AMERBICAN MUSEUM NOVITATES Published by Number 1101 THE AMERICAN MUSEUM OF NATURAL HISTORY January 30, 1941 New York City

RESULTS OF THE ARCHBOLD EXPEDITIONS. No. 31 NEW AND MARSUPIALS FROM NEW GUINEA BY G. H. H. TATE AND RICHARD ARCHBOLD

The large collection of assem- mouth of the Fly River. Thence stations bled by the Archbold Expeditions of 1936 were established all the way up the Fly and 1938 to New Guinea, contains many River and into the Central Mountain interesting novelties, some of which are range to an altitude of nearly 1000 meters. here described. Work was done also along the south coast The 1938 party made its base at Hol- of the Western Division of Papua at landia on the north coast of the island Mabadauan and Strachan Island, and in whence, using a large aeroplane, it estab- the Central Division northeast of Port lished field bases on the Idenburg River Moresby up to Kagi, 1600 meters. and on Lake Habbema (3300 meters), on The mammals brought back by the 1938 the northern slopes of Mt. Wilhelmina. Expedition were collected by W. B. From Bernhard Camp (60 meters), the Richardson; those of the 1936 Expedition base on the Idenburg, subsidiary stations by G. H. H. Tate. Both Richardson and were occupied up the slopes to the crest of Tate employed native collectors to help the range south of the river, the altitudes them. Measurements, unless otherwise of which were recorded as 850, 1200, 1800, stated, were taken by Richardson or Tate. 2100 meters, respectively. Colors printed with capitals are those of From Lake Habbema two main lines of Ridgway.2 camps were prepared. One chain of sta- tions reached downward, following the , NEW GENUS courses of the Bele and Balim Rivers This genus needs comparison only with Lep- (with drainage to the south coast of New tomys, from which it is separated by its normal Guinea)-through altitudes of 2800, 2200 hind foot (elongated in Leptomys), and by the absence of m3-.3 and 1600 meters. A second series of sta- GENOTYPE.-Paraleptomys wilhelmina. tions passed to the eastern side of Mt. Wilhelmina and then drove upward to new species collecting stations at 3560 and 3800 meters. Paraleptomys wilhelmina, TYPE.-No. 150512, Amer. Mus. Nat. Hist.; Thus the entire northeastern fall from Mt. adult d; 9 kilometers NE. of Lake Habbema, Wilhelmina was ransacked for species at near Mt. Wilhelmina, Netherlands New Guinea; all elevations between 60 and 4000 meters. altitude 2800 meters; October 15, 1938; collec- The new forms come chiefly from the tor, W. B. Richardson (field No. 5143), 1938 New Guinea Expedition. The type is a skin heavy lowland forests of the Idenburg with skull in good condition. River where they seem to have been rare GENERAL CHARACTERS.-Superficially very and are represented often by single speci- like certain brownish-gray mountain inhabiting mens; and from the high country above Melanomys. Differs sharply from Leptomys (with elongate feet and pure white ventral hairs), 2000 meters, in which case large series are from Pseudohydromys (shrew-like), and from available for study. (Mus-like). The 1936 expedition1 placed its head- DESCRIPTION OF TYPE.-Skin dorsally colored quarters on the Island of Daru near the near Bone Brown, very slightly intensified along the back; ventrally dull white, with gray bases ' For full report see Rand, A. L., and L. J. Brass, 1940, Bull. Amer. Mus. Nat. Hist., LXXVII, No. 29, 2 Ridgway, 1912. Color Standards and Nomen- pp. 341-380. clature. 2 AMERICAN MUSEUM NOVITATES [No. 1101 which amply show through the white tips. Ears connecting the Doorman Top ranges with large, gray; hands and feet thinly covered with whitish hairs. Tail light fuscous above, white the Wilhelmina uplands. beneath, its terminal 1.5 cm. white. Tail-scales MICROHYDROMYS, NEW GENUS simple, flat, rounded to slightly hexagonal, 14 per cm. Scale-hairs 3 per scale, about one scale- The smallest known Hydromyine, ahd unique length, whitish. Skull: in many respects very in possessing grooved upper incisor teeth. Pel- like Leptomys; rostrum, braincase, zygomatic age, ears, tail and lips without modifications for plate, palate, pterygoid region and bullae are aquatic habitus. Feet neither modified for essentially the same. Differs in the absence of swimming nor elongate as in Leptomys. Skull m' exhibiting the short, broad muzzle and widened (reduced but still 3-rooted in Leptomy8), and interorbital region and braincase of Para- the diminution of flare of the maxillary root of , but the bullae, although low, are very the zygoma. much larger and proportionally much closer to- MEASUREMENTS OF TYPE.-Skin: total gether (width of bullae, 2.9; distance apart of length, 261 mm.; tail, 135; hind foot (s.u.), 31; bullae, 2.0); zygomatic plate essentially as ear from crown, 17. Skull: condylo-incisive Hydromys and Pseudohydromys; palatal foramina length, 29.2; zygomatic breadth, 16.0; length of more as in Leptomys, less shortened than in nasals, 12; interorbital breadth, 5.7; breadth of Hydromys; nasals unshortened. interparietal, 7.3; breadth of braincase, 13.5; TEETH.