River Temperature. 289 
temperature and that of the mass of the river in great part, 
if not altogether, disappears. 
Although such marginal heating may form an insignificant 
feature in the thermal economy of the Thames, it can readily 
be perceived that in the case of a much larger river bordered 
by extensive shallows, this phenomenon might be observed 
on a far greater scale. It ought to be very pronounced in 
the instance of some tropical rivers. In an inland journey 
from the port of Guayaquil, during the period of the 
inundations when the rivers flowed over the vast savannahs, 
A. von Humboldt sailed on the flooded plains where the 
water was 4 to 5 feet deep and the temperature 33° to 
34° (C.), or 91°-4 to 93°2 (F.). When we contrast this with 
the usual temperature given by this traveller for the Orinoco, 
viz., 27° to 28° (C.), or 80°6 to 82°4 (F.), we recognise an 
important feature in the thermal regime of tropical rivers. 
The marginal heating of a river like the Thames, and the 
superheating of the waters of the flooded plains of a tropical 
river, are determining elements in the distribution of plants, 
and it is noteworthy that in the case of the Thames the 
same causes that favour the heating of the water at the edges 
favour also the collection of floating seed-drift, a fact of 
weight in the germination of floating seeds. This is a 
subject I have been working at for the last three years, and, 
though full of interest, it can only here be alluded to in 
passing. 
Coming to the question of the uniformity or otherwise of 
a river’s temperature in depth, I append the following data, 
which go to show that, except in the case of rivers that have 
an almost imperceptible current, the temperature varies but 
slightly with the depth. It cannot be said, however, that 
the data give full precision to this inference, materials 
relating to great rivers being scarce, and when present often 
scanty. We can, notwithstanding, accept the general con- 
clusion, based as it is on the testimony of several observers, 
and leave to time and to research to supply us with 
particulars. 
During the summer of 1834 Professor Merian, of the 
University of Bale, found no difference between the tempera- 
