on the very edge of the water, which it overhangs at high tide. The only deciduous trees of importance are alders, willows, and wild crab apples. All are abundant but are somewhat limited to the immediate vicinity of the coast or the borders of streams. The alder {Alnus oregona) grows to a relatively large size; individuals from 10 to 15 inches in diameter were frequently seen. The willow {Sallx Nouh American Fauna, No. 21. Plate II. Fig. 1 .—Moss-grown Spruce, Cumshewa Inlet. Fig. 2.— Salmonberry Thicket, Cumshewa Inlet. Habitat of Melozpiza f. ruflna. sei-t. I'.uu.i FLORA. 13 scauleriand) is also of fair size and is generally distributed. The Oregon era!) apple {Pyrus rivularis) is found about streams and along the borders of sandy beaches as well as in occasional open places in the forest, in company with elderberry {Sambucus racemo&us), dog- wood (Comu8 occidentalis), honeysuckle {Lonicerd in/vohtcrata), and wild currants {Ribes bracteosum and R. laxifiorwn). A hawthorn ( < '/uifd ,■< vispina) also occurs, but evidently is rare, as we found it at > >nt one locality, near a small stream on Louise Island. The underbrush is largely composed of several species of huckle- berry (Vaccinium), the sallal (Gaultheria shallon), and the salmon berry (Rubus spectabilis- see Plate II, rig*. 2). Menstesia, Viburnum, and Amelanchier are locally abundant. Throughout the dam]) forest are many ferns, mosses, liverworts, and numerous species of flowering plants that thrive in such an environment. The1 few and small open meadows that are occasionally to be found teem with grasses, sedges, buttercups, beach peas, vetches, monkey flowers, thistles, lilies, and large cow parsnips The flora of the higher parts of the mountains is much like that of similar altitudes on the mainland. As the forest becomes more open the character of the herbaceous plants changes, and numerous pretty glades are found carpeted with heathers {J'ln/lln- doc< and Cassiope) and sprinkled with dwarf laurel {Kalinin glauca), dwarf huckleberries (Vaccinium ccespitosum), and cowslips {Caltha paliisfris). Most of our time was devoted to collecting mammals and birds, but a small collection of plants, representing nearly all the species observed was also made. From these specimens the following list has been prepared through the kindness of Mr. Frederick V. Coville, Botanist of the V . S. Department of Agriculture and Honorary Curator of the U. S. National Herbarium. Mr. Coville has not only generously given his own time and that of his assistants to the identification of speci- mens, but has enlisted the very necessary aid of several eminent spec- ialists, who have authoritatively named specimens in special groups. The liverworts were very kindly determined by Prof. A. W. Evans, of Yale University; the mosses by Prof. J. M. Holzinger, of the Min- nesota State Normal School. wTith the exception of the twTo species of Dicranaceae. which are given on the authority of Dr. R. H. True; the ferns by Mr. William I\. Maxon, of the U. S. National Museum, and the flowering plants by Mr. Frederick V. Coville and Mr. W. F. Wight, of the U. S. National Museum. u NORTH AMERICAN FAUNA. Plants. [no. 21. JUNGERM ANSIAI'K .K. Porella navicularis (l. a i.. ) uxni;. Radula bolanderi gottsche. Scapania bolanderi aust. Diplophylleria albicans (l. ) trevis. Frullania nisquallensis si-li.iv. Herberta adunca (dicks.) s. f. gray. DICRANACK.K. Dicranum fuscescena turn. Dicranella heteromalla sen. BRYACK.K. Bartramia glaucoviridis c. m. & k. Mnium glabrescens kindb. HYPXAt'K.K. Eurhynchium oregonuni (sulliv. ) i.. & .t. Isothecium cardoti kindb. Plagiothecium undulatum ( l. ) sch. Hypnum calliehrouin bkid. H ylocomium squarrosum ( l. ) sch. Hylocomium loreuni (l. ) sen. Hylocomium splendens (l. ) sen. POLYPODIACEiE. Filix fragilis (l. ) undew. Dryopteris spinulosa dilatata (hofp. ) UNDEW. Polystichum munitum (kaulf.) undew. Athyrium cyclosorum eupk. Struthiopteris spicant ( l. ) weiss. Adiantum pedatum l. Polypodium falcatuni kellogg. Pinus contorta loud. Picea sitchensis (bong. ) trautr. & mey. Tsuga mertensiana (bong.) carr. Tsuga heterophylla (raf. ) sargent. Tlmja plicata don. Chamsecyparis nootkatensis ( lamb. spach. Taxus brevifolia nutt. SCHEUCHZERIACEJE. Triglocbin maritima l. Triglochin palustris l. Agrostis exarata trin. Deschampsea csespitosa (l.) beauv. Dactylis glomerata l. i'dack.k — continued. Puccinellia distans ( l. ) parl. Festuca rubra l. Bromus aleutensis trin. Elymus mollis trin. Elymus Bp. CYPERACE.E. Scirpus pauciflorus lightf. Carex sp. JUNCACEJE. Juncus balticus willd. Juncus bufonius l. Juncoides parviflorum (eiirh. ) covilli I.ILIACE.E. Fritillaria kamtschatcensis (l.) kkr. Unifolium bifolium (l. ) greene. Streptopus roseus michx. IRIDACE/E. Sisyrinchium littorale greene. ORCHIDACE.K. Habenaria hyperborea (l.) r. br. Peramiumnienziesii (lindl. ) morong. Corallorhiza mertensiana bong. SALICACE/K.- Salix scouleriana barratt. BETULACE2E. Abms sinuata (kei;el) rydb. Alnus oregona nutt. URTICACEJE. Urtica lvallii avats. POLYGONACEiE. Rumex sp. Polygonum sp. rilENoPODIACE.E. A triplex gmelini C. A. mey'er. Salicomia herbacea l. I'DHTrl.ACACE.E. Montia parvifolia (moc.) greene. Montia sibiriea ( l. ) iiowell. SEPT.. 1901 ] PLANTS. 15 J 'It ui is — Continued. A LSI X ACK.K. Cerastium sp. Sagina crassicaulis wats. Ammodenia peploides (l. | bi pr. Tissa marina 1 1.. ) britton. i; \xrxci l. ack.k. Caltha palustris i.. Ooptis asplenifolia salisb. Aquilegia Formosa Fischer. Anemone narcissiflora i.. Ranunculus occidentalia nutt. Ranunculus sp. HKASSIC.U K.K. Cochlearia oblongifolia dc. Brassica campestris i.. Cardamine angulata hook. Aral lis hirsuta scop. Arabia Bp. i i; \SSI1.ACK.K. Sedum roseum t l. i scop. Sedum spathulifolium hook. >AXIFKA<,ACEJE. Saxifraga mertensiana box*;. Saxifraga nutkana moc. Tiarella trifoliata l. Heuchera cylindrica dougl. Heuchera glabra willd. KIBACE.E. Ribes bracteosam dougl. Ribe.s lacustre (peks. ) poib, Ribes laxiflorum pubsh. rosacea. Lutkea pectinata (hook.) kuntze. Aruncua animus ( l. ) kar.st. Pyrus rivularis dougl. Crataegus brevispina dougl. Sorbus Bambucifolia (c. a s. ) eoem. Amelanchier alnifolia nutt. Rubus pedatus smith. Rubus gpectabilis pubsh. Rubus parviflorua nutt. Fragaria chiloenais (l. ) duchesne. Potentilla anserina l. Potentilla villosa pall. Geum ealthifdliuin menzies. Geuni sp. Sanguisorba sp. Rosa sp. V HI ACK.K. Lupinus nootkatensia donn. Lupinua nootkatensis unalaskensis wats. Trifolium involucratum wills. Lathyrua maritimua | l. ) bigel. Vicia gigantea hook. VIOLACEjE. Viola glabella nutt. (IXAlll! ACK.K. Epilobium glandulosum lehm. Epilobium bomemanni beichenb. Epilobium minutum lindl. AKAU ACK.K. Echinopanax horridtim (smith) ]>. & p. APIACE.dE. Washingtonia sp. Conioselinum gmelini coult. * bose Heracleum lanatum michx. C< IK X ACK.K. Cornua canadensis l. Cornusoceidentalis (tobb. age.) coville. PYROLACEJE. Moneses uniflora (l.) gray. vaccixiack.k. Vaccinium csespitosum michx. Vaccinium ovalifolium smith. Vaccinium parvifolium smith. Vaccinium uliginosum i.. KKICACK.K. Menziesia ferruginea smith. Cbamsecistus procumbens (l. i kuntze. Kalmia glauca ait. Phyllodoce glanduliflora ( hock. ) co- ville. Cassiope mertensiana ( bong, i don. Caasiope stelleriana dc. Gaultheria shallon pubsh. Arctostaphylos uva-ursi (l.) bpreng. PRIMUL \l K.K. Glaux maritima L. Dodecatheon viviparum greene. 16 NORTH AMERICAN FAUNA. [NO. 21. Plants — Continued. NEPETACEJE. Prunella vulgaris i.. Stachys sp. SCROPHULARIACE.2E. Collinsia parviflora dougi/. Mimulus langsdoffii don. Veronica americana schwein. Castilleja pallida kuntii. Castilleja parviflora bono. Pedicularis lanata willd. Pedicularis pedicellata bunge. PLANTAGINACE/E. Plantago maritima l. Plantago sp. RUBIAC'E.K. Galium aparine l. CAPRIFOLIACEJE. Loniceta involucrata (rich.) hanks. VIBURNACEJS. Viburnum pauciflorum pylaie. Symphoricarpos racemosus michx. Linnaea borealis l. CAMPANULACE^E. Campanula langsdorffiana fisch. CAKDUACE.K. Grindelia sp. Erigeron peregrinus (pursh. ) green Achillea borealis bong. Matricaria discoidea DC. Senecio resedifolius less. Carduua edulis (nutt. ) gkeene. FAUNA. The mammal fauna of the Queen Charlottes, in view of the prox- imity of the group to the mainland, may be said to be rather meager. There are only 11 species of indigenous land mammals, and 1 of these are bats. The genera represented are Peromyscus, Ursus, Lutra, Mmtela, Putorius, Sorex, Lasionycteris, and Ifyotis. Characteristic genera of the adjacent mainland, such as Odocoileus, Lutreola, Sciurtts, JE/votomys, and Microtus, are entirely wanting. The absence of these genera, which are common on the mainland and well adapted to all the conditions of the islands, indicates that the water barrier between the islands and the mainland is an effective one. The channel is cer- tainly too wide to swim, and small mammals could not easily be carried on driftwood, as the strong tides would be apt to sweep everything of that nature out at either end of Hecate Strait rather than across it.1 The presence of the marten, an animal which is terrestrial and arboreal, and the absence of the mink, which is semiaquatic, and the deer, which swims considerable distances, might be considered evidence that the islands must have been peopled with animals at a time of complete connection with the mainland. Rut if at that time the mainland fauna was approximately the same as at present, it is difficult to explain the present absence of the deer and mink. Whatever the case may have been, it is certain that the mammals have been isolated for a 1 This means of communication is possible, however, as is shown by the fact that the dead bodies of deer have been washed up on the shores of the islands (see p. 25), but the course of the tides is such that they could not have been carried directly across the strait. It is therefore probable that the journey would be too long and perilous for any living animal to survive. sept., ltwi.] FAUNA. 17 sufficient length of time to become differentiated into well-marked insular forms. All the land mammals that have been studied have proved distinct from the species of the same genera found on the mainland.1 Moreover, the larger, less mutable forms ( Ursus, Mustela, and Putorvus), which arc also the ones least likely to have been intro- duced through accident or human agency, are the most strongly charac- terized, thus affording additional evidence of isolation of rather long standing. Still more interesting than the general peculiarity of the entire mammal fauna as contrasted with that of the mainland, is the individ- uality shown among closely adjacent islands when contrasted with one another. A mouse (Peromyscus pr< vostensis) and a shrew {Sorex I. j>r< vosh nsis) arc peculiar to Prevost Island. The island is quite small, possibly 50 square miles in area, yet these mammals are found in great abundance, and do not occur on other islands which lie less than a mile away. The channels between these islands are said by Dawson to he accidental, but at least they can not have been formed very recently or this distribution of animals would not be found. Pero- myscus prt rust, nsis, though from the southernmost island of the group, is most closely related to /'. sitkensis which has been found only on islands some distance north of the Queen Charlottes, while 1\ heeni, of Graham and Moresby islands is not related to northern forms, but is scarcely distinguishable from the comparatively remote P. akeleyi of the Olympic Peninsula. Washington. The small mammals of the northwest coast are still so imperfectly known and they are all so interrelated, that it may be unprofitable to speculate at present on the derivation of various insular forms, and it is possible that the animals are so plastic that their present characters can not safely be taken as indicative of their true relationships. Birds are comparatively abundant. The species are much the same as those found throughout the whole district from Puget Sound to Glacier Pay. but individuals seem to be more numerous than on the mainland. The land birds frequent the thickets of salmon berry, alder, and willow bushes found bordering sandy beeches or small streams (see PI. II. fig. 2). In these restricted areas certain species are very common. Many such environments are afforded in Cumshewa and Skidegate inlets, and it was there that most of our collecting was done. The steamer which carried us to the islands dropped anchor in Cumshewa Inlet, about 50 yards from the shore at 5 o'clock in the morning of June 13, and through the little port of my stateroom sounded such an avian chorus as I have heard nowhere else on the northwest coast. The greatest volume of song came from song spar- lThe land otter, Lutra, is the only one known to occur ou the islands of which no specimens have been critically examined. 3505- No. 21—01— 18 NOKTH AMEBIC AN FAUNA. [no. 21. rows and fox sparrows, but the rich tones of the russet-backed thrushes contributed much to strength and quality; winter wrens bubbled and spluttered on all sides, the clear notes of a robin were heard now and then, and from farther back in the forest came the weird call of the varied thrush, while all the time several huge ravens carried on a lively conversation with one another. The deep forest is for the most part dark and quiet, and save for an occasional creeper or winter wren con- tains no birds. Water birds are reported to breed in large colonies on islets near the west coast of Moresby Island. We were unable to visit these, but observed small rookeries at the mouth of Gumshewa Inlet and in Houston Stewart Channel. Very few of the land birds are definitely separable from those of the Sitkan district, but the tendency to dark colors and heavy markings is extreme. Two forms are peculiar to the islands, a jay {Cyanocttta stelleri carlottce) and a woodpecker (Dryobatespicoideus), both of which belong to groups not known to be as variable as others, such asJunco, Melospisa, and Passerella, whose Queen Charlotte representatives are not strongly characterized. Ninety-six species are known to occur on the islands, 62 of which are found in the breeding season. For notes on the occurrence of many of the species, I am greatly indebted to Rev. J. H. Keen, whose observations at Massett covered a period of eight years. Besides the mammals and birds but one land vertebrate has been found on the islands, a toad (Bufo halophilus columbiensis) , which is common on the adjacent mainland. We noted no strictly fresh-water fishes, but trout are reported to inhabit some of the streams. A col- lection of several hundred littoral species of fish was made, chiefly by Mr. Edmund Heller, who has determined the species and found none of them peculiar. Conditions about the islands are exceedingly favor- able for marine invertebrate life and it flourishes in profusion, but no attempt at collecting such forms was made. The vertebrate fauna of the islands, as a whole, is very similar to that of the adjacent mainland, but is nevertheless peculiar in many respects. The vertebrates known to occur on or about the islands are as follows: MAMMALS. 1. Globicepbala scammoni. 11. Mustela nesophila. 2. Balaenoptera velifera. 12. Emnetopias stelleri. 3. Lagenorhynchus obliquidens. 13. Otoes alascanus. 4. Phocsena phocsna. 14. Phoca largha. 5. Oclocoileus columbianus sitkensis 15. Sorex longicauda prevostensis. (introduced). 16. Sorex longicauda elassodon. 6. Peromyscus keeni. 17. Lasionycteris noctivagans. 7. Peromyscus prevostensis. 18. Myotis yumanensis saturatus. 8. Ursus carlottse. 19. Myotis subulatus keeni. 9. Lutra canadensis subsp. 20. Myotis californicus caurinus. 10. Putorius haidarurn. SEP! '., 1901.] FAUNA. BIKDS. 1. ( iavia imber.1 50. 2. Gavia pacifica.1 51. 3. ( iavia luniiiR'.1 52. 4. Lunda cirrhata.1 53. 5. Fratercula corniculata. r>4. . (i. Synthliboramphus antiquus. 55. i . Brachyramphus marmoratus.1 56. s. Cepphus columba.' 57. 9. I'ria troile californica.1 58. 10. Rissa tridactyla pollicaris. 59. 11. Larus glaucescens.1 60. 12. Larus brachyrhynchus. 61. 13. Larus Philadelphia. 62. 14. Pufnnus griseus. 63. 15. Purtinus tenuirostris. 64. 16. ( toeanodroma furcata. 65. 17. Phalacrocorax pelagicus.1 66. 18. Merganser americanus.1 (17. 19. Merganser serrator.1 68. 20. Anas boschas.1 69. 21. Mareca americana. 70. 22. Nettion carolinensis. 71. 23. Dafila acuta. 72. 24. Aythya Bp.1 25. ( llangula clangula americana. 73. 26. Charitonetta albeola. 74. 27. Harelda hyernalis. 75. 28. Histrionicus histrionicus. 76. 29. Oidemia deglandi.1 77. 30. Oidemia perspicillata. ' 78. 31. Anser albifrons gambeli. 79. 32. Branta canadensis occidentalis.1 80. 33. Olor sp. 81. 34. Ardea herodias fannini. 82. 35. Porzana Carolina. 83. 36. < rallinago delicata. 84. 37. Tringa acuminata. 85. 38. Ereunetes occidentalis. 86. 39. Calidris arenaria. 87. 40. Totanus melanoleucus. 88. 41. Actitis macularia. 89. 42. Sqoatarola sn>alis calurus.1 Haliseetus Leucocephalus alascanus.1 Falco peregrinus pealei.1 Kalci i columbarius suckleyi.1 PandUon baliaetus carohnensis.1 Megascops asio kennicotti.1 Nyctala acadica scotsea.1 Nyctea nyctea. t Jeryle alcyon.1 Dryobates picoideus.1 Sphyrapicus ruber flaviventris.1 t lolaptes cafer Baturatior.1 Selasphorus rufus.1 Empidonax difficilis.1 ( !yan< »citta stelleri' carl< >ttse. ' ( Jorvue corax principalis.1 ( Jorvus caurinus.1 Pinicola enucleator flammula.1 Loxia curvirostra minor.1 Loxia leucoptera.1 Spinas pinus.1 Passerina nivalis. Ammodramus sandwirhensis alaudi- nus. Calcarius lapponicus alascensis. Zonotrichia coronata. Junco hyernalis oregonus.1 Melospiza melodia rufina.1 Passei'ella iliaca townsendi.1 Hirundo erythrogastra. ! Tachycineta bicolor.1 Tachycineta thalassina.1 Helminthophila celata lutescens.1 Dendroica estiva rubiginosa.1 Dendroica t< wnsendi.1 Wilsonia pusilla pileolata.1 Anthus pensilvanicus.1 Cinclus mcxicanus.1 Anorthura hiemalis pacifica.1 Certhia familiaris occidentalis.1 Sitta canadensis.1 Parus rufescens.1 Regulus satrapa olivaceus.1 Regulus calendula grinnelli.1 Hylocichla ustulata.1 Hyiocichla aonalaschkse verecunda.1 Merula migratoria propinqua.1 Hesperocichla naevia.1 BATRACHIAN. Bufo halophilus columbiensis. 1 Known, or supposed, to breed. 20 NORTH AMERICAN FAUNA. [NO. 21. FISHES.1 Squalus sucklii. Hydrolagus colliei. CI u pea pallasi. Oncorhynchus sp. Salmo sp. Gasterosteus cataphractus. Aulorhynchus flavidus. Cymatogaster aggregates. Sebastodes melanops. Sebastodes caurinus. Hexagrammus stelleri. Ophiodon elongatus. Artedius lateralis. Hemilepidotus hemilepidotus. Euophrys bison. Leptocottus maculosus. Oligocottus maculosus. Blenicottua globiceps. Ascelicbthya rbodorus. Caularchus maendricus. Apodichthys flavidus. Pholis ornatus. Anoplarchus atropurpureus. Xiphistes chirus. Xiphidion umcosum. Xiphidion rupestre. Theragra fucensis. Gadus macrocephalus. Hippoglossus hippoglossus. Psettichthys melano&tictus. Limander asper. LIFE ZONES. The zones of the Queen Charlottes are the Canadian and the Hud- sonian. The greater part of the group, at least all that which lies below an altitude of 2,500 feet, is Canadian, and the remainder above that elevation is Hudsonian. This zonal definition is determined almost entirely b}^ the plant life. The insular occurrence of various species of mammals, and to some extent of birds, may be due to acci- dent and unnatural agency; therefore the absence of certain forms obviously can not be considered significant in correlating island and continental zones. Moderate insular isolation restricts vegetation much less than animal life; so that in determining the faunal position of the Queen Charlottes the fact that practically all the trees and shrubs are those usually found in the Canadian and Hudsonian zones is of much more importance than that no mammals occur other than those of the Canadian zone. The latitude of the southern part of the group is about the same as that of the mainland where the Transition zone merges into the Canadian, and since the average temperature on the islands may be assumed to be slightly warmer than on the mainland, some Transition intrusions might be expected. These are comparatively few, however, and may safely be disregarded in a general definition of the zones of the group. The characteristic Transition tree, Pseudotsuga ?nucronata, which reaches its northern limit on the mainland in about this latitude, has not been found on the Queen Charlottes. None of the coniferous trees, with the possible exception of Taosus brevifolia, can be consid- ered unequivocal Transition species. Picea sitchensis^ Tsuga het&r- ophylla, Thuja plicata, and Finns contorta all range throughout the Sitkan district, which is Canadian, and, roughly speaking, extends 1 Species determined by Mr. Edmund Heller. North Ameiican Fauna, No. 21 PLATE III. Fig. 1.— Mouth of Stream, Cumshewa Inlet. Fig. 2.— Shore of Bare Island. Skidegate Inlet. SEPT.. 1901.] life zones. 21 from River Inlet, British Columbia, to Cross Sound, Alaska. Among the small shrubs of the Queen Charlottes that also range through this Canadian district may be mentioned Alnus sinuata, Ribes laxiflorum, Ribes bracteosum, Vacdnium ovalifolium, Mendesia ferruginea. Viburnum paudflorum, Arctostaphylos uva-ursi, and Echi/rwpa/nax horridum. The Hudsonian zone occupies those parts of the islands which are above an elevation of about 2,500 feet.1 The mountains which exceed this height are distributed in four principal groups, namely, those of the southwestern part of Graham Island, those of the head of Cumshewa Inlet, Moresby Island, those of the central part of Louise Island, and those lying along Darwin Sound, and known as the San Cristoval Range8 (see fig. 1). The only part of the Hudsonian zone actually traversed by us is that on the prin- cipal peak at the head of West Arm of Cum- shewa Inlet. There the timber on the lower slopes of the mountain was found to be very heavy and of much the same character as that of the shore, the principal difference being the absence of Taocus, which seems to be con- fined to the immediate border of the inlets. At an altitude of about 2,000 feet a more decided difference in the flora began to be noticeable. This change progresses rapidly. The giant cedar {Thuja) disappears entirely, and the spruce (Picea) and western hemlock (Tsuga hett rophylla) are much reduced in num- bers. The alpine hemlock (Tsuga mertensiana) and the yellow cedar (Chamceeyparis nooika- tensis) take a place in the forest and soon be- come well established as the predominating trees, the hemlock being slightly in the ascendancy. Many smaller Hudsonian plants appear simultaneously with these trees and continue with them nearly or quite to the summit. Among these may be men- tioned Cassiope steUeriana, Cassiope mertensiana, Phyllodoce glanduli- -tf<>r9-170, Aug. 10, 1898. Putorius haidarum sp. nov. Type from Massett, Graham Island. 1898. Dobsey, George A. A Cruise Among Haida and Tlingit Villages about Dix- on's Entrance. chooner for some time while we were sailing in Hecate Strait July 7. Phocsena phocaena Linn. Common Porpoise. A school of porpoises numbering at least 100 individuals was seen in Hecate Strait July 7; others were frequently seen in the strait. A skull, evidently of this species, was sent from the islands by Dr. Robert Brown in 1868 to the British Museum.1 [Odocoileus columbianus sitkensis Merriam. Sitka Deer. Deer have been introduced on the islands, but have not yet thoroughly established themselves there. I was told by Rev. Mr. Collinson, who was formerly in charge of the missionary work at Masse tt, that he was instrumental in the introduction of a few deer on Graham Island some years ago. Nine individuals from the vicinity of Port Simpson were liberated at Massett, and within a year .signs of them were seen near Skidegate. Mr. Tennant, of Skidegate, states that a deer was killed by Indians about two years ago on Moresby Island, near Skidegate, but that since that time no deer or tracks have been reported. According to Rev. Mr. Collinson the dead bodies of deer from the mainland are occasionally found washed up on the beaches of the islands. ] 'Ann. and Mag. Nat. Hist., hit. 4, VII, 64, 1871. This specimen is probably the one referred to by Flower as having come from Vancouver Island. (List Cetacea Brit. Mas., p. 16, 1885.) 25 26 NORTH AMERICAN FAUNA. [no.' 21. [Rangifer dawsoni1 ( = Rangifer areticus). The description of a caribou from the Queen Charlotte Islands was, to say the least, somewhat unexpected, so in visiting the islands I was particularly interested in obtaining information in regard to it. I could find no evidence, however, that native caribou ever existed on any of the islands. Rev. Mr. Keen, who lived at Massett for eight years, and who was specially interested in matters pertaining to natu- ral history, says that from his own experience and that of the oldest Indian hunters, whom he questioned closely, he is decidedly of the opinion that no caribou are to be found in any part of the islands. Rev. Mr. Collinson, who was one of the earliest missionaries at Mas- sett, has the same belief as Mr. Keen, though he did not express such definite conclusions. Besides the missionaries I also interviewed a Mr. Stevens, who has kept the general store at Massett for the past nine years, and obtained from him the same opinion. All these per- sons are familiar with the story of ' Mackenzie's caribou,1 which is doubtless the cause of the mistaken idea that a peculiar species is native to Graham Island. According to this story, which was told me independently and without essential variation by Messrs. Keen, Collinson, and Stevens, some fifteen or twenty years ago Mr. Alex- ander Mackenzie, a trader for the Hudson Bay Company at Mas- sett, conceived the idea that in such a favorable place as Graham Island there must be deer and caribou, though the Indians had never killed them or even seen their tracks. Accordingly he offered a reward to anyone who should kill one or bring him evidence of having done so. The offer remained open for a long time, but finally a claim- ant appeared with fragments of a caribou, including the head. This imperfect specimen passed through several hands and finally found its way to the Provincial Museum in Victoria, where it was unearthed to receive the name Mangifer dawsoni. If the reward was incident to such a statement the Indian who brought this specimen to Mackenzie no doubt solemnly averred that he killed it on Graham Island. An Indian's testimony in a case of this kind, however, would not hang very heavy in the balance, even against a small amount of circum- stantial evidence. Mr. Mackenzie is not now living, but the testi- mony of Mr, W. Charles, who received the caribou head from him, indicates that for its absolute origin we have the word of the Indians only. In response to a letter to Mr. Charles I received an answer from Mr. J. R, Anderson, deputy minister of agriculture at Victoria, from which the following is extracted: Some time ago Mr. W. Charles, who is an invalid, handed me your letter of the 10th January last regarding the occurrence of caribou on Queen Charlotte Islands. Mr. Charles asked me to communicate with you and say that the head referred to, and which had deformed antlers, undoubtedly came from Queen Charlotte Islands, hav- lSeton-Thompson, Ottawa Naturalist, XIII, 257-261, Feb., 1900. Sept., 1901.] MAMMALS. 27 ing been sent to him by the Hudson Bay Company agent there, and was equally that of a caribou. The animal, Mr. Charles has no reason to doubt, was actually killed by the Indians, and they being unacquainted with it, broughl the skull to Mackenzie, and reported more of the same kind in the interior of the island. From this it seems that till the information in regard to the Mackenzie specimen came from the Indians, and that no white man has given any direct first-hand testimony as to its absolute origin. At the instance of Mr. Anderson a brief request for information was inserted in the * Daily Colonist,' of Victoria, B. ('.. and several replies were received. One of these, from Mr. S. M. Harrison, of Massett, which is of considerable interest, was kindly forwarded to me. It is addressed to Mr. Anderson under date of April 30, 1901, and is as follows: Sir: I noticed a paragraph in the Colonist under the heading of "Who knows" re the existence of caribou on Queen Charlotte Islands, t have lived here twenty years, and know the account given is quite correct. I have made diligent inquiries amongst the Indians, and have gained the following information: (1) Three years ago an Indian named Shakwau saw a female caribou feeding near a lake up Virago Sound, but failed to kill it, although he fired twice. Yethgwonas, another Indian, was with him at the time This March a man named Stlinga with his two sons saw the tracks of a big herd near the headwaters of Malon River, near Virago Sound. (3) Men who were with the man who killed the two referred to in the Colonist are ready to show me the place where he killed them. This is near Lthum, up Virago Sound. (4) The Haidas refused to eat the flesh of the caribou and left their carcasses. Mr. Mackenzie then paid them to go and bring the meat in and kept it for his own use. (5) As the Indians are not interested in the killing of caribou, they refusing to eat the meat and there being no market for the antlers, etc., they consequently do not hunt them. They say they are afraid to go up the mountains and into much danger for no recompense, there being, according to their traditions, one-eyed mon- sters, hobgoblins, spirits, etc., to be met with on the mountains which they fre- quent. * * * This, though much more definite than any other report received, contains little which did not emanate from the Indians, and it is there- fore difficult to be certain that it contains any element of reliability. Surely men who believe in "one-eyed monsters, hobgoblins," etc., could easily indulge themselves with an imaginary caribou. How- ever. Mr. Harrison's statement that meat was brought to Mackenzie and used by him is much more worthy of consideration and might lead one to entertain a belief in the possibility that caribou were killed on Graham Island, but the probability that such was the case is still doubtful. If the type specimen of Rangifer dawsoni originally came from the mainland, as seems probable, instead of from Graham Island, it may either have been deliberately bartered for with the intention of obtaining a reward, or it may have been innocently brought to the islands to be used in the native arts. More or iess communication has always existed 28 NORTH AMERICAN FAUNA. [no.21. between the islands and the coast, and between the coast and the inte- rior, both in times of peace and during hostilities. Hence either explanation is probable. The fact that the Haidas used caribou horn for making implements and ornaments is not particularly pertinent to the question, since articles made of mountain goat and mountain sheep horns are even more commonly used by them. If they could obtain horns of elk, deer, mountain goats, and mountain sheep from the mainland, which they undoubtedly did, it certainly must have been just as easy to get the useful parts of the caribou. The Haidas, it is true, are better fishermen than hunters; but this is probably more on account of lack of game than otherwise, for they are physically and mentally a very superior tribe. In view of the conflicting nature of the reports it does not seem safe or advisable to recognize Rangifer da/wsoni as a distinct form, particularly as the specimen itself furnishes no indication that it rep- resents a peculiar species, even granting that it came from the islands, for it does not essentially differ from specimens of Rangifer arcticus, the only difference claimed being that of darker color, and this is merely an opinion, as the piece of skin was long since destroyed.] [Mus musculus Linn. House Mouse. Said to have been abundant at Clew and Skidegate, but recent liberal importations of cats seem to have cleared it out.] [Mus norvegicus Erxl. Norway Rat. A few have escaped from ships occasionally, but in most cases each individual was detected and hunted down immediately, so the pest has not yet obtained a foothold on the islands.] Peromyscus keeni (Rhoads). Keen's Mouse. This is the common white-footed mouse of Graham and Moresby islands and the small detached islets near them. We found it abun- dant about Cumshewa Inlet from sea level to timberline, and equally common in Skidegate Inlet. It infests the few inhabited houses in company with shrews, and elsewhere is found indiscriminately all over the islands. Most of our specimens were caught near the shore in rockj' or relatively dry places in the underbrush, but a few were taken high up in the mountains. In all, 98 specimens were secured, as fol- lows: Cumshewa Inlet, 40; Skidegate Inlet, 50; near Rose Harbor, south end Moresby Island, 8. I have not recently examined the type of the species which was taken at Massett by Rev. J. H. Keen, but several alcoholic specimens from that locality are at present in the Biological Survey collection. I have compared four good skulls of these Massett specimens with a series from Skidegate, at the other end of Graham Island, and found them identical. Those from Moresby Island average a trifle larger than those from Skidegate, but the 1901.] MAMMALS. 29 difference is extremely slight. The mainland species most similar to P. I,-,, in seems to be P. akeleyij from the Olympic Mountains, Wash- ington. Specimens from various parts of the Olympic Peninsula (Neah Bay. Lake Cushman, Queniult Lake, etc.) do not differ from Iceeni in color, or appreciably in cranial characters: The only dis- tinctions of consequence are the rather smaller ears and shorter tail of keeni. Ten specimens from Neah Bay. Washington, assumed to be /'. akeU y'. average as follows: Total length, 203.8; tail vertebrae, 1 14.7; hind foot. 23.3. Twenty males of keeni from Skidegate average L97; L02; 22.7; fifteen females of keeni from Skidegate average L99.8; L03.4; 22 A. Peromyscus prevostensis sp. now Prevost Island Mouse. Typeirom Prevost Island, Queen Charlotte Islands, British Columbia, 9 adult (old), X,.. 100818, l". S. Nat. Museum, Biological Survey collection. Collected July 5, 1900, by W. II. Osgood and E. Heller. Original No. 1135. ( Characters. Similar to P< romyscus macrorhinus, but larger and with shorter tail. Similar to Peromyscus sitkensis, hut with slightly shorter tail and cranial differences. Color'. Similar to P. sitkensis and 1\ macrorhinus, hut slightly darker. Lpperparts with dusky concentration on middle of hack, forming a wide, ill-defined dorsal stripe; space around and in front of eyes black; ears dusky, with faint pale edgings; under parts grayish- white, occasionally with a faint narrow stripe of pinkish buff down middle of breast; hind feet generally somewhat dusky; tail sharply bicolor. Skull. Similar to that of Peromyscus sitkensis, but slightly heavier; nasals decidedly shorter and not so distinctly attenuate posteriorly; posterior palatine foramina nearly or fully twice as long as in sitkensis. Measurements. — Average of forty -seven adults: Total length, 217; tail vertebrae, 104; hind foot, 26. (Average of ten adult topotypes of 1\ sitkensis: 224; 113.6; 26.5. Average of two adult topotypes of P. macrorhinus: 231; 128; 26.) Average of five skulls of pi-evostensis (adult males): Basilar length of Hensel, 23; zygomatic breadth, 15.2; nasals. 11.5. (Average of five skulls of sitkensis (adult males): 23.-A; L5.2; 12.7.) Remarks. — This large mouse is very abundant on Prevost Island, where forty-seven fine adults were easily trapped in the three nights of our stay. They were taken in the dark woods along the shore of the island, under old logs or roots and in damp, mossy places. They are stoutly built, and individuals often made their escape from the ordinary 'out o' sight' traps by beating about until they freed them- selves. Occasionally a sprung trap and a dead mouse would be found lying on the ground side by side or a foot or more apart. P. jjrerost- 1 Elliot, Field Columbian Mas. Zool. Ser., I, 220-227, Feb., L899. 30 NORTH AMERICAN FAUNA. (NO. 21. ensis appears to be entirely confined to Prevost Island, as it was taken nowhere else in the Queen Charlotte group. Exactly opposite Prevost Island, on the south end of Moresby Island, P. keen I pnly was caught, as was the ease elsewhere on Moresby Island. The dis- tance between the two islands at this point is less than a mile, but it is probable that the strong tides which sweep through the channel would carry logs or any floating debris out to sea rather than from shore to shore, and thus prevent small mammals from being transferred. P. prevostensis is so different from P. keeni that even if the islands were more closely connected it would be improbable that either species was derived from the other. Moreover, it is also questionable whether either is the immediate descendant of P. macrorhinus, the form of the adjacent mainland, as both are more similar to forms found at a greater distance. Notwithstanding its geographic position, P. pre- roxfensis is related neither to the mainland species of the same lati- tude nor to any of those farther south, but to 1\ sitkensis, which is found much farther north. In fact, considering its distribution, its resemblance to sitkensis is remarkable; it is characterized onty by a combination of slight peculiarities. It seems best, however, to treat it as a full species until the rather difficult subject of the distribution of the section of the genus to which it belongs is more clearlv understood. Ursus (Euarctos) carlottae sp. nov. Queen Charlotte Black Bear. Type from Massett, Graham Island, Queen Charlotte Islands, British Columbia. Skull only, No. S7620, U. S. National Museum, Biological .Survey collection, Nov., 1896, J. H. Keen. Characters. — Size slightly greater than Ursus americamus ; skull more elongate; rostrum relatively more slender; cranium less arched; teeth larger and heavier, particularly last molars; last upper molar with posterior 'heel' quite elongate. (See Plate IV.) Measurements. — The following table gives measurements of 6 full- grown old skulls of U. americanus from western North America and one adult (the type) and ti immature skulls of carlottce. The meas- urements of the type of carlottce are the onty ones fairly comparable with those given of americanus, but those of the young skulls are introduced to show the relatively large size of the last upper molar. No. 87618, for example, is the skull of a mere cub, yet its last upper molar is much larger than any of those of the full-grown specimens of nun rlcanus. North American Fauna, No. 21. Plate IV. SEPT., 1901.] MAMMALS. 31 LTRSUS CARLOTT I Num- ber. Locality. 87620 Masse tt, British Columbia. 78066 do 87619 do 75062 do 87617 do 87621 do 87618 do K - s ?. no 03 1 . >."> Oh n 82 L27 19:; in; 111 162 97 70 L39 101 146 88 76 130 100 137 81 66 i:;i 99 137 86 70 127 96 132 78 59 L15 85 119 7:; 55 "H fl o ° i) S bo« § :•- a 13 ^ a) a » *« ~ •_ 5 3"Eg ■C3.g.3 •4-, O S0 •_ 30 :;i 31.6 30 27 URSUS AMERICANTJS. 72332 31277 Shuswap, British Columbia . . . Jasper House, Alberta Salmon River Mountains, Idaho. Cook lnlii, Alaska smart Lake, British Co- lumbia. — do I ? & ad , old 260 111 116 168 91 78 81 i ad. 260 110 120 186 101 65 76 ? ad. 255 139 114 181 94 67 77 '.' ad. 254 138 116 172 109 73 81 .' ad. 260 142 118 1S7 113 78 77 ? cf ad. 266 143 123 186 101 75 77 26 25 25. 5 27 26. 5 Remarks. — Seven perfect skulls of the Queen Charlotte black bear are in the Biological Survey collection, and although most of them are those of immature animals there is no difficulty in distinguishing any of them from mainland specimens. In comparisons specimens from western British Columbia and Alaska have been used to repre- sent americanus, which, as far as known at present, ranges from the Atlantic to the Pacific. No specimens are available from the west side of the coast mountains on the mainland opposite the Queen Charlottes, but a few from the region immediately north (Cook Inlet and White Pass) and south (Olympic Peninsula) have been examined, and none of them show any approach to carlottce. The skull of Xfrsus hif.nl '//.v i> equal in size to that of carlottce, but the teeth, though as long as in earlottce, are much wider and heavier. In carlottce the brain case is fuller, the arch of the cranium much greater, and the interor- bital region wider. At the suggestion of G. S. Miller, jr., the skull of I . carlottce was compared with that of Ursus procerus, the fossil species from Ohio, and more or less superficial resemblance between them was found, though, of course, detailed diagnostic characters are numerous. Both agree in general form of skull, particularly in the elongate rostrum in contrast with that of U. arnericanus. The denti- tion of procerus, however, is much heavier and otherwise different, while other characters arc abundant, indicating that the resemblance 32 NORTH AMERICAN FAUNA. [no. 21. to carlottm is merely coincidental and not indicative of close relation- ship. No skins of carlottce have been examined, but they are said to be gloss}7 black at all times, the cinnamon form being- absolutely unknown on the islands. The skulls on which the foregoing descrip- tion is based were secured from the natives at Massett, in 1895 and 1896, by Rev. J. H. Keen, to whom we are indebted for so many other specimens from the Queen Charlottes. I saw signs of bear only on the mountain at the head of West Arm of Cumshewa Inlet, and there the indications were not fresh enough to raise any hopes of securing a specimen. The Haidas hunt bears to some extent, and also secure them in dead-fall traps. 1 noticed several of these traps near the head of Cumshewa Inlet, but they had not been used for some time. Mr. Tennant, of Skidegate, secures from the Indians 10 to 30 skins annually. Lutra canadensis subsp. ( Land Otter. Otters are rather rare on the Queen Charlottes, though perhaps no more so than on the mainland. Mr. Tennant, or Skidegate, says that his annual receipts from the Indians have seldom included more than a dozen otter skins. This is the only mammal known to occur on the islands of which I have not examined specimens. Putorius haidarum Preble.1 Haida Weasel. Three specimens of this weasel were caught about Cumshewa Inlet and one at Skidegate. Traps baited with tish and set along the shore in the rank grass or in the rocks were the most successful. The type of this species is in winter pelage with slight traces of the summer coat, indicating, as Mr. Preble noted, a probable color difference from Putorius kadiacensis. This surmise is amply borne out by our spec- imens in full summer pelage. They are much darker than kadia- censis, displaying very nearl}r the rich chocolate of P. streatori, and having much less of the lemon-yellowish wash on the underparts than kadiacensis. The color of the upperparts does not encroach on that of the belly, and the black on the tail is extensive, occupying nearly half its length; in these respects it is more like kadiacensis than streatori. The white of the underparts extends on the under side of the tail for about one-third of its length. The fore feet are entirely white; the toes and one-third of the upper side of the hind feet are white. Its skull is quite distinct from that of any other of the cicognan i group. In general terms, it is fat, short, and stocky. The most conspicuous point of difference from kadiacensis, cicognani, or strea- tori is the very broad rostrum and interorbital region. Young spec- imens entirely free from parasites show this to good advantage, and adult skulls wdiich are infested are so much wider than those of 'Proc. Biol. Soc. Wash., XII, 169-170, Aug. 10, 1898. sept.. 1901. 1 M A M SI A LS. 33 kadiacmsis in a similar condition that they are easily distinguishable. The flesh measurements of a fully adult male are as follows: Total length, 283; tail vertebrae 70; hind foot 39. Adult female: 252; 63; 31. Skull of adult male: Basal length 39; mastoid breadth L9.2; width across postorbital processes 14.5; palatal length 17.5; length of audita! buUse 13. Skull of adult female: 35.5; 17.5; L2; L6.5; 11.6. Mustela nesophila sp. now Queen Charlotte Marten. Type from Massett, Graham Island. Queen Charlotte Islands, British Columbia. Skull only. Male (?), No. 78066, U. S. National Museum, Biological Survey Col- lection. J. II. Keen. ( Characters. — Similarto Musk la caurina, but larger; rostrum shorter and heavier; dentition heavier; premolars larger and more crowded. Last upper molar similarto that of Mustela americana, internal length being more nearly equal to external length than in caurina. (See PI. V.) Measurements. — Type: Basilar length of Hensel 73; palatal length 39; postpalatal length 34; zygomatic breadth 45.5; width across post- orbital processes 23: interorbital constriction 19. TopotypeNo. 76429, female: Basilar length of Hensel 69; palatal length 37; postpalatal Length 32; width across postorbital processes 23; interorbital constric- tion 15. (Mustela caurina No. 87075, female adult, Port Moody, British Columbia: Basilar length of Hensel 64; palatal length 33; post- palatal length 31; zygomatic breadth 41; width across postorbital proc- esses 2d; interorbital constriction 15.) Remarks. — This form is represented by two skulls which Mr. Keen secured several years ago from natives at Massett. These are not sexed. but the difference in size and other slight characters make it evident that one is a young male and the other an adult female. The peculiarities shown by these skulls are so marked that there can be no doubt that they represent an insular species. In the Biological Survey series of nearly 500 skulls of Mustela america/na and its allies I have been able to find no others showing the characters of these individuals from the Queen Charlotte Islands. The molar teeth of nesophila are heavier than in any other form of the group. The audita! bulla* are actually about the same size as in caurina and thus relatively smaller. The maxillary region between the infraor- bital foramen and the alveoli of the upper molars is wider and heavier than in caurma. The most obvious cranial character, however, and the one which distinguishes nesophila from all other members of the americana group is the thick, heavy rostrum. When skins are avail- able for comparison they also will doubtless show some slight differ- ences. The fur traders say the Queen Charlotte martens are always light colored and short haired and do not command as high a price as tho>e from the mainland. The Ilaidas trap more or less for martens 3505— No. 21—01 3 34 NORTH AMERICAN FAUNA. [no.21. every winter, but the animal is evidently not abundant, for Mr. Ten- nan t's annual receipts seldom exceed fort}7 skins. Latax lutris (Linn). Sea Otter. Formerly very abundant, but quite rare at present. A few are occasionally taken on the west coast of the islands or off the southern end of Prevost Island. During his cruise about the islands in 1787 Dixon bartered with the Haidas for 1,821 sea-otter skins. He secured a great many in Cloak Bay, on North Island, and describes his expe- rience as follows:1 A scene now commenced which absolutely beggars all description, and with which we were so overjoyed that we could scarcely believe the evidence of our senses. There were 10 canoes about the ship, which contained, as nearly as I could estimate, 120 people. Many of these brought most beautiful beaver cloaks, others excellent skins, and, in short, none came empty-handed, and the rapidity with which they sold them was a circumstance additionally pleasing. They fairly quarreled with each other about which should sell his cloak first, and some actually threw their furs on board if nobody was at hand to receive them. * * * In less than half an hour we purchased near 300 beaver skins, of an excellent quality. * * * That thou mayest form some idea of the cloaks we purchased here I shall just observe that they generally contain three good sea-otter skins, one of which is cut in two pieces. Afterwards they are neatly sewed together so as to form a square, and are loosely tied about the shoulders with small leather strings, fastened on each side. At another time, when near either Skidegate or Cumshewa Inlet, under date of July 29, he writes: Early in the afternooon we saw several canoes coming from shore, and by 3 o'clock we had no less than 18 alongside, containing more than 200 people, chiefly men. This was not only the greatest concourse of traders we had seen, but what rendered the circumstance additionally pleasing was the quantity of excellent furs they brought us, our trade now being equal, if not superior, to what we had met in Cloak Bay, both in the number of skins and the facility with which the natives traded. * * * Besides the large quantity of furs we got from this party (at least 350 skins) they brought several raccoon cloaks, each cloak consisting of 7 raccoon skins neatly sewed together.2 ? Eumetopias stelleri (Lesson). Steller Sea Lion. A sea lion, probably this species, is reported. It was not seen by us. 0toes:i alascanus4 (Jordan and Clark) Alaska Fur Seal. Fur seals still occasionally stop on or near the Queen Charlotte Islands. In former days the natives secured a great many in the region off the south end of the group. 1 A Voyage Bound the World in the King George and Queen Charlotte, pp. 199-234, London, 1789. 2 Since raccoons are not found on the Queen Charlotte Islands, these skins were probably from Vancouver Island where the animals are common. 3 Fide Palmer, Proc. Biol. Soc. Wash., XIV, 133-134, Aug. 9, 1901. 4 Report Fur Seal Invest. 1896-1897, Pt. 3, pp. 2-3, 1899. North American Fauna, No. 21. Plate V. Skulls of Mustela 'Natural Size). l. 2. Mustela caurina. 3, 4. Mustela nesophila. sept.. 1901.] MAMMALS. 35 Phoca largha Pallas. Pacific Harbor Seal. Harbor seals are quite common. They bobbed up very often near our canoe as we were paddling about the inlets. Sorex longicauda prevostensis subsp. now Prevosi Island Shrew. Type from Prevost Island, Queen Charlotte [slands, British Columbia. $ ad. No. 100618, U. S. Nat. Mus., Biological Survey Collection. Collected July 3, L900, by W. H. Osgood and E. Heller. Orig. No. L089. Characters. Similar to Sorex longicauda; tail, relative to head and body, shorter; dental characters distinctive. Color. —Very slightly darker than S. longicauda; contrast between upper and lower parts less; otherwise similar. Skull. Size large, equal to . 17, 1898. 'Macoun, Catalogue of Canadian Birds, Part I, p. 61, 1900. 40 NORTH AMERICAN FAUNA. [no. 21. Anas boschas Linn. Mallard. A flock of about a dozen mallards was seen frequently about the head of Cumshewa Inlet, June 17-26. Mareca americana (Gmelin). Baldpate. Reported by Mr. Keen. Nettion carolinensis (Gmelin). Green-winged Teal. Given in Mr. Keen's Massett notes. Not seen by us. Dafila acuta (Linn.). Pintail. Reported by Mr. Keen. Not seen by us. Aythya sp. ? Scaup ducks were several times seen flying- at a distance. Clangula clangula americana (Bonap.). American Golden-eye. Included in Mr. Keen's Massett list. Not seen by us. Charitonetta albeola (Linn.). BufhVhead. Seen at Massett (Keen). Harelda hyemalis (Linn.). Old-squaw. A roughly stuffed skin of an adult male old-squaw was seen at an Indian village in Cumshewa Inlet. Histrionicus histrionicus (Linn.). Harlequin Duck. Said to occur at Massett (Keen). Oidemia deglandi Bonap. White-winged Scoter. Frequently seen in Cumshewa and Skidegate inlets. Oidemia perspicillata (Linn.). Surf Scoter. Common. Anser albifrons gambeli (Hartl.). American White-fronted Goose. Reported by Mr. Keen. Branta canadensis occidentalis (Baird). White-cheeked Goose. Six or seven were seen crossing Cumshewa Inlet June 16. Olor sp. ? Swans are said to have been taken frequently. Ardea herodias fannini Chapman.1 Northwest Coast Heron. Often seen feeding at low tide on the beaches and mud flats of Skidegate and Cumshewa inlets. No specimens taken. Porzana Carolina (Linn.). Sora Rail. Included in Mr. Keen's manuscript list of birds seen at Massett. ^ul. Am. Mus. Nat, Hist. XIV, 87-90, Apr. 15, 1901. sept., looi.] BIRDS. 