and Its closest Allies. 501 



The skull is veiy large and heavily built (see measure- 

 ments on p. 507), but the facial portion (anteorbital width) 

 comparatively narrow; nasal swellings well developed, 

 inflated^ making the upper surface of the facial portion very 

 conspicuously convex*; portion between nasal swellings 

 and sagittal crest convex or flattened, not distinctly concave, 

 as in //. d'ladema ; hinder part of orbits (temporal fossa) 

 markedly broader ; sagittal crest higher, more abruptly 

 descending in front ; mandible much longer, the rami 

 higher. The teeth considerably larger. 



The central projection of the jjosterior leaf is more 

 prominent than in any other form of the diadema type, the 

 upper border of the leaf, therefore, trilobate f, in so far 

 recalling the shape of the posterior leaf in H. armiger. In 

 H. diadema the central projection is but slightly prominent, 

 the upper border almost evenly convex, as "a segment of 

 the circumference of a circle "J. In one specimen of 

 H. lankadiva there is a minute fourth lateral leaflet, external 

 to the third ; I have seen a similar individual aberraiion in 

 two H. d. vicarius from Borneo. 



Range. — Cey\on. 



Technical name. — The type locality of H. lankadiva § is 

 Kandy, Ceylon ||. Two skins from Dr. Kelaart^s collection 

 are in the British Museum, one of them marked (in Kelaart's 

 handwriting?) " H. lankadiva ; Kelaart ; Kandy, 1850,^^ and 

 both of them quite as described above. But in his original 

 description of H. lankadiva, Kelaart gives as length of the 

 forearm 3 inches (76"2 mm.), of the third digit 4;^ inches 

 (108 mm.), measurements which, if they are correct (and 

 there is no reason to doubt them), would indicate a bat much 

 inferior in size to any " lankadiva " I have seen. There is 



* 111 this respect H. lankadiva is perfectly iu accordauce with the 

 other bats described in the present paper; I emphasize this important 

 point to prevent confusion with H. armiffe?; a species to which 

 jff. lankadiva bears a certain superficial resemblance in the general outline 

 of the posterior leaf, but from which it differs widely' in the shape of the 

 facial portion of the skull. See the " General Remarks," below. 



t The peculiar shape of the posterior leaf in specimens from Ceylon 

 was pointed out by Blyth in 1863 (Cat. Mamm. Mus. As-iat. Soc. p. 20) ; 

 it seems to have been overlooked by subsequent writers, who are 

 unanimous in puttinjf " lankadiva " down as a synonym of '' diadema." 

 The shape of the leaf is not always easy to ascertain in dried skins ; it is 

 quite pronounced in two spirit-specimens at mv disposal. 



X Dobson, Cat. Chir. Brit. Mus. (1878) p. 187. 



§ " Lanka," I am informed on good authority, is a Hindu name of 

 Ceylon. Is "-diva" the Sanskrit "deva" (cojinate with Zn/s (Aetrs), 

 5(Vc, deus) ? 



II E. V. Kel.iart. * Prodromus Faun.T Ze}lanir.T' ' (Colombo. 18.5-2).p. 19, 



