That which is the first movement of a thing-in the sense that it
supplies not "that for the sake of which" but the source of the
motion-is always together with that which is moved by it by "together"
I mean that there is nothing intermediate between them). This is
universally true wherever one thing is moved by another. And since
there are three kinds of motion, local, qualitative, and quantitative,
there must also be three kinds of movent, that which causes
locomotion, that which causes alteration, and that which causes
increase or decrease.

    Let us begin with locomotion, for this is the primary motion.
Everything that is in locomotion is moved either by itself or by
something else. In the case of things that are moved by themselves
it is evident that the moved and the movent are together: for they
contain within themselves their first movent, so that there is nothing
in between. The motion of things that are moved by something else must
proceed in one of four ways: for there are four kinds of locomotion
caused by something other than that which is in motion, viz.
pulling, pushing, carrying, and twirling. All forms of locomotion
are reducible to these. Thus pushing on is a form of pushing in
which that which is causing motion away from itself follows up that
which it pushes and continues to push it: pushing off occurs when
the movent does not follow up the thing that it has moved: throwing
when the movent causes a motion away from itself more violent than the
natural locomotion of the thing moved, which continues its course so
long as it is controlled by the motion imparted to it. Again,
pushing apart and pushing together are forms respectively of pushing
off and pulling: pushing apart is pushing off, which may be a motion
either away from the pusher or away from something else, while pushing
together is pulling, which may be a motion towards something else as
well as the puller. We may similarly classify all the varieties of
these last two, e.g. packing and combing: the former is a form of
pushing together, the latter a form of pushing apart. The same is true
of the other processes of combination and separation (they will all be
found to be forms of pushing apart or of pushing together), except
such as are involved in the processes of becoming and perishing. (At
same time it is evident that there is no other kind of motion but
combination and separation: for they may all be apportioned to one
or other of those already mentioned.) Again, inhaling is a form of
pulling, exhaling a form of pushing: and the same is true of
spitting and of all other motions that proceed through the body,
whether secretive or assimilative, the assimilative being forms of
pulling, the secretive of pushing off. All other kinds of locomotion
must be similarly reduced, for they all fall under one or other of our
four heads. And again, of these four, carrying and twirling are to
pulling and pushing. For carrying always follows one of the other
three methods, for that which is carried is in motion accidentally,
because it is in or upon something that is in motion, and that which
carries it is in doing so being either pulled or pushed or twirled;
thus carrying belongs to all the other three kinds of motion in
common. And twirling is a compound of pulling and pushing, for that
which is twirling a thing must be pulling one part of the thing and
pushing another part, since it impels one part away from itself and
another part towards itself. If, therefore, it can be shown that
that which is pushing and that which is pushing and pulling are
adjacent respectively to that which is being pushed and that which
is being pulled, it will be evident that in all locomotion there is
nothing intermediate between moved and movent. But the former fact
is clear even from the definitions of pushing and pulling, for pushing
is motion to something else from oneself or from something else, and
pulling is motion from something else to oneself or to something else,
when the motion of that which is pulling is quicker than the motion
that would separate from one another the two things that are
continuous: for it is this that causes one thing to be pulled on along
with the other. (It might indeed be thought that there is a form of
pulling that arises in another way: that wood, e.g. pulls fire in a
manner different from that described above. But it makes no difference
whether that which pulls is in motion or is stationary when it is
pulling: in the latter case it pulls to the place where it is, while
in the former it pulls to the place where it was.) Now it is
impossible to move anything either from oneself to something else or
something else to oneself without being in contact with it: it is
evident, therefore, that in all locomotion there is nothing
intermediate between moved and movent.

    Nor again is there anything intermediate between that which
undergoes and that which causes alteration: this can be proved by
induction: for in every case we find that the respective extremities
of that which causes and that which undergoes alteration are adjacent.
For our assumption is that things that are undergoing alteration are
altered in virtue of their being affected in respect of their
so-called affective qualities, since that which is of a certain
quality is altered in so far as it is sensible, and the
characteristics in which bodies differ from one another are sensible
characteristics: for every body differs from another in possessing a
greater or lesser number of sensible characteristics or in
possessing the same sensible characteristics in a greater or lesser
degree. But the alteration of that which undergoes alteration is
also caused by the above-mentioned characteristics, which are
affections of some particular underlying quality. Thus we say that a
thing is altered by becoming hot or sweet or thick or dry or white:
and we make these assertions alike of what is inanimate and of what is
animate, and further, where animate things are in question, we make
them both of the parts that have no power of sense-perception and of
the senses themselves. For in a way even the senses undergo
alteration, since the active sense is a motion through the body in the
course of which the sense is affected in a certain way. We see,
then, that the animate is capable of every kind of alteration of which
the inanimate is capable: but the inanimate is not capable of every
kind of alteration of which the animate is capable, since it is not
capable of alteration in respect of the senses: moreover the inanimate
is unconscious of being affected by alteration, whereas the animate is
conscious of it, though there is nothing to prevent the animate also
being unconscious of it when the process of the alteration does not
concern the senses. Since, then, the alteration of that which
undergoes alteration is caused by sensible things, in every case of
such alteration it is evident that the respective extremities of
that which causes and that which undergoes alteration are adjacent.
Thus the air is continuous with that which causes the alteration,
and the body that undergoes alteration is continuous with the air.
Again, the colour is continuous with the light and the light with
the sight. And the same is true of hearing and smelling: for the
primary movent in respect to the moved is the air. Similarly, in the
case of tasting, the flavour is adjacent to the sense of taste. And it
is just the same in the case of things that are inanimate and
incapable of sense-perception. Thus there can be nothing
intermediate between that which undergoes and that which causes
alteration.

    Nor, again, can there be anything intermediate between that which
suffers and that which causes increase: for the part of the latter
that starts the increase does so by becoming attached in such a way to
the former that the whole becomes one. Again, the decrease of that
which suffers decrease is caused by a part of the thing becoming
detached. So that which causes increase and that which causes decrease
must be continuous with that which suffers increase and that which
suffers decrease respectively: and if two things are continuous with
one another there can be nothing intermediate between them.

    It is evident, therefore, that between the extremities of the
moved and the movent that are respectively first and last in reference
to the moved there is nothing intermediate.