OF А REMARKABLE UREDINE. 147 
sporidia similarly obtained, and in the same way, upon a potted plant in a glass forcing- 
box outside, and on the 21st (6 days) eighteen distinct foci of attack could be counted. 
These went on in due course to complete development. Again, I inoculated similarly a 
small plant growing in the open air, in my garden, on the 5th July, and on the 10th I 
noticed several distinct spots of attack, which ultimately ran a normal course. Attack 
is therefore very rapid. I could give the details of other experiments; but these are, I 
think, sufficient to prove that the parasite is autcecious. 
The next question in the life-history of the fungus, which presents itself for considera- 
tion, is the fate of the гесійіоврогев. It was of course probable that they re-attacked the 
same host, and experimental evidence shows that this is the case. And, indeed, the 
natural history of the fungus pointed strongly to this result, for after a plant has become 
attacked in the first instance by sporidia from over-wintered teleutospores, and some 
eecidial patches have been thus formed, the subsequent spread of the fungus is ex- 
traordinarily rapid, leading irresistibly to the conclusion that it must have been due 
to attack by the numerous гесійіоврогев liberated. Неге, then, we have a marked case of 
the assumption by eecidiospores of the function of uredospores, and the consequent non- 
necessity for the production of the latter. We have, therefore, two directions in which 
we may look for the throwing out of uredospores from the developmental history of the 
JEcidiomycetes, namely (а) where the power of immediate germination of the teleuto- 
spores secures a wide distribution, as I have explained elsewhere in the case of ZEcidium 
Strobilanthis*, and (b) where, as in the present case, the secidiospores assume this 
function. On the 23rd July I dusted some recently unfolded leaves of Jasmine on a 
twig, the end of which had been cut off under water, and then kept in a moist 
atmosphere under a glass shade in my laboratory, with secidiospores. On the 30th I 
noticed several distinct spots of attack. Аз these could not go on to complete develop- 
ment, I dusted some leaves on the latter date, both young and old, of a potted plant 
outside with ecidiospores, and covered the whole with a glass shade. Оп the 
9rd August very numerous spots of attack could be detected on all the young leaves, 
and all these went on to further complete development. Attack, then, with these 
spores is even more rapid than with sporidia. The most striking thing in such 
experiments with ecidiospores is the immense attack of the host (Pl. L., fig. 1), 
and the extreme rapidity with which it takes place. But although the young leaves 
were thus largely attacked, the older darker leaves resisted completely. I could give 
the details of other experiments in this direction also, but they would be, I think, 
superfluous. 
In conclusion, I will offer a few remarks suggested by the peculiarities of this fungus. 
'These are mainly three: (a) the production of teleutospores within the peridia; (5) the 
assumption of a distributive function by the ecidiospores; and (с) the very peculiar 
germination of the zecidiospores. 
The unusual site of teleutospore-formation forms a link with Puccinia Berberidis, 
* ‘Scientific Memoirs by Medical Officers of the Army of India,’ pt. ii. p. 20. 