-Upper incisors with distinct groove height of muzzle at level of front of incisive separating anterior face of tooth into an outer foramina, 5.8; palatal length, 14.5; palatal third and an inner two-thirds. Molars reduced foramina, 4.8 X 2.3; diastema, 7.6; breadth to two in each row, as in most Hydromyinae, but between ml-1, 2.9; breadth of mesopterygoid the toothrows wider spaced in the palate and less fossa, 2.1; bulla, 4.9 X 3.6; width between convergent anteriorly than in Parahydromys and bullae, 4.0; mastoid width, 12.2. Teeth: Hydromys, the arrangement being essentially length mI + Mi2, 4.5; mi1, 3.0 X 1.8; Mi2, 1.6 X that of Pseudohydromys. 1.5; ml + m2, 4.6; ml, 2.7; m2, 1.8. GENOTYPE.-Microhydromy8 richardsoni. Assumption by m2 of the terminal posi- Microhydromys richardsoni, new species tion in the toothrow has so modified its TYPE-.NO. 152079, Amer. Mus. Nat. Hist.; form that the postero-external tubercle adult d; 4 kilometers SW. of Bernhard Camp, has been eliminated. In Paraleptomys ml Idenburg River, Netherlands New Guinea; differs from that of in altitude 850 meters; March 16, 1939; collector, Leptomys being W. B. Richardson (field No. 7677), 1938 New slightly broader. But a more important Guinea Expedition. The type is a skin in good difference is observed in the relatively condition, and skull with braincase badly broken. great depth of the fissures separating the GENERAL CHARACTERS.-A blackish murid laminae of seen from the outer about the size of a housemouse, externally un- three ml, specialized; with terminal centimeter of tail face of the tooth. Said fissures reach a white. depth of 0.5 mm.-virtually to the level DESCRIPTION.-Skin: pelage dorsally deep of the cingulum. In Leptomys the fissures fuscous, the hairs quite short (5 mm.), ventrally either do not at all or remain very slightly lighter and hairs slightly shorter. develop Ears neither reduced nor enlarged, gray. Feet minute (0.2). and hands slender, thinly clad with short blackish Paraleptomys wilhelmina is represented hairs. Feet unwebbed and the soles lacking the granular structure of Parahydromys and Hy- in the collection obtained by Mr. Richard- dromys. Tail blackish above and below, the son by almost 50 specimens, all from the terminal cm. white. Scales rectangular; per Lake Habbema region between 2200 and cm. 15. Scale-hairs fine, short, blackish, about 3000 meters. A second series of 30 speci- 1 1/2 scale-lengths. Skull: short and broad, with the general appearance of a tiny Para- mens from camps on the heights(1800-2100 hydromys skull. Muzzle short and heavy; meters) SW. of Bernhard Camp between nasals unshortened (but slightly exceeded by the Idenburg River and its affluent, the pmx.); interorbital width unconstrieted, at- Hablifoert River, may be separable as a taining half of zygomatic width; brainease broad and full (broken); zygomatic plate excised faintly differentiated race. This slight in front (as in most Hydromyinae); palate about difference is inconstantly manifested by as broad as Pseudohydromys (its posterior part the form of the anterior margin of the missing), and palatal foramina larger than in zygomatic plate; less excised in the Iden- Hydromys, their length 50 per cent of diastema; bullae larger in proportion than in any other burg , more excised in the Habbema known Hydromyine genus, width of bulla more race. The two areas of distribution are than one-third of combined width across bullae made continuous by way of the highlands and included basioccipital. RESULTS OF THE ARCHBOLD EXPEDITIONS. NO. 31 3

MEASUREMENTS.-Skin: total length, 172 the toothrows are somewhat more widely spaced, mm.; tail, 92; hind foot (s.u.), 20; ear from the width between ml' being rather more than crown, 8. Skull: length of nasals, 6.4; width twice the width of ml. Also the palatal foramina of nasals anteriorly, 2.6; width muzzle just an- are slightly less shortened, remaining nearly 50 terior to zygomatic plate, 4.2; least interorbital per cent of diastema. width, 4.4; mastoid width, 9; approximate MEASUREMENTS.-Skin: total length, 317 zygomatic width, 9.2; palatal length, approxi- mm.; tail, 170; hind foot, 36; ear (s.u.), 9.1 mately 10; palatal foramina, 2.5 X 1.5; dia- Skull: condylo-incisive length, 33.3; zygo- stema, 5; bullae, 3.8 X 2.9; distance apart of matic width, 17.4; least interorbital width, 5.6; bullae, 2; distance between ml-', 2.3; length width of brainease, 15.6; length of nasals, 11.2; ml + m2, 2.4; ml, 1.65 X 0.8; m2, 0.8 X 0.75. palatal length, 17.8; diastema, 8; palatal The discovery of this interesting "water foramina, 4.2 X 2.1; width between ml-', 4; width mesopterygoid fossa, 2.2; bulla, 4 X 3.4; rat" and of the allied form whose descrip- width between bullae, 5. tions follow, supports the already estab- The species is represented in Richard- lished view that New Guinea has been an son's collection by twenty-five specimens important focus of development for the from Lake Habbema (3225 meters) and Hydromyinae. four from a camp on the NE. slope of Mt. The species name commemorates the fine Wilhelmina (3560-3600 meters). work done for the New Guinea Expedition Hydromys habbema represents a new by its discoverer, Mr. W. B. Richardson. type of Hydromys-a very small, delicate Microhydromys richardsoni is founded form which in no way approaches known upon a single specimen. Hydromys. It seems to be the least spe- cialized of any species, and perhaps lies Hydromys habbema, new species near the base of that stem from which came TYPE.