1. An important exception to this general rule are two papers by Balme, D. M., ‘Aristotle's biology was not essentialist’, Archiv für Geschichte der Philosophie 62 (1980) 1–12CrossRefGoogle Scholar, and ‘The snub’, Ancient Philosophy 4 (1984) 1–8CrossRefGoogle Scholar, now revised and reprinted in Gotthelf, A. and Lennox, J. G., (eds.), Philosophical Issues in Aristotle's Biology (1987) 291–312CrossRefGoogle Scholar.

2. It is a matter of indifference for my discussion what the order of composition of these works may have been.

3. See Frede, Michael, ‘Substance in Aristotle's Metaphysics’, in his Essays in Ancient Philosophy (1987) 72–80Google Scholar and his and Günther Patzig's edition, German translation and commentary on Metaphysics Z (forthcoming 1988).

4. See his Substance, Form and Psyche (1988).

5. Although, as these passages witness, Aristotle often refers to both the menstrual fluid and the semen as sperma, he also often (most notably in GA I, 17, when raising the question whether both the male and the female contribute something that then works to structure and form the embryo) uses this word more narrowly, to refer to what does do the work of structuring the embryo (and on his view not the female fluid, but only the male, does that.)

6. But only if there is a συμμετρία, a suitable balance or ‘symmetry’ between its heat and movements and the wetness and quantity of the catamenia that it works upon. Aristotle emphasizes the need for this ‘symmetry’ as early as I, 18.723a28–31: see also 729a16–19, 743a26–34, 772a10–22, 777b27–9 and especially 767a13–35, discussed below p. 24. As Aristotle often makes explicit in these passages (729a 17, 767a17–20, 772a11–12), this is a symmetry between an active factor in generation (the male) and a passive one (the material provided by the female). His talk of symmetry in this connection does not imply any kind of interaction between the two factors, in the sense of a joint working together by two independent but proportionately coordinated agents with a view to a common product.

7. He says also that the same principles explain monstrous births – offspring that are animals but not regular human beings at all. I do not discuss this extension of the theory.

8. In both these passages I take it that μεταβάλλειν is being used intransitively, as elsewhere in this context (766a16, 23–4, 24; 768a14). A survey of Aristotle's usage shows that where he does use μεταβάλλειν transitively the object of the verb is almost always the respect in which the thing in question changes, almost never some other person or thing on which it effects a change. (The only clear instances of this latter kind that I have found are at HA 592a15 and Poet. 1459b41.) So Aristotle's words here do not mean that the male fluid ‘changes the female fluid to the opposite or to extinction’.

9. Taken out of context ‘the movements of the female’ might, I think, be given any of three interpretations: (1) the movements belonging to the menstrual fluid, (2) the movements (belonging to the semen) that are such as to fashion a female (compare 768a32, ) (3) the movements (belonging to the semen) that correspond to those both actually and potentially present in the mother's menstrual fluid. I argue in the next two paragraphs against opting for interpretation (1) in this context. I prefer (3) to (2) because I understand Aristotle's purpose in these lines (to 768a21) to be to explain how the semen can be responsible for fashioning not just a female (as (2) would permit) but one resembling her mother or any of her mother's forebears – the movements characteristic of whom are of course in the mother's menstrual fluid either actually or potentially. But on either of these two interpretations the basic principle is the same: Aristotle attributes to the semen, as a potential movement of its own, a movement it is going to put actually into the embryo in fashioning it.

10. Since Aristotle only explicitly mentions movements in the catamenia while discussing the phenomenon he calls λύσις, which only operates on movements (1), he never has occasion to attribute to the female fluid movements (2), (3) and (4). On general grounds we can infer he must suppose these other movements are present too.

11. The French translation of P. Louis agrees with Peck at this point.

12. One might think, as Geoffrey Lloyd has suggested to me, of Aristotle's remarks at e.g. 737a22–34, 738b3–4, 740b18–20, to the effect that the material contributed by the female is already potentially, though of course not actually, all the bodily parts that the male will fashion it into. But these remarks hardly yield a sense in which the movements in the menstrual fluid, that are in fact Aristotle's ground for saying that it is potentially all those bodily parts, are not present in it actually, but only potentially; indeed, they imply precisely the contrary.

