502 Sir David Brewster on the Existence of Crystals 



came of it while a part of the space which it occupied was 

 appropriated by the bubble B, and the addition to the bubble 

 A? An accidental circumstance enables me to answer this 

 question, which would have been otherwise a very perplexing 

 one. Having applied too strong a heat to the specimen, the 

 bubble A threw off beside B two or three smaller ones, which 

 moved along the upper edge AE. My attention having been 

 thus directed to this part of the specimen, I was surprised to 

 observe a great number of capillary lines or pipes PQ, rising 

 from the edge AE of the cavity, and into which the fluid was 

 forcing itself, oscillating in these minute tubes like the mercury 

 in a barometer, and sometimes splitting the laminae between 

 them. The force of cohesion, thus overcome by the expan- 

 sive efforts of the fluid, predominated over the capillary attrac- 

 tion of the tubes and surfaces, and pressed back all the fluid 

 into the cavity, when the body of fluid had contracted in 

 cooling. 



If we now consider the body which occupies the vacuity A 

 as a gas, and, consequently, the other bubble B as the same, 

 it follows that the whole of the gas in B was absorbed by the 

 fluid while cooling, and again given out by an increase of tem- 

 perature. The gas, when in the act of being discharged, took 

 its course to the locality of the speck at B, and to the bubble 

 A ; but to the bubble A alone when the speck had disap- 

 peared. 



Upon repeating these observations the cavity burst ; and I 

 have now before me its two halves, forming its upper and its 

 under surface. The portion of" the cavity at A has the same 

 depth as the portion below mno, all the restof the cavity being 

 much shallower. There was a fine doubly refracting crystal 

 at MN, which polarized the blue of the second order; and its 

 outline is still left on the cavity. There was a sort of crystal- 

 line powder disseminated round MN to a considerable di- 

 stance, and the roof of the bubble B, when the roof of the 

 cavity was entire, was always mottled with this powder. 



In a former paper, I have distinguished vapour cavities 

 from common cavities, by the manner in which the vacuity in 

 the expansible fluid disappears. In the one case, the vacuity 

 gradually enlarges by the degradation, as it were, of its mar- 

 gin, as the fluid passes into vapour ; in the other, the vacuity 

 gradually diminishes till it disappears. I have since found 

 cavities of an intermediate character, in which the vacuity, on 

 the first application of heat, diminishes, and then, when it has 

 contracted to a certain size, it begins to expand ; and its 

 margin becoming thinner and thinner, it finally passes into 

 vapour. 