-No. 110055, Amer. Mus. Nat. Hist.; adult d; Lake Habbema, 15 kilometers north the highly specialized Crossomys and Para- of Mt. Wilhelmina, Netherlands New Guinea; hydromys and the less specialized H. altitude 3225 meters; August 18, 1938; collec- chrysogaster and H. neobrittanicus.' tor, W. B. Richardson (field No. 4722), 1938 New Guinea Expedition. The type is a skin XENUROMYS,3 NEW GENUS with skull in good condition. A new genus of "giant rats." Dentition ap- GENERAL CHARACTERS.-The smallest known proximates that of Uromys, but toothrows widely Hydromys (s.s.); hind foot not exceeding 40 min.; spaced in the palate and palatal foiamina much while in all others 50 mm. is passed. (Para- larger. Audital bullae quite small as in Uromys hydromys and Crossomys both attain 50 mnm.). and Macruromys. Zygomatic plate as broad Pelage gray with silvery tips, underparts pale and as well thrown forward but slopes more than gray, feet and hands whitish, terminial third of in Uromys, and is unlike the straight to slightly tail, white. Ears small, partly hidden by fur, excavated plate of Macruromys. The mandible but not extremely reduced as in Crossomys. of this rat differs in its weak development of the DESCRIPTION.-Pelage of back dense, soft, of insertion of the masseter. In adult Uromys and aquatic type, about 12 mm., its color Deep Rattus dominator the masseteric ridge develops a Mouse Gray, the tips silvery. The weak guard prominent masseteric tubercle. hairs black. No trace of the rich brown colors The teeth of Xenuromys besides resembling seen in large Hydromys. Underparts soft and those of Uromys, recall those of certain Rattus, dense, near Smoke Gray, the bases rather especially the R. dominator group of Celebes. darker, approaching Mouse Gray. Feet and But from Rattus they differ, as in Uromys, by hands with very short, silvery hairs. Tail fus- the relative simplicity of the laminae of the cous above and below, its terminal half white. molars. In ml and m2 the third loph is propor- Scale-hairs about three scale-lengths, increasing tionately much thickened from front to back markedly toward the tips (as in all Hydromys). (also a Uromys character); and correspondingly Ears gray, partly concealed in pelt. Skull: in the lower molars the anterior loph of ml is small and delicately built, but with all the much reduced, while its posterior median tu- structural characteristics of Hydromys. Com- bercle (in Rattus dominator resting in an enamel pared with H. chrysogaster subspecies, that of bay behind loph 3) is expanded in Xenuromys the new species has the muzzle slightly more and to a less degree in Uromys to form a con- slender and tapered anteriorly (as in Crossomys), siderable functional feature. The posterior and the braincase fuller and more nearly as wide 1 Note.-In 1937 I remeasured the external ear of as the zygomatic expanse. The supraoccipital Crossomys while in London: the height from crown region is also fuller and less rugose. The equalled 1 mm. are very at their 2 I do not concur with Rilmmler's (Mitt. Zool. Mus. pterygoids low, particularly Berlin, XXIII, 1, pp. 15-17) synonymizing of Para- anterior end, so that the ectopterygoid recesses hydromys and Crossomys. have become virtually obsolete. In the palate 3 From tvos, strange; Uromys, a genus of rats. 4 AMERICAN MUSEUM NOVITATES [No. 1101 portion of m2 shows a corresponding modifica- width of ml is less than half of the width be- tion. tween mi-'. The relationships of this interesting new In the lower teeth the first lamina of ml is may be considered to be with Uromys and much reduced (as in Uromys, but not in Rattus) Rattus, possibly near the point where the Rattus and the posterior tubercle of the same tooth, and Uromys stems diverged from each other, but which lies in a median position behind the third after the development of Uromys-like dentition. lamina, is correspondingly enlarged and broad- The enlarged palatal openings, broad palate ened. The same condition is seen with respect and rather full brainease may be subsequent to the posterior tubercle of m2. In both teeth in modifications analogous to those observable in Xenuromys the tubercles when slightly worn pre- Macruromys, from which in dentition it differs sent a large triangular pattern. In Uromys the sharply. pattern is that of a narrow elipse whose long axis GENOTYPE.-Xenuromys guba. runs transversely across the tooth. While in Rattus the tubercle, though an elipse placed as in Xenuromys guba,I new species Uromys, is less accentuated. MEASUREMENTS.-Skin: total length, 591 TYPE.