506 Mr. M. A. C. Hinton on

gradually transformed into mixed forests with alder and oak, in which hazel formed an undergrowth."

It appears to the present writer very probable that this rich occurrence of nut-bearing hazel was of very great importance to the big-toothed Bear, the jaw of which has bcen described above and, therefore, I have taken the liberty of indicating this connection with the name given.

E X P L A N A T I O N OF P L A T E IV.

-Pig. 1. Left half of ramus of mandible of ~ young individual of Ursus pruinosus, Blyth, from Kansu, West China; from above.

]*7 a. 2. Left half of ramus of mandible ot ~ Brown Bears ( Ursus arctos, Linn.), from Sweden ; from above.

~/g. 3. Left half of ramus of numdible of an extinct Be~r (Ursus arctos nueifragus, subsp, n., from ltalland, South Sweden; from above. (Type-specimen.)

I'hq. 4. Ditto, from outer side.

LI. - -The Nomenclature and Subspecies of the Purple-faced Langur, with Notes on the other Langurs inhabiliug Ceylon. By MARTIN A. C. H1NTON.

(Published by permission of the Trustees of the British Museum.)

BY the kindness of nay friend 5~[r. W. W. A. Phillips and Of the I)irector of the Colombo Museum, the British Museum has lately received a fine series of specimens illustrating the variation of the Purple-faced Langur in South-western Ceylon. This material enables me to clear up some points ~'hieh have long been doubtful, and to direct attention to some others upon which fur ther information is required.

Pithecus vetulus, Erxleben. 1777. Cercopitheeus vetulu,, Erxleben. Syst. Beg. An. p. 25. 1780. Cereopithecus kephalo2terus, Zimmermann, Geograph. Geseb. ii.

p. 185. 1785. Simia cepitalopterus, Boddaert, Elencbus Animal. p. 58. 1792. Cercopithecus silenus pu,Turatus ~ Kerr, An. Kingd. 1 ). 65. 1795. SigMa po~Thyrops , Link, Beitr~ige, ii. p. 62. 1800. 8imia veter, Shaw, Gem Zooi. i. p. 36 (nec Linnaeus). 1807. 8imia latibarba, Temminck, Cat. Syst. Ornith. et Quadrum. p. 3. 1812. Cercopit]zecus latibarbatus, Geoffrey, Ann. du Mus. xix. p. 94. 1825. Cercopithecus leucoTrymnus , Otto, N. Acta Acad. C~es. Leop.

xii. p. 505, pl. xlvi. his. 1833. Semnopithecus nestor, Bennett, Prec. Zool. Soc. 1833, p. 67. 1847. _Pre.sbytes thersites, Elliot, Blyth~ J. Asiat. Soc. Bengal, xvi.

p. 1271. 1876. 8cmnopithecus ketaarti, Sehlegei~ Monogr. des Singes, p. 52.

tl~e Lanqurs inlmb;ting Ceylon. 507

The above is a" complete list of the specific names which, at divers times, have been applied to the Purple-faced Langur. In addition, Simia dentata, Shaw (Gen. Zool. i. p. 24, 1800), and Semnopithecus j'uh, ogriseus, Desmoulins (Dict. Class. d'Hist. Nat. xlviii, p. 4,39, 1827), are sometimes cited in the synonymy of this species; but I can find nothing in the original descriptions to connect either of these names with this animal.

The Purple-faced Langur was first clearly referred to t)y John Ray (Syn. Au. Otmdr. p. 158, 16t3) iu the futlowing passage : - - " Cercopithecus niger, barba in cana promissa. WANDUaU Zeylanensibus. Ejusdem musei" it. e., "D. lgobin- son ~ museo Leydcns i" ] ; for the present species is the only one inhabiting Ceylon to which such a description cau apply.

