I 



PLASTIC CHAKACTKRS. XIX 



a similar ridgo along the iuner side, the two ridges being separated 

 by a median longitudinal depression ; in the sirailarlj* shaped lo'vrer 

 niola.riform teeth the inner ridge is formed by cusps 1, 2, and 3, 

 tlie outer by cusps 4 and 5 : 



(4) ig and m' are lost : 



(5) The second and, particularly, the third phalanx of the 

 second finger are somewhat reduced in length. 



The principal plastic characters of the Megachiroptera are 

 discussed in the next paragraph. 



II. Plastic characters of Megachiroptera. 

 1. Rostrum. 



Varies considerably in length, much less so in sliape. 

 Length. — If measured from the front of the orbit to the extremitj- 

 of the nasal bones, the cranial rostrum is : — 



(1) Longest, i. e. from about f to nearly ^ of the skull, in 

 Hi/psir/nathits and some species of Ejiomophorus, in both genera 

 relatively longer in males than in females ; the male ffypsignafhus 

 is the longest-faced of all Fruit- bats : 



(2) Medium or somewhat shortened, i. e. from about g to |^ of 

 the skull, in : — (a) the genera of the Bousettus section ; (b) all 

 Macroglossince ; (c) within the EpomopiTiorus section in the genera 

 Plerotes and Epomops, and in some (the smaller) species of Epomo- 

 phorus; (d) within the C'ynopferus section only in Myonycteris 

 and Spharias : 



(3) Shortest, i. e. about I— I of the skull, in :— («) the following 

 four genera of the Epomoplwrtis section, Micro pteropus, Nano- 

 nycteri.'; (rostrum rather more than ^ of skull), Scotonycteris, and 

 Casinycieris : (6) all genera of the Cynopterus group, except 

 Myonycteris and SpTK^rias. 



Briefly summed up : a moderately long rostrum is characteristic 

 of all Eonsettine and Macroglossine Fruit-bats, a very short of 

 nearly all Cynopterine genera, while the Epomophorine is the only 

 section showing any degree of variation in the length of the 

 rostnira, from that of Plerotes (moderate), on the one hand to that 

 of Ilypsrgnathus (longest known), on the other to that of Casi- 

 nycieris and allied genera (shortest). 



No Fruit-bat with a very short rostrum (third stage above) has 

 the full Megachiropteran number of cheek-teeth (the formula being 

 eilher j, j, or A\ but this rule cannot be reversed, that is, a 

 reduction in the number of teeth is by no means always, not even 

 usually, associated with a shortening of the rostrum. Fruit-bats 

 with a moderately long or very long rostrum (second and first 

 stage above) exhibiting any variation of the dental formula from 

 the highest number of cheek-teeth (|) to the lowest (0. 



Shape.— As a general rule the rostrum is conspicuously tapering 



62 



