1 8 HETEROGENEITY. [introd. 



identical with that of the parent, observation of any parent and 

 its offspring is to the point ; but such a field as this is plainly too 

 wide to be studied with profit as a whole, and it is necessary from 

 the first, that attention should be limited to certain classes of such 

 phenomena. With this object certain limitations are proposed, 

 and though confessedly arbitrary, they will be found on the whole 

 to work well. 



The first limitation thus introduced concerns the magnitude 

 of Variations. We have seen above that the assumption that 

 Variation is a continuous process lands us in serious difficulties 

 in the application of a hypothesis which, on general grounds, 

 we nevertheless are prepared to receive. If then we can shew 

 that Variation is to some extent discontinuous, a road will be 

 opened by which these difficulties may perhaps be in part avoided. 



Species are discontinuous ; may not the Variation by which 

 Species are produced be discontinuous too ? It may be stated at 

 once that evidence of such Discontinuous Variation does exist, 

 and in this first consideration of the subject attention will be con- 

 fined to it. The fact that Continuous Variation exists is also none 

 the less a fact, but it is most important that the two classes of 

 phenomena should be recognized as distinct, for there is reason to 

 think that they are distinct essentially, and that though both may 

 occur simultaneously and in conjunction, yet they are manifesta- 

 tions of distinct processes. The attempt to distinguish these two 

 kinds of Variation from each other constitutes one of the chief 

 parts of the study. It will not perhaps be possible to find any 

 general expression which shall accurately differentiate between 

 Variations which are Discontinuous and those which are Con- 

 tinuous, but it is possible to recognize attributes proper to each 

 and to distinguish changes which are or may be effected in the one 

 way from other changes which are or may be effected in the other. 



For the present w r e shall treat only of the evidence of Dis- 

 continuous Variation. 



In order to explain the second limitation w T hich is to be intro- 

 duced it is necessary to refer to some phenomena which are 

 characteristic of the forms of organisms, and to separate from them 

 the group with which we shall deal first. 



It was stated above that perhaps no character of form is 

 common to all living things, but nevertheless there is one feature 

 which is found in the great majority. 



In the first place, the bodies of organisms are not homo- 

 geneous but heterogeneous, consisting of organs or parts which 

 in substance and composition differ from each other. This 

 heterogeneity in composition is of course an objective expression 

 of the process of Differentiation, and it is further recognized 

 that such structural heterogeneity of material corresponds with 

 a physiological Differentiation of function. This Differentiation 