-No. 152043, Amer. Mus. Nat. Hist.; mm.; tail, 281; hind foot (s.u.), 66; ear from adult d; Bernhard Camp, Idenburg River, crown, 25. Skull: condylo-incisive length, Netherlands New Guinea; altitude 75 meters; 62.2; zygomatic breadth, 29.0; least interorbital April 14, 1939; collector, W. B. Richardson breadth, 9.9; breadth of brainease, 22.8; (field No. 7847), 1938 New Guinea Expedition. breadth of interparietal, 12.5; mastoid width, The type is a skin with skull, the latter with 21.2; zygomatic plate, 7.6; palatal length, 35.0; rostrum broken, the former cut over the shoulder zygoma, 20.4; palatal foramina, 9.3. X 5.3; by the snare or trap. width between ml-', 7.6; width pterygoid fossa, GENERAL CHARACTERS.-Superficially like a 6.5; bulla, 6 X 3.5; width between bullae, 10.1; large gray Uromys with pure white underparts, length mI-3, 11.8; min, 5.7 X 3.3; mi2, 3.8 X 3.0; the tail coarsely scaled much as in Hyomys, but mi, 2.2 X 2.1; ml-3, 11.0; ml, 4.7 X 3.3; m2, the un-keeled scales in annular instead of spiral 3.7 X 3.1; ms, 2.7 X 2.4. arrangement. DESCRIPTION.-Skin with short, crisp pelage Xenuromys guba is described from a in which the guard hairs are few, short and little single specimen caught by Mr. Richard- differentiated. Dorsal color grizzled gray-the hairs dark gray with whitish tips, guard hairs son's native helpers in the heavy forest of black; ventrally white to the roots, with tint of the lowlands about the Idenburg River. light cinnamon under neck and throat. Feet The shattered state of the rostrum suggests and hands so thinly haired as to appear naked, it may have been killed by a blow with a white. Tail with basal half fuscous, terminal in half white. Tail-scales rectangular or with the stick while alive a snare. terminal margin slightly bowed, toward the distal part of tail becoming narrower and more Lorentzimys nouhuysii alticola, new sub- wedge-shaped; scale-hairs per scale, the median species one obsolete, white, about one scale-length. Ears small but not concealed in fur, naked, fus- TYPE.-No. 150592, Amer. Mus. Nat. Hist.; cous. Vibrissae very long as in or adultc?; 9 kilometers NE. of Lake Habbema, Mt. Hyomys. Skull: rather less massively built Wilhelmina, Netherlands New Guinea; altitude, than that of Uromys, with ampler muzzle and 2700 meters; October 28, 1938; collector, W. B. braincase, the latter approaching the fullness of Richardson (field No. 5261), 1938 New Guinea the braincase in Macruromys major. Zygomata Expedition. The type is a skin with skull in not much expanded; small supraorbital ridges good condition. developed; skull lacking the pronounced bend in GENERAL CHARACTERS.-A mountain in- profile to be seen in the naso-frontal area of habiting form, distinguished from the lowlands Uromys; palatal foramina large; palate broad; L. n. nouhuysii by the less extreme narrowing of mesopterygoid fossa broad, the ectopterygoid the ear conch to the peculiar wedge-shape of the wings unconnected with the alisphenoid at their typical form (Jentinck described them as outer, posterior tips; bullae quite small as in "pointed, Vespertilio-like"); by the much Uromys. longer, lax pelage (Jentinck wrote "the longest TEETH.-Upper incisors heavy, deeply pig- hair does not exceed 6 mm."); by the gray in- mented, slightly opisthodont. Molars about as stead of whitish cheeks; by the scale-hairs of heavy as in lowland Uromys with the posterior the tail being black instead of white. loph in m' and m2 enlarged, as indicated. Upper DESCRIPTION OF TYPE.-Color of back darker first molar proportionately narrower. The than in typical form; Mars Brown, as opposed toothrows parallel and so widely spaced that the to between Tawny and Russet-and much darker on neck and head where it approaches 1 The aeroplane, by means of which the 1938 Ex- Bone Brown. Underparts grayish white, with pedition was transported to Lake Habbema, was named "Guba." The word in Motuan dialect means fuscous bases. The dorsal pelage, 7-8 cm.; storm. ventral pelage, 7-8 cm. Ears naked; light RESULTS OF THE ARCHBOLD EXPEDITIONS. NO. 31 5 brownish fuscous. Hands and feet with short burg specimen is in fragments, but that of blackish hairs. Tail dark above, white be- neath, its scale-hairs whitish (scale-hairs black the mouse from the Fly River, though the in typical form). The distal half of the posterior braincase is much broken, has the rostrum margin of ear not excised, the tip of the pinna complete and the palate nearly so. thus far more obtuse than in L. n. nouhuysii. From that specimen I determined the Skull very similar to that of lowland form, but with slightly longer, narrower muzzle. Palate mammary formula in the field, namely: with rather prominent, rounded ridges extending 1-2 = 6. The specimen contained two from ml-' to the incisors along outside of palatal embryos. foramina. Nasals considerably narrower both in front and behind. That portion of maxilla anterior to the infraorbital canal nearly flat in POGONOMELOMYS Rb-MMLER new race, markedly inflated in typical form. Pogonomelomys RtMMLER, 1936, Zeits. f. MEASUREMENTS OF TYPE.-Skin: total length, Saugetierk., XI, p. 252. 211 mm.; tail, 128; hind foot (s.u.), 26; ear from "Melomyswith3-haired tailscales" TATE, 1936, crown, 19. Skull: condylo-basal length, 21.9; Bull. Amer. Mus. Nat. Hist., LXXII, p. 592. zygomatic breadth, 12.5; interorbital width, Pogonomelomys RtMMLER, 1938, Mitt. Zool. 5.6; breadth braincase, 11.0; mastoid width, Mus. Berlin, XXIII, heft 1, pp. 99, 166. 