[ 551 ] 
xn^ On the Forniafion of Definite Fujures Inj the Deposition of Dust. 
By John Aitken, F.R.S. 
Eeceived July 13, 1903. 
Owing to the kindness of Dr. W. J. Russell, F.R.S., I received in June an advance 
copy of his paper on the above sidiject.'^'' After reading this paper it appeared to me 
that all the figures illustrated in it could be exjilanied on well-known principles. I 
shall therefore do what I can to fulfil the hope expressed by Dr. Russell at the end 
of his paper that physicists from his descriptions may be enabled to explain their 
formation. 
The formation of these dust figures appeal’s to be due principally to three causes : 
(1) the convection currents set up by the hot plate; (2) to gi-avitation; and (3) to the 
repelling action of the liot surface. It seems trivial to remind the reader that 
gravitation plays a part in the formation of these figures, but it is to be feared that it 
is from not keeping the efiects of gravitation fully in view that difficulty has been 
experienced in explaining them. It is iirincipally owing to gravitation, or rather to 
an after-effect of gravitation, that no dust is deposited on certain parts of the plate. 
Gravitation acts on the dust under the plate as well as on the dust over it, thus 
causing the film of air flowing along the under surface of the plate to be dust- 
fiee, all the dust having fallen out of it. This dust-free film of air, after flowing 
along the under surface of the plate, turns round the edges and flows over the top 
sill face, presenting its dustless side to the plate, and the air has to travel some 
distance over the top surface before the dust falls througli the dustless film. That is, 
it takes some time for the upper current to undo the work of the under current, and the 
result is no dust falls on the plate till tlie current has flowed some distance from the 
edge. As stated, the third influence at work in the formation of these dust figures is 
O 
the rejielliug action of the hot surface. It is well known that a hot surface tends to 
keep itself free from dust while surrounded by dusty air. The hot surface may in a 
manner be said to repel the dust, the action being probably due to the air next the 
hot body being warmer than the air at a slight distance from it, and the dust 
particles, being more strongly bombarded by the hotter air molecules on the one side 
than by the colder ones on the other, are driven away from the hot surface. The 
energy of this action will probably be the greater the cpiicker the temperature 
gi'adient in the air in a direction at right angles to the hot surface. 
VOL. cci.—A 344. 
* ‘Phil. Trans.,’ series A, vol. 201, pp. 185-204. 
25.8.03 
