December 28, 1893] 



NATURE 



213 



of electricity. The great maiority of the small particles or- 

 dinarily formed consisted, I thought, of perhaps only a few 

 molecules, which were dispersed in the air, and again converted 

 into vapour without ever having become visible, while the larger 

 particles formed by their coalescence under electrical action were 

 of such dimensions as to impede the more refrangible waves of 

 light. Hence the brownish-yellow colour. 



Other explanations have been proposed. There is the mole- 

 cular shock theory of the late R. Helmholtz(who, as it turned 

 out, had studied electrified steam jets before I made my own 

 .experiments). I shall refer to his speculation later. And there 

 is the dust-nucleus theory, which no doubt appears a very 

 obvious one. 



Though I knew that my own hypothesis was not quite free 

 from objection, neither of these alternative ones commended 

 itself to me as preferable ; and so the matter rested until a few 

 months ago, when the steam jet phenomenon was discussed 

 anew in a paper communicated to the Royal Society by Mr. 

 Aitken. Mr. Aitken said that he did not agree with my con- 

 jecture as to the nature of the effect. This led me to investi- 

 gate the matter again, and to make some further experiments, 

 the results of which have convinced me that I was clearly in 

 error. At the same time it seems to me that the explanation 

 which Mr. Aitken puts forward is little less controvertible than 

 my own. Mr. Aitken's explanation of the phenomenon is, like 

 mine, based upon Lord Rayleigh's work in connection with 

 water-jets, but, unlike mine, it depends upon the experiment 

 which shows that water particles when strongly electrified are 

 scattered even more widely than when unelectrified. He believes, 

 in short, that electrification produces the effect, not by pro- 

 moting coalescence of small water particles, but by preventing 

 such coalescence as would naturally occur in the absence of 

 electrical influence. In the electrified jet, he says, the water 

 particles are smaller, but at the same time more numerous ; thus 

 its apparent density is increased. 



The chief flaw in my hypothesis lies in the fact that the mere 

 presence of an electrified body like a rubbed stick of sealing- 

 wax, which is quite sufficient to cause coalescence of the drops 

 in the water jet, has no action whatever upon the condensation of 

 the steam jet. There must be an actual discharge of electricity. 

 But it is by no means essential, as Mr. Aitken assumes, that 

 this discharge should be of such a nature as to electrify, posi- 

 tively or negatively, the particles of water in the jet. If, instead 

 of using a single electrode, we employ two, one positive and the 

 other negative, and let them spark into each other across the 

 jet, dense condensation at once occurs. So it does if the two 

 discharging points are removed quite outside the jet. A small in- 

 duction coil giving sparks an eighth of an inch in length causes 

 dense condensation when the electrodes are more than an inch 

 distant from the nozzle and on the same level. In one experiment 

 a brass tube two feet long was fixed in an inclined position with 

 its upper end near the steam jet, and its lower end above the elec- 

 trodes of the induction coil. In about three seconds after the 

 spark was started dense condensation ensued, and it ceased 

 about three seconds after the sparking was stopped. No test 

 was needed, though in point of fact one was made, to show that 

 the steam was not electrified to a potential of a single volt by 

 this operation. And the time required for the influence to take 

 effect showed that whatever this influence might be it was not 

 induction. 



The inference clearly is that in some way or other the action 

 is brought about by the air in which an electrical discharge has 

 taken place, and not directly by the electricity itself. The idea 

 has no doubt already occurred to many of you that it is a dust 

 effect. Minute particles of matter may be torn off the electrodes 

 by the dis,';harge, and form nuclei upon which the steam may 

 condense. The experiments of Liveing and Dewar have indeed 

 shown that small particles are certainly thrown off by electrical 

 discharge, and the idea that such particles promote condensation 

 appears to be supported by the fact that if a piece of burning 

 material, such as touch-paper, is held near the jet so that the 

 products of combustion can pass into it, thick condensation is 

 produced. 