13. The referred to here as being in the catamenia must be the semen, and not the female fluid itself, for two reasons, (i) This sentence supplements the statement at 767b10–13 just before about what causes a female birth, by stating the cause of the contrasting male birth. Since the female birth was said there to be caused by a defect in the hotness of the semen, we need a reference here to the well-concoctedness, and so hotness, of the semen as what causes the male birth.(ii) In the following sentence Aristotle makes a remark about the equivalency of speaking in such contexts of either the semen (γονή) or the movement (in it), and the relevance of that remark here is heightened by, if it does not actually require, a reference to the semen in the first part of this sentence. Read that way, Aristotle does precisely begin by speaking at 767b15 of the well-concoctedness of the semen only to conclude by saying (bl7–18) what, when it is well concocted, the movement of the male will effect; the remark at b18–20 is then fully in place, as indicating that no gap has in fact been left open in the explanation just given between the semen (and its features) and the formative movements. I have been unable to find any exact parallel for the expression (referring either to the semen or to the female generative residue), but Aristotle does of course speak frequently enough of a mixture of the two residues when conception occurs (e.g. 728a29–30), and he does occasionally mention semen being in the catamenia under such circumstances (e.g. 727a17–18). So neither Aristotle's usage nor his general theoretical position throws up any obstacle to finding here the reference to the semen that the surrounding context requires.

14. Notice that the unexpressed dative with ἔνεισι at 768a11 (‘the movements in…eare present’) is to be supplied from the dative at 767b35–6, , which in fact this sentence partially repeats. Since σπέρμασιν at 767b36 is the semen (only), so must the intended reference to seminal fluid at 768a11 be. This is added confirmation, if any were needed, for my interpretation of this sentence. (The fact that at 768b4 and 6 the implied dative with ἔνεισι includes both the semen and the female residue is irrelevant to the interpretation of ἔνεισι at 768a11. Once movements in the female fluid are introduced into Aristotle's account, at 768a18–21, they obviously fall within the scope of the general principles about actual and potential movements first set out for the male fluid alone at 767b35–7 and 768a11–14, and he simply takes note of this fact at 768b4 and 6.)

15. I translate Drossaart Lulofs' text. But I wonder whether one ought, instead of excising with Drossaart Lulofs, or excising this together with with other editors, to retain the manuscript reading, changing only the second καὶ of a 18 to ὤσπερ, thus giving the sense: ‘And in this way too, on the side of the females as on that of the males…’. Or mss. might have resulted from a scribe's having been misled by at a 11 to repeat that construction here by altering ὤσπερ to καί.

16. Since it says less about what in or about the semen enables it to do the required work, the first conception may seem less satisfactory than the second. On the other hand, it may be a virtue of the first conception that its commitments here are less substantial. At any rate, one might feel some discomfort with the second conception's idea that the semen has some sort, however general, of physically realized underlying movements for resemblances to any of the potential mothers’ families. If that is the alternative, then perhaps the more non-committal first conception is philosophically and scientifically preferable after all.

17. It is noteworthy that though in the first words of this paragraph Aristotle offers to explain the slackening of movements in general (and so, among others the slackening of the movements in the catamenia mentioned at 768a 18–21) in fact he goes on to discuss only the slackening of the movements in the semen – these are the only movements that are active, of anything. This confirms the secondary rôle assigned in his theory to movements in the catamenia. It also fits in well with the attribution to the semen of potential movements somehow corresponding to them: in effect, in discussing only the slackening of the semen's movements he will have covered the whole range of the phenomena.

18. One should not object to this analogy on the ground that it compares the semen, which Aristotle treats as merely a tool used by the father (the actual ‘artisan’), to an artisan (the sculptor) rather than to his tools (chisel etc.). For as I have pointed out Aristotle is evidently thinking of the semen as a highly refined, fully programmed and self-starting tool not controlled by the father himself once it is set loose, and so much more like the sculptor himself than his chisel (which is presumably why he refers to the movements in the semen as themselves 768a16). It should also be borne in mind that Aristotle's doctrine in Gen. et Corr. I, 7 apparently includes reciprocal effects not just on the doctor's or the sculptor's tools, but also on these practitioners themselves: 324b3–13.

19. In the Hippocratic treatise On Generation I can find no clear reference to the first type of resemblance. All the emphasis, at any rate, is on ways the offspring takes after one or the other parent in features distinctive of and more or less special to that parent. See sects. 8 and 9.

20. Balme (1984), reprinted in Gotthelf and Lennox (1987) 292.

21. I mean only to be reporting Aristotle's view, not defending it. Since for most individual animals for most of their life-spans eye-colour remains the same, through growth, etc., it must continue to be true that the combination of the auxetic movements in the animal's blood and the availability of-material for adding to and maintaining the eye has the effect of preserving this colour. It is not clear that he has a good explanation for why this happens, if the form is not aiming at this result.

22. In revising this paper for publication I have been aided by comments of Myles Burnyeat, Nicholas Denyer, Geoffrey Lloyd, and Robert Wardy when it was read to the Society. I have been greatly helped also by further written comments of Prof. Lloyd and by his paper ‘Aristotle's zoology and his metaphysics: the status quaestionis’ presented to a conference at Oléron, France in July, 1987, and forthcoming in its proceedings. Finally, sharp questioning by Jonathan Barnes, David Charles and Lindsay Judson when I read the paper at Balliol in May 1988 led to significant improvements in sections V and VI.