498 G. R. T. ROSS : and say species a is not B and hence it is C. The assump- tion came in, as we saw, in ranking species a under A ; the Statpeo-t? is presupposed, but apart from this there is genuine inference. What we have done is to limit our discussion to the genus or universe of discourse marked by A, but within this there is a genuine synthesis of content, a with c, and it is a mediated synthesis ; hence it is, so far, a genuine inference. The divisive disjunction is used chiefly perhaps in classi- fication. Now classification depends upon the mutual ex- clusiveness of species, and the distinction of species in a genus depends, I shall try to show, on certain indemonstrable negations. If our disjunction is used in classification then the modus ponendo tollens is valid (though a priori and pro- ducing no conclusion in itself valuable), but a divisive judg- ment need not be one of classification, and hence may not permit of the employment of the modus ponendo tollens. I
- shall give an example of this latter class and try to show
how such a judgment passes into a classificatory disjunc- tion. We may say that such and such a kind of fish is found either in streams or in fresh water lakes. This is more than the conjunctive proposition that the kind of fish in question is found both in streams and in lakes, for we have denied that it is found elsewhere (e.g., the ocean) than in the two places mentioned. But our assertion does not deny that the very same kind of fish is found in both. In this case we may very well understand the reverse, nor does our state- ment imply that the same individual may not live now in a lake at another time in a stream. Now, we may find that those individuals found in the streams have peculiarities distinct from the characteristics of the denizens of lakes. We may find that living in the different localities may accompany or may indeed have helped to produce differences of behaviour and structure in each case, and hence we may be able to divide our genus into two species or at least varieties of fish, according to the amount of difference between the two classes. The moment that we understand that the attributes river- and lake-inhabiting mark two different classes, then to the exhaustive judgment All not B is C we are able to add the exclusive statement No B is C. But this is a merely a priori judgment drawn from the general logical principle that co-ordinate species are distinct, which again is a corollary of the law that the individual cannot be predicated, for co-ordinate species are related to each other as individuals.
