58 THE ORIGINATION OF PARASITISM. 



Next in importance in the history of investigations upon the matter are 

 the cultures in which seedling-s of green plants have been made to send 

 their roots into the bodies of other species as into a substratum. Wheat 

 has thus been nourished on potato tubers for extended periods, and peas 

 have sent their root-systems through medulla of stems of the horse-bean 

 ( Vicia) in such manner as to obtain sufficient material to carry out com- 

 plete development. The reductions in both root and shoot members in 

 this case are indicative of a marked parasitism. 



Finally, there comes up for consideration the experimental arrangements 

 which justify the present paper, in which regenerated shoots of several 

 species were established in nutritive contact with various hosts. The de- 

 velopments of the shoots were of the restricted character sh own in parasitism ; 

 the roots, when present, were likewise atrophied, and in some instances 

 these organs were entirely lacking, absorption being carried on through 

 epidermal tissues of the bases of stems immersed in the bodies of the 

 hosts. Some of these xeno-parasites lived for extended periods, sustain- 

 ing dependent nutritive relations upon enforced hosts, a few being now at 

 the close of the second year of their existence. In addition to extending 

 the range of experimental research upon the matter, the number of pre- 

 parations employed has permitted some analysis of the conditions which 

 affect parasitism and which must be taken into account in a consideration 

 of its origination and development among the higher plants. 



The factors of importance are seen to be: coincidence of seasonal cycle, in 

 which extending roots or elongating shoots may encounter active stems and 

 roots of a possible host; supplementary roots and stem habits; succulency 

 or presence of some balance of solutions in the bodies of both possible 

 host and parasite; unequal osmotic activity of the cell-sap of two plants of 

 a possible couple, and equivalent range of variation with regard to this 

 feature, it being obvious that any plant would soon free itself from a de- 

 pendent which did not follow it in its extreme concentration, thus entail- 

 ing, of course, the power of automatic adjustment in the osmotic activity 

 of the dependent, as has been suggested. With regard to anatomical and 

 mechanical features, the ready formation of wound-tissues and copious 

 excretions of mucilage or formation of oxidases would act as a deterrent to 

 a possible parasite, while the penetrative power of submerged or aerial 

 roots and the formation of haustoria or of epidermal absorptive cells would 

 be an opposed factor in the possible parasite. It is thus to be seen that 

 given any two plants, knowledge of their capability for entering into a 

 nutritive couple may be put in the form of an algebraic equation the reduc- 

 tion of which should indicate with some certainty the possibility of their 

 adhesion. 



The unceasing distributional movements of plants would operate to 

 bring under test conditions a large number of pairs of species, and it seems 



