408 



BULLETIN OF THE BUREAU OF FISHERIES 



part around the cape from the eastward. But if 0. arctica occurs in the gulf chiefly 

 as an immigrant from the north, as seems probable at present, its quantitative dis- 

 tribution within the gulf in early spring does not parallel the distribution of tempera- 

 ture, for at the time of the winter minimum the water is coldest next the western 

 side of the gulf whde arctica is most abundant in the eastern side. 



The actual proportions in which the several species of Ceratium occurred during 

 the early spring of 1920 appears from the following list of actual counts of samples 

 at representative localities: 



Relative numbers of species of Ceratium in samples, "Albatross" cruise, March 1 to 19, 1920 



With the advance of the season and hand in hand with the augmentation of 

 diatoms, peridinians of all species so diminish in numbers that in 1920 they had 

 practically disappeared from the two productive centers for diatoms in the two 

 sides of the gulf by mid-April and were so scarce elsewhere that counts of the relative 

 numbers of the several species of Ceratium are no longer significant. But 

 when they reappear in the Massachusetts Bay region late in April or early in May in 

 the western side of the gulf, and in the Nova Scotian waters in the eastern, following 

 the eclipse of the diatom flowerings, a complete reversal has taken place in the 

 relative importance of the two leading species, for we have found longipes far more 

 numerous than tripos during the first week in May at every station where the genus 

 as a whole was sufficiently abundant for counts to be of value, only excepting the 

 southwestern edge of Georges Bank, where the two species were about equally 

 numerous (station 20129, May 18, 1920). In fact, C. tripos is then practically non- 

 existent within the gulf, or at best represented by occasional examples oidy. A 

 slight recrudescence of C. arctica (or perhaps a fresh wave of immigration) apparently 

 takes place during the first half of May, when occasional examples have been detected 

 at most of our stations (except among the diatom swarms) ; and on the seventh of 

 that month in 1915 C. arctica proved to be as abundant on German Bank (station 

 10271) as C. longipes, its area of abundance coinciding with the location of the cold 



