PARTS OF ANIMALS, I. i. 



or will be in the future." Nor, in a process of 

 reasoning of this kind, is it possible to trace 

 back the links of Necessity to eternity, so as to say, 

 Because A is, therefore Z is. I have, however, dis- 

 cussed these matters in another work,^ and I there 

 stated where either kind of Necessity applies, which 

 propositions involving Necessity are convertible, and 

 the reasons why. 



We must also decide whether we are to discuss 

 the processes by which each animal comes to be 

 formed — which is wliat the earlier philosophers 

 studied — or rather the animal as it actually is. 

 Obviously there is a considerable difference between 

 the two methods. I said earlier that we ought first 

 to take the phenomena that are observed in each 

 group, and then go on to state their causes. This 

 applies just as much to the subject of the process of 

 formation : here too we ought surely to begin with 

 things as they are actually observed to be when 

 completed. Even in building the fact is that the 

 particular stages of the process come about because 

 the Form of the house is such and such, rather than 

 that the house is such and such because the process 

 of its formation follows a particular course : the 

 process is for the sake of the actual thing, the thing 

 is not for the sake of the process." So Empedocles 

 was wrong when he said that many of the character- 

 istics which animals have are due to some accident 

 in the process of their formation, as when he 

 accounts for the vertebrae of the backbone by say- 

 ing ^ " the fetus gets twisted and so the backbone 

 is broken into pieces " : he was unaware (a) that 

 the seed which gives rise to the animal must to 

 ^ Emped. frag. 97 (Diels, Fragmented, 31 b 97). 



(31, 



