1903-4.] Prof. C. G. Knott on Ocean Temperatures, etc. 
181 
water beneath ; and if it could sink without loss of heat it would 
find its position of equilibrium at a depth of about 5 metres. 
This, of course, is a very crude description of what really occurs ; 
but it is sufficient to indicate the general nature of the convective 
process. The steady cooling by radiation of the surface waters 
must be accompanied by a steady vertical convection determined 
by the average temperature gradient and the viscosity of the 
liquid. This will go on steadily until an approximate equilibrium 
is reached, probably towards early morning ; and it is evident that 
by this process a considerable depth of surface waters will be 
cooled.* 
Of no small importance with respect to the question of the 
penetration of solar heat through the surface waters of an ocean or 
lake is the manner in which the temperature falls off as the depth 
increases. The curve shown in the figure represents the means 
given in Table B, and may be taken as typical of all cases in 
which the body of water is above the temperature of maximum 
density. 
It will be seen at a glance that the 
vertical distribution of temperature 
follows a somewhat complex law. As 
the depth increases the temperature 
falls off, first fairly rapidly, then more 
slowly until a depth of 20 metres is 
reached. Thereafter a rapid rate of 
diminution sets in, which attains its 
maximum at a depth of about 30 metres. The rate of decrease of 
temperature with increase of depth then begins to diminish, and con- 
tinues falling off till the greatest depths are reached. It is evident 
that this fairly permanent vertical distribution of temperature can- 
not be explained by conduction alone. Probably for depths greater 
than 40 metres the main factor is conduction of heat from the upper 
warmer layers to the cooler lower layers. But it is quite clear 
that some other factor powerfully affects the distribution of tem- 
* For an interesting discussion of similar phenomena in the fresh-water 
lakes of the Austrian Alps, see “ Seestudien,” by Professor E. Richter (Geo- 
graphische Abhandlungen, edited by Professor Penck, Vienna, Band VI.. 
Heft 2, 1897) — an important memoir. 
