1918] 



MicJtael : BcJiavior of Salpa democratica 



263 



according to this hypothesis, have to be restricted to that time interval 

 between, the death of many of the embryo-bearing salpae and pro- 

 trusion from the mantle cavity of the solitary forms of the first salpae 

 of the next generation — a time interval which is brief if it occurs at 

 all. If this were not the case, either no significant temperature rela- 

 tions would have appeared in the data, or aggregate forms as well as 

 solitary forms would have been most abundant in the warmer water, 

 both of which are contrary to fact. Moreover, the hypothesis requires 

 the death of a large number of aggregate forms prior to their maturity, 

 for, as stated on page 245, the testis does not develop until after 

 the embryo has matured and made its escape. Does this seem reason- 

 able ? 



But to speculate gains naught. The hypothesis is therefore sub- 

 jected to an empirical test. Table 7 gives the distribution of hauls 

 involved in tables 4, 5, and 6 with respect to each of the four months 

 concerned : June and July, 1908, and June and July, 1909. 



Temperature 



in 



centigrade 



15? 9-18? 3 

 ]8?4-20?8 



Table 7 



Distribution of hauls with respect to season 



June, 1908 June, 1909 July, 1908 



July, 1909 



Hauls Mean date Hauls Mean date Hauls Mean date Hauls 



Section A — Hauls involved in table 4 



11 18 14 2,3 2 19 3 



5 20 13 22 9 21 19 



Mean 

 date 



Section B — Hauls involved in table 5 

 lo?9-18?3 .... 9 23 



18?4-20?8 .... 9 23 8 22 



Section C — Hauls involved in table 6A 

 15?9-18?3 11 18 5 22 2 19 



18?4-20?8 5 20 4 23 1 8 



Section D — Hauls involved in table 6B 

 15?9-18?3 7 18 1 15 



18?4-20?8 3 19 .... 



15?9-18?3 

 18?4-20?8 



Section E — Hauls involved in table 6C 

 17 2 23 2 18 



27 1 29 1 8 



Section F — Hauls involved in table 6D 

 15?9-18?3 .... 6 24 .... 



18?4-20?8 .... 8 21 4 21 6 



