INTRODUCTION.

21

The method may be thus briefly indicated. All the
known facts, belonging characteristically to the class of
subjects which have to be arranged, are separated into
groups, each group containing those subjects which
have a manifest or demonstrable relation to each other.
Each of these groups is found on investigation to have
one primal or causal fact antecedent to all others belonging
 to that group. Now when the causal facts thus
isolated are compared together, and their relations carefully
 studied, it becomes apparent that certain of them are
particularly related, and thus a second series of groups
related to each other, fewer in number ana of a more
primal character than the first, can be formed. Each of the
groups belonging to the second series has its causal fact or
general antecedent. These antecedents being compared
together, as before, are found to admit of likewise grouping
 into a third series, each member of which contains
facts having particular and definite relations to each
other.

A second generalization has, therefore, been accomplished
 ; the number of groups has diminished, ‘whilst
their primal character has increased. By the continued
application and pursuance of this method, in the like
manner, it is evident that eventually the actual basis, or
general antecedent fact, of the divisional science itself
will become known.
This method may be described as the systematic
investigation of the manner in which a characteristic
section or division of compounded knowledge, is compounded.
