138 University of California Publications in Zoology. [Vol.8 



first salinity group is the lowest, it is certain that the abundance 

 of 105 is too large for comparison with the 54 of the second, and 

 the 16 of the third salinity group. Again, the hauls were 

 unevenly distributed with respect to light. Corresponding to the 

 first group, four hauls were made between 6 and 8 a.m.. the time 

 of morning maximum abundance, two between 4 and 6 p.m., and 

 six between 8 a.m. and 4 p.m. Corresponding to the second 

 group, five hauls were made between 8 a.m. and 4 p.m., one 

 between 4 and 6 p.m., and none between 6 and 8 a.m. Concern- 

 ing the third group, five hauls were made between 6 and 8 a.m.. 

 two between 4 and 6 p.m., and four between 8 a.m. and 4 p.m. 



From these facts it follows that the hauls included in the 

 above table were so distributed that both effect of light and 

 temperature were antagonistic to that of salinity. Therefore, if 

 the data were corrected with respect to these factors, the effect 

 of variations in salinity ought to be more apparent. By includ- 

 ing only data relative to hauls made between 8 a.m. and 4 p.m. 

 the antagonistic effect of light may be mostly eliminated. The 

 results of such elimination are shown in the following table. 



TABLE 50 



Effect of Salinity on the Surface Distribution of <S'. bipunctata based on 

 Hauls between 8 a.m. and 4 p.m. 



Because only four of the entire fifteen hauls were made when 

 the temperature of the water was between 17?6 and 19?5, it is 

 impossible to standardize the data with respect to temperature. 

 However, from the average temperatures recorded in the ' table, 

 we can be sure that the abundance of 16 corresponding to the 

 first salinity group is larger, and that of 8 corresponding to the 

 third salinity group smaller than it would be if the effect of 

 temperature could be eliminated. In spite of the antagonistic 

 effects of temperature the table shows that, when disturbing 

 light factors are eliminated, the maximum abundance and fre- 



