or it is by stating the species or the
genus that we appropriately define any individual man; and we shall
make our definition more exact by stating the former than by stating
the latter. All other things that we state, such as that he is white,
that he runs, and so on, are irrelevant to the definition. Thus it is
just that these alone, apart from primary substances, should be called
substances.

Further, primary substances are most properly so called, because they
underlie and are the subjects of everything else. Now the same relation
that subsists between primary substance and everything else subsists
also between the species and the genus to which the primary substance
belongs, on the one hand, and every attribute which is not included
within these, on the other. For these are the subjects of all such. If
we call an individual man 'skilled in grammar', the predicate is
applicable also to the species and to the genus to which he belongs.
This law holds good in all cases.

It is a common characteristic of all substance that it is never present
in a subject. For primary substance is neither present in a subject nor
predicated of a subject; while, with regard to secondary substances, it
is clear from the following arguments (apart from others) that they are
not present in a subject. For 'man' is predicated of the individual
man, but is not present in any subject: for manhood is not present in
the individual man. In the same way, 'animal' is also predicated of the
individual man, but is not present in him. Again, when a thing is
present in a subject, though the name may quite well be applied to that
in which it is present, the definition cannot be applied. Yet of
secondary substances, not only the name, but also the definition,
applies to the subject: we should use both the definition of the
species and that of the genus with reference to the individual man.
Thus substance cannot be present in a subject.

Yet this is not peculiar to substance, for it is also the case that
differentiae cannot be present in subjects. The characteristics
'terrestrial' and 'two-footed' are predicated of the species 'man', but
not present in it. For they are not in man. Moreover, the definition of
the differentia may be predicated of that of which the differentia
itself is predicated. For instance, if the characteristic 'terrestrial'
is predicated of the species 'man', the definition also of that
characteristic may be used to form the predi