PBEFACE. 



work was commenced in February, 1886, and completed 

 in March, 1888. One hundred and sixty-four pages and 

 twenty plates were issued to subscribers in monthly parts. The 

 issue then ceased, owing to the failure of the publishers, and I 

 was compelled to take the work into my own hands in order to 

 finish it. 



It was my original intention to treat of the sedimentary 

 rocks and crystalline schists in the same volume, but when I 

 approached the consideration of these rocks I found that the space 

 available was far too small for the purpose. I accordingly 

 determined, notwithstanding the fact that several plates had 

 been prepared, to postpone all detailed reference to these rocks, 

 for the present, in the hope that I might be able to return to the 

 subject on some future occasion. The plates that have been 

 prepared are, however, issued with the present volume. 



As regards the classification of rocks I am sorry to say that 

 increasing knowledge has not tended to bring about any clearness 

 of view. The more rocks are studied the less they seem to me 

 to adapt themselves to any classification at all comparable in 

 definiteness with the classifications of organic bodies and mineral 

 substances. Rock-masses often vary so much in composition and 

 structure that any scheme of classification based on work done 

 in the laboratory is unsuitable for the expression of broad geolo- 

 gical facts. It is absolutely impossible to map the different 

 varieties recognized by modern petrographers. The conclusion 

 at which I have arrived is that the necessity for giving names to 

 rocks arises rather from work done in the field than from work 

 done in the laboratory. Rock-specimens are mineral-aggregates 

 and may be described as such. Rock-masses are integral portions 

 of the earth's crust and possess a certain amount of individuality 

 in virtue of their mode of occurrence. 



It will be understood from these remarks that I attach very 

 little importance to any of the rock-names used in the present 

 work. My object has been to describe the structure and composi- 

 tion of the rocks ; not to introduce new names or to add to the 

 confusion already existing by giving new definitions of old names. 



