84 BULLETIN 57, UNITED STATES NATIONAL MUSEUM. 
proximal phalanx of third finger. In all the known genera the tail 
perforates the interfemoral membrane and appears on its upper sur- 
face distinctly back from the edge. 
Next to the Rhinopomidae the Emballonuridse appear to combine 
the greatest number of primitive characters with the least degree of 
specialization. The head of the humerus is nearly as in Rhinopoma. 
As in the Rhinopomidse the premaxillaries retain their Pteropine 
condition of freedom; while another character pointing in the same 
direction is seen in the long, curved postorbital processes. On the 
other hand, the index finger has completely lost its phalanges and a 
curious specialization appears in the flexing forward of the proximal 
phalanx of the third finger. The other portions of the skeleton as 
well as the general external structure are neutral. 
History. — This family was first recognized as a distinct group by 
Gervais in 1855. It was treated by him as a tribe of the Vesper- 
tilionidse under the name Emballonurina, and was composed of 
Urocryptus (=Saccopteryx ) , Diclidurus , Emballonura , Proboscidea 
(= Rliync his cus ) , C entronycteris , and Furia ( = Fuript erus) . Except 
for its subordinate rank it was therefore essentially the same as the 
present family Emballonuridse, the only difference being the inclusion 
of the genus Furipterus , a member of the family Furipteridse. Ten 
years later Peters enlarged the group by adding the genera Mysta- 
cina (=Mystacops) , representing the family Mystacopidse, and Noc- 
tilio , representing the family Noctilionidse. He also discarded or 
overlooked the name proposed by Gervais, and substituted the new 
term Brachyura. a Gray, in 1866, practically returned to the classi- 
fication of Gervais, but improved it by eliminating Furipterus. He 
also separated Diclidurus as a group Diclidurina, of equal rank with 
the Emballonurina, Furipterina, and his eight other subdivisions of 
the family Vespertilionidse. 6 Fitzinger placed the group, enlarged 
by the addition of Chilonycteris , Mormoops , and N octilio , as a sub- 
family Noctiliones of the family Vespertilionidse (“ Vespertiliones”) . c 
Gill returned to the more natural arrangement of Peters, but used 
Gray’s name Noctilionidse/ 7 In 1875 Dobson combined the Brachy- 
ura and Molossi of Peters to form the family Emballonuridse. This 
strikingly artificial assemblage has been accepted as a family by most 
recent authors. Both Winge 6 and Harrison Allen/ however, re- 
« Monatsber. k. preuss. Akad. Wissensch., Berlin, 1865, pp. 257-258. 
Ann. and Mag. Nat. Hist., 3d ser., XVII, pp. 92-93, February, 1866. 
c Sitzungsber. k. Akad. Wissensch., Wien, Math. Nat. Cl., LXI, Abtli. 1, pp. 
458-530, 715-828; LXI I, Abtb. 1, pp. 13-66. 
d Arrangement of the Families of Mammals, p. 17, November, 1872. 
® Jordfundne og nulevende Flagermus (Chiroptera) fra Lagoa Santa, Minas 
Geraes, Brasilien, p. 24, 1892. 
f Monograph of the Bats of North America (1893), p. 162, March 14, 1894. 
