546 



NATURE 



[April 9, 1903 



a circle, no deposit forms. It does not appear as if the 

 composition of the dust used to produce these figures is a 

 matter of importance, the dust from ashes, from ammonium 

 chloride, the fine spores of a fungus, all produce the same 

 figures, but the magnesia produced by burning magnesium 

 is, as before mentioned, more brilliant in colour, and more 

 readily produced than any other dust. There is, however, 

 one essential character necessary in whatever dust is used, 

 namely, that it be very fine. 



To obtain these figures perfectly regular in form, care has 

 to be taken that the atmosphere surrounding the plat.- shall 

 be fairly uniform in temperature. If the reservoir of dust be 

 a glass vessel, and an ordinary Bunsen burner be at a 

 distance of one to two feet from the plate and outside 

 the receiver it is sufficient to spoil the symmetry of 

 the cross by either making one limb of it much thicker 

 than the others or by pushing it more or less on one 

 side. Again, by placing a hot body under the plate 

 while the dust is depositing, curious modifications of 



F .;. 



Hi.- deposit are produced, but require the photographic 

 pii tures to show exactly what has taken place. At first this 

 extra heating causes an increase of deposit, but when the 

 temperature rises beyond a certain point, it gradually 

 diminishes the amount of deposit, and if the plate re I on 

 a metal support which is at a temperature of about [50° ( '., 

 no deposit takes place. In fact, for each different way that 

 the heat is applied a different form of deposit is produced. 

 F01 instance, if the plate be ii.it heated, but is placed on a 

 small metal cylinder which is heated, a remarkable deposit 

 is formed ; so again when a hot or cold metal cylinder is 

 placed on the top of the plate instead of below it. curious 

 and complicated figures are formed. When the plate is not 

 exactly horizontal, the figures formed on it are no longer 

 s\ mmetrical, but have the appearance of sliding down the 

 plate. Very remarkable effects are produced on these dust 

 <l. posits by proximity to the plate of different-sized bodies; 

 fen instance, stick up a piece of glass against the plate, and 



NO. 1745, VOL. 6/] 



then expose it to the dust. The glass screens the plate from 

 the currents of air formed, and a deposit takes place accord- 

 ing to the size of the obstruction. Fig. 2 shows what 

 happens when an ordinary pin is placed with its point on a 

 level with a square plate, and at a distance of 3 mm. from 

 it. The cross is still formed, but the pin has caused 

 .1 realm of calm from the centre towards the edge 

 of the plate. Again, Fig. 3 shows strikingly the deli- 

 cacy of this kind of action ; the fine deposit ending in 

 a fine point was produced by sticking a human hair 

 vertically against the side of the plate and exposing it to 

 the dust atmosphere. 



It is then unnecessary for this pin or post to be in con- 

 tact with the plate; it may be at a distance of some S to 

 10 mm. from the plate. It may be above the level of 

 the plate, on a level with it, or even below its level, and 

 still influences the deposit of dust. In all cases, as the pin 

 recedes from the plate, so does the deposit recede from the 

 edge, getting smaller and smaller, until at last it dis- 

 appears at the centre. It is difficult to 

 realise that a pin held so that its point 

 is at 6 mm. below the level of the plate 

 and 2 mm. away from it should be able 

 to induce on the plate a definite and 

 decided deposit, but such is the case. 

 In using glass plates for the figures to 

 deposit on, care must be taken that the 

 edges are quite smooth, for if not, the 

 small pieces forming the rough edge of 

 a cut piece of glass are sufficient to 

 1 ause spikes of deposit to shoot out 

 from the centre on other parts of the 

 figure. 



There still remains another way of 

 studying the formation of these singular 

 figures and influencing their formation, 

 l>\ offering obstructions to the free de- 

 position of the dust ; for instance, if a 

 Strip of glass be placed across a square 

 plate, and the strip be not more than 1 

 mm. high, the deposit takes no notice 

 of it, and the cross forms as if the strip 

 was not there ; but increase the height 

 of the strip, make it 4 or 5 mm. high, 

 and the figure becomes much altered, 

 and the form of the deposit is much 

 changed. Again, if the obstruction to 

 the free flowing of the currents be pro- 

 duced by hanging a strip of glass or a 

 point above the plate to receive the de- 

 posit, an interesting series of figures is 



i. 1 d, but these cannot be discussed 



without the illustrations. Fig. 4 may, 

 however, serve to give some idea of the 

 kind of changes which are produced. 

 This represents a square glass plate 

 with a strip of glass some 25 mm. high, 

 and longer than the plate, placed across 

 it, and a pin pressed against it at the 

 middle of the lower side. The influence 

 of the four corners of the plate, of 

 the pin and of the strip are all 

 clearly indicated ; also it will be 

 seen that the right hand ray at the top of the picture 

 has two points, the smaller one is produced by some 

 splint. -r of glass which was very near to the corner. In 

 the full paper to be printed in the Phil. Transactions, there 



,1 ne fifty pictures showing the formation of different 



figures. 



If the fine powder from burning magnesium is used on 

 a glass plate, it is, when first deposited, easily removed by 

 tli, slijjii.st touch, but if allowed to remain on the glass for 

 some time, say a fortnight, it becomes comparatively fixed 

 there, and may even be lightly rubbed without being re- 

 move. I. 



If mercury ir. a square vessel be used in place of a 

 solid plate, the same figure of a cross forms upon it. If 

 water be used, entirely different figures form, the sinking 

 of the powder gradually through the water producing other 

 changes. 



W. J. Kiss! II .. 



