20 SALMON: SUPPLEMENTARY 
It may be remarked here that little is known in connection with 
the germination of the ascospores. Wolff succeeded in germi- 
nating those of Erysiphe graminis, and in infecting grass leaves with 
them, but apparently no one since has repeated the experiment. 
Galloway germinated the ascospores of Unctnula necator, but failed 
in all attempts to infect leaves with them. In September, 1900, | 
placed a number of leaves bearing ripe perithecia of Arysiphe tortilis 
on the ground. These leaves were fully exposed to the weather, 
and were kept in every way under natural conditions. From time 
to time during the winter the dried remains of the leaves bearing 
still numerous perithecia were examined, but no opening of the 
perithecium or change in the appearance of the asci or ascospores 
was found. Finally, in the early summer of the next year, all the 
material was examined, consisting of about twenty perithecia on 
fragments of leaves, and it was then found that externally the 
perithecia appeared quite unchanged. On opening the perithecia 
it was found that many of the ascospores had collapsed, and in 
the surviving ones no germination could be induced. It may be 
worth noting that in Z. torts the appendages of the perithecium 
are very persistent, being found little changed at the end of eight 
or nine months of exposure. In U. xecator Galloway found that 
the appendages soon disappeared on exposing the perithecia to the 
weather. 
With regard to the sudden appearance of ‘‘ Ozdium”’ on plants, 
it has long been remarked that it is especially liable to occur after 
cold nights. Some experiments I have made ('71) in connection 
with the germination of conidia seem to give some direct evi- 
dence on this subject. In the case of S. Aumult on strawberries it 
was found that the conidia when sown in hanging drops at ordinary 
temperature showed only a feeble germination. On subjecting the 
conidia, however, before being sown to a low temperature—either 
by placing them on blocks of ice, or by subjecting leaves bearing 
conidia to a temperature of 0° C. the germinating power shown sub- 
sequently was very perceptibly greater. As I have already (71) 
remarked, it would appear, therefore, that in nature a decrease 
of temperature may cause plants to be attacked by mildew, not — 
only because the vitality of the plants may perhaps be weakened, 
and so rendered less able to withstand the attacks of parasites, but 
