Objection 1. It would seem that the semen is not produced from the surplus food, but from the substance of the begetter. For Damascene says (De Fide Orth. i, 8) that "generation is a work of nature, producing, from the substance of the begetter, that which is begotten." But that which is generated is produced from the semen. Therefore the semen is produced from the substance of the begetter.
Objection 2. Further, the son is like his father, in respect of that which he receives from him. But if the semen from which something is generated, is produced from the surplus food, a man would receive nothing from his grandfather and his ancestors in whom the food never existed. Therefore a man would not be more like to his grandfather or ancestors, than to any other men.
Objection 3. Further, the food of the generator is sometimes the flesh of cows, pigs and suchlike. If therefore, the semen were produced from surplus food, the man begotten of such semen would be more akin to the cow and the pig, than to his father or other relations.
Objection 4. Further, Augustine says (Gen. ad lit. x, 20) that we were in Adam "not only by seminal virtue, but also in the very substance of the body." But this would not be, if the semen were produced from surplus food. Therefore the semen is not produced therefrom.
On the contrary, The Philosopher proves in many ways (De Gener. Animal. i, 18) that "the semen is surplus food."
I answer that, This question depends in some way on what has been stated above (1; 118, 1). For if human nature has a virtue for the communication of its form to alien matter not only in another, but also in its own subject; it is clear that the food which at first is dissimilar, becomes at length similar through the form communicated to it. Now it belongs to the natural order that a thing should be reduced from potentiality to act gradually: hence in things generated we observe that at first each is imperfect and is afterwards perfected. But it is clear that the common is to the proper and determinate, as imperfect is to perfect: therefore we see that in the generation of an animal, the animal is generated first, then the man or the horse. So therefore food first of all receives a certain common virtue in regard to all the parts of the body, which virtue is subsequently determinate to this or that part.
Now it is not possible that the semen be a kind of solution from what is already transformed into the substance of the members. For this solution, if it does not retain the nature of the member it is taken from, it would no longer be of the nature of the begetter, and would be due to a process of corruption; and consequently it would not have the power of transforming something else into the likeness of that nature. But if it retained the nature of the member it is taken from, then, since it is limited to a certain part of the body, it would not have the power of moving towards (the production of) the whole nature, but only the nature of that part. Unless one were to say that the solution is taken from all the parts of the body, and that it retains the nature of each part. Thus the semen would be a small animal in act; and generation of animal from animal would be a mere division, as mud is generated from mud, and as animals which continue to live after being cut in two: which is inadmissible.
It remains to be said, therefore, that the semen is not something separated from what was before the actual whole; rather is it the whole, though potentially, having the power, derived from the soul of the begetter, to produce the whole body, as stated above (1; 108, 1). Now that which is in potentiality to the whole, is that which is generated from the food, before it is transformed into the substance of the members. Therefore the semen is taken from this. In this sense the nutritive power is said to serve the generative power: because what is transformed by the nutritive power is employed as semen by the generative power. A sign of this, according to the Philosopher, is that animals of great size, which require much food, have little semen in proportion to the size of their bodies, and generated seldom; in like manner fat men, and for the same reason.
Reply to Objection 1. Generation is from the substance of the begetter in animals and plants, inasmuch as the semen owes its virtue to the form of the begetter, and inasmuch as it is in potentiality to the substance.
Reply to Objection 2. The likeness of the begetter to the begotten is on account not of the matter, but of the form of the agent that generates its like. Wherefore in order for a man to be like his grandfather, there is no need that the corporeal seminal matter should have been in the grandfather; but that there be in the semen a virtue derived from the soul of the grandfather through the father. In like manner the third objection is answered. For kinship is not in relation to matter, but rather to the derivation of the forms.
Reply to Objection 4. These words of Augustine are not to be understood as though the immediate seminal virtue, or the corporeal substance from which this individual was formed were actually in Adam: but so that both were in Adam as in principle. For even the corporeal matter, which is supplied by the mother, and which he calls the corporeal substance, is originally derived from Adam: and likewise the active seminal power of the father, which is the immediate seminal virtue (in the production) of this man.
But Christ is said to have been in Adam according to the "corporeal substance," not according to the seminal virtue. Because the matter from which His Body was formed, and which was supplied by the Virgin Mother, was derived from Adam; whereas the active virtue was not derived from Adam, because His Body was not formed by the seminal virtue of a man, but by the operation of the Holy Ghost. For "such a birth was becoming to Him," [Hymn for Vespers at Christmas; Breviary, O. P.], WHO IS ABOVE ALL GOD FOR EVER BLESSED. Amen.
Source: Summa Theologica (New Advent)