The earliest technical name applied to the species is Erxleben's Cercopithecus velulus, diagnosed as " C. barbatus niger, barba alba. Hab. in Zeylona, Asia, Africa." Since the first determinable reference given by Erxleben is to Ray's description, quoted above, there can be no doubt that tile name vetulus primarily applies to tim present species, and therefore~ being valid and available for use in the genus Pithecus, this name, dating from 1777, must replace kepha- lopterus, Zimm. (1780), on the ground of priority. The fact that Erxleben, by error unavoidable in the )ear 1777, identified various other species with his vetulus does not, of course, affect the question. Many later writers, including Anderson (I878 and 188l) and Elliot (1913), misled by the secondary wider application attributed to vetulus by Erxlebeu have relegated the name to the synonymy of' Macaca silenus, and have not considered its bearing upon the nomenclature of P. "cephalopterus." Others, like Blanford (i889) and Forbes (1894), have rightly given the name a place iu the synonymy of the latter species, but have not applied the priority rule strictly.

I t will simplify matters to state at once that three weli- marked forms of the Purple-faced Langur are represented by the material now available. Indicating these forms as A, B, and C, they may be defined sufficiently for present purposes as follows :--

A. General colour of the body black. B. ,, ,, ~, dark brown. C. ,, ,, ,, pale, ashy to brownish grey. After carefully examining the literature, I find that all

the earlier names, viz., vetulus (based upon Ray's description of specimens iu the Leyden Museum); kel)haloplerus,

33 ~

508 Mr. M. A. C. t t inton on

cephahqderus~ silenus purpuralus, porphyro'ps, and latibarba (all based primarily upon Pennant 's Purple-faced Monkey) ; and latibarbatus and leueoprymnus (based upon more recently eolleeted material) refer to the form A. This is, therefore, the typical subspecies P. v. vetulus; and of it the British 5Iuseum possessed but two examples before the arrival of the fine series collected at Anasigatla by Mr. Phillips. These older specimens are : - -one presenled by the Colombo ~ u s e u m from the district of Pasdon Corola (79 .9 .5 .1 ) , and one collected by the Bombay Natural History Society's :Mammal Survey at Kottawa in the Southern Province (15 .3 .1 .7 ) .

On the other hand~ Bennett 's nestor (of which the type is now before me, B.M. no. 55.12.24,. 12) is based upon form B ; and of it SchlegeVs kelaarli would appear to be a synonym.

Witl, regard to Presbfles lhersites, Elliot and Blyth, based upon a living specimen obtained by Dr. Temi)leton at Nuera-kelawa to the west of Trincomali, there is some doubt, which can only be resolved by a re-examination of the type in the Calcutta Museum. By Blyth ~, Anderson ?., and more recently by D. G. Elliot :~ thersltes has been treated as a synonym of P. priamus; while Blanford §, finding nothing in the description to distinguish the type from " cephalopterus," could not " help suggesting that the so-called Presbyles thersiles was really a variety of Semno- pilhecus cephalopterus, perhaps approaching the variety called S. kelaarli hy Schlegel." Blyth l[ (who saw the type iu life, as well as after its death and preparation) describes thersiles as having no crest of hair upon the head ; and its "whiskers and beard more developed than in the other entclloid Indian species, and very conspicuously white, contrasting much with the crown and body, which are darker than in P. priamus.:" In these respects, no doubt, lhersiles, as described, resembles velulus; but Blyth states tha t the crown is darker and not lighter than other portions of the dorsal surface, and he makes no mention of the rump-patch so characteristic of all the known forms of velulus. There is no other evidence of the occurrence o[ vetulus in N.E. Ceylon, while P. priamus is there a common species. I t is quite possible that thersites is really

* Blyth, Catalogue, 1863~ p. 12. "~ Anderson, Cat. Mature. Ind. $lus. lSS1, i. p. 38.

Elliot, ]~ev. Primates, ii. p. 96 (1913). § Blanford, Proc. Zool. Soc. 18~,7, p. 62~J; Mammalia, 1889, p. 35. II Blyttb J. Ae. Soc. Ben~al~ xvi. p. I271.

the Langurs inhabili, j (5,ylon. 509

a distinct species; bnt even assuming, wit]l B]anford, it to be a form of veluhes, then it is clear that lhersites is a sub- species perfectly distiaet from any of the forms A, B, and C mentioned in tile above scheme.

