68 PKOCEEDINGS OF THE GEOLOGICAL SOCIETY. 



mud to form such a shale, would reduce the thickness to at least 

 one sixth. Such a change would be quite adequate to develop a 

 fissile structure nearly in the plane of bedding, which, however, 

 could not be so perfect as the cleavage in the best slates, on 

 account of the ultimate particles being chiefly granules and not 

 laminae. 



Slate Rocks in general. 



In studying cleaved slate rocks it is sometimes most important 

 to prepare and examine sections cut in several different directions. 

 I strongly suspect that the neglect to do this is the reason why 

 some statements made by other writers on this subject differ 

 materially from my own. If the cleavage be well developed, it is 

 best to select specimens in which it is at right angles to the plane 

 of stratification, since then there is less fear of confounding 

 together the original and the superinduced structures. In any 

 case the angle of their intersection should be known. It is 

 desirable to prepare sections at right angles to the cleavage, both 

 in the line of its dip and in that of its strike, since in many cases 

 there is as much difference in the structure in these two directions 

 as between a slate with very perfect cleavage and one with a much 

 less perfect. It is, however, still more important to prepare also 

 sections in the plane of cleavage, since the general appearance of 

 a section thus cut may be so very different to that from one at right 

 angles to the cleavage, that any one might easily mistake it for a 

 section of a totally different rock. 



When, nearly thirty years ago, I commenced the study of the 

 microscopical structure of slates, especially in connexion with their 

 cleavage, I was struck with the remarkable difference between the 

 constitution of many so-called cZa^Z-slates and modern deposits of 

 true clay. I found that in some slates the relative amount of 

 very fine granular material analogous to kaolin was very small, 

 whilst that of mica, or some closely allied laminar mineral, was 

 very large. This does not occur in plates of considerable size, like 

 the mica in many stratified rocks, but mainly in the form of flakes 

 so minute that in the best Welsh slates their average size may 

 be taken at y-oVo" ^^ ^^ ^^^^ broad, and -^-^ of an inch thick, 

 whilst many are considerably less ; separately they are altogether 

 invisible to the naked eye, and at most only serve to give a silky 

 lustre to the rock when they lie chiefly in the plane of the 

 fracture. 



Since it is a jloint of much importance, it appears to me desirable 

 to state clearly what is the evidence in favour of these conclusions. 

 The structure is best seen in black slates, like some near Llan- 

 beris, since the immense number of red grains in purple slates 

 interferes with accurate observations. When a section of this 

 black slate, cut at right angles to the cleavage in the line of the 

 dip, is examined with a power of about 250 linear, we can clearly 

 see that the great bulk of the rock is made up of what look hke 

 transparent fibres, which, so far as the mere appearance of this 



