LII 



the excessive variability of the ocelli but in the interpretation of this phenomenon 

 little progress has been made, because, as 1 have already pointed out, so little 

 attention has been paid to the fact that every animal organism is subjected to 

 continuous changes of form, appearing in a greater or lesser degree in various 

 individuals, and that changes of colour come under this category, from which it 

 follows that where in consequence of colour evolution special forms have been produ- 

 ced, as is the case in butterflies with ocelli, these changes are particalarly striking. 

 Individual differences exist invariably ; no two butterflies, any more than two men, 

 being completely alike ; some difference in colour or marking is, as a rule, to be noted 

 also between two butterflies of one species. With regard to the forms created by the 

 evolutional stress of a special process of change of form such individual differences 

 must consequently occur. Its greater or lesser progress is always governed by 

 the peculiar constitution of each individual whose origin is never identical with 

 that of the other individuals and also by the very different individual suscep- 

 tibility for this special stress of development. These must, therefore, of neccessity, 

 reflect the fact whether the process of the evolution occurring here has more 

 or less advanced in some individual and, therefore, the degree of its develop- 

 ment. In one the ocelli will consequently be larger or more numerous than 

 in another; in some they will exhibit a particular distribution while in others 

 again they will be scarcely apparent or altogether absent. This is, in fact, 

 what a study of the ocelli brings to light. In size they differ considerably 

 in individuals of the same species. Bateson, on page 295 of his important 

 work already quoted, figures four specimens of Satyrus Hyperanthus L. all 

 differing in the number and size of the ocelli ; three of these he notes as having 

 been taken on the same day in the same district in England, " so that here 

 no question of seasonal or local difference is necessarily involved". Indeed, 

 in view of the present tendency of basing so-called sub-species and even species 

 purely on such differences, this observation may usefully be reiterated. Besides, 

 how insignificant actually are these differences. The specimen marked n*^. Ill is 

 the only one of any importance because practically no ocelli are to be seen 

 in it. This form occurs very little and is evidently one in which the colour 

 development has been much retarded ; it clearly corresponds in the present 

 species, producing but a single generation in a year, to the form Levana in 

 the Vanessa species yielding two generations each year. In Cyllo Leda L. 

 such specimens occur not infrequently and here it may be observed how they 

 arise through retarded colour development and grade into one another with 

 more advanced forms. As regards the other three, which are in a further 

 advanced stage of development, there is only a slight difference in respect to 

 the size of the ocelli; in the one which is most developed, marked IV, by the 