41 Gallinago delicata (Ord). Wilson Snipe. Reported by Mr. Keen. Tringa acuminata (Horsf.). Sharp-tailed Sandpiper. Taken at Magsett, Graham Island. December 27, L897, by Rev. J. II. Keen.1 Mr. Keen kindly forwarded me the specimen on which this record was made. I have compared it with others of the same species and found it typical. Ereunetes occidentalis Lawr. Western Sandpiper. A small sandpiper supposed to be this species was seen on a beach in Cumshewa Inlet. .Inne 28. Mr. Keen reports its occurrence al Massett. Calidris arenaria (Linn.). Sanderling. Reported by Air. Keen. Totanus melanoleucus (Gmelin). Greater Yellowlegs. Two were seen and one of them taken on tin1 beach at Skidegate, July 17. Actitis macularia (Linn.). Spotted Sandpiper. One seen at Skidegate in July. Squatarola squatarola (Linn.). Black-bellied Plover. Reported by Mr. Keen. Charadrius dominicus Miiller. American Golden Plover. Reported by Mr. Keen. Arenaria melanocephala (Vigors). Black Turnstone. A flock of 6 was seen near Lina Island, Skidegate Inlet, July 12. Haematopus bachmani And. Black Oystercatcher. Abundant. The shrill cries of the oystercatchers were1 heard about the inlets at all hours of day or night. Nearly every outlying rocky islet was occupied by a pair of 'sandpipers,' as they are locally called, and whenever a boat approached both birds would circle about it for some time, flying close to the water and crying shrilly. Dr. Robert Brown, writing in 1869, 2 says of this species: About Queen Charlotte Islands it is very plentiful. In March, 1866, while row- ing along the narrow sounds among these islands we often saw it. It would sit on the rocks until we could almost touch it; then, uttering a low whistling cry, it would dart off to another skerry, repeating the same maneuver over and over again. A nestfoundin Cumshewa Inlet, June 17, was merely a hollowr about 2 inches deep and almost perfectly round, scooped out of a weedy turf a few feet above high-water mark. The bottom of the hollow was covered with bits of broken stone, evidently placed there by the old 'Fannin, Preliminary Catalogue Provincial .Museum, Victoria, B. C, ]>. 28, L898 2 Ibis, IV, 424, 1868." 42 NORTH AMERICAN FAUNA. [no. 21. bird. A few feet from the nest a downy young bird was discovered squatting- in the weeds and gravel. It may be described as follows: Upperparts chiefly mottled olive-gray and black, the gray predomi- nating and the black distributed mainly in an ill-defined patch on the back of the head and two prominent parallel stripes that extend from the nape down the middle of the back to the rump; middle of back with a little buffy- tipped down; wings like back but with more buffy; flanks spotted with black; throat and breast slate gray, darker on sides of neck and indistinctly patched with paler on middle of throat; a conspic- uous white spot far back on middle of breast with a white line extend- ing forward from it on each side to the vicinity of the axillars; abdomen paler than In-east and lightly washed with buffy, also having to some degree the vermiculated appearance of the upperparts. Dendragapus obscurus fuliginosus Ridgw. Sooty Grouse. Several were heard booming about Cumshewa Inlet early in June. A pair of adult birds was taken at an altitude of about 3,000 feet in the mountains at the head of Cumshewa Inlet June 23. [Bonasa umbellus sabini (Dough). Oregon Ruffed Grouse. In Macoun's Catalogue of Canadian Birds the following statement occurs under Bonasa u. sabini: "One of the most abundant birds of the coast region of British Columbia including all the islands in the Gulf of Georgia, Vancouver Island, and Queen, Charlotte Islands." We did not meet this bird anywhere on the Queen Charlottes and it is not mentioned in Mr. Keen's manuscript list, so it seems probable that this statement is erroneous.] ? Lagopus sp. I was told by Mr. Tennant, of Skidegate, that eight 'white grouse ' were killed several years ago by a party of prospectors in the moun- tains on Graham Island a few miles from Anchor Cove, Skidegate Inlet. We found conditions favorable for ptarmigan near the summit of the mountains about the head of Cumshewa Inlet, but did not see any during our short stay there. Zenaidura macroura (Linn.). Mourning Dove. Seen at Massett by Mr. Keen; not observed by us. Aceipiter velox (Wils.). Sharp-shinned Hawk. Two small hawks supposed to be this species were seen at Skidegate July 12. Mr. Keen reports its occurrence at Massett. Aceipiter atricapillus striatulus Ridgw. Western Goshawk. Seen at Massett (Keen). Buteo borealis calurus (Cassin). Western Red-tailed Hawk. A solitary red tail was seen Hying near the head of Cumshewa Inlet June 22; no others were seen during our visit. sept.. 1901.] BIRDS. 43 Haliaeetus leucocephalus alascanus Towns. Northern Bald Eagle. Very common; often seen in parties of from '2 to LO individuals, the majority being birds of the year. They seem to feed largely on inollusks and crustaceans, which are very abundant. At one time I saw seven huge birds clumsily hopping over the rocks on the shore evidently looking for crabs. Eagles' nests were occasionally noted in the tops of tall, partially dead trees. Falco peregrinus pealei Ridgw. Peale Falcon. An immature female was taken July 2. It was shot as it circled around the small schooner in which we were drifting- in Hecate Strait a few miles oti' Scudder Point, Burnaby Island. Several others were seen near Prevost Island, and while we were in Houston Stewart Chan- nel a pair of them had daily altercations with a bald eagle in the tops of the trees on an islet near our anchorage. Falco colnmbarius suckleyi Ridgw. Black Merlin. A small dark hawk was indistinctly seen flitting out from the top of a tall spruce in Cumshewa Inlet June 14. Mr. Keen reports the black merlin from Massett. Pandion haliaetus earolinensis (Gmelin). American Osprey. Ospreys were not seen by us, but they are evidently common in some parts of the islands, as we heard numerous reports of them. Mr. Keen has noted their spring arrival at Massett as follows: 1894, May 13; 1806, April i>4; 1897, May 12; 1898, April 30. Megascops asio kennicotti (Elliot). Kennicott Screech Owl. Seen at Massett (Keen). Nyctala1 acadica scotaea2 subsp. nov. Northwest Saw-whet Owl. Type from Massett, Queen Charlotte Island, British Columbia, $ ad. No. 168171, TJ. S. National Museum, Biological Survey collection. Collected December 19, 1896, by J- H. Keen. ( Tiaracti rs. -Similar to W. acadica, but darker both above and below, dark markings everywhere heavier; flanks, legs, and feet more rufes- cent. Color. — Cpperparts, including head, neck, back, and upper tail- coverts, mummy brown;3 head with light stripes on forehead and thence down sides of neck; neck and interscapulars irregularly marked with white; wings slightly lighter than back; five outer primaries with two to four white spots on outer and inner webs; inner prima- 1 Dr. C. W. Richmond has proposed the name Cryptoglaux as a substitute for Nyctala on the ground that the latter is preoccupied by Nyctalus (cf. Auk, XVIII, 193, April, 1901 ) . This dispositi< m of Nyctala seems reasonable, but Cryptoglaux is not used here, since it has nut been adopted by the American Ornithologists' Union. '-' Scotfea=dark, dusky. 3 The color name.- used are from Ridgway's Nomenclature of Colors. 44 NORTH AMERICAN FAUNA. [no. 21. ries and secondaries with white spots on inner webs only: tail pale clove In-own, narrowly tipped with white, webs of rectrices crossed by three white bars; auriculars Isabella color streaked with dusky; fore- head and superciliary region white; orbital ring and outer feathers of lores sooty; chin, throat, and upper breast white, interrupted by a collar of mars brown; lower breast and abdomen white, heavily streaked with walnut brown; sides, flanks, legs, and feet clear ochraceous buff. Measurements.— -Type: Wing 85; tail 69; tarsus 26. Remarks. — This dark-colored form of the Acadian owl doubtless ranges throughout the humid Pacific coast region. Its rarity prob- ably accounts for its having been previously overlooked, for its char- acters are in general the same as those of the numerous other forms peculiar to the same region, which have long been recognized in nomen- clature. The only specimens that I have examined beside the type are several imperfect ones from Puget Sound, which are in the National Museum collection. These agree with the type in richness of color and extent of dark markings. The type was collected by Rev. J. H. Keen, who very generously presented it to the Biological Survey collection. A small owl, apparently this species, flew over our vessel at 11 o'clock on the night of July 4, while we were at anchor in Houston Stewart Channel. This was the only owl seen at any time during our visit to the islands. Nyctea nyctea (Linn.). Snow}T Owl. Mr. Tennant says he has killed large white owls at Skidegate. Mr. Keen reports them from Massett. Ceryle alcyon (Linn.). Belted Kingfisher. Generally found along the larger streams. One specimen was taken June 21. Dryobates picoideus sp. nov. Queen Charlotte Woodpecker. Type from Cumshewa Inlet, Moresby Island, Queen Charlotte Islands, British Columbia. J ad., No. 166816, U. S. Nat. Mus., Biological Survey collection. Collected June 15, 1900, by W. H. Osgood and E. Heller. Orig. No. 386. Characters. — Similar in general to Dryobates v. ha/rrisi; bill slightly smaller; middle of back barred and spotted with black; flanks streaked with black. Description. — Top of head, wings, and tail black; middle of back from nape to rump white, heavily barred, or spotted with black (streaked in immature specimens), primaries and secondaries more or less spotted with white; coverts usually with two to four elongate white spots; unclerparts smoky brownish, deepest on breast; flanks streaked, barred, or spotted with black and dusky (this sometimes extending forward on sides to axillars); three outer tail-feathers white, the innermost always partially black, the others sometimes more or less barred with black. SEPT.. 1901.] BIRDS. 15 Measuri rru nts. The Queen Charlotte specimens have rather smaller bills than harrisi, as appears from the following table: DRYOBATES PICOIDEUS. Num- ber. Sox. Locality. Wing. Tail. Expi ised culmen. Tarsus. 166816.. 166821 9 ad Queen Charlotte Islands, British Columbia (type). 126 1120 L25 124 121 122 94 96 '.ic< 95 98 99 29 28. 5 27.5 24 28. 5 29. •". 23 22 166820 •)■> 22 166817 23 ltil'.M'.' 21 5 DRYOBATES VILLOSUS H &.RRISI. 157586.. 157619.. 157598.. 117689.. 151483.. 132504.. fad. fad. fad. 9 ad. 9 ad. 9 ad. Neah Bay. Washington do do Departure Bay, British Columbia . Comox, British Columbia Seattle. Washington L23 97 34 126 99 36 129 102 35 123 85 30 123 95 31 121 97 32 23 21 23 21 22.5 Remarks. -Adult specimens of Dryobates picoideus arc easily dis- tinguishable from all other members of the villosus group by the black markings <>n the back. Immature birds of harrisi and of other mem- bers of the group occasionally have a few median or lateral streaks of black on the back, but never the definite barring, as in picoidms. There is also an occasional tendency in young- harrisi to show dusky on the flanks, but neither this nor the black in the back persists in the adult. This woodpecker is not abundant on the islands; during our stay we saw but six, all of which were secured. These consist of two adult females, two immature females, and two immature males. The only adult male examined was one brought to me Iry a boy at Skide- gate, which was not preserved, as it was very much mutilated, but its characters, particularly the barred back, were noted. Sphyrapicus ruber flaviventris (Vieill.). Northern Red-breasted Sap- sucker. Picas flaviventris Vieillot, Ois. Am. Sept., II, 67, 1807. Sphyrapicus vceritu ruber Grinnell, Condor, III, 12, Jan. 15, 1901. Common. Represented by ten specimens, which agree perfectly with birds from Vancouver Island and the mainland of British Columbia. This bird was discovered by Captain Cook in Nootka Sound on the west coast of Vancouver Island. In the narrative of his famous voyage he makes special mention of it and gives a very good description. Later, in 1807, Vieillot named it Picus flaviventris ', and distinguished it from Picas ruber of Gmelin as follows: 'Diftere principalement par la teinte du ventre qui est d'un jaune olivatre.' Colaptes cafer saturation (Ridgw.j. Northwestern Flicker. Apparently quite rare, as we saw none. A few unmistakable tail feathers were found, however, by Heller in Cumshewa Inlet. Mr. Keen reports flickers abundant near Massett. 46 NOKTH AMERICAN FAUNA. [no. 21. Selasphorus rufus (Gmelin). Rufous Hummingbird. Common. During the month of June hummers were often seen visiting the abundant blossoms of Lonicera involucrata. Mr. Keen has observed their arrival at Massett for six years as follows: 1891, April 6; 1892, April 21; 1893, April 29; 1894, April 2; 1895, April 11; 1896, April 15. Empidonax difficilis Baird. Western Flycatcher. Rather common, but very slry and difficult to secure. Represented by two adult females from Cumshewa Inlet. Mr. Keen has noted its spring arrival at Massett as follows: 1892, May 15; 1894, May 20. Cyanocitta stelleri carlottae subsp. nov. Queen Charlotte Ja}". Type from Cumshewa Inlet, Moresby Island, Queen Charlotte Islands, British Columbia. $ ad., No. 166822, U. S. Nat. Mus., Biological Survey collection. Collected June 17, 1900, by W. H. Osgood and E. Heller. Orig. No. 400. Characters. — Similar to ('. stelleri, but larger and darker colored; abdomen and flanks deep Berlin blue instead of Antwerp or China blue as in C. stelleri; frontal spots much reduced; black of head extending on breast and merging into blue of abdomen without sharp demarcation. Color. — Head, neck, and back deep dull black (bluish black in some specimens and very slightly brownish in others); forehead with very slight blue spots or immaculate; upper parts of wings and tail deep Berlin blue; inner secondaries and tip of tail with black bars of vary- ing distinctness; rump and upper tail-coverts Paris blue; throat and neck black or brownish black; breast Berlin blue becoming slightly paler on flanks and crissum. Measurements. — Type: Total length 350; extent 483; (length and extent measured in flesh) wing 153; tail 155; exposed culmen 32.5; tarsus 49. Average of three adult males from the type localit}^: Wing 155; tail 154; culmen 32; tarsus 49. (Average of six adult males of C. stelleri from Puget Sound: Wing 147; tail 147; culmen 30; tarsus 45.) Remarks. — The large size and dark color of this jay were noticed in the field, and subsequent comparison of specimens in the museum showed these characters to be amply sufficient to distinguish it from the mainland form C. stelleri It is accorded only subspecific rank because its derivation from the mainland form is scarcely to be doubted, and because individual variation in C. stelleri occasionally approaches the condition of C. s. carlottce. It is represented in the collection by four adult and four immature birds, all but one of which were taken about Cumshewa Inlet. Jays are not very common on the islands. They were seen only occasionally and were generally in family parties of four to six adults and young. Corvus corax principalis Ridgway. Northern Raven. Very abundant. During June the majority of those seen were young birds of the year which were easily distinguishable by their juvenile sept., 1901.] BIRDS. 47 manners and ludicrous colloquial attempts. 1 frequently watched them feeding on crabs. The general method of procedure seemed to be for one raven to catch a small crab on the shore and then retire to a log or the top of a stump, a few rods back in the forest, to eat it. A\ nile he was enjoying the tidbit several of his companions would perch in the trees near by preening themselves and making vigorous comments now and then until it was time to return to the beach for another morsel. Corvus caurinus Baird. Northwest Crow. Not common. A flock of about thirty was seen several times near the head of Cumshewa Inlet. Pinicola enucleator flammula (Homeyer). Kadiak Pine Grosbeak. A small flock of pine grosbeaks was seen in Cumshewa Inlet June L6, and one immature male was taken. No others were seen during our stay. Loxia curvirostra minor (Brehm.). American Crossbill. Large flocks of crossbills were seen frequently, but as none came within range of our guns, no specimens were secured. Mr. Keen reports this species from Massett. Loxia leucoptera Gmelin. White-winged Crossbill. Mr. Keen found this species at Massett. We were unable to dis- tinguish species among the many crossbills that we saw in other parts of the islands. Spinus pinus (Wilson). Pine Siskin. Heard occasionally; no specimens taken. Seen at Massett by Mr. Keen. Passerina nivalis (Linn.). Snowflake. Seen at Massett by Mr. Keen. Ammodramus sandwichensis alaudinus (Bonap.). Western Savanna Sparrow. Reported from Massett by Mr. Keen; not seen by us. Calcarius lapponicus alascensis Ridgw. Alaska Longspur. Seen at Massett by Mr. Keen. Zonotrichia coronata (Pallas). Golden-crowned Sparrow Said to occur at Massett (Keen). Junco hyemalis oregonus (Towns.). Oregon Junco. Not common. It was very seldom seen near the coast and but few were noticed on the mountains. Although great pains were taken to secure every specimen seen, our total was but seven, and three of these were immature birds. If the junco that breeds at Sitka be con- sidered typical oregonvs, the Queen Charlotte birds are easily refer- able to this form. They seem to be identical in color, and the meas- urements differ too slightly to be of consequence. 48 NORTH AMERICAN FAUNA. [no.2L Melospiza melodia rufina (Bonap.). Sooty Song Sparrow. Very abundant. Their favorite haunts are the dense Rubus thick- ets along the shore, whence they occasionally wander out on the rocks and sandy beaches in search of insects and sand fleas. Hour after hour they sit swinging on the slender topmost twigs of the salmon- berry bushes and look out over the water while they pour forth a jubi- lant ringing song. In some of the few open grassy places they were particularly numerous, and in skulking through the weeds frequently came to grief by encountering our mouse traps. A nest which Heller found in Cumshewa Inlet June 24, was situated on the ground in a bunch of weeds near the water's edge. It contained two fresh eggs, which dissection of the female bird showed to be a complete clutch, though undoubtedly a second laying, as fledged 3Toung were abundant at that time. Another nest, which I stumbled upon near Skidegate Juty 14, was placed in much the same kind of situation and contained three fresh eggs. These eggs are slightly smaller than those of Melospiza m. insignis in the National Museum, but other- wise very similar. They measure as follows: 22.6 x 16.7, 22.7x16.8, 23x16.7, 22.8x16, 22x16.4. I have seen very few specimens of typical M. m. rufina from Sitka, but have little hesitancy in referring the Queen Charlotte bird to this form. There seems to be no appreciable difference in color and very little, if an3T, in size. The measurements of 12 males from the Queen Charlotte Islands average as follows: Wing 73, tail 69.5, exposed culmen 14.7, tarsus 24.6. Average of 6 females: Wing 68.5, tail 64, exposed culmen 14, tarsus 23.7. Passerella iliaca townsendi (Nutt.). Townsend Fox Sparrow. Common, but, as usual, exceedingly shy. Occasionally a bird would be seen pouring out a wealth of song from the top of an alder or willow near the shore, but more frequently they skulked away through the brush before one could get a fair sight of them. Represented by 10 specimens, 6 adult and 4 immature. These are not identical with breeding birds from Sitka, and perhaps should be considered inter- mediate between townsendi and fuligfinosa. The young particularly are more dusky than young from Sitka. In the adults the spotting on the lower parts is heavier and duskier and in general there is less of the deep rufescent shades than in typical townsendi. Hirundo erythrogastra Bodd. Barn Swallow. A few barn swallows were always found about the numerous deserted Indian villages and their nests were frequently noticed on the big cedar beams which are the framework of the Haida houses. Only one specimen was secured, and this is evidently not full grown, as the tail is not as long nor the color as rich as in the fully adult western birds I have examined. sept., 1001.] BIRDS. 49 Tachycineta bicolor (Vieill.). Tree Swallow. One or two swallows supposed to be this species were seen among the barn swallows in Cumshewa Inlet. Mr. Keen reports it from Massett, and has noted the time of its spring arrival as follows: 1891, April 30; L892, April 24: L893, May L2; 1895, May 3; L896, April 7; L89T, April 15; 1898, April 27. Tachycineta thalassina (Swains.). Violet-green Swallow. .V bird thought to be this species was seen by Heller m Cumshewa Inlet dime 30. Helminthophila celata lutescens (Ridgw.). Lutescent Warbler. Occasionally seen or heard. Two specimens were taken in Cum- shewa Inlet dune 15. Dendroica aestiva rubiginosa (Pallas). Alaska Yellow Warbler. Rather rare: seen twice in Cumshewa Inlet. Reported by Mr. Keen from Massett. Dendroica townsendi (Towns.). Townsend Warbler. One taken in Cumshewa Inlet dune 15, and rive at Skidcgate duly 14: very few others seen. Mr. Keen found it at Massett, and noted its spring arrival there as follows: 1891, May 30; 1893, April 28; 1894, May 15: 1896, April 20; 1898, April 17. Wilsonia pusilla pileolata (Pallas). Pileolated W^arbler. Two speeimens were taken and several seen in the mountains near the head of Cumshewa Inlet June 22-24. They were not seen else- where, but the species is noted in Keen's Massett list. ?Anthus pensilvanicus (Latham). American Pipit. A bird thought to be this species was seen on a snow field in the mountains of Moresby Island dune 23. Cinclus mexicanus Swains. American Dipper. A dipper was seen and beard several times along a stream emptying into AVest Arm of Cumshewa Inlet. Anorthura hiemalis pacifica (Baird). Western Winter Wren. Very common on all the islands. It is practically the onty bird to be found in the deep forest away from the seashore. On the occasions when we attempted to penetrate the labyrinth of undergrowth toward the interior of the islands, we were alwa}\s greeted, even in the dark- est places, by the tiny wren's bright bubbling song or scolding chatter. It is always in motion and utterly regardless of the weather. During continuous rains while we were camped at the head of Cumshewa Inlet a wren would appear every few hours near the front of the tent 35U5— No. 21— 01 4 50 NORTH AMERICAN FAUNA. [no. 21. and, after scolding us for awhile, move on through the wet brush cheerfully and oblivious of the drenching rain. I Hushed a bird from an empt}^ nest in the upturned roots of a large fallen cedar June 15. I visited this nest frequently and flushed the bird from it each time, but up to June 28 it still contained no eggs. Four specimens only were collected, two adults and one young from Cumshewa Inlet, and one young from Skidegate. These do not differ from specimens from the adjacent mainland of British Columbia and from Puget Sound near the type locality of Anorthura h. padjica. Certhia familiaris occidentalis Ridgw. Western Creeper. One specimen was taken and several others were seen in Cumshewa Inlet June 20. Sitta canadensis Linn. Red-breasted Nuthatch. Two specimens were taken in Cumshewa Inlet June 18 and June 22, respectively. No others were seen during our stay. Parus rufescens Towns. Chestnut-backed Chickadee. Abundant. Seven specimens were taken. Regulus satrapa olivaceus Baird. Western Golden-crowned Kinglet. Common. An adult male was taken in Cumshewa Inlet June 20. Regulus calendula grinnelli Wm. Palmer. Sitka Kinglet. Reported by Mr. Keen. Not seen by us. Hylocichla ustulata (Nutt.). Russet-backed Thrush. Common. Eight specimens were taken in various parts of the islands. It was very abundant at Clew on the north side of Cumshewa Inlet, but was not seen at all at our camp at the head of the inlet, where we found II a. verecunda. Mr. Keen found it common at Massett, and noted its annual arrival for seven years as follows: 1891, May 29; 1892, May 23; 1893, May 17; 1891, May 19; 1895, April 25, 1896, April 11; 1898, April 26. Hylocichla aonalaschkse verecunda Osgood.1 Coast Hermit Thrush. Rather rare. Two adult females were taken at the head of Cum- shewa Inlet, and one male at Prevost Island. These Queen Charlotte specimens have the extreme development of the characters of this form, being rich brownish olivaceous, although in breeding plumage. Merula migratoria propinqua Ridgw\ Western Robin. Common. No specimens preserved. Mr. Keen notes the spring- arrival of the robin at Massett as follows: 1891, March 12; 1892, March 16; 1893, March 6; 1894, February 20; 1895, March 1; 1896, February 21; 1898, February 21. Hesperocichla nsevia (Gmelin). Varied Thrush. Occasionally seen or heard. Seen at Massett (Keen). 1Auk, XVIII, 183, April, 1901. NATURAL HISTORY OF THE COOK INLET REGION, ALASKA. By Wilfred II. ( >sgood. INTRODUCTION AND ITINERARY. The region about Cook Inlet was, at the beginning of the Held season of 1900. the only general district of consequence on the Pacific roast of Alaska that had not been recently visited by naturalists. The important bearing- which collections from this region would have on problems connected with the general natural history of Alaska was strongly realized, and. accordingly, after the completion of work on the coast farther south. 1 was directed to proceed to Cook Inlet and make as thorough a biological reconnoissance of the region as time and circum- stances would permit. On this trip, as earlier in the season. I had the efficient assistance of Air. Edmund Heller. We entered the region August 21, making stops of a few hours each at Seldovia and Homci on the southwestern end of the Kenai Peninsula. From Homer we continued up the inlet and into Turnagain Arm. and landed at the mining camp of Hope City August 23. The lower coast country about Hope occupied us until August 31, when Ave moved on into the mountains at the head of # Bear Creek, a medium-sized stream that empties into Turnagain Arm near Hope. A week later we left Turn- again Arm for the northwest side of the inlet at Tyonek, and there spent the remaining time from September 13 to September 28. From this it may be seen that most of the work was done in but two general localities, the vicinity of Hope and the vicinity of Tyonek. Short stops at Seldovia. Homer, Kenai, and Sunrise, however, were of con- siderable value, and information received from prospectors gave some general information about the Knik and Sushitna districts. If more time had been available it could doubtless have been spent profitably in these districts. The vicinity of Seldovia also seemed promising, but we were obliged to pass it by on our way into the inlet and could not return to it. PHYSIOGRAPHY. Cook Inlet is the first important indentation of the Alaskan coast east of the Alaska Peninsula. It is a long narrow inlet bifurcated at its upper end into two large arms, Knik Arm and Turnagain Arm. The first of these, Knik Ann. i> about 15 miles long, and at its upper 51 52 NORTH AMERICAN FAUNA. [no. 21. end receives the waters of a large stream, the Matanuska. The other, Turnagain Arm, is 30 miles or more in length, and extends inland until within about 5 miles of the waters of Prince William Sound. West of Knik Arm is the delta of the Sushi tna River, the largest stream empty- ing into the inlet. South of Turnagain Arm, and connected with the mainland onl}T by the 5 miles of glacier between the head of the arm and Prince William Sound, is the Kenai Peninsula Numerous rela- tively small streams enter both sides of Turnagain Arm and both sides of the main inlet as well, so that in addition to the great volume received from the Sushitna there is a large secondary supply of fresh water. This abundance of fresh water, much of which carries quanti- ties of silt in suspension, makes the inlet unsuitable for an exten- sive marine fauna. The tides are very strong and the rise and fall very great, particularly in Knik and Turnagain Arms, where the flood is accompanied by a bore. Navigation by either large or small craft is difficult and often dangerous. Except in Turnagain Arm, the country bordering Cook Inlet is low and comparatively level, though high mountains from 10 to <>(» miles inland can be seen on all sides. The upper end of the main inlet, in the region of the Sushitna delta, is of course low and more or less swampy. The east shore along the Kenai Peninsula, from the month of Turnagain Arm nearly to Kachemak Bay, is also low and compara- tively flat, but is for the most part heavily wooded. The northwest side from Mount Sushitna, near the mouth of the Sushitna River, west- ward for nearly 100 miles consists of a slightly rolling coastal plain, varying possibly from 20 to (50 miles in width. This country is broken here and there by rather sluggish streams, most of which head in the mountains farther back or in the small lakes which abound between the coast and the mountains. The small trading station and native village of Tyonek is situated on a low sandspit at the base of this plain about 20 miles west of the mouth of the Sushitna. About Turnagain Arm the mountains reach the coast, and except for a few small tide flats at the mouths of relatively narrow valleys, the shore is steep. Hope City, in the vicinity of which our work in Turnagain Arm was done, is situated at the mouth of Resurrection Creek, a stream of suffi- cient size to have made quite a wide cut through the mountains. On each side of Resurrection Creek rugged mountains rise to an altitude of 5,000 feet or more and from their canyons many small rushing streams pour into Resurrection Creek or Turnagain Arm near Hope. Bear Creek (see PI. VI, fig. 2) and Palmer Creek, which come from the east side, are the most important of these. Although the region north and east of Cook Inlet is exccedingi}' mountainous and quite imperfectly known, it is evident that there is a gap of some consequence between the Coast or Saint Eiias Range, which practically culminates in the upper Matanuska region, and the North American Fauna, No. 21. Plate VI Fig. 1.— Peat Bog and mixed Woods near Tyonek. Fig. 2.— Looking toward Turnagain Arm from Head of Bean Creek. sect., 1901.] FLORA. 53 so-called Alaskan Range which lies north and northwest of the inlet and includes the lofty peak of Mount McKinley. The effectiveness of this gap in its relation to the geographic distribution of animals and plants is of great Interest. FLORA. The flora of the Cook Inlet region is quite different in its general character from that of the coast farther south, although many species are common to both regions. The difference is largely in the reduc- tion of the number of coniferous trees in the Cook Inlet region and the corresponding increase in deciduous trees; but other features somewhat transitional between the heavy saturated forest of the south- ern coast and the treeless tundra of the north are numerous. The flora of the mountainous district about Turnagain Arm is, of course, different from that of the coastal plains of other parts of the inlet. The low country near Hope consists of a grassy tide flat, about 50 acres in extent, and a few miles of forest and occasional small swamps along the lower part of Resurrection Creek. Balsam poplars, paper birches, alders, and willows abound near the streams, and spruces (Picea canadensis and Picea sitchensis) and hemlocks (Tsuga mertensi- ana) are common on the slopes and slightly elevated flats. A third species of spruce {Picea mariana) is found in the small peat bogs, where smaller Hudsonian plants, such as Labrador tea (Ledum), crow- berry (Empetrum), and dwarf birch (Betula glandulosa) are in profu- sion. The hemlock is much the most abundant of the large trees, but it is exceeded in individual size by the spruces. The conifers ascend the mountain slopes to about 2,000 feet but above that point rapidly disappear. Beyond this elevation are alder thickets, small patches of dwarf willows and birches, and vast stretches of waving grass from 1 to .". feet high. Still higher, the slopes and rounded backs of the ridges are cushioned with a mass of heather and heather-like shrubs, chiefly Empetrum nigrum. This extends up to an approximate alti- tude of 5,000 feet, above which there is very little or no plant growth. The whole country is characterized by the abundance of high grass; otherwise it is a typical Hudsonian-Alpine region. The flora on the northwest side of the inlet in the vicinity of Tyonek i- somewhat different in character. With the. exception of considerable areas occupied by lakes and peat bogs, the whole coun- try is covered with comparatively open forest (see PI. VI, fig. 1). Deciduous trees greatly outnumber conifers, of which but two species occur. Picea canadensis and Picea mariana, and one of these, P. mariana, is quite rare and local. The paper birch (Betula ]><']>!/- rifera) is by far the most abundant tree, and next in rank are the pop- lar-, of which there are two species, Popvlus balsamifera and Populus tremuloide8. Alders and willows are found along the streams and 54 NORTH AMERICAN FAUNA. [noM. sparingly through the forest. The underbrush is not heavy; it consists mainly of Menziesia and Viburnum, with an occasional clump of devil's club (JSckinopanax) in wet places. Long grass grows luxuriantly in numerous pretty open glades in the birch woods. The September aspect of the forest is very attractive. From a little distance the birches on the low, rolling slopes appear as a mass of golden and rusty yellow, punctured here and there bj^ the dark-green spruce tops. The foliage of many of the smaller plants, such as Viburnum, Cornus, Ribes, and Epilobium, is bright red, and adds greatly to the general effect. On the whole, it reminds one very much of the autumn woods of New England, and is quite unlike anything I have seen elsewhere in Alaska. Unfortunately, we made no collection of plants in the Cook Inlet region, hence an authoritative list can not be given here. The follow- ing list, with brief annotations copied from my field notes, gives a general idea of the important trees and woody plants that occur. Specimens of a few species were preserved, and these have been identi- fied by Mr. Frederick V. Coville; the remainder are field identifica- tions only. Tsuga mertensiana. Alpine Hemlock. This is the most abundant tree from the seacoast to timberline all about Turnagain Arm. It was not found elsewhere in the Cook Inlet region. Picea sitchensis. Sitka Spruce. A few trees of this species were found at Hope. Specimens preserved. Picea canadensis. White Spruce. Found at all points visited, including Homer, Hope, and Tyonek. It is very common about Turnagain Arm, and is found on the moun- tains up to an altitude of about 2,000 feet. It is practically the only conifer to be found at Tyonek. Specimens preserved. Picea mariana. Black Spruce. Found in limited numbers in peat bogs at fiope, Sunrise, and Tyonek. Specimens preserved. Empetrum nigrum. Black Crowberry. This is by far the most common heather-like shrub. It abounds in all the peat bogs in the low country, and there are miles and miles on the mountains where one could not walk without treading on it. Populus balsamifera. Balsam Poplar. Very abundant. Large groves stand on the flat near the mouth of Resurrection Creek and trees of smaller stature are numerous in all the Turnagain Arm country; also abundant at Tyonek. sept., 1901.] BXORA. 55 Populus tremuloides. Aspen. A few trees supposed to be this species were found with the balsam poplars at Tyonek. Not seen elsewhere. Salix sp. Willow. Four or more species of willows occur in the Cook Inlet region, including several dwarf species only found above or near timberline. Alnus sinuata. Alder. Abundant all about the inlet. In the low country it is found along streams, and on the mountains it forms into dense thickets extending in altitude far above the coniferous trees. Betula papyrifera. Paper Birch. Abundant all about the inlet. Its vertical range is about coexten- sive with that of the conifers. At Tyonek it .surpasses all other trees in point of numbers and grows to a slightly larger size than at Hope. .Judging from the size of some of the Indian birch baskets trees a foot or more in diameter are to he found. Betula glandulosa. Dwarf Birch. Common. In the low country it is most common in peat bogs; high on the mountains it is found on open slopes in company with the dwarf willows. Ledum groenlandicum. Labrador Tea. Rather common, but more or less contined to peat bogs and wet heather meadows. Specimen preserved. Ledum palustre. Dwarf Labrador Tea. Less common than the preceding species, with which it is found. Menziesia ferruginea. Mehziesia. A large percentage of the underbrush is composed ox this species. It was found at all points visited, but most commonly at Tyonek. Phyllodoce glanduliflora. Heather. Found in limited quantities above 2,000 feet altitude in the higher mountains near Hope. Specimen preserved. Cassiope tetragona. Cassiope. Rather rare. It was occasionally found iti the beds of Empetrum in the high mountains near Hope. Specimen preserved. Cassiope stelleriana. Cassiope. Quite common above timberline in the mountains near Hope. Speci- men preserved. 56 NORTH AMERICAN FAUNA. [no 21. Arctostaphylos uva-ursi. Bearberry. A few plants were found on a rocky point near Hope. It was not observed elsewhere about the inlet. Vaccinium vitisideea. Mountain Cranberry. Very abundant from the coast to the upper limit of plant growth. Vaccinium sp. Huckleberry. Several species are abundant. Sorbus sambucifolia. Mountain Ash. Common, both in the mountains near Hope and in the low hills at Tyonek. Viburnum pauciflorum. Highbush Cranberry. Very common at all points visited. Sambucus racemosus. Elderberry. Abundant in the mountains near Hope; occasionally seen near Tyonek. Cornus canadensis. Bunchberry. Excessively abundant. Echinopanax horridum. Devil's Club. A few clumps of under-sized devil's club were occasionally found in damp shady places about Cook Inlet. Bibes laxiflorum. Blue Currant. Occurs sparingly in Turnagain Arm. Bibes rubrum. Red Currant. Quite common in Turnagain Arm. Rosa acicularis. Wild Rose. Abundant all about the inlet; especially so at Tyonek. Amelanchier alnifolia. Serviceberry. A single bush was found at Tyonek. The species was not seen else- where about the inlet. Rubus strigosus. Raspberry. Abundant at Hope and Tyonek. Spiraea betulsefolia? Spiraea. Abundant. FAUNA. The mammals of the Cook Inlet region are essentially the same as those of the interior of Alaska. Nearly all the species of the lower SEPT., 1901.1 FAUNA. 57 Yukon Valley are found among them, and none show any marked peculiarities not possessed in their interior habitat. With the excep- tion of widely distributed species, such as the black bear, no species are common to the Cook Inlet region and the Sitkan region. Thus, while the mammals of Cook Inlet are not peculiar to the region, the mammal fauna, as a whole, is peculiar, as contrasted with that of the coast farther south. Two new species. Microtus mi/urus and Sorex eximiuS) were found, hut both are new. not only to Cook Inlet, hut to Alaska as well, and will undoubtedly be found in other parts of the Territory. Considering the latitude, both large and small mammals are numerous in species as well as individuals. Moose, bear, and mountain sheep are the principal big game, and although they have already been hunted to a considerable extent, it is probable that they are more abundant than in any equally accessible place in North America. Fur-bearing animals are well represented, but, as elsewhere in the North, have been much reduced in numbers. The smaller, less conspicuous mammals are such as are generally found throughout northern Alaska, and are well represented on account of the varied conditions offered by the mountains of the Kenai Peninsula and the low country on the northwest side of the inlet. Our collection of mammals from Cook Inlet numbers 240 specimens, the majority of which, of course, are species of small size, such as shrews and mice, since we made no special effort to secure big game. Birds were not found in great numbers. Owing to the lateness of the season at the time of our arrival in the inlet, those seen were per- manent residents or fall stragglers, the summer residents being missed almost entirely. Land birds, with the exception of grouse, which were fairly common, were not numerous in species or individuals. Water birds, particularly littoral or semi-pelagic forms, are notice- ably uncommon, probably on account of the brackish water of the inlet and the comparative absence of marine invertebrates. Ducks and geese, however, and birds which feed in fresh water are locally quite abundant. As in the case of the mammals, no birds are peculiar to the Cook Inlet region, but several interior species are found which do not occur on the Alaskan coast south of Cook Inlet. The only other land vertebrate is a frog, collected by Heller at Tyonek. The species has very kindly been determined by Dr. L. Stejneger as Rami eantabrigensis latiremis. The land vertebrates may be summed up as follows: MAMMALS. 1. Rangifer stonei. 8. Arctomys caligatus. i'. Aires gi.^as. 9. Castor canadensis. 3. Ovis dalli. 10. EvotomyH dawsoni. 4. Oreamnos kenned yi. 11. Microtus operarius kadiacensis. 5. Sriuropterus sp. 12. Microtus miurus. 6. Sciuras hudsonicus. 13. Fiber spatulatus. 7. Spermophilus ernpetra .sul)s]>. 14. Synaptomys dalli. 58 NORTH AMERICAN FAUNA. [NO. 21. 15. Zapus hudsonius alascensis. 16. Erethizon epixanthus myops. 17. Ochotona collaris. 18. Lepus americanus dalli. 19. Lynx canadensis. 20. Canis occidentalis. 21. Vulpes kenaiensis. 22. Ursus americanus. 28. Ursns middendorffi. 24. Lutra canadensis. 25. Lutreola vison. 26. Putorius kadiacensis. 27. Putorius rixosus. 28. Mustela ainericana. 29. Gulo luscus. 30. Sorex personatus. 31. Sorex alascensis. 32. Sorex eximius. 33. Myotis lucifugus. BIRDS. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. Gavia imber. 40. Gavia lumme. 41. Uria troile californica. 42. Stercorarius parasiticus. 43. Rissa tridactyla pollicaris. 44. Larus sp. 45. Larus Philadelphia. 46. Sterna paradisiea. 47. Diomedea albatrus. 48. Phalacrocorax pelagicus. 49. Anas boschas. 50. Datila acuta. 51. Aythya marila nearctica. 52. Somateria v-nigra. 53. Oidemia perspicillata. 54. Branta canadensis subsp. 55. Olor columbianus. Ardea herodias. 56. Grus canadensis. 57. Phalaropus lobatus. 58. Gallinago delicata. 59. Macrorhamphus griseus scolopaceus. 60. Tringa couesi. 61. Tringa bairdi. 62. Tringa alpina paciflca. 63. Ereunetes occidentalis. 64. Limosa haemastiea. 65. Totanus melanoleucus. 66. Actitis macularia. 67. Numenius hudsonicus. 68. Squatarola squatarola. 69. Canachites canadensis osgoodi. 70. Lagopus rupestris. 71. Lagopus leucurus. 72. Circus hudsonicus. 73. Accipiter atricapillus striatulus. 74. Halia?etus leucocephalus alascanus. 75. Falco columbarius. 76. Bubo virginianus saturatus. 77. Nyctea nyctea. Ceryle alcyon. Picoides americanus fasciatus. Selasphorus rufus. ( lontopus borealis. Pica pica hudsonica. Cyanocitta stelleri. Perisoreus canadensis fumifrons. Corvus corax principalis. Scolecophagus carolinus. Loxia curvirostra minor. Loxia leucoptera. Acanthis linaria. Spinus pinus. Calcarius lapponicus alascensis. Ammodramus sandwichensis alaudi- nus. Zonotrichia leucophrys gambeli. Zonotrichia coronata. Spizella monticola ochracea. Junco hyemalis. Melospiza melodia kenaiensis. Melospiza lincolni. Passerella iliaca annectens. Lanius borealis. Helminthophila celata lutescens. Dendroica coronata. Dendroica striata. Anthus pensilvanicus. Cinclus mexicanus. Certhia familiaris montana. Parus atricapillus septentrionalis. Parus hudsonicus. Regulus satrapa olivaceus. Regulus calendula. Hylocichla ustulatus almas. Hylocichla aonalaschka'. Merula migratoria. Hesperocichla nsevia. BATRACHIAN. Rana cantabrisjensis latiremis. sept., 1901.] LIFE ZONKS. 59 LIFE ZONES. IVo zones are evident in the Cook Inlet region, the Hudsonian and the Arctic-Alpine; All the low country about the inlet and also the mountain sides up to timberline may be considered as Hudsonian, and the region above timberline on the mountains as Arctic- Alpine. The Hudsonian region has the same general features as the great interior transcontinental Hudsonian belt, and is doubtless imperfectly eon- nected with it. This Hudsonian belt is not particularly marked by characteristic forms, since most of the species of plants and the genera of mammals and birds are also found in the Canadian zone; but its distinction consists in the absence of many of the forms which are characteristic of the Canadian zone or which range from the south up into that zone. A notable feature of the Hudsonian flora of Cook Inlet, which is to some extent an exception to the statement just made, is the abundance of Tsuga mertensiana at sea level on the shores of Turnagain Arm. This tree is exceedingly characteristic of the Hud- sonian zone, and except at this point has been found only high on mountains in the vicinity of timberline, as its name, the alpine hemlock, implies. The other large trees of Cook Inlet, Picea cana- densis, Picea sitchensis, Picea mariana, Populus balsamifera, Populus tremvloides, Alnus sinuata, and Betula papyrifera, are such as are generally found in a northern Hudsonian zone, but all are also found in the Canadian. Such trees as Pinus, Abies, Thuja, etc, which are represented in the Canadian zone of the Sitkan district, are entirely absent in Cook Inlet. The mammals and birds of the Hudsonian dis- trict of Cook Inlet are, like the trees, nearly all species which are found in the Hudsonian of the interior of Alaska, but which also range, at least to some extent, into the Canadian. All the genera of mammals belong to this category, as well as many species, such as Sciurus hudsonicus, Evotomys dawsoni, Fiber spatulatus, Synaptomys dalli, Erethizl>r< viatus}' skull distinctive. 1 Evotomys orca .Merriam, Proc. Wash. Acad. Sci., II, 24-2;",, .Mar. 14, 1900. 2 Miurus = cnrtailed. sEn.. 1901.1 MAMMALS. B5 Color. — Fresh faU pelage (No. 107167): Upperparts uniform pale tawny, lightly mixed with black; face, sides, and flanks exactly like back; underparts strongly washed with buffy; tail dusky above, bufly below and on sides. Warn summer pelage (Type, No. 107175): Upper- parts pale bufly gray with plumbeous under-fur showing through in places; underparts whitish gray faintly suffused with bufly. Young: Similar to adults but grayer, with buff somewhat intensified about the cars and base of tail. Skull. — Size small; dorsal outline nearly straight, very slightly depressed in interorbital region; nasals of moderate length, some- what expanded anteriorly; naso-frontal suture slightly emarginate; lachrymal shelf prominent, with decided dorsal depressions between fronto-rnaxillary sutures; z}Tgomata rounded anteriorl}T with scarcely any prezygomatic notch; interparietal slightly produced anteriorly, about twice as wide as long; audital bulla? full and rounded, their inner surfaces nearly parallel; lateral pits of palate shallow, inter- pterygoid fossa rather wide. Teeth rather light; upper incisors slightly sulcate; molar enamel pattern essential^ as in Microtus abbre- viatus; m3 with but 2 closed triangles; m1 with 5 closed triangles and o inner and 4 outer salient angles. Measurements. — Type: $ ad. Total length 153; tail vertebra? 31; hind foot 20. Average of three females: 133; 23; 19. Skull: Basal length 26; zygomatic breadth 15; mastoid breadth 11.4; nasals 7; alveolar length of upper molar series 6.2. Remarks. — In a general way Microtus miurus is a miniature of M. abbreviatus, which is undoubtedly its nearest known relative, but detailed differences are very numerous. External characters other than size are its relatively longer tail and its more ochraceous color; cranial characters most appreciable are its full, rounded, and nearly parallel audital bulla? and the nearly straight dorsal outline of the skull. The peculiar enamel pattern like that of M. abbreviates is sufficient to distinguish miurus from all other members of the sub- genus Microtus. The discovery of a mainland relative of the insular species abbreviatus is very interesting and indicates in a slight way how much is still to be learned of the small mammals of Alaska. In the high mountains of the interior other similar forms undoubtedly remain to be discovered. Eleven specimens were secured, including five adults and six young. These were all taken in ' meadows ' above timberline in the mountains near Hope on the south side of Turn- again Arm. In these places it was only by very careful and diligent trapping that they were secured, for though many runways were found but very few showed signs of recent use. The burrows instead of opening vertically in the side of a slight eminence, as usual, have entrances which are flush with the floor of the runway. The grass in which these runs are made is very short, as it grows in rocky soil near 3505— No. 21—01 5 66 NORTH AMERICAN FAUNA. [»o."2L the upper limit of vegetation in small hollows and basins. Snow lies in these places all the time except a few months in summer. Fiber spatulatus Osgood. Northwest Muskrat. Fiber spatulatus Osgood, N. Am. Fauna No. 19, pp. 36-37, Oct. 6, 1900. Muskrats are not known to occur about Turnagain Arm, but they are rather common about small ponds in the peat bogs near Tyonek. They also occur at Kenai, as is shown by two specimens from there collected by Bischotf in 1869 and now in the National Museum. These, as well as one that we secured at Tyonek, are typical Fiber spatulatus, having the small molars and expanded nasals exactly as in the type of the species. Synaptomys dalli Merriam. Dall Lemming Mouse. One adult male was taken in a small peat bog near Hope August 26, but persistent trapping in the same locality failed to secure more, and none were found elsewhere about the inlet. The single specimen secured is essential^ the same as an adult male from Lake Lebarge, Yukon Territory. The Cook Inlet specimen is slightly lighter, par- ticularly on the head and shoulders, where there is less admixture of black. The skull agrees perfectly with the Yukon specimen and both agree fairly well with the type of dalli, all having much larger audital bulla? than the coast form wtungeli. The flesh measurements of the Cook Inlet specimen are as follows: Total length 131; tail vertebrae 22; hind foot 20. Skull: Basal length 25; mastoid breadth 14.8; zygomatic breadth 15.6; nasals 7.5; alveolar length of upper molar series 7.9. Zapus hudsonius alascensis Merriam. Alaska Jumping Mouse. A jumping mouse in good condition was found floating in a sunken water barrel near Tyonek September 13. The entire vicinity was assiduously trapped, but no more could be obtained, from which it seems that the species is rare, though possibly it may have gone into early hibernation. It also occurs in Turnagain Arm, for a miner at Hope accurately described one to me that he had seen there several years ago. Erethizon epixanthus myops Merriam. Alaska Porcupine. Porcupines are abundant in the Turnagain Arm region, but are very rare at Tyonek. Mr. T. W. Hanmore, who has lived at Tyonek for eleven years, says that he has seen but one porcupine there in that time. The natives on the Kenai Peninsula use porcupine flesh as food and prize it very highly. They prepare the animals by first plucking out all the quills, then singeing off the hair, then roasting entire. I did not have an opportunity to taste the flesh cooked in this way, but found it very palatable when fried. Porcupines are eaten bj^ various SEPT., 1901.] MAMMALS. 