Footnote 591: Theophr. _Phys. Op._ fr. 3 (_Dox._ p. 477); _ap._ Simpl. _Phys._ p. 25, 21 (R. P. 166 b). The function of Love is to produce union; that of Strife, to break it up again. Aristotle, however, rightly points out that in another sense it is Love that divides and Strife that unites. When the Sphere is broken up by Strife, the result is that all the Fire, for instance, which was contained in it comes together and becomes one; and again, when the elements are brought together once more by Love, the mass of each is divided. In another place, he says that, while Strife is assumed as the cause of destruction, and does, in fact, destroy the Sphere, it really gives birth to everything else in so doing.[592] It follows that we must carefully distinguish between the Love of Empedokles and that “attraction of like for like” to which he also attributed an important part in the formation of the world. The latter is not an element distinct from the others; it depends, we shall see, on the proper nature of each element, and is only able to take effect when Strife divides the Sphere. Love, on the contrary, is something that comes from outside and produces an attraction of _unlikes_. Footnote 592: Arist. _Met._ Α, 4. 985 a 21; Γ, 4. 1000 a 24; b 9 (R. P. 166 i). [Sidenote: Mixture and separation.] 109. But, when Strife has once separated the elements, what is it that determines the direction of their motion? Empedokles seems to have given no further explanation than that each was “running” in a certain direction (fr. 53). Plato severely condemns this in the _Laws_,[593] on the ground that no room is thus left for design. Aristotle also blames him for giving no account of the Chance to which he ascribed so much importance. Nor is the Necessity, of which he also spoke, further explained.[594] Strife enters into the Sphere at a certain time in virtue of Necessity, or “the mighty oath” (fr. 30); but we are left in the dark as to the origin of this. Footnote 593: Plato, _Laws_, x. 889 b. The reference is not to Empedokles exclusively, but the language shows that Plato is thinking mainly of him. Footnote 594: Arist. _de Gen. Corr._ Β, 6. 334 a 1; _Phys._ Θ, 1. 252 a 5 (R. P. 166 k).