I t wouhl appear, therefore, that form C has not hi therto attracted the attention of the systematist, and it is described [)elc)w as a new subspecies, P. v. phillipsi. The characters of the three subspecies now recognised may he set forth as follows : ~

1. Pilhecus vetulus velulus, Erxl.

Hair of crown directed backwards, not radiating, whiskers very long, concealing the lower part of the ears. Black supraorbital hairs but moderately developed. Hair of body of moderate length.

General colour of body and limbs black. Crown of head rni!ous or yellowish brown, more or less strongly contrasting with the shoulders. Whiskers white, but sometimes tinged with yellow. A conspicuous silvery-white patch, of sub- triangular form, upon tile r u m p ; the base of the triangle bordering the callosities arid involving more or less of tile root of the tail. Buttocks, hinder parts of loins, and outer surfaces of thighs fi'osted conspicuously by thesilvery hair-tips. Lower limbs and upper surfaces of hands and feet black. Naked parts of face, eallosities, palms, and soles black. Basal half or two-thirds of tail dark grey above and below, its distal portion white or yellowish, with a more or less well- developed terminal pencil.

Skull showii~g a considerable differe~ee in size between the two sexes ; the oecipito-basilar length ranging between 88 and 93'6 ram. in adult males, and between 81"6 and 8~1~'5 ram. in adult females. Facial portion short a ,d steel) , making with the horizontal plane passing through the external occipital protuberance and the gnathi(m an angle which ranges between 45 ° and 48 ° in adult males and between 49 ° and 53 ° in adult females.

For external and cranial measurements, see tables at pp. 513 and 514.

tiab. S.~V. Ceylon ; " wet a r ea " (average rainfidl 100"- 200" per ammm).

,llaterial ea'ami~ed.--Twclve from |]1o following localities : Anasigalla, Matugama, Kalutara District (eolleeled by Mr. W. W. A. Phillips), 7 ; Kalutara District (Colombo Museum), 3 ; " Pasdon Coro ia" (B.M., I)rescnted by Colombo Museum), 1 ; Kottawa, S.P. (Bombay Nat. Hist. See. Mature. Surv.), 1.

510 Mr. M. A. C. IIinton on

2. Pi lhecus velulus nestor, Bonn.

[Essential characters as in v. velulus. General colour o~ body (lark brown. Crown of head paler than rest of back, but contrasting less noticeably with the other dors'd parts than in the typical form. Rump-patch normal, silvery grey, but usually less conspicuous than in the other subspecies. Frost ing of buttocks and thighs inconspicuous or entirely absent. Hands and feet dusky above. Tail dark grey at base, paling to white or cream-colour distally.

Skull apparently intermediate between those of v. velulus and v. phill/psi.

For cranial measurements, see table at p. 514. Hab. S.W. Ceylon. Material examined . - -S ix , viz., four specimens, including

the type (B.M, no. 55 .12 .24 . 12, ex Zool. Soc.), without any exact locality; and two (B.M. nos. 7 2 . 9 . 5 . 2 and 3, presented by the Colombo Museum) fi'om "Raygam Corola."

R e m a r k s . - - B e n n e t t ' s type is a mere baby with full milk- dentition. But the other specimens include fully adult and even old individuals; and tile material seems sufficient to show that P. v. nestor is a well-marked subspecies~ dis- tinguished from the others by its peculiar coloration. Unfortunately, none of the specimens is accompanied by collector's measurements, or by exact information as to locality. " Raygam Coro la" is a district immediately to the north of " Pasdou Corola" and the Kalutara district ; and a great part of it lies within the " wet a r e a " ~" Possibly the present form will be found to represent true vetulus in the northern part of the " wet area."

As to its coloration, dark brown though it be, it can scarcely be regarded as an intermediate between the black v. vetulus and the cold-grey v. phil l ipsi; for while in each of the latter forms the contrasts offered to the rest of thc dorsum by the crown and rump-patches are as strong as possible, in v. nestor these contrasts are softcued down, this subspecies thus showing a well-marked tendency to produce a general unifbrmity of colour.

3. Pithecus vetulus phillipsi, subsp, n.

[Essential characters as in v. vetulus ; distinguished by its much paler general colour and more conspicuous crown- p:~tch.

