mmmm^ LIBRARY OF THE UNIVERSITY OF ILLINOIS AT URBANA-CHAMPAIGN 590.5 FI v.59-> u u as 0) *j - I 3J o g >< 3:1 ^ w 1-H i-H a *- TO c 05 a 00 * a. -N := « 05 O t- """a OS bo 1^ oj O PQ J U 2 la So; e is 2 CO ■<-> lO 5 2 -To Oi § o e I-, ** Is u o O CO a 19 20 FIELDIANA: ZOOLOGY, VOLUME 60 three juveniles, two males and a female, was found here under a small pile of weeds surrounded by a thick stand of grass. These juveniles had lighter colored pelage than adults taken nearby. For each of 14 shrews we caught two house mice. Crocidura russula Hermann (Common European) White-toothed Shrew Sorex russulus Hermann, 1780, in Zimmerman, Geogr. Gesch., 2, p. 382. Type locality. — Eastern France, Bas-Rhin near Strasbourg. Distribution. — General: This species is widely distributed in the temperate and subtropical zone from Spain and Algeria to Japan and Korea. Afghanistan Records: Niethammer (1965, p. 22). 1939 Chaworth-Musters Expedition (Paghman, Shibar Pass) . 1965 Street Expedition: FMNH 102067-122, 102124-133; in alcohol 102082, -085, -088, -090, -118, -119, -121, -133; skull only 102095. Third Danish Expedition, 1948-1949 (Kabul). Afghanistan Habitat: This shrew was collected between 1,000 and 3,000 m. in montane Afghanistan; the largest number being taken in the watercourse and structure biotopes. Taxonomic Notes and Discussion. — The mean and extreme dimensions of 39 skulls are: greatest lenght of skull, 18.2-20.0 (19.1); condylo-premaxillary length, 16.6-18.5 (17.6); cranial width, 8.3-9.3 (8.7); width between infraorbital for- amina, 3.2-3.8 (3.5); and greatest antorbital width, 5.5-6.4 (6.0). Comparing these figures with similar measurements (c.f. page 23) for C. zarudnyi shows russula skulls to be larger, especially in the two characters herein italicized. As noted by Ellerman and Morrison- Scott (1951, p. 72), russula does not have the tail as much as 70 per cent of the head and body, which is, however, a characteristic of C. zarudnyi. The general color of the dorsal pelage ranges from Hair Brown to Chaetura Drab gradually changing to Drab-Gray ventrally. Comparison of C. russula with C. zarudnyi streetorum from the same locality shows the C. russula to be perceptibly darker; their backs appearing dark grayish-brown and not, as described for C. Z. streetorum, grizzled brownish-gray. 21 22 FIELDIANA: ZOOLOGY, VOLUME 60 Field Notes and Discussion. — Sixty-one of the 66 Crocidura russula trapped were taken from rocky terrain. All 66 came from places with free water accessible in the form of a nearby stream or irrigation ditch, but the kind of vege- tation had no apparent affect on their local or geographic distribu- tion. They were trapped in evergreen oak and coniferous forests in Nuristan, along alfalfa fields and in willow thickets in northeast Afghanistan, under mulberry trees in the Paghman oasis, and along irrigation ditches lined with camel thorn near Ghazni. They were consistently captured in sites with vegetation having a low homo- geneity. Traps set in this typical habitat near Jalalabad and Kandahar, however, failed to yield any shrews. Sixty-four shrews were trapped in museum special snap traps baited with small amounts of our usual (stock) bait, a mixture of pea- nut butter, oatmeal, and raisins. A single shrew was captured in a trap baited with the liver of a migratory hamster. This was an unsatisfactory bait because in dry Afghanistan meat desiccates rapidly, and no shrews were captured with dry meat. Crocidura zarudnyi Ognev East Persian White-toothed Shrew Crocidura zarudnyi Ognev, 1928, Mamm. E. Europe, N. Asia, I, p. 341, New name for tatianae Ognev, 1921, nee DoUman, 1915. Type locality. — East Persia near the Baluchistan border. Distribution. — General : C. zarudnyi has been collected from the Sistan Basin in Iran (Lay, 1967, p. 129, as pergrisea), Baluchistan (Siddiqi, 1961) and in Lyallpur, W. Pakistan (Taber, 1967, p. 369), and now in Afghanistan. Afghanistan Records: Hassinger (1970, pp. 5-8). 1965 Street Expedition: FMNH 102123, -34, -35, -36. Afghanistan Habitat: Ranging from 200 to 3,000 m. in the rock, and structure /clay-and-loess biotopes. This species was, without exception, found in dryer situations in Afghanistan, Iran (Lay, 1967, p. 129), and in W. Pakistan (Taber, 1967, p. 395) than was C. russula or C. suaveolens. Taxonomic Notes and Discussion. — Ellerman (1951, pp. 72, 83) includes C. zarudnyi as a subspecies of C. pergrisea saying: "we have not seen pergrisea, but from descrip- tions it is very like the Baluchistan form, zarudnyi, which it ante- HASSINGER: AFGHANISTAN MAMMALS 23 dates." At this writing I have before me both paratypical specimens of C. pergrisea, the three specimens of the C. zarudnyi placed in synonomy by Ellerman, a specimen from Lyallpur, W. Pakistan, six C. zarudnyi (pergrisea) collected by the earlier Street Expedition (Lay, 1967) from the Sistan Basin, and our four from montane habitat in Afghanistan. A number of measurable differences are observed between the samples of Crocidura zarudnyi and C pergrisea as follows: In total length (whether obtained as tip of snout to tip of tail excluding hair, or as head-and-body length plus tail length) means and extremes (in millimeters) of 14 zarudnyi are 103.9 (96-116) compared to the 125, 128, and 129 of three pergrisea. In greatest length of skull 11 zarudnyi are 18.1 (16.9-19.0) com- pared to 20.0 and 20.1 of two pergrisea. In greatest condylo-premaxillary length of skull 12 zarudnyi are (16.5-18.2) compared to 19.0, 19.1, and 19.3 of three pergrisea. In greatest cranial breadth 12 zarudnyi are 7.9 (7.5-8.3) com- pared to 8.7, 8.8, and 8.9 for three pergrisea. The length of the longest mystacial in 12 zarudnyi is 22 mm.; whereas that in two pergrisea is 27. The longest dorsal hairs in 12 zarudnyi are 6 mm., but in two pergrisea are 8 mm. The base of the dorsal hair of C. pergrisea (collected in October) from near Shigar, Baltistan, Kashmir, 2,900 m., is Dark Mouse Gray followed by a 1 mm. band of white and tips varying from Cinnamon Drab to Hair Brown, giving their pelage an extremely pilose, grizzled, pale grayish-hrown appearance. The base of the dorsal hair of C. zarudnyi (collected in November) from Iran, mon- tane Baluchistan and Lyallpur, 100-2,400 m., is Deep Mouse Gray (lighter than Dark Mouse Gray) followed by a narrow 0.5 mm. band of white and tips of Cinnamon Drab, giving their pelage a slightly grizzled, dull cinnamon-brown (Cinnamon Drab) appear- ance. Four C. zarudnyi (collected in July, August, and October) from mountains in Afghanistan, 2,250-2,650 m., have dorsal hairs similar to those of pergrisea but shorter and without Cinnamon Drab tips, giving their dorsal pelage a much less pilose, less grizzled, brownish-gray appearance. Thus, the very small sample of Crocidura pergrisea seems to be of a substantially larger form than zarudnyi, and one that differs markedly from C. zarudnyi in October pelage. These differences 24 FIELDIANA: ZOOLOGY, VOLUME 60 seem to me to be great enough to justify recording them here as separate species. I have not seen the habitat of C. pergrisea, but the site of its collection, in glacier-scarred, rocky mountains of Baltistan, has a mean elevation of approximately 3,000 m. and must be quite unlike the low (200 m.) saline semi-deserts of the Sistan Basin wherein we obtained these possible topotypes of Crocidura zarudnyi. The mon- tane specimens of C. zarudnyi from Afghanistan were each collected from topography surely more similar to that of Baltistan for being between 2,300 and 2,700 m. elevation: one was collected at the entrance to the Wakhan Corridor, about 400 km, northwest of the type locality (Skoro Loomba, Shigar, Baltistan) of C. pergrisea. However, this more "topographically related" form of C. zarudnyi is almost as different from C pergrisea as its low desert counterpart. So far I have been concerned to distinguish the several samples of C. zarudnyi from a single sample of C. pergrisea and have collectively referred to the four specimens from Afghanistan as a montane sample of C. zarudnyi. The measurements of these four specimens overlap those of typical C. zarudnyi from Iran and southern West Pakistan in every character, however, as described above, their average and extreme total length is greater, and their dorsal color is brownish-gray and not, as shown for other zarudnyi, dull cinnamon-brown. Judge- ment concerning size differences between lowland and montane zarudnyi cannot be conclusive until additional specimens become available, but the color difference is so pronounced and so consistent for the four localities (Ishkamish, Badakhshan; Shibar Pass, Central natural area; and Ghazni and Gardez of the Eastern natural area), that subspecific recognition must be considered. The montane sample of zarudnyi was only collected in the rock biotope, but lowland specimens reported from the Sistan Basin (Lay, 1967, p. 129) were taken in the structure and clay-loess biotopes, and from Lyallpur, W. Pakistan (Taber, 1967, p. 395) on a ". . . xeric tropical thorn plain." The Afghan sample ranges, as illustrated in Figure 4, from 33°37'N to 36°43'N, and of the other samples the northernmost is (Lyallpur) 31°25'N. Besides biotope and latitude the Afghan samples are separated from the more southern samples by altitude, the lowest Afghan one being 2,250 m. and the highest and lowest others being Kelat and Turbat (Ellerman and Morrison- Scott, 1951, p. 83) approximately 2,000 and 100 m. In summary, I find the size, color, and distribution of the four Afghan specimens to be unlike that of any other shrew population HASSINGER: AFGHANISTAN MAMMALS 25 presently known from the Iranian Plateau or West Pakistan. This difference is the basis for description of a new subspecies: Crocidura zarudnyi streetorum new subspecies. Type.