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



This table marks a distinction between frequency and abund- 

 ance of S. hipunctata at each depth. While it is most frequent 

 in 15-20 fathoms it appears most abundant on the surface. It is 

 fifth in frequency and sixth in abundance in 7-12 fathoms, 

 sixth in frequency and fourth in abundance in 40-75 fathoms, 

 and fourth in frequency and fifth in abundance in 25-35 fathoms. 

 However, in 4-6 and 100-150 fathoms the frequency and abund- 

 ance are parallel. 



"We should naturally expect complete parallelism between 

 frequency and abundance, but we find it incomplete. What, 

 then, does this lack of parallelism signify'? Is it due to an in- 

 sufficient number of hauls, to lack of correlation between hauls 

 made with surface and closing nets, to an uneven distribution 

 of hauls with respect to varying environmental conditions, or to 

 a combination of all these influences? 



It is obvious that accuracy of the percentage of successful 

 hauls varies with the total number made. Thus the chance of 

 error in the frequency of 100 based upon 13 hauls in 15-20 

 fathoms is much greater than that in the frequency of 70 based 

 on 136 surface hauls. Had 136 hauls been made in each depth 

 the frequencies noted in the table would undoubtedly be altered. 



On the other hand, it was previously stated that the surface 

 net was larger in orifice than the Kofoid closing net, and that 

 the former, therefore, filtered more water per unit length of 

 haul than the latter. Theoretically we estimated that the average 

 number of specimens obtained per hour from the surface ought 

 to be divided by 7 in order to correlate the number with that 

 obtained from each depth by the closing nets. If this were done, 

 however, the number would be reduced to 44, which is much too 

 small to represent the relative number of specimens obtained. 

 We can only be sure that the 307 specimens from the surface 

 should be smaller. Possibly it should be less than 195. 



Far more important than either an insufficient number of 

 hauls or lack of correlation between surface and closing net 

 hauls is the fact of their uneven distribution with respect to 

 varying environmental conditions. Variations in temperature, 

 salinity, currents, and light have all doubtless affected the 



