NATURAL AND ARTIFICIAL INFECTION. 55 



odour, e.g. Aec. odoratnm in America. The wind we must also 

 regard as the distributor of uredospores and of the sporidia of 

 germinating teleutospores. The Uredineae have typical spores 

 for distribution by wind with the exception of the so-called 

 spermogonia. These structures are produced by most Uredineae, 

 generally on the upper surface of the leaf and before the aecidia; 

 they are brightly coloured, and give out spermatia in a sticky 

 gelatinous slime, frequently with a distinct odour. Thus they 

 seem to be admirably adapted to transport by insects, and are 

 in fact visited by them. Their distribution, however, has little 

 importance, since they are, as far as known, incapable of 

 germination. They are regarded by many as degenerate forms, 

 either of male sexual organs, or of pycnidia. Some of the 

 spermatia have been made to germinate in artificial culture, 

 l»ut of their incapacity to germinate in natural surroundings 

 there can be no doubt. I am not aware of any one who 

 has succeeded in bringing about infection with these spermatia, 

 but I have tried it often with no result. 



It is much more difficult to ascertain how fungi, which 

 hibernate on the earth, find their way in spring to their 

 respective host-plants, in some cases even to the crown of very 

 large trees. Amongst such forms one frequently finds an 

 arrangement by which the spores are forcibly ejaculated. Thus 

 lihi/tisma accrinum, which reaches maturity only in spring after 

 liibernating on dead sycamore leaves, and Sckrotinia hctulac, 

 which does so on fallen fruits of birch, both have their spores 

 forcibly ejaculated and carried off' by wind. Klebahn states 

 that the ejaculation takes place in dry weather, and that the 

 spores of Rhytisma are prevented from drying up by a gela- 

 tinous covering. In a similar manner the hibernating spores of 

 Erysipheae on fallen leaves must be carried up again by wind ; 

 so also those of Polystirjma, which ripen on the ground and then 

 infect young leaves of plum and cherry trees. 



Infection by means of the mycelium generally occurs where 

 the mycelium lives in the earth. Thus, the hyphae of Tramctes 

 radicipcrda grow rapidly from one root to another, causing a 

 centrifugal spreading of the fungus, so that forests attacked by 

 it have the trees killed off in patches. Mycelial infection is still 

 more effective in fungi like Agaricus mcllcus which assume the 

 form of rhizomorplis. Infection by means of tlie mycelium may 



