ON THE SEXUALITY OF THE PL'XGI. 299 



provided with sexual organs, whichj so far as we can see, are 

 quite like those found in the saprophytic forms, we have two 

 points to notice. First, these forms are close to the parent 

 stock in phylogeny — i. e. they are not much modified from the 

 type of Pythium itself, which (as a comparison with Vau- 

 cheria shows), is no doubt derived from algal ancestors, and 

 with strongly inherited sexuality. Secondly, such forms are 

 probably not so highly parasitic as is commonly supposed. I 

 do not mean to say that their living hosts are not robbed by 

 them; but it is significant that the Peronosporse are often 

 saprophytes, and that even the most parasitic forms break 

 down the parenchyma of the hosts to a rotting, fetid mass, on 

 which they then flourish. ^Moreover, they are aided by bac- 

 teria in this process. In addition to this they are apt to be 

 omnivorous. I have cultivated Pythium De Baryanum^ 

 on the most various substances, as well as on more than a dozen 

 widely diflferent living plants. 



In all these cases the parasite appears to flourish in a variety 

 of substrata, and it has not got over the clumsy habit of de- 

 stroying its host forthwith. If we compare the highly developed, 

 almost intelligent, parasitism of a higher Ascomycete or 

 Uredine with this, it will be understood what I mean by 

 specialised parasitism. Instead of clumsily destroying its host 

 (like Phy tophthora infestans does the potato), a Puccinia 

 is adapted to live in isolated patches of carefully-sheltered leaf 

 tissue, ramifying in the lacunse filled with oxygenated air and 

 aqueous vapour. Here it taps the cells as they manufacture 

 organised substances in the sunlight, taxing them not too much 

 for their strength,^ and its mycelium keeping near the stomata. 

 Its spores are then protruded in centrifugal succession, and 

 shaken off from their advantageously high position on to other 

 leaves, &c. All such adaptations must imply long periods of 

 descent (and the fungus is therefore much further from the 

 parent stock in the phylogenetic scheme) , during which even the 



1 Cf. also De Bary op. cit., and this Journal, 18S3. 



- Many Uredinece appear to do no injury at all, unless in large amount 

 and for a long time — i.e. the host can pay the tax easily. 



