CHAP, v.] GENEIIAL CONCLUSIONS. 257 



gives the impression that a generally uniform temperature is 

 maintained by a belt of water at a depth of from seven to eight 

 liundred fathoms, and that this belt separates two bodies of wa- 

 ter which are under essentially different conditions. Above, 

 the vertical distribution of temperature differs greatly in dif- 

 ferent localities; while below the uniform belt there is a slow 

 and gradual cooling, which also differs both in rate and in 

 amount in different localities, but in another way. These va- 

 riations in temperature, whether in the superficial layers or in 

 the deeper, are undoubtedly in all cases connected wntli currents 

 or movements of the water, and may be regarded as evidences 

 of portions, modified by various causes, of a general system of 

 circulation of the water of the ocean. 



The movements of surface-water may usually be determined 

 with considerable precision by a comparison at the end of a 

 given time of the apparent course of a ship and her position by 

 dead reckoning with her actual position by observation. The 

 rate and direction of a surface-current may also be ascertained 

 by getting in some way a fixed point — by anchoring a boat, for 

 instance — and observing and timing the course of a body float- 

 ing past it. JSTeither of these methods can be satisfactorily ap- 

 plied to deep-sea currents ; indeed, it seems probable that the 

 movements of masses of underlying water are so slow, that, even 

 if we had some feasible method of observation, the indications 

 of movement within a limited period would be too slight to be 

 measured with any degree of accuracy. 



We can not, therefore, measure these currents directly, but 

 we have in the thermometer an indirect means of ascertaining 

 their existence, their volume, and, approximately, their direc- 

 tion. Water is a very bad conductor of heat, and consequently 

 a body of water at a given temperature, passing into a region 

 where the temperature conditions are different, retains for a long 

 time, without much change, the temperature of the place where 

 its temperature was acquired. To take an example : the bot- 



