Skey. — Oil as a x^ucleus in Super-saturated Saline Solutions. 409 



scientists, I will now affirm that such facts are irreconcilable with this 

 theory of theirs. 



The question I have started as to whether or no surface tension has an 

 effect, even of a mere subsidiary character, in producing these nuclear 

 effects described, I leave for solution to those physicists who for so many 

 years have been working at the mechanical properties of hquid surfaces. 



But I must remark, ere I leave this part of my subject, in relation to the 

 effect or non-effect of any specific oil in producmg crystallization, that in 

 regard to this, not only does the more or less resinous condition of an oil 

 affect it, but also the more or less proneness it has to become resinoid upon 

 exposure to air and water ; and this brings us face to face with another fact 

 elicited by these investigators. It is, that in dull, damp, and cloudy 

 weather an oil may not produce nuclear effect ; whereas on a fine bright 

 day it may be " particularly active." This they explain by supposing the 

 surface tension of the solution to be lower upon cloudy than upon bright 

 days. But in connection mth my theory (the chemical one) I would point 

 to the well-known fact, that light in conjunction with air favom-s the change 

 of oil generally to resmous matter, and hence an acquirement of nuclear 

 effect on bright days. 



Another and the last remark I have to make here is, that in all those 

 experiments with oil where a nuclear effect is produced by these scientists, 

 a flashing or spreading out of such oil from the lens-shape it may first 

 assume, is the constant concomitant or antecedent. Now, this uniformity 

 of action is misleading, for it impresses one with the idea that any oil which 

 spreads over the surface is one which will have, when faMy tried, a nuclear 

 action ; and so it has the effect of prepossessing one unduly in favour of the 

 tension theory. As a matter of fact, however, I find that gasoline, an oil of 

 very feeble tension, too, spreads rapidly over such solution without pro- 

 ducing any solidifymg action, so it is not nuclear, thus breaking the 

 uniformity hitherto observed. 



On the other hand, however, we may take it for granted that in every 

 case where a nuclear action is exercised by an oU, there is more or less of a 

 surface spread of the oil accompanying such action. 



And now in pm'suance of the next part of my paper : it is relative to 

 the mode of action of solid nuclei in super-saturated solutions, and is merely 

 a description of certain results I have obtained thereon, while the subject of 

 liquid nuclei was under consideration. Two of these results are, I beheve, 

 novel to scientists (i5^os. 4 and 6). The others are got by repetitious of expe- 

 riments by investigations which have often led to results diverging more or 

 less from one another. They are recorded, not under the idea of settling a 

 question about which there is great debate, but only as a small and possibly 



47 



