PHYSICAL OCEANOGRAPHY OF THE GULF OF MAINE 741 
a spread suggests that the bottom of the gulf had actually received much more 
water via the channel in 1920 than in 1919 during the whole winter. 
No cause can yet be assigned to annual differences of this sort, except that they 
do not result from local influences operative within the gulf, but from the state of 
the reservoir outside the edge of the continent, which supplies the indraft (p. 848). 
SALINITY IN MAY 
SURFACE 
The salinity of the gulf is especially interesting during the first half of May, 
because the two most important events in its vernal cycle — freshening of the surface 
by land water in the western side, and by the Nova Scotian current in the eastern 
side — culminate then. Unfortunately we have not been able to carry out a general 
oceanographic survey of the whole area of the gulf in any one May, nor have obser- 
vations been taken in its southeastern part during that month; but the data for 
1913, 1915, 1919, 1920, and 1925 afford a composite picture, which may be taken as 
representative for normal years because all are fairly consistent. 
In 1913 the surface salinity fell to its minimum (29.5 per mille) near the Isles 
of Shoals about May 5, followed by an increase to 30.9 per mille in the middle of the 
month; and while a northwest gale on the 10th, 11th, and 12th no doubt was 
partly responsible for this increase by bringing up more saline water from below, the 
spring influx of river water had evidently passed its peak by the first week of the 
month, to be gradually absorbed into the general circulation of the gulf thereafter. 
Unfortunately, close comparison is not possible between the years 1913 and 1920, 
for this region, because the locations of the stations do not coincide, which may cause 
a very considerable difference in salinity where the precise value depends so much on 
the proximity to the mouths of rivers. However, the surface again proved much 
fresher south of the Isles of Shoals on May 7 to 8, 1920 (station 20122, 28.26 per 
mille), than it had on April 9 (station 20092, 31.01 per mille) — a value even lower 
than any recorded for 1913. 
In 1920, too, the salinity of the surface of the northern part of Massachusetts 
Bay was almost as low as this on May 4 (stations 20120 and 20121, 29.1 to 29.16 per 
mille), but apparently this was close to the minimum for the month because followed 
by a considerable increase at this same general locality to about 29.9 per mille during 
the next 10 days (stations 20123 and 20124). 
In 1925 no observations were taken in Massachusetts Bay during the first 10 days 
of May, when salinity was probably at its lowest there; and the values recorded there 
on the 20th to the 22d (fig 119) were so high 87 that some increase may be assumed 
to have taken place during the second and third weeks of the month in that year, 
as it certainly did in 1920. 
Whether or not the surface salinity of the northern part of Massachusetts Bay 
fell below 30 per mille for a brief period in 1925, as April readings as low as 29 per 
mille in Ipswich Bay (p. 725) suggest, water of relatively low salinity was certainly 
drifting southward past Cape Ann as late as the third week of that May as a 
tongue less saline than 31.5 per mille directed toward Cape Cod (fig. 119). The 
87 31.1 to 31.9 per mille at the surface, averaging 31.6 per mille {Fish Hawk cruise 13). 
8951—28 48 