— FUNK No. 102123, adult female from Afghanistan: 30 km. northwest of Ghazni, approximately 33°43'N 68°15'E; collected October 8, 1965 by Jerry Hassinger, from dry, rocky habitat de- scribed in Hassinger (1968, fig. 8). Description. — The type and three paratypes (FMNH 102134- 102136) consist of skins and skulls. The body measurements for the type are HB, 58; T, 45; HF, 12; E-8. Its skull measures: greatest length, 18.3; condylo-premaxillary length, 17.5; cranial width, 8,0. The dorsal pelage is slightly grizzled, brownish-gray gradually chang- ing low on the sides to white or Pale Smoke Gray ventral pelage con- sisting of hairs with Dark Mouse Gray bases tipped with white. Tail is Pale Smoke Gray, the proximal third being almost unicolor, the distal two-thirds having a dorsal stripe of hairs resembling the dorsal pelage. The feet are Light Buff to white. Diagnosis. — Differs from the typical form in the following: 1) grayer color; 2) greater total length. The general hue of the dorsal pelage for typical zarudnyi and C. z. streetorum is, respectively: dull cinnamon-brown, and brownish-gray. The average total length for 12 C. z. zarudnyi and four streetorum is respectively: 102 mm. and 109 mm. Material examined. — Crocidura pergrisea, 2 — USNM 175917, 175919; Kashmir: Skoro Loomba; Shigar, Baltistan. Crocidura zarudnyi zarudnyi, 10— BM 19.11.7.15, 19.11.7.16, 19.11.83, 64.1195; West Pakistan : Panjgur, Turbat, Kelat, and Lyallpur, respectively. FMNH 96403 and 96409-13; Iran: Kerman (Seistan): 24 km. south- west of Zabol. Crocidura z. streetorum, 4 — FMNH 102123 and 102134-36; Afghanistan: see Figure 4 for localities. The author considers it a privilege to name this subspecies in honor of William S. and Janice K. Street. More specimens of the above forms would permit a more precise evaluation of their relationships. Series of specimens from the Pamirs, Kashmir, Tibet, and the Himalayas would be material for a significant contribution to the taxonomy of these species. Field Notes and Discussion. — Niethammer (1965, p. 22) identified parts of shrew skulls, found in pellets of Asio otus from Kang in Sistan, as C. russula. Dr. 26 FIELDIANA: ZOOLOGY, VOLUME 60 Niethammer kindly sent me his most nearly complete skull from this locality. It is virtually identical to skulls of zarudnyi collected near Zabol, Iran (about 50 km. west of Kang); its greatest length and cranial width being 18.2 and 7.9, respectively. (See C. russula, p. 20, for characters distinguishing these species.) C. zarudnyi streetorum is widely distributed, however, we never succeeded in catching more than a single specimen in any specific locality. This shrew was found in the habitat of the mouse-like hamster, Calomyscus bailwardi, and taken in traps set specifically for C. bailwardi, excepting for the one specimen from the Wakhan, which was caught in talus about 50 m. from a heavily vegetated gully having a large population of Alticola roylei. Suncus murinus Linnaeus House Shrew Sorex murinus Linnaeus, 1766, Syst. Nat., 12th ed., I, p. 74. Type locality. — Java. Distribution. — General: Southern Asia from southeastern China to Egypt. The range of this species has been modified by commensalism with man. It has at least a sporadic distribution throughout the Oriental Region reaching north of it to Japan, and it is also found in some southwest Asian Palearctic countries and some northeastern countries of the Ethiopian Region. Afghanistan Records: Horsfield (1851, p. 134), Niethammer (1965, p. 22), Gaisler et. al. (1967, p. 358). 1965 Street Expedition: FMNH 103137-42. Afghanistan Habitat: The house shrew was collected in the Jalalabad Vale and Chagha Sarai. It was taken between 700 and 900 m. in the structure and watercourse /rock biotopes. Taxonomic Notes and Discussion. — Suncus murinus griffithi was collected by William Griffith and described as Sorex griffithi, from Afghanistan, by Horsfield (1851, p. 134). Blyth (1856) states: "We suspect that S. griffithii Horsfield, of that naturalist's Catalogue of the specimens of Mammalia in the Honeurable Company's Museum is no other than our presumed Fig. 4. The records and estimated limits of distribution for Suncus murinus (S) and Crocidura zarudnyi (Z). Records for C. zarudnyi streetorum are circled and a zone of possible intergradation (wavy lines) between this subspecies (z.s.) and other zarudnyi is hypothesized. 27 28 FIELDIANA: ZOOLOGY, VOLUME 60 Murinus from the Arakan and Khasya Hills, although described from Afghanistan, because we saw a fine skin from Cherra Punji in the possession of the late Mr. Griffith, which was forwarded to the India House by Mr. M'Clelland; and we have previously had occasion to remark; that specimens of reptiles procured by Mr. Griffith in Afghanistan and in the Khasya Hills, had manifestly become mixed and confounded; whence certain important mistakes concerning habitats." Cherra Punji is in Assam, about 42 air kilo- meters N.N.W. of Sylket. Ellerman and Morrison-Scott (1951, p. 7) commenting on the type for Suncus murinus griiffithi, states the label of the type has "Afghanistan," but this has been crossed out and "Silket" substituted. They refer the reader to Lindsay (1929, p. 333). Lindsay, in reference to the above comments by Blyth remarks: "... Blyth's explanation seems so reasonable that this shrew had been inadvertently entered as from Afghanistan, rather than Assam, and since he himself saw a specimen in the possession of Dr. Griffith, which undoubtedly came from Cherrapunji, Assam, it is safe to assume that Blyth is right and so grifithi denotes the Hill species of Assam." I have examined the type of Suncus murinus grifithi (B. M. 79.11.- 21.471) and 20 specimens from the Arakan and Khasya Hills, five of which are from Cherrapunji, the proposed type locality for grifithi. My observations support those of Blyth (1856) and Lindsay (1929), in that the specimens from the Hills of Assam resemble the type specimen of grifithi, while our specimens from Afghanistan have lighter colored (Drab to Hair Brown) dorsal pelage and are smaller than grifithi. The greatest skull length for the type and two adults from Cherrapunji is over 34.0 mm. while the longest of six skulls from Afghanistan is 32.2 mm. If Cherrapunji is indeed the type locality for grifithi, we are left with the question : Is Suncus murinus grifithi a junior synonym of Suncus murinus soccatusi Ellerman and Morrison-Scott (1951, p. 66) following Lindsay (1929, p. 333) includes Sorex heterodon Blyth (1855, J. Asiat. Soc. Bengal, 24, p. 31) as a synonym of Suncus murinus soccatus Hodgson (1845, Ann. Mag. Nat. Hist., 15, p. 270). The type locality for Sorex heterodon is Cherrapunji, in the Khasi Hills, Assam. Thus, if heterodon and grifithi are so similar as well as geographically close, and heterodon is a synonym of soccatus, then for the hill of subspecies of Assam the name grifithi is also a junior synonym of soccatus from the Central Region of Nepal. This in- HASSINGER: AFGHANISTAN MAMMALS 29 quiry reaches a little beyond the projected scope of this paper, but a cursory comparison of 10 soccatus from the vicinity of Darjeeling with the aforementioned specimens from Cherrapunji did not reveal any characters which would consistently separate these two named forms. Field Notes and Discussion. — In the Jalalabad bazaar three house shrews were procured by pro- viding the owners of various stores with traps. We trapped a single specimen from one of a series of burrows in the weedy garden of a hotel. The other burrows there yielded three Nesokia indica. Two shrews were collected west of Jalalabad: one from Laghman and the other in a small seepage covered with rocks and dense vegetation, draining a steep slope about 11 km. west of Jalalabad. This is the first collection of this synanthrope from other than the structure biotope in Afghanistan. These shrews were trapped using museum specials baited with peanut butter, oatmeal, and raisins. Order PRIMATES Macaca mulatta Zimmerman Rhesus Macaque Cercopithecus mulatta Zimmerman, 1780, Geogr. Gesch. Mensch., 2, p. 195. Type locality. — "India." Distribution. — General: This sub-tropical, short-tailed macaque has its centers of distribution between 25 and 30° north latitude rang- ing from eastern China through northern Vietnam, Burma, and India, and along the southern flanks of the Himalayas to Monsoonal Afghanistan. The most north-western record of this macaque is Nuristan in Afghanistan. Afghanistan Records: Babur in the sixteenth century A. D. (1921, p. 213); Kullman (1965, pp. 15, 16). 