184 DIVISION II. COURSE OF DEVELOPMENT OF FUNGT. 



is in harmony with this. We should be able to connect the Ustilagineae more 

 closely with other groups, if we could do the same with Protomyces, and for this 

 purpose the approximation of Cladochytrium is important. In these two forms 

 indeed there is complete agreement except in two points. First, the propagative cells 

 formed in the resting-spores are swarm-cells in Cladochytrium, but not in Protomyces; 

 but this difference is quite unimportant for the present question, because strictly 

 homologous spores of most closely allied species may behave differently in this 

 respect. Secondly, and this is important, there is no conjugation or pairing, as far 

 as we know, in Cladochytrium. If further researches only confirm this fact, then 

 there is a gap in the chain of connection which is not bridged over, but this has yet 

 to be determined. According to the data before us it would appear that the best line 

 of connection of the Ustilagineae with other groups must lead through Protomyces 

 and Cladochytrium to the Chytridieae or to the group of the Chytridieae to which 

 Cladochytrium belongs (see page 165). The Ustilagineae therefore, considered 

 phylogenetically and in connection with what was said above of the Chytridieae, may 

 be regarded as a more highly developed group proceeding from the Chytridieae, 

 whose development has advanced from the simpler forms like Entyloma in two 

 diverging directions, and has reached its highest point on one side in the cluster- 

 forming Sorosporia and Urocystes, and on the other in the compound sporophores 

 of Sphacelotheca and other forms. The same considerations arise moreover if 

 we connect the Ustilagineae with other groups, such as the Entomophthoreae, 

 which again can only be effected through Protomyces and Entyloma. And 

 whichever view we adopt, if we accept our present data, we must look upon conju- 

 gation or pairing in Protomyces and the Ustilagineae as a special new phenomenon, 

 that is, as one which has nothing homologous with it in the groups supposed to be 

 allied to them. If it has its analogue in them in sexual processes or processes of 

 conjugation, if, in other words, it is a sexual process, this process makes its 

 appearance in the Ustilagineae at a place in the course of the development other 

 than that which it occupies in groups before considered ; in these the resting-spore 

 is always the direct result of the sexual process, while its homologue in the Ustila- 

 gineae is of asexual origin. 



It is superfluous perhaps to say in conclusion that the Ustilagineae cannot 

 at present be brought into closer connection with the succeeding groups, as the 

 following sections will show. 



Historical remarks. Brefeld (Schimmelpilze, IV) and myself (Beitr. IV) were the 

 first who attempted recently to settle the homologies and the position of the Ustilagineae 

 in the system in the sense indicated in the text. The old writers placed them next the 

 Uredineae on the ground of external resemblance. I pointed, out as early as 1853 

 (Brandpilze, p. 28), that this position was inadmissible. When I attempted, seventeen 

 years ago, in the preface to the first edition of this work, to give a short survey of the 

 groups of Fungi according to the knowledge then before us, I followed the old custom 

 of placing the Uredineae and Ustilagineae together, under Fries' name of Hvpodermii, 

 expressly quoting that writer, because the state of our knowledge at the time did not 

 allow of my substituting a new and safer view for the old one. I write this in cor- 

 rection of statements which make me the author of Fries' division Hypodermii, or which 

 even seem to make it a reproach to me that I retained that division when nothing else 

 was possible, and abandoned it as soon as our increased knowledge permitted me 

 to do so. 



