280 



DEPARTMENT OF THE NATAL SERVICE 



It may also happen that the water, through inertia, flows too far. This gives rise 

 to a system of solenoids the reverse of the original, and tending to turn the current, 

 and so drive the water back. If this reverse movement, in its turn, carry the water too 

 far, then a new solenoid system, corresponding to the original one, will again arise, turn- 

 ing the current once more in its original direction. This may be repeated several times. 



Fig- 55. — Alteration of circulation ac- 

 celeration on the appearance of a) a 

 permanent current b) a casual and 

 temporary current. 



The alteration in the acceleration of circulation will then become periodical, vide 

 fig. 56. 



From the foregoing, it will be seen that the sea is continually striving towards a 



dC 



state where 



dt 



approaches nil, i.e. the three terms on the right side of Bjerknes cir- 



culation equation endeavour to adapt themselves one to another in such a manner that 

 their sum shall be nil. This would also take place, were it not for the fact that one or 

 other of them is from time to time subjected to disturbance which alters its magnitude. 

 This disturbance may be of various sorts. A physical change in the water, as for in- 

 stance that brought about by the heat of the sun, or by the melting of ice, will alter the 



dS 

 value of ^. If a current veers off owing to the topography of the sea basin, then 2w — 



dt 



will be altered. The action of the wind exerts a great influence upon R, etc. After 



dC 

 each such disturbance -= — attains a maximum, decreasing afterwards, however, 

 dt 



towards nil. Where the stability of the water layers is very great, as in the gulf of St. 



dC 

 Lawrence, this decrease may take place so forcibly that — — - passes beyond zero and 



becomes negative, thereafter altering periodically with decreasing amplitude, vide fig 

 56. The greater the stability of the water layers, the more rapidly will these oscilla- 

 tions take place. 



Such oscillation also, of course, take place in the surface movement of the water, 

 and would appear to be particidarly frequent in surface currents occasioned by the 

 wind; also, other surface currents however arising from different causes, doubtless 

 oscillate in the same way. As the water at any point of the surface can, at a given 