1965 Street Expedition: FMNH 102839. Afghanistan Habitat: This species has a disjunct distribu- tion in Afghanistan. Kullman (1965) states: "Monkeys live in large numbers in the forests of Nuristan and Paktia." The Street Expedition obtained its single specimen by purchase alive in Kandahar, but it was said to have been captured north of Chagha Sarai. 30 HASSINGER: AFGHANISTAN MAMMALS 81 Discussion. — The specimen, a young female, had been captured as an infant and maintained in captivity for 15 months. I could find no measurements of Afghan macaques in the litera- ture, therefore the following skin and skull measurements are in- cluded: HB, 355; T, 182; HF, 123; E, 42; width of braincase— 61.1; zygomatic width — 63.7; greatest length — 84.6; distance from a line connecting the anterior alveolar edge of the upper incisors to the most posterior margin of the nasals — 28.1; skull length from the posterior margin of the nasals to the most posterior bulge of the braincase — 80.0. By its dentition (all deciduous still, except for the permanent first molars, of course) Dr. Jack Fooden, Research Asso- ciate, Field Museum, places its age at two years and advises that adulthood is attained at about five years. A specimen which became the type of a subspecies, Macaca mulatta mcmahoni Pocock (1932), was collected at Kootai in lower Chitral, W. Pakistan, between the Bashgul Valley in Kafiristan (=Nursitan) and the Chitral Valley at 3,600 ft. I can find no reference to an actual collection of this species from Afghanistan, although different authors state monkeys can be found, or were seen in Monsoonal Afghanistan. Kullman (1965, pp. 15, 16) gives no specific locality for any observation of a wild one. I was privileged to visit the Zoological Garden in Kabul, where Dr. Kull- man had three rhesus macaques in captivity, not unlike the specimen we purchased. I saw other live monkeys, all captives performing for monkey trainers, three in Kabul, one in Jalalabad, and three in Kandahar. When questioned, the trainers without exception said the monkeys came from either Nuristan or Paktia. Kullman (1965, p. 16) mentions that a long-tailed monkey was described to him in Mangall, Paktia. He speculates that it might be the langur, Presbytia entellus, which is distributed over India and Kashmir. Pocock (1939, p. 51) also speculates on a description (by Raverty, 1859, p. 332) of a large monkey found in the densely wooded district of warmer Kafiristan, concluding it might possibly be the langur. Order LAGOMORPHA Ochotona macrotis Gunther Large-eared Pika Lagomys macrotis Gunther, 1875, Ann, Mag. Nat. Hist., 16, p. 231 (Sep- tember). 32 FIELDIANA: ZOOLOGY, VOLUME 60 Type locality. — Doba, Kuenlum Mountains, extreme southern Chinese Turkestan. Distribution. — General: The Tianshan and Pamir mountains in Russia, and the mountains of southern Sinkiang, China, south to northern Nepal, Kashmir, and Nuristan. Afghanistan Records: Popov (1962, pp. 107-109). Third Danish Expedition, 1948-1949 (Stiewe, Nuristan), Niethammer (1966, in litt., Wakhan Region). According to Povolny and Daniel (1966, pp. 371, 373) the valley of Chap Darrah (3,700 m.) in the Wakhan Region: "... was frequently visited by pikas (Ochotona rufescens Gray) leaving there numerous traces of their activities." Subse- quently, and without explanation, they introduce a second species (0. roylei) from the Wakhan Region (the valley of Ishmurkh Darrah, 4,000 m.). This listing is immediately fol- lowed by a reference to: ". . . 11 pikas, which were trapped in the valley of Chap Darrah ..." These locality records would (for both roylei and rufescens) represent range extensions into the known range of 0. macrotis. It is possible that the Wakhan Region is a meeting ground for roylei from the southeast and macrotis from the north and northeast but collection of 11 rufescens ranging in from the south and west seems unlikely (see Ochotona distribution map). Since those authors present no evidence to support their identifications (and because they were less concerned with pikas for themselves as potential hosts of parasitic Diptera), their collections alleged to be roylei and rufescens are herein regarded as probably macrotis, but not included on the distribution map for Ochotona. Afghanistan Habitat: Montane biotopes between 2,000 and 4,000 m. According to Scully (1881, p. 207), this species frequents open stony ground near the snowline. Discussion. — On the average the three 0. macrotis from Stiewe (collected April 20, 1948) appear to be smaller than adult 0. rufescens, but with larger ears and different pelage. Ellerman and Morrison-Scott (1951, p. 448) distinguish macrotis as having larger ears (rarely less than 27 mm.) than rufescens (rarely reaching 27 mm. in length). The overall tone of the dorsum of the feet of macrotis is Pale Cinnamon Pink, Head and body Hind foot Ear 172,'167,'160 160-209 180-229(200) 30,32,30 31-36 34-39(37) 24,28,25 21-28 22-28(25) HASSINGER: AFGHANISTAN MAMMALS 33 Table 4. — Body measurements of Ochotona from Afghanistan, (Measurements in mm.) Species Record Sample O. macrotis 3rd Danish Exp. 3 O. rufescens Niethammer (1965)* 10 O.rufescens 1965 Street Exp.^ 16 ' Measurements recorded in the field; the remaining figures for this record are of dried skins. * Extremes only. » All adults. and the length and breadth of the ears of three dried specimens aver- ages 26 by 26 mm., in contrast with Pinkish to Cinnamon-Buff feet and 18 by 18 mm. ears of ten dried, adult specimens of 0. rufescens. A male and two females collected (June 20, 1948) in the valley between Stiewe and Weran Pass (3,350 m.) represent the south- western-most record of this species. One of these females had five embryos; the other was lactating. Paludan (Third Danish Expedi- tion field notes) lists sightings of pikas in the Weran Valley (3,720 m.) and between Tilli (2,675 m.) and Nau. He notes: "They stopped at about 3,900 m., perhaps 4,000 m. They are as a rule found where the boulders are adjacent to small, grassy areas." Popov (1962) reports this species from the mountains of Badakh- shan. Niethammer's collection of macrotis in the Wakhan is not unexpected, as they are found immediately to the north (see Bobrin- ski et. al., 1965) and south. Ochotona rufescens has not been collected east of 69°30'E longi- tude; in contrast, 0. macrotis has not been found west of Nuristan or Badakhshan. The ranges of these species probably meet in the vicinity of west Nuristan. If their ranges overlap, macrotis will most likely be found at higher elevations. Ochotona rufescens rufescens Gray Afghan Pika Lagomys rufescens Gray, 1842, Ann. Mag, Nat. Hist., 10, p, 266. Type locality. — Near Babers Tomb, Kabul, Afghanistan, Distribution. — General: From the Kopet-Dag Mountains in Iran through south-western Russian Turkestan to Afghanistan and Baluchis- tan. 34 HASSINGER: AFGHANISTAN MAMMALS 35 Afghanistan Records: Gray (1842, p. 266), Hutton (1846, p. 140), Griffith (1847, p. xx), Scully (1887, p. 75), Zimmerman (1955, p. 190), Akhtar (1944, p. 82; 1957, pp. 455, 456), Niethammer (1965, p. 23), Gaisler et al. (1968, p. 186). 1965 Street Expedition: FMNH 102840-102883; in alcohol 102856, -59, -60, -61; skull only 102876,-77, -80; nestlings 102860-64. 1939 Chaworth-Musters Expedition (Shibar Pass; B. M. 47.400-47.466, 62.775, 62.776), Third Danish Expedition. 1948-1949 Central Asia (Bamian). Afghanistan Habitat: The rock and structure biotopes be- tween 1,000 and 3,200 m. is the favorite haunt of this pika. Free water or at least vegetation with a high density seems to facilitate its occurrence. Pikas are numerous along the roads east and west of Shibar Pass, and north and south of Sauzak Pass. I counted 60 during the late afternoon of August 25 in less than 12 km. along a road stretching west from the summit of Shibar Pass. Taxonomic Notes and Discussioii .