January 9, 1902] 



NA TURE 



the forms originating in, and peculiar to, each of the various 

 cloud strata. These observations, and the data secured, 

 indicate that the temperature and the humidity of the air 

 at the earth's surface is a much less important factor than 

 is generally supposed in determining the form and size of 



humidity due to these ; the character of the storm, whether 

 local or general, and the portion of the storm region from which 

 the crystals come. To these must also be added the initial and 

 subsequent movement of the cry.stals within the clouds. If, as 

 must often be the case, the nuclear forms originating in the 



Fig. 6.— 1901, February 15. Wind no 



the crystals. We may easily con- 

 ceive this to be the case, because 

 at a given temperature, l\:c., at the 

 earth's surface, the temperature and 

 humidity of the air where the 

 crystals form might vary greatly, 

 one time from another, and would 

 depend largely upon the height of 

 the snow-producing clouds. The 

 height of these varies greatly at 

 different times, even when the 

 temperature at the earth's surface 

 remains the same. The data 

 secured have not revealed the great 

 mystery of the origin and cause of 

 the differences in the forms of the 

 nuclei ; why columnar forms pre- 

 dominate at one time, tabular 

 forms at another, or why both are 

 sometimes found associated to- 

 gether. Much has been learned, 

 however, of the conditions lending 



-;S, January 26. 



temperature i 



lower ascending clouds are carried 

 upward to much greater heights by 

 the strong ascending air currents, 

 which often occur within such 

 storms, until they become heavy 

 enough to fall back through them, 

 then the crystals will in all proba- 

 bility be greatly modified by passing 

 through atmospheric strata varying 

 greatly in density, temperature, 

 humidity, &c. That they are 

 greatly modified by these flights in 

 the clouds is clearly shown by the 

 interior structure of many of the 

 crystals outlining many of these 

 transitory states. Thus, crystals of 

 which the nuclear form was origin- 

 ally nearly perfectly hexagonal 

 sometimes become partly triangular 

 in outline, and vice versa. No. 8 

 is an example of such modifica- 

 tions. 



to modify their forms after the nuclear form is once organised. 

 These conditions are many, the chief among them being the 

 height, number and vertical dejith of the cloud strata and the 

 resultant variation in temperature, atmospheric pressure and 



NO. 1680, VOL. 65] 



Perhaps the most important facts of a general nature to be 

 gleaned from twenty years' study are these : 



(i) That the greater number of the more perfect and beautifu 

 tabular forms occur much more frequently in, and are confine 



