GENERATION OF ANIMALS, II. viii. 



ass, two animals which differ in species [and it was 

 laid dawn that an animal of a different species is 

 produced by two animals that differ in species]. Now 

 this argument is too general ; there is nothing in 

 it J because there is nothing in any argument which 

 does not start from the first principles belonging to 

 the particular subject. Such arguments may appear 

 to be relevant, but in fact they are not. For a geo- 

 metrical argument, you must start from geometrical 

 principles, and the same applies elsewhere ; that 

 which is empty, which has nothing in it, may appear 

 to be somewhat but in fact is nothing at all. But 

 also, this argument is false, because many of the 

 animals that are produced from parents of differing 

 species are fertile, as I have said earher. No ; this 

 method of inquiry is as wrong in natural science as 

 it is elsewhere. We shall be more Hkely to discover 

 the reason we are looking for if we consider the actual 

 facts with regard to the two species, horse and ass. 

 First, then, both horse and ass, when mated with 

 their own kind, produce only one at a birth ; secondly, 

 the females do not on every occasion conceive when 

 covered by the male, and that is why breeders after 

 an interval put the horse to the mare again [because 

 the mare cannot bear it continuously]. Mares do 

 not produce a large amount of menstrual discharge ; 

 indeed they discharge less than any other quadruped ; 

 she-asses too do not admit the impregnation, but pass 

 the semen out with their urine ; and that is why 

 people follow behind, flogging them." Further, the 

 animal is a cold subject ; and as it is by nature so 

 sensitive to cold, it is not readily produced in wintry 

 regions, such as Scythia and the neighbouring parts, 

 or the Keltic country beyond Iberia, which is also a 



255 