67. carnivorous animals, particularly wolves, though doubtless only in extreme hunger. Old trappers and hunters say that the majority of the wolves taken in this region have porcupine quills in their stomachs and under the skin about their heads. Ochotona collaris (Nelson). Alaska Pika. Pikas do not occur in the mountains on the peninsula side of Turn- again Arm, but I was told by a miner from Knik River that they had been seen in the mountains near there. Lepus americanus dalli Merriam. Hall Varying Hare. A few signs of rabbits were seen at Hope, but no specimens were secured; they are said to be very abundant in winter. Six specimens were obtained at Tyonek, but only by persistent and careful trapping-. All were caught in steel traps set in runways in the thickets or in the peat bogs. Rabbits arc very seldom seen here in the daytime, and dependence on gun alone would result in but a small bag. Although the summer pelage of dalli is unknown and there is some possibility that topotypes would be slightly different in color from the Cook Inlet specimens, I have little hesitancy in referring the latter to dalli, since the skulls are nearly typical, being but slightly smaller and shorter. Most of the specimens from Tyonek are slightly immature and the color of the upperparts is strongly mixed with black. In one of the oldest (No. 107611) the predominating color of the upperparts is cinnamon; the middle of the back is heavily mixed with black, which becomes less intense laterally until at the edge of the pure white under parts there is no trace of it, and a clear cinnamon lateral line remains. The throat is also cinnamon with very faint signs of black. The outer sides of the fore legs and fore feet are russet and dusky and the inner sides are creamy white. The hind feet are white well mixed with cin- namon and dusky. The ears are nearly white, except in front, where there is a strong cinnamon and dusk}' admixture. The flesh measure- ments of the six specimens average as follows: Total length 411; tail vertebrae 39; hind foot 139. Lynx canadensis (Kerr). Canada Lj-nx. Lynxes are evidently still fairly common. Mr. George Coon, a reliable woodsman of Hope, told me that in a season of two and one- half months' trapping in the winter of 1899 he secured fifteen lynxes near the mouth of Turnagain Arm. Canis occidentalis Richardson. Wolf. Wolves are considered rather common in the Cook Inlet region. Mr. Coon, of Hope, told me that during the winter of 1899 he secured fourteen with poison. Among these were six in the black phase. 68 NORTH AMERICAN FAUNA. [no. 21. Vulpes kenaiensis Merriam. Kenai Fox. Vulpes kenaiensis Merriam, Proc. Wash. Acad. Sci., II, 670, Dec. 28, 1900. Red, cross, and black foxes are taken annually in limited numbers. This species, which, as far as known, is the only one found in the region, is the largest fox known to North America. The skins secured here are usually of very good quality and blacks or ' silver grays ' com- manding high prices are not uncommon. Some attempt at 'farming' these large foxes has been made, but so far has generally proved unsuccessful on account of the vicious habit the males have of injur- ing or devouring the young. It seems possible, however, that while this might interfere with such extensive breeding farms as are feasible in the case of the blue fox, it might be controlled and the business made profitable if a few pairs were kept at each of a large number of localities. Ursus americanus Pallas. Black Bear. Black bears are moderately common in the Cook Inlet region. A few are killed about Turnagain Arm each year, but they are less com- mon in the lower country on the northwest side of the inlet. While returning from our traps on upper Bear Creek on the evening of Sep- tember 1, we discovered a bear crossing a grassy place between two alder thickets on a hillside near us. I hurried* to camp and returned with my rifle and after a long stalk succeeded in securing it. It proved to be a young female. The fur was short, but even and glossy black. The animal had been feeding on berries entirely and its stomach was found filled to distension, chiefly with black crowberries {Empetrum nigrum). These were clean and absolutely free from twigs and leaves and so tightly packed that there hardly seemed room for another mouthful. The feeling of satisfaction enjoyed by the possessor of this well-filled paunch was very evident. Before shooting it I had an opportunity to watch it feeding and was amused at its exhibition of exuberant spirits. It would browse leisurely for a few minutes then would suddenly give a bound and roll over and over down a little heather-grown glade to the bottom and then jump up to gallop at full speed up and down and around in a circle, apparently impelled by nothing but sheer joy. This bear measured in the flesh as follows: Total length 1,310; tail vertebras 166; hind foot 213. Ursus middendorffi Merriam. Kadiak Bear. Large bears are still very often seen both on the Alaska Peninsula side of Cook Inlet and on the mountainous Kenai Peninsula. Accord- ing to report they were very abundant about ten years ago, but in the short time since have been So constantly pursued that their numbers have been greatly reduced. Nearly every old prospector has one or more stories to tell about personal experiences with big brown bears, and often is able to show the skins as evidence of his truthfulness. sept., 1901.] MAMMALS. 09 Both whites and natives distinguish several varieties of large hears according to color. One of these, which is called the 'big white bear." and of which I examined specimens, is creamy white1 about the neck, shoulders, and back, and pale brownish about the haunches and legs. Nearly every degree of gradation from these ' white ' bears to the dark brown ones may be found, however, so that it does not seem probable that more than one species is represented. Mr. T. W. Hanmore, of Tyonek, says the brown bear generally goes into hibernation early in October, but that a few years ago he saw the track of one that had plowed through 2 feet of snow down to the beach near Tyonek in the middle of November. Latax lutris (Linnaeus). Sea Otter. Sea otters are said to have been seen in Cook Inlet, but owing to the very muddy water it is probable that they were never numerous there, even in times of their greatest abundance elsewhere. Lutra canadensis (Schreber). Land Otter. Apparently rather uncommon, though a few are said to be taken every winter. Lutreola vison energumenos Bangs. Pacific Mink. Moderately common. A few mink tracks were seen along some of the small streams. Several skins of poor quality were offered for sale by miners at Hope. One specimen, a male in good pelage, was taken on a small stream near Tyonek September 16. It is not fully adult, and its skull shows no characters of value, but its color is very dark. On this account it is referred to energumenos. Putorms kadiacensis Merriam. Kadiak Weasel. One specimen was secured at Hope August 30. It was shot while in the act of making away with some scraps of meat that had been thrown out near the door of our cabin. This specimen is not quite adult, but agrees with specimens of kadiacensis of the same age in size, color, and cranial characters. Its flesh measurements are as fol- lows: Total length 326; tail vertebne 91; hind foot 15. Putorms rixosus Bangs. Bangs Weasel. One adult female was taken in a swampy place near Tyonek Septem- ber 19. It was caught in a small mouse trap in a Microtias runway and doubtless would have escaped had it not thrashed into a pool of water and drowned. This specimen differs but very slightly from the type of rixo&uS, and shows no definite approach to subspecies eskimo. The color of the upperparts is uniform vandyke brown, slightly darker than the type of rixosus; the tail is the same color, with a faint paleness on the underside and no trace of black anywhere; the under 70 NOKTH AMEBIC AN FAUNA. [no. 21. parts, including the under and inner sides of the forelegs and the fore- feet, are pure white; the toes and one-third of the hind feet are white. The skull of the Cook Inlet specimen is about the same size as that of the type of rixosus; the braincase is slightly flatter and more elon- gate; the teeth are identical. The flesh measurements are as follows: Total length 165; tail vertebrae 18; hind foot 21. Skull: Basal length 27.5; palatal length 10.8; zygomatic breadth 14; breadth across post- orbital processes 9; length of audital bulla? 10. The natives regard the capture of one of these rare animals as a piece of great good fortune. One old Indian who frequently visited our cabin told us that his brother who had caught one when a small boy had in consequence become a 'big chief;' and he assured me that since I had caught one I must surely be destined to become a man of great wealth and power. Mustela americana Turton. American Marten. Martens are only moderately common. George Coon, a reliable trapper of Hope, told me that in a season of about two and one-half months in 1899, near the mouth of Turnagain Arm, he took but 15 martens. Two marten skulls in the Biological Survey collection, col- lected by Dall De Weese on the Kenai Peninsula, are not referable to either M. a. caurina or 31. a. actuosa, but seem to be very nearly like typical americana. The skulls and teeth are about the same size as in americana from the Adirondack Mountains, New York, and the shape of the last upper molar also agrees with that of americana, being of almost equal width internally and externally. Gulo luscus (Linnaeus). Wolverine. Apparently rather common, as a number of skins are said to be secured annually. All of these are shipped via St. Michael to trading posts on the Yukon River, where they are sold to the Yukon natives, who value them very highly for making trimmings for their fur clothing. The natives and older prospectors tell many stories of the wolverine's skill and cunning in discovering and securing caches of provisions. Sorex personatus Geoffroy. Common Shrew. Very common at both Hope and Tyonek. Twenty specimens were taken in the vicinity of Hope and 19 at Tyonek. These are smaller and lighter colored than topotypes of streatori from Yakutat, but can hardly be referred to arcticus. I have not been able to find any char- acters in which they differ from personatus oi the eastern United States. The measurements of 20 Cook Inlet specimens average as follows: Total length 101; tail vertebra? 40; hind foot 12. sept., 1901.] MAMMALS. 71 Sorex alascensis shumaginensis Merriam. Shumagin Shrew. Sorex alascensis shumaginensis Merriam. Proc. Wash. Acad. Sci., II, 18, Mar. 14, L900. Seventy-six specimens of a shrew almost indistinguishable from X. shumaginensis were taken, 27 near Hope and 49 near Tyonek. These are slightly smaller than S. alascensis and decidedly paler col- ored, thus approaching X shumaginensis. They do not show as much light peppery spotting as shumaginensis, but otherwise do not differ from it. The skulls are slightly smaller than those of alascensis and practically identical with those of shumagmensis. Sorex (Microsorex) eximius,1 sp. nov. Type from Tyonek, Cook Inlet, Alaska. 9 ad. No. 107126, U. S. National Museum, Biological Survey collection. Collected September 14, 1900, by W. H. Osgood and E. Heller, brig. No. 1395. ( 'haracters. — Similar to Sorex hoyi, but larger and paler colored; skull widely different. ( blor. — Head, back, and sides uniform pearly sepia, slightly paler than in S. hoyi; underparts pale drab, not strongly contrasted with upperparts; tail bicolor. Skull. — Rostrum and interorbital region narrow and elongate; brain- case much higher than in S. hoyi and more compressed anteriorly, dis- tinctly elevated above plane of rostrum; palate long, narrow, and excavated. Mandibles longer and relatively more slender than in S. hoyi. Dentition much heavier than in S. hoyi; relative size of fourth unicuspid, as compared with first and second, quite small; inferior cusp of first upper incisor long and slightly decurved. Measurements. — Type: Total length 98; tail vertebrae 31; hind foot 11. Skull: Basal length (inferior lip of foramen magnum to front of middle incisors) 15; palatal length 6.5; mastoid breadth 7.1; antorbital breadth 4.3. Remarks. — The specimen which is the basis of the foregoing descrip- tion is the only one of its kind among nearly 150 shrews caught at Hope and Tyonek. It is of extreme interest, not only as representing a very distinct new species, but as the only specimen of the subgenus Jficr<>xo/'e,r recorded from Alaska. Its dentition is essentially as in the only other species of the subgenus, Sorex hoyi, but the form of its skull is entirely different and much more like the general type found in the subgenus Sorex. Myotis lucifugus (Le Conte). Little Brown Bat. A few bats were seen at Hope, but no specimens were secured. 1 Eximius=excellent, extraordinary. 72 NORTH AMERICAN FAUNA. [no. 21. BIRDS OF THE COOK INLET REGION. Gavia imber (Gunn.). Loon. An old skin of a loon was seen at Hope; otherwise the species was not observed by us. Gavia lumme (Gunn.). Red Throated Loon. Five specimens of ' Colymbus septentrionalis ' are recorded in the catalogue of the National Museum among Bischoff's birds from Fort Kenai. I have been unable to find any of these in the National Museum. Una troile calif ornica (Bryant). California Murre. Bean records specimens taken at Chugachik Bay (=Kachemak Bay) June 30, 1880, and reports the species as abundant; 1 and a specimen taken by George Palmer at Knik Station is in the National Museum. The species was not seen by us in August and September. Stercorarius parasiticus (Linn.). Parasitic Jaeger. Several were seen at Homer August 22. Not seen elsewnere about the inlet. Rissa tridactyla pollicaris Ridgw. Pacific Kittiwake. A few were seen at Homer August 22. Not noticed elsewhere in the inlet. Lams sp. A few large gulls in immature plumage were occasionally seen, but they were noticeably uncommon. This scarcity I found was due to the fact that for the past two years gulls have been systematically slaughtered for millinery purposes. A trader offered from 10 to 20 cents each for them, and consequently the Indians and half-breeds have killed every one that has come within range of their guns. Larus Philadelphia (Ord). Bonaparte Gull. Seen in considerable numbers at Homer August 22, but not elsewhere about the inlet. Sterna paradissea Brunn. Arctic Tern. The National Museum catalogues show that Bischoff secured 1 speci- mens of this species in May and June, 1869. We did not meet with it. Diomedea albatrus Pallas. Short-tailed Albatross. In the summer of 1880 Dr. T. H. Bean found this species common about the mouth of Cook Inlet, and a specimen was secured near Fort ^roc. U. S. Nat. Mus., V, 172, 1882. sept., 1901.] BIRDS. 73 Alexander. We did not sec it when we were in this vicinity, in August and September. 1900. Phalacrocorax pelagicus Pallas. Pelagic Cormorant. A single, lonely-looking cormorant was several times seen at Tyonek flying up the inlet close to the shore. Others were seen at Homer. Anas boschas Linn. Mallard. Common at Tyonek, where 7 immature birds of the year were shot by E. Heller in September. Dafila acuta (Linn.). Pintail. A specimen was taken by Bischofl' at Fort Kenai. Aythya marila nearctica Stejn. American Scaup Duck. A flock of 6 scaup ducks was seen on a pond near Tyonek Septem- ber 17. Somateria v-nigra Gray. Pacific Eider. A young bird and four eggs were secured b}7 Dr. T. H. Bean at Chugachik Bay (=Kachemak Bay) July, 1880. Oidemia perspicillata (Linn.). Surf Scoter. Several flocks of scoters supposed to be this species were seen at Homer August 22. A specimen was taken by Bischofl' at Fort Kenai July, 1869. Branta canadensis subsp? Canada Goose. Small flocks were frequently seen at Tyonek, but no specimens were secured. Olor columbianus (Ord). Whistling Swan. The miners about the inlet say that swans are often seen there and that several have been killed. Ardea herodias Linn. Great Blue Heron. A great blue heron was seen at Hope hj E. Heller. Grus canadensis (Linn.). Little Brown Crane. Immense flocks of migrating cranes are said to pass over Cook Inlet annually. Three specimens were taken at Fort Kenai by Bischofl. Phalaropus lobatus (Linn.). Northern Phalarope. Six specimens were taken by Bischofl at Fort Kenai May, 1869. Gallinago delicata (Ord). Wilson Snipe. One specimen was taken by Bischofl' at Fort Kenai May 5, 1869. 74 NORTH AMERICAN FAUNA. [no.21. Macrorhamphus griseus scolopaceus (Say). Long-billed Dowitcher. Four specimens were taken at Fort Kenai May -±-9, 1869, and one July 20, 1869. Two of these are still in the National Museum. Tringa couesi (Ridgw.). Aleutian Sandpiper. An Aleutian sandpiper was .seen on the beach near Homer August 22. Tringa bairdi (Coues). Baird Sandpiper. A sandpiper, thought to be this species, was seen at Homer August 22. Tringa alpina pacifica (Coues). Red-backed Sandpiper. One specimen taken at Fort Kenai by Bischoff Majr 16, 1869. Ereunetes occidentalis Lawr. Western Sandpiper. Three specimens were taken at Fort Kenai by Bischoff May 12-16, 1869. One of these (No. 58170) has been examined; it is perfectly typical of the large-billed form, occidentalis. Limosa haemastica (Linn.). Hudsonian Godwit. Nine specimens were taken by Bischoff at Fort Kenai. At least two of these are still in the National Museum — one an adult in breeding plumage, the other in fall plumage. Totanus melanoleucus (Gmel.). Greater Yellow-legs. Taken at Fort Kenai by Bischoff May and June, 1869; specimen still in National Museum. Actitis macularia (Linn.). Spotted Sandpiper. Several were seen along Resurrection Creek near Hope, and one was taken. Numenius hudsonicus Lath. Hudsonian Curlew. A specimen was taken by Bischoff at Fort Kenai May 18, 1869, but can not now be found in the National Museum. Squatarola squatarola (Linn.). Black-bellied Plover. One taken by Bischoff at Fort Kenai May 6, 1869; specimen exam- ined in National Museum. Canachites canadensis osgoodi Bishop. Alaska Spruce Grouse. 'Black grouse' or 'fool hens,' as they are locally termed, are very common in all the Cook Inlet region. They are easily killed, and many thus find their way to the miner's frying pan. The Indians and half-breeds also hunt them to a considerable extent. When flushed from the ground, they rise quickly and fly swiftly, but only to light in the nearest spruce. When this is but a few yards away, they immedi- ately flounder into the thickest part of it; but if a long stretch of sept., 1901.] BIRDS. 75 birches, poplars, or small deciduous bushes intervenes, they continue winding in and out until they reach the requisite spruce. In Sep- tember at Tyonek they were often found in small Hocks of eight or ten individuals. When flushed each would betake itself to a separate tree and after a brief interval start a subdued clucking, so that all could be easily located. At Hope they were found ranging from sea level to timberline; on one occasion I flushed a flock of grouse and but a short distance farther on a flock of ptarmigan. Their crops were usually found to contain spruce needles and Vacdnium and Viburnum berries, and in one case heads of Equisetum. Cook Inlet specimens agree perfectly with typical osgoodl from the Yukon Valley. Lagopus rupestris (Gmelin). Rock Ptarmigan. A few small flocks were seen in the mountains on the north side of Bear Creek, and three females were taken. They have been hunted more or less by the miners in this vicinity, and we found them very wild and hard to secure. Lagopus leucurus Swains. & Rich. Northern White-tailed Ptarmigan. While setting small mammal traps in a few patches of grass at the extreme head of Bear Creek September 5, I suddenly became aware of a subdued clucking apparently from a rockslide not far away. Upon following up the sound I was soon able to distinguish some gray birds moving over the rocks, but so closely did they resemble the back- ground that I could not see them unless they moved. They were quite tame and allowed me to approach within a few yards, so that I was easily able to make out an old female white-tailed ptarmigan and a brood of seven nearly full-grown young. The old bird was almost as solicitous for her charges as if they had been downy chicks, and led them away very adroitly, keeping up a continuous purring cluck and making herself as conspicuous as possible. Although evidently much alarmed at my presence, flight as a means of escape did not seem to enter their heads, and it was not until I fired on the old bird that the young took wing. I had only my rifle with me, and so was obliged to shoot the ptarmigan with a load intended for bear, but fortunately the bird was not irreparably mutilated and I was able to make a fairly good specimen of it. This specimen was found to be different from the white-tailed ptarmigan of the Colorado mountains, and examina- tion of the original description of Lagopus leucurus showed that the northern bird was the one originally described by Swainson.1 Circus hudsonius (Linn.). Marsh Hawk. One was seen flying back and forth near Homer over a meadow thickly populated with Microtus. Another was seen at Hope. 1 Cf. Auk, XVII, 180, April, 1901. 76 NOETH AMEKICAN FAUNA. [no.21. Accipiter atricapillus striatulus Ridgw. Western Goshawk. Goshawks were frequently seen near Tyonek, and two immature birds were taken September 18. Remains of Sciurus were found in their craws. Haliaeetus leucocephalus alascanus Towns. Alaska Bald Eagle. Said to occur; not seen by us. Falco columbarius-Linn. Pigeon Hawk. Rather common; several were seen at Hope and also at Tyonek. An immature bird was shot by E. Heller at Hope August 30. Its craw contained parts of crossbills. Bubo virginianus saturatus Ridgw. Dusky Horned Owl. Very common; they were heard nightly at Hope and occasionally at Tyonek. One was seen on a dark day in the deep birch woods back of Tyonek and an adult female was shot at Hope August 30. A specimen taken by George Palmer at Knik Station is in the National Museum. These two specimens are quite different from typical satu- ratus, but they are nearer to it than to any other described form. They are considerably lighter than saturatus and do not have barring- extending down on the feet to the toes, as is usual in that form; also the bars on the sides are not blended, but separated by distinct light areas. Nyctea nyctea (Linn.) Snow}^ Owl. The miners and traders about Cook Inlet say that snowy owls have frequently been killed there in winter. Ceryle alcyon (Linn.) Belted Kingfisher. Common along streams. Its loud, clattering cry was heard fre- quently along Resurrection Creek, near Hope, when the thick growth of trees and shrubs prevented seeing the bird. Picoides americanus fasciatus Baird. Alaska Three-toed Woodpecker. Represented by eight specimens as follows: Hope, four; Tyonek, one; Fort Kenai, three. It was found to be quite common in the Turnagain Arm region, but at Tyonek, where coniferous trees are scarcer, only one bird was seen. Three specimens taken by Bischotf at Fort Kenai in 1869 are in the National Museum. Selasphorus rufus (Gmelin). Rufous Hummingbird. Mr. T. W. Hanmore, who has been stationed at Tyonek for eleven years, says that he has seen humming birds there several times. This is doubtless near the limit of the range of the species, as the bird has not been recorded farther north. sin., 1901.] TSIRDS. 77 Contopus borealis (Swains.). Olive-sided Flycatcher. A specimen from Fort Kenai is in the National Museum. It is an adult male taken by Bischoff May 26, 1869. Pica pica hudsonica (Sab.). American Magpie. The miners at Sunrise City told us that magpies had been seen in that vicinity frequently, but we did not observe them there or at any other point in the inlet. Specimens taken in Graham Harbor by C. H. Townsend in 1892 are in the National Museum. Cyanocitta stelleri (Gmelin.). Steller Jay. Several specimens taken at Graham Harbor in 1892 by C. H. Town- send and ft. W. Evermann are in the National Museum. This is apparently the northern limit of the species, as we did not find it farther up the inlet in Turnagain Arm, nor on the north side at Tyonek. Perisoreus canadensis fumifrons Ridgw. Alaska Jay. Occasionally seen. One morning, after a light fall of snow, a small party of jays visited our camp in the mountains near Hope. A few were also seen at Tyonek. A large series was taken by Bischoff at Fort Kenai. Corvus corax principalis Ridgw. Northern Raven. Only moderately common. The trappers say they are very abun- dant in winter and a great nuisance to them, since they systematically spring their traps or take the bait from them. Scolecophagus carolinus (Miiller). Rusty Blackbird. Two males were shot by Heller at Tyonek September 23. No others were seen during our stay, but the birds undoubtedly breed in the vicinity, for two specimens were taken by Bischoff May 28 and July 1. respectively. An examination of the material in the National Museum shows a slight difference in size between eastern and western birds of this species. The bill especially is constantly a trifle shorter and lighter in specimens from Alaska. Loxia curvirostra minor (Brehm). American Crossbill. A specimen taken at Graham Harbor in 1892 by C. H. Townsend and B. W. Evermann is in the National Museum. Loxia leucoptera Gmelin. White-winged Crossbill. Common. They were not seen in large flocks, however, but gener- ally in pairs. Four specimens were taken at Hope August 25-28. Acanthis linaria (Linn.). Redpoll. Large flocks were seen frequently both at Hope and Tyonek, and one immature specimen was taken at Hope. Two summer adults are 78 NORTH AMERICAN FAUNA. [no. 21. in the National Museum, one taken by Bean at Chugachik Bay ( = Kachemak Bay), and one by Bischoff at Fort Kenai. Spinus pinus (Wilson). Pine Siskin. Three specimens were secured from a large Hock at Tyonek Sep- tember 22. The}" were not seen elsewhere about the inlet. Calcarius lapponicus alascensis Ridgw. Alaska Longspur. An adult male in breeding plumage was taken by Bischoff at Fort Kenai in May, 1869. Ammodramus sandwichensis alaudinus (Bonap.). Western Savanna Sparrow. Evidently an abundant breeder, as numerous specimens were taken in summer by Bischoff and Bean at Fort Kenai and Chugachik Bay. At the time of our work in August and September very few were seen. Four specimens were taken, three at Hope August 26, 28, and 29, respectively, and one at Tyonek September 18. Zonotrichia leucophrys gambeli (Nutt.). Intermediate Sparrow. Evidently a common breeder, as Bischoff took a number of speci- mens at Fort Kenai in May, 1869, at least one of which is still in the National Museum. The species was not observed by us, but it may have been overlooked among the immature birds seen at Hope and all* supposed to be Zonotrichia coronata. Zonotrichia coronata (Pallas). Golden-crowned Sparrow. Common in the low second-growth brush about the village of Hope; also occasionally seen in the mountains near there. Four birds col- lected by Bischoff at Fort Kenai are recorded in the National Museum catalogue as Zonotrichia querula. None of these are now at hand, but the entries doubtless refer to Z. coronata. Spizella monticola ochracea Brewst. Western Tree Sparrow. A specimen is recorded taken by Bischoff at Fort Kenai May 19, 1869, but it can not now be found in the National Museum. As the occurrence of the species is altogether probable, however, there seems no reason to doubt the identification. Junco hyemalis (Linn.). Slate-colored Junco. Common. Three specimens were taken at Hope August 26-28. Melospiza melodia kenaiensis Ridgw. Kenai Song Sparrow. The type of this subspecies was taken by C. H. Townsend at Port Graham on the Kenai Peninsula April 9, 1892. Two specimens were also taken at this locality by Dr. Bean July 4, 1880. A specimen met., 1901.] BIRDS. 79 taken at Hope August 26 differs from the type of kenaiensis to such a degree that it hardly seems possible that it merely represents the difference between fall and summer plumage. It is characterized by verv sooty coloration: the dark markings about the head, neck, and breast are very intense and the streaks on the back are very prominent. In size it is intermediate between h nait nsis and cawma, and as no fall specimens of either are at band it seems best to refer it to kenaiensis, which is geographically near. It measures as follows: Wing 74; tail 7<>; exposed culmen 14; bill from nostril 10; tarsus 24.5. Melospiza lincolni (And.). Lincoln Sparrow. An adult male was taken at Hope August 28, and a few others were seen while we were there. The specimen taken shows none of the characters attributed to Melospiza i lincolni striata. Passerella iliaca annectens Ridgw. Yakutat Fox Sparrow. Rather common, but very shy. ;is usual, and hard to secure. Two specimens were taken at Hope and one at Tyonek. These seem to be intermediate between /'. !. cmnectens and P. i. msularis, as they have the smaller bill and more dusk}' underparts of annectens and the lighter upperparts of msularis. Lanius borealis Yieill. Northern Shrike. An immature bird in the brown plumage was shot by E. Heller at Hope September »'». Several were seen near Homer. An adult from Fort Kenai, collected by Bischotf, is in the National Museum. At present I am unable to find sufficient characters to warrant use of the name iiwictus1 for these birds. The question is further complicated by Lanius horeaMs sibi/rievs, which, judging from three specimens in the National Museum, differs from Alaskan birds only in having the •yermiculations on the breast nearly obsolete. Helminthophila celata lutescens (Ridgw.). Lutescent Warbler. Three specimens taken by Bischotf at Fort Kenai May 22-2. Amelanchier alnifolia, 13,56. Ammodramub alaudinus, it, 78. Anas boschas, 10,73. Anorthura pacifica, 19 50. Ansei gambeli,40. Anthus pensilyanicus, 49. 80. A] '!'lc, Oregon crab. 13. Arctic-Alpine district, 59-60. Arctomys caligatus, 63. Arctostapbylos uva-ursi, 21,56. Ardea fannini,-10. aerodias, 73. Arenaria melanocephala, 41. Ash, mountain. 56. Aspen, 55. Aythya,40. nearctica, 7:;. Balaenoptera velifera, 26. Baldpate,40. Bat, Keen's.:;;. little brown, 71. northwest, 37. silver-haired, 36. sooty big-footed. 36-37. Batrachian, 19, 58. Bear, black, 68. Kadiak, 68-69. Queen Charlotte black, 30-32. Bearberry. 56. Beaver. .American, 63-64. Betula glandulosa, 55. papyrifeta, 56. Birch, dwarf, 55. paper, 55. Bird list, 19, 58. Blackbird, rusty. 77. Blackflsh,25. Bonasa sabini, 42. BrachyramphiLs marmoratus, 38. Branta canadensis subsp.,73. occiden talis, 40. Bubo saturatus, 76. Bufflehead,40. Bufo columbiensis, is. Bunehberry, 56. Buteo calurns, 12. ( lalcarius alascensis, 47, 78. Calidris arenaria, 41. Caltha palustris, 13. Canacbites nsgoodi, 74-75. ( lanadian zone, 20. Canis occidentalis, 67. Caribou, 26-28. Stone's, 62. i lassiope, 13,55. mertensiana, 21. stelleriana, 21, 55. tetragona, 55. ( lastor canadensis, 63-64. Cedar, giant, 12. yellow, 12. Cepphus columba, 38. Certhia montana, so. occidentalis, 50. Ceryle alcyon,44, 76. Chamaecistus procumbens, 21. Chainaecy paris, 12. Charadrius dominicus,41. Cbaritonetta albeola, 40. Chickadee, chestnut-backed, 50 Hudsonian, 80. long-tailed, 80. Cinclus mexicanus, 49, mi. Circus hudsonius, 75. Clangula americana, 40. Colaptes saturatior, 45. Contopus boreahs, 77. Cormorant, pelagic, 39, 73. Cornus canadensis, 56. occidentalis, 13. Corvus caurinus, 47. principalis, 46-47, 77. Cowslip, 13. Cranberry, highbush, 56. mountain, 56. Crane, little brown, 73. Crataegus brevispina, 13. Creeper, Rocky Mountain, 80. western, 50. 1 rossbill, American, 47, 77. white-winged, 47, 77. S3 84 INDEX. Crow, northwest, 47. Cr< iwberry, black, 54. Curlew, Hudson ian, 74. Currant, blue, 56. red, 56. wild, 13. Cyanocitta carlottse, 18, 46. stelleri, 77. ■ Daflla acuta, 40, 73. Deer, Sitka, 25. Dendragapus fuliginosus, 42. Dendroica coronata,79. rubiginosa, 49. striata, 79. townsendi, 49. Devil's club, 56. Diomedea albatrus, 72-73. Dipper, American, 49, 80. Dogwood, 13. Dove, mourning, 42. Dowitcher, long-billed, 74. Dryobates picoideus, 18, 44-45. Duck, American scaup, 73. harlequin, 40. scaup, 40. Eagle, Alaska bald, 76. northern bald, 43. Echinopanax horridum,21, 56. Eider, Pacific, 73. Elderberry, 13, 56. Empetrum nigrum, 54. Empidonax difflcilis, 46. Erethizon myops, 66-67. Ereunetes occidentalis, 41, 74. Eumetopias stelleri, 34. Evotomys dawsoni, 64. Falco columbarius, 76. pealei, 43. suckleyi, 43. Falcon, Peale, 43. Fiber spatulatus, 66. Finch, rosy, 60. Fish list, 20. Flicker, northwestern, 45. Flora list, 14-16. Flycatcher, olive-sided, 77. western, 46. Fox,Kenai,68. Fratercula corniculata, 38. Frog, 57. Gallinago delicata,41, 73. Gaultheria shalloh, 13. Gavia imber, 38, 72. lumme, 38, 72. pacifica, 38. Globicephala scammoni,25. Goat, Alaska mountain, 62. Godwit, Hudsonian, 74. Golden-eye, American, 40. Goose, American white-fronted, 40. Canada, 73. white-cheeked, 40. Goshawk, western, 42, 76. Grosbeak, Kadiakpine, 47. Grouse, Alaska spruce, 74-75. Oregon ruffed, 42. Grouse, sooty, 42. white, 42. Grus canadensis, 73. Guillemot, pigeon, 38. Gull, 72. Bonaparte, 39, 72. glaucous-winged, 39. short-billed, 39. Gulo luscus, 70. Hamatopus bachmani, 41-42. Haliseetus alascanus, 43, 76. Hare, Dall varying, 67. Harelda hyemalis, 40. Hawk, marsh, 75. pigeon, 76. sharp-shinned, 42. western red-tailed, 42. Hawthorn, 13. Heather, 13, 55. Helminthophila lutescens, 49, 79. Hemlock, Alpine, 12, 54. western, 12. Heron, great blue, 73. northwest coast, 40. Hesperocichla nsevia, 50, 81. Hirundo erythrogastra, 48. Histrionicus histrionicus, 40. Honeysuckle, 13. Huckleberry, 13, 56. dwarf, 13. Hudsonian zone, 21-22, 59. Hummingbird, rufous, 46,76. Hylocichla aliciee, 80-81. almae, 81. aonalaschkse, 81. ustulata, 50. verecunda, 50. Jaeger, parasitic, 72. Jay, Alaska, 77. Queen Charlotte, 18, 46. Steller's, 77. Junco hyemalis, 78. oregonus, 47. Junco, Oregon, 47. slate-colored, 78. Kalmia glauca, 13. Kingfisher, belted, 44, 76. Kinglet, ruby-crowned, 80. Sitka, 50. •western golden-crowned, 50, 80. Kittiwake, Pacific, 39,72. Labrador tea, 55. dwarf, 55. Lagenorhynchus obliquidens, 25. Lagopus, 42. leucurus, 75. rupestris, 75. Lanius borealis, 79. Larus, 72. brachyrhynchus, 39. glaucescens, 39. Philadelphia, 39, 72. Lasionycteris noctivagans, 36. Latax lutris, 34, 69. Laurel, dwarf, 13. Ledum grcenlandicum, 55. INDEX. 85 Ledum palustre, 55. Lepus dalli, 67. Leucostdcte, 60. Life zone, Canadian, 20-21. Hudsonian, 21-22, 59. Limosa luvmastiea, 74. Lion, Steller sea, 34. Longspur, Alaska. 17,7s. Loniceia involucrata, 13. Loon, 38, 72. Pacific, 38. red-throated, 38, 72. Loxia leucoptera.47,77. minor, -17. 77. Luetkoa pectinata, 21. Lunda cirrhata, 38. Lutra canadensis, 32, 69. Lutreola energumenos, 69. Lynx. Canada, 67. Lynx canadensis, 67. Macrorhamphus scolopaceus,74. Magpie, American, 77. Mallard, 10,73. Mammal list^, 18, 57-68. Mareca americana, 40. Marmot, hoary, 63. Marten, American, 70. Queen Charlotte, 33-34. Megascops kennicotti,43. Melospiza kenaiensis, 7S-79. lincolni, 79. rutina, 4s. Menziesia ferruginea, 13, 21, 55. Merganser, American, 39. red-breasted, 39. Merganser amerieanus, 39. serrator, 39. Merlin, black, 43. Merula migratoria, si. propinqna, 60. Microtus kadiacensis. 64. miurus, 57,64-66. Mink, Pacific, 69. Moose, Alaska. 57, 61. Mouse. Alaska jumping, 66. Dall lemming, G6. Dawson red-backed. til. house, 28. Keen's, 28-29. Prevost Island, 29-30. Murre, California, 38-39, 72. Murrelet, ancient, 38. marbled, 38. Muskrat, northwest, 66. Mns musculus, 28. norvegicus, 28, 64. Mustela, 16. americana, 70. nesophila, 33-34. Myotis caurinus, 37. keeni,37. lucifUgus, 71. saturatus, 36-37. Nettion carolinensis, 40. Numenius hudsonicus, 74. Nuthatch, red-breasted, 50. Nyctala scoteea, 43-44. Nyctea nyctea,44,76. i iceanodroma furcata,39. Ochotona collaris, 67. Odocoileus sitkensis, 26. Oidemia deglandi, 40. perspieillata, 40, 7:;. Old-squaw, 40. Olor, 40. columbianus, 73. < ireaimicis kennedyi, C>2. ( Isprey, American, 43. Otoes alascanus, 34. Otter, land, 32, 69. sea, 34, 69. Ovis dalli, 62. Owl, dusky horned, 76. Kennicott screech, 43. northwest saw-whet, 43-44. snowy, 44, 76. Oystercatcher, black, 41-42. Parus hudsonicus, 80. rufescens, 50. septentrionalis, 80. Pandion carolinensis, 4H. Passerella annectens, 79. townsendi, 48. Passerina nivalis, 47. Pedicularis lanata,21. pedicellata,21. Perisoreus fumifrons, 77. Peromyscus keeni, 17, 28-29. prevostensis, 17, 29-30. Petrel, forked-tailed, 39. Phalacrocorax pelagicus, 39, 73. Phalarope, northern, 73. Phalaropus lobatus, 73. Phoca largha, 35. Phocsena phocsena, 25. Phyllodoce, glanduliflora, 13,21,55. Pica hudsonica, 77. Picea canadensis, 64. mariana, 54. sitchensis, 11,54. Picoides fasciatus, 7f>. Pika, Alaska, 67. Pine grosbeak, Kadiak, 47. Pine, northwest coast, 11, 12. Pine siskin, 47, 78. Pinicola flammula, 47. Pintail, 40, 73. Pinus contorta, 12. Pipit, American, 49, 59, 80. Plover, American golden, 41. black-bellied, 41,74. Poplar, balsam, 54. Populus balsamifera, 54. tremuloides, 55. Porcupine, Alaska, G6-C7. Porpoise, common, 25. striped, 25. Porzana Carolina, 40. Ptarmigan, rock, 75. northern white-tailed, 75. Puffin, horned, 38. tufted, 38. 86 INDEX. Puffinus griseus, 39. tenuirostris, 39. Putorius haidarum, 32-33. kadiacensis, 69. rixosus, 69-70. Pyrus rivularis, 13. Rat, Norway, 28, 64. Rail, sora, 40. Rana latiremis, 57. Rangifer dawsoni, 26-28. stonei, 62. Raspberry, 56. Raven, northern, 46-47,77. Redpoll, 77-78. Regulus calendula, 80. grinnelli, 50. olivaceus. 50, so. Ribesbracteosum, 13. laxiflorum, 13, 56. rubrum, 56. Rissa pollicaris, 39, 72. Robin, 81. western, 50. Rosa acicularis, 56. Rose, wild, 56. Rubus spectabilis, 13. strigosus, 56. Salix,55. scouleriana, 12-13. Salmon berry, 13. Sallal,13. Sambueus racemosus, 13,56. Sanderling, 41. Sandpiper, Aleutian, 71. Baird, 74. red-backed, 71. sharp- tailed. II. spotted, 41, 74. western, 41, 74. Sapsueker, northern red-breasted, 4f Sciuropterus, 63. Sciurus hudsonicus, 63. Scolecophagus carolinus, 77 Scoter, surf, 40, 73. white-winged, 40. Sea lion, Steller, 34. Seal, fur, 34. Pacific harbor, 35. Selasphorus rufus, 46, 76. Service berry, 56. Shearwater, dark-bodied, 39. slender-billed, 39. Sheep, Dall, 57, 62. Shrew, common, 70. Prevost Island, 35. Queen Charlotte, 35-36. Shumagin,71. Shrike, northern, 79. Siskin, pine, 47, 78. Sitta canadensis, 50. Snipe, Wilson, 41, 73. Snowflake, 47. Somateria v-nigra, 73. Sorbus sambucifolia, 56. Sorex elassodon, 35-36. eximius, 67, 71. Sorex personatus, 70. prevostensis, 17, 35. shumaginensis, 71. Sparrow: golden-crowned, 47, 78. intermediate, 78. Kenai song, 78-79. Lincoln, 79. sooty song, 48. Townsend fox, 48. western savanna, 47, 78. western tree, 78. Yakutat fox, 79. Spermophilus empetra subsp.,63. Sphyrapicus flaviventris, 45. Spinus pinus, 47, 78. Spiraea, 56. Spiraea betulaefolia, 56. Spizella ochracea, 78. Spruce, black, 54. Sitka, 11, 54. white, 54. Squatarola squatarola, 41, 74. Squirrel, flying, 63. ground, 63. Hudson Bay red, 63. Sterna paradisaea, 72. Stercorarius parasiticus, 72. Swallow, barn, 48. tree, 49. violet-green, 49. Swan, 40. whistling, 73. Synaptomys dalli, 60. Synthliboramphus antiquus, 38. Tachycineta bicolor, 49. thalassina, 49. Taxus brevifolia,12. Teal, green-winged, 40. Tern, Arctic, 72. Thrush, Alma, 81. coast hermit, 50. dwarf hermit, 81. gray-cheeked, 80-81. russet-backed, 50. varied, 50, 81. Thuja plicata, 12. Toad, 18. Totanus melanoleucus, 41, 74 Tringa acuminata, 41. bairdi,74. couesi,74. pacifica, 74. Trout, 18. Tsuga heterophylla, 12. mertensiana, 12, 54. Turnstone, black, 41. Uria califomica, 38-39, 72. Ursus americana, 68. carlottae, 30-32. middendorfri, 68-69. Vaccinium, 56. caespitosum, 13. ovalif olium, 21. uliginosum, 21. vitisideea, 56. INDEX. 87 Viburnum pauciflorum,21,56. Vole, Alaska mountain, 64-66. Kadiak,64. Vulpea kenaiensis, 68. Warbler, Alaska yellow, 19. black-poll, 79. lutescent.49,79. myrtle, 79. pOeolated, 49. Townsend,49, Weasel, Bangs', 69-70. Haida, 32-33. Kadiak, 69. Whale, finback, 25. Willow, 12-13,55. Wilsonia pileolata, 49. Wolf, (17. Wolverine, 70. Woodpecker, Alaska three-toed, 76, Queen charlotte, 18,44-15. Wren, western winter, 49-50. Yellowlegs, greater, n, 74. Yew. Pacific, 12. ZapUS alaseensis, 66. zenaidura macroura, 42. Zonotrichia, coronata, 47, Vs. gambeli, 78. o North American Fauna, No. 22. 4 112 no 103 106 I0< i02 100 36 91 9J 90 68 86 W 82 SO 78 76 74 7£ 70 68 66 Scale of Miles. 0 55 » 75 im ia [SO KEEWATIN. 'I he route toHowed by this expedition is shown by e dotted line. U. S. DEPARTMENT OF AGRICULTURE DIVISION OF BIOLOGICAL SURVEY NORTH AMERICAN FAUNA No. 3 2 [Actual date of publication, October 31, 1902] A BIOLOGICAL INVESTIGATION OF THE HUDSON BAY REGION EDWARD A. PREBLE ASSISTANT BIOLOGIST, BIOLOGICAL SURVEY Prepared under the direction of Dr. C. HART MERRIAM CHIEF OF DIVISION OF BIOLOGICAL SURVEY WASHINGTON GOVERNMENT PRINT IN (J OFFICE 1902 CjtM LETTER OF TRANSMITTAL. United States Department of Agriculture, Washington, D. C, August 18, 1902. Sir: I have the honor to transmit for publication, as North American Fauna No. 22, a paper on the natural history of the Hudson Bay region, by my assistant, Edward A. Preble. C. Hart Merriam, Chief, Biological Survey. Hon. James Wilson, Secretary of Agriculture. PREFATORY NOTE. A century or more ago the employees of the Hudson's Bay Com- pany sent collections of birds and mammals from Hudson Bay to London. Some of these specimens came from the shores of Hudson Bay. others from trading- posts in the distant interior; but many were not labeled to show where they were obtained. They were examined by the naturalists of the time and a number of species were described and named as new. In most cases the original specimens have dis- appeared and modern naturalists have been greatly perplexed in attempting to ascertain just what the species really were. No mod- ern museum possessed anything approaching a representative collec- tion of the mammals and birds of Hudson Bay, and specimens for comparison with related forms from other parts of Boreal America were not to be had. The resulting embarrassment was most keenly felt when the Biological Survey secured large collections from Alaska. In many instances it was impossible to tell whether certain Alaska species were identical with or distant from related forms previously described from Hudson Bay. In order to obtain the long-needed material it was determined to send an expedition to Hudson Bay. Edward A. Preble was placed in charge of this expedition; his report shows how well and faithfully his duties were performed. His suc- cessful trip, in an open boat, in inclement fall weather, from Fort Churchill to the Barren Grounds near Cape Eskimo, in search of topo- types of the Hudson Ba}r ground squirrel (often known as Parry's marmot) deserves special commendation. I take pleasure also in referring to the uniform courtesies and facil- ities extended by the officers and employees of the Hudson's Bay Company, particularly by Mr. C. C. Chipman, commissioner of the companv, at Winnipeg. C. H. M. CONTENTS. Page. Introduction ami itinerary 9 General account of the region traversed 11 Life zones of the region 22 Previous work 2o Bibliography 27 Note on boundaries of the region treated 38 New species 38 Ann< itated list of mammals 39 Annotated list of birds 75 Anm itated list of batrachians 133 ILLUSTRATIONS Page Plate I. Map of Keewatin Frontispiece. II. Fig. 1. General view of Norway House. — Fig. 2. Shore of channel near Norway House 10 III. Fig. 1. Spruce thicket near Norway House. — Fig. 2. Aspen thicket near Norway House 10 IV. Fig. 1. Rocky shore near Norway House, showing store of fur- trader. — Fig. 2. Sea River Falls (looking down ) 12 V. Fig. 1. Oxford House (photographed by William Campbell). — Fig. 2. Indian Camp, Oxford House 12 VI. Fig. 1. Rapid below Windy Lake. — Fig. 2. Canoe entering rapid, Trout River 14 VII. Fig. 1. Rapid, Trout River.— Fig. 2. Shore of Knee Lake near south end 14 VIII. Fig. 1. White Mud Rapid, Hill River.— Fig. 2. Hill River near See- ing Portage; Brassy Hill in distance 16 IX. Fig. 1. Left bank of Hill River from Rock Portage.— Fig. 2. Clay banks, lower Hill River 1G X. Fig. 1. Fort Churchill. — Fig. 2. Meadows, looking southwest from Fort Churchill. — Fig. 3. Seal {Erignathusbarbalus), Fort Churchill. 18 XL Fig. 1. Ledge of quartzite at Fort Churchill. — Fig. 2. Low tide at Fort Churchill 18 XII. Fig. 1. Camp on Barren Grounds, 50 miles south of Cape P^skimo. — Fig. 2. Burrow of lemming (Lemmus trimucronatus), Barren Grounds 20 XIII. Fig. 1. Camp on Barren Grounds, 25 miles south of Cape Eskimo. — Fig. 2. Eskimo guide at northern camp 20 XIV. Fig. 1. Dragging canoe up rapid, Trout River. — Fig. 2. Hell Gate Gorge ( looking up ) 22 7 No. 22. NORTH AMERICAN FAUNA. October, 1902. A BIOLOGICAL INVESTIGATION OF THE HUDSON BAY REGION. Bv Edward A. Preble. INTRODUCTION AND ITINERARY. In 161<> Henry Hudson, while searching for a northwest passage, entered and partially explored the great inland sea that bears his name. In 1670 the Hudson's Bay Company was organized to trade for furs with the natives of the great unexploited territory adjacent to the Bay. This company first established several trading posts at the mouths of the rivers tributary to the Bay and then gradually extended its field of operations inland. By this means the southern and western shores of Hudson Bay and the principal rivers emptying into it on the west had become fairly well known at a time when immense areas in North America, apparently more favorably located and more accessible, were still unexplored. As a natural result the birds and mammals of this semiarctic region were early brought to the attention of naturalists, and many species whose ranges extend over a very large area were first described from specimens sent to Europe from Hudson Bay. This was mainly due to the labors of the employees of the Hudson's Bay Company, who, residing at trading posts and coming in contact for purposes of trade with practically all the natives of the region, were able to secure natural history speci- mens with comparative ease, especially the larger species. The many collections thus made were conveniently sent to .England by the ships which paid annual visits to the posts. In this way a number of mam- mals and many birds, mainly littoral and pelagic species, first became known to science. As time went on, however, less attention was given to the fauna of this region, while most other parts of North America were ransacked for natural history material, so that the close of the nineteenth century found Hudson Bay one of the most neglected fields of modern zoological research. Some species, orig- inally described from poor specimens, and in the loose and inaccurate style of a hundred years ago, were knoAvn by these descriptions alone, while others were represented in museums only by poorly 9 10 NORTH AMERICAN FAUNA. [No. 22, stuffed and faded .specimens, entire]}- inadequate to meet the require- ments of modern scientific methods. This lack of material, in con- nection with the absence of definite knowledge as to the boundaries of the life zones, made it desirable that a collection, as thorough as possible, be made in the region. Early in the summer of 1900, therefore, I was detailed to make a biological reconnaissance of as much of the region immediately to the west of Hudson Bay as it would be practicable to cover in a single season. My brother, Alfred E. Preble, of Tufts College, Massachu- setts, accompanied me as assistant. The Hudson's Bay Compan}^ still maintains trading posts through- out the region we were to visit, and the officials of the company com- pose almost its entire white population. These posts are situated on the usual lines of travel, and constitute the only bases of supplies available; hence it was considered advisable to arrange with the com- pany for food and means of transportation. This we did on our arrival at Winnipeg on June 13, and obtaining a canoe from the company, set out the next day down the Red River. The following morning we took the Northwest Navigation Company's steamer Prince.™ at West Selkirk, and on June 17 arrived at Norway House, near the north end of Lake Winnipeg, where we were to begin operations. Here we collected until June 23, when our northern trip was resumed. We took two Indians for guides, boatmen, and camp hands, and a large Peterborough canoe, in which our collecting and camp outfit and provisions were carried. We passed down the east channel of Nelson River, and ascending the Echimamish, followed the usual boat route to York Factory, stop- ping to collect at favorable points. At the head of the Echimamish proper, which terminates abruptly at a rock about 30 yards broad called the Painted Stone, we made a portage and launched our canoe in a small lake. A stream flows eastward from this lake and we thus had the advantage of the current for the remaining distance to Hud- son Bay. Beyond Painted Stone Portage the route passes successively through the Robinson lakes, Franklin River, and Pine, Windy, Oxford, Knee, and Swampy lakes. These different lakes vary from a few miles to forty in length, and the channels connecting them con- tain numerous rapids. Hill River forms the outlet of Swampy Lake, the last of the chain, and unites with Fox River to form Steel River. This in turn unites with the Shamattawa, and the resulting stream, known as Hayes River, empties into Hudson Bay at York Factory. On reaching the Bay we exchanged our canoe for a sailboat and made our way up the west coast to Fort Churchill, at the mouth of the river of that name. Here, after a few days' stay, 1 left my brother to com- plete the collection, while 1 pushed northward well into the Barren Grounds. 'Phis trip consumed three weeks, and on my return to Fort Churchill we immediately started on the homeward journey in order North American Fauna, No. 22. Plate II. Fig. 1.— General View of Norway House. Fig. 2.— Shore of Channel near Norway House. North American Fauna No. 22. Plate III. Fig. 1.— Spruce Thicket near Norway House. Fig. 2.— Aspen Thicket near Norway House. October, 190J] ACCOUNT OF REGION TRAVERSED. 