6 
HARKER : PRESIDENTIAL ADDRESS 
classification of igneous rocks, in which the numerical element will 
enter in the form of eutectic ratios or in some other way prescribed by 
Nature herself ; but that is at present only an aspiration. Very 
different is the plan to which I have referred. Starting from a list 
of standard minerals, itself arbitrary and artificial, the authors have 
built up a classification depending on the relative proportions in 
which these minerals enter, or are assumed to enter, into the com- 
position of the rocks. The dividing lines are not based upon the study 
and comparison of actual rocks, but merely on arithmetical symmetry, 
the system being, to all appearance, invented a priori. We are re- 
minded of Macaulay's trenchant criticism of James Mill's Essay : 
" an elaborate treatise on Government, from which, but for two or 
three passing allusions, it would not appear that the author was aware 
that any governments actually existed among men." We find ourselves 
again among the pre-Baconian philosophers. 
I have selected for notice these two Trans-Atlantic attempts to 
introduce quantitative considerations into Petrology, because they are 
instructive as exemplifying, in my opinion at least, how to do it and 
how not to do it. It would probably exhaust your patience, were I 
to prolong these rather rambling remarks by citing numerous other 
illustrations of the same general purport. I will therefore devote the 
rest of the time at my disposal to the consideration of a single problem, 
that of Geological Chronology. 
Of all geological questions involving the numerical element, none 
has been more frequently canvassed than this, and none has excited 
more general interest. Since, moreover, it introduces several points 
germane to my subject, a brief glance at its history and present state 
will not be wasted. I suppose it has happened to most of us, when 
relating how in past times the mammoth roamed the plains of Holder- 
ness, or how coral-reefs once flourished where the Craven hills now 
stand, to be met by the inquiry : How long ago was that ? The answer 
was perhaps to the effect that geology does not deal with the ordinary 
measures of time, but has its own system of chronology, not trans- 
latable into years and centuries. I must confess, however, to a sense of 
inadequacy in such a reply, and some sympathy with the lay-inquirer 
who is thus silenced but not satisfied. It seems a matter of reasonable 
regret that a science which deals with the history of past events should 
have no definite time-scale, by which those events could be ranged in 
a correct perspective. 
No such reflection, it is safe to say, disturbed the minds of the 
