642 MR E. M. WEDDERBURN 
will be seen later, the theory of the oscillations accounts for this discrepancy by 
taking into account the raising of the surface temperature. 
The Station III. curve, from 8th to 10th August, gives a fairly pure example of the 
binodal seiche from which a period of 8:2 hours is obtained. 
For a study of the effect of meteorological conditions the diagrams in figs. 11-15 
are not very satisfactory, and accordingly the observations for Station I. are represented 
in a more convenient form in fig, 17*, which shows the relative positions of the various 
isotherms throughout the month. 
We shall first consider the effect of a change in the direction of the wind. The 
loch runs nearly due east and west, so that it 1s the easterly or westerly components of 
winds which are of chief importance from the point of view of temperature seiches. 
About 4 a.m. on 5th August the wind changed from a westerly to an easterly direction, 
but returned to a westerly direction at noon. The Station I. thermometer was not work- 
ing satisfactorily on 4th and 5th August, but the diagrams in fig. 7 show that the west 
wind, enduring for eight hours, was sufficient to establish a typical east wind distribution, 
and that within eight or nine hours after the change back to a west wind we again have - 
a west wind distribution, although the winds were very moderate in strength. In the 
afternoon of the 6th, during a lull of the wind, there was for a very short time an easterly 
breeze, but it was too short in duration and too light to produce any noticeable effect. 
Thereafter, until midnight between 8th and 9th August, the wind was westerly, 
then, for nine or ten hours, there was an easterly wind, but not very strong. Its 
effect is, however, at once seen on all the isotherms higher than 11° C. (see fig. 17). 
The surface isotherms at Station IL. sink, indicating the accumulation of warm water 
at the lee end of the loch. There is also noticeable a slight decrease in the amplitude 
of the oscillations. But about noon on 9th August the wind was blowing strongly — 
from the west, and the change of wind, in this case, was so timed as to considerably 
increase the amplitude of the oscillation. But this increase was of short duration, for 
about 8 p.m. on 10th August the direction of the wind again changed to the east. 
This wind was of considerable strength, and commenced when the isotherms at Station 1. 
should have been nearly at their lowest. The effect of the wind was to bring the 
isotherms close together and to prevent them rising, and in this way the oscillation 
which was in progress was practically damped out. 
Thereafter, until 17th August, the wind was easterly, but was generally light. On — 
the 17th there was a change to westerly winds, but, except for short periods, the 
winds were light and had little effect on the general distribution, although there is 
an appreciable lessening of the depth at which the higher isotherms are found. 
On 19th August the wind again changed from west to east, and the effect is at 
once seen on the surface isotherms, and a small oscillation which was in progress 
was damped out and the isotherms brought closer together. From 22nd to 24th 
August the wind was rather variable in direction, yet was never strong; but on the 
* See explanation of diagrams, p. 652. 
