954 
BULLETIN OF THE BUREAU OF FISHERIES 
high in surface density throughout the summer, because of the strong tidal currents 
that constantly mix the surface stratum, as it warms, with colder and more saline 
water from below (p. 928), and because the indraft of slope water of high salinity is 
directed into this side of the gulf. Consequently, the regional variation in the density 
of the upper 40 meters is wider in summer than at any other season, with the 
fundamental west-east gradation reappearing from year to year in essentially the 
same spacial relationship. 
In April, and especially in May, the reader will recall, simple projection of the 
density contours at the surface mirrors the general dynamic tendency for the whole 
body of water in the gulf, regional distribution being essentially similar downward 
through the whole column. This, however, is not the case in summer, because the 
Fig. 201. — Distribution of density at the surface, for the inner part of the gulf, August, 1913 
surface pool of low density in the offing of Massachusetts Bay is a superficial 
phenomenon. In fact, the surface contour lines run almost at right angles to those 
at 100 meters (fig. 202), which more nearly preserve the character of the preceding 
months. The actual surface drift in this side of the gulf is therefore the component 
of a rather complex screwing motion. In the northeastern part of the gulf, however, 
the surface state more nearly mirrors the regional distribution of density for the 
whole column. 
Unfortunately no one of our summer cruises has afforded the data needed for a 
satisfactory mapping of density for the whole area. In the only summer (1914) 
when the southeastern part of the area was surveyed, the coastal belt (more impor- 
tant dynamically) was neglected. In every case, too, allowance must be made for 