Gcneral co]our of upper parts, including" the outer surfaces

* See p. 512 below,

t]te Langurs inhabiting Ceylon. 511

of the arms and thighs, cold brownish grey, darkest towards the shouhters and withers, l ighter and greyer upon the loins and thighs, where the whitish hair-tips increase in length. Crown of head covered by pale yellowish-brown hairs~ which are much lighter in colour than in the other subspecies. These pale hairs are continued backwards over the nape, instead of ending just behind tile oeeitmt. Whiskers large and white, in very sharp contrast with the surrounding parts. Tri- angular rump-patch wetl defined ; it, together with the root of' the tail, silvery white. Underparts, including the inner surfaces of the limbs, dark brownish g r e y ; but the chin and throat dirty white. Colour of outer surfaces of limbs gradually deepening distally; the hands and feet dusky or almost black above. Tail light brownish grey throughout, with the exception of the tip, which is white.

Skull with the facial port!on longer and forming a more acute angle (42 ° in an adult male, 41 ° in an adult female), with a horizontal plane passing through the external occipital protuberance and the gnathion, than in v. vetulus. The orbits also are slightly deeper antero-posteriorly in phillipsi; and correlated with this, firstly, the postorbital region is more constricted and the place of constriction some- what more posterior; and, secondly, the distances between m ~ and the pterygoid fossa and between m s and the condyle are greater. The sexual differences seen in the skulls of v. vetulus are not apparent in the skulls of the present form in the material before me. Possibly the only adult male skull now available belongs to an unusually small individual ; but if so, then there must be some considerable difference in size between v. vetulus and v. phillipsi, as will be seen by comparing the measurements of the female skulls ot ~ both forms in the table at p. 514.

For external and cranial measurements, see tables at pp. 513 and 514.

Type. Adult male. B.M. no. 2 3 . 1 . 1 9 . 1 . Original no. 60. Collected at Gonapola, Panadura District, Ceylon, Dee. 12, 1914, and presented to the British Museum by the Colombo Museum.

Hab. S.W. Ceylon, Panadura District (drier zone, average rainfall 75'/-100 '' per annum).

Remarks . - -We are greatly indebted to Mr. W. W. A. Phillips, who for several years has been engaged in collect- ing and working out the mammals of Ceylon. From the extensive material which he has gathered in the south- western part of the island lie has most generously enriched the national collection with long series of specimens accom-

512 M r . M . A . C . I l i n t o n o n

partied by most valuable notes. I I is labours will clear up many difficult problems left unsolved by pioneers like Kelaart, or by the re)re reecz~t work o~ tile Bombay Natura l I l i s to ry Soeiety's Mammal Smwev. I have, there(ore, great pleasure i t t " " " assomatmg Mr. Phi[lips's name with this fine mo,~key.

I t is a remarkable fact that in a small island like Ceylon the mammal ian fauna should be broken up into so many subspecies. Of course, the high relief in which part of the island is carved may account in some measure for this, al- though usually the differences between the highland and low- land forms of the island are specific ra ther than subspeeifie. But in the ease now befoce us, as in several others which I think could be cited, relief has nothing to do with the mat te r , for all three subspecies oceur on the low ground within qnite a small area, althou,,'h not in association with • . I D

each other. Some admirable maps of the rainfall of Ceyhm, published in the l{.eport of the Colombo Observatory for 1920, seem to throw a good deal of light upon the matter , suggesting that different subspecies have arisen in response to differences of humidi ty , in the south-west there is an ovate area, exteml'ing between latitudes 6 ° 3' N. and 7 ° 23' N., and through about 45' of' longitude, in which the annnal rainfall is not less than 100", and in places more than 200'C This tract forms what is termed above tile " wet a r e a " ; it is inhabited by the typical black subspeeies v. vetulus, and in the north probably by the slightly less saturate v. ,~eslor. This humid area is in emltaet with tlle eoast only between tile latitudes 6 ° 5' and 6040 ' N. Elsewhere it is surrounded by a narrow belt in which the annual rainfall wtries hetween 75" and 100" ; to the east this belt stretches aeross the island as a tongue, which ahnost reaches the eastern eoast. The narrow coastal strip to the south of Colombo belonging to this less humid belt is the eountry inhabited by the pallid P. v. phillipsi. The northern half of the island has for the most part an ammai rainfidl of between 50" and 75"; and a narrow belt of similar slight humidity oeeurs in the south- east. I~ the extreme north-west and iu the extreme south-east are still drier narrow eoastat strips in which the annual rainfall varies between 25 ~ and 50". Unless 1 °. thersiles proves to he a subspecies of' vetulus, these drier parts of Ceylon (25" -75" ) are inhabited by P . lwiamus alone ; the la t ter species, in turn, al though the most-widely distributed Langur of the island, is quite unknown from the humid area in the south-west. There is, of course, nothing novel in the idea that humidity may play a part in con- trolling either subspecifie var ia t ion or the distribution of