10.2; width of interparietal, 8.2; palatal length, 11.8; diastema, 5.9; zygomatic plate, 1.85; When studying Melomys mayeri, the palatal foramina, 2.6 X 1.4; breadth between genotype of Pogonomelomys designated by ml1, 2.6; mesopterygoid fossa, 1.0; bulla, 3.4 X nor I 2.6; width between bullae, 4.1; molars, ml3, Riummler, neither he wrote of the 2.8; ml, 1.4 X 0.9; m2, 0.9 X 0.9; m3, 0.6 X significant modification of the terminal 0.7. 3 cm. 4 of the skin of the dorsal surface Besides the type specimen of this race, of the tail (though his use of the term Richardson found three paratypes, three Pogonomelomys suggests he suspected it). additional specimens from the Bele River The presence of an additional specimen in (2200 meters), 10-15 miles NE. of the type Richardson's collection, plus his notes on locality, and seven others from the camps the manner of prehension with the tail in on the ridge just south of the Idenburg the case of a further species described be- River at altitudes between 1500 and 1800 yond, whose actions while alive were meters. watched by Richardson, has brought this To this race I refer a specimen (No. character sharply to my attention. 108454) which I trapped at Kagi (1600 In both the new specimen of mayeri meters) on the Kokoda Road, Papua, in (No. 152364) from the Idenburg River at March, 1937, and tentatively the two 400 meters and in the co-types collected by animals collected in the Astrolabe Moun- Shaw Mayer in the Weyland range at 1600 tains, which Rilmmlerl referred to the low- meters this caudal modification is expressed land form. through replacement of the large, obvious In our collections we have two specimens scales by a nearly smooth, less pigmented which I believe referable to true nouhuysii: tactile surface. This new character, No. 105036, a female from the upper Fly namely, the ability to flex the tip of the tail River, altitude 80 meters, and No. 152087, dorsally and probably to use it for prehen- a young male from near Bernhard Camp, sion, added to those already set forth by Idenburg River, 850 meters. Both have Riimmler and myself as distinguishing M. the clear cinnamon-colored, short pelage mayeri from Melomys (s.s.), warrants in mentioned by Jentinck. Ears and face my opinion full generic separation of (damaged by trap) cannot be studied in the Pogonomelomys. The great difference be- Idenburg specimen. But in the Fly River tween the dentitions of and the cheeks are white and the ears Pogonomelomys seems to insist that the extraordinarily narrowed and wedge- development of such peculiar type of caudal shaped, the angle formed by anterior and prehension has come about independently. posterior margins only slightly exceeding Pogonomelomys must be regarded as a 45°. Unfortunately the skull of the Iden- highly specialized offshoot of the stock from which Melomys, Xenuromys and 1 Rummler, 1938, Mitt. Zool. Mus., Berlin, XXIII, heft 1, p. 57. Uromys have risen. 6 AMERICAN MUSEUM NOVITATES [No. 1101

At Genoa in 1937, with the kind permis- length, 20.2 (22.6; 17.7); length palatal fo- sion of Dr. de Beaux I studied the of ramina, 5.1 (5.1; 5.2); breadth mesopterygoid type fossa, 3.3 (3.5; 3.1); bulla, length, 5.8 (-; bruijnii. That animal, a female from 4.7); m'l3, 7.1 (8.1; 6.5); width ml, 1.9 (1.95; Salawatti is very large, far exceeding in 1.75). size mayeri and a new form from the Fly P. b. brassi is based upon the type and a River whose description follows. part-grown female topotype without skull. Rummler included, with the climbing The close kinship of bruijnii, mayeri and bruijnii and mayeri, two smaller, terres- brassi is emphasized here. They are large trial-footed species, sevia from Huon Pen- Melomys-toothed rats with broad feet, very insula and fraterculus from Ceram. It is large claws and dorsally prehensile tails. questionable whether they ought now to be included; sevia at least, of which Richard- The new form is named for Mr. L. J. son took three closely related specimens at Brass, the botanist who accompanied all Lake Habbema, has no trace of specialized three of the Archbold Expeditions to New tail structure. But a large series of Guinea. another small species having prehensile tail and scampering feet is tentatively referred Pogonomelomys rummleri, new species to Pogonomelomys. TYPE.-NO. 150669, Amer. Mus. Nat. Hist.; young adult d; Lake Habbema, north slope of Mt. Wilhelmina, Netherlands New Guinea, Pogonomelomys bruijnii brassi, new sub- altitude 3225 meters; July 26, 1938; collector; species W. B. Richardson (field No. 4523), 1938 New TYPE.-NO. 105033, Amer. Mus. Nat. Hist.; Guinea Expedition. The type is a skin with adult d; Oroville Camp, Upper Fly River, 30 skull, in good condition. miles above d'Albertis Junction; altitude 60 GENERAL CHARACTERS.-A small dark brown meters; collector, G. H. H. Tate, 1936 New rat with Melomys-like dentition, feet unmodified Guinea Expedition. The type is a skin with for climbing (compare P. mayeri) but with dorsal skull in good condition. 