FISHERY BULLETIN: VOL. 72. NO. 2 



that slopes gently seaward to about the 100- 

 fathom (fm) isobath, the point sometimes desig- 

 nated as the shoreward limit of the continental 

 slope. The slope itself declines much more steeply 

 to the abyssal plain below. The distance between 

 the 100- and 1,000-fm isobaths is 17 miles off 

 New York City, 16 miles off Martha's Vineyard, 

 and no more than 20 miles along the southern 

 edge of Georges Bank. The width of the conti- 

 nental shelf "out to" the 100-fm isobath is 104 

 miles off New York City and 76 miles off Martha's 

 Vineyard. Most of the southern New England 

 subarea is contained within the 60-fm isobath. 

 The Gulf of Maine is a basin of irregular relief 

 within the continental shelf, surrounded by a 

 shallow sill formed by Georges Bank on the south- 

 east, Browns Bank on the east, and the Seal 

 Island Banks on the northeast. The sill in turn is 

 pierced by three narrow channels, which have 

 been named according to their locations: the 

 Great South Channel, 40 fm maximum depth; 

 the Eastern Channel, 128 fm maximum depth; 

 and the Northern Channel, 65 fm maximum 

 depth. Most of the Gulf of Maine is deeper than 

 60 fm, and the central Gulf, occupied by a Y- 

 shaped trough, is deeper than 100 fm. The topog- 

 raphy of the Gulf of Maine has been discussed 

 in detail by Bigelow (1927). 



Temperature 



The following description of seasonal thermal 

 regimes is compiled from Bigelow (1927, 1933) 

 and original data collected during the Alba- 

 tross IV surveys. The southern New England 

 subarea has extreme seasonal temperature 

 changes related to its shallowness. Bottom tem- 

 peratures reach an annual minimum in late 

 February or early March, and are lowest in the 

 shoal waters close to shore and on Georges Bank 

 and highest at the edge of the continental shelf 

 (Figure 2). Spring warming proceeds most rapidly 

 in the shallowest water. By early summer, 

 thermal stratification occurs and prevents the 

 bottom water from warming at intermediate 

 depths (20 to 60 fm) which remain at 4° to 8°C 

 (cooler than both the shoaler depths inshore and 

 the deeper slope water offshore) (Figure 3). 



Bottom temperatures in shallow areas attain 

 the annual maximum by the beginning of fall. 

 However, in intermediate depths the maximum 

 is not reached until the thermocline is broken 

 down, usually during October or November 



AVERAGE WINTER BOTTOM TEMPERATURES CO 

 ALBATROSS IV GROUNDFISH SURVEYS 64-1.65-284.66-1 





Figure 2. — Average winter bottom temperatures compiled 

 from RV Albatross IV groundfish surveys 64-1, 65-2, 65-4, 

 and 66-1. 



(Figure 4). Winter cooling proceeds more rapidly 

 in shoal water than at intermediate depths. 

 Depths beyond 60 fm under the influence of slope 

 water have relatively little fluctuation in bottom 

 temperatures throughout the year. 



In the Gulf of Maine, seasonal thermal changes 

 on the bottom are most pronounced in the shoaler 

 areas. Also, the banks on the Nova Scotian shelf 

 are generally cooler by a few degrees than those 

 to the west, because of the influence of the cold 

 coastal current. Bottom temperatures over most of 

 the Gulf are more stable than those off" southern 

 New England, because the deeper Gulf water is 

 made up in part of warm moderating slope water 

 which enters by way of the Eastern Channel. 



Procedures of Analysis 



Life History Stage by Sampling Strata 



An analysis was performed to determine geo- 

 graphical and bathymetric distribution of U. 

 chiiss and U. tenuis by life history stage and 

 season. Taylor (1953), Moyle and Lound (1960), 

 and Roessler (1965) demonstrated that fish are not 

 randomly distributed, but that the sampling 

 distribution of the number of individuals of a 

 species taken per sample in a series of col- 

 lections is contagious and may be best approxi- 

 mated by the negative binomial distribution. The 

 natural log transformation y = In (x + 1), where 

 jc represents the number of individuals of a species 

 taken at each station, tends to "normalize" the 

 negative binomial distribution and substantially 



484 



