1918] Michael: Behavior of Salpa democratica 289 



between 6 a.m. and 10 a.m. are considered, when only hauls 

 made between 10 a.m. and 2 p.m. are considered, when only 

 hauls made between 6 p.m. and 10 p.m. are considered, when 

 all hauls are excluded except those made during June, 1909, 

 and when all are excluded except those made between June 15 

 and June 29, 1909, and within one mile of each other. 



2. When the data are tabulated with respect to three, four, five, 

 and six temperature groups each having an equal range between 

 15?9C at one extreme to 20?8C at the other, they show that: 



a. Solitary forms increase in abundance as the temperature 

 increases from its lowest, or next lowest, value to its highest 

 value. 



b. The frequency of both solitary forms and aggregate 

 forms decreases as the temperature increases from its lowest to 

 its middle values, and then increases as the temperature 

 increases. 



c. Both solitary forms and aggregate forms are more fre- 

 quent in the coldest than in the warmest water, more frequent 

 in the next coldest than in the next warmest water, more fre- 

 quent in the third coldest than in the third warmest water, and 

 so on. 



d. The abundance of aggregate forms parallels the fre- 

 quency of both generations. 



3. When each of the June and July surface hauls is examined 

 on its own merits, it is found that : 



a. Forty-six of the forty-nine hauls that captured solitary 

 forms also captured some aggregate forms. 



b. Twenty-four of the twenty-seven hauls that failed to 

 capture solitary forms also failed to capture aggregate forms. 



From these facts it is concluded that : 



1. Differentials in distribution of solitary forms and aggregate 

 forms are as definite and pronounced as are the differentials in their 

 structure. 



2. Seasonal succession of the two generations can not explain 

 the observed differentials. 



3. Greater production of solitary forms in stratified areas of warm 

 surface water than in those of cold surface water, if occurring at all, 

 can not explain the observed differentials. 



