Heredity and Natural Selection. 303 



the apricot, which belongs to a closely allied genus, dif- 

 fer from each other in nearly the same parallel manner. 

 Darwin gives many similar instances, and we must ac- 

 knowledge that in these cases we have homologies which 

 are not due to inheritance from a common ancestor, but 

 to secondary modification. 



It is true that, in all the cases which Darwin gives, the 

 parallelism exists between forms which are very much 

 alike, and which have quite recently diverged from a 

 common ancestor, but there is reason to believe that 

 this is not always the case. Morphologists assume that 

 homology or morphological resemblance is, in itself, evi- 

 dence of community of descent, and when two widely 

 separated organisms present features which show funda- 

 mental similarity of plan, they take it for granted that 

 they owe their resemblances to inheritance from a com- 

 mon ancestor, which exhibited all the characteristics 

 which they share in common. 



This is no doubt true as a general rule, and even if it 

 were not true it would usually be extremely difficult to 

 prove its falsity; but there are a few cases where great 

 groups of animals are related to each other in such a pe- 

 culiar way that the view that all their homologies are 

 due to descent is untenable. 



The Medusae present such a case. These animals re- 

 semble each other in many particulars. They have a 

 muscular contractile gelatinous umbrella by the pulsa- 

 tions of which they swim through the water. The di- 

 gestive organs are suspended from the concave centre of 

 the umbrella, and they give rise to diverticula which 

 penetrate its gelatinous substance. Their reproductive 

 organs are developed upon the digestive tract or upon 

 its diverticula, while their organs of sense are placed 

 around the margin of the umbrella. They usually pass 



