808 MR. H. MARSHALL WARD ON THE 
On Dec. 12 I treated six seedlings in the same manner, sowing 
spores on the young leaves of five and on the cotyledon of the 
sixth specimen. All the seedlings were grown in the same pot 
and soil, and may be considered equal in all essential respects. 
A very faint spot appeared on each of the leaves of four speci- 
mens on Dec. 21st; a similarly indistinct spot was developed 
next day on the leaf of the fifth; but the spot on the cotyledon 
did not become visible till the 23rd. 
Here, therefore, appeared a less successful attempt to prediet 
the rate of development of the “ disease-spot ;” whereas I had 
expected at least 14 days to be occupied in the whole process, 
there were, in fact, only 9-10 and 11 days respectively so occu- 
pied. One more example :— 
On Oct. 25 a sowing was made on a large, dark, tough leaf 
of Coffea arabica treated as usual. According to calculation, I 
searched for the “ disease-spot” on Nov. 8 and 9; but no 
spot appeared, and no trace could be discovered till the morning 
of Nov. 11, when the yellow spot appeared as usual. Here 
about 16 days had been occupied in the formation of the given 
mycelium and spot. 
In Table II. are summarized the results of further experiments 
of the same nature ; these should be compared with those already 
published in the Third Report to the Ceylon Government (Ses- 
sional paper xvii. 1881). 
Although in the above-cited examples, which have been pur- 
posely selected to illustrate all the chief points, it is evident that 
an absolutely accurate prediction did not always occur, some 
stress should be laid on the fact that the time allowed for the 
production of the “ disease-spot"' (viz. 14 days) proved correct in 
& majority of experiments. 
It is now time to inquire if any explanation of the differences 
is forthcoming. During the progress of the investigation it 
became clear, as already stated, that spores which had been 
matured some time and become quite dry took longer to germi- 
nate than fresh spores which had just ripened: experiment also 
showed that spores which were produced during the moist weather 
could be made to germinate in a few hours, and might produce a 
normal and complete germinal tube within 12 hours from the 
moment of sowing. I had indeed found the uredospores germi- 
nating while still on the damp parent rust-patch ; and the teleuto- 
spores commonly do so. 
