Joly — On the Permanency of Frost-Marks. 157 



How the marks are caused, it is not hard to understand. If a 

 surface consisting of loose small particles, holding water in the 

 interstices, be exposed to a low temperature, certain of the more 

 prominent particles, exposing a capillary surface of water more 

 freely than their neighbours, become centres of crystallization, 

 from which crystallogenesis is propagated, the molecular forces at 

 work being sufficient to disturb the loose sand particles, so that 

 they shall take up a position accommodating to the form and 

 direction taken by the ice spicules. These spicules, or rays, 

 would, if forming freely, extend, indeed, ever as straight lines ; 

 but here, hampered by the jamming or fixity of occasional par- 

 ticles, they wander minutely, now diverted a little in one direction, 

 and again in another, so that the sharp definition of crystalline 

 shape becomes modified into a straggling growth, resembling the 

 radiate straggling of Oldhamia radiata. 



There is another conspicuous variety of Oldhamia, known as 

 Oldhamia antiqua. I traced, indeed, some marks remotely resembl- 

 ing this ; but, although we might a priori expect such a form to 

 occur, I have not succeeded in finding anything fairly resembling 

 it since, nor have I, in the few experiments made, succeeded in 

 reproducing it. These experiments consisted in washing out the 

 finer constituents of some earth, and exposing this, while saturated 

 with water, to frost. I also froze a slab artificially, by placing 

 immediately above it, in a well-padded box, a metal tray contain- 

 ing a freezing mixture : freezing was produced by radiation from 

 the surface of the mud to the bottom of the tray, which was 

 coated with lamp-black. In this way, it was hoped, the conditions 

 obtaining in nature would be preserved. In general, marks more 

 or less resembling the Oldhamia radiata 1 were easily obtained, but 

 the Oldhamia antiqua could hardly be said to be reproduced. I 

 said that we might expect a different result. This will appear if 

 we consider the simple arrangement of such marking — a zigzag of 

 nearly straight lines, with tufts at the bends or meeting-points. 

 How such an arrangement might occur in the case of fine sand, 

 interspersed with larger particles, is quite conceivable. Finally, 

 anyone who has observed closely the symmetrical forms of frost 



1 This is by far the more common variety. 