-- The 30 specimens (FMNH 102840—869) collected near Pagh- man are virtual topotypes of Ochotona rufescens rufescens Gray, be- ing collected less than 40 km. from the type locality. Six (FMNH 102842, 102849, 102851, 102857, 102858, 102866) are considered to be adults, A specimen is here termed adult if the greatest length of the skull minus the least distance across the top of the skull between the orbits is more than 40.0 mm. Subadults have a shorter skull length and a wider interorbital width than adults. Data from 55 Ochotona rujescens skulls indicate that at no time during the life of this pika is the interorbital width greater than during the first month after birth. As the skull elongates and widens the frontals constrict. For example, a nestling pika with its eyes closed had a skull length of 24.2 mm. and an interorbital width of 5.0 mm., while an adult with a skull length of 45.8 mm. had an interorbital width of 3.0 mm. This definition of adulthood represents a morphological maturity, made for taxonomic purposes, and does not take sexual maturity into consideration; one growing female considered subadult by the above criterion had six embryos. The dorsal pelage of four adults captured near Paghman in July appears darker and the individual hairs are shorter, 10-15 mm., when 36 FIELDIANA: ZOOLOGY, VOLUME 60 compared with the longer hair, 20-25 mm., and the black-tipped, Cinnamon to Sayal Brown dorsal-pelage of four adults collected in late September from Sauzak Pass. Field Notes and Discussion. — Hunting proved to be the most effective method of collecting pikas. They were not attracted to our stock bait. Traps placed at fresh burrows failed to yield pikas. Twenty-eight specimens were shot. More than 75 per cent of them were living in the crevices of rocks supporting terraces or mountain-side irrigation ditches. At Paghman we purchased two adults, seven subadults and five nestlings between July 12 and 18; of these specimens one subadult female was lactating, a second contained six embryos. Small boys had no trouble catching pikas. Pikas living near wheat fields store the stalks of this grain in rocky crevices; those without this source of food store grasses and weeds in relation to availability. Pikas subsist on these food-stores during the winter; they are not known to hibernate. Niethammer (1965) mentions that excavated burrows had one or two exits and ended after about 2 m. at a depth of half a meter. In winter near Paghman he also observed that they had dug tunnels up through the snow and were sunning themselves at noon. Lepus capensis Linnaeus Cape Hare Lepus capensis Linnaeus, 1758, Syst. Nat., 10th ed., I, p. 58. Type locality. — Cape of Good Hope. Subspecies recorded for Afghanistan : Lepus tibetanus (= Lepus capensis tibetanus) Waterhouse, 1841, Proc. Zool. Soc, London, 7. Lepus hhmanni (=Lepus capensis lehmani) Severtzov, 1873, Mem. Soc. Amis. Sci., Moscou, 8, 2, pp. 62, 83. Distribution. — General: Widely distributed over Africa and Eurasia from the Cape of Good Hope north to Spain, east through Arabia and Iran to Kashmir; and through parts of the dry belt of Central Asia from Russian Turkestan east through China to approxi- mately 120°E longitude. This species is therefore widely distributed in both the Palearctic and Ethiopian Faunal Regions and marginally in the Oriental Region. Afghanistan Records: Irwin (1839, p. 1,007), Hutton (1845, p. 141), Griffith (1847, XX, pp. 366, 370), Scully (1887, p. 76), c B 'S at bfi 3 O be 3 O 37 38 FIELDIANA: ZOOLOGY, VOLUME 60 Aitchison (1889, p. 61), Niethammer (1965, p. 22), Gaisler et al. (1968, p. 187). 1965 Street Expedition: FMNH 102884 to 102895. Third Danish Expedition, 1948-1949 (Baqrabad and Faizabad in South Afgh. and Sauzak Kotal). Afghanistan Habitat: Found between 500 and 3,000 m. throughout Afghanistan excluding alpine habitat and the forests of Monsoonal Afghanistan. We collected or saw hares in the clay and loess, slope and plateau, and rock biotopes. Hares were characteristically found in terrain with vegetation having a medium or high density, low shade, one stratum, medium to high homogeneity, and a clumped distribution, Taxonomic Notes and Discussion. — An examination of the obliquely truncated apex of the large maxillary incisor tooth of specimens collected near Jalalabad revealed a bell-shaped or slightly bifurcated pattern of enamel infolding. This comparatively complicated groove is filled naturally with soft cement. Skulls of other Afghan hares had a simple, shallow groove with no cement filling. The skull length, from the most posterior point on the occiput to the anterior-most surface of the upper incisors ranges and averages: for three specimens from Jalalabad, 89.0-92.2 (91.1); for seven specimens from Wakhan, 85.2-90.5 (88.5) ; and one from Kandahar 82,4, The greatest distance between the outsides of the zygomata opposite the molar tooth row ranges from 40,1-42.5 for the Jalalabad hares, but only from 35,5-38.0 (36.8) for seven Wakhan skulls, and 34.3 for the Kandahar specimen. Table 5. — Body measurements of L. capensis from Afghanistan. (Measurements in mm.) Record Sample Head and body Tail Hind foot Ear^ Scully (1887) 1 — — — 109 1965 Street 3 470-540(502) 75-105(87) 115-128(120) 122-130(127) Exp., Jalala- bad^ 1965 Street 9 Exp., Wakhan and Kanda- har^ 408-487(448) 55-97(84) 117-133(128) 98-120(107) ' From the notch of orifice to the tip of the ear in this genus. 2 Specimens FMNH 102892, -93 and -94. ■' Specimens FMNH 102884 to 102891 and 102895. HASSINGER: AFGHANISTAN MAMMALS 39 If we follow Fetter's (1961, p. 30) classification of the European and Asiatic hares of the sub-genus Lepus, which is based solely on the structure of the groove on the anterior surface of the first upper incisor, hares with a complex gi'oove are assigned to the Indian species L. nigricollis, those with a simple groove to six other species of which L. capensis has the largest range. Lay (1967, pp. 153, 154) reports valid objections to uncritical use of this character, notably that Fetter bases his revision on too small a sample which fails to account for age or geographic variation within the key character, and that Hall (1951, pp. 181-182) presents evidence that within a single species of this holarctic subgenus Lepus (Lepus) the groove may be simple or complex dependent upon whether the specimen came from east or west of a particular range of mountains. I have examined the incisive grooves of 89 Eurasian hares in Field Museum of Natural History, dividing them into: 1) simple grooves with no cement filling or single cement-filled invaginations similar to the one illustrated by Fetter (1961, p. 34) for L. nigricollis dayanus; and 2) complex grooves, usually bi- or trifurcated, more complex than the one illustrated by Fetter for L. nigricollis dayanus. Simple gi'ooves were found in each of the 32 skulls from the Falearctic Region (Iran, Iraq, Manchuria, Mongolia). Of 48 hares collected near or in the Himalayan portion of the transition between the Pale- arctic and Oriental Region (south China, north India, Kashmir, and Afghanistan), 18 have simple grooves, 19 have a single invagination, and 11 have complex grooves. One hare from the Oriental Region (Indochinese Feninsula) had a single invagination, eight had complex invaginations. Fetter (1961) classified Lepus oiostolus and L. ruficaudatus as having a simple gi'oove. In seven of ten L. oiostolus from Szechwan Province in western China, I do find a single invagination, but the other three have complex grooves. Complex grooves were also found in each of the sample of four L. ruficaudatus from north India (Assam), supporting Tate (1947, p. 202) who distinguished this species as having complex grooves. Fetter acknowledges Tate's con- tradictory observations with a footnote but, from an examination of the type specimen, proceeds to include L. ruficaudatus with hares having simple gi'ooves. It is probably noteworthy that the ranges of both L. oiostolus and L. ruficaudatus extend into the transition be- tween faunal i-egions, and that Fetter considers ruficaudatus to be related to Lepus capensis. 