22 BULLETIN 43, UNITED STATES NATIONAL MUSEUM. 



elongated rods. As a rule the second digit possesses a single rudi- 

 mentary phalanx which may be free or semianchylosed to the metacar- 

 pal. The highest degree of development is attained in the pteropines 

 and in the genus BMiiopoma^the former having three and the latter two 

 phalanges. In the pteropines the third is ordinarily furiushed with a claw. 

 The j)halanges vary greatly in the range of motion, those of the second 

 and fifth digits being the least mobile; in their relative lengths in the 

 pteropines and the genera N'octUio and Mimopierus^ these forms being 

 remarkable for the degrees present of lateral and dorsal flexion. It 

 has been noted on p. 5 that the disposition and relative sizes of the 

 phalanges vary in the scurrying and pendant forms. In the position 

 of tiight the row of first phalanges is flexed downward, but the row 

 of second phalanges is at the same time deflected laterally; i. e., to- 

 ward the body. In the position of rest tlie parts either remain axially 

 disposed or the row of the first i^halauges is laterally or dorsally flexed, 

 as in the molossines and emballanouriues. The terminal cartilages are 

 apparently absent in pteropines and rhinolophines. When present they 

 remain in axial line with the phalanges, as in phyllostomines (excepting 

 Tampyrus), or they are deflected from that line, as in vesper tilionines 

 and molossines. These little rods appear to be indices of the amount and 

 direction of strain to which the membranes are subjected, and point, 

 therefore, to distinctions in methods of flight. It may be said that 

 they are absent, or, if present, axially disposed in the broad-winged 

 forms, but are deflected in the narrow winged.* In vespertilionines 

 (excepting Gorynorhinus) the fifth digit is provided with an accessory 

 cartilage, which lies to the outer side of the terminal cartilage. It 

 slightly projects from the margin of the wing membrane.t 



The usual number of phalanges to each digit is two. In Phyllosto- 

 midjc the number is three to the third digit, the fourth and fifth having 

 two each. Yet in a specimen (apparently normal) of PhyUostonia has- 

 fatum I note three phalanges in the fourth digit as well as the third. 

 The third phalanx is probably a segmentation of tlie second rather than 

 a distinct joint added peripherally to the series. At least it so appears 



* The signiiicance of the cartilaginous tips to the digits is not clear. At first 

 sight they appear to be of the value of distinct phalanges, and I have until recently 

 held to this view, but exaniination_of sections under the microscope yields no indi- 

 cation of segmentation, nor does inspection of embryos show stages in which separ- 

 ate elements exist. Nevertheless I note in Macroin!^ excess of segmentation of the 

 last phalanx, and Dobson observes the same in the Molossi. Interesting properties 

 are noted in the elements of the metacarpal and i)halangeal series. As a rule, they 

 are rigid and brittle. In Fiiria these peculiarities are so marked as to make it 

 difficult to handle a specimen without incurring a chance of breaking these exceed- 

 ingly delicate structures. In the I'hyllostomidiE, on the other hand, they are apt 

 to be yielding and elastic, and suggest (at least in the case of the second and third 

 digits) that the bones are imperfectly ossified. Mr. C. Percy Moore, of the Uni- 

 A^ersity of Pennsylvania, who has kindly investigated the subject for me, states 

 that after attempts at decalcification the digits of bats everywhere yield tough 

 matrices. 



tPr. Acad, of Nat. Sci. 



