December i, 1881.] THE TROPICAL 
AGRICULTURIST. 
sudden appearance of the "disease " is closely connected 
with wind, and this connection is of exactly the same 
nature as we should expect if the wind blows spores 
about. That the wind does cany spores of Hemileia 
as well as other fungi, is now proved. How far a single 
spore may be borne on a high wind is partly answered 
by my experiments in July, &c. Here were clearly spores 
caught in transit under such circumstances that we may 
safely conclude they would have travelled at least twice 
as far if thoy had not been caught by the obstruction 
— that is to say, the spores would travel 50 feet in one 
journey. But it must be remembered that the very 
light dry spore torn by the wind from a leaf on a hill 
or other elevation might he carried much further. 
But even if the spore only moves per saltum, it may 
be home some distance on a dry, gusty day over such 
open country as the coffee districts, and there can be 
no doubt that, once it has reached an area of coffee, 
the fungus has every chance of distributing its spores 
rapidly over the trees to leeward ; the progeny having 
equally good chances, on the whole, of travelling hack 
again when the wind blows in the opposite direction. 
It is in accordance with these facts that wind-blown 
areas of coffee become so badly diseased. 
But another sories of events suggested an equally im- 
portant connection with atmospheric moisture. Of two 
series of plants in the same exposed verandah, I found 
that those placed on the edges of the verandah, and 
kept wetter on the whole (from drip, driving rain, &c), 
appear to become worse " diseased " than more shelt- 
ered ones. Experiments then proved that plants strewn 
with spores and placed in wardiau cases became diseased 
in a fortnight to three weeks, if the interior of the 
cases was kept wet and the atmosphere surcharged with 
moisture ; whereas in very dry cases no such infection 
took place. These and other experiments now before 
you led to the establishment of the connection between 
rain, dew, &c, and the outbreak of disease, and is in 
agreement with the known facts that Hemileia flourishes 
especially in damp, steamy districts, and breaks out 
any influence in producing or diminishing attacks of 
" leaf disease " have not been corroborated by investi- 
gation. I find nothing to warrant any direct connec- 
tion between the phenomena of disease and thunder- 
storms. 
But the admitted fact, that the great outbursts of 
Wkmileia occur during the S. W. monsoon, and about 
the break of the N. E. monsoon, is also in accordance 
with the foregoing. It is in June and July especially 
that the winds and saturating rains first occur near 
Kaudy, for instance, and it is in July and August that 
the " disease " is at its height : clearly this occurs be- 
Cause the wind distributes spores which the rain then 
finises to germinate. So, also, durinu' the other periods 
of wind and rain. It is difficult to say of any two 
seasons that the outbreaks of Hemileia are more or less 
virulent, because no standard of comparison is given. 
Ill continuously wet weather the spores are less readily 
mftlibuted by wind currents through the air, but each 
one stands a better chance of germinating on a neigh- 
Hraring leaf : this I believe to be the reason why tho 
"disease" comes and goes in virulent and sudden 
attacks in a dry year such as this ; but " bangs about," 
to use a popular expression, during a wet season like 
that of lsso. Tho same, of course, applies to damp and 
Sheltered portions of an estate, or whole estates, or 
even districts, and a little reflection will convince any 
one with the competent knowledge that siicb is the case. 
Hut there ore other conditions to bo taken into account 
in comparing the "attacks of leaf discus. " in different 
district,, or on two estates, or even in different yean. 
It 18 plain that since llrmil in only attacks the leaves, 
there . ail be none of these extensive outbursts of disease 
On denuded coffee, and hence, from circumstances of 
soil, treatment, climate, &c, it may occur that two areas 
of coffee differing in this respect — i.e., in the proportion 
of foliage on the trees — may differ in the amount of 
"rust" and in the apparent " virulence of the disease." 
It is because a large surface of food material is offered 
to the fungus that it is so successful, and it is because 
a large area of leaf surface is offered to the spores, 
that they affect their inroads on the coffee with such 
facility. 
Speaking of the coffee near Peradeniya for illustra- 
tion, then, (each valley, and many estates differing in 
various ways as regards climate), the chief reasons that 
there was so little "leaf disease" noticeable during 
February, March, and April of this year (1881) may be 
thus summed up :■ — (1) There were at that time but 
few leaves formed on the trees, and therefore but little 
surface on which the wind could sow spores ; (2) there 
was but little wind to cany spores, and hence few fly- 
ing spores to be sown ; and (3) the rain was chiefly in 
showers with long dry intervals, and so the under sur- 
face of the leaf would rarely be wet for more than a 
few hours at a time. 
But with the April rains came the conditions required 
for the geiTnination of the comparatively few spores 
which had been successfully placed in the interval, and 
these gave rise to the still (comparatively) few "rust- 
patches " of May- June. Then came the heavy winds and 
scattered the " rust " produced by these patches on to 
the now numerous leaves ; rains, moist air, and dews 
occompanying or following the wind ; the consequence 
and climax being a widespread attack in July. On July 
15th the weather cleared up, and it was hot until the 
end of the month, and, although a number of spores 
would geiminate in the dews here and there, there was 
a lull in the disease as the already-destroyed leaves fell 
to the ground. Much distribution of spores occurred in 
this dry interval however, and the rains in August en- 
abled them to fonn mycelia, and so on. 
It is therefore evident that no general statement can 
be made as to the time of year an " attack of leaf- 
disease " may be expected, beyond saying that, since it 
is at tho burst of the south-west monsoon (or north- 
east as the case may be) that the estate is in full leaf, 
tho winds carrying spores about and shaking the trees, 
and the rains arc affording moisture for the gemiination 
of the spores, so it is some time within three weeks 
to a month afterwards that the outbreak of rust may 
be expected. 
That pruning, manuring, shelter and other such pro- 
cesses of cultivation should be directed in accordance 
with these principles, will be obvious, but as this is 
important, I shall refer to it in more detail shortly. 
" Attacks of Leaf Disease." 
§ 8. Since illustrations from the estates themselves 
are in some respects more intelligible, I propose to give 
one or two examples further showing the tmth of the 
above remarks. 
On an estate in Dirabula the experiences as regards 
the outbreaks of leaf disease this year were as fol- 
lows : — The disease commenced to show in May, increas- 
ing during June, and becoming very severe about July 
1st, when tho attack was at its height ; before or about 
the 15th of July an immense quantity of loaves were 
falling, and the disease was " passing away." Very 
little more occurred, except in damp corners of the 
estate until August, when a milder attack was noticed 
towards the end. On examining the weather and state 
of tho coffee from February, it is clear that this illus- 
tration confirms what has been already said. 
During tho April ruins the spores (which had been 
slowly accumulating bs the leaves formed) germinated, 
and produce I a large stock of spores in time for the 
monsoon winds and rains of May and June. The latter 
month being very wet, we cannot donbt that immense 
numbers of spores germinated, their tubes entered tho 
leaves, and tho resulting myoelia would be rampant In 