11 to complete it before navigation closed. We passed down the coast to York Factory in a sailboat and retraced our way to Norway House in our canoe. The trip up the rapid streams with our heavily loaded boat was a very arduous one, but we reached Norway House without accident or delay on September 16, having completed a journey by canoe and sailboat of more than 1,200 miles, much of it through very difficult water. We took a steamer from Norway House on Septem- ber 10 and arrived at Winnipeg on September 22. During our trip to Hudson Bay we were placed under many obliga- tions to a number of officers of the Hudson's Hay Company, to whom our cordial thanks are hereby extended. Through the courtesy of C. C. Chipman, commissioner of the Hudson's Bay Company at Win- nipeg, we were able to make arrangements to secure supplies and trans- portation at the different trading posts of that company on our route; without this aid the trip could hardly have been accomplished. Among others who assisted us in various wa}7s are Messrs. William Clark, W. C. King, and Roderick MacFarlane, of Winnipeg; J. K. Mac- Donald, of Norway House; William Campbell, of Oxford House; G. B. Boucher, of York Factory; Ashton Alston, of Fort Churchill; and especially Dr. Alexander Milne, of Winnipeg (formerly of York Factory), who has given me many notes on the distribution of the larger mammals. To the Rev. Mr. Chapman, a missionary at Fort Churchill, we were also indebted for information and various courtesies. The Key. W. A. Burman, of Winnipeg, kindly furnished us with a list of the principal trees and shrubs occurring about Winnipeg. From Colonel Scobell, C. E., of Winnipeg, we obtained much detailed and valuable information concerning the boat route to York Factory. In preparing this report 1 have receiyed many courtesies from Mr. Robert Ridgway, curator, and Dr. C. W. Richmond, assistant curator of birds in the U. S. National Museum, who have not only permitted the unrestricted use of the collection under their charge but have helped me in many other wajs. Mr. Gerrit S. Miller, jr., assistant curator of mammals in the National Museum, has extended similar favors in regard to the mammals; and Dr. Leonhard Stejneger, curator of rep- tiles in the National Museum, has aided in identifying the frogs col- lected. Thanks are also due to Frederick V. Coville, botanist of the Department of Agriculture, and his assistants, for identifying the plants collected. Finally, I am indebted to Dr. C. Hart Merriam and Dr. T. S. Palmer, chief and assistant chief of the Biological Survey, for various courtesies extended during the progress of the work. GENERAL ACCOUNT OF THE REGION TRAVERSED. Winnipeg is situated at the junction of the Red and Assiniboine rivers, on the site of old Fort Garry. To the westward stretch the plains, but the vicinity of the rivers is well wooded with elm (Uhnus americana), mossy-cup oak (Quercus maerocarpa), basswood {Tilia 12 NORTH AMERICAN FAUNA. [No. 22, americana), ash-leaved maple (Ace?' negtmdo), and other species, with an undergrowth composed principally of viburnums, hazel (Corylus americana and rostrata), wolfberry (Symphorica/rpos occidentalis), hawthorn (Crataegus coccmea), etc. The Red River below (to the north of) Winnipeg is very winding and is inclosed between banks of clay and limestone which at first are rather high and steep and are fairly well wooded, though the woods seldom extend far back from the river. But just beyond West Sel- kirk (a village about 20 miles below Winnipeg, near the site of the historic Selkirk Settlement) the banks become lower and the woods gradually yield to willow thickets. Farther down, a few miles from the mouth of the river, these willow thickets in turn disappear, and Lake Winnipeg is approached through a marsh which extends as far as the eye can reach, and where numberless coots and other marsh- loving birds find a congenial home. Soon after we entered the waters of Lake Winnipeg, about 42 miles from our starting point, our course carried us too far from shore to permit observations as to forest conditions, and such was the case throughout much of our voyage up the lake, though a few oppor- tunities for notes were offered. At The Narrows we could see that the western shore was well wooded with birch and conifers, a charac- ter of forest which, we were told, continues south nearly to the mouth of Red River. At Bull Head, off which we anchored early on the morning of June 16, the forest consisted mainly of spruce, tamarack, a species of pine (probably Pinus divaricata), birch (Betula papyrlfera), and poplar (Popufats halsamifera), the deciduous species predominating. The shores of the northern part of the lake are low and sandy with numer- ous outcrops of gneiss, and many low islands of the same rock occur. Great Playgreen Lake, the body of water next traversed, lies just east of the northern part of Lake Winnipeg, with which it is connected at its southern end by a rocky channel. This channel is entered at Mossy Point, the southern extremity of a slender strip of land separating the two lakes, on which Norway House was originally situated. Nelson River issues from Great Playgreen Lake by two main channels, known as East and West rivers, which, coming together at Cross Lake, inclose Ross Island, 50 miles in length. East River, on leaving Great Play- green Lake, divides into several minor channels encircling small islands, then expanding forms Little Playgreen Lake, about 25 miles from the outlet of Lake Winnipeg, and 800 miles from the southern end. On one of these islands, at the southern margin of Little Playgreen Lake, stands Norway House (PI. II, fig. 1). Two miles distant, on the east- ern shore of the hike, is Rossville Mission. Like most of the region between Lake Winnipeg and Hudson Bay, the country about Norway House consists hugely of swamps, mainly North American Fauna, No. 22. Plate IV. Fig. 1.— Rocky Shore near Norway House, Showing Store of Fur Trader. Fig. 2.— Sea River Falls, Looking Down. North American Fauna, No. 22. Plate V. 4 i i J > j. ^« < ^^ Fig. 1.— Oxford House. Photographed by William Campbell. Fig. 2.— Indian Camp, Oxford House. October, 1902.) ACCOUNT OF KEG ION TRAVER8ED. 13 grown up to willows and tamaracks. Numerous elevated places occur, rocky 'islands' on which has accumulated a rather thin covering of soil, supporting a moderately heavy growth of black and white spruce (Picea nigra and /'. alba, PI. Ill, tig. 1), balsam fir (Abies balsamea), Banksian pine (Pinus divaricata), aspen poplar (Pqpulm tremvloides, PL III. tig. 2), balsam poplar (Populus baisamifera), canoe birch {II tul, i papyrifera), and tamarack (Larix larid/na). These species form the bulk of the forest between Lake Winnipeg and Hudson Bay. From the vicinity of Norway House to the Sea River Falls (PI. IV, tig. 2), about 2<> miles below, the shores are rather low (see PI. II, tig. 2; PI. IV, tig. 1). Then for the few remaining miles before the mouth of the Echimamish is reached the route lies through a channel bordered by rather high banks and forested with birch and poplar. The water of the Echimamish, which flows into Nelson River from the eastward, is very dark and contrasts markedly with that of the Nelson, which carries the whitish waters of Lake Winnipeg. The course here leaves East River and ascends the Echimamish, a short distance from the mouth of which it passes through Hairy Lake, a broad, shallow sheet of water a few miles in length, in which grow extensive patches of bulrush ( Scirpvs laoustris). Above this lake for more than 20 miles the Echimamish is a winding, sluggish stream, with an east and west trend, and averages about 50 feet in width. Its banks are low and marshy, and on the lower part extensive swamps border it on either side. Occasional outcrops of gneiss occur, dry ' islands,' which form the only available camping places. The forest consists mainly of spruce, tamarack, and willow, the latter usually predomi- nating. Alosquitos, which swarm over the entire region, are here almost unbearable, and as the shallowness of the water, w7hich is barely deep enough to float a canoe, makes paddling very difficult, the ascent of this river was perhaps the least pleasant part of our journey. Three dams, at one of which — the second — we did some collecting, are kept up for the purpose of holding back a sufficient amount of water to permit the passage of boats. The stream flows through a flat country and in several places in the upper part of its course, broadens and forms small ponds. In its comparatively still waters the yellow pond lily {XympJixa) grows abundantly. At the Painted Stone, about 36 miles from Norway House, the stream comes to an abrupt termination, and boats are carried across a rock and launched in a small lake wTith high, rocky shores. From this lake issues a stream generally considered a part of the Echimamish, which in the Cree language signifies ' the river that flows each way.' It would appear that the small lake is fed from some underground source, and that some of its wTaters escape into the western part of the Echimamish. The vicinity of Painted Stone Portage proved a very good collecting ground. The eastern part of the Echimamish is 14 NORTH AMERICAN FAUNA. [No. 22, deep and bordered by high, rocky banks, on which Potentilla triden- tata grows abundantly. Seven miles from the Painted Stone the stream unites with White Water River, which discharges the waters of Little Lake Winnipeg, and from this point to Oxford Lake the stream is called Franklin River. The Robinson lakes, the southern shores of which are rather marshy and the northern shores higher, are next passed, and then 12 miles from the junction of the two streams Robinson Portage is reached. Here a portage of about three-fourths of a mile is necessary to avoid Robinson Rapids, where the river plunges through a deep ravine in a series of falls and rapids, with a total descent of about 50 feet. Deep mossy woods border this gorge, the excessive moisture from the rapids causing a luxuriant growth. From the lower or northern end of Robinson Portage, which lies nearly north and south, extends a line of lakes with marshy shores and supporting an abundant growth of sweet flag (Acorns calamus). A short stop was made and some col- lecting done at the north end of the portage. Immediately below these lakes the river plunges with considerable rapidity through a rocky gorge called Hell Gate. A short portage is made at its entrance, and after being launched in a surging pool at the foot of the rapid, the canoe is borne swiftl}T through the gorge. In some places the rocky walls rise nearly perpendicularly without a break; in others the bank consists of a succession of steep mossy terraces, the homes of several eagles." Throughout most of its course of 7 miles through the gorge, the river is confined within narrow limits, and the smooth but impet- uous current bears the voyager rapidly onward, constantly bringing fresh vistas to his view. In a few places a portion of the rocky walls has fallen, partiall}7 damming the stream, and the canoe is run through short, rapid chutes, the perpendicular walls preventing a landing, however desirable it might appear. Farther down the rocky banks are not so high and the surrounding countiy is seen to consist of rugged rock masses scantity clothed with Banksian pines. Here the voyager may land to see the ' kettles'1— deep, rounded potholes of various sizes, which have been worn in the rock during past ages. A short distance beyond the lower end of Hell Gate Gorge, 23 miles from Robinson Portage, lies Pine Lake, a small, irregularly outlined body of water containing numerous islands and environed by rocky but fairly well wooded shores. Ten miles farther on, below a succes- sion of small ponds and channels with marshy shores, lies Windy Lake. Here the banks are moderately high and formerly were well wooded; but within the past few years they have been partially denuded by fire. The head of Windy Lake is 12 miles distant from Oxford Lake, near ^Compare Franklin, who says, in speaking of thisgor^e, "The brown fishing-eagle, had built its nest on oneoftbe projecting cliffs." Narrative of a Journey to the Polar Sea, p. 39, 1823. North American Fauna, No. 22. Plate VI. Fig. 1.— Rapid Below Windy Lake. Fig. 2.— Canoe Entering Rapid, Trout River. North American Fauna, No. 22. Plate VII. Fig. 1.— Rapid, Trout River. Fig. 2.— Shore of Knee Lake near South Enc October, 1902.] ACCOUNT OF REGION TRAVERSED. 15 the northern end of which, 30 miles farther, is situated Oxford House, the only post or habitation of any kind on this route between Norway House and Hudson 'Bay (PI. V, figs. 1 and 2). In the short stream connecting the two lakes four rapids occur, at two of which portages are necessary (PI. VI, tig. 1). Oxford Lake extends southwest and northeast, and its 30 miles are marked by irregular short's and many islands. The shores are mainly of rock and are generally well forested. The locality about Oxford House we found favorable for collecting and a stop of a few days was made. The promontory on which the post is situated was probably well wooded originally, but its western half has been entirely cleared and is mainly covered with grass, wTith a few patches of wTillows and other shrubs. East of this cleared area the ground slopes gently to Back Lake, about a quarter of a mile distant, and is fairly well covered with spruce, fir. tamarack, poplar, and willow. The soil is a stiff clay, and potatoes and other garden vegetables of fine quality are raised. The waters of Oxford Lake flow into Back Lake through a short, narrow channel, and those of Back Lake find their outlet in Trout River (PI. VI. tig. 2; PI. VII, fig. 1), which runs southeastward to Trout Falls. 9 miles from Oxford House. At Trout Falls the river makes a plunge of about 12 feet into a deep pool. A short distance below is Knee Lake, a rather narrow body of water 4() miles in length and extending in a general southwest and northeast direction, with two somewhat abrupt bends. The shores of* the southern part are high and well wooded (PI. VII, fig. 2), and many wooded islands of various sizes dot the surface of the lake. At about a third of its length from the upper end the lake contracts and its shores become low and swampy. In this narrow part is Magnetic (or Magnetite) Island, a low, bare, irregular rock which offers considerable attraction to the magnetic needle. The shores and islands of the northern or larger part of the lake are low by comparison with those of the southern part, but are, like those, well wooded. The next lake in the series is Swampy Lake, which is connected with Knee Lake b}r Jack River, a stream about 10 miles in length, containing four rapids in its brief course. Swampy Lake is 13 miles long and has low shores, especially on the eastern side, where the Hudson's Bay Company formerly had a post, long since abandoned. Various species of water milfoil (Myriophyllum) and pondweed (Pota- mogeton) grow in the shallow portions of all these lakes, sometimes so profusely as to seriously interfere with navigation; and the beautiful water arum ( ( 'alia pahistris) is frequently seen near the margin of lake or river. The larger species of birds are noticeably scarce and wild, in consequence of the incessant warfare waged by the natives, who eat an}r- thing wearing fur or feathers, and never willingly allow a large bird of any kind to escape. For this reason the gulls, terns, and ducks, 16 NORTH AMERICAN FAUNA. [No. 22, which were nesting as we passed, were excessively shy. On the lakes and rivers off the main route these conditions probably do not prevail. Swampy Lake finds its outlet in Hill River, a rapid, winding stream, containing a great many willow -covered islands, and characterized during the first 30 miles of its course by numerous rapids. These necessitate frequent portages, half the entire number on the route, but fortunately they are all short. Each of these portages has its significant name— White Mud Portage (PI. VIII, fig. 1), Mossy Port- age, Seeing Portage, etc. The particular significance of the last named is that from the portage thus known Brassy Hill, a notable landmark, is seen for the first time on the way to the Bay (PI. VIII, fig. 2). This hill, which is also responsible for the name of the river, is a remarkable gravelly elevation 390 feet high and three-quarters of a mile cast of the river. As it is the highest point of land anywhere in the whole region between Lake Winnipeg and Hudson Bay, the natives naturally regard it as a veritable mountain. About 15 miles below the ' Hill,'' Rock Portage, the last on the route, is reached. Here a large flat rock divides the channel, and on each side is a fall of about 5 feet. Boats and baggage are carried over the rock. The Hudson's Ba}r Company formerly maintained a trading post near this point, but abandoned it many years ago. Between Brassy Hill and Rock Portage banks of clay gradually make their appearance. These, at first low, increase in size and in the vicing of the Rock have attained considerable height (PI. IX, fig. 1). From this point to Hudson Bay the character of the country and of the river remains much the same. The clayey banks continue on both sides nearly all the way and vary from a few feet to two hundred in height. They are marked by numerous gullies, cut b}T the many small streams that enter the main river, and, owing to frequent landslides, are continually giving way, precipitating uprooted trees into the river (PI. IX, fig. 2). In many places they are covered with a rank growth of willows and grasses, amid which are various orchids, violets, polygonums, and other small plants. Several species of scouring-rush {Equisetiwm) grow abundantly in the shallow water and often on the banks. Along upper Hill River sweet gale {Myrica gale) is common, and at the, mouth of Fox River, 30 miles below Rock Portage, buckthorn (Rha/m,us alnifoUa), honey- suckle {Lonicera glaucescens), silverberry {JElaeagnus argentea), small- flowered viburnum ( Viburnum pauciflorwm), and Canadian buffalo- berry {Lepargyr&a canadensis) were collected. None but the last two were noted farther north. Banksian pine {Pinus divaricata) and canoe birch (Betula pa/pyrifera) also find their northern limit in this region mar the confluence of Hill and Fox rivers; and the aspen poplar (Populvs tremuloides) was not noted beyond this point, though it may possibly extend farther north on this route. In some places Nortft American Fauna, No. 22. Plate VIII. Fig. 1.— White Mud Rapid, Hill River. Fig. 2.— Hill River near Seeing Portage; Brassy Hill in Distance. North American Fauna, No. 22. Plate IX. Fig. 1.— Left Bank of Hill River from Rock Portage. Fig. 2.— Clay Banks, Lower Hill River. October, 1902] ACCOUNT OF REGION TRAVERSED. 17 the valley of Hill River is narrow and the view of the traveler is con- fined to the immediate banks; in others it is broad, and its gradual, well-wooded slopes afford more extended prospects. The stream resulting from the junction of Hill and Fox rivers is called Steel River. After a course of 30 miles this in turn unites with the Shamattawa to form what is known as Hayes River, a broad, shal- low stream on whose shores gravelly beaches, absent on the deep and narrow Steel River, alternate with high clay banks. In the next 50 miles the character of the country varies but little. Back from the river are mossy swamps, which support a growth of black crowberry {Empetrum nigrum), Labrador tea (Zedum), dwarf birch (Betida glandvlosa), and associated species. Spruces and tamaracks are rather stunted. This characteristic Hudsonian country first makes its appearance in large areas on the east side of Swampy Lake, about 100 miles from Hudson Bay in a direct line, and is probably continu- ous from that point to the Bay, although for some distance the imme- diate banks of the river continue to show a more southern element. In the lower part of Hayes River are a number of low, sandy islands nearly devoid of vegetation from being- overflowed and ice swept dur- ing a large part of the year. Arctic terns and semipalmated plovers, which breed abundantly on some of the islands, were here met with for the first time. Six miles from the point where Hayes River empties into the Bay is York Factory, a post of the Hudson's Bay Company. It is on a strip of land, here 3 or 4 miles wide, lying between the Hayes and the Nelson. In the old days it was an important and well-peopled post, and was formerly the base from which all the supplies for the great interior region, brought from England by fleets of sailing vessels, were distributed by boats. The ground is low and swampy and is covered by the usual rather stunted growth of spruces, tamaracks, and thickets of willow. The soil is of a spongy character and remains frozen a few feet below the surface throughout the year.'- The woods extend about 5 miles beyond the post and are succeeded by a mile of very wet marsh inter- sected by many sloughs and channels. This marsh, which is called Point of Marsh or Beacon Point, supports a rank growth of grass and water plants, among which bog-bean [Menyanthes trifoliate!) and various species of pondweed (Potamogeton) are especially conspicuous. "Concerning this subject Richardson says: "At York Factory, on Hudson's Bay, in lat. 57°, in October, 1835, recent frosts had penetrated eight inches into the soil; the thaw due to the summer heat extended twenty-eight inches beyond this, beneath which a frozen bed seventeen and a half feet thick reposed on thawed mud which had a temperature of 33° F. The mean annual heat of this place is 25 \ F., being equal to that of Fort Simpson, which lies five degrees further north." Arctic Searching Expedition, p. 217, 1851. (See also Richardson, Edin. New Phil. Joum., XXX, p. 117, Jan., 1841.) 7165— No. 22 2 18 NOETH AMERICAN FAUNA. [No. 22, Various ducks and marsh sparrows and the elusive yellow rail find here a congenial habitat, and here, during their semiannual migra- tions, the various geese, ducks, and shore birds which breed in myriads to the northward stop for rest and food. Mosquitos become more abundant as the Bay is neared and are extremelv troublesome at this point. During our stay at York Factory — July 11 to 17 — collecting was difficult, owing to the almost incessant rain. More time was needed, but the short season and the distance still to be covered impelled us to proceed. Temporarily abandoning our canoe, therefore, we left in a sailboat for Fort Churchill, 150 miles up the coast. Contrary winds and periods of calm conspired to delay us, and the trip occupied six days. On the afternoon of the second day, being- unable to proceed, we pushed in as far as possible toward the shore at high tide, and during the ebb were able to go ashore by taking a 3-mile walk over the bouldery, weed-strewn beach, where, on every hand, flocks of shore birds of various species were hastity seeking a feeding place on the broad expanse left bare by the ebbing tide. On reaching the shore we found the Barren Grounds on a small scale lying before us. Gravelly ridges, the remains of old sea beaches, extended in various directions at a few feet above the general level, the intervening depressions occupied by small ponds or marshes. Oc- casional stunted spruces on the ridges and dwarf birches and straggling willows on the lower ground were the only fair-sized shrubs, though various small shrubby plants were abundant. Hundreds of curlews, godwits, phalaropes, plovers, and sandpipers of different species -swam or waded about the shallow ponds in their never-ending search for food. A den on a gravelly hillock a foot or two higher than the gen- eral level was occupied by a litter of half -grown Arctic foxes, and not far away was seen a pair of willow ptarmigan with young just able to fly. These patches of tundra are found all along the coast between York Factory and Fort Churchill. They seem to be roughly semi- circular in shape, the woods that bound them extending much nearer the coast on the banks of the rivers than elsewhere. At the point where we landed, between Stony and Owl rivers, the forest was just visible from the shore of the Ba}^. Similar conditions are said to exist farther south toward the Severn, though in all probability fewer Barren Ground animals are found in that region. No other stop was made until we reached the mouth of Churchill River. T -e the physiographic conditions are different from those found at any other points visited on the shore of the Bay. A ridge of greenish-gray sandstone or quartzite (PI. XI, fig. 1) extends to the coast on each side of Churchill River, and on the eastern side stretches eastward along the coast several miles toward Cape Churchill. These North Ameiican Fauna, No. 22 Plate X. ^^^^^Mf-^-H^. 1^^^a ^V ««m Jbe ^■0 ■••■ ^^^^^^5 n*^ - §5*9 '.' ?w^ HMC**5s8jE '•*!►*► ■. ^i^r - Fig. 1 .— Fort Churchill. Fig. 2.— Meadows, Looking Southwest from Fort Churchill. Habitat of Calcarius pictus. Fig. 3.— Seal iErignathus barbatusj, Fort Churchill.. North American Fauna, No. 22. Plate XI / /■■* Fig. 1.— Ledge of Quartzite at Fort Churchill. Fig. 2.— Low Tide at Fort Churchill. Feeding ground of various shore birds. October, 1902.] ACCOUNT OF REGION TEA VERSED. 19 ridges, particularly on the western side of the Churchill, consist of a succession of rounded hills, which attain a maximum altitude of about 100 feet, and support a shrubby, herbaceous growth with many mosses. Over this rocky area are scattered numerous shallow ponds with outlets flowing to the sea through narrow ravines that are scantily clothed with dwarfed spruces and willows. In these sparsely wooded ravines the Harris sparrow was common, the parent birds accom- panying young just from the nest. A low. gravelly point extending seaward from the hills forms the western bank of the river immedi- ately at its mouth. On this point lie the ruins of old Fort Prince of Wales, destroyed by the French in 1782. The bank of the river immediately opposite is composed of high rocks rising abruptly from the water. Fort Churchill (PI. X, fig. 1) is situated on the west side of the tidal lagoon which comprises the lower part of Churchill River. It is about 4 miles from the mouth of the river. To the south and west extends a broad, level meadow, only a foot or two above high-water mark, clothed with a low, shrubby growth in which appears an occa- sional dwarfed spruce or tamarack (PI. X, tig. 2). This meadow is a favorite place for Smith longspurs and horned larks, and on its drier portions we found a few burrows of lemmings (Dicrostonyx). Numbers of seals (PI. X, fig. 3) of several species frequent the mouth of Churchill River, attracted by the abundance of fish at that point. As it was very desirable to do some collecting on the Barren Grounds. I left Fort Churchill on July 30 in a small sailboat, accom- panied by three Indians, my brother remaining at Churchill to com- plete the collection. On account of the low coast, the tide in many places going out from 6 to 8 miles (see PI. XI, fig. 2), traveling in a small boat is very difficult. We could not land except at high tide, and were obliged to embark at the same stage of water. Owing to the build of our boat, sailing was impossible unless the wind was fair or nearly so, and rowing was very difficult. On the afternoon of July 31 a few hours were spent on the shore of Button Bay. Here the spruce woods nearly reach the shore at one point. North of the woods a broad grassy plain, intersected by many channels connecting small, shallow ponds, extends for several miles along the shore. Over this area a great many shore birds and ducks were feeding, some species accompanied by young, evidently reared in the vicinity, but by far the greater number associated in large, restless flocks, showing that the southward movement had commenced. That evening we rowed several miles along the coast and encamped after dark on a small, sandy islet, just above high-water mark, where Arctic terns were breeding. The next evening our camping place was a sandy point near the mouth of Seal River, the position of which is indicated by a conspicuous rounded mound that stands near its banks. 20 NORTH AMERICAN FAUNA. [No. 22, Along- the coast here the woods are visible from the Ba}T, and scattered dwarf spruces and tamaracks extend to the shore. Before Hubbart Point is reached, however, the tree limit curves inland so rapidly that the forest disappears from view altogether, although, according to Tyrrell, it can be seen with a glass from the summit of Hubbart Point. a This point, which we passed on the afternoon of August 2, is a high, grassy headland used as a burial place by the Eskimos, and is the most conspicuous landmark on this part of the coast, the mound near Seal River, just mentioned, being next in importance. Egg Island, which is mentioned as a breeding place for many sea birds, is apparently not conspicuous, for we failed to identify it either time we passed, prob- ably being too far off shore to see it. At dark on August 2 we anchored behind a small, rocky islet somewhat north of Egg Island, and at day- light next morning were again on our way. By noon we had reached a sandy point near Thlewiaza River, which proved so favorable a spot for collecting that I remained there several days (PI. XII, tig. 1). From the shore to a number of rocky and gravelly ridges a few feet in height, which were several miles inland, the country was nearly level, and was mostly wet and tilled with small hummocks. Near the shore were many broad, shallow ponds and muddy flats. Occasional dry areas, apparently raised sea beaches, were covered with rounded boulders of various sizes, and were inhab- ited by lemmings of the genus Lemmus, the burrows of which also occurred in the drier portions of the adjacent grassy meadows. A large species of meadow mouse {Microttis) was also found here, but was more abundant in the patches of coarse beach grass {Elymus mollis) which grew on the sandy ridges near the shore. On the gravelly ridges back from the immediate shore, pied lemmings (Dicrostony.i) were fairly abundant, and a number were secured. Dwarf shrubs, none of them exceeding a few inches in height, abounded; the most conspicuous were black crowberry (jEinpetrum nigrum), dwarf birch (JBetula ncma), Labrador tea {Ledum palustre), and several species of dwarf willows, including Salix angla/rum and S. phylicifolia. The scene was one of absorbing interest. On the beach and mud flats and about the shallow ponds thousands of shore birds of a dozen species circled and fed, the larger kinds, mainly Hudsonian curlews and god- wits, keeping at a little distance, the smaller kinds almost oblivious of my presence. In the deeper ponds among the ridges back from shore red-throated and Pacific loons, which later made night hideous by their cries, were feeding their unfledged young. Pomarine and parasitic jaegers harried about the tundra or sat motionless on the knolls, apparently asleep. Willow ptarmigan led their broods about in search of food, and horned larks, Lapland longspurs, tree and savanna spar- rows, and redpolls flitted from boulder to boulder. "Ann. Kept. Can. Geol. Surv., 1896 (new ser.), IX, p. 90F. (1897). North American Fauna, No. 22 Plate XII Fig. 1.— Camp on Barren Grounds, 50 Miles South of Cape Eskimo. Fig. 2.— Burrow of Lemming (Lemmus trimucronatus), Barren Grounds. North American Fauna, No. 2 Fig. 1.— Camp on Barren Grounds, 25 Miles South of Cape Eskimo. Fig. 2.— Eskimo Guide at Northern Camp. OCTORER.1902.] ACCOUNT OF REGION TRAVERSED. 21 Leaving on the afternoon of August 8, I pushed northward, accom- panied by an Eskimo (PI. XIII, fig. 2) who had promised to guide me to a place where ground squirrels could be secured. After a great deal of hard work and exposure we landed, on the evening of August it, in a small, shallow bay at the mouth of a stream about 25 miles south of Cape Eskimo (PI. XIII, fig. 1). Here I remained until August 13. The country was similar to that farther south and supported a similar flora, but wet ground was less common and sandy ridges were more frequent. The Barren Ground caribou had commenced their southward movement and one was secured. Some of the shore birds had departed, and the daily lessening numbers of other species had an obvious significance. On the sandy ridges and hillocks were scatter- ing burrows of ground squirrels, but the absence of colonies indicated that 1 had reached only the border of their range. Two polar bears were seen in the vicinity, but we were unable to secure either. The distance to be covered and the difficulties of navigation to be overcome admonished me that my return journey must be begun, and on the morning of August 13 I started southward. The wind was against us the first day. and at the close of the second we had pro- gressed no farther than my former camp, about 25 miles. Two more days of very laborious work took us a few miles below Hubbart Point. Here willows attain a respectable size, and on the boulder- covered ridges the buffalo-berry {Lepargyraea canadensis) is a con- spicuous shrub. A few Arctic hares were found here, and on the morning of August IT two were secured. On August 17 and 18 slow progress was made, and we were only able to reach a point about 15 miles above Seal River, the high mound near its mouth being visible from the ridges close to camp. Tama- racks grew in the more sheltered places, and the edge of the forest was only a few miles back from the coast. On the morning of August 19 a fair, strong wind that lasted until noon carried us within sight of the Beacon and the high rocks near Fort Churchill, which we reached that evening. During my absence my brother had made a good collection, and we left Fort Churchill on the afternoon of August 21, arriving at York Factory on the evening of August 26. In the course of this uneventful voyage we spent a few hours ashore on the afternoon of August 21 about 20 miles below Cape Churchill, where the conditions were somewhat similar to those between Stony and Owl rivers (see p. 18). Lemmings (Dierostonyx) inhabited the sandy ridges near the shore. AVe left York Factory on the afternoon of August 28 in our canoe, which was loaded with the outfit and the entire summer's collection, and, making further collections on the way, arrived at Norway House September 16, after a very laborious but pleasant trip. 22 NORTH AMERICAN FAUNA. [No. 22, Hayes, Steel,- and Hill rivers as far as the Rock Portage were ascended by tracking — the men walking along the shore pulling the canoe by a line, while our efforts were directed toward steering and avoiding the rocks, though we sometimes relieved them on the line. The passage of the various rapids on Hill River was accomplished with much difficulty. Many we were able to ascend by poling and paddling; at others it was necessary to wade waist deep in the seething water, dragging the canoe by hand (PI. XIV, fig. 1); and often neither of these methods was possible and we were obliged to unload the boat and carry both canoe and baggage around the rapids. To avoid the ascent of the river through Hell Gate Gorge (PI. XIV, fig. 2), we made a detour, as is usual on the upstream journey. Leaving Pine Lake by a narrow passage, we followed the windings of a tortuous lake for several miles in a direction approximately parallel to the course of the river, and then made a portage over a low divide to the shore of an arm of one of the lakes a short distance above Hell Gate Rapids. About twenty -five portages in all were required on the return journey, during which we retraced, with the exception of the slight deviation just mentioned, the exact route followed on our northward journey. The trip down Lake Winnipeg was made by steamer, and we arrived at Winnipeg September 22. LIFE ZONES OF THE REGION. Our route from Winnipeg to the northernmost point reached, a short distance south of Cape Eskimo, passed successively through the Canadian and Hudsonian zones and entered well into the Arctic. In the fauna and flora of Winnipeg the Canadian element seems to predominate, though the presence of Quercus macrocarpa, Acer negundo, and Ulmus americanus among trees, Blarina hrevicauda among mammals, and Zenaidura macroura and Icterus galbi/la among birds, indicate that there is a strong tinge of Transition. It is probable that this Transition element disappears a short distance to the north- ward, but no very definite data regarding the country bordering Lake Winnipeg is available. In the region about Norway House the fauna is pure Canadian, as evidenced by the presence of Paras hudsonicus, Perisoreus canadensis, Hylocichla a. swainsoni, Canachil* s canadensis, Picoides arefieas. Regal iik satrapa, Zonotrichia albicollis, Mustela jyennanti, Mustela a/men cana, Lynx canadensis, and other character- istic Canadian species. In the cleared and burnt tracts Chordeiles virginianus is common. Between Norway House and Swamp}^ Lake there is little change in the fauna. On the eastern shore of Swampy Lake the true Hudsonian begins and is probably continuous to Hudson Bay. Here Zonoti ichia albicollis is gradually replaced by Z. leucophrys, Chordeiles occurs only as a straggler, and Peromyscus is much reduced in numbers. This North American Fauna, No. 22. Plate XIV. •«i, !si ■tyfoj. k IS l^AttJkJifcAL MkAi x Gk ft™ _ Fig. 1.— Dragging Canoe up Rapid, Trout River. Fig. 2.— Hell Gate Gorge, Looking Up. October, 1902.] PREVIOUS WORK. 23 area appears to be a sort of coastal plain, and occupies a strip at this point about 100 miles wide on the borders of the Bay. It is compara- tively level, and the beds of the streams are free from rocks. The spruce, tamarack, balsam poplar, aspen poplar, Banksian pine, and canoe birch, which have formed the bulk of the forest, are here reduced in size, and the last three practically disappear near this point. These conditions prevail until the Bay is reached at the mouth of Hayes River. How far this Hudsonian strip extends southward along the coast of Hudson Bay is not apparent, but the presence at Moose Factory of Am/petis cedrorum,, Dendroica maculosa, Wilsonia cana- densis, Sialia sialis, Oarpodaeus purpv/r&us, and Condylura cristata would seem to indicate that at that point the Canadian element must predominate over the Hudsonian. To the northward from York Factory along the coast, patches of tundra, where the fauna is typi- cally Arctic, are first found not far above Nelson River. Similar barren areas occur on the coast between York Factory and Severn River, but too little is known about the fauna of that region to deter- mine whether the Arctic or the Hudsonian element predominates. These barren areas, from York Factory as far north as Fort Churchill, lie mainly between the mouths of the numerous rivers. Above Cape Churchill the Arctic Zone is continuous on the coast, except in the sparsely wooded area about the mouth of Churchill River and on part of the shores of Button Baj, where a considerable admixture of Hud- sonian forms occurs. Among the birds and mammals which breed in the Arctic zone may be mentioned various species of Tringa, Phalaropus lobatux, Lagopus lagopus, Nurnerums hudsonicits, Limosa ]i;ijmZoological appendix by Joseph Sabine. 1 vol., 4to, pp. i-xv, 1-783. London. A few species from York Factory mentioned in the appendix. The narra- tive contains a few references to the natural history of Keewatin. 1824. Lyon, G. F. The Private Journal of Captain G. F. Lyon, of H. M. S. Hecla, during the recent Voyage of Discovery under Captain Parry. [Parry's Second Voyage.] 12mo. London. Contains numerous notes on natural history, relating especially to the vicinity of Winter Island and Igloolik, Melville Peninsula. 18241. Macgillivray, William. Description, Characters, and Synonyms of the Dif- ferent Species of the Genus Larus, with critical and explanatory Remarks. Vertebrata, by Edward Sabine. Collections and observations made mainly about Melville Island. 1825. Lyon, G. F. A Brief Narrative of an unsuccessful attempt to reach Repulse Bay through Sir Thomas Roe's "Welcome," in His Majesty's Ship Griper, in the year MDCCCXXIV. 8vo, pp. 198. London. Contains a few notes on the natural history of Southampton Island and of the mainland in the vicinity of Cape Fullerton. 1825 ( 1827). Richardson, John. Appendix to Captain Parry's Journal of a Second Voyage for the discovery of a north-west passage from the Atlantic to the Pacific. 4to, pp. 1-432. London. > Zoological Appendix. Account of the Quadrupeds and Birds, by John Richardson. Refers mainly to natural history of Melville Peninsula, particularly Winter Island and Igloolik. 1826. Parry, William Edward, and Ross, James Clark. Journal of a Third Voy- age for the discovery of a North-west passage from the Atlantic to the Pacific, performed in the years 1824-25, by William Edward Parry. 1 vol., 4to, pp. i-xxviii, 1-186, 1-151. London. Appendix, Zoology. By Lieut. James Clark Ross. The zoological appendix refers mainly to the natural history of Port Bowen, where the exj:>edition wintered, and other points about Prince Regent Inlet. A few notes on natural history occur in the narrative. 1828. Richardson, John. Short Characters of a few quadrupeds procured on Cap- tain Franklin's late expedition. f a Second Voyage in search of a North-west Passage, etc. By Sir John Ross. 1 vol., 4to, pp. i-xii, 1-120, i-ciii. London. > Report on mammals and birds by J. C. Ross, pp. vii-xlv. A systematic account of the collections and observations made on Ross's second voyage. 1836. Back, George; and Richardson, John. Narrative of the Arctic Land Expe- dition to the month of the great Fish River, and along the shores of the Arctic Ocean, in the years L833, L834, and 1835; by Captain Back, R. X., Commander of the Expedition. 1 vol., 8vo, pp. i-x, 1-663. uondon. Appendix No. 1. Zoological Remarks by John Richardson, pp. 477-.11S. In the narrative mention is made of the observation of a number of species within the present boundaries of Keewatin. The Zoological Appendix con- tains a nominal list of the mammals collected, and a list of the birds in which the localities of a few species are indicated, followed by a commentary '"respecting those which are objects of chase to the Indian hunter, either for food or for the sake of their fur," with "a few hrief remarks on the speci- mens of the other species when they serve for the elucidation of doubtful points of their history." 1836. Kino, Richard. Narrative of a Journey to the shores of the Arctic Ocean in 1833, 1834, and 1835; under the command of ('apt. Back, R. N. 2 vols. 12mo. Vol. I, pp. i-ix, 1-312. Vol. II, pp. i-viii, 1-321. London. Contains many notes on the natural history of the Great Fish River and on the route from the headwaters of that river to York Factory via the Slave, Athabasca, Clearwater, Churchill, Saskatchewan, Nelson, Echimamish, Hill, Steel, and Hayes rivers. 1837. Richardson, John. Report on North American Zoology. B. hufchinsii. 1861. Barnston, George. Recollections of the Swans and Geese of Hudson's Bay. HalPs Second Arctic Expedition, Chapter VII, pp. 199-268. A concise account, prepared chiefly from official sources, of Hall's expedi- tion in northern Keewatin and adjacent lands in 1864-1869, in search of rec- ords, relics, and other traces of Sir John Franklin's last expedition. The account contains many notes on the game animals of the region. 1884. Ridgway, Robert. Note on the Anas hyperboreus, Pall., and Anser albatus, Cass. i - 1 ; i : . Report on the Doobaunt, Kazan and Ferguson Rivers ami ilif North- West Coast of Budson Bay and on two overland routes from Hudson Bay to Lake Winnipeg. Ann. Kepi. Geol. Surv. Canada, IX (new series), L896, Part F. (1897.) Chiefly geographical and geological. Notes on fauna of the region, pp. 164F to 167F, ami passim throughout the report. Octobbb,19(B.] BIBLIOGRAPHY 37 1898. Allen. .1. A. Revision of the Chickarees, or North American Red Squir- rels (Subgenus Tamiasciurus). 49-l>98. [Author's edition issued July 22, 1898. ] Sdurus hudsonicus recorded from Hudson Bay. 1898. Russell, Frank. Explorations in the Far North. Being a Report on an Expedition under the Auspices of the University of Iowa during the years 1892, '93 and '94. Published by tin- University. 8vo., pp. 290. EowaCity. Contains observations on natural history of Grand Rapids, at the mouth of the Saskatchewan River. 1898. Tyrrell, .1. \Y. Across the Sub-Arctics of Canada. 8vo, pp. 280. New- York. A narrative of a journey of exploration by way of Athabasca Pake, Telzoa River, and Chesterfield Inlet to Hudson Bay, returning by way of Fort Churchill, York Factory, Oxford House, and Norway House. Contains many notes on the natural history of the region. 1899. Howe, Reginald Heber, jr. North American Wood Frogs. miles south of Cape Eskimo August 10 to 13 we frequently heard wolves howling in pursuit of caribou, and occasion- ally saw a few of the latter, of which we killed two, a rather young- doe and a buck about three years old. The animals were evidently just commencing their usual fall jour- ney to the southward. They showed a tendency to seek the vicinity of the shore on account of the protection gained from mosquitos and other insects, which were less numerous there, owing to the wind. Even under favorable conditions they were attended by swarms of insects, and when feeding were almost constantly moving. Soon after leaving our camp on the morning of August 13 we saw several near the shore. A young buck on a point of land was approached as closely as the depth of water would permit — about 200 yards. He showed little fear, trotting along the shore abreast of our a Across the Sub-Arctics of Canada, p. 226, 1898. 42 NORTH AMERICAN FAUNA. [No. 22, boat for about a quarter of a mile. He would frequently stop and wade a short distance toward the boat, at short intervals spreading and contracting the white patch on his throat laterally into an oval disk, so abruptly as to give the appearance of flashes of light. He finally grew tired of following us and dropped behind. This was the last one seen. James Clark Ross recorded reindeer from Cape Warrender, north shore of Barrow Strait, and from the coast of North Somerset, a and observed them in great numbers on the Isthmus of Boothia.* Dr. Rae observed them migrating northward about the 1st of March, near Repulse Bay,c and found them on the west coast of Melville Peninsula as far as Fraser Bay.'7 Lyon recorded them from Duke of York Ba}r, Southampton Island/ Schwatka's party killed large numbers between Camp Daly and King William Land in 1879 and 1880./- During the summer of 1893 the Tyrrell brothers, of the Canadian Geological Sur- vey, saw on the shores of Carey Lake, about 150 miles northwest of Fort Churchill, a herd which they estimated to contain from one to two hundred thousand individuals.'7 On their exploring trip northward through the interior of Keewatin, in 1891, they first met with Barren Ground caribou, near Ennadai Lake, on August 1-1. The animals were then moving southward in large numbers. h The southern range of Barren Ground caribou, on the west coast of Hudson Bay, ma}r be said to be limited by Churchill River. Even in former 3Tears these caribou were seldom known to cross that river,*' and they are still killed within a few miles of Fort Churchill. Farther inland they reach the south end of Reindeer Lake/ Description. — Adult male in summer pelage, killed on the Barren Grounds about 25 miles south of Cape Eskimo August 10: General color of upperparts and head dull brown; face dull reddish brown; legs dusky brown with an indistinct ashy stripe on inner side of fore leg, and of hind leg below the heel; a lateral stripe of dusky brown where the hairs of the belly and sides meet, separated from the color of upperparts by an indistinct ashy stripe; chest dusky; belly and ven- tral surface of tail white; a small white patch on rump, divided by a narrow stripe of brown extending from dorsal area to tip of tail. A white disk on throat 15 inches long and (when spread) about 10 inches « Parry's Third Voyage, Appendix, p. 94, 1826. ^ Appendix to Ross's Second Voyage, p. xvii, 1835. c Narrative of an Expedition to the Shores of the Arctic Sea, p. 93, 1850. d DM., p. 149, 1850. e Lyon's Private Journal, p. 46, 1824. /Gilder, Schwatka's Search, Introduction, p. viii, and elsewhere, 1881. ?Ann. Rept. Can. Geol. Surv., 1896, IX (new ser.), p. 165F (1898). /'Ibid., p. 19F (1898). « Hearne, Journey * * * to the Northern Ocean, p. 225, 1795. J Tyrrell, Forest and Stream, XLII1, No. 4, p. 70, July 28, 1894. October, 1902.] MAMMALS OF KEEWATTN. 43 wide, consisting of hairs 5 inches in Length; an indistinct stripe of ashy extending from this white area forward to chin; feet around edge of hoof white; oose and edges of lips whitish; ears mainly ashy. The winter skins seen were more or less suffused with white or hoary, which in some eases was the predominating color. Alces americanus Jardine. Eastern Moose. This species occurs in suitable places throughout the region traversed from Lake Winnipeg nearly to Hudson Bay. While ascending the Echimamish, which is a noted locality for moose, we frequently saw places where the animals had crossed the stream, and the bordering swamps were intersected by a network of their tracks. Many of the tracks in the bed of the stream were so recent that the mud had not 3Tet settled; but no animals were seen, though we were constantly on the lookout for them. A young one was killed by a party of Indians at the outlet of the Echimamish a few days before we ascended the river, and the cranium of a female, probably killed during the pre- ceding winter, was found at the middle dam. The extensive swamps below Robinson Portage are also much frequented by the animals, one of which was trailed for some distance b}T my Indian guides. At Oxford House, 60 or 70 miles beyond Robinson Portage, the moose was formerly almost unknown, according to information received independently from several officers of the Hudson's Bay Company, but is extending its range toward Hudson Bay and is now frequently killed near that post. On Steel River, as we were ascend- ing it September 1, we saw a fresh track at the mouth of a small stream a few miles below Fox River, and at York Factory I was shown a skin which had been brought from Shamattawa River, where the York Factory Indians now go regularly to hunt moose. Farther inland the moose ranges to the northwest. I saw a head at Norway" House from the vicinity of Split Lake, and J. B. Tyrrell met with moose on Stone River about 40() miles west of Fort Churchill. a Ovibos moschatus (Zimm.). Musk-Ox. The musk-ox was first described from the region between Seal and Churchill rivers and formerly ranged in winter more or less regularly to about that latitude. Hearne saw its tracks within a few miles of Fort Churchill, and on his first journey toward the northwest met with it within a hundred miles of that place.6 Edward Sabine recorded musk-oxen from Melville Island, where they arrived in Ma}r from the southward. They crossed on the frozen sea, and recrossed it on leaving in September/ Ross recorded them from Felix Harbor, « Ann. Rept. Can. Geol. Surv., 1896, IX (new ser.), 165F (1897). b Journey * * * to the Northern Ocean, p. 135, 1795. cSuppl. to Appendix to Parry's First Voyage, p. clxxxix, 1825. 44 NORTH AMERICAN FAUNA. [No. 22, Boothia." Dr. Rao found them in the vicinity of Repulse Bay/' east of which they seem never to have been recorded. Frederick Schwatka found a herd between Wager and Back rivers in the winter of 1879/ Tyrrell saw fresh skins in the possession of the Eskimos near the head of Chesterfield Inlet in the summer of I893.d A skull that had been brought by Eskimos from the vicinity of the head of Chestertield Inlet was obtained by us at Fort Churchill. The tips of the horns of this specimen, an adult male, curve forward in nearly parallel planes, instead of diverging, as is usually the ease. Through the kindness of Dr. Alexander Milne, of York Factory, I obtained what is probably the most southern authentic record of the occurrence of this animal. Several times during the summer of 1897 parties of Indians reported seeing a pair of musk-oxen on the ' Bar- rens' about halfway between York Factory and Fort Churchill. The male was finally killed in August and the head was brought by the Indians to York Factory. This pair had probably wandered beyond the limits of their normal range during the preceding winter, and for some reason failed to return northward. The Eskimos who trade at Fort Churchill hunt the musk-ox in the Barren Grounds several days1 journey northwestward from Cape Eskimo. Sciuropterus sabrinus (Shaw). Hudson Bay Flying Squirrel. Said to be found throughout the region traversed between Lake Winnipeg and Hudson Bay. It has been recorded by Bell from Nelson River House, on Churchill River/ and by Bangs from Moose Factory. -f We obtained several hunters' skins at Norway House and one at Oxford House, and learned that the species is rather common on Shamattawa River. It sometimes becomes a pest to trappers on account of the frequenc}7 with which it is caught in traps set for mar- tens and other fur-bearing animals. Several winter skins from the vicinity of Norway House differ as follows from skins of Sciuropterus s. macrotis in corresponding pelage from New England: Upperparts slightly darker (loss yellowish); tail with duskier color toward tip; face and cheeks darker; foot similar in color, but much more heavily furred; color beneath not noticeably different; hind foot (measured dry), sabrinus, 38-40 nun.; macrotis, about 30 mm. Breadth of tail with hairs spread naturally, sabrinus, 60-65 mm.; macrotis, about 45 nun. These specimens from Norway House and Oxford House can probably be considered typical sri/ins, which was described from Severn River. An imperfect skin in the a Ross's Second Voyage, p. :'.:;7, is:;:,. ''Narrative of an Expedition to the Arctic Sea, \>. 