3 cm. of tail provided with tactile surface for GENERAL CHARACTERS.-A large brownish- prehension. gray rat with large tail-scales, scandent, large- DESCRIPTION.-Skin with rather long fur (12 clawed feet and dorsally prehensile tip to tail. mm.), colored above near Bone Brown, becoming It was present with true Pogonomys mollipilosus Clove Brown along the back, the face with a and in the field was confused with it. grayish cast; underparts grayish white, the long DESCRIPTION.-Dorsal pelage slightly crisp, pelage with fuscous bases which show through. not short, essentially as in mayeri, the color Hands and feet clothed with light yellow-brown grayer, less rufous; ventral color white to the hairs. Ears fuscous. Tail brown; the scales base as in bruijnii and mayeri; ears fuscous; small and lacking the prominent keeled structure both hands and feet whitish, a narrow brown of true Melomys, their scale-hairs blackish, 3 per streak extending down the back of foot to be- scale and from 2 to 3 scale-lengths (compare P. tween bases of 3rd and 4th digits; tail longer mayeri). Skull much smaller than that of P. than body, fuscous above and below, with dis- mayeri, the rostrum much compressed and less tinctive dorsally prehensile tip; tail-scales large, shortened; supraorbital ridges undeveloped; hexagonal, 9 per cm.; scale-hairs 3 per scale, braincase fuller; zygomatic plate much narrower, about 1/3 scale-length, becoming near tip 2/3 of and nearly straight as in Macruromys; pal- scale-length; ears small. Skull, massive, well ate with elongate anterior foramina and quite arched in profile, with the well-developed supra- large posterior foramina; back of palate even orbital ridges, general palatal and dental struc- with back of m3; bullae small; angular process ture and the full, rounded bullae of bruijnii and of mandible quite short. Incisors narrow, un- mayeri. It diverges from bruijnii chiefly by its broadened, orange. Molars very small and of much smaller size, and from mayeri by being de- characteristically simple Melomys type. cidedly larger, particularly in the enlargement of MEASUREMENTS.-Skin: total length, 242 the bulla and the development of a strong pos- mm.; tail, 138; hind foot, 25; ear from crown, terior palatal spine. 17. Skull: condylo-incisive length, 26.1; MEASUREMENTS.-Skin: total length, 370 zygomatic width, 16.2; nasals, 10.3 X 2.8; mm.; tail, 197; hind foot (s.u.), 33. Skull': interorbital width, 4.5; width of braincase, 14.0; condylo-incisive length, 37.5 (-; 34.0); zygo- interparietal, 10.5; mastoid width, 10.8; zygo- matic width, 20.4 (23.4; 19.9); interorbital matic plate, 2.7; palatal length, 13.9; diastema, width, 6.7 (7.2; 5.9); breadth of braincase, 15.4 7.3; palatal foramina, 6.0 X 2.1; mesopterygoid (17.4; 15.0); diastema, 11.0 (12.7; 9.4); palatal fossa (taken between pterygoids), 1.9; bulla, 1 Measurements of type of bruijnii and of co-type 4.2 X 3.1; distance apart of bullae, 3.6; ml-3, (No. 101954) of mayeri follow in parentheses. 4.4; ml, 2.0 X 1.4. RESULTS OF THE ARCHBOLD EXPEDITIONS. NO. 31 7

Pogonomelomys riimmleri is founded American Museum numbers 109794 9 and upon an exceptionally fine series of thirty 105854 9: incisors, canines, pi, p3, in13m1-3arare specimens from the type locality and fifteen fully in place; p4 changing (several of the four more from Wilhelmina Top, 300 meters minute milk premolars still in place and the tips of the permanent set plainly visible); m4-4 vir- higher. Many of the old specimens show a tually in place, but m4 - 4only half erupted. distinct tawny wash over the tips of the American Museum numbers 105795 c, ventral hairs. Adult specimens not neces- 152034 9, and 108562 9 (younger): the sarily old also show great wear of the minute, 2-rooted milk p4- undisturbed, m3 half molars, probably accountable by conditions erupted; m4 undeveloped. In this specimen milk P4 iS vestigial and single-rooted, and re- of soil and vegetation in their habitats. mains, like certain bats' teeth excluded at the outer margin of the toothrow. PLANIGALE TROUGHTON at a Planigale TROUGHTON, 1928, Records Aus- These dental changes take place tralian Mus., XVI, p. 282. very early age-when the animal has Wood Jones' drew attention to many reached only about one-half of its adult characters of the Australian Phascogalinae, weight. among then distinguishing between "Phas- The vestigial condition of the milk p '4 cogale" (Phascogale and Antechinus) and especially of the single-rooted tooth of the Sminthopsis. His illustrations particularly lower jaw may account for L6nnberg's be- showed characters of P. flavipes, type of lief that the absence of the "lower secator" Antechinus and of S. crassicaudata, type of of his Phascogale subtilissima (a Planigale) Sminthopsis, comparing the short foot is a generic character. On the other hand, with striated pads of the former with the photographs of the skull of his type in my slender foot with granulated pads of the possession show the upper permanent latter. In the skulls he noted distinctions fourth premolar to be large and completely in the interorbital area and the nasals. erupted, though