962 BULLETIN OF THE BXJRIJAU OF FISHEKIES 



surrounded by lower densities to the south, and so separated from equally heavy 

 water to be expected near the Eastern Channel and through the trough of the latter, 

 just as was the case in July and August, 1914. The available data thus suggest that 

 the dynamic tendency toward circulation continues regularly anticlockwise from 

 summer to summer in the northern and northwestern parts of the gulf, though dif- 

 ferences in the location of its center of revolution and in the regional distribution 

 of density off the western shore are correspondingly reflected in the stream lines. 



AUTUMN AND WINTER 



Progressive equalization of temperature taking place in the shoaler strata of the 

 gulf during the autumn obliterates the pool of low density that characterizes the 

 offing of Massachusetts Bay in summer. As a result, the distribution of density 

 comes to conform more and more closely to that of salinity. In the midwinter of 

 1920-1921 (apparently a representative season), the upper 100 meters were less dense 

 around the coast than in the basin offshore, with the transition more abrupt in the 

 western side than in the eastern, and the values highest in the oflBng of Cape Ann 

 (station 10490). 



A regional inequality of this sort must cause a dynamic tendency for the 

 coastal belt to drift parallel with the land anticlockwise around the gulf, much as in 

 spring (p. 942), producing a northerly set along Nova Scotia, westerly along the coast 

 of Maine, and southerly from the offing of Cape Elizabeth past Cape Ann to Mas- 

 sachusetts Bay, relative to the underlying water mass. This latter (as represented by 

 the 150-meter level) then proved nearly uniform in density horizontally (i. e., was 

 nearly stationary) . Unfortunately, no data are available for the southern or south- 

 eastern parts of the area for midwinter. 



The progressive mixing of the water that takes place as winter advances makes 

 the upper stratum more and more uniform, both horizontally and vertically, with 

 respect to density as well as in temperature and salinity, until by February it becomes 

 nearly homogeneous, as described above (p. 522), and the annual cycle is complete. 



WIND CURRENTS 



Seafarers have Icnown, from the dawn of history, that the wind sets up surface 

 currents often so strong that they must be taken seriously into account in navigation; 

 and many a good ship has been wrecked from ignorance of the wind current. 



In the Gulf of Maine the motive effect of the wind is made most apparent to the 

 oceanographer by the upwellings of colder and salter water from below, which take 

 place along its western margin when the surface water is driven offshore (p. 550). 

 Every fisherman along our coasts knows from first-hand experience that strong winds, 

 blowing from one quarter or another, strengthen the ebb at the expense of flood — or 

 vice versa, as the case may be. 



The dynamic principle according to which wind currents are produced is extremely 

 simple: The wind drives the surface water before it, the motion of the latter being 

 propagated to underlying strata by the internal friction of the water. Once in motion, 

 the water, as Nansen (1902) and Ekman (1902) have pointed out, must be deflected 

 by the effect of the earth's rotation. Nansen's (1902) observations on the drift of 



