68 PARASITIC PLANTS [CHAP. 



already some stronger capacity for producing 

 them in contact with a host-plant. But this 

 capacity is a general one in protoplasm so far as 

 one can presume from the existence of parasites 

 in widely distant families of no affinities, and 

 one may add the case of the Passion flower, 

 described below. 



So that parasitism must have been acquired 

 and become hereditary solely by self -adapta- 

 tion to other plants. But, further, the normal 

 method of absorption by roots is by means 

 of epidermal root-hairs, so that their presence 

 must also be somehow explained by natural 

 selection. When, however, a seedling of cow- 

 wheat is made to germinate where no other 

 roots are present, no haustoria are formed, but 

 the root-fibres are densely clothed with hairs. 

 When, however, some other roots are present, 

 abundance of haustoria are formed, but scarcely 

 any root-hairs are produced ; so that it would 

 seem impossible to account for these two, totally 

 different organs, being on the same rootlets by 

 natural selection and compensating for each 

 other, 1 just as stated with regard to vine- 

 tendrils which are metamorphosed flowering 

 branches, when they "revert" and bear a few 

 flower - buds, the acquired sensitiveness tends 

 to disappear. 



1 For the details of structure of Melampyrum, I am indebted to 

 M. Leclerc du Sablou :Butt. de la Soc. Bot. de Fr., 1886, p. 154 ; 

 and Ann. des. Sci. Not., vi., 1887, p. 00. 



