352 Mr. G. E. Dobson on a Natural 



strengthened by an examination of the minute structure of 

 the hairs composing the fur, which will be found to agree in a 

 very remarkable manner in these families, and to differ not less 

 remarkably from that in the allied genera united in what I 

 have termed 



The Emballonurine Alliance. 



(Emballonuridse, Phyllostomidae.) 



Analysis of Natural Affinities. 



The Vespertilionidse are connected with the Emballonu- 

 ridse by Miniopterus, which agrees with all the genera in the 

 shortness of the first phalanx of the middle finger ; with the 

 genus Furia in the great elevation of the crown of the head, in 

 the position and form of the upper incisors, and in the tail, which 

 ends in the interfemoral membrane, not even the extreme tip 

 projecting. Natalus also resembles Furia closely in the very 

 short thumb, in the peculiar form of the tragus, and in the 

 tenuity and venation of the membranes. 



The very peculiar genus Rhinopoma, which I have placed 

 in a separate group, is connected apparently more closely with 

 Taphozous than with any other genus. The very long tail, 

 produced nearly double its length beyond the short truncated 

 interfemoral membrane, really perforates the membrane (as in 

 other species of the subfamily Emballonuridse) near its pos- 

 terior margin ; and the short first phalanx of the middle finger 

 is imperfectly flexed upon the dorsal surface of the metacarpal 

 bone, as in Furia — thus also agreeing with other species of this 

 subfamily except Noctilio. If Rhinopoma Hardwickii and 

 Taphozous nudiventris, which inhabit the same regions, be 

 compared, the general resemblance of these generically very 

 distinct forms is sufficiently evident. Both agree in the frontal 

 depression between the eyes, in the quality and distribution of 

 the fur, in the great deposits of fat about the base of the tail ; 

 while Rhinopoma agrees generally with the genus Taphozous 

 in the form of the foot and in the presence of a few scattered 

 long hairs near the extremity of the tail. This last-named 

 character may appear to be a superficial one ; but I consider it 

 very indicative of affinity. The position of this very curious 

 genus (which differs from all the Microchiroptera in possess- 

 ing two distinct phalanges in the index finger, as in the Mega- 

 chiroptera) among the Emballonuridse is further shown by the 

 microscopical characters of the fur, in which it agrees with 

 that family and with the Phyllostomidse, and differs from all 

 species of the Vespertilionine alliance. The presence of a 

 small nose-leaf, the longitudinal fissure on the muzzle, the 



