222 Miss C. A. Raisin. On certain Structures formed 



similarity, although, there it is the rise of gas hindered by material 

 instead of the lateral spread of air, causing the retreat of the water- 

 film which opposes it. Many peculiar forms assumed by concretions 

 may be similarly explained. 



The deposit of an opaline layer with cusped points around a cavity 

 might be caused by the uniform drying of a smooth, homogeneous 

 material. Modified by crystallisation, it might help in tbe formation 

 of agates and chalcedonic deposits.* 



Igneous rocks have formed from a molten mass which may gene- 

 rally be considered (as pointed out by Lagorio) like a solution of the 

 less fusible in the more fusible constituents, or in some cases may 

 consist actually of two magmas imperfectly mixed. As it becomes 

 solid the two parts may mutually react in a manner analogous to 

 the phenomena already discussed. The forms shown in the secondary 

 silicification of certain rocks (e.g., cherts, silicified rhyolites, &c.) 

 perhaps afford another illustration. Though this process is not 

 exactly comparable with the deposits of sediment described above, 

 yet colloid silica might penetrate in different directions, and 

 encountering constant obstacles, give rise to an irregular, almost 

 felsitic structure. 



4thly. In the specimens which show cracking and expansion of 

 cracks to canalsf or bubbles, we may see a resemblance to other 

 forms in igneous rocks (fig. 4). Some of the stellate or cruciform 

 cracks remind us of the kind of contraction which, in a semi-solidified 

 pyromeride, develops fissures and inlets, afterwards filled by chal- 

 cedonic or other deposits. Further, the grouping of cracks and 

 bubbles in the drying film of mud, imitating roughly some of those 

 found within spherulites, would suggest that in certain of these a 

 crust may be formed before their complete development. This would 

 accord with the view which Professor Bonney has often advanced, in 

 conversation with me, for certain Boulay Bay pyromerides,J and 

 with the hypothesis which seemed to be suggested by certain Welsh 

 pyromerides and variolites : that the spherulite sometimes follows, 

 as it were, an initial formation of a nodule which has arisen by flow- 

 brecciation, or other process. 



Thus it seems possible that in several points the simple forms I 



* A comparison similar to that with the landscape marble might be made with 

 the process for the artificial formation of agates described by Messrs. J. I'Anson and 

 E. A. Pankhurst, ' Min. Mag.,' 1882, vol. 5, p. 34. 



f Compare with these the branched canals in gelatin described by Professor 

 Sollas, as due to the formation of ice spicules (' Trans. Roy. Irish Acad.,' 1890, 

 vol. 29, p. 427). 



J I believe that this view has the support of Mr. J. Parkinson, F.G-.S., but I 

 leave the sentence as it was written about two years ago (see ' Quart. Jo urn. Geol. 

 Soc.,' 1898, vol. 54, p. 101). 



* Quart. Joum. Geol, Soc./ 1889, vol. 45, p. 268 j and 1893, vol. 49, p. 152, 