the Lany~trs 5dmtdgbLq (Teylon. 513

particular species; the object of the foregoing remarks is simply to point out that Ceylon offers a very f'~vourable site fi}r the investigation of such problems. I ts mammal iau fauna is extremely ,'ieh and varied, so that Nat,we has there 1)rovidedan abundance of controlexper imeuts . I t is great ly to be hoped that some resident in tits northern half of tim ishmd, and auother in the south-east will take up the mat ter wit, It Mr. Phillips, and by similar processes of- intensive eolleetiou and accurate observation gather materials from which all the Cil}ghalese inforalat iou may eventually be read.

In eonc'hlding this paper, brief mentioa may be made of the two other Langurs inhabiting Ceylon. In my opinion, P, ursb~u.% Blyth, of the highlands of Ceylon, is a perfectly distinct species from P. vetu&s, being distinguished by its much more robust build (weight to 20 lbs.), mush longer fur, different coloration (although traces of a silvery rural}- patch are more or less evident), and peculiar skull. Widt regant to P. priamus, of which a very long series, eollested by the Mammal Survey in the northern, eastern, and southern parts of the island, is before me, the existence of a crest of hair upon the vertex, as figured by Blanford ~, is shown to be a natural and constant eharaeter. I t will be remembered that Anderson suspceted this crest to be the result of preparation. Whether the island form of priamus is distinct sul}speeitically f rom that inhabit ing the south of Ind ia will remain doubtful pending the arrNal of fur ther Indian speeimeus-- for at present we have only two examples from the nminhmd.

External measurements (in millimelres), taken in flesh by tim collectors.

l'ithecus vebdus vetulus. Anssig,dla C W. W. i

A. Phillips) . . . . I 1,2alutara Dish'iet {

(Colombo Mus.). (

l ~. v. lgdllipsi, [ Pmmdura District

(Colombo 3[us.), I

r

~- Head &i ~xo., body. I Tail"

[

;62 ,540 730 •. 525 7..50 5 595 710 5 550 700 5ai 6~0 690

60 495 7 ] 2 63 560 770 61 830 7O0 62 4f}O 640 ,59 490 700

I

I Ilind b l o t ,

161 150 ]60 ]45 150 161 ]53 152 15a 140

E!rs

o 8

32 25 30 33 86 31 31 32

\Vei;~ht (lb.).

11.}

" ~ Blanford, Mammalia, 1839, p. 32, fig. 8.

514 0~ the Langurs i~habiting Ceylon,

Skull-measurements (in millimetres).

I I

~ctulus

?:

7e8i 93"6' 74"5

76 '5 58 '2 44'3

61 39"I 32'6 34'3 19'1

Condylo-basih~r length. Occipito-basilnr leugth. Zygomatic breadth .. Cranial length (orbit to

occiput) . . . . . . . . . . Cranial width . . . . . . l)ostorbitM constriction Greatest width across

orbits . . . . . . . . . . . . Condyle to m a . . . . . l)a|at[lar length . . . . . Canine to m a . . . . . . .

veO'dus'I" ~" ~'i

c~ °q- ~

64"7 . . 73.3 84.5 96'1 93.1 69.~ 71.4 75.1

70'5 75'5 71 55'4 56 57 41 43 41

54'9 61 60'2 35"7 .. 42 28 33"7 33'4 3l"1 3 5 " 5 33"1 18'1 19'9 18"5

Pithecus

' T "e.V's['hillipsi" 63.~.

