HARKER : PRESIDENTIAL ADDRESS 
5 
problems, however, imitative experiment labours under the same 
disadvantage as mathematical analysis. Any concrete problem can 
be treated only in an arbitrarily simplified form, and among the 
conditions which cannot be realised in the laboratory may be some 
which in nature are of vital importance. Especially will this be the 
case where the time element enters. 
There is, however, another department of experimental geology 
in which we are justified in expecting results of very high value. I 
allude to the study of the conditions of formation and stability of 
different minerals, with the object of elucidating the mode of origin 
of igneous and other rocks. The artificial reproduction of many of the 
rock-forming minerals has engaged the attention of chemists, especially 
in France, during the last hundred years. Fouque and Michel-Levy 
succeeded even in imitating some of the simpler types of igneous rocks. 
These researches have furnished the petrologist with useful informa- 
tion, but it is information mostly of a very general kind. The laborious 
investigations now being carried out, more particularly in the Geo- 
physical Laboratory of the Carnegie Institution at Washington, are 
of a different order, systematic and precise to the highest degree 
attainable. Their chief object is to apply to the cystallization of 
igneous rock-magmas the methods which have proved so fruitful in 
other branches of physical chemistry. This necessitates working 
over a far wider range of temperatures than is usual in laboratory 
operations, and must sometimes include high-pressure work also. It 
involves too, other practical difficulties, arising especially from the 
slowness with which equilibrium is established in many of the trans- 
formations investigated. Owing perhaps to these obstacles, and 
partly, it may be, to the scarcity of enlightened millionaires — for 
expense is here a weighty consideration — research on these lines has 
not yet been widely taken up. Meanwhile it is scarcely too much to 
say that Dr. Day and his colleagues at Washington are already laying 
the foundations of an exact science of Petrogenesis. 
Concerning another American development of recent years, the 
quantitative classification of igneous rocks, I am constrained to speak 
in terms much less laudatory. We must all wish to see more of 
quantitative precision in this and other branches of geological science ; 
but it must be a precision depending on principles to be educed from 
the subject-matter itself, not arbitrarily imposed from outside. With 
increasing knowledge, from the experimental side as well as the 
descriptive, it may some day become possible to construct a logical 
