278 CAUSES OF DEGENERATIVE EVOLUTION 



ment, is that a smaller petal is always associated 

 with a larger stamen (fig. 82, B, c). 



In Viburnum opulus, the wild guelder-rose, there 

 are also two kinds of flowers, the central flowers 

 which are sexual and hermaphrodite, and the 

 peripheral flowers which are sterile and possess a 

 very large corolla. The five petals are, however, 

 equal in size, and there are no forms transitional 

 between the sterile and sexual flowers. The 

 cultivated guelder-rose is a variety of the wild 

 form in which all the flowers are sterile and 

 possess a large corolla : the plant has completely 

 lost the power of sexual reproduction. 



4. Atrophy without apparent cause. 



In many cases it is impossible to determine the 

 cause of atrophy. Why, for instance, has half the 

 inflorescence disappeared from a unilateral cyme ? 

 Why have some composites like Artemisia or 

 Kosacese like Poterium lost their habit of being 

 fertilized by the agency of insects, and become 

 anemophilous without conspicuous perianth ? Nor 

 can we explain why many Myriapods are blind, al- 

 though they live in association with species possess- 

 ing eyes and in conditions where vision apparently 

 would be useful. Semper discovered, in brackish 

 water in the Philippines, Crustacea (Cymothoe) 

 which were completely blind, although they lived 

 in light. The cause of such atrophies is unknown. 



