1918] Michael: Behavior of Salpa (I emocratica 251 



ences: (1) normal sequence in the life cycle; (2) the effect of higher 

 temperatures during July in increasing the abundance of solitary 

 forms; and (3) a similar effect of higher salinities during July. To 

 what extent and how each of these factors has operated in causing 

 the difference noted it is impossible to say, but it seems probable that 

 temperature has played an important part.- This is not only made 

 evident in the ensuing discussion but also by the fact that twenty-one 

 surface hauls made between August 9 and 23, 1911, in the vicinity 

 of San Diego when the temperature averaged 20?2C failed to catch a 

 single specimen of either generation, wiiile on August 21, 1911, three 

 surface hauls made with the same net in the vicinity of Santa Rosa 

 Island when the temperature averaged 16?4C were all successful, 

 obtaining 115 solitary forms and 1409 aggregate forms. 



In apparent contradiction to this fact, the table shows that aggre- 

 gate forms occurred to some extent during August although solitary 

 forms did not and judging from the high temperature average, this 

 is the reverse of what might have been expected. These facts indicate 

 the complexity of causes leading to the seasonal appearance and 

 disappearance of S. democratica. 



When all the hauls in all depths are examined, June and July 

 still stand out as the months of maximum abundance, the number of 

 specimens obtained during other months being relatively few. 



4. Brief Discussion of Vertical Distribution 



Considering horizontal closing net hauls from all depths, made 

 during the months of June and July, 1908 and 1909, a total of 6,889 

 solitary forms and 17,091 aggregate forms were obtained. Table 3 

 shows liow the}^ were vertically distributed. 



This table shows that solitary forms decrease in abundance as 

 the depth increases, disappearing entirely below 100 fathoms, while 