40 FIELDIANA: ZOOLOGY, VOLUME 60 The foregoing evidence indicating that the probability of finding all complex or all simple grooves in any single taxon (species or sub- species) of hare (near the Himalayas) may be greater in either faunal region than in the transition zone between these regions, suggests that at least in the area considered the complexity of the groove on the first maxillary incisor is a geographic character having subspecific rather than specific significance. Ellerman and Morrison-Scott (1951, pp. 421, 422) tentatively retain L. nigricollis as a species albeit: "... they have all the essential characters of [the europaeus] group including very large size of skull." Ellerman speculates that nigricollis might represent an eastward extension of europaeus {= capensis) : "However, the Southern Indian nigricollis with its Ceylon representative is remarkable for its black- streaked neck, and the remaining forms, which would be races of ruficaudatus if further specific division were required, have the upper part of the tail normally brown and white rather than black and white." Our Jalalabad series have incisive grooves and size and shape of skull like those of Field Museum series of L. nigricollis ruficaudatus from Sikkim and Assam, but their pelage resembles that of capensis in having Light Buff instead of Ocraceous Buff neck (as in the Sikkim one) and upper part of the tail blackish and white instead of brown and white. But the Jalalabad series is intermediate between the Afghan L. capensis and the L. nigricollis ruficaudatus in having darker, warmer dorsal body pelage than the former. Finally, the dorsal pelage of the feet and legs of the Wakhan series is almost white (Pale Pinkish Buff) whereas that of the Jalalabad series is Cream Buff or Chamois which seems much closer to the Ochraceous Tawny of the Sikkim one. Thus, based on the evidence of intergradation provided in this paper and by Lay and on the contradictions associated respectively with Ellerman and Morrison Scott's (1951) and Fetter's revisions of the subgenus Lepus, all hares examined from Afghanistan are con- sidered to be Lepus capensis.^ Further it is suggested that ruficauda- tus is an Indian subspecies of L. capensis. Field Notes and Discussion. — Specimens were usually collected at night using a spotlight and shotgun, although one hare was shot in the early morning. They 1 A young hare from Bisut, near Jalalabad, is reported as (sic) Lepus"! dayanus, by Gaisler et al. (1968, p. 187; on the authority of Angermann, in litt.). HASSINGER: AFGHANISTAN MAMMALS 41 seemed most numerous in the Wakhan (sparsely populated by man) usually seen on small, gently sloping, alluvial plateaus between the steep mountains and the Amu Darya. We observed these hares to take refuge among rocks and boulders during the day. Hares are hunted by the Afghans. Niethammer (1965) mentions: "L. capensis has become so rare because of too much hunting, that it is still not represented in the collection of the Institute of Zoology." Order RODENTIA Spermophilopsis leptodactylus Lichtenstein Long-clawed Ground Squirrel Arctomj/s /cptodacij/Zws Lichtenstein, 1823, Evermann. Reise, 119. Type locality. — Karata, 140 versts northwest of Bokhara, Rus- sian Turkestan. Distribution. — General: Russian Turkestan, from the east side of the Caspian Sea northeast to steppes south and east of Lake Balkash ; and south to Afghanistan and northern Iran. Afghanistan Records: Scully (1887, pp. 70-71), Niethammer, (1965, pp. 23-24) in alcohol 102906, -07; skull only 102915. 1965 Street Expedition: FMNH 102896-915. Third Danish Expedition, 1948-1949 (Maimana and near Shibarghan). Afghanistan Habitat: Collected below 1,000 m. in the clay- and loess biotope in North and Northwest Afghanistan. The prevalent vegetation of its usual habitat was characterized by: medium density, low shade, one stratum, a clumped distribu- tion, and high homogeniety. Discussion. — Two species of this genus have been described from near the Afghan-Russian border, and subsequently have been reduced to sub- species. Scully (1887, p. 70) described S. I. bactrianus from Kamiab in North Afghanistan. Satunin (1908, p. 255) described a larger foiTTi, schumakovi, from Kushka Turkmeniya S.S.R. (see fig. 8). The adult Afghan specimens studied by me all fit the description of S. I. bactrianus, none having the length of the head and body as long as 290 mm., the figure given by Satunin (1908, p. 257) for the smallest of six adult S. I. schumakovi. 42 HASSINGER: AFGHANISTAN MAMMALS 43 This steppe-dwelling rodent was quite active near 10:00 A.M. and 4:00 P.M. in August and September. Roadside hunting, using shot- guns, during periods of its peak activity was our most successful method of collecting these squirrels. They were frequently caught foraging more than 50 m. from the nearest burrow. Marmota caudata Jacquemont Long-tailed Marmot Ardomys caudatus Jacquemont, 1844, Voy. dans L'Inde, 4, Zool. 66. Type locality. — Kashmir. Distribution. — General: From the Hindu Kush in Afghanistan through southeast Russian Turkestan and Kashmir eastward to Sinkiang, China. Afghanistan Records: Griffith (1847, pp. 388. 483), Anderson (1875, p. 283), Furse (1963, p. 22), Niethammer (1965, p. 24), Povolny and Daniel (1966, p. 371). 1965 Street Expedition: FMNH 102916, a patch of dorsal fur. Third Danish Expedition, 1948-1949 (Pashki, Puistagoli). In addition to obtaining specimens as cited above, Paludan (Third Danish Expedition field notes) observed marmots at Pashki, 2,700-3,600 m.; Stiewe; near Sanglich, 3,500 m. ; Weran Pass. 3,460 m.; and Shibar Pass, 2,440 m., between May 18 and July 25 in 1948 and 1949. He notes the absence of marmots on a subsequent return to the Shibar Pass in October, albeit: "there were many dens and we had encountered them before." Afghanistan Habitat: Above 2,500 m. in sub-alpine biotopes. Niethammer (1965, pp. 25, 26) reports: ". . . it occurs in the Hindu Kush above 2,800 m. and probably also in the Pamirs, where it is numerous on the Russian side. The most southei'n record is in the Dasht-i-Newar, where its extensive burrows were found, especially on the borders of wet plains which have a rich vegetation in summer." Paludan (supra cit) saw red- dish-brown marmots southwest of Zebak (near Sanglich, 3,500 m.) on an "artemisia steppe." He also notes seeing and hearing marmots near Pashki in: "juniper-habitat above the forest." Povolny and Daniel (1966) collected marmots on an "Alpine artemisia-steppe" in the valley of Chap Darrah, 3,700 m. 44 FIELDIANA: ZOOLOGY, VOLUME 60 Discussion. — Povolny and Daniel (1966) collected marmots on Aug. 15 and 18, 1965, in the Wakhan Region. On Aug. 17 and 18, 1965, the Street Expedition was also collecting in the Wakhan Region, approximately 800 m. lower than Povolny and Daniel, but encountered no marmots. Subsequently, our expedition never entered good marmot habitat until close to or after the onset of hibernation, and then only briefly. Consequently, our only record consists of a patch of fur said to have come from a marmot collected on the Anjumin Pass. Spermophilus fulvus Lichtenstein Fulvous Ground Squirrel Arctomys fulvus Lichtenstein, 1823, Evermann Reise, 119. Type locality. — ^U.S.S.R., Kirghizia, River Kuwandzaliur, east of the Mugadsharz Mountains, north of the Sea of Aral. Distribution. — General: South Russia to Chinese Turkestan with the south- ern limits of its distribution in north Iran and Afghanistan. Afghanistan Records: Niethammer (1965, p. 24). Third Danish Expedition, 1948-1949 (Ab-i-Istada). Afghanistan Habitat: Niethammer (1965, pp. 24, 25) states: "It is sporadically distributed in more sandy steppes between 1,000 and 3,000 m. altitude. . . ." Discussion. — At the time of our visit to localities where this species has been collected or is thought to occur, the animals had already entered aestivation or hibernation. Kashkarov and Lein (1927, p. 65) give the time of activity for this species as February to June. Their obser- vations were made in Russian Turkestan, north of Afghanistan. Because few burrows large enough to house this squirrel were seen, this species is probably rare or absent beyond areas from which they were collected in South and East Afghanistan. Although too few collecting excursions have been made to Spermophilus habitat — during their short burst of activity — to ascertain their range in Afghanistan, they appear to have a sporadic distribution with, per- haps, two disjunct populations. Lay (1967, p. 157) indicates a dis- junct distribution for this species in Iran. This sporadic and pos- sibly insular distribution along the southern limit of this species range probably reflects more favorable ecological conditions and a wider distribution in the past. I 00 o^ o 00 s s 'a JS rt 00 ^ ;c «o f-( eo 00 00 ^ e« . r2 E £ 'S ,c WJ 2 *o c g £ 2 £ 3 •§ c -^ o o Tf t~ t- C^ (M (M O Oi ^ Tj< -H s ^ n ea < a £ -H Tf w •PSg ^ 00 „ C/3 in lo 2* -^ c- .« — ' 00 W ^^ e .2 C S e C la OS S c5 .2 c c o 5 £ £ CO es e« es -o "^ -- ^3 .S .S e«3 Z Z Oj CO £ u £ ^ z :? "o i "i o i2 3 ^ 5 c "o /! c £ 2J ^ £ m •-< * CO CO C£> U3 Oi 00 o:) t- CO 1— I 1— I »-l T-( O t-H °° be ^ § "" 00 O ,^ hZ ca i* iX w t— " -n* "3 in lO H I I 00 o o T3 c W £ ^ 1-1 (M O -* (M (M o o 00 CO CO w I I I o oa r-i (M (M !M Tt CO CD o ^ i^ o o o IM IM O ^ rH 1— I T— 1 1-H i-H 1-H ,^^ Ovl n lO lO g^ i-H T-H CO S '^-' 2 ^ n' Pi .2 g ^ s CO o CJ .