182 DIVISION II.—COURSE OF DEVELOPMENT OF FUNGI. 
is true that conjugation of more than two primary sporidia has been observed in the 
plants which we are considering; in Tilletia, for instance, and Ustilago Tragopogonis; 
but these exceptional cases are comparatively rare and would only confirm the rule, 
even if similar exceptions in typical sexual conjugation, as in Zygnema and Acetabu- 
laria, were unknown, 
The reasons which Brefeld alleges against my conception of the pairing, as far as 
I understand them, are these. Pairing occurs in germination in water; it does not 
occur when the primary sporidia of Tilletia, for example, are placed in nutrient 
solutions before pairing can begin; it also does not occur in Ustilago Carbo and 
similar germinating species, if they germinate and sprout in a nutrient solution. But 
when the supply of food is exhausted, pairing sets in in many species and is followed 
by the formation of incipient mycelium. A fresh supply of food again sets up 
gonidial sprouting. This shows that the uniting cells are not sexual organs in 
themselves incapable of germination, for ‘according to our present ideas with 
respect to the essence of sexuality” no such change in the properties of sexual cells 
could be brought about by changes in the nutrition. The answer to this is, that, 
according to our present knowledge, the only processes of conjugation which can be 
here brought into comparison, and which are to be distinctly regarded as simplified 
analogues of sexual processes, consist solely in the characteristic coalescence in 
each case of two cells into one capable of further development. What further 
‘essence ’ is involved in the matter we do not know. Many of these conjugating 
cells are not capable of further development without conjugation, the gametes, for 
example, of many Chlorophyceae; but this is not the case with all. I will not 
mention the Phaeosporeae, which supply examples of the opposite behaviour, because 
the circumstances in their case are not quite clear and are to some extent open 
to question. But Rostafinski and Woronin found that the swarm-spores developed 
from recently matured resting-spores of Botrydium granulatum? do typically and 
necessarily conjugate, and that the same swarm-spores germinate without con- 
jugation when they proceed from resting-spores which are more than two years 
old. A change therefore takes place in these cells, affecting those properties in them 
with which we are at present concerned, through the influence of external causes 
which must be closely connected with processes of nutrition. Even highly differen- 
tiated sexual cells may be capable of further development when fertilisation is 
prevented, as is shown by the oogonia with unfertilised oospheres of Pythium 
megalacanthum? which send out strong germ-tubes. We see therefore that the 
exceptional cases to the pairing which is the general rule in the Ustilagineae, whether 
artificially produced or spontaneous, are not opposed to phenomena known to occur in 
organs which are undoubtedly sexual. 
That there are species which do not pair cannot affect our decision with regard 
to those which do; moreover it is well known, that in other groups of Fungi organs 
strictly homologous have a distinct sexual function in some species and may be 
asexual in others, as we learn by comparison of the Peronosporeae and those Sapro- 
legnieae which are without antheridia, 

* Bot. Ztg. 1877, p» 662. 
® Bot. Ztg. 1881, p. 543. 