41), 1850. cGilder, Schwatka' s Search, \>. \>. II. p. 48DD (1885). /Proc. Biol. Soc. Wash., X, p. 163, L896. October. 1902.] MAMMALS OF KKKWAT1N. 45^ National Museum from Hudson Bay closely resembles those from Norway House, but is slightly more tinged with yellowish brown beneath. An albinistic specimen from Norway House is nearly white above, slightly tinged with light brown, with a brownish area on the middle of the back: the tail is very light brownish above, and lighter, about normal, beneath. No skulls from the Hudson Bay region are available for comparison. Sciurus hudsonicus (Erxleben). Hudson Bay Red Squirrel. W<> found the red squirrel abundant and generally distributed throughout the region to within a few miles of York Factory, where the trees dwindle to such an insignificant size as not to afford the ani- mals a congenial home. At Fort Churchill I saw a specimen which had been secured a few miles up the river, where good-sized spruce trees occur and the animals are said to be fairly abundant. An adult male taken at Norway House June 18 is in nearly full winter pelage, the summer coat just beginning to appear in patches on the face and back, the lower parts still showing- the heav3T grayish vermiculations, and the red median dorsal stripe of the winter pelage being- present. The tail is concoloi-with the back, and is edged with yellowish gray. All the other specimens taken at Norway House and at various points between that post and Oxford House late in June and early in July, and on the return trip between Steel Kiver and Norway House during the first half of September, are in summer pelage and appear to repre- sent two phases of color. By far the greater -number are very dark reddish above, the red suffusion tingeing the lower parts quite appre- ciably in many cases; the tails are bordered with grayish or reddish indiscriminately. Tamias striatus lysteri (Richardson). Northeastern Chipmunk. A specimen collected by C. Drexler at James Bay is recorded in the U. S. National Museum catalogue. It can not now be found, but notes regarding it. made years ago by Dr. C. Hart Merriam, indicate that it is referable to the present form. Eutamius neglectus (Allen). Lake Superior Chipmunk. "We obtained about a dozen specimens which were collected by Mr. William Campbell at Oxford House during July. August, and the early part of September. Chipmunks were reported to be rather common about Oxford Lake, but we failed to see any at that point, or at any other on our entire trip except the shore of Pine Lake, where we observed one September 13. The specimens secured were preserved in formalin and skinned on our return in September. The average measurements of <; are as follows: Total length 217; tail vertebra- 101; hind foot 31.a This series agrees essentially with a series taken by Gen-it S. "All measurement* are in millimeters unless otherwise stated. 46 NORTH AMERICAN FAUNA [No. 22, Miller, jr., in September and the early part of October atNepigon and Peninsula Harbor, Ontario, which are in the type region of neglecbus. The Oxford House series shows some seasonal variation, those taken in July having grayer rumps than the September specimens. Dr. Bell reported ' Tamias quadnri/vittatus'' to be common along the Nelson and Churchill rivers," but it is probable that it is not found on the lower portions of these rivers. Spermophilus parryi (Richardson). Hudson Ba}T Spermophile. Arctomys Parryii Rich., Appendix to Parry' s Second Voyage, p. 316, 1825 (1827). The name empetra, usually since 1877 applied to this species, was apparently based on a specimen of the Canadian form of Arctorm/8 monax (see p. 47), and will have to be replaced by parryi of Richard- son, based on specimens collected at Five Hawser Bay, Lyon Inlet, Melville Peninsula, on Parry's second voyage, which seems to be the next available name. Ground squirrels have been stated to inhabit the coast of Hudson Bay south to Fort Churchill, but I ascertained that the animal does not occur on the coast south of a point about 25 miles below Cape Eskimo — about 150 miles to the northward of Fort Churchill. I was unable to find any in the vicinity of my camp near Thlewiaza River. On starting northward in search of them I was fortunate in securing as guide one of the most intelligent of the Eskimos of the region, who was perfectly familiar with the entire coast and who promised to take me to the nearest point at which the animals were to be found. We left camp near the Thlewiaza on the afternoon of August 8 and on the evening of the next day succeeded in reaching our objective point — the mouth of a stream which enters the sea at the head of a shallow bay thought to be about 25 miles south of Cape Eskimo. One of the animals was seen a few minutes after we landed, but they proved to be rather rare, though the Eskimo gave me to understand that farther north they were more common. They inhabited the scattered sandy ridges and hillocks, but with the exception mentioned none were seen abroad, and careful trapping during the next three days at all the burrows found yielded only four specimens. The apparent inactivity of the animals and the fact that most of the burrows were closed from the inside led me to conclude (somewhat hastily) that the animals hud commenced to hibernate, but from the fact that they have been taken at points farther north much later in the season it is probable that such was not the case. The burrows usually had several entrances, and, judging from the amount of earth removed, were quite extensive. The foiu1 specimens taken agree closely in color and may be thus described: Fur of back, sides, breast, and belly dusky plumbeous at «Rept, Prog Can. Geol. Surv. 1882-3-4, App. II, p. 481)1) (1885). October, 1902.] MAMMALS OF KEEWATIN. 47 base, on breast, belly, and sides tipped with rather bright rufous; Legs, feet, and throat rufous to base of hail's, this color deepest on upper side i^ legs; sides and thighs more or less flecked with black hairs; fur on back from nape to base of tail with a broad zone of yellowish-brown, which forms the ground color, succeeded by a subter- minal zone of yellowish gray and tipped with black. The black tips and tht> wholly black hairs which are interspersed through the fur predominate in places, causing the grayish zone of color to be broken up into more or less distinct spots; top and sides of head varying from light rufous to dark chestnut rufous, deepest on cheeks and top of head and interspersed with many black hairs; ears light rufous; hairs of upper side of tail dark gray at base, this color succeeded by a narrow zone of black, a narrow zone of dull yellowish, and a broad zone of black, and narrowly tipped with rufous; pencil of tail, com- prising about a third of its length, nearly all black; most of under side of tail bright rufous. The three males average: Total length 88(5. <5; tail vertebrae 117; hind foot 64.3. The female measured: 392; 114; 57. Arctomys monax empetra (Pallas). Canadian Woodchuck. Mus empetra Pallas, Nova' Species Quad, e Glirium Online, p. 75, 1778. The name Mus < mpt tra of Pallas, usually of late years applied in a general way to the spermophiles inhabiting the Barren Grounds and the northern Rocky Mountains, was evidently based on a specimen of the Canadian woodchuck, as has been recognized by Sabine, a Rich- ardson.-'and others. Pallas cites (as Marmota quehekana) the 'Que- bec Marmot' of Pennant,' qualifying the reference to Pennant's figure with the word bona. Pennant's "Quebec Marmot," apparently based on a live specimen in the possession of Mr. Brooks, is unquestionably a woodchuck.'7 Forster's 'Quebec Marmot,' e which Pallas also cites, is as certainly a spermophile. Pallas's description of Mm empetra follows his citations. Dr. C. Hart Merriam and I have gone care- fully over this description with specimens of both Arctomys and SpermophUus from Hudson Ba}T before us, and he agrees with me that the description applies to the woodchuck. The description of the color of those parts which are nkost unlike in the two animals (the head and feet) applies unquestionably to the woodchuck, and the entire description tits that animal far better than it does the spermo- phile. The 'wenusk,' as the animal is called in the Hudson Bay country, "Trans. Linn. Sue. London, XIII, p. 584, 1822. 6 Fauna Boreali- Americana, I, p. 147, 1829. 'Synopsis of Quad., p. 270, pi. 24, fig. 2. tf Pennant, Hist. Quad., 3d ed., 1793, p. 129, says: "It has lately been described by Pallas under the name of Mm empetra." «Phil. Trans., LXII, p. 378, 1772. 48 NOKTH AMERICAN FAUNA. [No. 22, is apparently found throughout the region as far north as York Fac- tory, but is said to be more common to the southward. A hunters' skin, taken near Oxford House in the spring of 1900, was obtained from Mr. William Campbell, who reported the animal as not common in the vicinity. At York Factory we obtained an immature specimen and a hunters' skin, taken on Hayes River. Several skins from Trout Lake, where the animals are said to be numerous, were also obtained. The skins secured, representing immature individuals and adults of both sexes, vary but little in coloration. The prevailing color of the upperparts is rufous, the subapical zone of black being restricted and contributing little to the general color; the lower parts are rufous, this color varying somewhat in intensity; the tail is very dark brown or black, the hairs tipped with rufous; the feet are nearly black. The onl}r skull available for comparison was taken at Oxford House June 10, 1901. It is that of a female, and though the teeth are not much worn, the development of the occipital crest and the general appearance indicate maturity. It is much smaller than skulls of wood- chucks of about the same age from Washington, D. C, and Maryland (the type locality of monax), and shows other differences as follows: Rostrum less deflected, the nasals flatter and more narrowed poste- riority; basal portion of zygomatic process of squamosal broader and flatter, its posterior edge, when viewed from above, being practically continuous with the general curve of the zygoma and lacking the con- spicuous concavity usually seen in typical monax. The teeth are proportionally weaker, the incisors especially being more slender. This skull measures as follows: Occipito-nasal length 80; zygomatic breadth 55; breadth of rostrum immediately in front of zygoma 15; length of nasals 33; breadth of posterior end of nasals 7. Allen recorded specimens of Arctomys monax from James Bay and Nelson River/' Castor canadensis Kuhl. Canadian Beaver. Owing to persistent trapping, the beaver is becoming scarce through- out the region, but skins are annually traded at all the posts visited. We saw the remains of a beaver house between Pine and Windy lakes and a comparatively recent dam on a small stream which empties into Hayes River about 15 miles above York Factory. A number of skins were seen at Fort Churchill. These had been taken on the Lower Churchill River. Several Mack pelts were among the furs at Norway House. A line large ligantentarv skeleton was obtained at York Factory from Dr. Milne, who said that the animal it belonged to had been taken in the vicinity.. Its measurements (approximate) are as follows: ".Mon. \. Am. Rod., i>. '.Mil, 1877. October, ion.-.] MAMMALS OF KKKWATIN. 49 Total length L,000; tail vertebrae 380; hindfoot L90. Skull: Occipito- nasal length FIT ; zygomatic breadth 99; interorbital constriction 25; Length of nasals.")!; breadth of nasals 24; length of upper molar scries (crowns) 30. Nine adult skulls of both sexes, collected in the vicinity of Oxford House, average as follows: Occipito-nasal length L31.3; zygomatic breadth 93.5; interorbital constriction 25; length of nasals 47.3; breadth of nasals 23.6; length of molar series (crowns) 28.2. Dr. Bell reports that a family of beavers was found by Indians on North River, a stream that flows into the Bay about 15 miles above Fort Churchill." Ilcarne relates that the Indians who accompanied him on his first exploring trip killed beavers on Seal River, the mouth of which is about 40 miles north of Fort Churchill.''' The point at which llearne seems to have crossed Seal River, and presumably where the beavers were taken, is not far from the Bay. [Mus musculus Linn. House Mouse. Referring to this species. Richardson says: "I have seen a dead mouse in the storehouse at York Factory."' J . F. Gray records a specimen of the house mouse collected by Dr. Rae at York Factory, with the remark that it was probably introduced from Europe."' The species does not seem to have ever become established, though individ- uals should apparently be landed occasionally with goods from England. We trapped large numbers of mice about the various buildings, but took only the native species.] Peromyscus canadensis umbrinus Miller. ' Clouded White-footed Mouse. We found this form common and apparently quite generally dis- tributed between Norway House and York Factory, and took speci- mensat the following localities: Norway House, Sea Falls, Echimamish River (Middle Dam and Painted Stone Portage). Robinson Portage. Pine Lake, Oxford Lake (south end), Oxford House, Trout Falls, Hill River (mouth of Fox River and near Rock Portage), Steel River, Hayes River, and York Factory. I have compared this series with the type series of /'. canadensis umbrwms from Peninsula Harbor, Ontario. The specimens of the type series are mainly in fall pelage, while most of our Keewatin specimens were taken during the summer; but enough are comparable "Kept. Prog. Can. Geol. Surv. 1882-3-4, App. II, p. 49DD ( L885. ) & Journey from Prince oHVales Fort * * * to the Northern Ocean, p. 9, 1795. •Fauna Borcali-Amerieana, I, p. 141, 1829. f the Arctic Sea, Appendix, p. 199, 1850. «Proc. Bust. Soc. Nat. Hist., 28, No. 1, p. 23, April, L897. 7165— No. 22 4 50 NORTH AMERICAN FAUNA. [No. 22, to make it reasonably certain that they are all referable to the same form. In detailed measurements the two series agree very well. Seven adults from Painted Stone Portage average: Total length 186; tail vertebrae 92; hind foot 20. Average of seven adults from Oxford House: 186, 92, 20. Average of seven adults from York Factory: 1S3, 91, 20.7. This mouse constitutes the house mouse of the region, and infests all the houses and stores within its range. At York Factory almost all our specimens were taken in or about the post buildings; the animal seemed to be rare in the surrounding swamps. [(?) Neotoma drummondi (Richardson). Drummond Wood-Rat. A wood-rat received about I860 from W. MacTavish, and supposed to have been taken at Fort Churchill," is in the U. S. National Museum. It was mounted for many years, but is now made into a study skin. The color has become so changed by exposure that no dependence can be placed on it, and the skull lacks the greater part of the braincase and is otherwise defective. Comparison of this imperfect specimen with specimens taken at Jasper House in the type region of drummondi shows no essential differences. As Jasper House seems to be the nearest point to Fort Churchill that the genus Neotoma has been recorded, and as the animal is unknown to the inhabitants of Fort Churchill, it seems most likely that the specimen came from some point in the interior, and was erroneously included in a lot of specimens from Hudson Bay.] Phenacomys mackenzii Preble. Mackenzie Phenacomys. Phenacomys mackenzii Preble, Proc. Biol. Soe. Wash., XY, p. 182, August 6, L902. (Fort Smith, Mackenzie.) Found only at Fort Churchill, where an adult female and two imma- ture individuals were secured July 20 to 28. Two of these were taken on a dry hummock in the meadow west of the post, and the third in a grassy place among the rocks. The rostral portion of a skull, with a part of the skin attached, was found in a fissure among the rocks, where the animal had evidently been eaten by a weasel. Evotomys gapperi (Vigors). Common Red-backed Mouse. We found this species to be rather common throughout the region between Norway House and Hudson Bay, and took a large series embracing specimens from the following localities: Norway House, Sea Falls, Echimamish River, Robinson Portage. Pine Lake, near head of Oxford Lake, Oxford House, Knee Lake (near outlet and on island "In a letter to Sir John Richardson, extracts of which were published, Professor Baird says: "From * * * our other northern collectors and correspondents we have ahead)' received many interesting specimens, though the best are still on the way. We have already had Neotoma cinerea from Fort Churchill." (Edin. New Phil. Journ. (newser.), XIII, p. 164, 1861.) OcroBEE.1902.] MAMMALS OF KKKWATIN. 5] near "Narrows"). Swampy Lake, Hill River (at various points), Steel River, Haves River, and York Factory. One was also taken by mv brother on Churchill River, 15 miles above Fort Churchill. Mossy spruce woods seemed to be their favorite habitat, hut we also fre- quently trapped them in deciduous or mixed woods, and occasionally in willow thickets in swamps. They inhabited the larger wooded islands in Knee Lake. On lower Hill River as well as on Steel and Hayes rivers we took only an occasional one, and at York Fac- tory a week's trapping yielded but three, all of which were caught about the post buildings. None were found at Fort Churchill, or any- where on the tundra, and several nights' careful trapping by my brother on Churchill River above the post yielded but one. Of this series of nearly seventy specimens all but six are in the normal red-backed pelage. The exceptions may be briefly referred to. One. from Robinson Portage, has a few black hairs scattered over the back, perceptibly darkening the dorsal stripe. One, from Oxford Lake, taken September L2, has the dorsal stripe dusky, sides only slightly darker than normal, and belly yellowish white. One, taken on upper Hill River September .'». has the dorsal stripe dusky brown, the belly and sides about normal, and the back nearly concolor with sides; the dorsal stripe is indicated only by a slight admixture of red. Of the three specimens taken at York Factory, one is normal; another, probably in left-over winter pelage, differs from the average only in being slightly brighter. The third is yellowish white beneath, with nearly normal sides, and dorsal stripe indicated only by a slight darkening of the dorsal area. One. taken on Churchill River, 15 miles above Fort Churchill, has the dorsal area dull red and only faintly indicated, with dull-brown sides and ashy-brown lower parts. It is immature and measures: 138; 30; 20. A skin from Fort Churchill in the L. S. National Museum (No. 420t>) has a seal-brown dorsal stripe, which extends forward nearly to the eyes, and dull, yellowish- brown lower parts and side-. Three adults from Norway House average: Total length 134; tail vertebra- 39; hind foot 18. Four from Robinson Portage average: 135. :>: 38; 18. Five from Knee Lake average: 138; 12; 19. Three from York Factory average: 1-11; 40; 19. Microtus drummondi (And. and Bach.). Drummond Vole. We found this vole rather common in suitable locations throughout the region between Norway House and Hudson Bay. It seemed most abundant in the partially cleared areas near the posts, and as it readily takes up a residence in the houses and stores, it becomes a decided pest. A large series was collected, including specimens from the fol- lowing localities: Norway House. Echimamish River, Robinson Port- age, Oxford House, Steel River, York Factory, Cape Churchill, and Fort Churchill. 52 NORTH AMERICAN FAUNA. [No. 22, This series averages a little larger than typical drurnmondi, but on the whole is fairly typical. Seven adults from Norway House aver- age: Total length 152; tail vertebra? 12; hind foot 19. Seven from Oxford House average: 153; 17; 19. Seven from Fort Churchill (selected as being nearly typical of this form): 156; 10; 19. Microtus aphorodemus sp. nov. Barren Ground Vole. Type from Barren Grounds, about 50 miles south of Cape Eskimo, Keewatin (near mouth of Thlewiaza River). £ ad. (skin and skull), No. 106422, U. S. Nat. Mus., Biological Survey collection. Collected August 5, 1900, by Edward A. Preble. Original number, 3208. General characters. — Similar to Microtus drurnmondi, but larger, with much larger and heavier skull. Color. — Upperparts dark yellowish bister, as in drurnmondi, but usually with an admixture of yellowish-tipped hairs, imparting a coarse appearance to the pelage; adults varying beneath from nearly white to light plumbeous, sometimes tinged with light brownish. Young: Varying but little from adults, lighter in color than young of drurnmondi, especially beneath. Skull. — Much larger than that of drurnmondi; rostrum propor- tionally heavier; molar series longer. Compared with that of penn- syl/vanicus the skull is much larger, less arched, and with more widely spreading zygomata; teeth about as in pennsylwanicus. Measurements. — Type: Total length 190; tail vertebrae 50; hind foot 20. Average of six adults from type locality: 182; 19; 20.3. Average of four adults from near Hubbart Point: 181; 19; 20.2. Skull of type: Basal length 2S; nasals 7.6; zygomatic breadth 16.5; mastoid breadth 12.6; alveolar length of upper molar series 7. Skull, No. 106271, 9 ad. (largest in series): Basal length 29; nasals 8.5; zygomatic breadth 17; mastoid breadth 12.5; alveolar length of upper molar series 7. lu marks. — I found these voles common on the Barren Grounds near the mouth of Thlewiaza River early in August, and collected a large series. They frequented patches of wild rye {Elymus mollis) on the sandy raised sea beaches near the shore, and burrowed exten- sively in the sand and to some extent in the semidecayed masses of seaweed which had accumulated at high-water mark. They were most active in the morning and evening, but were taken at all hours without difficulty in traps set in their well-worn trails. Their prin- cipal food seemed to be the culms of the wild rye. I found many cavities beneath boulders which they had utilized for storehouses and had tilled with the stalks of this coarse grass, cut into short sections. Along the coast north of Foil ( 'liiiichill at several points where the wild rye grew luxuriantly I found abundant traces of these voles. A series of twenty-two was collected on a point about 1<) miles north of Hubbart Point on the morning of August 16, while 1 was waiting for October. 1902.] MAMMALS OF KEEWATIN. 53 the tide to come in that we might resume our journey southward. This series includes young and adults of all ages and both sexes, the females slightly predominating. Several females contained embryos which varied in number from seven to ten. Ten specimens taken on the "Barrens/ near Cape Churchill, August •J.*), are referable to this form, and the large series of drummondi taken at Fort Churchill includes some specimens that seem to incline toward aphoroeU mus. Microtus xanthognathus (Leach). Chestnut-cheeked Vole. This species was originally described from Hudson Bay. During our trip, though constantly on the lookout for the species, we saw no traces of voles which we could attribute to xanthognathus and secured no specimens. Cones recorded Armcola xanthognathus from Fort Churchill." The fact that the table in which this specimen is listed contains only skins with chestnut cheeks, as shown by the context, makes it evident that the identification is correct, a fact further supported by the length of the hind foot (1 inch). This specimen can not now be found. Bailey records Microtus xanthognath/us from Nelson River/' Fiber zibethicus hudsonius subsp. nov. Hudson Bay Muskrat. Type from Fort Churchill, Keewatin, Canada. $ ad. (skin and skull), No. 106881, C. S. Nat. Mus., Biological Survey collection. Collected August 9, L900, by Alfred E. Preble. Original number, 3081. General character*. — Rather small, about the size of Fiber spatulatus, but differing in color and cranial characters; smaller than and other- wise different from Fiber zibethicus. Color. — Above, dull yellowish brown, darker on middle of back and head; fur of lower parts tipped with yellowish brown, throat yellow- ish white; middle of chin dusky; cheeks yellowish fawn. Compared with spatulatus, hudsonius is more yellowish brown, the buck lacking the dusky tinge of spatulatus, and the sides are uniformly lighter. Skull. — Smaller and more angular than in Fiber zibethicus, with highly developed interorbital crest and small molars, as in spatulatus, but differing from skull of spatulatus as follows: Bullae more inflated; parieto-squamosal suture irregular; lateral face of zygoma, especially jugal, broader; nasals narrower anteriorly, but otherwise similar. Measurements. — Type: Total length 539; tail vertebrae 225; hind foot 74. Average of four adults from Echimamish River: Total length 542.5; tail vertebra* 23d. 2; hind foot 75.7. Average of two adults from York Factory: Total length 556; tail vertebrae 246.5; hind foot 74.5. Skull of type: Occipito-nasal length 56; zygomatic breadth 38; inter- orbital breadth 6; greatest length of nasals 2<>; greatest breadth of nasals 9.5; length of upper molar series (crowns) 13.5. «Mon. X. Am. Rod., p. 201, 1877. (No. 8366 in table.) &N. A. Fauna, No. 17, p. 58, 1900. 54 NORTH AMERICAN FAUNA. 0*0.22, Remarks. — This species is common and quite generally distributed throughout the region between Lake Winnipeg and Hudson Hay. Though its skin is worth but a few cents in trade, the abundance of the animal and the ease with which it is trapped cause it to be much sought for, and many thousands are annually taken. While ascending the Echimamish we found muskrats abundant and tame, and we also saw many in the marshy lakes in the neighborhood of Robinson Portage. On account of the abundance of ' rats ' these two localities are favorite trapping grounds of the Indian hunters. The sweet flag (Acorns calamus), a favorite food, abounds in the lakes below Robinson Portage, and the great number of muskrat houses seen there showed the locality to be a favorite resort. We saw many musk- rats while we were descending Hill and Steel rivers. Here they live entirely in burrows in the banks, as the deep water and swift cur- rent render house building impracticable. Piles of mussel shells, showing where the animals had been feasting, were frequently noticed on the banks. The point of marsh below York Factory seemed to be the home of a considerable number of muskrats, but as it is frequently overflowed, they are said to seldom survive the winter. In the vicinity of Fort Churchill they were found in but one place — a stream entering the Churchill River a few miles above the post, and there they were rather uncommon. A winter following a dry season is said to be very destructive to muskrats. On account of the low water the animals construct their burrows and houses correspondingly low and are often forced out by floods at a season when they are unable to procure food and have no protection from their enemies. We collected specimens on the Echimamish, and at Robinson Port- age, York Factory, and Fort Churchill. Young of the year in fresh pelage are darker than adults, but the entire series we collected shows much less rich brown than zibethicus from New England. Synaptomys (Mictomys) bullatus Preble. Northern ■Lemming Volt1. Synaptomys(Mictomys)bvMatus Preble, Proc. Biol.Soc. Wash., XV, p. 181, August6, 1902. (Troul Rock, near Fort Rae, Mackenzie.) We trapped only two specimens of this lemming vole one at Nor- way House June 21 and one, a female with six embryos, in the swamps bordering Echimamish River June 25. Lemmus trimucronatus (Richardson). Back Lemming. We. found this tine species at but one locality, near the mouth of Thlewiaza River, where it was common and where a series of about seventy, comprising adults and young of both sexes, was secured August 4 to 8. A succession of low, lint, boulder-covered areas, which lay between (he shore and some shallow lagoons a few hundred yards inland, was occupied by the animals. The ground was dry and well October, 1902.] MAMMALS OF KEEWATIN. 55 covered with short, thick grass, through which their runways extended in every direction. They burrowed extensively, sometimes beneath boulders, but as often in the sides of tiny terraces or from a Hat sur- face. Their holes seemed to be connected in an endless labyrinth. We captured several by suddenly overturning some of the boulders, but most wen' taken in traps set in their well-trodden roads. They paid no attention to bait, but were readily caught in runway traps. When taken alive they showed considerable ferocity for animals of their size, snarling and biting vigorously. The breeding season seemed to be nearly over, but a few females contained from four to six embryos. The teats are eight in number, four inguinal and four pectoral. No specimens of Lemmus from the vicinity of Point Lake, the type locality of trtmucronaftus, are available for comparison, but it is nor likely that the Hudson Bay animal differs appreciably. A few speci- mens, in poor condition, from the Anderson River region, show no apparent differences. Color of adults. — Fur everywhere plumbeous at base, on head and shoulders tipprd with black and grayish in about equal proportion; lower parts and sides rusty ochraceous, this color extending forward on cheeks and lips and sometimes tingeing slightly the head and shoulders; lower part of back bright reddish brown, the fur very thick and long; tail usually distinctly bicolor, yellowish below, and dusky brown above; pencil about 9 mm. in length; feet dusky brown above, lighter below. In some of the half-grown specimens in fresh pelage the head and shoulders are considerably flecked with the color of the sides and there is an indistinct dusky stripe extending from between the ears to the middle of the back. Younger specimens are nearly unicolor throughout, the plumbeous fur tipped with yellowish brown. Measurements. — One of largest specimens: Total length 1(50; tail vertebra' 22; hind foot 20. Average of ten adults: 151; 21.8; i!i>. Skull: Average of four adults: Basal length 30.6; zygomatic breadth 21.5; interorbital breadth 3.1; mastoid breadth 15.5. Remarks. — Armcola trimucronatus was recorded from Igloolik, Mel- ville Peninsula, by Richardson/' and was said by J. C. Ross to be com- mon on the shores of Boothia Felix. h Dicrostonyx richardsoni Merriam. Richardson Lemming. Dicrostonyx richardsoni Merriam, Proc. Wash. Acad. Sci., II, ]>. 26, March 14, L900. About 120 specimens of this interesting lemming were collected. These ranged in age from young a few days old to fully adult individ- uals. About 80 were taken, mainly l>3' my brother, at Fort Churchill, a Appendix to Parry's Second Voyage, p. ::i 1, L825 ( 1827). ''Appendix to Ross's Second Voyage, p. xiv. is:;."). 56 NORTH AMERICAN FAUNA. $to.&, the type locality of the species, and the remainder at two localities on the Barren Grounds, and on the 'Barrens' below Cape Churchill. Many burrows were untenanted and there was usually nothing in their appearance to indicate the fact. On this account trapping was found to be less satisfactory in securing specimens than digging. A few minutes' digging usually disclosed whether or not a burrow was occupied. Gravelly ridges, the remains of old raised sea beaches, occur through- out the country bordering the Bay, and are found inland many miles from the present coast line. Richardson lemmings frequent mainly these ridges. Their burrows differ widely from those of Lemmtts and Microtus. Each seemed to be distinct and occupied by only a single individual, except in the case of females accompanied by young. The hole sometimes has its entrance beneath a piece of driftwood or at the base of a dwarfed spruce. It usually proceeds at an angle of about 45° for a foot or so, and then extends nearly horizontally for 2 or sometimes 3 feet to the nest of grass and moss, which occupies a circular cham- ber 4 inches in diameter. A side gallery a foot or two in length usually branches off from the main burrow not far from the nest. This is without a terminal chamber and is evidently used as a place of refuge. In this retreat we usually found the owner of the burrow if he was at home. The sand and gravel dug from the burrow is usually pushed out into :i long pile extending sometimes 2 feet from the entrance. No runways are made, even when the burrows are near soft ground; the animals evidently range indiscriminately over the ground. The only food observed in the burrows was a few leaves of bear-berry {Arctostaphylos u va-ursi). Three young at a birth seems to be the usual number. Every litter we found consisted of three, and in each pregnant female we secured were three embryos. The breeding season seemed to be nearly over, however, so that pregnant females were not common. One, captured on the Barren Grounds August 12, besides containing the usual three small embryos, was suckling three young. Several young were kept for a few days by my brother. They were very readily tamed and took food (rolled oats and crumbs of bread) within a few hours of their capture, allowing him to hold them on his hand while they ate. They sat on their haunches and held the food in their fore feet like squirrels. These young lemmings were very gentle and interesting in their ways, but the old ones fought viciously when captured, and their sharp incisors and strong jaws made them somewhat formidable. In the immediate vicinity of the post at Fort Churchill we found only a few lemmings, but the sandy ridges on tin1 south side of the river and on the point near the rums of Fort Prince of Wales proved October, 190'J.J MAMMALS OF KKHW ATIN". 57 fruitful collecting grounds. At my two camps on the Barren Grounds south of Cape Eskimo they were abundant in their favorite locations; and on a long sandy ridge below Cape Churchill, where we landed on the afternoon of August 24, we found many burrows and captured about 20 lemmings. Only one was actually seen away from its bur- row a female which, when surprised by us some distance from her home, ran into a shallow deserted burrow for refuge. rriie animals seem mainly nocturnal in their habits, though a few were taken at the mouths of holes during the daytime. They are known by the natives of Churchill as "Husky' or 'Huckey' (i. e., Eskimo) mice, because of their northern habitat. Compared with skulls of Dierostonyx hitdsonius from the coast of Labrador, those of D. richardsoni exhibit the following conspicuous differences: Braincase broader; interparietal broader and squarish in outline (in hitdsonius small and triangular); parietals broader; bullae much more inflated. In color I), richardsoni is much darken- than hudsonius, owing to a much greater proportion of chestnut- and brown-tipped hairs in the pelage. The specimens taken by us vary remarkably, but the grayest show more brown than any specimens of hudsonius examined. Topotypes of D. richardsoni in summer pelage may be described as follows: Fur light plumbeous at base; that of back and sides tipped with chestnut or rich yellowish brown, occasionally with these colors and gray, sides usually lighter than back. A dusky stripe, rarely continuous but usually interrupted on face, extending from nose to tail: this stripe nearly obsolete in some very old specimens, but very distinct in young, and in some about half grown more than 5 mm. in width. Orifice of ears surrounded by black hairs tipped with chestnut. Lower parts varying from yellowish white to rich yellowish brown, variation according with the richness of color of upperparts; throat and chest always darker than rest of lower parts. Soles and forelegs lighter than rest of lower parts, usually nearly white; tail usually unicolor. yellowish white, but sometimes dusky above; pencil yellowish white. The young vary but little in color. Those perhaps a week old are grayish brown above, with the very distinct dusky dorsal stripe; beneath practically naked, but with a trace of white down. Others a little older are grayish brown above, varying a little even in the same litter, in the same way as the adults, but in a less degree; beneath yellowish white, brownish between forelegs. Young half grown and larger resemble adults, but are generally graver, and never show the extreme richness of coloring exhibited by adults. Measurements. — Average of 10 adults: Total length 139.5; tail ver- tebra3 IT. 8: hind foot is. 4. One of the largest specimens: Total length 150; tail vertebras 20; hind foot 20. Average of 8 adult skulls: Basal 58 NORTH AMERICAN FAUNA. [No.Jfc, length (incisor to occipital condyle) 28.3; zygomatic breadth 19.2; interorbital breadth 1; mastoid breadth 14.4. A large skull measures: Basal length 29; zygomatic breadth 20; interorbital breadth 4; mastoid breadth 15. This lemming has been several times recorded from this region, usually under the name Arvicola kudsonius. J. C. Ross reported it from Fort Bowen, Prince Regent Inlet," and obtained specimens from Boothia Felix, where it was active throughout the winter; b Richardson states that it inhabits Melville Peninsula; c specimens were taken by Parry's party on Melville Island on June 13, on which date the pelage was turning brown; d and Lyon observed it on Duke of York Bay, Southampton Island/' Rae says, probably referring to this species: Occasionally large numbers of lemmings are found drowned along the shores of James's Bay, but as they are generally seen after a very high tide, it is uncertain whether they are then migrating, or merely caught by the high tide on their native grounds./ Zapus hudsonms (Zimm.). Hudson Bay Jumping Mouse. We took Zapus at Norway House, on Echimamish River, at Rob- inson Portage, at Oxford House, on Steel River (near the mouth of the Shamattawa), and at York Factor}'. It was especially common in the grass}" thickets about Oxford House, where the greater part of the series was taken. At York Factory our traps yielded but two specimens, and at Fort Churchill none, though at the latter place we obtained an imperfect skin from the natives. The last one taken was trapped on Steel River near the mouth of the Shamattawa August 31. At Oxford House, where these jumping mice had been very plen- tiful in July, we set out traps on September 10 and 11 in the same places as before, without success: but as heavy frosts had occurred for some time it is probable that the animals were hibernating. An adult female taken at Oxford House June 30 is suffused above with bully clay color and is lighter and less bright than the others secured; but the series taken as a whole agrees essentially in color with a series from northern Minnesota, and no differences are noticeable in the skulls. Six adults from Oxford 1 louse average: Total length 209.3; tail vertebra1 L26; hind foot 30.3. Two from York Factory average: "■A\'2; 129.5; 30. Embryos from five to eight in number were noted in several instances. Zapus kudsonius was originally described from a specimen sent Prom Hudson Bay, probably from Severn River, by Mr. Graham. It has been recorded from James Bay and Fort Churchill. " « Parry's Third Voyage, Appendix, p. 93, L824. b Appendix to Ross's Second Voyage, i>. xiv, L835. c Fauna Boreali-Americana, p. L32, L829. tf Journal of Parry's ( Firsl ) Voyage, p. 202, 1821. ' Lynn's Private Journal, p. 47, isi'4. /Journ. Linn. Sue. London, Zool., XX, ]>. 144, 1888. g Preble, N. A. Fauna, No. L5, i>. 17, is1.)1.). October, 1902.] MAMMALS OF KEEWATIN. 59 Erethizon dorsatum (Linn.). Canada Porcupine. The porcupine occurs throughout the region between Lake Winni- peg and Hudson Bay. but is nowhere abundant. In a country where the life of tht1 native is a constant struggle lor food, the ease with which tliis animal is taken is a sufficient reason for its scarcity. According to Hearne, the porcupine was formerly found north of Churchill River, but was scarce." Bell says: Mr. Isbister, of the Nelson River House on the Churchill, informs me that it was once abundant there. It is ran- between Lake Winnipeg ami Hudson's Bay, but an individual is occasionally found as far north as York Factory.'' Forster recorded it from Severn River.' Dr. Milne informed me that he had known it to occur but once at York Factory, hut that it is reported by the Indians to be common in certain places about the headwaters of the Shamattawa. Lepus americanus Erxleben. Hudson Bay Varying- Hare. This species is quite generally distributed throughout the region between Lake Winnipeg and Hudson Bay. Unmistakable evidences of its presence were observed all along our route. We obtained a pair of adults and two immature specimens at Oxford House early in July. At York Factory the animal was stated to be rather uncommon. Bell reports it to be common some years in the neighborhood of Fort Churchill.'' An adult female taken at Oxford House July 1 is yellowish gray dorsally, much flecked with black, which predominates along center of back: color of sides extending down on upper part of thighs; head and face yellowish brown; outer side of legs yellowish fawn; inner side of legs dull white: upper side of feet dull white, slightly tinged with yellowish. A male taken at Oxford House July 3 has much less dusky on the back, and the back and sides are much suffused with grayish fawn. The measurements of these specimens are as follows: Male: Total length 430; hind foot 117; female: Total length 450; tail vertebrae 13; hind foot 133. Skull of female: Occipito-nasal length 71; zygomatic breadth 36; breadth across postorbital processes 25; length of nasals 30.5; breadth of nasals 26.5. Lepus arcticus canus subsp. nov. Keewatin Arctic Hare. Type from Barren Grounds near Hubbart Point, about 75 miles north of Fort Churchill, Keewatin. $ yg. ad. (skin and skull), No. 106860, U. S. Xat. Mus., Biological Survey collection. Collected August 17, 1900, by Edward A. Preble. Original number, 3347. General Characters. Differing from Lepus arcticus in assuming a "Journey * * * to the Northern Ocean, p. 381, 1795. &Rept. Pro-. Can. Geol. Surv. L882-3-4, App. [I, ]). 49DD (1885). 'Phil. Trans., LXII, p. 374, 1771'. <*Rept. Preg. Can. Geol. Surv., 1882-3-4, App. II, \>. 49DD | 1885). 60 NORTH AMERICAN FAUNA, [No. 22, gray summer pelage; similar to Lepus labradorius, but differing slightly in color and in cranial characters. ( 'olor. — Fur of upperparts light plumbeous at base, succeeded by a broad zone of yellowish fawn and a narrow zone of dusky and tipped with white; throat and rump light plumbeous without the fawn-colored zone and white tips; legs and feet white, but the light plumbeous of rump extending down a short distance on thighs; soles clay-color — probably stained; chin grayish white, lighter than throat; head grizzled fawn, brightest on nose and around eyes; nape grayish; occiput dusky; ears dusky grajnsh anteriorly, edged, except at tip and anteriorly toward the base, with white, dusky tips about 15 mm. long, mesial surface white; fur on inside of ears dusky, tipped with white. An imperfect winter skin obtained near York Factory has the entire fur pure white to base of hairs, except at extreme tips of cars, where the fur is black, becoming yellowish brown toward base. Lepus labradorius presents a more bluish appearance than Lepus arcticus canus, as the plumbeous element of its pelage is more con- spicuous. In labradoriiis the plumbeous extends down on outside of leg from thigh nearly to heel. Si, ill. — No skulls of adult arcticus are available for comparison. Compared with skulls of labradorius, those of canus differ as follows: Bullae natter, rising but slightly above level of basioccipital; jugal broader. A skull of a fully adult animal obtained at Fort Churchill shows the cranial characters of the species better than the Hubbart Point specimens, which are younger. The Fort Churchill skin resem- bles very closely those taken north of that post. Measurements. — Type: Total length 570; tail vertebra1 SO; hind foot 154. A topotype: Total length 550; tail vertebra1 83; hind foot L55. Skull of adult from Fort Churchill: Greatest length 98; zygomatic breadth r>(); breadth across postorbital processes 39; alveolar length of upper molar series 19. Remarks. — These tine hares occur sparingly in summer throughout the Barren Grounds from Fort Churchill northward. A few breed near Fort Churchill, and one was obtained there August L2. In win- ter they migrate to a slight extent, reaching tin1 neighborhood of York Factory and perhaps farther. Though a few signs of this species were1 noted on a rocky area near my camp 50 miles south of Cape Eskimo early in August, I saw none of the animals themselves until the morning of August 17, when an opportunity was afforded for hunting over a number of low morainic ridges just below Hubbart Point. Tracks made during the previous night were found along tin1 sandy beach, from which they led toward the ridges where I was sun1 the animals would be found. Carefully scanning the ground, which was covered with gray rounded boulders with occasional bunches of dwarfed willows, I hunted back October, 1902.] MAMMALS OF KEEWATIN. <>1 and forth over the ridges. Hundreds of Lapland longspurs flitted from boulder to boulder, but otherwise few signs of life were evident on the semibarren tracts, and I had walked several miles before my attention was attracted by what at lirst appeared to be a boulder on which a small restless bird was perched. A second glance showed that the object was an Arctic hare whose ears, twitching slightly, completed the resemblance that had deceived me. Another was after- wards started from beneath a dwarfed willow near by. Both were secured and proved to be males, evidently young of the year, but full grown. Lynx canadensis Kerr. Canada Lynx. Found throughout the region between Lake Winnipeg and Hudson Bay. but not common in the vicinity of York Factory. The abundance of lynxes from season to season is said to depend on the abundance of rabbits. We obtained a summer skin from the vicinity of Island Lake and a number of skulls from Cross Lake and Oxford House. The skin is apparently that of an adult. It differs considerably from the winter pelage, and may be described as follows: General color on back and sides yellowish brown, the hairs tipped with light grayish brown: a median dorsal stripe reaching from between ears nearly to tail dusky brown, the hairs tipped with light brown; a few obscure spots on sides; beneath dirty yellowish white, with a few spots of dusky on chest and belly; head and neck colored like sides, but tips of hairs more whitish; 'chin beard' white, with a conspicuous black blotch; ears grayish, edged and tipped with black with a few white hairs intermixed: legs concolor with sides; tail yellowish brown above, indistinctly banded with whitish, lighter below and tipped with black. The average measurements of six adult skulls from Oxford House are as follows: Occipito-nasal length (measured to anterior point of nasals) L23; zygomatic breadth 93.3; breadth of braincase 58.6. Dr. Bell says: This animal in its apparently erratic migrations does not reach the verge of the forest. A few skins are obtained at Fort George on the East-main coast ami at York Factory. It has been occasionally rather numerous aboul Oxford House." Hearne saw its tracks near Fort Churchill.6 Canis albus (Sabine). Barren Ground Wolf. Wolves were fairly common in the vicinity of our camp on the Bar- ren Grounds 25 miles south of Cape Eskimo, and several were seen. They were dirty yellowish white in color, and were conspicuous on the barren ridges. Their howling frequently reached our ears, especially at night, when their wild cries seemed peculiarly in keeping with that "Rept, Prog. Can. Geol. Surv., 1882-3-4, App. II, p. 49DD (1885). h Journey * * * to the Northern Ocean, p. 366, 1795. 62 NORTH AMERICAN FAUNA. [No. 22, lonely and desolate waste. Sometimes they were heard pursuing caribou. Owing- to the limited time, I was unable to obtain specimens, though 1 saw many skins, mostly nearly pure white, at Fort Churchill, where they are traded by the Eskimos and Chippewyans. The Barren Ground wolf has been recorded by the different Arctic expeditions from various points in northern Keewatin. Canis occidentalis Richardson. Gray Wolf. Gray wolves are found more or less commonly throughout the region between Lake Winnipeg and Hudson Bay, and numbers are traded at all the posts. We saw tracks in several places on Steel River, and a wolf was seen a few miles above Fort Churchill during our stay there. A skull of a female from the vicinity of Norway House was obtained from Mr. MacDonald, who informed me that the animal was one of a pair which for some time during the winter of 1900-1901 lived on small fish which they caught at a certain place in the river where ice did not form. Mr. Campbell, of Oxford House, informed me that during the winter of 1899-1900 several were killed within a few rods of the buildings. Vulpes fulvus (Desmarest). Common Red Fox. Foxes occur rather plentifully throughout the region between Lake Winnipeg and Hudson Bay. Many skins were seen in the store- houses of the company at Norway House, Oxford House, and York Factory. A few are collected at Fort Churchill. The 'cross' phase of pelage seems to predominate, and the normal or red phase to be next in abundance; but many black, or 'silver.' foxes are taken. Certain districts are said by the traders to produce black foxes almost exclusively; in others these are rarely taken. Large numbers of skins seen at the different posts exhibited every possible degree of variation from the normal red phase to almost pure black. A series of skulls, including specimens from Oxford House, Cross Lake, and Split Lake, was obtained from Mr. Campbell and Mr. MacDonald. Vulpes lagopus innuitus Merriam. Continental Arctic Fox. Vulpes lagopus innuitus Merriam, Proc. Biol. SocWash., XV, p. L70,Augus1 6, L902. We first met with this species July 19 on the 'Barrens' between Stony and Owl livers, about 75 miles north of York Factory. Here, on a slightly elevated part of the tundra, we found a burrow occupied by a family of Arctic foxes. This burrow was typical an under ground labyrinth with several entrances. In the vicinity were scat- tered the bones and feathers of various birds, principally ptarmigans, and well-worn trails leading in various directions gave evidence of the activity of the mother fox in providing for her family. A young one was enticed from the depths of the burrow and secured. Its color may Octobbb, 1902.1 MAMMALS OJ? KEEWATIN. 