16 W. W. Watts— Per lit ic Structure. 



contain inclusions of magnetite, zircon, and apatite, but none of the 

 metallic-looking plates which are so common in this mineral; the 

 magnetite grains, which occur at times in the glass of the rock and 

 at times partly in the hypersthene and partly in the glass, Mr. Smeeth 

 regards as derived from the hypersthene by solution. The presence 

 of this mineral in so acid a rock, one which contains 75 - o per cent, 

 of silica, is a remarkable feature. The bulk anatysis shows that 

 more than twice as much potash as soda is present, but the analysis 

 of the matrix, after removal of the phenocrysts, proves that most of 

 the latter oxide is in the glass, of the former in the porphyritic 

 crystals. 1*3 per cent, of soda will be in the sanidine, and 1-9 of 

 lime in the albite. The albite occurs in the glass as straight and 

 forked microlites, the other constituents in the glass being crystal- 

 lites, mainly in the form of a central axis with two, three, or four 

 rows of globulites arranged along it. 



The quartz crystals are traversed by curved cracks, so like those 

 described by myself as perlitic that the author regards them as 

 identical, and it seems to me quite rightly. He remarks, however, 

 that "if 'perlitic' simply implies that a crack is curved and may 

 reasonably be inferred to be due to contraction on cooling, then the 

 majority of these cracks are undoubtedly perlitic," but adds later 

 that he hopes to show " that typical perlitic cracks present well- 

 marked characteristic features, which give them the right to be 

 considered as forming a distinct and definite rock-structure, and 

 which features are markedly different to those exhibited by the 

 cracks in the quartz." 



Mr. Smeeth then gives in some detail the method of producing 

 artificial perlites in balsam, and insists on the necessity of using 

 a ground glass surface in order to produce the cracks in perfection. 

 On watching the process under the microscope, when it is allowed 

 to take place slowly, a set of polygonal cracks is seen to make its 

 appearance first, "and then each individual perlite springs suddenly 

 into existence, producing the impression of a tremor in the little 

 mass of balsam. The higher the balsam is heated and the more 

 quickly it is cooled the smaller is the interval of time between the 

 two sets of cracks, until it becomes imperceptible to the eye." It 

 appears, though this is not clearly stated, that the crack begins at 

 the exterior and travels to the interior, an important point being 

 that the crack always tends to be a spiral. Artificial perlites being 

 produced on the flat, Mr. Sir^eth next inquires into the nature of 

 the crack-surface in the solid, and shows that the lines revealed by 

 thin sections are either spirals or closed curves resembling circles 

 touching one another tangentially. From this he infers that as a 

 coiled plane cannot possibly be such that all its sections shall be 

 spirals, the simplest case will be where the plane is coiled about a 

 stationary axis, and this case will become more complex when the 

 axis shifts in the course of the growth of the perlite. From what 

 will shortly be said it will be seen that sections of perlites conform 

 to a great extent to these ideals, and Mr. Smeeth's illustrations give 

 admirable examples. 