From a recent paper by Prof. Barus, published in the 

 American Meteorological J ournaliox March, it appears that he 

 also is of opinion that such condensation is in all cases due to 

 the action of minute dust particles. Yet it is remarkable that 

 Mr. Aitken, the high priest and chief apostle of the philosophy 

 of dust, gives no countenance to the nucleus theory. He doss 

 not even advert to its possibility. I imagine that his experi- 



NO. I 26 I, VOL. 49] 



ments have led him, as mine have led me, to the conclusion that 

 it is untenable. And this not only in the case of electrical dis- 

 charge, but also in the case of burning matter. 



If we cause an electrical discharge to take place for some 

 minutes inside a suitably arranged glass bottle, and then, ten or 

 fifteen seconds after the discharge has ceased, blow the air from, 

 the bottle into the steam jet, the condensation is not in anyway 

 affected. Yet the dust could not have subsided in that time. 

 And again, if we fill another large bottle with dense clouds of 

 : smoke by holding a bundle of burning touch-paper inside it, 

 and almost immediately after the touch-paper is withdrawn, 

 force out the smoke-laden air, through a nozzle, upon the jet — 

 ! you can all see the black shadow of the smoke upon the screen 

 — nothing whatever happens to the jet. Yet a mere scrap of the 

 paper which is actually burning, though the ignited portion may 

 not be larger than a pin's head, at once darkens the jet. Dead 

 smoke (it I may u.se the term) exerts little or no influence by 

 I itself: there must bj incandescent matter behind it. The 

 question naturally arises, whether incandescent matter may not 

 [ be sufficient of itself, without any smoke at all. \Ve can test 

 j this by making a piece of platinum wire red-hot and then holding 

 j it near the jet. It is seen to be quite as effective as the burning 

 touch-paper. Yet here there can be no nuclei formed of pro- 

 ducts of combustion, for there is no combustion ; there is simply 

 \ ignition or incandescence. 



One other point I may mention. It is stated by Barus, in the 

 ! paper above referred to, that the fumes given off by a piece of 

 phosphorus constitute a most efficient cause of dense conden- 

 sation. This is true if they come directly from a piece of phos- 

 phorus ; but if phosphorus fumes are collected in a bottle and 

 then directed upon the jet, all traces of unoxidised phosphorus 

 being first carefully removed, they are found to be absolutely 

 inoperative. Phosphorus in air can hardly be said to be incan- 

 descent, though it is luminous in the dark ; but it appears to 

 act in the same manner as if its temperature were high. 



All these facts seem to indicate that the several causes men- 

 tioned, electrical, chemical, and thermal, confer upon the air 

 in which they act some temporary property — certainly not due 

 to mere inert dust — in virtue of which it acquires an abnormal 

 power of promoting aqueous condensation. 



I thought that possibly some clue as to the nature of this 

 property might be obtained by observing how some other gases 

 and vapours behaved ; but though the experiments I made 

 perhaps tend to narrow the dimensions of the mystery, I cannot 

 say that they have completely solved it. Indeed some of the 

 results only introduce additional perplexities. 



One of the most natural things to try is hydrochloric acid, 

 which is known to have a strong affinity for water. If we heat 

 a little of the acid solution in a test tube, closed with a cork, 

 through which a glass tube is passed, and direct the issuing 

 stream of gas upon the jet, the densest condensation results. 

 The vapours of sulphuric and nitric acids also cause dense 

 condensation, and I suppose both of these have an affinity 

 for water. But so also, and in an equally powerful degree, does 

 the vapour of acetic acid ; yet the aflinity of this acid for water, 

 as indicated by the heat evolved when the two are mixed, is very 

 small. 



Ammonia gas, when dissolved in water, causes the evolution 

 of much heat. Yet a stream of this gas directed upon the jet 

 has no action. 



Ozonised air, which Mascarl found so effective in his experi- 

 ments with the closed vessel, is quite inoperative with the 

 steam jet. Equally so is the vapour of boiling formic acid, 

 which I believe is chemically a much more active acid than 

 acetic, and has a lower electrical resistance. (See Table.) 



Condensation of Steam Jet. 



Active. 



Air, oxygen, or nitrogen, in which electrical discharge is 

 occurring. 



Burning and incandescent substances. 

 Fumes from phosphorus. 

 Hydrochloric acid. 

 Sulphuric acid vapour. 

 Nitric acid vapour. 

 Acetic acid vapour. 



