SAPROPHYTISM AND SYMBIOSIS 773 



while the more parasitic species have few root hairs and abundant 

 haustoria, culminating in forms like Tozzia and Lathraea, which have 

 no root hairs, and in which haustoria develop even in early seedling 

 stages. Thus there is a reciprocal relation between root hairs and 

 haustoria that probably indicates the degree of parasitism. 



While the parasitism of the Euphrasieae is undoubted, the substances obtained 

 from the host plants are unknown, except in such forms as Tozzia and Lathraea; 

 from analogy with the mistletoe, it might be thought that they absorb water and salts 

 from the host plants, a view that is supported by the reciprocal relation of root 

 hairs and haustoria. In favor of the view that carbohydrates are absorbed is the 

 fact that some forms (as Melampyrum pratense) are supposed to have a capacity 

 for saprophytism, and that many ordinary green plants are partial saprophytes. 

 It has been claimed that in some species (as Euphrasia officinalis) the chlorophyll 

 has almost if not quite lost its food-making power; this view, however, seems some- 

 what improbable, since all forms except Lathraea require light for their optimum 

 development, and since most species grow more luxuriantly as autophytes in the 

 light than as parasites in the shade; furthermore, careful experiments show that 

 starch formation in these partial parasites bears quite the same relation to abun- 

 dant light and carbon dioxid and to open stomata as in autophytes. In any event, 

 the fact that all of the Euphrasieae studied, even the most autophytic, grow more 

 luxuriantly with than without parasitic attachment shows that something is gained 

 through parasitism, but it must remain for further experimentation to determine 

 its exact nature. 



The influence of external factors upon development. The seeds of 

 autophytes germinate in the presence of oxygen under proper conditions 

 of temperature and moisture, but those of extreme holoparasites, such 

 as Orobanche, require still another condition, namely, contact with the 

 proper host plant; this condition probably means a particular kind of 

 chemical stimulation. Cuscuta, though commonly a holoparasite, germi- 

 nates readily in ordinary soil, and thus its parasitism appears to be 

 less complete than is that of Orobanche; this view is supported also by 

 the occasional presence of chlorophyll and by its less specialized haus- 

 toria. Even in Cuscuta there are degrees of parasitism, some species 

 (as C. lupuliformis) germinating readily and living independently for 

 some time, whereas other species (as C. europaea) germinate slowly 

 and soon die if a host plant is not present. The seeds of partial para- 

 sites, as in the Euphrasieae, germinate generally in ordinary soil, though 

 host contact seems to facilitate germination; Tozzia, however, requires 

 host contact for germination, in this respect alone appearing to surpass the 

 dodder in parasitic specialization. The seeds of Aeginelia, a relative of 

 Orobanche, require host contact for germination, the roots of many 



