River Temperature. 293 
at a time when the river temperature is approaching its 
maximum; they disappear nearly or completely at night. 
As the morning advances, the whole bulk of the water 
gradually heats up, but the surface at a quicker rate than 
the bottom; in the afternoon, however, the bottom begins to 
pick up its lost way, and by 8 or 9 P.M. the difference is 
very small. The bottom reaches it maximum temperature 
rather after the surface, and falls short of the surface 
maximum by about half a degree (F.), but both the surface 
and the bottom have the same or nearly the same minimum 
at night. That a river should exhibit diurnal fluctuations 
through its mass naturally follows from the experiments of 
the several observers, who have found little or no difference 
between the surface and the bottom temperatures. I have 
no data for rivers of great depth, but to rivers ordinarily 
deep this remark apples. Professor Plantamour speaks of 
the Rhone at its exit from the Lake of Geneva as having 
a scarcely appreciable diurnal variation at the depth of a 
metre. My facts go to show that this is not true of rivers 
generally, and it certainly would not be correct for the 
Thames or the Senegal, the last-named river displaying 
diurnal variation at a depth of 5 metres. This subject, how- 
ever, will be more fitly discussed when treating of the daily 
changes of river temperature. 
In winter, owing to the more rapid current, one would 
expect the Thames water to be always well mixed, and in 
fact my experiments gave little or no difference. One 
December morning, when the surface was 37° (F.), the 
bottom temperature, where the depth was 11 feet, proved to 
be not warmer. I got the same result at the end of the 
month, when the surface was 34°°3, and by carefully carrying 
out the observation, the possibility of the bottom being more 
than a small fraction of a degree cooler was excluded. On 
8th January 1894, I found at the same place a surface tem- 
perature of 33° and a bottom temperature not higher, the 
river being in great part frozen over. In the afternoon of 
4th January 1893, when the air in the shade was 23° (F.), 
the river having been frozen over for two days, I found 
in a depth of 10 feet, just above the upper lock-gates at 
