110 Dr. A.Waller's Microscopic Observations on Hail. 



turn again to a solid condition. The globular forms of solid 

 particles were in consequence very numerous. The cause of 

 this difference was, that in one case the quantity of moisture 

 in the air was small, the deposition gradual, and the transi- 

 tions of temperature from one extreme to the other very small ; 

 in the other, the moisture was abundant, the temperature of 

 the air greatly lowered, and the crystalline action rapid. We 

 find the same difference between the crystals obtained from a 

 concentrated solution in a liquid exposed to a rapid decrease 

 of temperature, and those which are formed in a solution less 

 concentrated and progressing more gently. A still greater 

 analogy is found in the crystals of naphthaline, iodine, cam- 

 phor, and other substances volatile at the common tempera- 

 ture, when allowed to crystallize in empty bottles without the 

 application of artificial heat. When either of these bodies is 

 driven off* in a state of vapour, the fumes that are formed 

 when condensed on glass, present the form of either minute 

 globules or of crystals, according to the temperature in which 

 they are exposed. When the fumes are given off' from either 

 of these substances in a state of ebullition, it is found that the 

 globules are of much larger dimensions than any which are 

 formed at a temperature below the boiling-point. On a 

 former occasion I have fully demonstrated that such is like- 

 wise the case with regard to the fumes of water. 1 will not 

 attempt at present to offer any explanation of this fact, which 

 is so invariably the case, that at any time, by the microscopic 

 inspection of the globules, it is easy to decide whether they are 

 formed from boiling water or at a temperature below. 



The above observations confirm what we might have al- 

 ready anticipated, that the production of solid globular parti- 

 cles arises from the same cause as that which produces con- 

 fused crystallization ; and that in fact it is simply confused 

 crystallization of separate particles. The most efficient cause 

 of this is rapid congelation, which apparently does not allow 

 the constituent particles sufficient time to group themselves in 

 geometrical forms, but obliges them to assume that which 

 their molecular attraction tends to give them. Between the 

 two extremes, of a globular form and a perfect hexagonal 

 prism, frequent shapes which the crystals assume in frost, 

 we have various intermediate forms which may be considered 

 as the resultants of the crystalline and molecular powers. 

 Although by our experiments we may observe the circum- 

 stances in which these globules are formed, it would be im- 

 possible to ascertain by their means, with any precision, the 

 condition of temperature requisite in one instance to form 

 small crystals, and in the other minute globules. But in 



