PARTS OF ANIMALS, I. iii. 



" pale " and " dark " : neither " tame " nor " pale " 

 is a differentiation of " winged," but the beginning 

 of another line of differentiation, and can come in 

 here only hy accident. Therefore, as I say, in dividing 

 we must distinguish the one original group forthwith 

 by numerous differentiae ; and then too the privative 

 terms will make valid differentiae, which they will 

 never do in the system of dichotomy. 



Here are further considerations to show that it is 

 impossible to come at any of the particular species by 

 the method of dividing the group into two, as some 

 people have imagined. 



Obviously it is impossible that one single differentia 

 is adequate for each of the particular species covered 

 by the division, whether you adopt as your differentia 

 the isolated term or the combination of terms ° (for 

 this is intended by the continuity of the series of 

 differentiae throughout the division from the original 

 group, to indicate that the whole is a unity ; but, in 

 consequence of the form of the expression, the last 

 one comes to be considered as the sole differentia). 

 And it is evident that there cannot be more than one 

 such differentia ; for the division proceeds steadily until 

 it reaches the ultimate differentia, and — supposing the 

 division is aiming at " Man " — this is either " cloven- 

 footed " alone, or else the whole combination, e.g. 

 if one combined " footed," " two-footed," " cloven- 

 footed." ^ If Man were merely a cloven-footed 

 animal, then this would be the one differentia, arrived 

 at by the right method. But as he is not merely 



" i.e. the last term of any series, or all its terms together, 

 as he goes on to say. Cf. 643 b 15 f. 



* This definition appears also in Met. 1037-1038. 



91. 



