688 Ge Te VAIN RETESTED 
rocks to rich potassium rocks, from massive lavas to tuffaceous 
forms, from the freshest rocks to the most altered, from the old- 
est rocks to the newest, from the igneous to the aqueous rocks ; 
between all of the various forms of the sedimentary rocks. More- 
over, many of the important stages of these gradations have 
been noted. Still further, in one region the gradation from one 
kind of rock is in one direction, and in another region is in a 
different direction. For instance, here a granite grades into a 
diorite; there grades into a syenite. 
What shall be the criteria for naming rocks ? —These being the 
facts, the question arises, what is the most important consider- 
ation which shall determine whether or not a certain kind of 
rock shall be assigned a name. It appears to me clear that the 
most important consideration is the relative abundance of the rocks. 
We must have names for the common things. It is well known 
that certain of the multifarious kinds of rocks which have 
been named are more abundant than others. Probably the 
igneous rocks, which can be included under twenty names, com- 
prise nine tenths or more of the mass of the igneous rocks. 
It follows that a philosophical method of rock nomenclature 
involves a knowledge of the relative abundance of the different 
kinds of rocks. 
However, it is not meant to imply that abundance shall be 
the only consideration in the naming of rocks, but merely that it 
shall be a fundamental one. While abundance ought to be the 
first consideration, this idea must not be pushed to an extreme. 
All rocks which must be assigned names will not be found in 
equal abundance. Some rocks which do not have specific names 
assigned to them may be more abundant than some other rocks 
deserving of aname. For instance, olivine-gabbro may be more 
abundant than some of the leucite rocks to which names must be 
assigned. This but illustrates the well-known principle that 
great variations due to secondary factors shall have weight in 
proportion to their range, and that therefore they may have an 
important modifying influence upon the application of the pri- 
mary considerations. 
