3G 



BULLETIN 3(1, UNITED HTATES NATIONAL MUSEUM. 



the skull. The proportional length of the beak iu the ditierent speci 

 mens, arranged in an ascending scale, is as follows : 



It ai)pears that, excepting in the skull with relatively longest beak, 

 the proportion of the length of the beak rises by gradations of four- 

 teuths of 1 per cent, and less. 



The proportion of the width of the beak at its base to its length is as 

 follows : 



The gradations here are 1.7 percent., 1.1 per cent., and less. 

 The proportions of the width of the beak at its middle compared with 

 its length rise by gradations of 1 per cent, and less, as follows: 



In all three cases the greatest variation is at the extremes of the 

 series. 



What do these proportions show ? Apparently that the relative 

 length and width of the beak give no indication of the sex. Unless 

 these twenty-one specimens are all of the same sex, which is very im- 

 probable, for the reasons stated, the gradation of proportions is such 

 that ib would be impossible to divide the females from the males. 



The skulls of greatest absolute length have not relatively longest 

 beaks. The beaks which are relatively longest, as compared with the 

 absolute total length of the skull, are, generally speaking, narrowest at 

 the base and middle in proportion to their absolute length. 



The length of the mandible as compared with the length of the skull, 

 minus the beak, is greater iu all of Dr. Fischer's males than in his fe- 

 males. In both of my young males (20901 and 10504), on the contrary, 

 it is shorter than in the young female, 20962 ; and in one of the former 

 (20901) it is shorter than in the old female, 2230-1. 



