55G 
MR. JOIiN AITKEN ON THE FORMATION OF 
Take, for instance, fig. 20. Here most of the air from beneath the j^late flows to the 
higher edge and but little curves round the lower one, while the side streams keep about 
the usual strength. The current, however, round the higher edge being warmer and 
stronger than usual, does not flow to the centre of the plate, which is in a dovmward 
direction, but rises in an easy curve, with the result that over a large area of the plate 
the air is nearly motionless and the dust is free to deposit itself on the plate. In fio-. 21 
no dustless film seems to have come from tlie lower edge owing to the high angle of the 
plate, and all the hot air from the under side has flowed to the higher edge ; where the 
rising current has been so strong it has curved in but little, and as the side currents 
are weak, as most of the hot air has flowed to the upper edge, the greater part of the 
plate is therefore exposed to the dusty air flowing over it from the lower edge. 
The figure shown in fig. 22 seems to be due to the obstruction placed on the plate 
interfering with the regular flow and causing eddies and deposition of dust, while 
the dustless film enters the holes in the obstruction and, as usual, protects the surface 
in front of them over which they flow. 
The curious eftect of cutting a re-entering angle out of the plate, as shown in 
fig. 22a, is very interesting, and shows that the cutting out of that angular piece 
has in some way introduced new conditions which have interfered with the protecting 
action of the dustless film. Keferring this to experimental observation, it is seen at 
once how this peculiar deposit is produced. The beam of light shows that the air 
streaming up through the angular opening does not turn over and flow over the plate 
but rises straight up, owing to the large quantity of hot air drawn to the one point. 
This upward-moving current induces another current over the plate moving towards 
it from the centre, and as this current flows slowly and is composed of dusty air 
without a dustless film, the particles settle out of it and cause the peculiar markino’ 
extending from the centre to the angular opening. 
The next series of figures, from fig. 23 to 29a, produced by the action of a piece 
of glass, a pin, a hair, or other obstruction touching, or even near the edge of, the 
plate, are most curious and unexpected. On putting these conditions to the test of 
observation, it was seen that all these oljstructions cause deposits to form by the 
interference they offer to the stream-lines of air moving over the surface of the plate. 
Where the obsti'uction cuts the stream the current is slaclcened, and more or less 
eddying probably takes place, enabling the dust to settle. What is seen when the 
air is examined with the lens while illuminated by a narrow beam of light is as 
follows:—While the beam of light is moved about on either side of the obstruction 
the air is seen to flow in well-defined stream-lines, the lower surface of the dusty air 
being distinct and clearly defined, but wlien the light shines on the air that has 
passed the obstruction, the upper limit of the dustless air has lost its definition. 
And further, if the beam of light were moved backwards and forwards, fi'om one side 
to the other of the obstruction, it was observed that not only the upper limit of the 
dustless air on each side of the obstruction was well defined, but there was alwavs a 
