the same causes must be held responsible for the following groups of facts. (1) some children resemble their parents, while others do not; some being like the father and others like the mother, both in the body as a whole and in each part, male and female offspring resembling father and mother respectively rather than the other way about. (2) they resemble their parents more than remoter ancestors, and resemble those ancestors more than any chance individual. (3) some, though resembling none of their relations, yet do at any rate resemble a human being, but others are not even like a human being but a monstrosity. for even he who does not resemble his parents is already in a certain sense a monstrosity; for in these cases nature has in a way departed from the type. the first departure indeed is that the offspring should become female instead of male; this, however, is a natural necessity. (for the class of animals divided into sexes must be preserved, and as it is possible for the male sometimes not to prevail over the female in the mixture of the two elements, either through youth or age or some other such cause, it is necessary that animals should produce female young). and the monstrosity, though not necessary in regard of a final cause and an end, yet is necessary accidentally. as for the origin of it, we must look at it in this way. if the generative secretion in the catamenia is properly concocted, the movement imparted by the male will make the form of the embryo in the likeness of itself. (whether we say that it is the semen or this movement that makes each of the parts grow, makes no difference; nor again whether we say that it ‘makes them grow’ or ‘forms them from the beginning’, for the formula of the movement is the same in either case.) thus if this movement prevail, it will make the embryo male and not female, like the father and not like the mother; if it prevail not, the embryo is deficient in that faculty in which it has not prevailed. by ‘each faculty’ i mean this. that which generates is not only male but also a particular male, e.g. coriscus or socrates, and it is not only coriscus but also a man. in this way some of the characteristics of the father are more near to him, others more remote from him considered simply as a parent and not in reference to his accidental qualities (as for instance if the parent is a scholar or the neighbour of some particular person). now the peculiar and individual has always more force in generation than the more general and wider characteristics. coriscus is both a man and an animal, but his manhood is nearer to his individual existence than is his animalhood. in generation both the individual and the class are operative, but the individual is the more so of the two, for this is the only true existence. and the offspring is produced indeed of a certain quality, but also as an individual, and this latter is the true existence. therefore it is from the forces of all such existences that the efficient movements come which exist in the semen; potentially from remoter ancestors but in a higher degree and more nearly from the individual (and by the individual i mean e.g. coriscus or socrates). now since everything changes not into anything haphazard but into its opposite, therefore also that which is not prevailed over in generation must change and become the opposite, in respect of that particular force in which the paternal and efficient or moving element has not prevailed. if then it has not prevailed in so far as it is male, the offspring becomes female; if in so far as it is coriscus or socrates, the offspring does not resemble the father but the mother. for as ‘father’ and ‘mother’ are opposed as general terms, so also the individual father is opposed to the individual mother. the like applies also to the forces that come next in order, for the offspring always changes rather into the likeness of the nearer ancestor than the more remote, both in the paternal and in the maternal line.

some of the movements exist in the semen actually, others potentially; actually, those of the father and the general type, as man and animal; potentially those of the female and the remoter ancestors. thus the male and efficient principle, if it lose its own nature, changes to its opposites, but the movements which form the embryo change into those nearly connected with them; for instance, if the movement of the male parent be resolved, it changes by a very slight difference into that of his father, and in the next instance into that of his grandfather; and in this way not only in the male but also in the female line the movement of the female parent changes into that of her mother, and, if not into this, then into that of her grandmother; and similarly also with the more remote ancestors.

naturally then it is most likely that the characteristics of ‘male’ and of the individual father will go together, whether they prevail or are prevailed over. for the difference between them is small so that there is no difficulty in both concurring, for socrates is an individual man with certain characters. hence for the most part the male offspring resemble the father, and the female the mother. for in the latter case the loss of both characters takes place at once, and the change is into the two opposites; now is opposed to male, and the individual mother to the individual father.

but if the movement coming from the male principle prevails while that coming from the individual socrates does not, or vice versa, then the result is that male children are produced resembling the mother and female children resembling the father.

if again the movements be resolved, if the male character remain but the movement coming from the individual socrates be resolved into that of the father of socrates, the result will be a male child resembling its grandfather or some other of its more remote ancestors in the male line on the same principle. if the male principle be prevailed over, the child will be female and resembling most probably its mother, but, if the movement coming from the mother also be resolved, it will resemble its mother’s mother or the resemblance will be to some other of its more remote ancestors in the female line on the same principle.

the same applies also to the separate parts, for often some of these take after the father, and others after the mother, and yet others after some of the remoter ancestors. for, as has been often said already, some of the movements which form the parts exist in the semen actually and others potentially. we must grasp certain fundamental general principles, not only that just mentioned (that some of the movements exist potentially and others actually), but also two others, that if a character be prevailed over it changes into its opposite, and, if it be resolved, is resolved into the movement next allied to it — if less, into that which is near, if more, into that which is further removed. finally, the movements are so confused together that there is no resemblance to any of the family or kindred, but the only character that remains is that common to the race, i.e. it is a human being. the reason of this is that this is closely knit up with the individual characteristics; ‘human being’ is the general term, while socrates, the father, and the mother, whoever she may be, are individuals.

