Our discussion so far results in the establishing of a sharp antithesis between the ‘ego-instincts’ and the sexual instincts, the former impelling towards death and the latter towards the preservation of life, a result which we ourselves must surely find in many respects far from adequate. Further, only for the former can we properly claim the conservative—or, better, regressive—character corresponding to a repetition-compulsion. For according to our hypothesis the ego-instincts spring from the vitalising of inanimate matter, and have as their aim the reinstatement of lifelessness. As to the sexual instincts on the other hand: it is obvious that they reproduce primitive states of the living being, but the aim they strive for by every means is the union of two germ cells which are specifically differentiated. If this union does not take place, then the germ cell dies like all other elements of the multicellular organism. Only on this condition can the sexual function prolong life and lend it the semblance of immortality. Of what important happening then in the process of development of the living substance is sexual reproduction, or its forerunner, the copulation of two individual protozoa, the repetition? That question we do not know how to answer, and therefore we should feel relieved if the whole structure of our arguments were to prove erroneous. The opposition of ego- (or death-) instincts and sexual (life-) instincts would then disappear, and the repetition-compulsion would thereupon also lose the significance we have attributed to it. Let us turn back therefore to one of the assumptions we interpolated, in the expectation that it will permit of exact refutation. We built up further conclusions on the basis of the assumption that all life must die from internal causes. We made this assumption so light-heartedly because it does not seem to us to be one. We are accustomed so to think, and every poet encourages us in the idea. Perhaps we have resolved so to think because there lies a certain consolation in this belief. If man must himself die, after first losing his most beloved ones by death, he would prefer that his life be forfeit to an inexorable law of nature, the sublime Ἀνάγκη, than to a mere accident which perhaps could have been in some way avoided. But perhaps this belief in the incidence of death as the necessary consequence of an inner law of being is also only one of those illusions that we have fashioned for ourselves ‘so as to endure the burden of existence’. It is certainly not a primordial belief: the idea of a ‘natural death’ is alien to primitive races; they ascribe every death occurring among themselves to the influence of an enemy or an evil spirit. So let us not neglect to turn to biological science to test the belief.