306 Proceedings of the Royal Physical Society. 
tropical rivers, where the plane of water temperature is 
much higher than that of the air, and the range is sometimes 
outside the air’s range altogether. We learn from Schom- 
burgk that the Essequibo is usually 8° to 10° (F.) warmer 
than the air at 6 A.M., 1° to 2° cooler at 2 P.M, and 2° to 3° 
warmer at 6 P.M.; whilst it would appear from his observa- 
tions that the daily rise would be 1° or 1°°5. By expressing 
these facts in curves for the whole twenty-four hours, we 
at once perceive what a high plane of temperature this 
river must possess. Then, again, Humboldt states that at 
San Borja, in the middle of April, the Orinoco maintained 
its habitual temperature of 27°-7 (C.), whilst the air ranged 
from 23°-7 at night to 26° or 27°°5 during the day. In our 
own ponds in summer, as well as in our rivers, we find 
a considerably higher plane of temperature than for the air, 
and it is obvious that we have the same principle at work 
here as in the tropics. Heated by the sun during the day, 
and cooled by radiation at night, the temperatures of the air 
and of the pond and the river range about different means. 
Thus, on August 24th, when the air range was 49° to 74° (F.), 
and the Thames ranged from 64°5 to 66°°5, the ponds in 
Bushey Park ranged from 64° to 72°. The ponds possessed 
the highest plane of temperature, then the Thames, and 
lastly, the air, as represented by the means 68°, 65°5, 
and 61°°5. 
Postponing the detailed treatment of this complex matter 
of the relation to each other of the air and water tem- 
peratures, I come to the question of the influence of the 
water temperature on the air. This is a matter of practical 
importance in river observations; but at present I have data 
only for the effects of water much cooler than the air, as 
given in the following table :— 
