264 F. M. JEWELL. 



Therefore, the results in the 20-30 group are very peculiar and 

 can be interpreted only as due to chance, even though the number 

 within that group is 152 individuals. It would be impossible 

 for the males to reach again such a preponderance as 127 to 100 

 females after such a differential elimination of the sexes as would 

 appear to be indicated in the 20-30 group, if there were such a 

 high mortality of males in that group. 



TABLE II. 



SHOWING SEX RATIOS ACCORDING TO LENGTH. 

 Length. Sex Ratio. Number of Individuals. 



o- 10 275.0 15 



io- 20 136.5 97 



20- 30 102.6 152 



30-40 . . .137-7 145 



40- 50 113-1 130 



50- 60 126.9 1 18 



60- 70 133.3 140 



70- 80 104.6 133 



80- 90 132.1 65 



90 100 400.0 5 



If the sex ratios are computed for individuals up to a certain 

 length on the one hand and then for all over that length, a 

 comparison can be made of the ratios for relatively younger and 

 older foetuses. In this way if there is any difference in the viability 

 of the sexes we should expect to find that in the younger foetuses 

 the sex ratio would be high and that there might perhaps be a 

 critical age, as indicated by length, in which there would be a 

 greater mortality for one sex than for the other. We should 

 also expect to find that after this critical stage in development 

 is past the sex ratio in the remaining groups would not vary so 

 greatly from the total of 123 as it did up to that group. 



Table III. gives such a grouping with the number of individuals 

 and the sex ratio for each group. The advantage of greater 

 numbers within the group is also gained in this way. The sex 

 ratios in this table were computed from the complete tabular 

 list. 



Thus, it is noted that for 112 individuals up to 20 cm. the sex 

 ratio is 148.88 and that for the remaining 888 individuals from 



