adopt any fresh terms as preferable, but should employ the usual expressions about it. Nor need we predicate anything else of time, as if this something else contained the same essence as is contained in the proper meaning of the word ' time ' (for this also is done by some). We must chiefly reflect upon that to which we attach this peculiar character of time, and by which we measure it. No further proof is required : we have only to reflect that we attach the attribute of time to days and nights and their parts, and likewise to feelings of pleasure and pain and to neutral states, to states of movement and states of rest, conceiving a peculiar accident of these to be this very characteristic which we express by the word ' time.' [He says this both in the second book " On Nature " and in the Larger Epitome.] (^After the foregoing we have next to consider that the worlds and every finite aggregate which bears a strong resemblance to things we commonly see have arisen out of the infinite. For all these, whether small or great, have been separated off from special conglomerations of atoms ; and all things are again dissolved, 6 some faster, some slower, some through the action of one set of causes, others through the action of another. [It is clear, then, that he also makes the worlds perishable, as their parts are subject to change. Elsewhere he says the earth is supported on the air.] *£And further, we must not suppose that the worlds have necessarily one and the same shape. [On the contrary, in the twelfth book "On Nature " he himself says that the shapes of the worlds differ, some being spherical, some oval, others again of shapes different