63 be described us follows: Head, hack, base of tail dorsally, and stripe extending down on outer side of legs, seal brown; face and legs seal brown, flecked with white; shoulders and thighs seal brown, flecked with fawn; sides and lower parts, including ventral surfaceof tail, light fawn, deepest on sides; proximal two-thirds of tail above, hair brown, strongly overlaid with fawn. The color of the adults is stated to he essentially the same as that of the young. On the Barren Grounds north of Fort Churchill tracks of Arctic foxes were frequently seen. Their dens were found at both of my camps south of Cape Eskimo, and an immature individual, which closely resembled the specimen above described, was trapped at the southernmost of these camps. The Continental Arctic fox occurs throughout northern Keewatin and the adjacent islands of the Arctic Sea. In summer it seems to be found chiefly on the seacoast, and breeds on the west coast of Hudson Hay as far south at least as the vicinity of York Factory." It was recorded by Edward Sabine from the North Georgia Islands, where it remained throughout the year/' and by J. C. Ross from Port Bowen.c Richardson described a specimen in the blackish-brown phase, which was killed December It',, on Winter Island, Melville Peninsula.'7 Lyon recorded the specie- from Duke of York Pay. Southampton Island.' In winter large numbers of these foxes pass down the coast of Hud- son Bay. Many are still taken at Fort Churchill, though fewer than in former years. Dr. Milne informs me that at York Factory a few were taken every winter during his residence there, and that the spe- cies reaches Severn River, a fact formerly recorded by llearne.'' I saw a skin which had been taken during the winter of L899-1900 near a large lake about 75 miles north of Oxford House, and Mr. William Campbell informed me that he trapped one some years ago at the out- let of Oxford Lake. The winter of 19* >0-1901 was remarkable for the great number of these animals which came southward, and for their wide dispersion in the interior. Mr. J. K. MacDonald, of Norway House, wrote me that on account of the light fall of snow the foxes followed up the rivers from the Bay to the vicinity of that post. The normal phase of color seems to predominate throughout the region. Among the many winter skins seen at Y'ork Factory and Fort Churchill was only one 'blue' one — probably the winter pelage of the sooty phase — and the color is said to be of very rare occurrence in the region. "Joseph Saliine, Narrative of a Journey to the Shores of the Polar Sea, Appendix, p. 658, 1823. ''Suppl. to Appendix Parry's First Voyage, p. clxxxvii, 1824. •Parry's Third Voyage, Appendix, p. 92, 1826. ,l Fauna Boreali- Americana, I, p. 80, 1829. 'Lyon's Private Journal, p. 46, 1824. /Journey * * * to the Northern Ocean, p. 3(54, 1795. 64 NOKTH AMERICAN FAUNA. [No.22, TJrsus americanus Pallas. Black Bear. Black bears arc rather numerous throughout the region between Lake Winnipeg and Hudson Bay, and many skins are annually traded at all the posts. Toward the northward they become less common. One was seen near Robinson Portage by Mr. W. C. King, who passed this point a day or two ahead of us on his way toward York Factory. This bear was feeding on the piles of Mayflies (Ephemerkhv), which perish in myriads and are washed up on the shores in long ' wind- rows.' These arc said to constitute a favorite food of the bear A number of skulls were obtained from Oxford Lake, where the black bear is rather common. On our return journey we obtained an immature skull at York Factory, and saw a bear's track on the banks of Steel River a few miles below the mouth of Fox River. Dr. Bell killed a black bear on Churchill River, about 100 miles from its mouth." TJrsus richardsoni Swainson. Barren Ground Bear. While at Fort Churchill I made inquiries in regard to the Barren Ground bear, but the official in charge, Mr. Alston, knew nothing of such a species. Dr. Bell, speaking of the Barren Ground bear, says: In the barren grounds to the northwest of Hudson's Bay, I have been told that a large bear is found, which the Eskimo consider a variety of the polar bear, which has adopted a terrestrial life, and to which they have given the name of "blue" <>r "grey" bear. * * * This bear is found in the barren grounds south of Hudson's Strait, a While on his journey of exploration Hearne saw the skin of an enormous grizzled bear at the tents of the Indians on the Copper River.6 This was probably the skin of this species. If the animal extends its range to the vicinity of Hudson Ba}^ it must be very rare. Thalarctos maritimus (Phipps). Polar Bear. We obtained several skulls at York Factory through the kindness of Dr. Milne and Mr. G. B. Boucher. While we were traveling between York Factory and Fort Churchill a party of Indians reported seeing a white bear, which swam out to sea on perceiving them. On August 9, below Cape Eskimo, while we were awaiting the rise of the tide SO that we could land, we saw a polar bear on tin; tundra, lie was rapidly quartering the rolling ground in search of food. A few days later another was seen several miles north of our camp. While traveling down the coast we saw several places where the animals had lain in the rank beach grass. Mr. Boucher killed a female and her two cubs on the coast between York Factory and Cape Tatnam about the middle of August. aRept. I'm-, fan. Geol. Surv., 1882-3-4, App. II, ]>. 511)1) (1885). b Journey * * * to the Northern Ocean, p. 372, 1795. OCTOBER, 1902.] MAMMALS OF KEEWATTN. 05 Polar bears occur regularly all along the coast of Keewatin as far south at least as Severn River, and probably to the head of James Bay. Bell reports that a tVw have been seen at Moose Factory.0 The female 'dens up' in a snowdrift in the winter, brings forth her young about March, and soon afterwards lends them to the sea. The male is said to pass the winter at sea. The animals are frequently seen during late summer swimming in the Bay, and Bell records that one was killed by the captain of one of the Hudson's Bay Company ships in open water about the middle of the Bay/' Polar hears have been met with by various expeditions to the north of Hudson Bay. Lutra canadensis (Schreber). Canada Otter. Otters seem to he found throughout the region, as we saw skins at all of the posts visited. They are not found in the immediate vicinity of Fort Churchill, hut are said to he taken not far up the Churchill River. Hearne states, probably referring to the interior, that they used to frequent the rivers to the north of Churchill as far as latitude 62 . We frequently saw their tracks while we were ascending the upper part of Hill River, and the Indians often spoke of the abundance of otters in the vicinity. At York Factory we obtained the skin of a young'one about the size of a small mink and dull dark brown in color. Skulls composing a series recently obtained from Oxford House, Norway House, and ( Iross Lake are apparently not separable from skulls from Godbout, Quebec, assumed to be typical canadensis. Mephitis mephitis (Schreber). d Canada Skunk. A large number of. skins were seen in the storehouse at Norway House, and the animal is occasionally taken about Oxford House, where we obtained a hunters' skin. While paddling up the channel between Windy and Pine lakes on September 12 we saw a skunk swimming across the stream a hundred yards in front of our canoe. On seeing- us he redoubled his exertions, hut we overtook and shot him just as he reached the shore. This was the only one seen on our trip. This specimen, as well as the Oxford House skin, has been recorded by Howell.' Dr. Bell gives the skunk as occurring on both sides of dames Bay." Forster records one sent from Severn River Uy Mi-. Graham/ Howell records a specimen from Moose Factory.' Among the large number of Bkins at Norway House were several which seemed to he referable to the Northern Plains skunk (J/, hudsonica). Kept. Prog. Can. Geol. Surv., L882-3-1, App. II, p. 50DD (1885). ''Ibid., p. 50DD (1885). ^Journey * * * to the Northern Ocean, p. :i74. 1795. <*See Science mew ser.), XVI, No. 394, p. 114, July is, 1902. • X. A. Fauna, No. 20, p. 23, 1901. /Phil. Tran<., LXII. p. 374, 1772. 71»;:.— No. 22 5 66 NOKTH AMERICAN FAUNA. [No. 22, Lutreola vison lacustris subsp. nov. Keewatin Mink. Type from Echimamish River (near Painted Stone) Keewatin, Canada. $ yg. ad. (skin and skull), No. 106872, U. S. Nat. Mus., Biological Survey collec- tion. Collected September 14, 1900, by Alfred E. and Edward A. Preble. Original number, 3518. General characters. — Similar to Lutreola vison but larger; .skull more angular. Skull. — Compared with skulls of vison from the Adirondack Moun- tains and New England (assumed to be typical) a large series from Oxford House and the surrounding region differ as follows: More angular and much larger, those of males of vison hardly equaling those of females of lacustris; sagittal ridge highly developed; zygo- mata strongly bowed outward; dentition heavier than in vison; bulla? large and rather flat, inner anterior corner produced toward pterygoids and ending in a rather acute angle. Color. — Type: Upperparts very dark chocolate brown; underparts lighter; chin and a narrow, irregular, and partially broken stripe extending to lower part of breast, white; a little white between hind legs. Measurements. — Type: Total length 560; tail vertebrae 11)0; hind foot 67. Adult female from Swampy Lake: 540; 180; 00. Average of five immature but well-grown individuals of both sexes: 540; 153; 64.8. Skull: Average of ten adults (probably males) from Oxford House: Occipito-nasal length 62.5; zygomatic breadth 41.25; inter- orbital constriction 11.9; breadth across postorbital processes 18.25; mastoid breadth 36.1. Remarks. — Minks seem to be common and quite generally distrib- uted throughout the region between Lake Winnipeg and Hudson Bay. They are annually traded at all the posts, and at all except Fort Churchill in large numbers. A few skins were seen in the storehouse at Fort Churchill, and Mr. Chapman, the resident missionary, saw a mink a few miles up Churchill River late in July. We trapped an adult female at Oxford House and another on the shore of Swampy Lake, and while traveling through the lakes on our return journey early in September shot several that were swimming across rivers or between the islands of the lakes. These were mainly immature. The Indians say that when traveling through the lakes at tins season they always see minks, and that these are young of the year that are leav- ing the place of their birth and starting out for themselves. We were usually able to approach quite near before they took alarm and dived, and as they came to the surface within 1 5 or 20 yards they fell an easy prey. The specimens composing the series show more white than is usual in vison, the average excess appearing in the type. October, 1902.] MAMMALS OF KKKWAT1N. 67 Putorius cicognani (Bonaparte). Bonaparte Weasel. Weasels, or 'ermines,' as they are usually styled in the north, wore reported as common throughout the region between Lake Winnipeg and the Bay, and many arc caught in traps set for more desirable species. We failed to trap any while traveling through the district. but saw many skins at the trading posts and obtained a number of skulls from the natives at Oxford House. The skulls prove to be typ- ical cicognani, and as two skins in the U. S. National Museum collec- tion from Fort Albany. Hudson Bay. are referable to this form, the Bonaparte weasel is probably the common weasel of the region. Two winter skins from York Factory seem referable to this species, but in the absence of skulls I find it impossible to decide whether they are nearest to cicognani or to richardsoni. Putorius cicognani richardsoni (Bonaparte). Richardson Weasel. This form probably replaces cicognani throughout the northern parts of the region as far north as the limit of trees and perhaps farther. An immature male taken on Churchill River a few miles above Fort Churchill is dark brown above, as in cicognani, and white beneath, very slightly tinged with yellow. The black on the tail occu- pies its terminal third, inclusive of the hairs. Another male, also immature, shot on the edge of the Barren Grounds below Hubbart Point August 17. resembles the Churchill River specimen, but is strongly tinged with sulphur yellow beneath. Putorius arcticus Merriam. Tundra Weasel. Weasels collected by Ludwig Ivunilien at Cumberland Gulf in 1878 prove to be referable to this species, the type of which came from Point Barrow. This indicates that the range of tin1 animal may extend across the northern part of the continent, and makes it probable that the following notes referring to erminea properly relate to arcticus. J. ('. Ross mentions the occurrence of the animals on the shores of Boothia Felix. He states that they assumed their winter dress in Sep- tember and turned brown again toward the end of May. Their tracks were seen at intervals throughout the winter, and the accompanying tracks of lemming's showed that the weasels were in pursuit of what was found to be their chief prey." They are also recorded from the west side of Baffin Bay/' and were found on the east side of Melville Peninsula on Parry's second voyage.' Putorius rixosus Bangs. Bangs Weasel. A specimen of this diminutive weasel from Moose Factory is in the U. S. National Museum collection, and has been several times recorded. ^Appendix to [loss's Second Voyage, p. \, 1835. ^ Appendix to Ross's First Voyage, p. xliii, 1819. 'Lyon's Private Journal, p. 54 (and elsewhere), 1824. 68 NORTH AMERICAN FAUNA. [No. 22, I made inquiries of the fur traders in regard to the species, but none knew it. It is probable, however, that it is found throughout the southern part of the region, and that their ignorance of it results from its fur being considered valueless and- consequently not offered in trade. Forster recorded a small weasel sent from Hudson Bay by Mr. Graham, which was undoubtedly this species. Its total length is given as 7 inches, and that of its tail as about 1 inch." Bangs recorded a specimen from Fort Albany.'' Mustela amerieana abieticola subsp. now Hudson Bay Marten. T>jpe from Cumberland House, Saskatchewan. $ ad. (skin and skeleton), No. iffff, U. S. Nat. Mus. Collected February, 1890, by Roderick MacFarlane. General cha/racters. — Much larger than Mustela . 293, 1825 (1827), • Appendix to Boss's Second Voyage, p. viii. L835. /Bept. Prog. 'an. Geol. Surv., 1877-78, j». 29c | L879). 70 NORTH AMERICAN FAUNA. [No. 22, The animal was formerly abundant at Walrus Island and other points to the north of Fort Churchill. Hearne relates that in 1767 while passing- Sea Horse 'Island' he and his party ''saw such num- bers of those animals lying on the shore, that when some swivel guns loaded with ball were fired among them, the whole beach seemed to be in motion." a J. C. Ross recorded walruses from the northern part of Prince Regent Inlet.* According to the same author they were unknown about Boothia, but abounded at Repulse Bay.' Dr. Rae mentions seeing many on a small island near Whale Island, below Wager Bay.'' Phoca hispida Schreber. Rough Seal. A number of skins were seen at Fort Churchill and the animal was said to be rather common in the vicinity. J. C. Ross reported the species from the seas on both sides of the Isthmus of Boothia/ and from Port Bo wen.' Phoca groenlandica Erxleben. Harp Seal. Dr. Bell gives this species as the commonest seal in all parts of Hudson Bay at all seasons// 1 saw no skins, but the species was reported at Fort Churchill. J. C. Ross reported it from the west side of the Isthmus of Boothia/' Cystophora cristata (Erxleben). Hooded Seal. Dr. Bell reported this seal from Hudson Straits, where a few skins were seen in the possession of the Eskimos.'' Its range may extend to other parts of the Bay. Erignathus barbatus (Erxleben). Bearded Seal. An adult female was killed July 20 about 75 miles north of York Factory. Its total length was 2,300 mm. (about 7£ feet), and it was so heavy that we had great difficulty in getting it on board. Its general color is grayish; lighter, almost silvery on sides and head, darker on nape and middle of back. We saw numbers of these seals both to the north and south of Fort Churchill, and the species probably reaches all parts of the Bay. The so-called 'clapmatch' lines, used in place of heavy rope for various uses, are made largely from the skins of this species in the region we ".Journey * * * to the Northern Ocean, p. 388, L795. ''Appendix to Ross's Second Voyage, p. x\i, L835. elbid, p. xxii, is:;r,. '' Expedition <<> the Shores <>f the Arctic Sea, p. 180, L850. e Appendix t<> Ross's Second Voyage, p. \ix, L835. /Appendix in Parry's Second Voyage, \>. 94, 1825 (1827). f/Kept, Prog. Can. Geol. Surv., L882-3-4, App. II, p. 52DD (1885). October, 1902.] MAMMALS OF KEEWAT1N. 71 visited. A long- line of surprising strength is made by cutting the skin of the animal in a spiral manner. Dr. Bell saw skins of the hoarded seal in the possession of the Eski- mos in Hudson Strait in L884, and a few of the animals were observed by his party. He also saw bearded seals on the Eastmain coast in IsTT. and killed a Large one at the mouth of Moose River." J. C. Ross reported the species to be found in summer on the shores of Boothia.'' Phoca vitulina Linn. Harbor Seal. Doubtless found in all parts of the Bay. We saw it in numbers all along the coast visited, especially at the mouth of Churchill River. We saw one August 28, several miles above York Factory, as we were ascending Hayes River. A specimen of a rather small female that measured 1,500 mm. in length was secured 50 miles south of Cape Eskimo August 14. Its general color is light yellowish, irregularly and obscurely spotted on the back with brownish. Neosorex palustris (Richardson). Marsh Shrew. This species is represented in our collection by a series of seven col- lected between Echimamish River and the upper portion of Hill River. We trapped them in the grass}7 margins of marshes or in wooded swamps. As this series shown considerable variation in color some of the specimens may be noted in detail. Four taken at Robin- son Portage June 27 agree very closely in color — back dusky, very finely flecked with silveiy white; beneath gra}Tish white with a tinge of brown, grading insensibly on sides into color of upperparts; throat and chin noticeably lighter than remainder of lower parts, this light color sharply divided from the dusky of the face; inner surface of forelegs and fringes of feet silveiy white; tail rather sharply bicolored nearly to tip; hind feet dark on outer side, lighter on inner side. A specimen taken on Hill River, near Swampy Lake, September 5, agrees closely with the June specimens. One taken at Robinson Port- age September 14 is evidently in winter pelage, the fur being very soft and full, and glossy black dorsally, much darker than the June speci- mens. One taken on Echimamish River September 15 agrees with the June specimens dorsally, but the ventral surface is strongly clouded with brownish. The skulls, compared with skulls from Min- nesota and South Edmonton, Alberta, assumed to be typical palustris, show no differences of value. The series shows little variation in size. The seven specimens aver- age: Total length 157.3; tail vertebra? 72; hind foot 19.4. A speci- men received from Mr. J. K. MacDonald, of Norway House, who obtained it from the Indians, is in full winter pelage. It resembles "Rept. Prog. Can. Geol. Surv., 1882-3-4, App. II, p. 52DD (1885). '' Appendix to Ross's Second Voyage, p. xxi, 1835. 72 NORTH AMERICAN FAUNA. [No. 22. the .specimen from Robinson Portage taken in September, except that the fur of the back is tipped with whitish. The species, it is stated, is locally called the 'beaver mouse,' from being found in the houses of the beaver. Sorex personatus I. Geoff roy. Common Eastern Shrew About twenty-live specimens were collected from various localities between Norway House and York Factory. These average lighter in color beneath and slightly larger than specimens of -personatus in corre- sponding pelage from the Northeastern States. The males average a little darker dorsally than the females. Sometimes there is a distinct dark area on the back separated from the ashy of the lower parts by a lighter lateral stripe. Among my specimens this dark area seems to increase in intensity with age, and a very old male with excessively worn teeth from York Factory has a conspicuous tricolor pattern. This specimen has an abnormally short tail and measures: Total length 90; tail vertebras 32; hind foot 12. Eight of the largest speci- mens average: 101.1; 40.6; 12. King records a specimen of Sorex forsteri which was found near the mouth of Great Fish River.'1 Sorex (Microsorex) alnorum sp. now Keewatin Shrew. Ti/jir from Robinson Portage, Keewatin, Canada. J a^- (skin and skull), No. 107014, U. S. Nat. Mus., Biological Survey collection. Collected June 27, 1900, by Alfred E. and Edward A. Preble. Original number, 2662. Ge?ie?'al characters. — Larger than Sorer hoyi with larger skull; lower parts not tinged with buffy. ( blor. — Type: Upperparts sepia brown as in S. hoyi/ lower parts ashy gray, not tinged with buffy; hind foot dusky on outer and whit- ish on inner half; tail dusky above, whitish below, becoming dusky toward tip. Compared with Sorex (Microsorex) eximius from Cook Inlet, Alaska, the type is slightly darker above and slightly grayer beneath. Skull. — Compared with skulls of S. hoyi from Elk River, Minne- sota, and Red River Settlement, Manitoba, the skull of the type is much larger; rostrum about the same; braincase much broader and higher, both actually and proportionally; mandible longer and slen- derer; teeth more heavily pigmented. Compared with Sorex (Micro- sorex) eximius from Cook Inlet, alnorum has rostrum slightly broader, longer and higher; unicuspids more crowded. Measurements. — Type: Total length 98; tail vertebra1 35; hind foot 12. Skull: Greatest length It!. 5; width of braincase 7; length of mandible 10. Remarks. — We found this species only at Robinson Portage, where the type was trapped, and on the Echimamish. At the latter point a "Narrative of a Journey to the Shores of the Arctic < >cean, 1 1, ]». 17, 1836. October, 1902.] MAMMALS OF KEEWATIN. 73 Sorex was bo oaten in the trap as to be unlit for a skin, but the skull was saved and proves referable to Microsorex. It was badly injured and is valueless for comparison, but it is assumed to belong to this species. Sorex richardsoni Bachman. Richardson Shrew. This shrew is represented in our collection by a series of eleven specimens. Excepting one taken in June at Robinson Portage and one trapped on Swampy Lake, near its outlet, September 6, all are from Norway House, where the species was abundant. Four of these Nor- way House specimens, taken June 18 to 22, are still in the dark-backed winter pelage; all the others have the dorsal area less distinctly indi- cated. One taken at Norway House dune 22 contained six embryos. The Swampy Lake specimen, which was trapped in spruce woods, has an abnormally long tail and measures: Total length 1_!<>; tail vertebra? 48; hind foot 14. The rest of the series average: 1 L6.6; 41.5; 14. Under the name Sorex araneus, Forster recorded a specimen sent from Hudson Bay. probably referable to the present species.0 Sorex sphagnicola Cones. Cones Shrew. A shrew collected by Dr. Bell on Shamattawa River was referred to this species by Dr. C. Hart Merriam.6 Sorer sphagnicola is very imperfectly known, but is undoubtedly most nearly related to richard- son i. Condylura cristata (Linn.). Star-nosed Mole. Dr. Bell speaks of this species as being- common at Moose Factory/ A specimen in the U. S. National Museum, collected at that post in 1881 by Dr. Walton Hay don. and one taken at Rupert House. James Baj% have been recorded by True.'' I have recently examined the Moose Factory specimen. It was kept in alcohol, but has spoiled in the fluid so that only the skull and part of the skeleton are intact. The skull resembles closely specimens from Pennsylvania and the Northern States. Myotis lucifugus (Le Conte). Little Brown Bat. Two specimens from James Buy are recorded by Miller.' We saw no bats of any kind during- our trip. "Phil. Trans., LXII, p. 380, 1772. '-X. A. Fauna, No. 10, p. 65, 1895. 'Kept. Pro<_'. Can. Geol. Surv. 1882-3-4, App. II, \>. 48DD (1885). <*Proc. U. 8. Xat. Mus., XIX, p. 84, 1896. ' X. A. Fauna, No. 13, p. 63, 1897. BIRDS OF KEEWATIN. Colymbus holboelli (Reinh.). Holboell Grebe. The Bolbcell Grebe was lirst described by Reinhardt from Green- land. Joseph Sabine mentions having received specimens from Hudson Bay." MacFarlane took its eggs in the vicinity of Fort Anderson;6 and as the species is recorded from other northern points and breeds in Manitoba it probably occurs throughout the Hudson Bay region. Colymbus auritus Linn. Horned Grebe. Several grebes thought to belong to this species were seen near the the mouth of Red River June 15, and a specimen was collected at Sea Falls. 2<» miles north of Norway House, September 16. Andrew Murray recorded the species from Trout Lake," and Dr. Bell has collected it at Fort Severn and York Factory, and found it breeding at Fort Churchill,'' from which point Clarke records a speci- men in summer plumage.' A specimen collected by Dv. Rae at Repulse Bay is recorded in the British Museum Catalogue. Podilymbus podieeps (Linn.). Pied-billed Grebe. Bell records a specimen from York Factory collected and presented to him by Dr. Matthews. ' This seems to be the only published record of the capture of the pied-billed grebe in the Hudson Bay region. Nutting records specimens taken at Chemawawin, Saskatchewan, an Indian village about GO miles west of the mouth of the Saskatchewan, where the species was breeding abundantly.9 Gavia imber (Gunn.). Loon. We found this species rather common on the lakes and deeper parts of the rivers on our route between Norway House and York Factory, and its wild notes were heard nearly every night. Several were seen near the mouth of Churchill River July 30, and several more near North River Jury 31. On our return trip we saw one or more on nearly every lake between Oxford House and Norwa}' House. '« Franklin's Narrative of a Journey to the Polar Sea, Appendix, ]>. 692 i Podieeps rubricottis), 1823. &Proc. U. S. Nat. Mas., XIV, p. 415, 1891. eEdin. New Phil. Journ., (new per.), IX, p. 231, 1859. rfProc. Royal Soc. Canada, 1882, I, Sec. IV, p. 49 (1883). 'Auk, VII. p. 320, 1890. /Kept. Prog. Can. Geol. Surv., 1882-3-4, App. Ill, p. 56DD (1885). r/Xat. Hi.-t. Bull. Univ. Iowa, II, p. 249, 1893. 75 76 NORTH AMERICAN FAUNA. [No. 22, King- recorded the 'great northern diver' from the mouth of Back River/' and Murray received a specimen from Severn House/' The catalogue of birds in the U. S. National Museum collection con- tains the record of a specimen collected at Moose Factory in 1881 by Walton Haydon. Gavia adamsi (Gray). Yellow-billed Loon. Under the name Colymbus glacialis, James Clark Ross records three loons, which from his description were undoubtedly of this species, obtained about Boothia during (John) Ross's second voyage.0 In speaking of a loon given him by James Clark Ross, which had been procured in a very high latitude, and which had a yellow bill. Audubon probably refers to one of these specimens. d This record extends the previously recognized range of the species considerably to the eastward. Gavia arctica (Linn.). Black-throated Loon. Swainson and Richardson speak of this species as common on Hudson Bay,' and Murray recorded it from Severn House. -; Clarke recorded an adult in summer plumage from Fort Churchill/7 Some of these records may refer to pacijlca. Gavia pacifica (Lawr.). Pacific Loon. First seen on Hudson Bay about 25 miles north of York Factory July IT, and rather common northward. A tine adult male was secured at Fort Churchill July 25. On the Barren Grounds below Cape Eskimo, August 4 to 13, the species was abundant on the shallow ponds on the tundra, where the young are raised. The old birds were often seen flying to and from the Bay, where most of their food seemed to be secured. The howl of a wolf, or any unusual sound, was generally followed by a chorus of their wild, weird calls, lasting for several minutes. The species was last seen by us below Cape Churchill August 24. Edwards's plate and description of the 'Speckled Diver, or Loon,' « Narrative of a Journey to the Shores 'if the Arctic Ocean, II, p. 21, is:;ii. &Edin. New Phil. Journ. (new ser.), IX, p. 231, L859. Severn House does not appearon most, maps. Thompson, who lias had access to the maps ami records of the Hudson's Bay Company in London, locates it "on Severn Lake, at .r>4° 5' north lati- tude and 92° 30' wesfc longitude." (Proc. U. S. Nat. Mus., XIII, p. 463, L890). In Stieler's Hand Atlas it is located at the same place. If this information is correct (as it probablyis) Severn House should not he confounded with Fori Severn, the post at the mouth of Severn Liver, though it is barely possible that Murray's records refer to Fori Severn. c Appendix to Ross's Second Voyage, p. xlii, 1835. •i Birds of America, VII, p. 291. - Fauna Boreali-Americana, II, p. 475, 1831. / Edin. New Phil. Journ. (new ser.), IX, p. 231, 1859. '/ Auk, VII, |>. 320, 1890. Octobek, l'JOJ.] BIRDS OF KKEWAT1N. 77 plainly represent this species. He gives the length of the bill to the angle as ."» inches, and the color of the hinder part of head and neck as light ash." Specimens from Winter Island, Melville Peninsula, and from Repulse Bay are recorded in the British Museum Catalogue. Gavia lumme (( iunn. ). Red-throated Loon. First seen July 21 at Fort Churchill, where it. was fairly common. It was abundant August 4 to L3 on the shallow ponds on the Barren Grounds below Cape Eskimo, where the old birds were feeding young that were still unable to fly. An adult male and a young one in the dusky downy plumage were collected 50 miles below Cape Eskimo August 4. At our camp 25 miles south of Cape Eskimo the species was more abundant than G. pacijlca, and fully as noisy. While returning we saw two and shot one on Knee Lake September 9. Cepphus mandti (Lieht.). Mandt Guillemot. Two were seen on Hudson Bay about 50 miles south of Cape Eskimo August ■">. One of these sat motionless on the water while the boat passed by within a few feet. On August 19 1 saw one on Button Bay near Fort Churchill. Two specimens from Melville Peninsula are recorded in the British Museum Catalogue. Andrew Murray records the black guillemot ( O. grylle) from Severn House/' and Richardson states that it was very numerous oil' Melville Peninsula.' In his narrative Dr. Rae speaks of finding it nesting on an island off Knaps River, Hudson Bay.'' These records of grylle probably refer to the present form. TJria troile (Linn.). Murre. Richardson describes a specimen collected at York Factory, Hudson Bay.' Una lomvia (Linn.). Bri'mnich Murre. Swainson and Richardson say that this species frequents Hudson Bay.-' A specimen from Hudson Bay, received from the Hudson's Bay Company, is recorded in the British Museum Catalogue. Though I have been unable to find any other record, there seems to be no reason that this murre should not occur there regularly. The species is recorded by Ross from Port Bowen, Prince Regent Inlet, where the birds arrived earlv in .June.'' "Nat. Hist.. Ill, pi. 146, L750. '-Edin. New Phil. Journ. (new ser.J, IX, p. 231, 1859. c Appendix to Parry's Second Voyage, p. 377, 1825 (1827). '^Narrative of an Expedition t<> the Shores of the Antic Sea, p. 22, 1850. ' Fauna Boreali-Amerieana, II, p. 477, 1831. /Ibid., p. 477, 1831. U Parry's Third Voyage, Appendix, p. 107, 1826. 78 NOKTH AMEKICAN FAUNA. [No. 22, Alle alle (Linn.). Dovekie. This species is recorded as abundant in Baffin Bay, Davis Strait, and other parts of the Arctic seas, and it winters southward on the coast, hence it probably occurs, at least in migrations, about the northern part of Hudson Bay. Megalestris skua (Bri'inn.). Skua. Larus Tceeask of Latham," stated to inhabit Hudson Bay, and based on Hutchins's 'Esquimaux Keeask,'^ probably refers to the present species. Though it has apparently not since been recorded from Hudson Bay, the facts of its presence in Hudson Straits and its breeding- at Lady Franklin Island, north of the straits, render its occurrence on the waters of the Bay probable. Stercorarius pomarinus (Temm.). Pomarine Jaeger. Several seen near the mouth of Churchill River July 21. They were pursuing the terns and small gulls with remarkable agility, evidently to rob them of their prey. A male was collected below Capo Eskimo, where the birds were common August 3 to 8. They were generally living about over the tundra somewhat after the manner of a marsh hawk, frequently hunting in small, noisy companies; but sometimes several would be seen standing on a small knoll, apparently asleep. They were easily decoyed by the imitation of the cry of a bird in distress. Despite the frequency of melanism in this genus, I saw no black jaegers of any species, though my Indian guide reported seeing one at our camp 25 miles below Cape Eskimo. Richardson records the species from Igloolik, Melville Peninsula.6" Stercorarius parasiticus (Linn.). Parasitic Jaeger. First met with about 50 miles north of York Factory, where several were seen, and a female was shot July 19. The species was common on the Barren Grounds below (ape Eskimo, August 1 to 13. In habits it resembles S. pomarinvs. Clarke records a melanistic specimen from Fort Churchill.** Stercorarius longicaudus Vieill. Long-tailed Jaeger. Edwards's plate of the 'Arctick Bird' represents this species. In his description he says the tail feathers are 13 inches long.' Dr. Bell records a specimen which was shot near York Factory by Dr. Matthews, who presented it to him.'' The catalogue of birds in the U.S. National Museum collection records a specimen of S. ricluwckoni taken at Fort "Index- Ornithologicus, II, p. sis, L790. '-Latham, Synopsis, III, Part 2, p. 389, L791. '■'Appendix to Parry's Second Voyage, p. 361, L825 (1S137). ''Auk, VII, p. 320, is'. io. ' Nat. Hist., Ill, i»l. 148, 1750. /Kept. Prog. Can. Geol. Surv., L882-3-4, App. Ill, p. 56DD (1885). October, 1902.] BIRDS OF KEEWATIN. 79 Churchill by W. \Y. Kirkby. A specimen from Dukeof York Bay is recorded in the British Museum Catalogue. Pagophila alba (Gunn.). Ivory Gull. Richardson describes a specimen of the ivory gull killed at Hudson Bay.*1 This gull has also been recorded from Port Bowen and from other localities to the north of Keewatin. Rissa tridactyla (Linn.). Kittiwake. Sabine says this species abounds in Hudson Bay,6 and Richardson gives a description of one killed on Melville Peninsula in July/ A specimen from the Savage Islands. Hudson Bay, is recorded in the British Museum Catalogue. Larus glaucus Brunn. Glaucous Gull. Doubtless found in all parts of Hudson Bay. It has been recorded from Melville Island. Felix Harbor, and other places in the Arctic regions, and has been found breeding in James Bay and at various points on the east coast of Hudson Bay.6* Larus leucopterus Faber. Iceland Gull. Undoubtedly occurs on Hudson Bay during migrations, since many winter on the Great Lakes, and Arctic expeditions have obtained specimens in Davis Strait and Baffin Bay and at Melville Island. Larus mariiius Linn. Great Black-backed Gull. Common on the coasts of Greenland and Labrador and frequent in winter on the Great Lakes. It is, therefore, like the Iceland gull, very probably to be found, at least during migrations, inhabiting Hudson Bay. Larus argentatus Brunn. Herring (lull. Common on Lake Winnipeg, Hudson Bay. and all the intermediate lakes and larger rivers, and breeding throughout the region. It is usually very shy. however, as it is shot for food by the natives when- ever opportunity offers. At Fort Churchill, where we collected a specimen July 28, the eggs are gathered in large numbers in May and packed in salt, to be eaten in the late autumn and early winter. On our return trip during the latter part of August and fore part of September, many birds in the dark, immature plumage, sometimes almost sooty, were seen. Larus delawarensis Ord. Ring-billed Gull. Gulls referred to this species were rather common in June and July on the inland waters from Lake Winnipeg to Hudson Bay and north- " Fauna Boreali-Ainerk-ana, II, j>. 419, 1831. & Franklin's Narrative of a Journey to the Polar Sea. Appendix, \<. 095, 1823. 'Fauna Boreali-Amerieana, II, p. 423, 1831. Appendix to Parry's Second Voyage, p. 360, 1825 (1827). -Appendix to Rose's Second Voyage, p. xxxvi, is::."). a Proc. Zool. Soc. of London, p. Ill, 1871. e Transactions Norwich Soc, IV, p. 351, 1887. October, 1902.] BIRDS OF KEEWATIN. 81 Sterna caspia Pallas. Caspian Tern. A specimen procured at Moose Factory by J. McKenzie is recorded by Turner." and Nutting records one taken near Grand Rapids, Sas- katchewan, in the summer of 1891. 6 As the species occurs in the interior as far as Great Slave Lake, it is probable that it regularly reaches southern Keewatin. Sterna hirundo Linn. Common Tern. Common on Lake Winnipeg June 15 to 17, and seen on nearly all the lakes on our route to Hudson Bay. One was collected near Rob- inson Portage June 28. It may occur on Hudson Bay north to the Barren Grounds, in association with the Arctic tern, but was not identified with certainty this far north. It was recorded from Hudson Bay by Andrew Murray." Sterna paradissea Bri'mn. Arctic Tern. Common on Hudson Bay. A specimen was collected 50 miles north of York Factory July 19, and the species was seen daily at Fort Churchill. Young- just able to fty were observed on the meadows bordering Button Bay Jul}T 31, and still smaller young on a sandy islet a few- miles farther north August 1. When I started south from near Cape Eskimo August 13, the species had apparently commenced to migrate, and old and young were common on the Bay until we reached York Factory August 26. The catalogue of birds in the U. S. National Museum collection contains the record of an Arctic tern collected at Moose Factory in 1881. Hydrochelidon nigra surinamensis (Gmel.). Black Tern. Abundant on the marshes at the mouth of Red River June 15. Many were seen June 24 on a marsh near Hairy Lake and many more June 27 at the north end of Robinson Portage, where two specimens were collected. Murray recorded the species from Severn House and Moose Factory.' Fulmarus glacialis (Linn.). Fulmar. Stated by James Clark Ross to be peculiarly numerous in Hudson Bay. Davis Strait, and Baffin Bay.' Phalacrocorax dilophus (Swain.). Double-crested Cormorant. A single immature bird was taken near Pine Lake September 12. It was swimming about in some rapids and was easily approached. «Proc U. S. Nat. Musi, VIII, p. 245, 1885. ''Nat. Hist. Bull. Univ. Iowa, II, p. 252, 1893. cEdin. New Phil. Journ. (newser.), IX, p. 231, 1859. a Appendix to Boss'* Second Voyage, p. xxxviii, 1835. 7165— No. 22 0 82 NORTH AMERICAN FAUNA. [No. 22, We saw cormorants, probablj" referable to this species, on Lake Win- nipeg in June, and our Indian guide said he had often seen them at Norway House. Pelecanus erythrorhynchos Gmel. White Pelican. We saw several June 16 near Bull Head Point, Lake Winnipeg. Forster recorded the species (as a variety of P. onocrotalus) from York Fort, Hudson Bay/' Andrew Murray also recorded a specimen from Hudson Ba},* though Blakiston says later that Mr. Murray was not sure where the specimen was killed/' Merganser americanus (Cass.). Merganser. A female, apparently with a brood of young in the vicinity, was seen near Sea Falls, about 20 miles north of Norway House, June 23. Several individuals were observed on Oxford Lake June 30. A female with moulting wing quills was taken on Hayes River a few miles above York Factory July 10. Though unable to fly, this bird dived so adroitly that we had some difficulty in securing it. Merganser serrator (Linn.). Red-breasted Merganser. A number were killed for food by the Indians at our camp on the Barren Grounds 50 miles south of Cape Eskimo August 3 to 8. While ascending Hill River September 2 we observed a female accompanied by young unable to fly, and near Pine Lake September 13 we met with a flock of the birds and obtained from it several specimens. This species is recorded by Murray from Trout Lake and Severn House;6 and a specimen collected at Repulse Ba}r by Dr. Rae is recorded in the British Museum Catalogue. Lophodytes cucullatus (Linn.). Hooded Merganser. A female was collected by my brother, Alfred E. Preble, on Churchill River about 15 miles above Fort Churchill August 6. The species was recorded by Murray from Trout Lake and Hudson Bay;6 and Bell saw the young going south on Nelson River in September. d Anas boschas Linn. Mallard. First seen near Pine Lake June 28, where a female followed by a brood of young was noted, and one of the brood collected. An adult female was taken near Oxford House July 3, and between this point and York Factory the species was several times observed. One was "Phil. Trans., LXII, p. 419, 1772. &Edin. New. Phil. Journ. (new Her.), IX, p. 231, 1859. 'I his, V, p. 151, L863. ''Kept. Prog. Can. Geol. Surv., L878-79, App. VI, p. 69c (1880). October, 1902.] BIRDS OF KEEWATIN. 83 soon on the marsh at Beacon Point, near York Factory, July L3, one on a small pond on the Barron Grounds north of Soal River August 18, and a female with a brood of unfledged young on Steel River August 31. During the first half of September the species was seen almost daily. A large flock was noted on Trout River September 9, and many were observed on the Echimamish September 14 and 15. During the first part of our journey, when we were going north, we saw several females with broods, and it was noticeable that the young, when startled, invariably took to the woods, where they easily con- cealed themselves. On our return we started up several large flocks, but more commonly found just three in a flock. The name given to this cluck throughout the whole region visited is 'stock duck."1 The catalogue of birds in the U. S. National Museum collection shows that specimens were collected at Moose Factory in 1881, and the bird undoubtedly occurs throughout the wooded portion of the region. Anas obscura rubripes Brewst." Red-legged Black Duck. An adult male was collected at Fort Churchill July 28 and another had been shot by an Indian a day or two previously. Blakiston records a specimen received from York Factory. h A specimen taken at Moose Factoiy and one from Cape Hope, Severn River, have been recorded Y>y Brewster/ The species is called throughout the region the 'black stock duck' to distinguish it from the mallard. Chaulelasmus streperus (Linn.). Gadwall. One was taken near Cape Churchill August 24, but was not pre- served. Bell collected the species on Hudson Bay.'' Mareca americana (Gmel.). Baldpate. Recorded by Murray from Hudson Bay.' A specimen from Nelson River is recorded by Baird as being in the Smithsonian collection/ Clarke records an adult male from Fort Churchill collected many years ago. fJ Nettion carolinensis (Gmel.). Green -winged Teal. An immature male was preserved from a lot killed for food by Indians at Fort Churchill July 25. Several green-winged teals were seen in small ponds about 15 miles north of Seal River August 18, and large flocks were started up from ponds between Fort Churchill and «Auk, XIX, p. 184, 1902. ''Ibis, V, p. 146, 1863. Rept. Prog. Can. Geol. Surv., 1878-79, A pp. VI, p. 69c ( L880). ■ Nat. Hist. Hull. Univ. Iowa, II, p. 257, IS!):!. d Edin. New Phil. Journ. (new ser.), IX, p. 229, L859. ■ Auk, IN. p. 14M, L886. /Kept. Prog. Can. Geol. Surv. L882-3-4, App. Ill, p. 55DD (1885). g Edin. New Phil. Journ. (new ser.), IX, p. 230, L859. October, 1902.1 BIRDS OF KKKWATIN. 85 Factory by James McKenzie are recorded in the catalogue of birds in the I . S. National Museum. Aythya americana (Evt.). Redhead. Nutting recorded this species and obtained specimens near Chema- wawin, Saskatchewan, in the neighborhood of which "countless num- bers of the red-head breed/'" Aythya vallisneria (Wils.). Canvas-back. Nutting obtained a specimen near the mouth of the Saskatchewan in the summer of L891.6 Aythya marila (Linn.). Scaup Duck. A specimen was taken August ~2'2 from a large flock on a pond near the shore of the Bay about midway between Fort Churchill and Cape Churchill. These birds were moulting their wing- quills, for which purpose they had evidently congregated on this pond, where they were safe from the attacks of predatory mammals. Aythya affinis (Evt.). Lesser Scaup Duck. The lesser scaup duck has been reported from a number of localities in the Hudson Bay region, though on account of its close resemblance to the greater scaup the accuracy of some of the records is doubtful. Baird recorded a specimen from Nelson River and considered Forster's record of marila from Severn River'' to refer to the present species. d Murray records it from Severn House,' and Bell reports it from Fort Churchill and York Factory. ' Clangula clangula americana (Bonap.). Golden-eye. Rather common on the larger lakes between Norway House and Oxford House. At the latter place we took an adult male July 3, but did not shoot any more on our northward trip, although wTe saw a flock containing about 200 near the outlet of Knee Lake July 6, and daily noted a few on our way to York Factory. After this we did not meet with the species again until we reached Hill River on our return, September 5. Here we found many, and on Knee Lake September 6 noted a large flock, probably the same we had observed there two months before. We saw a few more on the Echimamish September 15, one of which we collected, and found a large flock on Hairy Lake. Macoun records specimens taken at Fort Churchill/ and Forster notes the occurrence of the species on Severn River.A a Nat. Hist. Bull. Univ. Iowa, II, p. 257, 1893. &Ibid., p. 258, 1893. cPhil. Trans., LXII, p. 413, 1772. •i Birds of North America, p. 791, 1858. - Edin. New Phil. Journ. (new ser.), IX, p. 230, 1859. /Rept. Prog. Can. Geol. Surv. 1882-3-4, App. Ill, p. 55DI) (1885). 0 Catalogue Canadian Birds, Part I, p. 95, 1900. k Phil. Trans., LXII, p. 417, 1772. 86 NORTH AMERICAN FAUNA. [No. 22, Clangula islandica (Gmel.). Barrow Golden-e3re. Bell intimates that this .species, as well as the common golden -eye, is found on Hudson Bay." Seton, on the authority of R. H. Hunter, records it from Lake Manitoba, Shoal Lake, and the mouth of Red River, Manitoba.6 Charitonetta albeola (Linn.). Buffle-head. We saw an adult male on Red River, between Winnipeg and West Selkirk, June 11, and an immature bird on Hill River, near Swampy Lake, September 5. The species is recorded by Forster from Severn River/' by Murray from Severn House, Moose Factory, and Trout Lake,'7 and by Clarke from Fort Churchill/ It probably ranges throughout southern Keewatin. Harelda hyemalis (Linn.). Old-squaw. Though not observed by us, this species has been recorded from a number of localities on the west shore of Hudson Ba}\ Edwards figured a 'Long-tailed Duck from Hudson's Bay;'-7" Forster recorded the species from Churchill River ;g and Richardson described speci- mens killed at Winter Island, Melville Peninsula, on Parry's second voyage.7' Rae speaks of finding this species breeding on an island off Knaps River,* and collected a specimen at Repulse Bay which is recorded in the British Museum Catalogue. Murray recorded the species from Severn House ;'; Clarke, an adult pair in summer plumage from Fort Churchill;-? and Bell, a specimen from York Factory/" Histrionicus histrionicus (Linn.). Harlequin Duck. Forster recorded a specimen from the Hudson Bay region ;* Blakis- ton says he examined one at York Factory;"* and the catalogue of birds in the U. S. National Museum collection records a specimen taken in James Bay August 3, 1860. Camptolaimus labradorius (Gmel.). Labrador Duck. Joseph Sabine includes Anas lahradorla among the species which at that time were found on Hudson Bay and its vicinity, but which were a Proc. Royal Soc. Canada, 1882, I, Sec. IV, p. 50 (1883). 1> Auk, III, p. 328, 1886. cPhil. Trans., LXII, p. 417, 1772. dEdin. New Phil. Journ. (new ser.), IX, p. 230, 1859. « Auk, VII, p. 320, 1890. /Nat. Hist. Ill, pi. 156, 1750. 9 Phil. Trans., LXII, p. 418, 1772. A Appendix to Parry's Second Voyage, p. 373, 1825 (1827). * Narrative of an Expedition to the Shores of the Arctic Sea. p. 22, 1850. J Auk, VII, p. 320, 1890. fcRept. Prog. Can. Geol. Burv., L882-3-4, App. Ill, p. ttDD (1885). ZPhil. Trans., LXII, p. 419, 1772. wlbis, V, p. 149, 1863. October, 1902 BIRDS OF KEEWATIN. 87 not obtained on Franklin's Expedition." An adult male, perhaps from Hudson Bay, was presented to the British Museum by the Hudson's Bay Company about the year 1835;6 and it is likely that a specimen to which no definite locality is assigned that is recorded in the British Museum Catalogue is this same bird. Somateria mollissima borealis (C. L. Brehm). Northern Eider. This species was obtained on Parry's second voyage at Winter Island, being recorded as mollissima;0 and Blakiston, writing of S. mollissima, speaks of having received specimens from Hudson Bay.'' A specimen collected by Dr. Rae at Repulse Bay is recorded in the British Museum Catalogue. Somateria dresseri Sharpe. American Eider. First seen in small numbers about 50 miles north of York Factory July 19, and rather common from there northward. Flocks of a hundred or more were often seen north of Fort Churchill, and a female with young two or three weeks old was observed August 3, 50 miles south of Cape Eskimo. The species is said to breed in large numbers on certain rocky islands north of Fort Churchill. Quantities of the eggs are gathered by Indians and Eskimos in late April and earl}' May. and brought to Fort Churchill, where they are packed in salt for later consumption. The bird is called by the Indians 'Husky' (i. e., Eskimo) duck. Two specimens, collected by Dr. Bell at Fort Churchill, are recorded by Macoun/ Somateria v-nigra Gray. Pacific Eider. Murray recorded this species from Severn House/ As the species occurs on Great Slave Lake, it should be occasionally found in Keewatin. Somateria spectabilis (Linn.). King Eider. Edwards figures this species from Hudson Bay, calling it the 'Gray- Headed Duck.*? Linnaeus cites Edwards's figure in his description of spt otabilis, but does not base his description exclusively on it. It was recorded by Blakiston, who speaks of having seen specimens from Hudson Bay.7' and by Richardson, who states that it was observed in numbers about Melville Peninsula on Parry's second voyage.* A " Franklin's Narrative of a Journey to the Polar Sea, Appendix, p. (398, 1823. &See Dutcher, Auk, VIII, p. 203, 1891. 'Appendix to Parry's Second Voyage, p. 370, 1825 (1827). ''Ibis, V, p. 150, 1863. « Catalogue Canadian Birds, Part I, p. 105, 1900. /Edin. New Phil. Journ. (new ser. ), IX, p. 229, 1859. 9 Nat. Hist, III, pi. 154, 1750. ''Ibis, V, p. 150, 1863. 1 Appendix to Parry's Second Voyage, p. 373, 1825 (1827). 88 NORTH AMERICAN FAUNA. [No. 22, specimen from York Factory, collected by Dr. Bell, is recorded by Macoun/' Oidemia americana Swains. American Scoter. Svvainson based his description of this species on Hudson Bay speci- mens;6 a specimen collected by Dr. Rae at Repulse Bay is recorded in the British Museum Catalogue; Andrew Murray recorded the species from Trout Lake;e Blakiston says that he "received specimens from the west coast of Hudson's Bay;""' and an adult male is recorded from Fort Churchill by Clarke/ Oidemia deglandi Bonap. White-winged Scoter. We observed this species twice on Knee Lake — a small flock on July 7 and a few on September 8 as we were returning-. We also noted a few about fifty miles north of York Factory July 19. Many breed about the borders of small ponds throughout the interior, and large numbers are killed for food before they are able to fly. Murray recorded the species from Trout Lake, Moose Factory, and Severn House. c Oidemia perspicillata (Linn.). Surf Scoter. Rather common in Hudson Bay south of Cape Eskimo August 4 to 13. My brother took a specimen on Churchill River near Fort Churchill August 11, and we obtained another near Pine Lake Sep- tember 13, as we were returning. On September 11 we saw several near Robinson Portage. Edwards figured this species, which he called the ' Great Black Duck from Hudson's-Bay,'-; and on this figure Linnaeus based his description of Anas perspicillata. The species has since been several times recorded from Hudson Bay by different writers, and a specimen from Repulse Bay, collected by Dr. Rae, is mentioned in the British Museum Catalogue. Erismatura jamaicensis (Gmel.). Ruddy Duck. Blakiston speaks of having examined a specimen at York Factory, on Hudson Bay/ and Bell records a specimen from the same place.9. The species seems to be of regular occurrence about Lake Winnipeg and other points in Manitoba, and probably in the adjacent parts of Keewatin. aCatalogue Canadian Birds, Part I, p. 108, 1900. ''Fauna Boreal i- Americana, II, p. 450, 1831. cEdin. New Phil. Journ. (new ser.), IX, p. 231, 1859. ''Ibis, V, p. 150, 1863. ^Auk, VII, p. 320, 1890. /Nat. Hist., Ill, pi. 155, 1750. srRept. Prog. Can. Geol. Surv., 1878-79, App. VI, p. 69e (1880). Octobbb.1902.] BIRDS OF KKKWATIN. 