414 



BULLETIN OP THE BUREAU OF FISHERIES 



The hauls listed above are further interesting as showing that C. arcticum, so 

 widely distributed in spring (p. 407) but not detected in the gulf in late summer or 

 autumn, reappears there in small numbers in midwinter, but curiously enough along 

 its northern and western shores and not in the eastern side. 



There is no reason to suppose that any notable alteration takes place in the 

 relative numbers of the several species of Ceratium during the months of January and 

 February; certainly not off Gloucester during the winter of 1913, where C. tripos 



Number April May June July Aug. Sept. Oct. Nov. Dec. 



ever [Q 20 30 1 ZQ 30 10 20 30 10 20 30 10 20 30 10 20 30 10 Z0 30 10 20 30 10 20 30 

 Specimens 



50 



48 



46 



44 



42 



40 



38 



36 



34 



SZ 



30 



28 



26 



24 



22 



20 



18 



16 



14 



12 



10 



8 



6 



4 



2 







Fig. 111.— Proportionate numbers ol C. longipes (solid curve) and C. tripos (broken curve) in samples of 100 Ceratium of 

 all species in the Massachusetts Bay region at different seasons, 1913 to 1922 



continued predominant until the diatom flowerings commenced in March. Doctor 

 McMurrich's plankton lists also show this to have been the case at St. Andrews during 

 1916. 



The mutual fluctuations of C. tripos and C. longipes in the northern part of 

 Massachusetts Bay are represented in the accompanying diagram (fig. Ill) based on 

 the combined data for the years 1912, 1913, 1915, 1916, and 1920, which will serve 

 equally for the offshore parts of the gulf if the reversal of dominance be imagined as 