unworn. So permanent Iredale and Troughton2 have separated P4 of his species may really never develop. Phascogale Ternminck and Antechinus It has yet to be shown that the tooth Macleay. The former contains only the succession indicated for melanura prevails brush-tailed tapoatafa (the genotype) and among other species. Nevertheless, mi- calura, in both of which the bulla is rela- nute, obsolescent, or seemingly obsolete tively greatly enlarged and the periotic is fourth premolars should be viewed with also inflated. Thus Antechinus contains suspicion. For normally, in adult animals generalized forms while species referable to 44are the largest or next to largest of the Phascogale (s.s.) and Sminthopsis show premolars. specializations. Planigale appears to be related to An- The majority of the Phascogalinae from techinus but differs by its much flattened, New Guinea have striated pads, rather widened skull; short, wide palate and ros- short feet, small bullae and moderately trum; heavy feet, with pads faintly stri- broad interorbital region, thus fitting in ated.3 with Antechinus and Phascogale rather than The form described beyond had the skull with Sminthopsis. markedly flattened. Unfortunately, by The tooth succession in the "phasco- some accident the entire braincase behind gales" is an important feature which must the broad interorbital region has been be understood when working upon their broken and destroyed, and specific charac- classification. Usually material for study ters can be based only upon what remains of this process is lacking but in the case of and upon skin characters. Antechinus melanura several young speci- mens in our collection demonstrate it per- Planigale novaeguineae, new species fectly: TYPE.-NO. 108561, Amer. Mus. Nat. Hist.: adult d; Rona Falls, near Port Moresby, Cen- 1 Wood Jones, 1923, The Mammals of South Australia, pp. 94-122. 3 Troughton describes the pads of P. i. brunneus aF 2 1934, A check list of the mammals recorded from granulated, "the granulations comparatively large." Australia, p. 7. I found faint striations, not granulations. 8 AMERICAN MUSEUM NOVITATES [No. 1101

tral Division, Papua; altitude 250 meters; rock. The hillside was comparatively bar- Jan. 20, 1937; collector, G. H. H. Tate, 1936 New Guinea Expedition. The type is a skin in ren of vegetation. good condition and skull with posterior half Planigale novaeguineae is decidedly larger missing than any hitherto described (ingrami, brun- GENERAL CHARACTERS.-A small brownish- neus, tenuirostris and subtilissima) -all from gray marsupial mouse, with very low ears deeply Australia. notched at their posterior margins; heavy short feet; and tail shorter than head plus body. DESCRIPTION.-Skin: dorsal color brownish Antechinus tafa centralis, new subspecies gray, formed by a mixture of short, light brown, TYPE.-NO. 109823, Amer. Mus. Nat. Hist.; black-tipped hairs and inconspicuous black guard adult d; Bele River, 18 kilometers north of hairs; ventral color light tan, gray-based, except Lake Habbema, north of Mt. Wilhelmina, Neth- chest and throat where the hairs are light brown erlands New Guinea; altitude 2200 meters; to bases. Hairs of scrotum vinaceous; of hands Nov. 17, 1938; collector, W. B. Richardson, 1938 and feet pale brown; those of tail fuscous above, New Guinea Expedition. The type is a skin light brown beneath. Ears peculiar, very low, with skull in good condition. projecting 7 mm. above crown, but rather ample GENERAL CHARACTERS.-Separable from true posteriorly, the opening from meatus to outer tafa only by cranial characters described below. pinna very high (7 mm.). (Note that the ears DESCRIPTION.-Skin: above dark gray, with shown by Troughton, plate 39, have been re- a glint of light brownish, grayer on head; be- volved. Their vertical height parallels a di- neath dull grayish white with gray bases; transi- agonal line from upper left to lower right across tion from dorsal to ventral color not well defined. the page.) The posterior margin of the pinna in Ears fuscous. Hands and feet with light gray novaeguineae is deeply notched. Ear finely hairs. Tail brownish gray, becoming yellowish haired internally and externally. Faint traces gray beneath-the tip with slight hair pencil. of striations on pads (rest of foot normally Skull, distinguished from that of true tafa by the granulated as in Antechinus and other genera). following characters: enlargement of the pos- Skull: when newly caught the specimen had terior palatal fenestrae; marked increase in the braincase extremely flat and low. Destruc- internal width of the mesopterygoid fossa just tion of the braincase precludes demonstration of behind palate. that fact. Rostrum heavy, broad and flat ex- MEASUREMENTS.1-Skin: total length, 281 tending backward to broad interorbital region; mm. (279); tail, 146 (145); hind foot (s.u.), 26 nasals twice as broad proximally as distally, (24); ear from crown, 15 (17). Skull: condylo- without notch between them at tips (as in incisive length, 34.3 (32.0); zygomatic width, Antechinus flavipes); palate much broken, its 19.2 (17.1); least interorbital width, 7.8 (7.7); incisive foramina reaching back only to middle of width braincase, 13.9 (13.4): palatal length, 18.0 canine alveoli (posterior vacuities destroyed). (17.0); posterior palatal fenestrae, 4.0 (3.4); Incisors: il-' twice the size of other incisors, greatest inside width of mesopterygoid fossa, 4.5 not contiguous; i3-4 subequal, slightly larger (3.8); bulla, length, 3.9 (3.1); toothrow, il-m4, than i2; lower incisors absent (alveoli only). 