116 
Proceedings of the Royal Society 
great interest, but it is almost impossible to get tlie time to make 
certain ones necessary to connect them or to work them up. I hope, 
however, now that we have got a new Mathematical Professor, that 
I will get a little more time for such work. 
By the way, I have discovered that the “green sun” spectrum 
can be exactly represented by combining the spectrum of the sun 
seen through a fairly thick mist — the sun’s disk being still visible — 
and the spectrum of a moist atmosphere, showing the rain band 
strongly. This would explain all cases in which the sun has been 
seen green. 
There is one point in regard to measuring atmospheric electricity 
on which I am not quite clear. In using the water-dropping 
collector, Thomson seems to say that you must have the reservoir 
inside, so that part may be on one side and part on the other of the 
neutral line. I have tried comparisons between the lighted match 
and the water-dropper entirely outside, and they seemed to agree 
fairly well. I can get no good exposure here if I simply put the 
water-dropper at a window, and I would like to have it on the roof 
of the house, but will that do ? — I remain, yours very sincerely, 
C. MICHIE SMITH. 
7. On a Mew Form of Chromotrope. By John Aitken, Esq. 
8. On an Application of the Atmometer. By Professor Tait. 
The Atmometer is merely a hollow ball of unglazed clay, to which 
a glass tube is luted. The whole is filled with boiled water and 
inverted so that the open end of the tube stands in a dish of 
mercury. The water evaporates from the outer surface of the clay 
(at a rate depending partly on the temperature, partly on the dry- 
ness of the air) and in consequence the mercury arises in the tube. 
In recent experiments this rise of mercury has been carried to nearly 
25 inches during dry weather. But it can be carried much farther 
by artificially drying the air round the bulb. The curvature of the 
capillary surfaces in the pores of the clay, which supports such a 
column of mercury, must be somewhere about 14,000 (the unit 
being an inch). These surfaces are therefore, according to the 
curious result of Sir W. Thomson (Proc. R. S. E. 1870, p. 63), 
specially fitted to absorb moisture. And I found, by inverting over 
