in men, then, and in other kinds, as said before, such deficiency occurs sporadically, but the whole of the mule kind is sterile. the reason has not been rightly given by empedocles and democritus, of whom the former expresses himself obscurely, the latter more intelligibly. for they offer their demonstration in the case of all these animals alike which unite against their affinities. democritus says that the genital passages of mules are spoilt in the mother’s uterus because the animals from the first are not produced from parents of the same kind. but we find that though this is so with other animals they are none the less able to generate; yet, if this were the reason, all others that unite in this manner ought to be barren. empedocles assigns as his reason that the mixture of the ‘seeds’ becomes dense, each of the two seminal fluids out of which it is made being soft, for the hollows in each fit into the densities of the other, and in such cases a hard substance is formed out of soft ones, like bronze mingled with tin. now he does not give the correct reason in the case of bronze and tin —(we have spoken of them in the problems)— nor, to take general ground, does he take his principles from the intelligible. how do the ‘hollows’ and ‘solids’ fit into one another to make the mixing, e.g. in the case of wine and water? this saying is quite beyond us; for how we are to understand the ‘hollows’ of the wine and water is too far beyond our perception. again, when, as a matter of fact, horse is born of horse, ass of ass, and mule of horse and ass in two ways according as the parents are stallion and she-ass or jackass and mare, why in the last case does there result something so ‘dense’ that the offspring is sterile, whereas the offspring of male and female horse, male and female ass, is not sterile? and yet the generative fluid of the male and female horse is soft. but both sexes of the horse cross with both sexes of the ass, and the offspring of both crosses are barren, according to empedocles, because from both is produced something ‘dense’, the ‘seeds’ being ‘soft’. if so, the offspring of stallion and mare ought also to be sterile. if one of them alone united with the ass, it might be said that the cause of the mule’s being unable to generate was the unlikeness of that one to the generative fluid of the ass; but, as it is, whatever be the character of that generative fluid with which it unites in the ass, such it is also in the animal of its own kind. then, again, the argument is intended to apply to both male and female mules alike, but the male does generate at seven years of age, it is said; it is the female alone that is entirely sterile, and even she is so only because she does not complete the development of the embryo, for a female mule has been known to conceive.

perhaps an abstract proof might appear to be more plausible than those already given; i call it abstract because the more general it is the further is it removed from the special principles involved. it runs somewhat as follows. from male and female of the same species there are born in course of nature male and female of the same species as the parents, e.g. male and female puppies from male and female dog. from parents of different species is born a young one different in species; thus if a dog is different from a lion, the offspring of male dog and lioness or of lion and bitch will be different from both parents. if this is so, then since (1) mules are produced of both sexes and are not different in species from one another, and (2) a mule is born of horse and ass and these are different in species from mules, it is impossible that anything should be produced from mules. for (1) another kind cannot be, because the product of male and female of the same species is also of the same species, and (2) a mule cannot be, because that is the product of horse and ass which are different in form, [and it was laid down that from parents different in form is born a different animal]. now this theory is too general and empty. for all theories not based on the special principles involved are empty; they only appear to be connected with the facts without being so really. as geometrical arguments must start from geometrical principles, so it is with the others; that which is empty may seem to be something, but is really nothing. now the basis of this particular theory is not true, for many animals of different species are fertile with one another, as was said before. so we must not inquire into questions of natural science in this fashion any more than any other questions; we shall be more likely to find the reason by considering the facts peculiar to the two kinds concerned, horse and ass. in the first place, each of them, if mated with its own kind, bears only one young one; secondly, the females are not always able to conceive from the male (wherefore breeders put the horse to the mare again at intervals). indeed, both the mare is deficient in catamenia, discharging less than any other quadruped, and the she-ass does not admit the impregnation, but ejects the semen with her urine, wherefore men follow flogging her after intercourse. again the ass is an animal of cold nature, and so is not wont to be produced in wintry regions because it cannot bear cold, as in scythia and the neighbouring country and among the celts beyond iberia, for this country also is cold. for this cause they do not put the jackasses to the females at the equinox, as they do with horses, but about the summer solstice, in order that the ass-foals may be born in a warm season, for the mothers bear at the same season as that in which they are impregnated, the period of gestation in both horse and ass being one year. the animal, then, being, as has been said of such a cold nature, its semen also must be cold. a proof of this is that if a horse mount a female already impregnated by an ass he does not destroy the impregnation of the ass, but if the ass be the second to mount her he does destroy that of the horse because of the coldness of his own semen. when, therefore, they unite with each other, the generative elements are preserved by the heat of the one of them, that contributed by the horse being the hotter; for in the ass both the semen of the male and the material contributed by the female are cold, and those of the horse, in both sexes, are hotter. now when either hot is added to cold or cold to hot so as to mix, the result is that the embryo itself arising from these is preserved and thus these animals are fertile when crossed with one another, but the animal produced by them is no longer fertile but unable to produce perfect offspring.

and in general each of these animals naturally tends towards sterility. the ass has all the disadvantages already mentioned, and if it should not begin to generate after the first shedding of teeth, it no longer generates at all; so near is the constitution of the ass to being sterile. the horse is much the same; it tends naturally towards sterility, and to make it entirely so it is only necessary that its generative secretion should become colder; now this is what happens to it when mixed with the corresponding secretion of the ass. the ass in like manner comes very near generating a sterile animal when mated with its own species. thus when the difficulty of a cross contrary to nature is added, (when too even in the other case when united with their own species they with difficulty produce a single young one), the result of the cross, being still more sterile and contrary to nature, will need nothing further to make it sterile, but will be so of necessity.

we find also that the bodies of female mules grow large because the matter which is secreted in other animals to form the catamenia is diverted to growth. but since the period of gestation in such animals is a year, the mule must not only conceive, if she is to be fertile, but must also nourish the embryo till birth, and this is impossible if there are no catamenia. but there are none in the mule; the useless part of the nutriment is discharged with the excretion from the bladder — this is why male mules do not smell to the pudenda of the females, as do the other solid-hoofed ungulates, but only to the evacuation itself — and the rest of the nutriment is used up to increase the size of the body. hence it is sometimes possible for the female to conceive, as has been known to happen before now, but it is impossible for her to complete the process of nourishing the embryo and bringing it to birth.

the male, again, may sometimes generate, both because the male sex is naturally hotter than the female and because it does not contribute any material substance to the mixture. the result in such cases is a ‘ginnus’, that is to say, a dwarf mule; for ‘ginni’ are produced also from the crossing of horse and ass when the embryo is diseased in the uterus. the ginnus is in fact like the so-called ‘metachoera’ in swine, for a ‘metachoerum’ also is a pig injured in the uterus; this may happen to any pig. the origin of human dwarfs is similar, for these also have their parts and their whole development injured during gestation, and resemble ginni and metachoera.