17.6 (17.3); ml4, 7.3 (7.5); ml-3, 6.3 (6.3); Canines straight, slender. Premolars: appear- width ml, 1.7 (1.7). The smaller size which ing crowded, becoming successively and sharply shows throughout is due to the type of true tafa larger from pl to p3 to p4, all with high triangular being a female. In a female (No. 109809) of A. blades and 2-rooted; each lower premolar with a t. centralis the width of the mesopterygoid fossa "heel," especially well developed in p3; pi absent is 4.3, and the length of the fenestrae, 4.1. (2-rooted) but unquestionably larger than re- The new race is represented by a total of duced P4, and much smaller than p3; P4, though 22 specimens from localities near Mt. Wil- very small and single-rooted, is apparently a permanent tooth, not to be confused with the helmina, varying from 1500 to 2800 meters. minute milk p4 of the deciduous dentition (m4 completely in place). Antechinus habbema, new species MEASUREMENTS.-Skin: total length, 170 TYPE.-NO. 109812, Amer. Mus. Nat. Hist.; mm.; tail, 78; hind foot (s.u.), 14; ear from adult d; 9 kilometers NE. of Lake Habbema, crown, 7. Skull: nasals, 9.3 X 4.1; inter- north slope of Mt. Wilhelmina, Netherlands New orbital width, 6.1; anterior palatal foramina, Guinea; altitude, 2800 meters; collector, W. B. 1.8 X 1.5; length toothrow, 10.8; ml-3, 4.3; Richardson, 1938 New Guinea Expedition. The crown of ml, 1.7 X 1.2; m2, 1.4 X 1.6; M3, 0.75 type is a skin with skull in good condition. X 1.7. GENERAL CHARAcTERs.-Astonishingly like about the rock- A. tafa centralis, whose upper distributional Intensive trapping great range it shares. Differing by the much smaller, strewn slope where this animal was caught more delicate feet and hands, and much smaller failed to produce any more. It was taken 1 Measurements of A. tafa tafa are placed in paren- in a dryish place beneath an overhanging theses. RESULTS OF THE ARCHBOLD EXPEDITIONS. NO. 31 9 teeth. Its nearest relative must nevertheless be Rona, near the Laloki River, Central Divi- regarded as A. tafa. DESCRIPTIoN.-Skin: color and quality of sion, Papua. pelage above and below almost exactly that of A. While working the savanna country of tafa centralis. Ears fuscous. Tail above as A. the middle Fly River at Lake Daviumbu t. centralis, beneath darker, the hairs fuscous. a large series of more than 60 specimens of Feet and hands much smaller and more deli- this interesting little marsupial was se- cately formed; the width of foot across base of 5th metatarsal, 3.7, in A. t. centralis, 5.1, and in cured. Eight more were trapped at Ta- the type of A. t. tafa ( 9 ), 4.3. Skull: general rara, Wassi Kussa. The big series at character of tafa but smaller. Back of palate Daviumbu could be secured because of the well behind m4-4, and supplementary fenestrae interdigitation of tongues of low grass- appearing between the posterior palatal fenestrae and the back of palate. Teeth smaller, molars savanna with swamps and flooded exten- also narrower than in tafa. sions of the lake. Burning of some of the MEASUREMENTS.-Skin: total length, 252 grass areas resulted shortly in a high con- mm.; tail, 140; hind foot (s.u.), 21; ear from crown, 13. Skull: condylo-incisive length, centration of a previously well-dispersed 29.6; zygomatic breadth, 6.5; least interorbital fauna among the unburned vege- width, 7.7; width of braincase, 13.0; palatal length, 15.7; posterior palatal foramina (pri- tation at the waterside. mary), 3.9, (secondary) 0.8; width mesoptery- In the summer of 1937 I was privileged goid fossa, 3.7; length bulla, 2.9; toothrow, the British Museum il-m4, 14.6; ml'4, 6.1; 5.3; width ml, 1.4. to visit (Natural ml-3, History) and study among other types that This species (it must be so considered) is I was struck with the close founded upon five specimens. It was taken of rufigenis. from 2200 to 2800 meters in the neighbor- relationship of my large series from hood of Lake Habbema. Daviumbu to the types of rufigenis and rona. Further study establishes that all Sminthopsis in South New Guinea have nasals unwidened behind, narrow feet Thomas' described a species rufigenis and pads distinctly granulated instead of from the Aru Islands which he compared striated, and are thus members of the genus with S. virginiae Tarragon from North Sminthopsis-atypical members it is true, Queensland. A few years ago I published do not become swollen. a description2 of Phascogale rona from since the tails Many of the females had young in the 1 Thomas, 1922, Ann. Mag. Nat. Hist., (9) IX, p. 265. 1922, Nova Guinea, XIII, pp. 739-740. pouch. Mammary formula invariably 3 2 Tate and Archbold, 1936, Amer. Mus. Novitates, No. 823, p. 2. pairs = 6.