276 DR. KNUD ANDERSEN ON BATS. [AjJr. 7, 



lower, less convex, and more spreading brain-case, broader palate 

 and heavier dentition, the upper tooth-row (canine and molar- 

 premolar series) having a length of 11-5 mm. against 10-2 in 

 A . jxchnarum ; also in more prominent face stripes and darkei' 

 coloration." Forearm 71 mm. ; third metacarpal 69 mm. 



I have had no adnlt specimen of the Artibeus jamaicensis 

 section from Bolivia*, but nine from Ecuador, S. and Central 

 Colombia ; they accord in every respect with Allen's description 

 of A. rusb't/i, being at the same time perfectly indistinguishable 

 from A. j. lituratus. Allen gives as length of the maxillaiy tooth- 

 row in his single specimen of A. rusbyi 11-5 mm., as against 10'2 

 in " A. 'palmarum" ; in so far he had good reason to separate the 

 former. But in 6 skulls of adult individuals of A . i . lituratus 

 from the type locality, viz. Paraguay, the average length of the 

 maxillary tooth-row is 11 -6 mm. ; in 19 adult skulls of the same 

 race from various localities in Brazil 11-7 mm. ; thus precisely as 

 in the type of A. rusbyi. — The explanation, why Allen regarded 

 A. rusbyi as an undescribed species, is evidently this: Allen 

 compared A. ribsbyi not with a series of A. j. lituratus from 

 Paraguay or Brazil, but with an xmusually small-toothed A. j. 

 palmaru7n. I have examined 28 skulls of A. j. palmar um, and 

 10-2 mm., given by Allen as the length of the upper tooth-row 

 in this form, is precisely the actual minimum in the whole series. 

 Allen found the skull of A. rusbyi "lower" (less convex, &c.) 

 than in palmarum ; it only mea.ns that the skull of the type 

 of A. rusbyi is in the adolescent stage, whereas the skull (or 

 skulls) of p)almarihm with which he compai'ed it must have been 

 in the stage characteristic of adult and aged individuals ; that 

 these differences in the shape of the skull ai'e dependent on 

 age has been pointed out elsewhere in this paper (p. 248, text- 

 figs. 50, 51). The colour characters mentioned by Allen are 

 of no taxonomic value ; individuals of lituratus and palmarum 

 may be dark or light, their facial stripes strong or obsolete ; 

 these variations are pai-tly dependent on age, partly pui'ely 

 individual. 



In the table of measurements below, p. 277, I have arranged 

 the adult specimens of A. j. lituratus in three sections according 

 to their geographical habitat, viz. Paraguay, Brazil, and Ecuador- 

 Central Colombia. In each of these sections are given the abso- 

 lute minimum and maximum and the average, chiefly in order to 

 show that individuals from these three regions are in every 

 respect identical in size, i. e. vary within the same limits. It 

 will be noticed that the average measurements of skulls (including 

 teeth) fi-om the three regions are perfectly alike ; that averages 

 of external measui-ements of individuals from Pai-aguay and 

 Brazil are alike; and that the actual measurements of skulls, 

 teeth, and external dimensions, within all these categories are 



* Specimen "*i" in Dobson's Catalogue (p. 520), stated to be from Bolivia and 

 "adult," is quite young; the correctness of the locality is questionable. 



