b. In general the circulation in the Gulf of Maine comprises 
a counterclockwise eddy. The drift along the Maine-Massachusetts 
coast is southerly, on the order of 2-6 miles per day (4-13 cm/sec). 
The southerly part of the circulation, in Massachusetts Bay, is com- 
prised of two drifts, one counterclockwise around Cape Cod Bay, and 
one across the mouth of Massachusetts Bay toward the outer coast of 
Cape Cod and thence southerly. The drift from Georges Bank is gener- 
ally west during the spring and summer but more offshore and perhaps 
even easterly during the autumn and winter. 
c. South of Massachusetts, Rhode Island, and New York the 
coastal drift tends to be westerly, 3-5 miles per day, (6-11 cm/sec) and 
from the offing of New Jersey southward to Cape Hatteras the set is 
southerly with speeds varying from 3-15 miles per day (6-32 cm/sec). 
d. From Cape Hatteras to Georgia the surface non-tidal 
drift tends northeasterly at speeds of 0.2 to 12 miles per day, (.4 to 26 
cm/sec) with the highest concentrations of drift bottle strandings on the 
southern side of the Capes. From Georgia to Daytona Beach the set 
appears to be southerly and from Daytona Beach south the drift appears 
to be northerly (unpublished data). 
e. Redfield and Walford (11) noted that ''. . wastes likely to 
be transported to beaches in the surface layers should be carried at 
least 10 miles to sea if contamination of beaches is to be avoided.'' The 
percentage frequency of strandings of drift bottles from areas off the 
U.S. coast, figure 2, indicates the variation in what may be construed as 
onshore or offshore areas of drift. The contours extend farthest off- 
shore in the Gulf of Maine and south of Nantucket. They bend in towards 
the mouths of the Hudson River, Delaware and Chesapeake Bays. On 
the other hand, on either side of the mouths of these estuaries the fre- 
quency of returns is high. Note also the high percentage of returns from 
bottles dropped from Georgia southward. 
Diffusion Processes 
In addition to the gross transportation of contaminants by perma- 
nent and semi-permanent ocean currents, the mechanism by whicha 
contaminant will move from the bottom if current velocities are zero 
at the water-bottom interface is obviously important. This mechanism 
is diffusion controlled. Similarly, once the contaminant is contained in 
water its dispersion and subsequent dilution will also be controlled by 
turbulent diffusion processes. So far as can be discovered, there are no 
direct measurements of diffusion-controlled processes within the areas 
of interest that can be used to evaluate the relative importance of these 
mechanisms to the movement of the contaminants. It is possible, how- 
ever, to treat the problem theoretically, by imposing a number of sim- 
plifying assumptions which give a result that is not an accurate descrip- 
tion of the diffusion process, but does give limiting values of the 
concentration of contaminant in the sea. Reid (12) has examined several 
cases in which the following assumptions were made: 
18 