89 Chen hyperborea nivalis (Forst.). Greater Snow Goose.0 This bird was first described by Forster from Severn River, Hudson Bay.6 Swainson and Richardson speak of its occurrence at Albany Fort and York Factory;' Murray records it from Moose Factory and Severn House;** and Hell characterizes it as abundant at Fort Churchill and York Factory during migrations,' and says that it began to arrive at the former place September 5/ Barnston says that at Martin Falls. Albany River, the species was generally passing in large flocks about May 10. g A specimen collected by Dr. Rae at Repulse Bay is recorded in the British Museum Catalogue; and one from Black Island, Lake Winnipeg, is recorded by Macoun.7' Chen caerulescens (Linn.). Blue Goose. Ed wards figured the 'Blue- Winged Goose1 from Hudson Bay,' and on this figure Linnseus based his description of the species. A specimen from Repulse Bay is recorded in the British Museum Catalogue, and one taken at Moose Factory in August, 1860, by J. McKenzie, and one from Fort Churchill, appear in the catalogue of birds in the U. S. National Museum. It is said to be found principally about the south- ern part of the Bay. and according to Indian information breeds in northern Labrador. Chen rossi (Cassin). Ross Snow Goose. A specimen procured at Fort Churchill, Hudson Bay, is recorded by Macoun.-' The 'Horned Wavey' of Hearne is doubtless this species. Anser albifrons gambeli (Hartl.). White-fronted Goose. Edwards figured a specimen procured from Hudson Bay, which he called the ' Laughing-Goose. '* Barnston says that this species is sel- dom seen in the southern part of Hudson Bay, but is less rare at York Factory, and is frequent at Fort Churchill. l Blakiston received a specimen from Hudson Ba}V and one from Repulse Bay, collected by Dr. Rae, is recorded in the British Museum Catalogue. a In some of the cases cited the species has been recorded as C. hyperborea, but I have assumed that the eastern form is referred to. 6 Phil. Trans., LXII, pp. 413, 433, 1772. c Fauna Boreali-Americana, II, p. 467, 1831. tfEdin. New Phil. Journ. (new ser.), IX, p. 225, 1859. eRept. Prog. Can. Geol. Surv., 1878-79, App. VI, p. 69c (1880). /Ibid., 1882-3-4, App., Ill, p. 55DD (1885). ?Edin. New Phil. Journ., XXX, p. 254, 1841. ^Catalogue Canadian Birds, Part I, p. 114, 1900. *Nat. Hist, III, pi. 152, 1750. ^Catalogue Canadian Birds, Part I, p. 115, 1900. fcNat. Hist., Ill, pi. 153, 1750. I Ibis, II, p. 257, 1860. ™Ibid., V, p. 141, 1863. 90 NORTH AMERICAN FAUNA. [No. 22, Anser fabalis (Lath.). Bean Goose. Under the name segetum, Richardson mentions this species as one of the geese which are known to visit the Hudson Bay region, but are rarely seen, being accidental visitors." Branta canadensis (Linn.). Canada Goose. An island in the northern part of Lake Winnipeg, on which this bird is said to nest in considerable numbers, was pointed out to us. While descending Steel River July 9 we took a half -grown bird from a flock of five. These had probably come down Fox River, where the species is said to breed. Young geese unable to fty, probably of this species, were seen by Alfred E. Preble on Churchill River August 11. While ascending Steel and Hill rivers, August 31 to Sep- tember 4, we saw one or two flocks daily. We shot one bird on Hill River September 1, but found its preservation impracticable. Murray recorded the species from Moose Factory and Severn House,6 and Bell says it breeds on Churchill River. '' In former times, when the posts on Hudson Bay supported a much larger population than at present, geese constituted a staple article of food, and this species and B. c. hutchinsi especially were shot in grpat numbers, both for immediate consumption and to be salted for winter. Barnston, from the number recorded at the different posts, estimated that at least 57,500 were annually killed on James Bay and the west coast of Hudson Bay/ At present the demand for the birds is less and their numbers are diminished; hence fewer are killed. Branta canadensis hutchinsi (Rich.). Hutchins Goose. Several flocks of geese referred to this species were seen on the Barren Grounds near Hubbart Point August 16. Dr. Rae saw a female with a brood of young near Neville Bay/ Macoun records two specimens taken at Fort Churchill by Dr. Bell, and an egg obtained at Repulse Bay/ The species has been recorded from other points in northern Keewatin. Branta bernicla (Linn.). Brant. Said by Swainson and Richardson to breed in great numbers on the coast and islands of Hudson Bay/ A specimen from Repulse Bay, collected by Dr. Rae, is recorded in the British Museum Catalogue; and Dr. Bell records a specimen killed at York Factory/' « Appendix to Parry'H Second Voyage, p. 364, 1825 (1827). &Edin. New Phil. Jonrn. (new ser.), IX, p. 225, L859. cRept. I'rog. Can. Geol. Surv., 1878-79, App., VI, p. 69c ( L880). <*Ibis, II, p. 258, 1860. 'Narrative of an Expedition to the ShoreH of the Arctic Sea, |>. 24, 1850. /Catalogue Canadian Birds, Part 1, p. 120, 1900. 9 Fauna Boreali-Americana, II, p. 469, 1831. /'Proc. Royal Soc. Canada, L882, I, Sec. IV, i>. 51 (1883). October, 190_>.] BIRDS OF KEEWATIN. 91 Branta leucopsis (Beehst.). Barnacle Goose. Said by Richardson to be rare and accidental in Hudson Bay." A specimen in the U. S. National Museum collection was obtained near Rupert House, James Bay, by Bernard R. Ross.6 Olor columbiamis (Ord). Whistling Swan. Whistling swans visit the western shores of Hudson Bay in great numbers in the spring and fall, and their assembled thousands are said to present a beautiful and imposing spectacle during their semiannual visits. The broad expanse of Churchill River near its mouth is a favorite place of resort. A specimen collected at Fort Churchill and presented by J. R. Spencer is recorded by Dr. Robert Bell,'' who also states that the species breeds on Nottingham Island, Hudson Bay/* Olor buccinator (Rich.). Trumpeter Swan. Richardson describes a specimen killed on Hudson Ba}V Barnston speaks of the breeding of swans near Eastmain Fort, on James Bay, and of eggs being brought to him from a nest on the banks of a lake near Norway House. -f His remarks probably refer to this species, as it has a more southern breeding range, especially in the interior, than O. oolurnMcmw. Botaums lentiginosis (Montag.). Bittern. We saw several flying over the marsh at the mouth of Red River June 15, heard the notes of one near Norway House June 19, and saw one near Sea Falls June 23. At Beacon Point, near York Factory, I started one from the marsh July 13. On our return trip we saw sev- eral near Oxford House September 10, and on the Echimamish Sep- tember 15. The species is recorded from Severn River by Forster// and from Fort Churchill by Clarke/' Ardetta exilis (Gmel.). Least Bittern. Under the name Botaurus minor Dr. Robert Bell records this species from York Factory,* and later says, "Of the least Bittern {Ardetta exilis Gray) I have specimens from Manitoba and York Factory.""' It has been taken near Winnipeg, but can scarcely reach Keewatin except as a straggler. « Appendix to Parry's Second Voyage, p. 364, 1825 (1827). *Baird, Brewer, and Ridgway, "Water Birds of North America, I, p. 475, 1884. cRept. Prog. Can. Geol. Surv., 1878-79, App. VI, p. 70c (1880). tfRept. Prog. Can. Geol. Surv., 1882-3-4, p. 30DD (1885). « Fauna Boreali-Americana, II, p. 464, 1831. /Ibis, II, p. 253, 1860. r. Rae at Repulse Bay is recorded in the British Museum Catalogue. Tringa maculata Vieill. Pectoral Sandpiper. First seen on the marshes about 50 miles north of York Factory July 19, where the species was rather common, and where a male was shot. Common on the meadows bordering Button Bay July 31, and abundant on the Barren Grounds south of Cape Eskimo August 3 to 13. Several seen near Oxford House September 10. A specimen collected by Dr. Rae at Repulse Bay is recorded in the British Museum Catalogue, and the catalogue of birds in the U. S. National Museum collection records a specimen taken at Moose Factory in 1881 by Walton Haydon. Tringa fuscicollis Vieill. White-rumped Sandpiper. Rather common on the meadows bordering Button Bay July 31, where a specimen was collected. A number were noted on the Barren Grounds below Cape Eskimo August 3 to 8. Tringa bairdi (Coues). Baird Sandpiper. Macoun says: " Spreadborough saw three on a small island in James Bay on June 16, 1896, and believes they were breeding. "ffi The species probabty occurs regularly in western Keewatin Tringa minutilla Vieill. Least Sandpiper. First met with at Fort Churchill, where adult birds accompanied b}^ young were seen on the meadows July 26 and an immature bird was taken. Large numbers were observed on the shores of Button Bay July 31, and immense flocks were seen on the Barren Grounds south of Cape Eskimo August 3 to 13, though at the latter date their num- bers had greatly diminished. The species was noted at nearly every place at which we landed on the way back to Fort Churchill, August 13 to 19; and it was doubtless present in flocks of small sandpipers seen between Fort Churchill and York Factory, though it was not again positively identified. Tringa alpina Linn. Dunlin. Blakiston speaks of having seen a specimen from Hudson Bay/' and the American Ornithologist Union Check List records it as accidental on the west side of Hudson Bay, though J have been unable to discover on exactly what information the record is based. Tringa alpina pacifica (Coues). Red-backed Sandpiper. First seen on the beach about 50 miles north of York Factory July 19, where numbers were observed and several specimens were taken. The birds seemed to bo just commencing their southward migration. a Catalogue Canadian Birds, Part 1, p. 162, 1900. ''Ibis, V, p. 132, 1863. October, 100-'.] BIRDS OF KEEWATIN. 97 They were soon in large numbers wherever we Landed on the coast north of Fort Churchill, wore present by thousands on the Barron Grounds south of Cape Eskimo August 3 to L3, and were still noon, though in diminishing numbers, wherever we landed on the way back to York Factory. Ereunetes pusillus (Linn.). Semipalmated Sandpiper. Rather common at Fort Churchill, where adults and young were taken on the meadows July 25. Common along- the coast north of Fort Churchill, and seen in immense flocks on the Barron Grounds south of Cape Eskimo. As with the red-backed sandpipers, we found the species common on the beach wherever we landed on our return down the coast to York Factory August 13 to 26. One taken at Moose Factory in L881 is registered in the catalogue of birds in the V. S. National Museum collection. Calidris arenaria (Linn.). Sandorling. A number were soon at my camp 25 miles south of Cape Eskimo on the morning of August 13. They were flying southward with other species. A specimen from Repulse Bay, collected by Dr. Rae, is recorded in the British Museum Catalogue, and Clarke records an adult in sum- mer plumage from Fort Churchill/' According to Swainson and Richardson, Hutchins reported that the species bred on Hudson Bay as low as the fifty -fifth parallel.* Limosa fedoa (Linn.). Marbled Godwit. Edwards figured a specimen brought from Hudson Bay by Mr. Isham/ His figure formed the basis of Linmeus's description of Scolopaxfi doa. Murray recorded the species from Hudson Bay/' and a specimen from Hayes River is recorded in the British Museum Cata- logue. The catalogue of birds in the U. S. National Museum collec- tion records a specimen from Moose Factory. Limosa haemastica (Linn.). Hudsonian Godwit. We first met with this species on the beach about 50 miles north of York Factory July 19, where a number wore seen and two were taken. It was common on the Barren Grounds south of Cape Eskimo August 4 to 8, and several were seen between Fort Churchill and Cape Churchill August 22, and below Cape Churchill August 2-i. Edwards figured a specimen brought from Hudson Bay by Mr. Isham. calling it the 'Red-breasted Godwit, '* and on his figure Lin- "Auk, VII, p. 321, 1890. >> Fauna Boreali-Americana, II, p. 336, 1831. 'Nat. Hist., Ill, pi. 137, 1750. <'Edin. New Phil. Journ. (new set-. |, IX, p. 225, 1859. 'Nat. Hist., Ill, pi. 138, 1750. 7165— No. 22 7 98 NORTH AMERICAN FAUNA. [No. 22, nseus based his description of Scolopax hsemastica. Forster recorded the species from Churchill River/' and Murray from Severn House.6 A specimen collected by Dr. Rae at Repulse Bay is recorded in the British Museum Catalogue. Totanus melanoleucus (Gruel.). Greater Yellow-legs. One was seen at Oxford House July 3 and one on upper Hill River July 7. I saw several on the beach near York Factory July 13, and at Fort Churchill July 24. The species was very common on the shores of Button Bay July 31, and about 50 miles below Cape Eskimo August 4: to S, and was noted wherever we landed on the way back to Churchill. While ascending Hayes River August 29 and 30 we observed a great many, usually in flocks of about a dozen, but after passing the mouth of the Shamattawa, which is in reality the main river, we saw none for several da}rs. Steel and Hill rivers, which compose the other branch of Hayes River, are not so well adapted to the habits of the species and it seems probable that most of the yellow-legs, and probably other sandpipers, migrate up the Shamattawa, where low, sandy shores more often occur. We saw two or three on Knee Lake September 7, one at Oxford House September 10, and several on Playgreen Lake September 17 to 20. The catalogue of birds in the U. S. National Museum collection con- tains the record of a specimen taken at Moose Factory in 1881. Totanus flavipes (Gmel.). Yellow-legs. First met with June 27 at Robinson Portage, where a pair were seen on the marsh and the male was collected. They probably had a nest in the vicinity, as they were very uneasy and frequently alighted on neighboring dead trees. The species was common at Fort Churchill July 22 to 30, and abundant at Button Bay July 31, and 50 miles below Cape Eskimo August 3 to 8. A few were noted on Hayes River August 30. The species is recorded by Andrew Murray from Severn House,6 and the catalogue of birds in the U. S. National Museum collection records a specimen taken at Moose Factory in 1881. Totanus totanus (Linn.). Common Redshank. The only evidence that this is a North American species is a descrip- tion by Swainson and Richardson of a specimen from Hudson Bay of ' Totcmus calidris, the Redshank or Gambct,1 which they said existed in the British Museum.'' The species occupies a place in the 'Hypo- thetical List' of the American Ornithologists' Union Check List on the strength of this record. a Phil. Trans., LXII, p. 411 (Scolopax lapponica), 1772. Mvlin. New Phil. Journ. (newser.), IX,]). 225, 1859. cFauna Boreali-Americana, II, p. 391, 1831. (See also Coues, Auk, XIV, p. 211, 1897. Octobeb, 1902. BIRDS OF KEKWATIN. 99 Helodromas solitarius (Wils.). Solitary Sandpiper. We met with this species only while descending the Echimamish, where several were seen September 14 and L5 and one was taken. A specimen collected at Repulse Bay by Dr. Rao is recorded in the British Museum Catalogue, and A. McKenzie took one at Moose Fac- tory. The species probably occurs throughout the region. Helodromas ochropus (Linn.). Green Sandpiper. Swainson and Richardson described a specimen of Totcmus ochropus, the 'White-tailed Tattler.' from Hudson Ray." This specimen is recorded in the British Museum Catalogue. Heteractitis incanus (Gmel.). Wandering Tattler. Bell says: "I obtained a specimen of the Wandering Tattler {Ilet- trnr, Jus brevipes, Vieill.), supposed to be a western species, on the Eastmain Coast.* Bartramia longicauda (Bechst.). Bartramian Sandpiper. Several were seen and one was shot on the Barren Grounds 50 miles below Cape Eskimo August 8, and several more were noted 25 miles farther north August 10 to 13. Tryngites subruficollis (Vieill.). Buff-breasted Sandpiper. I saw a number on the higher parts of the tundra 25 miles south of Cape Eskimo August 10 to 13, and noted others (collecting two) August 21 on some sandy ridges, the remains of old shore lines, below Cape Churchill. The birds had a soft, plaintive call, and were rather tame and unsuspicious. A specimen taken by Dr. Rae at Repulse Ba}T is recorded in the British Museum Catalogue. Actitis macularia (Linn.). Spotted Sandpiper. This widely distributed species is found throughout the region as far north at least as Fort Churchill. It was common on Red River June 11, and seen daily between Norway House and York Factory June 23 to July 10. A deserted nest was found on the rocky island that constitutes Rock Portage, at the lowest of the Hill River rapids. While we were descending Steel and Hayes rivers, July 9 and 10, one or more pairs of this species were almost constantly in sight, and voung ones unable to fly were often seen running along the shelving banks. The species was not seen on the shores of Hudson Bay, but we collected a specimen on Churchill River, near Fort Churchill, July 25, and my brother found it rather common on the same stream, about 2<» miles from its mouth, early in August. While ascending "Fauna Boreali-Americana, II, p. 392, 1831. &Proc. Royal Soc. Canada, 1882, I, Sec. IV, p. 51 (1883). 100 NORTH AMERICAN FAUNA. [No. 22, Hayes, Steel, and Hill rivers on our return, August 28 to September 3, we found that many were still lingering there, and I saw several at Duck Point, Playgreen Lake, on September 20. The catalogue of birds in the U. S. National Museum collection contains the record of a specimen taken at Moose Factory in 1881 by Walton Hay don. Numenhis hudsonicus Lath. Hudsonian Curlew. ' First observed about 50 miles north of York Factory, where a number were seen and one was shot July 19. Another was shot at Fort Churchill July 28. Quite a number were seen on the shores of Button Bay July 31, and to the north of that place on the following da}r. We saw a few daily while we were encamped on the Barren Grounds south of Cape Eskimo, August 10 to 13; and noted a few more below Hubbart Point August 18, between Fort Churchill and Cape Churchill August 22, and below the cape August 24. The birds were usually seen at low tide, when they flew about in small compa- nies over the broad boulder-strewn flats in search of feeding grounds, uttering a loud, clear whistle. During high tide the}7 resorted to the tundra and were less observable. Latham based his description of this species mainly on a specimen received from Mr. Hutching, probably from Severn River/' Murray recorded the species from Severn House.6 Numenhis borealis (Forst.). Eskimo Curlew. This species was first described by Forster from a specimen taken at Albany Fort, Hudson Bay/ It is recorded by Dr. Bell as abundant at Fort Churchill in August, 1879/* Squatarola squatarola (Linn.). Black-bellied Plover. A specimen in winter plumage, evidently a bird of the year, was killed on a small rocky island in Swampy Lake September 5. Forster recorded the species from Severn River;'' Richardson from Melville Peninsula, where he says it breeds/ and Bell from York Factory/ Clarke recorded two adults in summer plumage from Fort Churchill/' In former years the various plovers, particularly this species and the golden plover, were much hunted at the various posts on the Bay. especially during their southward movement in August. "Index Ornithologicus, [I, p. 712, 1790. (Sec also Latham, Syn. Suppl., I, p. 243, L787. I &Edin. New Phil. Journ. (new ser.), IX, p. 225, 1859. 'Phil. Trans., LXII, pp. 411, 431, 1772. d Rept. Prog. Cani Geol. Surv., 1878-79; App. VI, p. 69c ( L880). « Phil. Trans., LXII, p. 412, 1772. /Appendix tp Parry's Second Voyage, p. 352, L825 (1827). '/Kept. Prog. Can. Geol. Surv., L882-3-4, App. Ill, p. 55DD (1885). /'Auk, VII, p. 321, 1890. October, wol\] BIRDS OF KKKWATIN. 101 Charadrius dominicus Miill. Golden Plover. I found this species moving southward in small Hocks at a point 50 miles south of Cape Eskimo August 4 to 8, and also, though in diminished- numbers, 25 miles to the northward. August 1" to L3. A specimen was taken at the first point. It has the black of the lower parts varied by a number of whitish feathers, which predominate on the throat and the sides of the chest. This species was recorded by J. C. Ross from Port Bowen, Prince Regent Inlet, where it arrived the middle of May;" by Murray f rom Trout Lake,6 and by Bell from York Factory.'' Clarke recorded an adult in summer plumage from Fort Churchill.'' The catalogue of birds in the V . S. National Museum collection contains the record of a specimen taken at Moose Factory in 1881 by Walton Havdon. .ffigialitis vocifera (Linn.). Killdeer. Several were seen on the grassy meadows about midway between Fort Churchill and the mouth of Churchill River on July 24. Though shy. they seemed very solicitous and probably had young in the vicinity. .ffigialitis semipalmata Bonap. Semipalmated Plover. First noted as we were descending Hayes River July 10. Here, on a sandy island about 25 miles above York Factory, the birds were breeding and downy young were running about. We saw a number of old birds at Beacon Point, York Factory, July 13, and six days later, on landing at a spot 50 miles farther north, found the species common. At Fort Churchill it was abundant, and we collected, on July 24, both adults and downy young. As I passed up the coast, July 30 to August 8, it continued rather common; and it was still represented, though sparingly, at the northernmost point of the trip, about 25 miles south of Cape Eskimo, during my four days' stay there (August 10 to 13). On our return trip it had evidently migrated, as we saw nothing of it. Andrew7 Murray recorded the species from Trout Lake and Severn House:' and a specimen from Repulse Bay, collected by Dr. Rae, is recorded in the British Museum Catalogue. iEgialitis meloda circumcincta Ridgw. Belted Piping Plover. Under the name Charadrius //telodus, King records a bird, probably referable to the present form, which he shot on Lake Winnipeg, near the northern end, while he was traveling between the mouth of the Sas- « Parry's Third Voyage, Appendix, p. 102, 1826. &Edin. New Phil. Journ. (newser.), IX, p. 229, 1859. cRept. Prog. Can. Geol. Surv., 1882-3-4, App. Ill, p. 55DD (1885). ''Auk, VII, p. 321, 1890. *Edin. New Phil. Journ. (new Ber.), IX, p. 225, 1859. 102 NORTH AMERICAN FAUNA. [No. 22, katchewan and the outlet of the lake/' This specimen is described by Richardson in the appendix to Captain Back's narrative.6 Speci- mens referable to circumcincta have since been taken on Lake Winni- peg on several occasions. Arenaria morinella (Linn.). Ruddy Turnstone. The hrst turnstones we met with were feeding on the beach at the 'whale fishery,? Fort Churchill, July 30. On August 1 we again noted the species a few miles north of Fort Churchill, and August 10 to 13 observed many small Hocks about 25 miles south of Cape Eskimo. On August 14 I took a specimen just after starting down the coast on my return, and from Fort Churchill to York Factory, August 21 to 26, we saw many flocks daily. Whenever the birds perceived our boat they would approach and circle about it very slowly, sometimes coming within a few yards of us. Whether they mistook the boat for a rock on which they designed to alight, or were impelled merety by curiosity, 1 could not tell. Under the specific name of mterpres this bird has been recorded by Forster from Severn River/ by Bell from York Factory, (/ and by Clarke from Fort Churchill/ Swainson and Richardson say it breeds on Hudson Bay/ as it probably does to the northward. The British Museum contains a specimen from Repulse Bay, collected by Dr. Rae. Canachites canadensis (Linn.). Hudsonian Spruce Grouse. A pair in worn breeding plumage were collected at Oxford House July 3. A flock was started as we were ascending Hayes River, August 30, and others were seen on Hill River September 3 and 4. A female that was shot September 4 was preserved, and also an adult male of several that were killed on the Echimamish, September 14 and 15. Linnaeus based his description of Tetrao canadensis on Edwards's figures of a male and female from Hudson Bay, probably from the west coast;-*7 Forster recorded the species from Severn River;7' Murray received it from Trout Lake;' Bell reported it from York Factory;'' and a specimen from Fort Churchill is recorded by Clarke.' Speci- " Narrative of a Journey to the Shores of the Arctic Ocean, 1 1, p. 229, 1836. '' Narrative of the Arctic Land Expedition to the Mouth of the Great Pish River, App., p. 509, 1836. '• I'hil. Trans., LXII, p. 412, 1772. dRept. Prog. Can. Ueol. Surv., L878-79, App. VI, p. 68c (1880). ' Auk, VII, p. 321, 1890. ./'Fauna Boreali-Americana, II, p. 371, 1831. '/Nat, Hist., II, j)I. 71, female, 1747; III, pi. IIS, male, 1750. i< Phil. Trans., LXII, p. 389, 1772. < Edin. New I'hil. Joum. (new Her.), IX, p. 224, 1859. /Kept. Prog. Can. Geol. Surv., 1882-3-4, App. Ill, p. 551)1) (1885). October, 1902.] BIRDS OF KKEWATIN. 103 mens collected at Moose Factory are recorded in the catalogue of birds in the V . S. National Museum collection, and the species doubtless ranges throughout the wooded portions of Keewatin. Bonasa umbellus togata (Linn.). Canadian Rutted Grouse. We heard grouse drumming on the Echimamish during the night of June 24, and near the south end of Oxford Lake June 30. We did not note the species again until September s. on our return trip, when a female was taken on Knee Lake. While making a portage on Trout River, September !». I heard a grouse drumming, and entering the thicket soon located the bird on its drumming stand, a low rock, and shot it. Within a few minutes, as 1 carried the dead body on my hand toward our embarking place, I noticed that its wings had become raised over the back until they were nearly in contact and were firmly fixed in that position, apparently owing- to the contraction of the muscles of the wings, which had been exercised so violently. As the body cooled the wings dropped to their normal position. Near the south end of Oxford Lake, September 11, a few were seen, and near Hairy Lake, September 15, several were shot, a male and a female of which were preserved. Forster recorded the species from Albany Fort and Severn River," and Bell reported it rare as far north as York Factory.6 Lagopus lagopus (Linn.). Willow Ptarmigan. First seen and a line pair shot on the tundra about 50 miles north of York Factory. Specimens were also taken Juhr 24 in the vicinit}7 of Fort Churchill, where the species was rather common. While encamped on Churchill River, about 15 miles above Fort Churchill, early in August, Alfred E. Preble observed them in considerable num- bers. They were rather common on the Barren Grounds south of Cape Eskimo August 3 to 13, where two immature specimens were secured. Forster recorded this species from Severn River,'' and Swainson and Richardson state on the authority of Hutchins that 10,000 were cap- tured in a single season at that place. rf James C. Ross recorded it from Port Bowen, Prince Regent Inlet, where it remained as late as November 16, and returned about the middle of March.' Barnston states that it arrived at Martin Falls, on Albany River, about October 20/ and that it began to assume its summer plumage about March 20, at which time also it departed northward.? In the interior of Keewatin this species regularly goes south in winter as far at least as Norway a Phil. Trans., LXII, p. 393, 1772. &Rept. Prog. Can. Geol. Surv., 1878-9; App. VI, p. 68c (1880). cPhil. Trans., LXII, p. 390, 1772. '■' Fauna Boreali-Americana, II, p. 351, 1831. e Parry's Third Voyage, Appendix, p. 99, 1826. /Edin. New Phil. Journal. XXX, p. 256, 1841. Plbid., p. 253, 1841. 104 NOKTH AMERICAN FAUNA. [No. 22, House, and forms a welcome addition to the winter bill of fare at the Hudson Bay posts. Definite data are lacking- regarding- the southern limit of its breeding range on the west coast of Hudson Bay, but on the east coast it breeds as far south as .Tames Bay, where Spread- borough found it nesting at a point a short distance north of Fort George. a Lagopus rupestris (Gmel.). Rock Ptarmigan. First described by Gmelin, who based the name on Pennants 'Rock Grouse1 from Hudson Bay.6 We did not meet with the species, as its summer home lies to the north of the region visited. The people of Fort Churchill say it occurs at that post regularly in winter, and Clarke has recorded from there a pair in winter plumage.'' Swainson and Richardson, quoting Hutchins\s manuscript, say that the species reaches York Factoiy^ and Fort Severn in very severe winters/7 J. C. Ross says that the species left Port Bowen, Prince Regent Inlet, in October and returned in March. e Dr. Rae saw a rock ptarmigan with a brood of young near Cape Fullerton/ and specimens collected by him at Repulse Bay are recorded in the British Museum Catalogue. Pedicecetes phasianellus (Linn.). Sharp-tailed Grouse. Two males in worn breeding plumage were taken at Norway House June 18, and downy young at Oxford House early in July. We did not again note the species until September 11, when we saw a pair on the upper Echimamish. It was common at Norway House September 16 to 19, and a specimen in fall plumage was taken at that point. The tracks of a large flock were seen on the sandy shore at Duck Point, Playgreen Lake, September 19. Edwards figured and described a specimen from Hudson Bay, call- ing it the 'Long-Tailed Grous from Hudson's Bay,'0' and this figure and description formed the basis of LinnsBUs's description of Tetrao p/umanellttx. Forster recorded it from Albany Fort and Severn River,7' and Murray from Trout Lake.' J. B. Tyrrell saw it near York Factory.-' The catalogue of birds in the U. S. National Museum collection contains the record of one taken at Moose Factory in 1881. «Macoun, Catalogue Canadian Birds, Part I, i>. 206, 1900. '-Arct. Zool., II, p. 312, 1785. f North America, p. 16, L858. 'Auk, VII, p. 322, 1890. /Kept. Prog. Can. Geol. Surv., L878-79, App. VI, p. 67c (1880). sTbid., 1882 :: 4, App. Ill, p. r>4I>I> (1885). /'I'roc. U. S. Nat. Mils., VIII, p. 244, L885. October, 1902. ] BIRDS OF KEEWATIN. 107 This species is recorded by Murray from Severn House and Trout Lake," and an apparently immature specimen, collected at Fort Churchill many years ago, is recorded by Clarke.6 Aquila chrysaetos (Linn.). Golden Eagle. We saw several of these birds as we were passing through Hell Gate Gorge June 28, and noticed at least two of their nests on its rocky walls. Edwards figured a specimen from Hudson Bay,'' and Linnaeus based his description of Falco canadensis on Edwards's figure. Sir John Franklin, who passed through Hell Gate Gorge in October, 1819, speaks of a nest of the brown fishing-eagle on one of the projecting elitfs.'' Haliaeetus leucocephalus (Linn.). Bald Eagle. We were informed that white-headed eagles were occasionally seen near Fort Churchill and nested in the vicinity, and we obtained the upper mandible of one that had been killed there. Mr. Campbell, of Oxford House, reported having seen the species near Oxford Lake. Falco islandus Bri'mn. White Gyrfalcon. The residents of Fort Churchill spoke of a white hawk, probabty this species, which is sometimes seen there. Richardson described a mature bird from Hudson Bay;' Murray recorded F. candicans from York Factory \a Ridgway described a specimen from Moose Factory •/ and a specimen collected by Dr. Rae at Repulse Bay. is recorded in the British Museum Catalogue. Falco rusticolus gyrfalco (Linn.). Gyrfalcon. Forster described Falco sacer, probably identical with gyrfalco, from a specimen taken at Severn River; !/ Bell records a specimen (as Falco sacer) from York Factorv;/' and Clarke records two specimens from Fort Churchill.6 Falco peregrinus anatum (Bonap.). Duck Hawk. One was seen on the Barren Grounds below Cape Eskimo about August 5. The species is recorded by Murray from Trout Lake and Severn House,' and by Bell from York Factory/' and also from Marble "Edin. New Phil. Journ. (new Her.), IX, p. 221, 1859. 6 Auk, VII, p. 322, 1890. Nat. Hist., I, pi. 1, 1743. . 262, 1892. October, 1902.] KIRDS OF KEEWATIN. 109 Asio wilsonianus (Less.). Long-eared Owl. Thompson, quoting Hutchins's manuscript, says that this species was found at Severn Settlement, presumably Fort Severn, where Hutchins resided.'' Asio accipitrinus (Pall.). Short-oared Owl. Rather common and quite generally distributed throughout the region wherever favorable ground occurs. One was seen at Beacon Point, near York Factory, July 13, and two were taken at Fort Churchill. One was seen on the Barren Grounds south of Cape Eskimo August -k and one at Oxford House September 10. Murray recorded the species from Trout Lake Station, h and Bell from York Factory and F\>rt Churchill.'' The catalogue of birds in the U. S. National Museum collection contains the record of one taken at Moose Factory in 1881 by Walton Haydon. Syrnium varium (Barton). d Barred Owl. Strix varius Barton, Frag. Nat. Hist. Penna., p. 11, 1799. Syrnium nebulosum authors (not Stri.r nebulosa Forster, Phil. Trans., LXII, pp. 386, 424, 1772, which is based on the great gray owl. Several specimens of the barred owl taken at Moose Factory are recorded in the catalogue of birds in the U. S. National Museum. Nutting records the species from Chemawawin, Saskatchewan/ Scotiaptex nebulosum (Forst.). Great Gray Owl. stri.r nebulosa Forst., Trans. Phil. Soc. London, LXII, pp. 386, 424, 1772. (Severn River.) stri.r einerea Gmel., Systema Naturae, I, p. 291', 1788; and of authors. Forster based the name Strix nebulosa on a specimen of the great gray owl sent by Mr. Graham from Severn River, Hudson Bay.-' His description, in part, is as follows: Strix capite la'vi, corpore fuseo, albido undulatim striate, remige sexto longiore apice, nigricante. Description. — Rostrum fusco flavum, mandibula superius magis flava. Oculi magna iridibus flavis. * * * Pectus albidum maculis longitudinalibus trans versisque fuscis. Abdomen album superius uti pectus maculis longitudinalibus sed inferius striis trans- versis notatum. Latitudo pedum quattuor. «Proc. U. S. Nat. Mus., XIII, p. 540, 1890. 6 Edin. New Phil. Journ. (new ser.), IX, p. 222, 1859. <-Rept. Prog. Can. Geol. Surv., 1878-79, App. VI, p. 67c (1880). d stri.r nebulosa Forster plainly having been based on the great gray owl, the name Strix varius Barton seems to be the next name available for the barred owl The barred owls will, therefore, stand as follows: Syrnium varium (Barton), Frag. Nat. Hist. Penna., p. 11, 1799. Syrnium varium attend (Ridgw. ), Proc. U. S. Nat. Mus., Ill, p. 8, March, 1880. Syrnium varium Jielveolum (Bangs), Auk, XVIII, p. 299, 1901. 'Nat. Hist. Bull. Univ. Iowa, II, p. 270, 1893. /The fact that Forster's description of Strix nebulosa refers to the great gray owl was brought to my attention by Mr. R. Ridgway. 110 NORTH AMERICAN FAUNA. [No. 22, Blakiston received two specimens from York Factory," and Turner has recorded the species from Moose Factory, where a specimen was collected bjr J. McKcnzie.* Seton records it as resident in the woodo about Lake Winnipeg. c Nyctala tengmalmi richardsoni (Bonap.). Richardson Owl. A specimen collected at Repulse Bay by Dr. Rae is recorded in the British Museum Catalogue. Fielden records the species from near York Factory, where it was obtained by Captain Markham in 1886. d It probably occurs throughout the Hudson Bay region. Nyctala acadica (Gniel.). Saw-whet Owl. Striae passerina, recorded by Forster from Hudson Bay, probably from Severn River,'- is in all likelihood referable to this species. Turner recorded a specimen, which is still in the National Museum, collected at Moose Factory by J. McKenzie.* Megascops asio (Linn.). Screech Owl. George Barnston, writing in 1810, mentions "The small owl (Scops)" as being heard in April at Martin Falls, Albany River/ Bubo virginianus arcticus (Swains.). Arctic Horned Owl. The characteristic notes of the great horned owl were heard during the night of September 11, while we were encamped near the south end of Oxford Lake, September 13 at Robinson Portage, and the next day on the Echimamish. The catalogue of birds in the U. S. National Museum collection records two specimens of the great horned owl collected at Moose Factor}7 by J. McKenzie. One of these has been examined and is referable to this form. Clarke recorded a specimen probably referable to this form collected many years ago at Fort Churchill. 0 Fielden records a line specimen obtained near York Factory in 1886 by Captain Markham.'' Nyctea nyctea (Linn.). Snowy Owl. We did not meet with this species but saw wings at several posts, and learned from a young man at Fort Churchill that he had seen a snowy owl late in July. Forster early recorded it from Churchill River,' and the reports of a Rept. Pro«. Can. Geol. Surv., V, p. 50, 1863. &Proc. U. S. Nat. Mus., VIII, p. 24.S, 1885. '•Auk, III, p. 155, 1886. d Transactions Norwich Society, IV, p. 349, 1887. < Phil. Trans., LXII, p. 385, 1*772. / Edin. New. Phil. Journ., XXX, p. 253, L841. P. Auk, VII, p. 322, 1890. October, 1902.] BIRDS OF KEEWATIN. Ill various Arctic expeditions aote its occurrence at several points to the north and northwest of Hudson Bay. Its presence throughout the region in winter is attested by various observers. Surnia ulula caparoch (Mull.). Hawk Owl. "The Little Hawk Owl'" of Edwards, from Hudson Bay. formed the basis of Muller's Strix caparoch. Forster recorded the hawk owl, under the name Strix funerea, from Severn and Churchill rivers;6 Murray received it from Trout Lake and Severn House;' Swainson and Richardson mention a specimen from York Factory;"' and one col- lected by Dr. Rae at Repulse Ray is recorded in the British Museum Catalogue. The catalogue of birds in the U. S. National Museum collection contains the record of a specimen taken by Walton Haydon at Moose Factory in L881. Ceryle alcyon (Linn.). Belted Kingtisher. We found the belted kingfisher common throughout the region between Norway House and Oxford House. A large clay bank on the shore of Oxford Lake near Oxford House, which we passed June 30, was perforated by the nesting holes of a numerous colony of bank swallows. In this bank were also several larger holes, which were probably those of kingfishers, judging from the anxiety mani- fested by at least four pairs of kingfishers that were flying about. During our return trip we saw several kingfishers on Steel and Hill rivers early in September, on Trout River September 9, and on the Echimamish September 14. Clarke recorded an adult male from Fort Churchill,'' and several have reported the species from York Factory. The catalogue of birds in the IT. S. National Museum collection contains the record of a specimen taken at Moose Factory in 1881 by Walton Haydon- Dryobates villosus leucomelas (Bodd.). Northern Hairy Woodpecker. Forster recorded villosus from Severn River,' and Bell reported it from York Factory/' both probably referring to the present form. An adult male of this species from Fort Churchill is recorded by Clarke.- It is likely the bird ranges throughout the wooded portion of the region. Dryobates pubescens medianus (Swains.). Downy Woodpecker. A specimen collected at Moose Factory by Walton Haydon is in the U. S. National Museum. The downy woodpecker undoubtedly ranges throughout southern Keewatin. a Xat. Hist., II, pi. 62, 1747. (See Stejneger, Auk, I, p. 362, 1884.) 6 Phil. Tnm*., LXII, p. 385, 1772. cEdin. New Phil. Journ. (new Ber. ), IX, p. 221, 1859. d Fauna Boreali-Aniericana, II, p. 94, 1831. *Auk, VII, p. 322, 1890. /Phil. Trans., LXII, p. 388, 1772. ?Rept. Prog. Can. Geol. Surv., 1882-3-4, App. Ill, p. 54DD (1885). 112 NORTH AMERICAN FAUNA. [No. 22, Picoides arcticus (Swains.). Arctic Three-toed Woodpecker. We collected two males in a tract of burnt spruce woods at Norway House June 19, and while ascending- fiayes River August 30 saw another in spruce woods. Baird recorded a pair from Hudson Bay obtained from John Gould." Fielden recorded a specimen obtained near York Factory in August, 1886.6 The catalogue of birds in the United States National Museum collection records a specimen taken at Moose Factory in 1881 by Walton Hay don. Picoides americanus Brehm. Striped -backed Three-toed Woodpecker. Forster recorded this species from Severn River; c Murray received one from Severn House ;'; Baird recorded a male from Hudson Bay,e supposed to be the one figured by Audubon, and Clarke recorded several adults of both sexes collected at Fort Churchill many years ago/ Sphyrapicus varius (Linn.). Yellow-bellied Sapsucker. An adult female collected at Fort Churchill more than fifty years ago is recorded by Clarke. Proc. U. S. Nat. Mus., Kill, p. 551, 1890. I Birds of North America, p. 107, 1858. ./Phil. Trans., LXII, p. 387, 1772. *Edin. New Phil. Journ. (new ser.), IX, p. 223, 1859. October, 1902.] BIRDS OF KKKWATIN. 113 Antrostomus vociferus (Wils.). Whip-poor-will. Boll says: "The Whippoorwill was not seen nor heard north of Norway House,"0 implying its occurrence at that point. Bendire, probably referring1 to the same record, says: wk As far as I have been able to ascertain, this species roaches the extreme northern limits of its range on the north shore of Lake Winnipeg, near Norway House.6 The species probably occurs regularly in extreme southern Keewatin. Chordeiles virginianus (Ginel.). Night-hawk. Rather common at Norway House June 17 to 23, and several seen at Robinson Portage June 27, Oxford Lake June 30, and about the shores of Knee Lake July 5. They seemed to frequent entirely the districts which had been swept by tire. As we saw none on our return early in September, they had undoubtedly migrated by that time. Murray recorded the species from Trout Lake.' Bell reported it from York Factory." and a specimen collected years ago at Fort Churchill is recorded by Clarke. d Edward Sabine recorded one that was picked up dead on Melville Island.' The catalogue of birds in the U. S. National Museum collection contains the record of a speci- men taken at Moose Factory in 1881 by Walton Haydon. Muscivora forficata (Gmel.). Scissor -tailed Flycatcher. Bell recorded a " specimen of Milwtlus forficatm in the Govern- ment Museum shot at York Factory in the summer of 1880. 'V The species has also been reported from Manitoba. '■' It evidently occurs only as a rare straggler. Tyrannus tyrannus (Linn.). Kingbird. The catalogue of birds in the LT. S. National Museum collection records a specimen collected at Moose Factory July 11, 1881, by Wal- ton Haydon. Nutting found the species abundant at Grand Rapids, at the mouth of the Saskatchewan. h It should occur regularly in southern Keewatin. Sayornis phcebe (Lath.). Phoebe. A pair had a nest beneath the wharf at Norway House, and several more nests were observed on June 28 as we were passing through Hell Gate Gorge. They were placed on the face of cliffs overhanging the water, and contained young nearly ready to fly. Though the bird "Kept. Prog. Can. Geol. Surv., 1878-79, App. VI, p. 68c (1880). ''Life Hist. N. A. Birds (U. S. Xat. Mus. Special Bull. 3), p. 140, 1895 (1896). ' Ivlin. New Phil. Journ. (newser.), IX, p. 222, 1859. ''Auk, VII, p. 322, 1890. • Suppl. tu Appendix, Parry's First Voyage, p. exciv, 1824. /Proa Royal Soc. Canada, 1882, I. Sec. IV, p. 52, (1883). !/Seton, Auk, II, p. 218, 1885. ''Xat. Hist. Bull. Univ. Iowa, II, p. 271, 1893. 7165— No. 22 8 114 NORTH AMERICAN FAUNA. [No. 22, should occur throughout southern Keewatin, I find no published rec- ords of its occurrence in the Province. Contopus borealis (Swains.). Olive-sided Flycatcher. Observed but once, on July 4, in a swamp bordering Trout River, between Oxford House and Knee Lake. The bird was perched on a tall dead tree, uttering at intervals its characteristic note. Murray recorded the species from Hudson Bay." Empidonax trailli alnorum Brewst. Alder Flycatcher. Fly catchers referred to alnorum were several times observed by us at Norway House, and while we were ascending the Echimamish, but various causes, including their extremely wary habits, conspired to prevent their collection. Empidonax minimus Baird. Least Flycatcher. A nest containing well-incubated eggs was collected near the south end of Oxford Lake on the morning of June 30. The female parent was secured at the same time. I have compared this bird with most of the specimens in the large series in the U. S. National Museum collection, and it proves to have a smaller bill than any of them. Turner recorded a specimen collected by C. Drexler at Moose Factory. h Otocoris alpestris (Linn.). Horned Lark. An adult male horned lark, collected at Moose Factory June 18, 1863, and now in the U. S. National Museum collection, has been sev- eral times recorded. This specimen proves, on examination, to be referable to the typical form. Otocoris alpestris hoyti Bishop. Hoyt Horned Lark. We first met with this form on the 'Barrens'1 about 50 miles north of York Factory July 19, and found it common from there north as far as we went, especially at Fort Churchill, where adult birds and a young one not long from the nest were taken July 24 to 2(>. A speci- men was collected 50 miles south of Cape Eskimo August 4. During our return Ave saw many horned larks on the clearing at Norway House, September lt'» to 19, feeding in company with Lapland long- spurs. Bishop has recorded specimens of this form from Depot Island, ' Hudson Strait' | Hudson Bay].' J. C. Ross took a specimen near Felix Harbor, Boothia.'7 Richardson recorded a specimen taken July 10, "Ivliu. New Phil. Journ. (newser.), IX, p. 223, 1859. b I'.nc. r. S. Nat, Mus., VIII, p. 242, LS85. 'Auk, XIII, p. 132, 1896. '' Appendix to Ross's Second Voyage, p. xxvi, is:;,"). October, 1902.] BIRDS OF KEEWATIN. 115 L822, near Cape Wilson. Melville Peninsula;" and Murray received specimens from York Factory and Severn House/' Forster recorded horned larks from Albany Fort.' but in the absence of specimens, it is impossible to decide whether alpestris or hoyti is referred to. Pica pica hudsonia (Sab.). Magpie. Forster recorded the magpie from Albany Fort under the name Corviis pint.1' Thompson, quoting Hutchins's manuscript, says that one was caught in a marten trap at York Factory;' and Joseph Sabine, who described hudsonia from Cumberland House, speaks of having had a specimen from Hudson Bay in his possession some time before.'' Fielden records a specimen procured at York Factory/' Cyanocitta cristata (Linn.). Blue Jay. Several were seen in the groves of Quereus macrocarpa at West Selkirk on the morning of September 22. According to Chamberlain, the species has been taken at Moose Factory/' Russell records it from Grand Rapids/ Perisoreus canadensis (Linn.). Canada Jay. This species was reported about Norway House, and was seen nearly every day on our journe}' between there and York Factory. Speci- mens were collected on the Echimamish, at Oxford House, and at York Factory; and Alfred E. Preble saw several near Fort Churchill. Murray recorded the species from Severn House,7' and Clarke an adult from Fort Churchill/ The catalogue of birds in the U. 8. National Museum collection records a specimen taken at Moose Fac- tory in 1881 by Walton Haydon. Corvus corax principalis Ridgw. Northern Raven. We saw several between Robinson Portage and Pine Lake June 28, and while descending Hill River July 8 noticed a pair flying about the face of a high clay bank. Except for one seen at Fort Churchill July 30, we did not again note the species. James Clark Ross speaks of a pair which wintered at Port Bowen, Prince Regent Inlet/ Bell reported this species as breeding through- a Appendix to Parry's Second Voyage, p. 343, 1825 (1827). *>Edin. New Phil. Journ. (new ser.), IX, p. 398, 1859. cPhil. Trans., LXII, p. 398, 1772. ''Ibid., p. 387, 1772. ' Proc U. S. Xat. Mus., XIII, p. 565, 1890. /'Franklin's Narrative of a Journey to the Polar Sea, Appendix, p. 071, 1823. 9 Transactions Norwich Society, IV, p. 348, 1887. '' Catalogue Canadian Birds, p. 75, 1887. » Explorations in the Far North, p. 2f>4, 1898. JEdin. New Phil. Journ. (new ser.), IX, p. 222, 1859. * Auk, VII, p. 322, 1890. 'Parry's Third Voyage, Appendix, p. 97, 1826. 116 NORTH AMERICAN FAUNA. [No.22, out the region between Norway House and Forts Churchill and York." Corvus americanus And. Crow. A few seen about Lake Winnipeg and Norway House, and small numbers notid nearly every day between Norway House and York Factory. They were rather common at York Factory July 10 to IT. A few were seen about the mouth of Churchill River July 30, and another was noted 50 miles south of Cape Eskimo on the morning of August 14. On our return trip they were several times met with. Murray recorded the crow from Trout Lake and Hudson Bay.6 Xanthocephalus xanthocephalus (Bonap.). Yellow-headed Blackbird. Recorded by Murray from Hudson Bay.'' Nutting found it breed- ing abundantly at Chemawawin, Saskatchewan."' Agelaius phceniceus (Linn.). Red-winged Blackbird. Common in the Red River Valley, and abundant about the marshes below Robinson Portage, where two specimens were collected June 27. A number were seen near Oxford House July -1, in the marsh between Oxford and Back Lakes. Recorded by Murray from Hudson Bay.6 Sturnella magna neglecta (Aud.). Western Meadowlark. A number seen along Red River between Winnipeg and West Selkirk June 11. Specimens procured at Winnipeg have been exam- ined and prove referable to this form. Icterus galbula (Linn.). Baltimore Oriole. One seen flying across Red River, about midway between Winnipeg and West Selkirk, June 11. Bell records a specimen from York Factory, collected and presented by Dr. Matthews.' Scolecophagus carolinus (Midi.). Rusty Blackbird. Several seen and a female taken .June 25, near the head of the Echimamish, where they were undoubtedly breeding. One was taken from a large flock at Fort Churchill July 26. While encamped on Churchill River, about 15 miles above Fort Churchill, Alfred F. Preble found the species abundant, and took several specimens August 8. We found the species common as we ascended Hill River September 3 to 5, and saw several between Oxford and Windy lakes September 12. It was very common along the Echimamish September 11 and 15. «Rept. Prog. Can. Geol. Surv., 1878-79, A.pp. VI, p. (17c ( L880). &Edin. New Phil. Journ. (newser.), IX, p. 222, L859. elbid., |>. 222, 1859. ''Nat, Hist. Bull. Univ. Iowa, II, p. 274, 1893. - Kept. Prog. Can. Geol. Surv., lssi>-:;-4, A pp. Ill, p. 54DD I 1885). October, 190-'. 1 BIRDS OF KKKWAT1N. 117 Forster recorded it from Severn River," his record being- probably the earliest notice of the bird, which was then undescribed. Murray received specimens from Severn House and Trout Lake/' and Hell reported it from York Factory." The catalogue of birds in the U. S. National Museum collection contains the record of specimens taken at M.OOS6 Factory in LSS1 by Walton Ilaydon. Scolecophagus cyanocephalus (Wagl.). Brewer Blackbird. Common in the Red River Valley, between Winnipeg and West Sel- kirk. June 14. Nutting records this species and S. carolmus from the lower Saskatchewan, where both apparently breed.'' Quiscalus quiscula seneus (Ridgw.). Bronzed Grackle. Several seen near Sea Falls and on the lower Echimamish June 24; rather common June 26 and 27 at Robinson Portage, where two speci- mens were collected; common at Oxford House June 30 to July 4; and a single bird seen on upper Hill River July 7. On the return trip several were seen on Trout River, near Oxford House, September 9. Bell reported Q. purpurms from York Factory/ and Clarke recorded an adult male from Fort Churchill, collected many years ago/' The catalogue of birds in the U. S. National Museum collection contains the record of a specimen taken at Moose Factory in 1881 by Walton Haydon. Coccothraustes vespertinus (Coop.). Evening Grosbeak. Thompson, on the authority cf R. H. Hunter, reports that this species was seen in autumn, winter, and early spring at Big Island, Lake Winnipeg/ Pinicola enucleator leucura (Midi.). Pine Grosbeak. A male was seen perched on a tree overhanging Hill River, near the mouth of Fox River, Jul}7 8. The species was reported to us by the residents of Fort Churchill. Edwards figured a pair of these birds brought from Hudson Bay by Mr. Isham, who informed him that they wintered there.