304 THE DEPTHS OF THE SEA. [CHAP, vn 



There can be no doubt that this view, which oi 

 late years has received almost universal acceptance, 

 is entirely erroneous. It has been shown by M. 

 Despretz, 1 as the result of a series of carefully con 

 ducted experiments which have since been frequently 

 repeated and verified, that sea-water, as a saline 

 solution, contracts and increases steadily in density 

 down to its freezing-point, which is, when kepi 

 perfectly still, about -3'67 C. (25'4P.), and whei 

 agitated -2'55 C, 



The temperature observations of Sir James Clark< 

 Ross during his Antarctic voyage in 1840-41, 

 seemed to give support to the theory of a constanl 

 temperature of 4'5 C. for deep water, but these obser 

 vations have as evidently been made with unguarde 

 instruments, as those of Sir John Ross in 1818 wit] 

 instruments defended from pressure ; and althougl 

 I believe they must be taken as proving that ii 

 high southern latitudes the surface temperature 

 sometimes lower than the temperature of the watei 

 at a considerable depth beneath, still the amount oi 

 correction for pressure is uncertain, depending upon 

 the construction of the thermometers used, and in 

 any case it must reduce the difference considerably. 



A large number of thermometers of the ordinary 

 Hydrographic Office pattern were sent out with us, 

 as I have already mentioned, in the 'Lightning,' 

 and these were of course the instruments used by 

 Staff-Commander May for his temperature obser 

 vations. There was an opportunity of testing these 

 thermometers, however, on the return of the vessel, 



1 Recherches sur le Maximum de Densitc des Dissolutions aqueuses. 

 Loc. cit. 



