PLANKTON OF THE GULF OF MAINE 249 



on the second occasion with 100 specimens in a total of only 400 copepods of all 

 kinds. Willey (1919) also records it much more often from vertical than from 

 surface hauls in the Gulf of St. Lawrence and off Nova Scotia. 



I can offer no data on its presence or absence at the surface in the Gulf of 

 Maine during the summer months; but Willey's (1919) tables, which show that a 

 larger proportion of the records of it obtained by the Canadian Arctic expedition 

 were from the surface in May and June than in July and August, suggest that 

 it tends to sink down into cooler strata as the seasonal warming of the top of 

 the water progresses. 



The vertical distribution of this species in other seas makes it probable that it 

 ranges right down to the deepest levels in the Gulf of Maine, but the data are not 

 sufficient to show whether it tends to gather at any particular level or is more evenly 

 and indifferently distributed vertically. 



When the locality records for M. longa are plotted (fig. 75) it is evident that 

 in the years when it is most plentiful in the gulf it becomes generally distributed 

 over the entire area of the latter, indifferently in the peripheral zone, in the central 

 basin, and over the offshore banks as far west as Marthas Vineyard. It should be 

 noted that the absence of summer and autumn records on Georges and Brown's 

 Banks, and in the southeastern part of the gulf generally, is actually not a contra- 

 diction, because there were no, or at least very few, M. longa in the gulf during 

 1914, the year when we made our chief midsummer cruise in this region. The 

 apparent predominance of records in the western side of the gulf is equally deceptive, 

 due simply to the fact that we have worked more there than elsewhere. 



Immigration and breeding. — The periodic appearances and disappearances of 

 M. longa in the Gulf of Maine, coupled with its Arctic nature in general, identify 

 it as primarily an immigrant to the gulf from the north, depending on frequent 

 accessions from more prolific centers to maintain the local stock. But the fact that, 

 unlike most of the immigrant species, it is not localized in the eastern side and around 

 the peripheral belt of the gulf is evidence either that the visiting specimens come 

 in such abundance and live so long that they spread universally over the entire 

 extent of the latter before they perish, or that they succeed in breeding within the 

 gulf to an extent sufficient for the dispersal of the resulting generations to hide 

 the routes of entrance of their parents. In this connection it is instructive to find 

 the distribution of M. longa paralleling the spring status of Calanus hyperboreus, 

 a species similarly of northern affinities but for which a certain amount of local 

 reproduction within the gulf seems sufficiently demonstrated. 



The locations of the stations (fig. 76) where more M. longa have been taken 

 than the average numbers per square meter for their respective months (in which 

 respect M. longa closely parallels Calanus hyperboreus) are further evidence of this. 

 In spring and early summer (the season when the influx of northern water is at its 

 height, and when consequently the greatest invasions of M. longa are to be expected) 

 two distinct lines of immigration are suggested by the rich catches — the one inward 

 into the eastern side of the gulf via the northern and eastern channels, and the other 

 westward along the continental edge of Georges Bank. The rich spring catches made 

 in the western side of the gulf in 1920 might have been the result either of local 