the reason why the movements are resolved is this. the agent is itself acted upon by that on which it acts; thus that which cuts is blunted by that which is cut by it, that which heats is cooled by that which is heated by it, and in general the moving or efficient cause (except in the case of the first cause of all) does itself receive some motion in return; e.g. what pushes is itself in a way pushed again and what crushes is itself crushed again. sometimes it is altogether more acted upon than is the thing on which it acts, so that what is heating or cooling something else is itself cooled or heated; sometimes having produced no effect, sometimes less than it has itself received. (this question has been treated in the special discussion of action and reaction, where it is laid down in what classes of things action and reaction exist.) now that which is acted on escapes and is not mastered by the semen, either through deficiency of power in the concocting and moving agent or because what should be concocted and formed into distinct parts is too cold and in too great quantity. thus the moving agent, mastering it in one part but not in another, makes the embryo in formation to be multiform, as happens with athletes because they eat so much. for owing to the quantity of their food their nature is not able to master it all, so as to increase and arrange their form symmetrically; therefore their limbs develop irregularly, sometimes indeed almost so much that no one of them resembles what it was before. similar to this is also the disease known as satyrism, in which the face appears like that of a satyr owing to a quantity of unconcocted humour or wind being diverted into parts of the face.

we have thus discussed the cause of all these phenomena, (1) female and male offspring are produced, (2) why some are similar to their parents, female to female and male to male, and others the other way about, females being similar to the father and males to the mother, and in general why some are like their ancestors while others are like none of them, and all this as concerns both the body as a whole and each of the parts separately. different accounts, however, have been given of these phenomena by some of the nature-philosophers; i mean why children are like or unlike their parents. they give two versions of the reason. some say that the child is more like that parent of the two from whom comes more semen, this applying equally both to the body as a whole and to the separate parts, on the assumption that semen comes from each part of both parents; if an equal part comes from each, then, they say, the child is like neither. but if this is false, if semen does not come off from the whole body of the parents, it is clear that the reason assigned cannot be the cause of likeness and unlikeness. moreover, they are hard put to it to explain how it is that a female child can be like the father and a male like the mother. for (1) those who assign the same cause of sex as empedocles or democritus say what is on other grounds impossible, and (2) those who say that it is determined by the greater or smaller amount of semen coming the male or female parent, and that this is why one child is male and another female, cannot show how the female is to resemble the father and the male the mother, for it is impossible that more should come from both at once. again, for what reason is a child generally like its ancestors, even the more remote? none of the semen has come from them at any rate.

but those who account for the similarity in the manner which remains to be discussed, explain this point better, as well as the others. for there are some who say that the semen, though one, is as it were a common mixture (panspermia) of many elements; just as, if one should mix many juices in one liquid and then take some from it, it would be possible to take, not an equal quantity always from each juice, but sometimes more of one and sometimes more of another, sometimes some of one and none at all of another, so they say it is with the generative fluid, which is a mixture of many elements, for the offspring resembles that parent from which it has derived most. though this theory is obscure and in many ways fictitious, it aims at what is better expressed by saying that what is called ‘panspermia’ exists potentially, not actually; it cannot exist actually, but it can do so potentially. also, if we assign only one sort of cause, it is not easy to explain all the phenomena, (1) the distinction of sex, (2) why the female is often like the father and the male like the mother, and again (3) the resemblance to remoter ancestors, and further (4) the reason why the offspring is sometimes unlike any of these but still a human being, but sometimes, (5) proceeding further on these lines, appears finally to be not even a human being but only some kind of animal, what is called a monstrosity.

for, following what has been said, it remains to give the reason for such monsters. if the movements imparted by the semen are resolved and the material contributed by the mother is not controlled by them, at last there remains the most general substratum, that is to say the animal. then people say that the child has the head of a ram or a bull, and so on with other animals, as that a calf has the head of a child or a sheep that of an ox. all these monsters result from the causes stated above, but they are none of the things they are said to be; there is only some similarity, such as may arise even where there is no defect of growth. hence often jesters compare some one who is not beautiful to a ‘goat breathing fire’, or again to a ‘ram butting’, and a certain physiognomist reduced all faces to those of two or three animals, and his arguments often prevailed on people.

that, however, it is impossible for such a monstrosity to come into existence — i mean one animal in another — is shown by the great difference in the period of gestation between man, sheep, dog, and ox, it being impossible for each to be developed except in its proper time.

this is the description of some of the monsters talked about; others are such because certain parts of their form are multiplied so that they are born with many feet or many heads.

the account of the cause of monstrosities is very close and similar in a way to that of the cause of animals being born defective in any part, for monstrosity is also a kind of deficiency.