^ Forster recorded it from Severn River; h Murray recorded it from Severn House;6 Bell reported that it was frequently seen on the Churchill in the latter part of July; * Clarke recorded adults of both sexes, collected oPhil. Trans., LXII, p. 400, 1772. &Edin. New Phil. Journ. (new ser.), IX, p. 223, 1859. ^Rept. Prog. Can. Geoi/Surw, 1878-79, App. VI, p. 68c (1880). >> Nat. Hist. Bull. Univ. Iowa, II, p. 274, 1893. «Auk, VII, p. 322, 1890. /Pruc U. S. Nat. Mus., XIII, p. 584, 1890. PNat. Hist., Ill, pis. 123 and 124, 1750. A Phil. Trans., LXII, p. 402, 1772. «Rept. Prog. Can. Geol. Surv., 1878-79, App. VI, p. 70c (1880). 118 ". ORTH AMERICAN FAUNA. [No.22, at Fort Churchill many years ago;a and Fielden recorded a specimen obtained near York Factory in 1886.6 The catalogue of birds in the U. S. National Museum collection contains the record of a specimen taken at Moose Factory in 1881 by Walton Haydon. Carpodacus purpureus (Gmel.). Purple Finch. The song of the purple finch was several times heard early on the morning of June 16 at Bull Head Point, Lake Winnipeg, where the steamer stopped for wood. Turner recorded the species from Moose Factory, c and Walton Hay- don collected specimens there in 1881. Loxia curvirostra minor (Brehm). Red Crossbill. A small flock was seen at our camp on the Echimamish June 25. Forster recorded two specimens of 'Loxia curvirostra* from Severn River. d Loxia leucoptera Gmel. White-winged Crossbill. Murray received this species from Hudson Bay, Severn House, and Trout Lake;'' Baird recorded a specimen in the U. S. National Museum from Nelson River;-' and adults of both sexes from Fort Churchill were recorded by Clarke." Acanthis hornemanni (Holb.). Greenland Redpoll. Clarke recorded two adults collected many years ago at Fort Church- ill/' Murray's record of Linota horealis from Severn House may be referable to the present form or to exilipes. e Acanthis hornemanni exilipes (Coues). Hoary Redpoll. Three specimens from York Factor}7 and one from Fort Churchill, collected in July, are referable to this form. Acanthis linaria (Linn.). Redpoll. Eight specimens, including one in juvenal plumage, were collected July 12 to lb' at York Factory, where the birds were abundant, and a very bright male was taken at Fort Churchill July 23. Forster recorded this species from Severn River.'7 Acanthis linaria holboelli (Brehm). Hollxell Redpoll. A specimen (No. 81)311) taken at Moose Factory in 1881 by Walton Haydon, and now in the U. S. National Museum collection, seems to be a typical example of this form. a Auk, VII, p. 322, is', in. 6 Transactions Norwich Society, IV, p. 348, L887. cProc. U. S. Nut. Mus., VIII, p. 239, 1885. dPhil. Trans., LXII, p. 402, 1772. «Edin. New Phil. Journ. (new ser.), IX, p. 223, 1859. /Birds Of North America, p. 42S, L858. f/Phil. Tntns., LXII, p. 405, 1772. October, 1H02.] BIRDS OF KEEWATIN. 119 Spinus pinus (Wils.). Pine Siskin. Nutting records the pine siskin from Grand Rapids, Saskatchewan, where two specimens were taken.0 It probably occurs throughout southern Ceewatin, but I find no published records. Passerina nivalis (Linn.). Snowflake. Edwards figured the 'Snow-Bird from Hudson's-Bay,' b and on this figure Linnaeus partially based his description of Emberisa nivalis. Forster recorded migrants from Severn River;' Richardson says the species breeds on Melville Peninsula,d and records it as usually arriv- ing at Fort Churchill from March 26 to April 6, and being very rarely seen in midwinter; Dr. Rae saw young near Neville Bay;6 Swainson and Richardson speak of its breeding about Chesterfield Inlet ■/ Murray received specimens from Hudson Bay, Severn House, and Trout Lake;r/ Bell reported it from York Factory;7' Clarke from Fort Churchill; ' the British Museum Catalogue has a record of one collected by Dr. Rae at Repulse Bay; and the catalogue of birds in the U. S. National Museum collection records a specimen taken at Moose Factory in 1881 by Walton Hay don. Calcarius lapponicus (Linn.). Lapland Long-spur. Rather common 10 miles north of Fort Churchill, on the shores of Button Bay, where an immature bird was taken July 31. Abundant on the Barren Grounds south of Cape Eskimo August -1 to 13, at which time the old birds were moulting and were almost invariably destitute of tail feathers. They were seen in great numbers near Hubbart Point August IT, and were common at Norway House Sep- tember 19, where one was secured. At the latter point they were feed- ing in company with horned larks. Forster recorded the species from Severn River;7 Murray received specimens from Trout Lake and Severn House;9, Clarke recorded adults and young collected at Fort Churchill;' and two specimens col- lected at Repulse Bay by Dr. Rae are mentioned in the British Museum Catalogue. The catalogue of birds in the U. S. National Museum collection contains the record of a specimen taken at Moose Factory in 1881 by Walton Hay don. « Nat. Hist. Bull. Univ. Iowa, II, p. 275, 1893. 6 Nat. Hist., Ill, pi. 126, 1750. cPhil. Trans., LXII, p. 403, 1772. ''Appendix to Parry's Second Voyage, p. 344, 1825 (1827). e Narrative of an Expedition to the Shores of the Arctic Sea, p. 24, 1850. /Fauna Boreali-Americana, II, p. 246, 1831. f/E.lin. New Phil. Journ. (new ser.), IX, p. 222, 1859. ^Rept. Prog. Can. Geol. Surv., 1878-79, App. VI, p. 68c (1880). 'Auk, VII, p. 322, 1890. JPhil. Trans., LXII, p. 404, 1772 120 NORTH AMERICAN FAUNA. [No. 22, Calcarius pictus (Swains.). Smith Longspur. Rather common on the meadows at Fort Churchill July 23 to 30. They were quite tame, but were hard to see on the mossy hillocks, their coloring rendering them very inconspicuous. Often the first intimation I had of their proximity was their note, consisting of several sharp chips uttered in rapid succession. Three adult males and a young male just from the nest were taken. The young male, which was secured July 24, may he described as follows: Back dusky, the feathers edged with deep buff and whitish; feathers of head and neck dusky, mostly edged with buff, collar of adult being indicated by white edgings; wing quills strongly edged with brown, coverts tipped with white; lower parts buffy, about as in adults, chest conspicuously streaked with black, and sides marked with obscure spots of dusky; white markings of head plainly indicated. Murray recorded specimens from Severn House;" and a specimen collected by Dr. Rae at Repulse Bay is recorded in the British Museum Catalogue. Pooecetes gramineus (Gmel.). Vesper Sparrow. Nutting reports the vesper sparrow common at Grand Rapids, Sas- katchewan, where specimens were taken in the summer of 1891. -; Ammodramus sandwichensis savanna (Wils.). Savanna Sparrow. Common throughout the region wherever suitable ground occurred, especially in the vicinity of the posts. At Norway House June 22 we found young just beginning to fly, and took several specimens of these and the old birds. We collected other specimens at Oxford House June 30 to July 1, and at York Factory, where they were especially common on the marsh at Beacon Point. We collected two more at Fort Churchill and two in the juvenal plumage at my camp 50 miles south of Cape Eskimo August 4 to 8. The last were taken in traps set for voles. The catalogue of birds in the U. S. National Museum collection contains the record of a specimen taken at Moose Factory in 1881 by Walton Haydon. Ammodramus bairdi (And.). Baird Sparrow. Nutting records a specimen taken at Grand Rapids. Saskatchewan, in the summer of 1891. c Zonotrichia querula (Nutt.). Harris Sparrow. A number of specimens, including adults of both sexes and young just from the nest, were collected, July 23 to 30, at Fort Churchill, where the birds were rather common. They frequent the scattered patches of "Kdin. New Phil. Journ. (newser.), IX, |>. 223, 18.59. ''Nat. Hist. Bull. Univ. Iowa, II, p. 275, 1898 elbid., ).. 275, L893. October, 1902.] BIRDS OF KEEWATIN. 121 dwarfed spruce that grow in the small valleys and ravines intersecting the extensive expanse of precipitous ledges along Churchill River in t ho vicinity of the post. They undoubtedly nest among these spruces, hut no nests attributable to the species were found. We heard no song, hut they had a loud metallic chip which was audible and easily recognized at a distance of several rods. Young- just from the nest, taken July 24 and 25, may bo thus described: Upperparts dusky black, the feathers edged with deep burly and brown, the black predominat- ing on crown, the brown on hind neck, and the black and brown about equally divided on back; outer wing quills edged with deep buffy, inner with brown; tail feathers edged and tipped with whitish; sides of head and lower parts buffy; chest and side streaked with black, which is most conspicuous on sides of chest and forms a promi- nent malar stripe; upper throat grayish white, with fine dusky markings. Several were seen in a thicket bordering upper Hayes River August 30, and the species was rather common in afire-swept tract between Robinson Portage and the Echimamish September 14. Zonotrichia leucophrys (Forst.). White-crowned Sparrow. First seen at York Factory, where it was abundant and where a small series, including adults of both sexes and young birds not long from the nest, was taken July 12 to 11. About the post at Fort Churchill it was extremely abundant July 23 to 30, but after passing that point we saw no more of it until we had repassed the post on our return and were ascending- Hayes River August 3<>, when we observed a number in a thicket bordering- the river. This species was first described by Forster from Severn River, Hudson Bay. He also received it from Albany Fort, where it was said to breed;" and Murray received it from Severn House/' It prob- ably breeds throughout the northern wooded portions of Keewatin, being confined mainly to the Hudsonian zone. Zonotrichia albicollis (Gmel.). White-throated Sparrow. Abundant throughout the region between Norway House and York Factory. It was especially numerous in the extensive tracts which had been devastated by fire, where its simple but beautiful song- lent a charm to the gloomy surroundings. A few were noted, one of which was collected, at York Factory in July; and on our return trip a few were seen at Oxford Lake September 11. Murray reported the species from Hudson Bay/' and Clarke recorded an adult pair from Fort Church ill. '" It wTas taken at Moose Factoiy "Phil. Trans., LXII, p. 403, 426, 1772. &Edin. New Phil. Journ. (new ser.), IX, p. 223, 1859 cAuk, VII, p. 322, 1890. 122 NORTH AMERICAN FAUNA. [No.22, in 1881 by Walton Haydon. Its breeding range in Keewatin is probably nearly co-extensive with the forest, though it is rare in that part that lies in the Hudsonian zone. Spizella monticola (Gmel.). Tree Sparrow. First seen at York Factory, where the species was rather common and a young bird not long from the nest was taken July 12. It was abundant at Fort Churchill July 21 to 30, and we took a series at that point. Many were noted on the Barren Grounds, 50 miles south of Cape Eskimo, August 1 to 8. On our return trip several were seen at Duck Point, Playgreen Lake, September 19. Murray received specimens from Severn House*/' and a specimen collected by Dr. Rae at Repulse Bay is recorded in the British Museum Catalogue. The catalogue of birds in the U. S. National Museum col- lection contains the record of a specimen taken at Moose Factory in 1881 by Walton Haydon. Spizella socialis (Wils.). Chipping Sparrow. A few chipping sparrows were seen about the post at Norwa}7 House, one of which was collected. We also met with them about the post buildings at Oxford House, and saw one or two on an island in Knee Lake July 5. It was peculiarly pleasant to meet with this friendly little bird in these wild northern forests; and it was inter- esting to note that it showed no absence of that social trait to which its name is due, but was usually found nesting near dwellings Spizella pallida (Swains.). Clay-colored Sparrow. Nutting records three specimens, evidently taken on the lower Saskatchewan.6 The species probably barely reaches southwestern Keewatin. Junco hyemalis (Linn.). Slate-colored Junco. A specimen was taken June 20 at Norway House, where the species was common; another was secured July 3, one of a number seen in the spruce and tamarack woods about Oxford House, and the species was again observed while we were ascending Steel River, September 1. Specimens were received from Severn River by Forster, who, sup- posing the species to be undescribed, renamed it Wrvagilla hudsonia.6 Two specimens are recorded from Fort Churchill by Clarke. & Speci- mens were taken at Moose Factory in 1881 by Walton Haydon. Melospiza melodia (Wils.). Song Sparrow. Common at Norway House and in the shrubbery about the clearing at Oxford House, and a number observed about Knee Lake -July 5 "Ivlin. New Phil. Journ. (new ser.), IX, p. 223, L859. ''Nat. Hist. Bull. 1'i.iv., Eowa, 1 1, p. L'7r>, L893. ' Tliil. Trans., IAN, pp. 406, 428, 1772. ''Auk, VII, p. 322, L890. Octobeb.1902.] birds of keewaten. 123 and »>. Not noted between Knee Lake and York Factory on our inward trip, though we found the species rather common on Hill River when we ascended it early in September. Melospiza lincolni (And.). Lincoln Sparrow. Rather common July 13 to 16 at York Factory, where three speci- mens were collected. Melospiza georgiana (Lath.). Swamp Sparrow. A few were seen in the shrubby woods back of the post at Oxford House, and an adult was taken July 3. At York Factory, where the species was rather common, two young, not long from the nest, were taken July 13 and 16. Passerella iliaca (Merr.). Fox Sparrow. First noticed on the afternoon of July 10, when its beautiful song- was heard in the willow thickets bordering Hayes River a few miles above York Factory. While at York Factory we found fox sparrows fairly common in willow thickets, and took a pair July 16. Zamelodia ludoviciana (Linn.). Rose-breasted Grosbeak. We heard the song- of this bird while descending- Red River, a few miles below Winnipeg, June 11. King took one near the north end of Lake Winnipeg in the summer of 1835/' Progne subis (Linn.). Purple Martin. Edwards figured a bird brought from Hudson Bay by Mr. Isham, calling it the 'Great American Martin'/ Linnaeus based his descrip- tion of Hi I'u ndn subis on Edwards's figure. Petrochelidon lunifrons (Say). Cliff Swallow7. Forster recorded a specimen sent from Severn River as ' Hirundo No. 35.' '' This is probably the earliest notice of the species, which was not formally described until many years afterward. Baird records a specimen taken at Moose Factory May 27, 1860; '' and Barnston mentions the species as arriving at Martin Falls, Albany River, by May 15/ Hirundo erythrogastra Bodd. Barn Swallow. On the morning of August 13, while I was encamped at the mouth of a river on the Barren Grounds, about 25 miles south of Cape Eskimo, a barn swallow that had evidently been following the course of the stream flew past the camp. When it reached the Bay it turned south- ward and soon disappeared from sight down the coast. "Narrative of a Journey to the Shores of the Arctic Ocean, II, p. 225, 1836. ''Nat. Hist., Ill, pi. 120, 1750. 'Phil. Trans., LXII, p. 40s, 1771'. ''Review of American Birds, p. 290, May, 1805. eEdin. New Phil. Jonrn., XXX, p. 254, 1841. 124 NORTH AMERICAN FAUNA. [No. 92, Tachycineta bicolor (Vieill.). Tree Swallow. A few seen on lower Red River June 15, and two at Norway House June 19. Common between Norway House and Oxford House, and many nests observed in deserted holes of woodpeckers in trees at the water's edge, June 23 to 30. Common also at Oxford House, and noted on Steel River July 9. On our return through their haunts early in September we saw none. The catalogue of birds in the U. S. National Museum collection records specimens taken at Moose Fac- tory, and the species probably ranges throughout southern Keewatin. Riparia riparia (Linn.). Bank Swallow. A small colony was found on Red River, a few miles below Winni- peg, June 14, and a large one in a high clay bank on the shore of Oxford Lake, near Oxford House, June 30. Several colonies were seen July 10 on Hayes River, a few miles above York Factory. The catalogue of birds in the IT. S. National Museum collection records a specimen taken at Moose Factory in 1881 by Walton Haydon. Ampelis garrulus Linn. Bohemian Waxwing. Three were observed by Alfred E. Preble in the stunted spruce woods near Fort Churchill July 25. Tyrrell speaks of seeing a flock " in a grove of birch trees near the shore of Theitaga Lake, on their breeding grounds."" This lake is situated about 300 miles slightly north of west of Fort Churchill. Ampelis cedrorum (Vieill.). Cedar Waxwing. Recorded by Baird from Moose Factory, where it was collected by Drexler August 26, I860.6 Walton Haydon took specimens at the same place in 1881. Nutting found it breeding abundantly at Grand Rapids and Chemawawin, Saskatchewan/ Lanius borealis Vieill. Northern Shrike. Two specimens were taken at Fort Churchill, where the birds wore rather common July 23 to 30, and one was taken and another noted near Painted Stone Portage September 14. Forster recorded La/nivs excubitor, referring to the present species, from Severn River;'' Murray received specimens from Trout Lake and Severn House;' and Bell reported it from York Factory.' The species was collected at Moose Factory in 1881 by Walton Haydon. Vireo olivaceus (Linn.). Red-eyed Vireo. Abundant about Lake Winnipeg, at Norway House, and between Norway House, and Oxford House. In the vicinity of Oxford House " Ann. Rept. Can. Geol. Surv., 1896 (new ser.), IX, |». L65F ( L897). ''Review of American Birds, p. 408, May, L866. 'Nat. Hist, Bull. Univ. Iowa, II, p. -277, L893. ■i Phil. Trans., LX II, p. 386, 177L'. ' Edin, New Phil. Journ. (new ser.). IX, p. 223, 1859. /Kept. Prog. Can. Geol. Surv., 1878 79, App. VI, p. 68c. (1880). Octobbb,19Q2.] BIRDS OF KKKWATIN. 125 its song was heard almost continually. After leaving that point we saw nothing more of the bird. Vireo philadelphicus (Cass.). Philadelphia Vireo. A peculiar vireo song heard on Hill River July 8 was probably the song of this species, but I was unable to secure the bird. Baird recorded the species from Moose Factory, where specimens were taken June 2, L860, by 0. Drexler." Walton Haydon took specimens at the same place in 1881. Vireo solitarius (Wils.). Blue-headed Vireo. One (a male) taken at Oxford House July 3, and one heard singing in a swamp bordering Knee Lake July 5. Mniotilta varia (Linn.). Black and White Warbler. Recorded by Turner from Moose Factory, where Drexler took specimens May 13 and 31, I860.6 Nutting' reported one specimen from Grand Rapids, Saskatchewan.'' Helminthophila celata (Say). Orange-crowned Warbler. One was taken in a willow thicket at York Factory July 16, and the species was again noted near Pine Lake September 13, and at Duck Point, Playgreen Lake, September 19. Helminthophila peregrina (Wils.). Tennessee Warbler. Two females taken at Oxford House July 3. The species was fairly common at York Factory, where specimens were taken .July 13, 14, and 16. Baird recorded specimens from Fort George and Moose Factory, collected by C. Drexler in I860,'7 and the species was taken at Moose Factory in 1881 by Walton Haydon. Dendroica tigrina (Gmel.). Cape May Warbler. Recorded by Turner from Moose Factory, where one was taken May 28, 1860, by Drexler/ The collection of the U. S. National Museum contains other specimens from the same place. Dendroica aestiva (Gmel.). Yellow Warbler. Rather common at Norway House, Oxford House, and York Fac- tory. Specimens taken at these three points have a slightly darker crown than is usual in eastern examples, but are referable to the typ- ical form. Specimens taken at Fort Churchill many years ago are recorded by Clarke,'" and the catalogue of birds in the U. S. National Museum col- " Review of American Birds, p. 341, May, 1866. ';Proc. U. 8. Nat. Mua., VIII, p. 236, 1885. <-Nat. Hist. Bull. Univ. Iowa, II, p. 277, 1893. ^Review of American Birds, p. 179, April, 1865. 'Proc. U. S. Nat. Mus., VIII, p. 237, 1885. /Auk, VII, p. 322, 1890. 12(3 NORTH AMERICAN FAUNA. [No.22, lection records a specimen taken at Moose Factory in 1881 by Walton Haydon. The species probably breeds throughout the wooded por- tions of Keewatin. Dendroica coronata (Linn.). Myrtle Warbler. One was seen, in company with kinglets and chickadees, in the spruce woods bordering Hill River, September 2. According to Turner, Drexler took the species at Moose Factory/' Clark records specimens from Fort Churchill/ Dendroica maculosa (Gmel.). Magnolia Warbler. One or two seen at Norway House, and one taken at Oxford House July 3. Turner recorded the species from Moose Factory/' where specimens were also taken in 1881 by Walton Haydon, and the British Museum Catalogue records one collected by G. Barnston on Albany River. Dendroica castanea (Wils.). Bay -breasted Warbler. One was taken at Oxford House July 3. Turner recorded the species from Moose Factory, where it was taken June 2, 1860, by Drexler." Dendroica striata (Forst.). Black-poll Warbler. At Oxford House, June 30 to July 4, they were rather common and a female was collected. On July 10 we noticed a pair in a thicket that bordered Hayes River, a few miles above York Factory, and on arriv- ing at that post we again found the birds rather common. At Fort Churchill, where they were also common, we took another specimen July 24. It proved to be a young one not long from the nest and still in the speckled plumage. This species was first described by Forster from Severn River,' and was recorded by Murray from Trout Lake."1 Dendroica blackburniae (Gmel.). Blackburnian Warbler. This species is recorded by Murray from Severn House and Trout Lake under the name Syhricola par us. d Dendroica palmarum (Gmel.). Palm Warbler. One was seen on a small willow-covered island in Hill River Sep- tember 4. Clarke recorded an adult specimen from Fort Churchill.6 «Proc. U. S. Nat, Mus., VIII, p. 237, 1885. '-Auk, VII, p. 322, 1800. hil. Trans., LXII, pp. 406,428, 1772. c*Edin. New. Phil. Journ. (new ser.), IX, p. 222, 1859. Octobbb.1902.] BIRDS OF KF.KWATIN. 127 Dendroica palmarum hypochrysea Ridgw. Yellow Palm Warbler. Turner recorded a specimen taken ;it Moose Factory in July, L860, by Drexler,0 and Baird records one (under the name palma/mm) taken at Fort George, on the east coast of James Bay, in 1861. b Seiurus aurocapillus (Linn.). Oven-bird. A specimen taken at Moose Factory by Walton Haydon in the sum- mer of 1881 is recorded in the catalogue of birds in the U. S. National Museum. Seiurus noveboracensis notabilis (Ridgw.)- Grinnell Water-Thrush. We saw our first water-thrush at Painted Stone Portage. It was close to the edge of the water and was running through the under- growth which fringed the foot of a cliff. On the afternoon of the same day. June 26, we noted another at Robinson Portage, and on June 30 a third at Oxford Lake. When we arrived at Oxford House we found the species rather common, and from there to York Factory, as we descended the rivers, its sprightly song was heard daily. A pair seen at a portage on Hill River July 7 were feeding young just from tin1 nest. Three specimens were taken in the marshy woods about York Factory, where the species was common July 11 to 17, and one was taken August 8 by Alfred E. Preble on Churchill River about 15 miles above Fort Churchill. These prove to be intermediate between nove- horacensis and notabilis, but nearer to notabilis. Murray received the species from Severn House/ Geothlypis Philadelphia (Wils.). Mourning Warbler. Nutting records a specimen from the lower Saskatchewan (exact locality not stated) that is apparently intermediate in characters between tolmiei and philaddphia, but seems nearer to tolmdei.d Thompson, on the authority of Macoun, records the mourning war- bler from Waterhen River and Swan Lake/ Manitoba, which are not far to the southward of where Nutting collected. As the species has been recorded also from various points to the southeast of Keewatin, there is little question that it breeds in the southern part of the Province. Wilsonia pusilla (Wils.). Wilson Warbler. Rather common in the undergrowth bordering a swamp at Robinson Portage June 27. Several were seen at York Factory July 10 to 17, one being taken on July 14. «Proc. IT. S. Nat. Mus., VIII, p. 237, 1885. b Review of American Bird*, p. 208, April, 1865. cEdin. New Phil. Journ. (new set.), IX, p. 222, 1859. ''Nat. Hist. Bull. Univ. Iowa, II, p. 278, 1893. *Proc. U. S. Nat. Mas., XIII, p. 622, 1890. 128 NOETH AMERICAN FAUNA. [No.23, The British Museum Catalogue records a specimen from Hudson Bay collected by Captain Herd. A specimen taken at Moose Factoiy in the summer of 1881 by Walton Hay don is in the U. S. National Museum collection. Wilsonia canadensis (Linn.). Canadian Warbler. The catalogue of birds in the U. S. National Museum collection records a specimen taken at Moose Factoiy in the summer of 1881 by Walton Haydon. Nutting records the species from Grand Rapids, Saskatchewan." Setophaga ruticilla (Linn.). Redstart. Baird recorded a specimen taken at Moose Factory by Drexler,ft and Thompson quotes Hutchins's manuscript to the effect that one was shot at Fort Albany. c Anthus pensilvanicus (Lath.). Pipit. Rather common July 21 to 30 on the rocky hills at Fort Churchill, where a pair were taken. A large flock was seen on lower Hayes River as we were ascending it August 29. Murray recorded the species from Hudson Bay.rf Olbiorchilus hiemalis (Vieill.). Winter Wren.' The catalogue of birds in the U. S. National Museum collection records a specimen taken at Moose Factoiy by Walton Haydon. Cistothorus stellaris (Licht.). Short-billed Marsh Wren. A male was taken in a wet meadow at Norway House June 20. From its actions it probably had a nest in the vicinit}r, but despite a careful search none was found. Sitta carolinensis Lath. White-breasted Nuthatch. An adult female taken at Fort Churchill many years ago is recorded by Clarke/ Sitta canadensis Linn. Red-breasted Nuthatch. Heard on the Echimamish June 25, and a few seen on an island in Knee Lake on the forenoon of July 5. In his narrative Dr. Rae speaks of taking one above Broad River, between York Factoiy and Fort Churchill/ Parus atricapillus Linn. Chickadee. Baird recorded a specimen taken at Moose Factory/ A chickadee in the l'. S. National Museum that was taken at the same place by a Nat. Hist, Hull. Univ. Iowa, II, p. 279, 1893. & Review of American Birds, |>. 256, May, 1865. cProc. I'. S. Nat. Mus., XIII, p. 624, L890. 'Mvlin. New Phil. Journ. (newser.), IX, |>. 222, L859. ' Luk, VII, p. 322, 1890. J Narrative of An Expedition to the Shores of the Arctic Sea, p. 11, 1850. (/ Review of American Hinls, p. si, July, 1S64. October, 1902.' BIRDS OF KEEWATIN. 1*29 Walton Haydon is referable to the eastern form. Forster recorded Pants atricapillus from Fort Albany.0 Parus atricapillus septentrionalis (Harris). Long-tailed Chickadee. One taken on the lower Echimamish June 24. Nutting reported a specimen from Grand Rapids. Saskatchewan.6 Parus hudsonicus Forst. Hudsonian Chickadee. We first met with this species on the Echimamish June 24. We noted it again at Robinson Portage three days later, and found it com- mon at Oxford House, where we secured a male July 3. We saw several on an island in Knee Lake July 5, and a number near York Factory July 13, collecting- two on the latter date. On our return we saw several on Hill River September 3. The species was first described by Forster from Severn River.' It is recorded from Fort Churchill by Clarke/' and from Moose Factory by Rhoads/ Regulus satrapa Licht. Golden-crowned Kinglet. One was seen on the lower Echimamish J une 24. Regulus calendula (Linn.). Ruby-crowned Kinglet. I saw one at Norway House .June 17 and took a specimen on the Echimamish June 24. It was common at Oxford House June 30 to July 4. One was observed as we were ascending Hill River Sep- tember 1. Forster recorded a specimen probably sent from Severn River;" Bell recorded the species from York Factory/ and Clarke an adult from Fort Churchill.'' The catalogue of birds in the U. S. National Museum collection records a specimen taken at Moose Factory in the summer of 1881 by Walton Haydon. Hylocichla fuscescens salicicola Ridgw. Willow Thrush. The characteristic 'veery' call note and song were heard several times, and the singer was seen once as we floated down Red River between Winnipeg and West Selkirk June 14. None were .taken, but specimens from the region seem referable to the western form. Hylocichla alicise (Baird). Gra}T-cheeked Thrush. Not met with until we reached York Factory, where a female and two young just from the nest were taken in a dense willow thicket «Phil. Trans., LXII, p. 407, 1772. '-Nat. Hist. Bull. Univ. Iowa, II, p. 279, 1893. fPhil. Trans., LXII, p. 408, 430, 1772. tf Auk, VII, p. 322, 1890. < Auk, X, p. 328, 1893. /Rept. Prog. Can. Geol. Surv., 1882-3-4, App. Ill, p. 54DD (1885). 7165— No. 22 9 130 NORTH AMERICAN FAUNA. [No. 22, July 13. The young birds may be described as follows: Back and head dark olive-brown, each feather tipped with dusky and with a longitudinal spot of brown; rump and upper tail-coverts brownish spotted with rusty; lower parts white, slightly tinged on breast and sides with buffy, each feather tipped with a dusky bar, those on breast heavily marked, the marking decreasing in size posteriorly; throat almost unmarked; cheeks grayish, spotted with dusky; wings and tail olive-brown, the wing quills lighter on outer edges. 1 again met with the species July 24 at Fort Churchill, where I saw several in stunted spruce woods. Hylocichla ustulata swainsoni (Cab.). Olive-backed Thrush. The song of this species was heard at Bull Head Point, Lake Win- nipeg, on the morning of June 16. The birds were rather common at Norway House, and were seen or heard daily between there and Oxford House. They were common at Oxford House, and a specimen was taken at that point. While descending the streams between Oxford House and. York Factory we found them abundant; every wooded islet in the lakes seemed to be the home of a pair, and wherever we camped we heard their songs, which began soon after midnight. A nest found in a bush overhanging Jack River, between Knee and Swampy lakes, July 6, contained eggs on the point of hatching. At York Factory, where we took two specimens, the species was apparently less abun- dant, and beyond that point we did not meet with it. Baird recorded a specimen collected at Moose Factory in July, 1860, by Drexler/' Hylocichla guttata pallasi (Cab.). Hermit Thrush. Nutting found the hermit thrush abundant at Grand Rapids, Sas- katchewan, in the summer of 1891/' Though it doubtless occurs in southern Keewatin, I find no published records of such occurrence. Merula migratoria (Linn.). Robin. Found throughout the region, but seldom seen elsewhere than in the vicinity of the posts, where, however, they were rather common. Man}7 old and young were seen at Fort Churchill during the latter part of July. On our return trip we noted the species on Hayes River August 30, Steel River August 31, Hill River September 4, and between Oxford and Windy lakes September 12. Forster recorded it from Severn River,' and Murray from Severn House and Trout Lake.'' " Review of American Birds, p. 21, June, 1864. ''Nat, Hist. Bull. Univ. Iowa, II, p. 270, L893. 'Phil. Trans., IAII, p. .'!!)(), 1772. (/ Edin. New Phil. Journ. (new eer.), IX, p. 222, 1859. OcroBBK,1902.] BIRDS OF KEEWATIN. 131 Saxicola oenanthe leucorhoa (Gmel.). Greenland Wheatear. James Clark Ross recorded a specimen of Sylvia cenanthe obtained at Felix Harbor, Gulf of Boothia," which is probably referable to the race recently recognized by Stejneger.6 The British Museum Cata- logue records from Albany River an adult male cmanthe, collected by Barnston, which is also probably referable to the Greenland race. If, as is probable, this bird inhabits the country to the north of Hudson Ray, its most natural route of migration would seem to be along the borders of the Bay, and it is not unlikely that it is a regular breeder about the northern shores. Sialia sialis (Linn.). Bluebird. The U. S. National Museum collection contains a specimen taken at Moose Factory in the summer of 1881 by Walton Haydon. "Appendix to Ross's Second Voyage, p. xxvi, 1835. &Proe. U. S. Nat. Mus., XXIII, p. 476, 1901. BATRACHIANS OF KEEWATIN. We made a small collection of the species of frogs noted during the trip, and extended their previously recorded ranges. A gradual short- ening of the hind legs as the northern limit of the ranges of these species is approached seems to be the rule, and is evidently correlated with the shorter period of activity. In addition to this collection I include references to several species of frog-sand salamanders recorded b}T Cope from the region about James Bay, and from the mouth of Nelson River. Rana pipiens Gmel. Leopard Frog'. We noted this species at two points— in the meadows near Norway House, where it was fairly common, and at Sea Falls, about 20 miles farther north, where we saw several individuals while we were making a portage. We collected two specimens at each place. Rana palustris Le Conte. Le Conte Leopard Frog. Cope records specimens from James Bay, collected by C. Drexler.a These specimens, which are preserved in the U. S. National Museum, have been re-examined in connection with this report. Rana cantabrigensis latiremis Cope. Northern Wood Frog. We collected a series of wood frogs that includes specimens from TatVs Fishery (on Great Playgreen Lake), Norway House, York Fac- tory, and Fort Churchill. According to measurements, this series is referable to latiremis as restricted by Howe.6 Eleven well-grown specimens from Norway House average: Length of body (nose to anus) 46.3; femur (measured from central line) 19.1; tibia 19.9. Seven specimens from York Factor}' average: Length of body 43.9; femur 18.7; tibia 17.7. A single specimen taken at Fort Churchill measures: Length of body 50; femur 19; tibia 19. Cope records cantabrigensis from the mouth of Nelson River, and from James Bay.c The James Ba}T specimens have not been critically examined, and may be nearer to cantabrigensis than to latwemis. This frog was by far the most abundant species throughout the region between Lake Winnipeg and York Factory, and many were «Batrachia of North America (Bull. 34, U. S. Nat, Mus.), p. 409, 1889. &Proc. Boston Soc. Nat. Hist., vol. 28, No. 14, p. 373, Feb., 1899. ^Batrachia of North America (Bull. 34, U. S. Nat. Mus.), p. 437, 1889. 133 134 NORTH AMERICAN FAUNA. [No. 22, October, 1902.] taken in our traps as well as by hand. At Fort Churchill, though frogs were reported to be frequently heard in the marshes, we were able to obtain but one specimen. This is characterized b}^ extremeh7 short legs, and differs further from any others collected in being heavily vermiculated on the sides with black. Most of the specimens collected have a grayish median dorsal stripe. The color from life of a York Factory specimen is as follows: Back grayish green, sparingly spotted with black, and with a greenish-white median dorsal stripe; beneath greenish white, darkest on sides. Rana septentrionalis Baird. Specimens recorded b}^ Cope under the name Rana cantabrigensis evittata from Moose River, Ontario, are referable to Rana septen- trionalis, as stated by Howe.a Chorophilus septentrionalis Boulenger. Northern Chorophilus. Chorophilus septentrionalis Boulenger, Cat. Batrachia Salientia Brit. Mus., p. 335, 1882. We found this species throughout the region between Lake Win- nipeg and York Factory, and took specimens at the following localities: Taft's Fishery (Great Playgreen Lake), 1; Norway House, 3; Oxford House, 1; York Factory, 3. These specimens appear to be referable to septentrionalis, the type locality of which is Great Bear Lake, as the tibio-tarsal joint does not reach the tympanum when the leg is stretched forward. The fol- lowing brief description of the color was taken from a live specimen from Oxford House: Body light green above, greenish white beneath; bod}7 stripes bronzy lavender; tympanum brownish; hind legs light green above, flesh color beneath. Ambly stoma jeffersonianum platineum Cope. Cope records a specimen collected by C. Drexler on Moose River.6 Chondrotus microstomia Cope. Cope records a specimen from Hudson Bay, probably from James Bay, collected by F. W. Haydene (probably mistake for W. Haydon). Plethodon cinereus Green. Cope records four specimens from 'Hudson's Bay Territory. ' prob- ably from the region about James Bay, collected by C. Drexler.'' «Proc. Boston Soc. Nat. Hist., vol. 28, No. 14, p. 374, Feb., 1899. 6 Batrachia of North America (Bull. 34, U. S. Nat. Mus.), ]»■ 94, 1889. elbid., p. 103, 1889. ''Ibid., p. 135, 1889. INDEX [Names of new species in bold-face type. Page. Abies balsamea 13 Aran this hornemanni lis horaemanni exilipes 118 ltnaria 118 linaria holbcelli 118 Aceipiter atricapillus 100 i m iperi 106 pennsyl vanieus 100 velox 105-ioc Acer negundi i 12, 22 Acorns calamus 14 Actitis macularia 99-100 iEgialitis meloda circumcincta 101-102 Bemipalmata 101 vocifera 101 Agelaius phoenicens lit; Aix sponsa 84-85 Alces americanus 43 Alio alio 78 Amblystorua jeffersonianum platdneum 134 Ammodramus bairdi 120 sand wiehensis savanna 120 Ampelis cedrorum 23, 124 garrulus 124 Anas In ischas 82-83 labradoria 80 obscura rubripes 83 Ansei albifrons gambeli 89 fabalis 90 segetum 90 Anthus pensilvanicus 23,128 Antrostomus vociferua 113 Aquila ehrysaetos 107 Archibuteo lagopus sancti-johannis 100-107 Arotomys monax empetra 47-48 Parryii 40 Arctostaphylos uva-ursi 50 Ardea herodias 92 Ardetta exilis 91 Arenaria interpres 102 morinella 102 Arvicola hudsonius 58 Asio accipitrinua 109 wilsonianus 109 Ay thy a affinis 85 amerieana 85 marila 85 vallianeria 85 BalaL-na niystieetus 39 Baldpate 83 Bartramia longieauda 99 Baaswood 11 Bat. little brown 73 Batrachians 13:5-134 Page. Hear, Barren Ground 64 Mack 04 polar 21,64-65 Beaver, Canadian 48-49 Betnla glanduloaa 17 nana 20 papyrifera 12,13,10 Bibliography 27-38 Birch 12 canoe 13, 10, 23 dwarf 17, 20 Birds 75-131 Bittern 91 least 91 Blackbird, Brewer 117 red-winged 110 rusty 1 10-117 yellow-headed lid Blarina brevicauda 22 Bluebird 131 Bog bean 17 Bonasa umbellus togata 103 Botaurus lentiginosus Branta bernicla canadenaia canadensis hutchinsi . leucopsis 91 90 9 J 90 91 Bubo virginianus arcticus 110 10 21 16 80 13 100 100 Buckthorn Buffalo-berry ( 'anadian Buffle-head Bulrush Buteo borealis Iineatus platypterus 100 swainsc »ni 100 Calcarius lapponicus 23, 119 pictus 23,120 Calidris arenaria 97 Calla palustris 15 Camptolaimus labradorius 86-87 Canaehitea canadenaia 22, 102-103 Canis albus 01-02 occidentals 02 Canvas-back 85 Caribou. Barren Ground 21,23,41-43 woodland 40-41 Carpodacua purpureus 23, 1 18 Castor canadensis 48-49 Ceophloeua pileatus abieticola 112 Cepphus mandti 77 Ceryle alcyon Ill Cbaradrius dominicus 101 135 136 INDEX. Page. Charadrius melodus 101 Charitonetta albeola 86 Chaulelasmus streperus 83 Chen crerulescens 89 hyperborea nivalis 89 rossi 89 Chickadee 128-129 Hudsonian 129 long-tailed 129 Chipmunk, Lake Superior 45-46 northeastern 45 Chondrotus microstomas 134 Chordeiles virginianus 22, 113 Chorophilus, northern 134 Chorophilus septentrionalis 134 Circus hudsonius 105 Cistothorus stellaris 128 Clangula clangula americana 85 islandica 86 Coccothraustes vespertinus 117 Colaptes auratus luteua 112 Colymbus auritus 75 holbcelli 75 Condylura cristata 23, 73 Contopus borealis 114 Coot 93 Cormorant, double-crested 81-82 Corvus americanus 116 corax principalis 115-116 Corylus americana 12 rostrata 12 Crane, little brown 92 whooping 92 Cratagus coccinea 12 Crossbill, red 118 white-winged 118 Crow 116 Crowberry, black 17, 20 Crymophilus fulicarius 93-94 Curlew, Eskimo 100 Hudsonian 100 Cyanocitta cristata 115 Cystophora cristata 70 Dafila acuta 84 Delphinapterus catodon 40 Dendroica sestiva 125-126 blackburnise 126 castanea 126 coronata 126 maculosa 23, 126 palmarum 126 palmarum hypochrysea 127 striata 126 tigrina 125 Dicrostony x 19, 20, 21 riehardsoni 55-58 Dove, mourning 105 Dovekie 78 Dowitcher, 1 our billed 95 Dryobates pubescens medianus ill villosus leucomelas n l Duck, harlequin 86 l ,abrador 86-87 lesser scaup 85 red-legged black 83 ruddy 88 Page. Duck, scaup 85 wood 84-85 Dunlin 96 Eagle 14 bald 107 golden 107 Ectopistes migratorius 106 Eider, American 87 king 87-68 northern 87 Pacific 87 Elsagnus argentea 16 Elm 11 Elymus mollis 20, 52 Emberiza nivalis 119 Empetrum nigrum 17, 20 Empidonax minimus 114 trailli alnorum 114 Equisetum 16 Erethizon dorsatum 59 Ereunetes pusillus 97 Erignathus barbatus 70-71 Erismatura jamaicensis 88 Eutamias neglectus 45-16 Evotomys gapperi 50-51 Falco columbarius 108 hudsonius 105 islandus 107 peregrinus anatum 107-108 riehardsoni : 108 rusticolus gyrfalco 107 sparverius 108 Fiber zibethicus hudsonius 38, 53-54 Finch, purple 118 Fir, balsam 13 Fisher 69 Flicker, northern 112 Flycatcher, alder 114 least 114 olive-sided 114 scissor-tailed 113 Fox, Continental Arctic 62-63 red 62 Fringilla hudsonia 122 Frog, leopard 133 northern wood 133-134 Fulica americana 93 Fulmar 81 Fulmarus glacial is 81 Gadwall 83 Gallinago delicata 94 major 95 (ia via. adamsi 76 arctica 76 iinbcr 75 76 I mnine '. 77 pacifica 76-77 Geothlypis Philadelphia 127 God wit, Hudsonian 97-98 marbled 97 red-breasted 97 Golden-eye Barrow Goose, barnacle bean blue INDEX. 137 Page. Qoose, Canada 90 greater snow 89 Hutchins 90 Ross snow 89 white-fronted 89 Grackle, bronzed 117 Grebe, Holbcell 75 horned 75 pied-billed 75 Grosbeak, evening 117 rose-breasted 123 pine 1 1 7-1 IS Grouse, Canadian ruffed 103 Hudson ian spruce 102-103 sharp-tailed 104 Grus axnericanus 92 canadensis 92 Guillemot, Mandt 77 Gull, Bonaparte 80 Franklin 80 glaucous 79 great black-backed 79 herring 79 Iceland 79 ivory 79 ring-billed 79-80 Ross NO Sabine 80 Gulo luacus 09 Gyrfalcon 107 .white 107 Ealiseetus leucocephalus 107 Hare, Arctic 21 Hudson Bay varying 59 Keewatin Arctic 59-61 Harelda hyemalis 86 Hawk, broad-winged 106 Cooper 106 duck 107-108 marsh 105 pigeon 108 red-tailed 106 rough-legged 106 sharp-shinned 105-106 sparrow 108 Swainson 106 Hawthorn 12 Hazel 12 Helminthophila celata 125 peregrina 125 Helodromas ochropus 99 solitarius 99 Heron, great blue 92 Heteractites incanus 99 Heterocelus brevipes 99 Hirundo erythrogastra 123 Histrionieus histrionicus 86 Honeysuckle 10 Hydroehelidon nigra surinamensis 81 Hylocichla aliciae 129-130 fuscescens salicicola 129 guttata pallasi 130 ustulata swainsoni 22,130 Icterus galbula 22, 116 Jaeger, long-tailed 7s-79 parasitic 20, 78 Page. Jaeger, pomarine 20,78 Jay, blue 115 Canada 115 J unco hyemalis 122 Junco, slate-colored 122 Killdeer 101 Kingbird 113 Kinglet, golden-crowned 129 ruby-crowned 129 Kittiwake 79 Knot 95 Lagopus lagopus 23,103-104 rupestris 104 Lanius 1 » irealis 121 Larix laricina 13 Lark, horned 19, 20, 1 1 1 Hoyt horned 114-1 15 Larus argentatus 79 delawarensis 79-80 franklini 80 glaucus 79 keeask 7s leucopterus 79 marinus , 79 Philadelphia 80 roseus 80 rossii 80 Ledum 17 palustre 20 Lemming 19, 20, 21 Back 54-55 pied 20 Richardson ." 23, 55-58 Lemmus 20 trimucronatus 54-55 Lepargyrrea canadensis 16, 21 Lepus americanus 59 arcticus canus 38, 59-61 Life zones 22-23 Lily, pond 13 Lime isa fedoa 97 h»mastica 23, 97-98 Longspur, Lapland 20,61,119 Smith 19, 120 Lonicera glaucescens 16 Loon 75-76 black-throated 76 Pacific 20, 76-77 red-throated 20, 77 yellow-billed 76 Lophodytes cucullatus 82 Loxia curvirostra minor 118 leucoptera 118 Lutra canadensis 65 Lutreola vison lacustrls 3s, 66 Lynx, Canada 61 Lynx canadensis 22, 61 Macrorhampus scolopaceus 95 Magpie 115 Mallard 82-83 Mammals 39-73 Maple, ash-leaved 12 Mareea americana 83 Marten, Hudson Bay 68 Martin, purple 123 Meadowlark, western 116 138 INDEX. Page. Megalestris skua 78 Megascops asio 110 Melospiza georgiana 123 lincolni 123 melodia 122 Menyanthes trifoliata 17 Mephitis mephitis 65 Merganser 82 hooded 82 red-breasted 82 Merganser amerieanus 82 serrator 82 Merlin, Richardson 108 Merula migratoria 130 Micropalama himantopus 95 Mierotus 20 aphorodemus 38, 52-53 drummondi 51-52 xanthognathus 53 Milfoil, water 15 Milvulus forflcatus 113 Mink, Keewatin 66 Mniotilta varia 125 Mole, star-nosed 73 Monodon monoeeros 39-40 Mc ii isc, eastern 43 Mosquito 13,18 Mouse, clouded white- footed 49-50 house 49 Hudson Bay jumping 58 meadow 20 red-backed 50-51 Murre 77 Briinnich 77 Mus empetra 47 musculus 49 Muscivora forflcata 113 Musk-OX 23, 43-44 Muskrat, Hudson Bay 53-54 Hustela americana abieticola 38, 68 pennanti 22, 69 Myotis lucifugus 73 Myrica gale 16 Myriophyllum 15 Narwhal 39^0 Neosorex palustris 71-72 Neotoma drummondi 50 Nettion carolincnsis 83-84 New species 38 Night-hawk 113 Numenius borealis 100 hudsonicus 23, 100 Nuthatch, red-breasted 128 white-breasted 128 Nyctala acadica 110 tengmalmi richardsoni no Nyetea nyetea no-ill Nymphsea 13 Oak, mossy cup 11 Odobenus rosmarug 69-70 Oidemia americana 8s deglandi 88 perspicillata 88 Olbiorchilus hiemalis 128 Old-Squaw *80 (tli ii- buccinator oi Page. Olor columbianus 91 Oriole, Baltimore 116 Osprey 108 Otocoris alpestris 114 alpestris hoy ti 114-115 Otter, Canada 65 Oven-bird 127 Ovibos moschatus 13—14 Owl. Arctic horned 110 barred 109 great gray 109-110 hawk Ill long-eared 109 Richardson 110 saw-whet 110 screech no snowy 110-111 Pagophila alba 79 Pandion haliaetus carolinensis 108 Parus atricapillus 128-129 atricapillus septentrionalis 129 hudsonicus 22, 129 Passerella iliaca 123 Passerina nivalis 119 Pedicecetes phasianellus 104 Pelecanus erythrorhynchos 82 Pelican, white 82 Perisoreus canadensis 22, 115 Peromyscus canadensis umbrinus 49-50 Petrochelidon lunifrons 123 Phalacrocorax dilophus 81-82 Phalarope, northern 94 red 93-94 Wilson 94 Phalaropus lobatus 23, 94 Phenacomys, Mackenzie 50 Phenacomys mackenzii 38, 50 Philohela minor 94 Phoca grcenlandica ' 70 hispida 70 vitulina 71 Phcebe 113-114 Pica pica hudsonia 115 Picea alba 13 nigra 13 Picoides amerieanus 112 arcticus 22, 112 tridactylus 112 Pigeon, passenger 105 Pine, Banksian 18, 16,23 Pinicola enucleator Ieucura 117-118 Pintail M Pinus divaricata 13,16 I'ipit 128 Plethodon cinereus 134 Plover, belled piping loi 102 black-bellied loo golden 1 01 semipalmated 17,101 Podilymbus podiceps 75 Pond-weed 15, 17 Pocecetes gramineus 120 Poplar 12 aspen 18,16,23 balsam 13, 23 Populus balsamifera 12, 13 INDEX. V6\) Page. Popnlus trcmuloidcs 13, 16 Porcupine, Canada 59 rorzana Carolina 92-93 no vein iracensia 98 Potamogeton 15, 17 I 'ot en til la tridentata II Progne subis 128 Ptarmigan, rock 104 willow 20, 103-104 Putoriua arcticus 07 cioognanj 67 cicognani richardsoni 67 rixosus 67-68 Rail, yellow 18,93 Virginia 92 Rallus virginianus 92 Rana cantabrigensis latiremis 133 palustris 133 pipiens 133 septentrionalis 134 Rangifer arcticus 41-13 caribou 40-41 Raven, northern 115-116 Redhead 85 Redpoll 20, 118 Greenland 118 hoary ll.s Holboell 118 Redshank 98 Redstart 128 Regulus calendula 129 satrapa 22, 129 Rhamnus alnifolia 16 Khodostethia rosea 80 Riparia riparia 124 Rissa tridaetyla 79 Robin 130 Querelas macrocarpa 11, 22 Querquedula discors 84 Quiscalus quiscula seneus 117 Salix anglarum 20 phylicifolia 20 Sanderling 97 Sandpiper, Baird 96 Bartramian 99 buff-breasted 99 green 99 least 96 pectoral 96 purple 95-96 red-backed 96-97 semipalmated 97 solitary 99 spotted 99-100 stilt 95 white-rumped 96 .Sapsucker, yellow-bellied 112 Saxicola eenanthe leueorhoa 131 Sayornis phoebe 113-114 Scirpns lacustris 13 Sciuropterus sabrinus 44-45 Sciurus hudsonicus 45 Scolecophagus carolinus 116-117 cyanocephalus 117 Scoter, American 88 surf 88 Page. scoter, white-winged 88 Scotiaptex nebulosum 109 Scouring-rush 16 Seal, bearded 70-71 harbor 71 harp 70 hooded 70 rough : 70 Si iu ii is aurocapillus 127 aoveboracensis notabilis 127 Setophaga ruticilla ]2s Shoveller 84 Shrew, Coues 73 eastern 72 Keewatin 72-73 marsh 71-72 Richardson 73 Shrike, northern 124 Sialia sialis 23, 131 Silver-berry 16 Siskin, pine 119 Sitta canadensis 128 carolinensis 128 Skua 78 Skunk, Canada 65 Snipe, greater 95 Wilson 94 Snow flake il9 Si (materia dresseri 87 mollissima borealis 87 spectabilis 87-88 v-nigra 87 Sora 92-93 Sorex (Microsorex) aluorum 38,72-73 personatus 72 richardsoni 73 sphagnicola 73 Sparrow, Baird 120 chipping 122 clay-colored 122 fox 123 Harris 120-121 Lincoln 123 savanna 20, 120 si mg 122 swamp 123 tree 20, 122 white-crowned 121 white-throated 121-122 Spatula clypeata 84 Spermophile, Hudson Bay 46-47 Spermophilus parryi 46-47 Sphyrapicus varius 112 Spinus pinus 119 Spizella monticola 122 pallida 122 socialis 122 Spruce, black 13 white 13 Squatarola squatarola 100 Squirrel, Hudson Bay flying 44-45 Hudson Bay red 45 Steganopus tricolor 94 Stercorarius longicaudus 78-79 parasiticus 78 pomarinus 78 140 INDEX. Page. Stercorarius richardsoni 78 Sterna caspia 81 hirnndo 81 paradissea 81 Strix einerea ¥ 109 nebulosa 109 Sturnella magna neglecta 116 Surnia ulula caparoch Ill Swallow, bank 124 barn 123 cliff 123 tree 124 Swan, trumpeter 91 whistling 91 Sweet-flag 14 Symphoricarpos occidentals 12 Synaptomys ( Mictomys) bullatus 38, 54 Syrnium nebulosum 109 varium 109 Tachycineta bicolor 124 Tamarack 13, 21, 23 Tamias quadrivittatus 46 striatus lysteri 45 Tattler, wandering 99 Tea, Labrador 17,20 Teal, blue-winged 84 green-winged 83-84 Tern, Arctic 17, 19, 81 black 81 Caspian 81 common 81 Tetrao canadensis 102 phasianellus 104 Thalarctos maritimus 64-65 Thrush, gray-cheeked 129-130 hermit 130 olive-backed 130 willow 129 Tilia americana 11-12 Totanus flavipes 98 , melanoleucus 98 totanus 98 Tringa alpina 96 alpina paeifica 96-97 bairdi 96 canutus 95 fuscieollis 96 maculata 96 maritima 95-96 minutilla 96 Tryngites subruflcollis 99 Turnstone, ruddy 102 Tyrannus tyrannus 113 Ulmus americana 11,22 Una lorn via 77 troile 77 rrsiis americanua 64 richardsoni 64 Viburnum 12 small-flowered 16 pauclflorum 16 Vireo, blue-headed 125 Page. Vireo olivaceus 124-125 philadelphicus 125 solitarius 125 Vireo Philadelphia 125 red-eyed 124-125 Vole, Barren Ground 52-53 chestnut-cheeked 53 Drummond 51-52 northern lemming 54 Vulpes fulvus 62 lagopus innuitus 38, 62-63 Walrus, Atlantic 69-70 Warbler, bay-breasted 126 black and white 125 blackburnian 126 black-poll 126 Canadian 128 Cape May 125 magnolia 126 myrtle 126 orange-crowned 125 palm 126 Tennessee 125 Wilson 127-128 yellow 125-126 yellow palm 127 Water-thrush, Grinnell 127 Wax wing, Bohemian 124 cedar 124 Weasel, Bangs 67-68 Bonaparte 67 Richardson 67 tundra 67 Whale, Greenland 39 white 40 Wheatear, Greenland 131 Whip-poor-will 113 Wilsonia canadensis 23, 128 pusilla 127 Wolf, Barren Ground 61-62 gray 62 Wolf berry 12 Wolverine, Hudson Bay 69 Woodchuck, Canadian 47-48 W< todcock 94 Woodpecker, Arctic three-toed 112 downy Ill northern pileated 112 striped-backed three-toed 112 Wood-rat, Drummond 50 Wren, short-billed marsh 128 winter 128 Xanthoeephalus xanthocephalus L16 X i 'ina sabinii 80 Yellow-legs 98 greater 98 Zamelodia ludoviciana 128 ZapUS Hudson ins 58 Zenaidura macroura 22, 105 Zonotrichia albicollis 22, 12] 122 leucophxys 22, 121 querula.. 120-121 O B. P. L. Bindery, JUN 3 1904