^ 00 IM '^ 03 O 0) jS := -- H W J?: e in CD ^ I' 5 -^ ,.4 w h:^:^!' oj CD -2 .2 ^ a 00 J3 >» 00 _m OS ^ '=« > W 3 T3 Xfl 00 a a «- c »o ^ CD 73 OS u 1-H CO ;2 CO Q < CO C rt^ OS OJ MH ^ £ M e O -C >'3l O " o u a; =« t- o *" ^.S £ o CO •-• 58 HASSINGER: AFGHANISTAN MAMMALS 59 mm. deep, with a series of lateral passages and exits. This subter- ranean passage ran parallel to the edge of a fallow wheat field. A female trapped on July 25 contained four embryos. Twenty subadults were trapped in July, but none thereafter, although adults were trapped in August, Septembei*, and October. Microtus arvalis Pallas Common Vole Mus arvalis Pallas, 1779, Nov. Spec. Quad. Glir. Ord., p. 78. Type locality. — Germany. The following species previously recorded from Afghanistan is considered to be a subspecies of Microtus arvalis. Microtus transcaspicus Satunin, 1905, Verz. Saug. Transkaspiens (Russ), 25, p. 30. Tschuli Gorge, near Ashabad, Transcaspia. Distribution. — General: Temperate zone of Europe and Asia reaching its southern limit of distribution in Afghanistan. Afghanistan Records: Ellerman (1948, p. 789), Niethammer (1965, p. 36; 1970, p. 11). 1965 Street Expedition, FMNH 103126-150; in alcohol 103140, -46. Afghanistan Habitat: Found on mountains in the water- course and clay-and-loess biotopes. The Street Expedition collected 25 specimens in approximately 25 sq. m. of a seepage on Shibar Pass shown in Hassinger (1968, fig. 13). They were not taken beyond the lush grass of this biotope, but Microtus afghanus were collected in the drier perimeter of this seepage, less than 8 m. beyond the center of M. arvalis activity as indicated by runways and trapping success. Niethammer (1965) states: "In contrast to Central Europe they [arvalis] occur in Afghanistan only sporadically in relatively humid localities and rarely on cultivated land." He gives their ver- tical distribution as 2,500-3,000 m. Discussion. — Ellerman (1948, p. 789) provisionally identified three large voles collected by Chaworth-Musters on the Shibar Pass as Microtus transcaspicus. Ognev (1950, p. 184) considers transcaspicus to be a subspecies of M. arvalis. Niethammer (1965, p. 36) concuning, considers the specimens collected by Chaworth-Musters to be: "typical field voles, differing little from other large subspecies of 60 FIELDIANA: ZOOLOGY, VOLUME 60 M. arvalis." A superficial comparison of teeth, bullae, and skull dimensions of M. arvalis from Iran with specimens from Afghanistan supports Niethammer's judgement, i.e., adult voles from Afghanistan are similar to adult M. arvalis from Iran (Lay, 1967), but, as shown in Table 8, larger. The overall appearance of the dorsal pelage of adults collected in July is Buffy to Olive Brown. Subadults have darker pelage with more black hairs than adults, which, in turn, have darker pelage than neighboring Microtus afghanus. Six subadults, seemingly from the same litter, were phenotypi- cally alike but different from the other 19 voles. These six appear to have a black moustache in that the pelage around the base of the mystacial vibrisaae and above the nose is black contrasting with adjacent Buffy Brown pelage. We collected six adults and 19 subadults in July. The grass of the swale inhabited by these voles was etched with well-worn trails. Voles were collected by placing traps in these runways. Setting our traps at dusk produced 15 specimens within an hour. Burrows and runways in drier soil adjacent to the swale yielded three Microtus afghanus in the same period of time. Microtus socialis Pallas Social Vole Mtis socialis Pallas, 1773, Reise Russ. Reich., 2, p. 705. Type locality. — U.S.S.R. : Grassy regions of desert by Ural River. Arvicola guentheri Danford and Alston, 1880, Proc. Zool. Soc, London, p. 62. Type locality. — Turkey: Marash. Distribution. — General: From the Ukraine south to Lybia and east to Rus- sian Turkestan and Afghanistan. Afghanistan records: Scully (1887, p. 72). Third Danish Expedition, 1948-1949 (Gilzai— in the Koh-i- Baba Mountains). AFGHANISTAN HABITAT: Montane and Steppe biotopes. Ognev (1950, p. 322) remarks: "The animal [M. socialis] does not avoid low and sometimes moist valleys . . ., dry watershed steppes, semi-deserts and deserts, river valleys, and finally high plateau steppes ..." 3 'B .9 O S E e I u O T3 hi O O E-" 61 62 FIELDIANA: ZOOLOGY, VOLUME 60 Discussion. — A single specimen collected by C. E. Yates of the Afghan Boun- dary Commission in Afghan Turkestan (= North Afghanistan) was identified as Arvicola guentheri by Scully (1887, pp. 72-73). A second vole was taken in August, 1948, in the Koh-i-Baba Moun- tains northwest of Kabul. This specimen represents a range exten- sion of approximately 200 km. southeast of the nearest previously recorded locality for this species. Ognev (1950, p. 342) refers specifically to the specimen collected by Yates and described by Scully, and postulates that M. guentheri is a subspecies of M. socialis, a synonym of the subspecies M. socialis paradoxus Ognev and Geptner (1928). Lay (1967, p. 167) presents much further evidence from specimens of Iran and Turkey supporting conspecificity of M. guentheri with M. socialis. Microtus juldaschi Severtzov Pamir Vole Arvicola juldaschi Severtzov, 1879, Zapiski Turkest. Ot. Obs. Lub. Estest, I, p. 63. Type locality. — Lake Karakul, in Pamir Mountains, Tadzhik S.S.R. Distribution. — GENERAL: The Pamir Mountains in Tadjikistan, and Hindu Kush Ranges in northeast and central Afghanistan. Afghanistan Records: Niethammer (1970, p. 10). 1965 Street Expedition, FMNH 103118, -19, 20 miles SW of Eshkashem along the Faizabad-Eshkashem road. Third Danish Expedition, 1948-1949 (Panjao) AFGHANISTAN HABITAT: Pamir Voles were collected above 2,000 m. in montane habitat. Ognev (1950, p. 301), citing M. P. Pozanov, writes: "This Pamir endemic lives where wet mea- dows and alpine glens occur along rivulets and streams." I collected two at approximately 2,700 m. in the watercourse biotope which supports his generalization. However, he adds that this vole is not highly discriminating in choosing stations. The perennial stream along which these two were collected was large enough that it supported trout. The clay banks of this stream and the surrounding meadow were covered with alfalfa. HASSINGER: AFGHANISTAN MAMMALS 63 Discussion. — The Third Danish Expedition's record represents a range exten- sion of approximately 400 km. southwest of the nearest previously recorded locality for this species. An adult and juvenile were collected by the Street Expedition August 18 at the same spot, one at 10:00 p.m., the other around 7:00 A.M. The adult had two embryos. Mice of the genus Microtus were not generally attracted to our usual bait. The two specimens of M. juldaschi were each captured in a trap placed in a streamside runway. Ellobius fuscocapillus Blyth Afghan Mole- Vole Georychus fuscocapillus Blyth, 1842, Jour. Asiat. Soc. Bengal. 10, p. 928, nom. nud. 1843, Jour. Asiat. Soc. Bengal. II, p. 887. Type locality. — Quetta, Baluchistan. EUobius intermedius Scully, 1887, Jour. Asiat. Soc. Bengal, 56, p. 7;J. Type locality. — Herat, Afghanistan. Distribution. — general: Eastern Turkey and east throughout the moun- tainous periphery of the Iranian Plateau. AFGHANISTAN RECORDS: ScuUy (1887, p. 73), Aitchison (1889, p. 59), Niethammer (1965, p. 35; 1970, p. 21). 1965 Street Expedition: FMNH 103167-173. AFGHANISTAN HABITAT: They are vertically distributed between 500 and 2,600 m. in mountains and adjacent steppes. Common in the structure biotope, particularly on fallow terraces; also found in the clay-and-loess biotope. They are conspicuously absent in South Afghanistan. A search extending more than 300 km. in Monsoonal Afghanistan and the Jalalabad Vale failed to reveal any of the characteristic mounds of earth which indicate the presence of this vole. Taxonomic Notes and Discussion. — Scully (1887, p. 73) described Ellobius intermedius from three specimens collected at Bokun and Kila Wali (Qala Vali35°47'N, 63°44'E) in Herat Province. EUerman and Morrison-Scott (1951) and Ognev (1950) erroneously give Herat as the type locality for this alleged species. Scully (1887) used three characters to distinguish between E. fuscocapillus and his E. intermedius: the color of the base of the fur, 64 HASSINGER: AFGHANISTAN MAMMALS 66 Table 9. — Body measurements of Ellobius from Afghanistan. (Measurements in mm.) Head and Species Records Sample body Tail Hind foot E.fuscocapillus Scully (1887)' 3 114-127 10-12 20-23 I Aitchison (1889) 4 98-115(111) 12-16(14) 19.4-20.6(20.4) ' 1965 Street Exp. 7 114-127(119) 8-16(11) 21-25(23) E.talpinus 1965 Street Exp. 16 98-124(109) 6-10(8) 21-24(22) ' No mean given ; original measurements are in inches. the shape and structure of the zygoma, and the presence or absence of a posterior lobe behind the last outer angle of MS. Two years subsequent to Scully's description of intermedius, Aitchison (1889, p. 59) remarked: "The cranial and dental characters Scully gives, however, although at first sight they would prove to be of specific importance, prove to be so variable within the present series that I feel I must adhere to the above determination, [E.fuscocapillus] . . ." Ognev (1950, p. 609) is more to the point, noting: "For our part we are