APPENDIX. 803 



to be met with at the commencement of the frost, and very 

 seldom perfect, being fractured in their descent. Towards the 

 conclusion of the frost, and when the severity of the weather 

 had been long continued, the prevailing character of these bo- 

 dies might be said to change : they were less often to be seen 

 in the intermediate stages of crystallization, and were of more 

 solid structure and richly incrusted than those of any other 

 period, and presented on investigation combinations of solid 

 figures cut into facets, which glistened with crystalline trans- 

 parency, according to the inflections of the light upon their 

 surfaces. 



" The observation of these bodies may not improbably assist 

 in the solution of higher problems ; at present we are greatly 

 in doubt respecting their origin, and I regret that I have been 

 able to throw so little light on this part of the subject, although 

 I had frequent facilities of observing the method of their 

 change from one figure to another, which was accomplished 

 with inconceivable rapidity and kaleidoscope movement very 

 beautiful to witness. 



" My method of observing these bodies was to receive them 

 upon a piece of plate-glass, either plain or coloured, which had 

 been previously exposed, and thus cooled down to several de- 

 grees below the temperature of the air. I was by this means 

 able to examine them in their progress towards dissolution 

 with much minuteness, and it was interesting to observe the 

 groups of prisms nearest the apex dissolve, in so doing thicken- 

 ing and elongating the spiculse which had served as axes to the 

 prisms. Beyond recording the forms of these bodies, I have 

 been able to determine very little that is satisfactory concern- 

 ing them. I came however to infer that the greater the cold 

 the greater their departure from the figure of the simple star, 

 and the further removed they were in their component parts 

 from any similitude to the common icicle or to the incrusta- 

 tions on the surface of water at its first congelation. 



" In the early period of the frost I traced much analogy be- 

 tween these figures and such incrustations, with the great ex- 

 ception, that the latter never assumed the stellar form, but 



