204 ON SIMILARITY OF FORM IN SNOW CRYSTALS. 



One main difference between these and the figures of snow, 

 is that they exhibit an entire want of angularity, and only 

 approximate, even when at their greatest perfection, to the 

 snow crystal just as it appears before finally dissolving. 

 Moreover, they have at all times a watery, uncertain, fluctuat- 

 ing outline and appearance generally, as though viewed 

 through a watery medium ; even the disc into which they 

 generally subside, presents the same indefinite wavy outline, and 

 the transparency of the parts is rather that of globules of water 

 than of crystals. 



These peculiai'ities attach to them at all times ; but the 

 appearance of the field differs very considerably, under, to 

 all appearance, similar circumstances of observation. Some- 

 times it is covered with minute globular bodies (crystals in 

 embryo), which quickly settle close to each other in clusters, 

 and never go beyond the figure of a well-defined star. Some- 

 times they may be seen in fewer numbers, swiftly travelling 

 over the field ; some single, but the greater number double, 

 for they share in this respect a peculiarity of the snow 

 crystals, but differ in their being united by a point of contact 

 common to the two, instead of being united by a slender axis 

 as in the crystal of snow. As they roll over, their conforma- 

 tion is distinctly visible ; and it is curious to watch the double 

 process of development as the star emerges from the ele- 

 mentary globule, which is every instant perfecting itself to the 

 complete figure, when it settles down and remains stationary 

 till the moment of its final departure. Sometimes they unite 

 in rows, sometimes in clusters so intermixed that the in- 

 dividual forms are not distinguishable. 



The greater number of tliese figures, indeed I may say it 

 is the rule with them, crystallize at an angle of 60° ; there 

 are, however, exceptions, and on one occasion I failed to 

 discover any other than eight-rayed crystals, arborescent and 

 differing only in regard to the additional radii. They have 

 all of them, with few exceptions, a nucleus generally circular 

 but sometimes star-like, with parallel and inner markings. 



I have made mention of a flattened disc into which 

 these bodies disperse. This disc is in itself a most interest- 

 ing study ; it is curiously intersected in different directions 

 by sinuous channels of different densities. It forms quickly ; 

 and when the room has been overheated I have seen the field 

 covered with discs with instantaneous rapidity, without ex- 

 hibiting a trace of stellar crystals. Sometimes the disc forms 

 round one-half of the crystal, dissolving the parts in contact 

 with itself. 



This is all that I have been able to collect respecting these 