certain that E. intermedius Scully (from Herat) [sic] is a synonym of E. fuscocapillus; the analysis of features given by Scully proved fruitless, as none had any validity." In our seven specimens the variation of Scully's supposed taxonomic characters supports the quoted views of Aitchison and of Ognev. Scully's specimens from Qala Vali, a village in the Murghab River drainage, represent the northeastern most record (or fuscocapil- lus. Further north, but in this same drainage, Bobrinskii et al. (1965) illustrates U.S.S.R. distribution records for E. talpinus, and about 100 km. east of Qala Vali (in an abutting drainage) the Street Expedi- tion collected 15 E. talpinus. The geographical proximity of these records, the relative continuity of suitable habitat, and the absence of geographical barriers suggests that the ranges of these species may meet or overlap in the environs of Qala Vali. The overall appearance of the dorsal pelage of this species, unlike the duller, grayer looking pelage of talpinus, is Avellaneous to Pink- ish Cinnamon. As shown in Table 9, its body dimensions are on the average larger than those of talpinus. Fig. 15. The records and estimated limits of distribution for Ellobius fusco- capillus (F'.) and Ellobius talpinus (T'). Three records within the zone of dis- tribution overlap are for fuscocapillus. 66 FIELDIANA: ZOOLOGY, VOLUME 60 Field Notes and Discussion. — The mole-vole is aptly named. Earth is ejected upon the sm-face from the mouths of lateral tunnels of usually shallow burrows, form- ing small mounds about 300 mm. wide and 100 mm. high; subse- quently the lateral tunnel leading from the mound to the main pas- sage is plugged with soil. During the course of our expedition we excavated approximately 150 such lateral tunnels, and succeeded in finding the open main passage in about 75 per cent of them. I counted over 100 separate mounds of ejected earth on a 30 by 50 m. terrace between Kabul and Paghman in November. Approxi- mately 10 per cent of these mounds were fresh, indicating recent digging. Although abundant where they were found, we noticed some peculiarities in their geographic distribution. Mounds were locally abundant near the eastern edge of collecting locality 9, but we saw nothing suggesting the presence of mole voles near the western end of this same locality. Two roads separated by a hill connect Paghman with Kabul. No Ellobius mounds were seen along the route which traverses the southwestern exposure, but I saw over 1,000 mounds on fallow terraces and earthen banks along the alter- nate route, which traverses the northeastern exposure of this hill. Plugged burrows leading from fresh mounds were excavated with a spoon and small shovel on Sauzak Pass (Sept. 21) and near Herat (Sept. 16-20), Gardez (Oct. 8), and Paghman (Nov. 22-26). In most instances the plug ended within a third of a meter revealing the main burrow system. Mole voles were collected by so unplugging their burrow systems, hollowing out a chamber (150 mm. deep), placing a mouse trap baited with melon, onion, or stock bait in the chamber, making sure there was enough vertical space for the trap to operate properly, and, finally, excluding light and drafts from the excavation by placing a piece of cardboard covered with soil or a rock over the chamber. Approximately 50 trap nights equally dis- tributed between the above localities yielded four specimens. Of the 46 remaining sets at least 35 were rendered useless with fresh earth delivered by a resident mole vole. This experience stimulated me to repeat the procedure but to place a mole trap at the entrance of each. Doubtless to replug the lateral tunnel, a vole pushed earth into the trigger of the trap and was caught. Using this method at dusk (near Paghman) I collected three specimens from ten traps within an hour. We encountered what seems a difference in response from E. talpinus. Approximately 50 trap nights (near Mazar-i-Sharif, Sept. HASSINGER: AFGHANISTAN MAMMALS 67 6; and in Maimana, Sept. 8-14) using the chamber set yielded 16 individuals in this case, and only about 25 traps were covered with earth. These results, when compared with the above figures for a similar number of trap nights for E. fuscocapillus, suggest that the sampled segment of each species reacted differentially to the same stimulus. An environmental difference such as declining food re- source could have facilitated the successful use of the chamber set in North Afghanistan. The mean air temperature recorded at camp was 20.5°C or above during the period when talpinus were collected and below 15.0°C when fuscocapillus, without exception, ignored the proffered bait. Perhaps lower temperatures stimulated fuscocapillus to re-establish environmental equilibrium by shoving earth into the chamber to close off the leak of humid and constant temperature air, and incidentally by shoving earth into the chamber to cover the trap and food. Ellobius talpinus Pallas Northern Mole-Vole Mus talpinus Pallas, 1770, Nov. Comm. Acad. Petrop., 14, 1, p. 568. Type locality. — Kostytchi, west bank of River Volga, Russia. Distribution. — General: The southern Ukraine through southern Russia to Chinese Sinkiang and western Mongolia; reaching its southern limit of distribution in North Afghanistan. AFGHANISTAN RECORDS: Niethammer (1965, p. 35; 1970, p. 8). 1965 Street Expedition; FMNH 103151-166. AFGHANISTAN HABITAT: Found below 1,300 m. in the clay-and- loess and structure biotopes in North Afghanistan, it is com- mon on fallow terraces and alluvial fans. Ognev (1950, pp. 584-590) summarizes the natural history of this species, in- dicating that it is found in a variety of habitats. Discussion. — Niethammer (1965, p. 35) comments: "Dr. Meyer-Oehme re- ported it in Afghanistan from remnants in pellets collected near Balkh." Subsequently, Niethammer (1970) collected four talpinus near the Amu Darya (river) north of Taliq-an. The first collections of whole specimens were made by the Street Expedition. A descrip- tion of the method used to trap this species is given in the foregoing account on E. fuscocapillus. 68 HASSINGER: AFGHANISTAN MAMMALS 69 Gerbillus cheesmani Thomas Cheesman's Gerbil Gerbilhts cheesmayii Thomas, 1919, Jour. Bombay Nat. Hi.st. Soc. 26, p. 748. Type locality.— Iraq, near Basra. Distribution. — GENERAL: Iraq, Arabia, Iran, and Afghanistan. AFGHANISTAN RECORDS: 1965 Street Expedition: FMNH 103206-246, in alcohol 103226-35, skin only 103206, -07, -09, -10, -12-17, -19-22, -24, -25, -36. AFGHANISTAN HABITAT: Found below 1,000 m. in the sand biotope in South Afghanistan. Vegetation: medium to high density, low shade, one or two strata, clumped distribution, low to high homogeneity. Taxonomic Notes. — The overall tone of the dorsal pelage is sandy orange. There is a small white spot above each eye and behind each ear. Flanks, belly, and feet (including the soles) covered with white hair. Tail similar to back, sometimes white ventrally, and dorsally tipped with Light Drab hairs. The mean and extreme skull length for 11 skulls: 29.1-32.4 (31.3) . This compares with 27.3-28.4 mm. for three G. gleadowi from the Sind, W. Pakistan as given by Siddiqi (1961, p. 216) for the occipitonasal length. The mean and extreme dimensions for nine topotypical G. gleadowi skins given by Murray (1886, p. 246) in his original description of this Oriental gerbil are: HB 82-89 (83.9); T 117-133 (126.8) ; HF 28-30 (29.3) ; and E 9-11 (10.5). Apparently, cheesmani from Afghanistan (table 10) is a larger species than the typical gleadowi from the Sind. Whether these two species can be maintained as such, on the basis of size alone, remains to be seen. Field Notes. — The most northwestern record for this species was made by Mr. Atallah about 5 km. due west of the Kandahar International Airport terminal. This species was numerous in the sand dunes west of Spin Baldak. A census of active burrows revealed there were at least 100 animals per hectare in certain areas. In one such area we caught 30 speci- mens within two hours after dusk using 150 museum specials baited with our stock bait. Their burrows are from 40-60 mm. in diameter and usually located in sandy soil, often under a thorny shrub. They were easily trapped. I I i-H O o Oi -^ 00 o 03, ss CO c 1^ t> CO CD 00 lo is c CO (M (M ' CO Tl< ■k^l 1 1 1 1 1 .2 X o (M l-H lO -^ 5 CO (M (M CO CO CS pC be < oT B d CO 1— 1 s ,_^ s s >, T3 ci? o •r-i £3 O t> ^ — ^ in rO ,Q S- 05 rH ■to 03 73 oj^ t-^ 03 lO 00 o c Ej T-H '^ «o (M in o «4-l 0) a tH 05 00 (M T— 1 c^ o CO S 0) OS 1 o 1 00 1 ID 1 o CO 1 00 C3 w T3 pq El W Pi in 0) (^ 05 X in [iq ^ S S^ o ^ 5^ tj +j in Q, -« ej e! 70 HASSINGER: AFGHANISTAN MAMMALS 71 This species has at least one activity peak, at dusk. On October 31 we were setting traps in low ridges of sand located along a wadi about 15 km. west of Spin Baldak near the Afghan-Pakistan border. There was an appreciable amount of daylight remaining when Dr. Lewis called my attention to a small gully (about 1 m. from where we stood) with shrubby clumps of Haloxylon obscuring about 50 per cent of the sandy substrate. For approximately 3 m. to the right and left of our vantage point we were able to count at least ten G. cheesmani retrieving seeds from near the tops of the shrubby Haloxy- lon, some plants a half meter high. The climbers appeared to be agile. Approaching to within a meter of one gerbil-adorned shrub sent them scurrying in all directions. At this point there appeared to be twice as many as I had counted before. In any event, at least ten gerbils were feeding within 2 sq. m. Gerbillus nanus Blanford Baluchistan Gerbil Gerbillns nanus Blanford, 1875, Ann. Mag. Nat. Hist., 16, p. 312. Type locality. — Gedrosia, further specified as Baluchistan (W. Pakistan), Saman Dasht by Blanford (1876, p. 72). Distribution. — GENERAL: South of the Mediterranean from Algeria through parts of Arabia, Iran, Afghanistan, and Baluchistan eastward to the Punjab in northwest India. AFGHANISTAN RECORDS: Niethammer (1965, p. 33). 