Birge—On the Evidence for Temperature Seiches . 1013 
rents set up by the northeast wind of Aug. 19 and which con¬ 
tinued to move slowly after the wind died away. It must be 
remembered that conditions on Aug. 19 were very different 
from those on Aug. 17. When the northeast wind began on 
the former date, the southwest wind, extending through the 
preceding twenty-four hours, had blown all of the warm water 
from the southern end of the lake. Several hours and much 
wind were, therefore, necessary to bring it back. But on Aug. 
18 there was little wind; a large mass of warm water remained 
at the yacht and there was doubtless much more in-shore. It 
required, therefore, only a small amount of northeast wind to 
put the deeplying warm water into motion again. I suppose 
that the mass of warm water, accumulated by the wind of 
Aug. 17, moved out to the observing station, causing the rise of 
temperature before midnight of Aug. 19-20, and then was 
either crowded back or pushed past the station, causing the 
fall of temperature in the early morning of Aug. 20. I am 
very glad to admit that this movement looks like a temperature 
seiche, but the evidence that it is such is by no means complete, 
or even adequate. We do not know how much warm water lay 
in the southwest end of the lake on Aug. 19, nor do we know 
the extent or shape of the mass. We do not know that there 
was any corresponding movement of temperature in a reverse 
direction on Aug. 19-20 at the other end of the lake, or indeed 
that there was any movement of temperature at all. So far 
as the facts at hand give evidence, there is no reason which 
requires us to conclude that the disturbance of Aug. 19 was a 
different kind of phenomenon from that of Aug. 15, 17, or 22; 
in each of which not only is the qualitative relation to the wind 
distinctly marked but the quantitative relation seems adequate. 
Thus after a review of the clearest evidence for temperature 
seiches, adduced by Wedderburn, I am forced to give my vote 
for the Scotch verdict of “not proven”. I am quite ready to 
accept seiches of any period or amplitude that can be proved, 
but I cannot see that we ought to accept seiches of 200 feet 
amplitude, or even of any amplitude, on the evidence presented. 
It is unfortunate for the establishment of the seiche theory 
that Loch Hess is so long. The period of its oscillation, about 
