5i8 
THE TROPICAL AGRICULTURIST. [December i, 1881. 
seen that the rapidly-germinating spores are continually | 
being brought by the wind, and the tubes produced as 
continually entering the stoinata of the leaf during the 
rains, &c. 
And here a fresh difficulty arises. To meet the re- 
quirements of the case, the reagent must be soluble, 
and yet the time when it is most required to be active 
is when the rains arc continuously dissolving it, and 
rurining off with it in solution to the ground. This 
being so, one must also be doubly careful in selecting 
nothing which (or products of which) may do harm to 
the tree on being washed into the soil around the roots. 
All the mineral or metallic poisons, such as salts of 
copper, zinc, antimony, &c, are therefore useless for 
the purpose, though fatal to the fungus. Either on 
account of their prime cost, or danger and difficulty of 
application, &c, the following chemicals, though readily 
lolling the fungoid tubes, must also be placed out of 
account ; acetic acid, osmic, chromic, and other metallic 
acids ; bromine and iodine and compounds, hydrocyanic 
and other organic acids of like nature.* This rapidly 
brings the list of available reagents down to a limit, 
and it appears clear that the possible chemical must 
be selected from among the following : sulphuric acid 
and compounds of sulphur, nitric acid, hydrochloric acid 
and compounds of chlorine and carbolic acid. Amnionic 
potassic, and sodic hydrates also kill the tubes 
effectually, as do strong solutions of common salt and 
other similar compounds. 
Nevertheless, very few of the above are available, since 
most of them would have to be applied in a liquid 
form, or are hi other ways unsuited. Some compound 
which will continuously evolve an easily diffused, readily 
soluble gas, forming with the water on the leaves a 
solution sufficiently strong to do the maximum of injury 
to the germinating spores, and the minimum of harm 
to the coftee leaf — the after-products of which are not 
injiuious to the tree when washed to the roots — such 
is the desideratum. 
Carbolic acid is not sufficiently soluble and diffusible 
for the purposes required, and its highly caustic and 
irritant characters make it a very dangerous reagent to 
put on living coffee trees. Nevertheless, I shall refer to 
experiments where it has been largely used in a certain 
form. