218 ME. J. PARKINSO]Sr ON THE HOLLOW SPHERT7LITES [May I9OI, 



endeavoured to obtain information concerning the latent heat of 

 liquefied felspar, but without success.^ 



In a cooling magma we may conclude that the temperature 

 gradually falls to the saturation -point when crj'stallization com- 

 mences, and the rise in temperature produced, if any, would do 

 little more than counteract the loss of heat due to the cooling of the 

 magma as a whole. 



If the part under consideration became supersaturated, some rise 

 of temperature, no doubt, would ensue, though to what extent is 

 doubtful. Supposing, however, heat to be produced by crystalliza- 

 tion in a supersaturated solution, one deduction must be made for 

 leakage into the surrounding rock, a second to counterbalance the 

 fall of temperature in the whole mass, and it is only the balance 

 which can be devoted to ' lowering the saturation of the surrounding 

 mother-liquor.' 



We find, then, some probability, at least, that the means which 

 Prof. Iddings suggests are inadequate to the required end. In 

 my opinion, that author's earlier work is more nearly in accord 

 with the facts, and it is with this that the following remarks 

 closely agree. No doubt can exist that when crystallization com- 

 mences we have to deal with a hydrous patch in the magma, and 

 that in this, as the temperature falls, anhydrous minerals develop. 

 It is, therefore, clear that the remaining liquid — no doubt in a very 

 viscous state — would become more hydrous, and the occluded vapour 

 would be pushed away from the crystallizing zone, though part 

 would be entangled in it and produce the characteristic porous struc- 

 ture. We have at this stage a comparatively solid portion succeeded 

 by a vapour-laden belt, followed in turn by another area in which 

 solidification is commencing. 



The solid state of the part crystallized, and the viscous state of 

 that in the act of crystallization, would prevent diffusion ; and the 

 process would be continued, until the whole of the hydrous patch 

 presented the characteristic concentric structure. 



[Part II — Great Britain.] 



YI. BouLAT Bay (Noethern Jersey). 



These peculiarities of structure in the spherulites and lithophysaB 

 of the Yellowstone can often be paralleled in those at Boulay Bay 

 and elsewhere in Great Britain, and thus afford great help in any 

 investigation concerning the latter. Not infrequently in the old 

 lavas we find concentric arcs composed o'f quartz, and roughly parallel 

 with the periphery of the nodule. Occasionally the outermost is 

 broad and well-defined, those nearer the centre progressively being 

 less well marked (PL YIII, fig. 1). This tendency to fade away is 

 shown (i) by the quartz-filled arcs becoming smaller ; (ii) by the 

 various portions of the arc becoming disjointed the one from the 

 other, so as to produce a segmented appearance ; and (iii) by the 



^ I wish to record my indebtedness, for invaluable help in this matter, to 

 Prof. T. Gr. Bonney and Principal E. Carey Foster. 



