AJfXIVERSARY ADDRESS OF THE PRESIDENT. 83 



It may also be important as preveiitiiig dissociation, especially of 

 water. 



Let us now examine the mode of crystal-building in a volcanic 

 glass, putting aside for the present the consideration of the crystals 

 which occur porphyritically ; because, as will be shown, there is no 

 reason to associate their formation with this last stage of consoli- 

 dation. A volcanic glass, Avhen molten, may be either homogeneous 

 throughout or not homogeneous. The former might produce a 

 homogeneous solid, of which a piece of window-glass would be a 

 perfect type ; the latter a glass streaky from the occurrence of 

 different substances, like various slags and very many glassy lavas. 

 This is obviously due to the imperfect mixture of two materials (how 

 mixed, matters not for our present purpose) of slightly different 

 chemical composition, the masses of which during motion are drawn 

 or " teazed " out into shreds. 



Considering for a while the former case only, we see that the 

 molten mass may solidify without marked separation of any of its 

 chemical constituents, though this is rare. Commonly, numerous 

 microliths are formed, and the history of these, if traced, throws 

 much light on the process of crystal-building. Por this purpose 

 no better examples can be found than some of the well-known 

 pitchstones of Arran. On examining a slide from one of these with 

 an objective of low power, we see that the clear glass of the rock 

 appears full of a minute spicular dust ; on applying a higher power 

 (say I" objective) the particles of this dust are seen to be very small 

 pale-green belonites, disseminated pretty uniformly and without 

 orientation. Taking another slide, we perceive a number of larger 

 belonites, and in parts of the same or in a third slide we find curiously 

 tufted groups of the belonites, or aggregations of the smaller on the 

 larger, like miniature spruce-fir trees. 



In'ow each one of these — larger belonites, tufted groups of all kinds 

 — will be surrounded by a lacuna of perfectly clear glass, while be- 

 yond that, there will be interspaces crowded, as above, with the 

 spicules. Moreover a closer examination of the larger belonites 

 will often show that they are compound in stiiicture, built up by 

 the laying side by side of the spicules ; and further that in the fir- 

 tree-like groups the branches, where they inosculate with the stem 

 (to use a simile), sometimes make with it at first a comparatively 

 small angle, and then stretch out more nearly at right angles, 

 exactly as we see the young branches start at an acute angle with 

 the upper part of a fir-stem, but afterwards drawn down by the 

 increasing weight of the bough (a botanical fact of which I may 

 remark, by the wa}-, many artists take no note). It appears, then 

 pretty clear that either the increasing viscosity of the surrounding 

 material, or the resistance of the tufts to which they were already 

 attached, prevented these spicules from being incorporated into the 

 main stem. Why in parts of the rock we have a uniform distribution 

 of the spicules, and why in others they are able to aggregate as 

 above, we cannot say ; but probably it is due to some very slight 

 irregularity — an almost infinitesimal difierence might suffice— in the 



