of Edinhurgh, Session 1883-84. 
671 
radiation, and radiate heat to all surfaces in every direction within 
their view. All surfaces, therefore, though shaded from the sun, 
are receiving a great amount of radiant heat ; and if this heat was 
not carried away by the air, the surfaces in shade would acquire 
nearly the same temperature as those in sunshine. But the supply 
of heat comes in slowly from the sunlit surfaces, and the passing 
air carries it away nearly as quickly as it is received. We see from 
this that the temperature of bodies, even in the shade, is greatly 
determined by the rate at which this radiant heat is carried away. 
Now, very little consideration is necessary to show us that the air 
will carry away the heat much more quickly from a small than from 
a large surface. When a current of air comes into contact with a 
hot surface, the air where it first touches the surface is rapidly 
heated, and a film or layer of hot air is formed near that surface. 
Now, if the body is small, this hot layer is at once swept away, and 
gives place to another cold layer, which in turn abstracts more heat ; 
whereas in a large body, the hot film or layer formed where the 
current first touches the body keeps near the surface in its passage 
over it, and forms, as it were, a hot j)rotecting coat to all parts of 
the surface at a small distance to the leeward of the point where 
the current first touched it. No doubt this hot layer of air will 
thicken as it passes over a large surface, but the rate at which it 
thickens is very slow after it has passed over a very small distance. 
This I have observed in my late experiments on dust. The effects 
due to the heat in the air in front of a hot flat vertical surface are 
almost all accomplished at the lower edge of the plate, where con- 
tact is first made, the upper part of the plate producing but a slight 
increase in the effect. The result of the air acting in this way is, 
that a large body parts with its heat to the passing air at a much 
slower rate per unit of area than a small body does; and as both are 
receiving heat at the same rate, the temperature of a small body is 
lower than that of a large one. The protection of large bodies 
will be far from being so perfect when the currents are rapid and 
the air ceases to move in stream lines. 
Experiments were made to see if this reasoning was correct, 
to determine whether large bodies were really hotter than small 
ones, and if so, to what extent. For this purpose I pre- 
pared three tin boxes, all 8 inches long, but of different 
