58 Proceedings of the Royal Society of Edinburgh. [Sess. 
naents have been discussed by Professor King in his book The Soil, where 
he gives particulars of their influence on the height of water in wells and 
on the flow of springs.* If the downward expansion of air — heated by 
the action of the sun on the surface — is sufficient to bring about these 
movements of soil water, it is only to be expected that it should also 
affect the underground temperatures. 
When we examine closely the temperature curve for the 4-inch depth 
on sunny and overcast days, two points of difference are at once apparent : — 
(1) On sunny days, as soon as the surface begins to rise rapidly in 
temperature a slight convexity appears in the 4-inch depth curve which 
is not noticeable on an overcast day. This convexity cannot be due to 
conduction, as it appears very soon after the first rapid rise of the surface, 
whilst any change due to conduction could only show after a number of 
hours equal to the lag on the day in question. Moreover, the surface 
temperature curve is almost invariably concave, and so could not give 
rise to any convexity in the 4-inch depth curve. A corresponding but 
slighter concavity appears early in the afternoon, when the surface falls 
rapidly in temperature. 
It appears that these changes in the shape of the 4-inch depth tempera- 
ture curve can be explained in part by the fact that as soon as the surface 
becomes hot the soil air expands and pushes downwards, displacing the 
air round the 4-inch depth thermometer and causing a small rise of 
temperature there ; whilst in the early afternoon, when the surface cools 
again and the soil air contracts, this displaced air rises and regains its 
former position. This would only be noticeable on sunny days, when the 
surface changes of temperature are large and rapid. 
Recent observations seem to point to the possibility of these variations 
being brought about in part by renewed percolation of water due to the 
reduced viscosity and surface tension of the soil water consequent on the 
rapid rise of temperature of the soil. 
Under strong sun a temperature of 30° C. at the surface, and from 
15°-20° C. at the 4-inch depth, may often be attained, and the very consider- 
able reduction in viscosity entailed must result in renewed percolation of 
soil water, with a corresponding rise in the 4-inch depth temperature. 
This will be further discussed in Section V. 
(2) On overcast days or in shady places, owing to the absence of the 
above-mentioned convexity, the range of temperature at the 4-inch depth 
R 
is diminished, and the value of — 4 — the ratio of the ranges at the 4-inch 
K o 
* The Soil, by F. H. King. The Rural Science Series. Macmillan, 1917. 
