Ili])|io.si<li'rn.s armigor (;«7 (^omnuMsoiii 'i)//"'*- -'''J 



Dentition nuuli as in diadcmn, bnt slightly more ad- 

 vanced : — the eingula o^ ju and p^ generally overlapping each 

 other, often very htrongly so, rarely in .simi)le contact withont 

 overlapping ; in none of tiie twelve skulls cxaniincd, repre- 

 senting all the foinis known, is there any interspace between 

 7^3 and p^. ;>-' always external to the tooth-row, often 

 nnnsually small ; the upper canine and /^* in contact, some- 

 times slightly overlapping eacli other at base, rarely separated. 

 A small, but always venj distinct, cusp-like prujection on the 

 hinder cnftint/ rdt/c of the upper canines, a little above the 

 middle of the tooth (no trace of a similar " cusp " in diadema 

 nor in arniiyer) ; front face of upper canines more or less 

 distinctly furrowed, in H. giyas very deeply so (practically 

 smooth in diadema and arniiyer). 



AVing-structure * on a considerably higher level of develoj)- 

 ment than in diadema and its allies : — third and fourth 

 metacarpals slightly shortened, jifth lengthened, making as a 

 final result the third metacarpal a little longer than the 

 fourth, but the ionrth and fifth practically equal iti length 

 (individually the tilth metacarpal can be even a trifle longer 

 than the fourth) ; broadly speaking, these three metacarpals 

 might be called approdimateli/ equal in length (in diadema 

 the fifth decidedly much shorter than the fourth and third). 

 Distal phalanx' of the third Jiixjer much lemjthened (much 

 longer than the proximal phalanx; in diadema only about 

 equal to, or shorter than, the proximal phalanx). These 

 modifieations combined make a broader and, especially, more 

 pointed wing, i. e. an increased power of flight. 



The nose-leaves are, in their more essential characters, of 

 the diadema type. Four supplementary leaflets on either 

 side, external to the horseshoe, the fourth always the 

 smallest, but very rarely [H. yiyas (jainbiensis) completely 

 ^^ anting; I never found more than four leaflets f. The 



* Compare tbe wing-indices Lolow, on p. 48. 



t The statements in literature about the number of supplementary 

 leaflets in these bats are very confut^ing : — two (Temminek, in his 

 " J'h. rittdta''); three (Warmer and others); "three, sometimes witli 

 the rudiment of a fourth"; four; "four, often with the rudiment of u 

 tiftli." Suppo.«ed dillerences in this respect have been used as a means 

 to discriminate subspecies or specie.*. The facts, from the material 

 examined by me, aie these: — JI. Commersoui ti/jncus: four leaflets, 

 the fourth quite small (one alcohol specimen ; one skin innuitablo 

 for this purpose); H. C. marunyensis: four leaHets, the fourth small, 

 but distinctly larger than in the .Madaj:ascar form (two in ale, one skin) ; 

 II. thomensis: c^uite as iiiarutKjittsis (one skin; two other skins unsuit- 

 able) ; //. {/iyan (i// icils: rjuite as miininynisis (four in ale, three -kins, a 

 fourth skin unsuitable) ; //. //. yamhiensis: three leaflets, without any trace 

 of a fourth (one, in ale). — The divergent statement- iu literature are, I 