74"7 9[ 72"2

69"5 58 41

58"9 42 31'7 33'5 19

zrsinus.

" ~ c ~

82.21 80 102 101 81'4 ..

80"71 77 / (;3"61 5~'3

42"3 4P9

64"2 62"9 47"31 43'4 35"5 I 34'5 36'1 I 36"5 19'8 20'3

P.S. , Feb. 1, 1923 : - Since drawing up the foregoing paper I have found two

impor t an t le t ters f r o m Mr. Phil l ips to me, dated December 5, 1921, and J a n u a r y 11, 1922, which by inadver tence I had over looked and forgot ten . I n the earlier le t ter Mr. Phil l ips says : - - " While examining the Pithecus monkeys in tile Colombo M u s e u m the o ther day I found tha t the P. kepha- lopterus of this district [Ma tugama] is apparent ly different f rom that of the Panadura distr ict nearer to Colombo. The form f r o m down here [i. e., P. v. velulus of this paper] , and f rom ttle U d u g a m a Hills, is very dark, a lmost black, with a very wel l -marked grey sacral region ; while tha t f rom nearer Colombo is dark brown, sacral region no t near ly so well marked, and, in fact, it seems almost in te rmedia te between the Kalu ta ra f o r m of P. kephalol)terus and the .P. ursinus of up-coun t ry . . . . . The boundary between the ranges of tile two forms seems to be the Kala Ganga River, which flows into the sea at Kalutara . I am hav ing a series of about a dozen fl'esh skins of each fo rm collected, which I hope to be able to submit to you before long . "

I n the second letter Mr. Phillips says : - - " P e r s o n a l l y I am quite satisfied that the two forms are d i s t i nc t - - inhab i t i ng different areas. No. 1 [P. v. velulus], which I regard as tile typical P. kephalopterus, ran ges f rom about R a n n a (S. Ceylon), t h rough the U d u g a m a Hills, in the Galla district, and up the wesLern coast to the Ka la Ganga River . . . . and it is

On t}~e Use of the Generic Name Braehycercus. 515

probably found in the Ratnapura district. No. 2 ~i. e., P . v . phillipsi and probably P. v. nestor~ is found only to the north of the Kala Ganga, inhabit ing the Panadura and Horana districts, and ranging up the west coast to a little wav north of Colombo, and inland up the Kelani Valley, probably to about the foot of the Adam's Peak Range. Both of them have a very restricted range, with the Kala Ganga as their mutual boundary, but [ am not as yet able to say exactly how far inland each form extends. I t eannot, however, be tar, as both forms arc eo~lfine(l to the wet z o n e " [ that is, the area in which the rainfall exceeds 75" per annum] .

" There can be no question of seasonal change, as the two series of skins have been collected at the same t ime of the year [December] . . . . and no question of age . . . . As far as I am aware, the two /brms are never found together ."

I have thought it bet ter to add these extracts f rom Sir. Phillips 's letters in a postscript rather than to remodel a paper which has already been suiticiently diflieult to write. i t will be seen that Mr. Phillips deserves the fullest possible credit for an interesting discovery, and tha t the results arrived at in this paper have beeu rendered possible by his enterprise alone. His description, in the earlier letter, of the fi)rm occurring north of the Kala Ganga seems to me to apply bet ter to P. v. nestor than to the specimens upon which I have based P. v. phillipsi. TMs, I think, tends to confirm the suggestion made above that the habitat of I-'. v. neslor is in the northern half of the wet zone proper (rainfall exceeding 100") ; while phillipsi inhabits the ra ther drier coastal belt (rainfall 75" to 100 't) to the south of Colombo.

L I l . - -On the Use of the Generic Name Braehycereus iu Plectoptera and Ortholotera. By HEanEar CAMrION.

Two recently-published generic names owe their origin to forgetfulness in the one case and long-continued neglect in the other of a genus established as long ago as 1834. The first of the two modern names to be considered is Euryceenis, employed by Dr. S. Bengtsson for a new genus of Mayflies, containing a single species removed by him from the genus Cwnis (Ent. Tidskr. xxxviii. 1 ). 186; 19t7) . That species