340 University of California Publications in Zoology [Vol.22 



characteristics are somewhat variable, especially as to general appear- 

 ance of the chain and of the setae. It was probably confused some- 

 times with Ch. constriction Gran and some other forms. Its maximum 

 was reached on September 16 at 8 a.m. in a catch numbering 14,800, 

 the average for the month being 1674. It is not recorded after the 

 middle of October. 



Chaetoceras sp. is a catchall for a large number of species difficult 

 to identify, or rare. In the four months under consideration its main 

 components were small forms with poorly marked characters, of which 

 Ch. a/fine Laud, was probably most prominent. Numerically and 

 probably volumetrically this assemblage was distinctly the most 

 important of the diatoms in September, October, and December. The 

 maximum, 339,376, was recorded for September 24, 8 p.m., and there 

 was an average for that month of !)0,069 per liter. High production 

 of this group began about September 11 and lasted almost a month. 

 Its numbers in the catches then became very small until the latter 

 part of November. While not very abundant, considerable numbers 

 were found through most of December. 



Dactyliosolen was represented apparently by only one species, 

 D. tenuis CI., and identification in many cases was not very certain. 

 It was fairly well represented in October and November, but very 

 slightly in the other two months. Maximum, 5700 per liter, October 

 21, 8 a.m., average in October, 1464. 



Leptocylindrus also showed only one species, Leptocylindrus 

 danicus Schiitt. This form attained considerable prominence in late 

 September and early October, but was almost or quite absent at other 

 times. Maximum production, 17,600, was reached on October 6 at 

 8 a.m. and the average in October was 2187. This form does not 

 appear to be of great importance in our region. 



The genus Nitzschia was probably represented by several species 

 but only two attained much numerical importance. N. seriata CI. and 

 N. puncjens Grun. Their distribution as recorded was essentially 

 similar and there were very many cases in which it was uncertain 

 whether the specimen was rightly called N. puncjens or whether it was 

 merely a small specimen of A T . seriata. Such cases, together with the 

 similarity in distribution, lead one to think that perhaps the two forms 

 may both be properly included under N. seriata. There is, however, 

 some advantage in keeping the records separate, since a marked and 

 fairly consistent deviation within the species may help to show the 

 influence of certain environmental factors, such as temperature in 

 relation to growth, shape, etc. 



