PLANKTON OF THE GULF OF MAINE 261 



in spring than in summer, and its excursions upward to the top of the water are not 

 so closely confined to the hours of darkness in spring as they are during July and 

 August. 



The vertical hauls shoaler than 100 meters yield further evidence of a diurnal 

 migration of M. lucens, for the catches have averaged decidedly larger between 

 6 p. m. and 8 a. m. (average 4,246 per square meter for 26 stations) than between 

 8 a. m. and 6 p. m. (average 896 per square meter for 21 stations); and if further 

 separated into two groups by months — February to May and September to October — 

 the same holds good, as follows: 



6 p. m. to 8 a. m. to 

 " a. m. p. m. 



Average number per square meter, February to May 



Average number per square meter, September to October. 



1.601 

 7,854 



287 

 4,553 



To compensate for this, smaller averages might be expected by night and larger 

 by day in the deeper hauls as the Metridia swim up and sink back. Interpretation of 

 these and comparison of the deeper hauls with the shoaler is complicated by the fact 

 that we have one unusually rich catch of almost 23,000 per square meter in a vertical 

 haul from 200 to meters (station 10304, August 6, 1915) by night, but it is obvious 

 that if the specimens in question were concentrated near the surface, as is perfectly 

 possible, a shoal haul would have caught nearly or quite as many. This applies 

 to any individual haul, but when deep hauls consistently average more productive 

 than shoal, with a greater difference than can be accounted for by the longer column 

 of water fished through, it is safe to say that the animals are concentrated in the 

 lower levels. 



The greater the number of hauls, the greater the dependence which can be 

 placed on the average results. In the present case the number of hauls is not large 

 enough to warrant definite conclusions. If the one very rich deep haul just men- 

 tioned be omitted, we have 1,190 as the average number per square meter in vertical 

 hauls from deeper than 200 meters from 8 a. m. to 6 p. m. and 1,200 from 6 p. m. to 

 8 a. m. This does not suggest anj' diurnal migration as deep as 200 meters. 



It is obvious that the contour of the bottom of the gulf largely determines the 

 depth range of this copepod or of any other animal, for such of the stock as inhabit 

 the coastal zone are necessarily confined to a very shoal stratum. No copepod can 

 sink as deep in the Gulf of Maine, where the greatest depth is only about 330 meters, 

 as it can off San Diego. Apart from this limitation by topography, however, the 

 level of plurimum abundance of this species is about the same in the gulf as in the 

 eastern North Atlantic — namely 50 to 150 meters. Thus all but one " of the verti- 

 cal hauls which have yielded 5,000 or more per square meter have been from depths 

 of 200 meters or less, more than half of them shoaler than 100 meters, irrespective 

 of the time of day or part of the gulf in which the stations were located. The depths 

 of the five richest catches of all (those yielding M. lucens at the rate of more than 

 15,000 per square meter) have likewise varied from shallow to deep. 



•I The exception is station 20087, Mar. 24, 1U20, from 250 meters. 