1965 Street Expedition: FMNH 103174-205, in alcohol 103174, -78, -83, -84, skin only 103175, -76, -77, -79, -81, -85, -86, -87, -88, -90, -91, -93, -94. (see: Discussion). AFGHANISTAN HABITAT: Typically found in the clay-and-loess, watercourse and structure biotopes, and to a lesser extent in the alluvial ecotone between the rock and clay-and-loess biotopes below 1,150 m. in South Afghanistan. We found their burrows in vegetation characteristic of the southern semideserts having a low to medium density, low shade, one stratum, a clumped distribution, and usually high homogeneity. Discussioji. — This species was previously listed by Sclater (1891, p. 51) as being collected from Gulistan in Afghanistan. J. A. Murray was cited as the collector. Murray (1887, pp. 50-69, 106-124) reports on the "Zoology of Beloochistan and Southern Afghanistan"; how- 72 FIELDIANA: ZOOLOGY, VOLUME 60 ever, the specimens he lists from southern Afghanistan are, with few exceptions, from the deserts and mountains surrounding Quetta in W. Pakistan. Accordingly, Gulistan is north of Quetta, but not in Afghanistan. Gerbillus was easily collected from mouse-size burrows using museum special snap traps baited with peanut butter. Three specimens were captured with the aid of a spotlight and net. Tatera indica Hardwicke Indian Gerbil Dipus indicus Hardwicke, 1807, Trans. Linn. Soc. London, 8, p. 279. Type locality. — United Provinces between Benares and Hardwar, India. Distribution. — GENERAL: Syria and northern Arabia through parts of Persia, Afghanistan, W. Pakistan, and India south to Ceylon. AFGHANISTAN RECORDS: Hutton (1845, pp. 137-139), (Accord- ing to Sclater, 1891, p. 48, J. A. Murray collected Gerbillus indicus = Tatera indica from Gulistan, Afghanistan, but as mentioned before, the "Gulistan" in question is in W. Pakis- tan.) Niethammer (1965, p. 33). Gaisler et al. (1967). 1965 Street Expedition: FMNH 103247, 103250-273, in alcohol 103252, -65, -80, -81, skin only 103272, -76, -77. AFGHANISTAN HABITAT: This subtropical species reaches the northern limit of its distribution in Afghanistan. It was usually found below 1,000 m. in the clay-and-loess, water- course, and structure biotopes in the steppes and semi-deserts of dry Afghanistan. Vegetation was frequently characterized by high density and low homogeneity in the vicinity of their burrows. Discussion. — The northernmost record of this species is Herat where we purchased four specimens that had been captured in local bazaars near our camp. The fact that our traps set beyond the city of Herat in Tatera habitat failed to yield this species suggests that man may Fig. 17. Records and estimated limits of distribution for Tatera indica. Four specimens caught within the city of Herat may be accidental introductants and as such may not reflect a continuous distribution this far north. 78 74 FIELDIANA: ZOOLOGY, VOLUME 60 have facilitated their distribution this far north. In fact, Gaisler et al. (1967) calls Tatera a hemisynanthrope in Afghanistan, but he adds, it does not invade continuous human habitations. As shown in Figure 17 their distribution is disjunct in Afghanistan. Nietham- mer (1965, p. 33) remarks: "Its distribution in Afghanistan is the same as that of Gerhillus nanus." I found no evidence which would suggest G. nanus occurred as far north as Herat or in the Jalalabad Vale. These nocturnal rodents were commonly found in colonies, occasionally in a single burrow. The highest density that we encoun- tered, as shown by trapping records, burrows, and well-worn trails, was along the steep banks of irrigation ditches which supplied the pomegranate orchards of Kandahar with water. As one example, six rat traps set at 2 or 3 m. intervals along a well-worn trail at the base of a mud wall separating two pomegranate orchards caught five adult Tatera. Meriones crassus Sundevall Sundevall's Jird Meriones crassus Sundevall, 1842, K. Sv. Vetensk. Akad. Handl., p. 233. Type locality. — Sinai: Fons Moses (Ain Musa). Distribution. — GENERAL: Western Sahara to West Pakistan reaching its north- ern limit of distribution in Iran and Afghanistan (see com- ments below) . AFGHANISTAN RECORDS: ScuUy (1881, p. 228), Ellerman (1948, p. 797). Niethammer (1966-67, in litt., Kandahar, and owl pellets from Herat.) 1965 Street Expedition: FMNH 103285-326, in alcohol 103324, skin only 103285-289, -291-294, -296-299, -301, -302. AFGHANISTAN HABITAT: Common below 1,500 m. in the clay- and-loess, and ephemeral watercourse (wadi) biotopes, and in the ecotone between the clay-and-loess and sand biotopes. Characteristic vegetation: low to medium density, low shade, one stratum, a clumped distribution and medium to high homo- geneity. Taxonomic Notes and Discussion. — Scully (1881, p. 228) described Gerhillus swinhoei (=M. crassus swinhoei) from Gatai between Kandahar and Kojak Pass, about Ji "be a 5 c T3 I ♦3 $ o 75 76 FIELDIANA: ZOOLOGY, VOLUME 60 10 miles (16 km.) north of the Afghan-Baluchistan border. He re- marks: "The animal, though not very old, appears to be quite full- grown." Measurements of the type compared with 42 specimens of adult crassus (including 22 virtual topotypes) contradict Scully's estimation of full growth. None of our material is as small as the type. I was able to examine the type skull which, contrary to Eller- man (1948, p. 797), is unbroken. Its occipitonasal length is 30.1. The occipitonasal length for 10 skulls from Afghanistan ranges from 31.1-35.5. See also Table 11. The type for M. crassus swinhoei was described from a growing subadult. Chaworth-Musters and Ellerman (1947, p. 487) and Ellerman and Morrison-Scott (1951, p. 647) make M. zarudnyi a subspecies of M. crassus, thus M. crassus zarudnyi. The recent comments by Lay (1967, p. 182) and the observations reported below in the species account of M. zarudnyi both support Heptner's (1937, pp. 189, 191) original assessment of the status of M. zarudnyi, that it is quite dis- tinct from M. crassus and other species of Meriones. The inclusion, as by Ellerman and Morrison-Scott (1951, p. 647), of: "extreme south Russian Turkestan," in the range of M. crassus is therefore no longer justified. Field Notes and Discussion. — When located, this species was easily trapped using museum special snap traps or rat traps, preferably the latter. Over 75 per cent of our catch was made between dusk and 10:00 p.m., although the traplines were run again after ten. A small, roadside guUey south of Kandahar had 100 burrow entrances in as many meters, which we recognized to be those of this species. Burrow orifices were often under thorny shrubs, although many were also found in areas devoid of vegetation. A wadi skirting a pebble desert southwest of Qala Bust contained burrows recognized as of this species, evenly distributed, about one per 15 sq. m. for 1 km. Meriones hurrianae Jerdon Indian Desert Gerbil Gerbillus hurrianae Jerdon, 1867, Mamm. India. Type locality. — India: Punjab: Hissar. Distribution. — General: Northwestern India and West Pakistan. Afghanistan Records: Sclater (1891), Thomas (1919, p. 72). — ^ 12 o5 00 - t2 """ »-l 1-H CJ '^ c: 5^ 2 2 ■* "ti OS «^ "=^ W « CO I ; « TO =2 M o t- « I I ^ 2 0 ctf CO ^ — o .5 I <-H tre I— I ^ t- 00 .53 aj TS O o. _ ^, .-H -H oi OS eg .-( i£ 00 "SR ^' i >>=" 3 = w u OS M — OS (O OS d X ^, , — . — ^ El) ?4 00 00 00 c £ £ E 4-> b. ■*^ s_-' a> C3 >> _>. £ X ■u 3 £ OS I— I .-H 0-1 « OS T3 C J3 C o •— C C .si — c bC o OS ,_, u 50 .*' OS u a; d :::^ d 2 a 00 X o c o X C- X sS no 4-> 2 *- a> ^ zS o £ t 0) £ §5 •H b. o 5 £ 2 m in to OS Z2 C 1 ? T3 s ^ 1 •c -ss "S, «*^ vl >»^ ss *5 «^ •< «*. "« ss s fc. u u k _s o o o s 0) V ^ a> a a 3 a ^ fti >> >>T3 >. H E-