NO. 1186. VARIATIONS OF HYLA BEGILLA— TEST. 487 



very careful ineiisureinents were made of 304 specimens throuoliout tlic 

 entire range, the results of which are given below. The method pur- 

 sued was as follows : With a pair of sharp pointed dividers the length 

 from the tip of the snout to the end of the pelvis, the breadth of the 

 head at the iiosterior edges of the tympana, and the length of the tibiic, 

 were determined in millimeters and half millimeters. Where tlie two 

 tibiae were found to be of different lengths, and quite often the left one 

 was a very little shorter, the average was taken as nearly as i)ossible. 

 Then the proportion of the head to the body was found, and of the 

 tibia} to the body. To establish a constant standard, the length of the 

 body in all cases was fixed at 100, and the breadth of the head com- 

 puted in percentage of this. In most cases this proportion varied 

 from 32 to 30 jier cent. In like fashion in the second proportion, the 

 length of the body being still fixed at 100, the length of the tibia' was 

 calculated to suit it, the figures usually being found to range from 48 

 to 50 per cent. In order to insure as complete accuracy as possible, 

 in the great majority of cases each measurement was taken twice, and 

 the result obtained by the proportion was calculated to two decimal 

 places, thus indicating the more delicately any variation, and i)roving 

 of advantage when the averages were struck. The uniformity with 

 which, approximately, the same figures reapi)eared was a matter of 

 some little surprise at first, in the light of the proportions claimed by 

 Cope in distinguishing his three varieties, but the persistence removed 

 all doubts as to its correctness. As was to be ex])ected, some few speci- 

 mens gave extreme figures, which taken alone might indicate almost 

 a subspecilic difference, but the gaps are so completely filled in, usually 

 by specimens from the same locality even, that the discrepancies are 

 unmistakably due to individual variation and exaggeration, and have 

 no real significance. Since the proportion of the head and body is the 

 one used by Cope in separating his varieties, especial notice has been 

 paid to that, but it is found to differ no more than does that of the 

 body and tibia, which is very little. 



The narrowest heads observed give the proportions of 30.9G per cent 

 in the total length, or entering it not quite three and a quarter times. 

 But this is found in only 3 specimens, 2 from Puget Sound and 1 from 

 Fort Tejoii, California, and as the specimens are quite small, only 21 

 mm. in length, and since the head appears to be normally slightly nar- 

 rower in younger individuals, it is probable that this accounts for the 

 proportion. At all events, it seems clear that it is a case of individual 

 variation, and of no value as a varietal distinction. 



The broadest head belongs to a specimen from the Panamint Moun- 

 tains, Johnson Canyon, with the proportion of 39.48 per cent, but 

 another specimen from that locality has a head 34.07 per cent, and the 

 average of 34 specimens is 30.83 per cent. This seems an approach to 

 the variety laticeps, but laticepfi was described from Cape St. Lncas 

 specimens, and besides really does not exist, to Judge from the i)r()por- 

 tions of the type specimens. To them Cope ascribes a head breadth of 



