448 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 9» 



(c); maxillary width across first molars, 7.8; width across cingula of 

 canines, 5.1; maxillary tooth row, 6.1; length of mandible to front of 

 incisors, 12.7. 



Eemarks. — Andersen's classification (Proc. Zool. Soc. London, 

 1908, pp. 285-315) of the pygmy forms of Artibeus is based mainly 

 on the presence or absence of a rudimentary third lower molar and, in 

 the first upper molar, the degree of development of the posterointernal 

 lobe, the hypocone ("cusp 7" of Andersen). The diagnostic value of 

 the presence or absence of the third lower molar may be ruled out as 

 in the case of the large species of Artibeus (see foregoing "Remarks" 

 under headings A. lituratus palmarum and A. jamaicensis) . In his 

 treatment of the pygmy "species" normally lacking the third lower 

 molar, Andersen indicated the presence of this structure in one jaw, 

 absent in the other, of individuals of rosenbergi and toltecus. 



The form of M* in the pygmy Artibeus is similar to that of the large 

 Artibeus of the jamaicensis-lituratus group. Sufficient specimens of 

 the latter group were available to Andersen to demonstrate the variable 

 structure of M^ On the other hand, the few small localized series of 

 pygmy Artibeus examined by him caused Andersen to attach undue 

 significance to the structure of the same tooth in the pygmy species. 

 A comparison of the molars of 8 skulls of toltecus from Mexico and of 

 10 skulls identified as A. toltecus ravus from Ecuador with those of 

 10 skulls of A. cinereus from Venezuela shows complete intergradation 

 in the size and structm-e of the first upper molar. In one specimen 

 from Pambilar, northwestern Ecuador, the hypocone of this tooth is 

 as little developed as that described for the type of rosenbergi from 

 Cachavi, a locality in the same general area. The hypocones of M^ 

 in the remaining specimens from Ecuador (Pambilar and Carondolet) 

 show increasing development to the condition described by Andersen 

 as distinctive of toltecus. Similarly, the appearance of the hypocone 

 of M^ in Venezuelan series of cinereus ranges from the condition typical 

 of toltecus to the greatly reduced one of rosenbergi. The specimen 

 (U. S. N. M. No. 62635) from La Guaira, Venezuela, the only one other 

 than the type identified by Andersen as rosenbergi, is simply one of 

 the same series of bats collected by Robinson in northern Venezuela, 

 which were assigned to A. cinereus bogotensis by Andersen. Addi- 

 tional material in the collection of the Chicago Natural History Mu- 

 seum confirms the specific identity of cinereus, toltecus, and rosenbergi. 

 Of four specimens from San Jose, Manavi Provmce, Ecuador, not far 

 from the type localities of rosenbergi and ravus in Esmeraldas Province, 

 one shows "cusp 7" of M^ as described for ravus, another as described 

 for rosenbergi, the first upper molars of the remaining two specimens 

 being intermediate in structure. 



