CHAPTER V.—COMPARATIVE REVIEW.—USTILAGINEAE. 181 
conjugation of the segments of the promycelium, or primary sporidia, and that in 
Urocystis Violae, whose development was also fully observed by Kühn, conjugation 
must be supposed to take place from Prillieux’ figures, though it is not expressly de- 
scribed. On the other hand there are species, which, according to our present observa- 
tions, do not conjugate. But we have no complete observations of their development; 
only from the data which we possess, and the positive result of Kiihn’s extended re- 
searches into the infection of host-plants with Urocystis Maidis, it must be supposed 
that the endophytic sporogenous mycelium is in this case developed directly from the 
resting-spore or from interpolated gonidia which do not conjugate. Accordingly the 
course of development, apart from certain special formations and complications which 
vary in the different species, would be the same in its main features in the entire 
group of the Ustilagineae up to the difference which lies in the presence or absence 
of conjugation. 
What importance is to be assigned to this difference is open to discussion. The 
pairing I have formerly called conjugation and do so still, thus giving expression to 
the view, that it may or ought to be considered as analogous with a sexual process. 
Brefeld does not admit this and explains it to be an analogous phenomenon to the 
coalescence of vegetable cells, especially young germ-tubes (see on page 2). In 
such cases as these no decisive argument for or against is to be found in the phe- 
nomenon in itself, but we must look about for indirect grounds of probability. Mine 
are as follows, if I confine myself to cases that are thoroughly known. First, the 
almost invariable occurrence of pairing under the normal conditions of germination ; 
the conditions, I mean, to which the species is actually adapted in nature. The 
conditions are those for germination in water in the case of Tilletia, the species of 
Entyloma, Urocystis, and Tuburcinia Trientalis, as has been already shown; and here 
pairing takes place so promptly that it is not easy to keep primary sporidia from 
uniting, if we disregard some special exceptional cases mentioned above. But these 
cases are of the kind which prove the rule. Secondly, the great predominance of 
union in pairs. The sporidia which are placed close to one another in whorls in 
Tilletia, Entyloma, Urocystis, and other genera unite, almost without exception, in 
pairs only, and where there is an odd sporidium it usually does not conjugate, 
though its union with some pair would be easy, one might almost say would be very 
natural. The segments of the promycelium of Ustilago Carbo conjugate when they 
are immediate neighbours in a mode similar to the ‘ clamp-connection’ of page 2; 
under different local relationships pairing is effected in a different way, as by loop- 
unions between two segments of a promycelium which are separated by a pair which 
have already united. These facts show that a change usually takes place in a pair 
after conjugation which renders a second union difficult or impossible, while it 
introduces the further development. All these are phenomena which find their 
analogy, as far as we know at present, only in sexual processes, or, as it may be 
briefly expressed, in sexual processes of conjugation, and must be interpreted by 
them, till we obtain further knowledge. The case is different in the coalescence of 
germ-tubes, which are the nearest comparable. A glance at Fig. ı shows that in 
that case coalescence may take place between any number of spores or may even be 
omitted, while the further development is the same in kind, but shows different 
degrees of strength according to the number of the germs which have coalesced. It 
