¹ Lloyd, G. E. R., Polarity and Analogy (Cambridge, 1966) 17 and 50Google Scholar. It is interesting to note the change in wording from Lloyd's article in JHS lxxxii (1962) 60 where he uses the word ‘apparently’ instead of ‘probably’. Other discussions on the problem of Parmenides' sexual theories within the last ten years include that of Guthrie, W. K. C., History of Greek Philosophy, vol. ii (Cambridge, 1965) 78 ffGoogle Scholar. and Tarán, L., Parmenides (Princeton, 1965) 263–6Google Scholar. Tarán indeed asserts (264, note 98) ‘sex, according to Parmenides, was determined by the female and not by the male’. Earlier work of importance in this field has been done by Lesky, E., Die Zeugungs- und Vererbungslehren der Antike und ihr Nachwirken, Akademie der Wissenschaften und der Literatur, Mainz, Abhandlungen der Geistes- und Sozialwissenschaftlichen Klasse, Jahrgang 1950, Nr. 19, 1272 ff.Google Scholar

² References to Censorinus, Aëtius, Lactantius, Caelius Aurelianus and Galen are based on the texts of Diels, H. and Kranz, W., Die Fragmente der Vorsokratiker (Berlin, 1964)Google Scholar and Diels, H., Doxographi Graeci (Berlin, 1879)Google Scholar. I have abbreviated the former to DK and the latter to Dox.

³ Guthrie, op. cit., 467, note 1, gives a summary of Lesky's views on this subject and the origin of the semen in general in Presocratic philosophy.

⁴ The views of Censorinus can be expressed as follows, with y as male (father and offspring), x as female, z as F¹ offspring and r, l as right and left parts of the father.

⁵ Tarán, op. cit., 264, note 98, claims that the text is wrong to have θήλεα after γίνεσθαι since this ‘contradicts our other sources’. He does not however follow Diels, , Dox., 194Google Scholar, and read but changes the order because certain other sources (e.g. Galen and Censorinus) ‘show that the sex, according to Parmenides, was determined by the female and not the male’. I prefer to reject Tarán's logic here (as will become clear) and retain the original reading.

⁶ γόνος might mean ‘child’ or ‘seed’ here, which makes deductions about sex differentiation hypothetical.

⁷ Aëtius of course may very well not be interested in giving credit for originality, but one can see no reason why Parmenides' name should occur after that of Anaxagoras, unless Aëtius was following the passage in Aristotle, GA 763b30 and the phrase ‘Anaxagoras and others of the physicists’; for a discussion of this point see below.

⁸ These first two points of Aëtius may be expressed as follows, using the same notation as for Censorinus but with r, l also meaning right/left of the womb:

(a) Sex differentiation

(b) Heredity

⁹ See Guthrie, op. cit., 78/9 for an interesting discussion on density and rarity for Parmenides.

¹⁰ On the scheme as already used, Lactantius' views are as follows (noting that y and x may apply either to the single semen with masculine and feminine characteristics or to two semina, one for the father and one for the mother):

(a) Heredity

(b) Sex differentiation

¹¹ DK 28A54; Diels, Dox., 194.

¹² Tarán, op. cit., 264, note 97 and 265, note 99, gives a fair summary of the various views on the fragment.

¹³ Tarán, op. cit., 172, translates ‘the power which is formed in the veins…’, i.e. makes informans intransitive. This is incomprehensible.

¹⁴ Tarán, op. cit., 263–5.

¹⁵ Peck, A. L., Aristotle's Generation of Animals (London, 1943) 373Google Scholar, note c, brackets the phrase and comments: ‘These words must be an interpolation, as they are inconsistent with the view just described’. If we analyse the paragraph in two sections schematically however we can see that there is no inconsistency, merely amplification—which might or might not be genuine:

(a)

(b)

¹⁶ Guthrie, op. cit., 77/79.

¹⁷ Grateful acknowledgement is due to my colleague G. D. Field and to the editorial board of JHS for many helpful suggestions. Needless to say all the faults of this paper are entirely my own responsibility.