950 
BULLETIN OF THE BUREAU OF FISHERIES 
and the general distribution of density in the upper 20 meters (fig. 199) makes it 
probable that the calculated directions are a close approximation to the actual 
dynamic tendency toward circulation at the time. 
