58 BULLETIN : MUSEUM OF C0MPAKAT1VE ZOOLOGY. 



Of the specimens from St. Kitts, two lack the tbird lower incisor on both sides, 

 and one has lost that of the left side. The upper anterior premolar is so minute 

 that except in one of the bats from St. Bartholomew's it cannot be seen until it is 

 uncovered by scraping away the gums. 



Molossus obscurus E. Geoffeot. 



There still exists some confusion regarding the identity of Geoffroy's 31. obscurus 

 and its relationship with other closely allied forms. The original description reads 

 as follows: "Pelage brun-noiratre en dessus, obscur en dessous; les poils etant 

 blancs a leur origine. . . . Le poil n'est brun qu'a sa pointe ; en quoi elle differe 

 principalement des deux precedentes [31. rufus, 31. ater] qui out leurs poils d'une 

 seule couleur. Longueur du corps, m , 060 (2 p. 2 lig.) ; — de la queue, m , 030 

 (1 p. 1 lig.) ; — de la membrane interf. m , 023 (8 lig.)-" This was one of four 

 species described by Geoffroy from actual specimens in the Paris museum, while 

 the five other species named in the same paper, are based wholly on Azara's de- 

 scriptions of Paraguayan bats. Geoffroy states that the specimens from winch 

 he describes his four first species are from " l'Amerique du nord, de Surinam, et 

 principalement de Ca'ienne; " he further adds, under 3Iolossus obscurus: " Je rap- 

 porte a cette espece la petite chauve-souris obscure, ou la 9? chauve-souris de 

 M. d'Azzara." The type locality of 31. obscurus has currently been accepted as 

 Paraguay, where Azara's work was done. A reference to the latter's description, 

 however, makes it clear that his ninth bat (Essais sur l'Hist. Nat. des Quadr. de 

 Paraguay, 1801, vol. 2, p. 288), or "petite chauve-souris obscure" was a Nycti- 

 nomus, since " la levre superieure a des plis verticaux " and " les canines, les in- 

 cisives, et les molaires, sont comme dans la Chauve-souris huitieme," identified 

 by Thomas (Ann. Mag. Nat. Hist., 1901, ser. 7, vol. 8, p. 441) as Nyetinomus 

 laticaudatus. Of this species Azara says the upper incisors are " avec une sepa- 

 ration," whereas in Molossus they are so closely approximated medially that Azara 

 was misled into supposing that his tenth bat, Molossus crassicaudatus, had in the 

 upper jaw "deux canines avec une seule incisive au milieu." Temminck, who 

 examined Geoffroy's type specimen, recognized that it was not Azara's ninth bat, 

 and added that " M. Desmarest partage mon opinion" (Mouogr. de Mammalogie, 

 1827, vol. 1, p. 236 and footnote) ; his description of 31. obscurus was based on 

 individuals from Surinam, and he mentions that he had seen others from Brazil. 

 The type locality of 31. obscurus is therefore probably Surinam (one of the local- 

 ities mentioned by Geoffroy), not Paraguay. The Museum has a series of alcoholic 

 specimens from various localities in northeastern Brazil that are thus practically 

 topotypes, and these agree in measurements with a skin and skull from Goya, 

 Argentina, kindly loaned by the U. S. National Museum. In this skin the hair 

 above is a broccoli brown, paler at the base ; and below, a lighter tint of the same. 

 In alcoholic specimens the bases of the hairs are not so conspicuously light as in 

 31. crassicaudatus. West Indian representatives of this species from Dominica 

 and Sta. Lucia are a very little smaller than those from Brazil, but the material 



