M. Sanna-Solaro on the Artificial Production of Hail. 185 



of pressure results, to which the crust at length gives way. The 

 shock or impulse consequent on this determines a fresh act of 

 congelation and the formation of a new layer. The latter must 

 consist of two distinct portions ; one devoid of air-bubbles and 

 consequently transparent, the other opake by reason of the 

 intermingling of such bubbles within it. The same pheno- 

 menon recurs in each successive congelation. If the hailstones 

 fall on the ground before their congelation is complete, their 

 centre may be found liquid, or contain at the same time air- 

 bubbles, water, and needles of ice. These last will occur when 

 the internal liquid has cooled down very slowly, for threads of 

 ice are only seen in water under such circumstances. If the 

 congelation be suddenly brought about, the nucleus will have 

 the whiteness of snow. If the cold which acted on the mass of 

 water be very intense, the crust will be more solid and thicker : 

 the internal pressure caused by the dilatation of the liquid may 

 be so augmented that it will cause the bursting of the hailstones 

 into fragments, especially at the moment when congelation is ac- 

 complished. Hence it can be understood why hailstones fall 

 in a pyramidal form. 



I have frozen water in different quantities in perfectly trans- 

 parent caoutchouc capsules, and have obtained artificially all these 

 phenomena. Between the natural and the artificial hailstones 

 there is no difference except in the number of layers, which for 

 equal dimensions is greater in the latter. This proves that the 

 cold which produces the natural hailstones is much more intense 

 than that,°viz. — 17° C, by the aid of which I have formed them 

 artificially. It- is true indeed that, other conditions remaining 

 the same, the number of layers decrease if the portion of water 

 sought to be congealed be strongly agitated; but in that case 

 the opake zones present a fibrous or radiated structure. In my 

 experiments I have compared artificial hailstones with natural 

 ones of equal dimensions which did not present this structure, 

 and have always found the number of zones greater in the arti- 

 ficial. Besides, it is evident that the congelation of the crust 

 must take place almost instantaneously, for otherwise we should 

 never meet with ether than small hailstones. But in artificial 

 congelation more or less time is needed to obtain an envelope 

 of ice capable of containing the unfrozen water ; and therefore 

 it follows that the cold which seizes on the collections of water 

 in the atmosphere must, as we have said, be much greater than 

 -17° C. 



In artificial congelation a phenomenon transpires which eluci- 

 dates a similar occurrence which often accompanies the forma- 

 tion of hail. After the liquid has remained in the freezing- 

 mixture for a few minutes, slight cracking or detonating noises 



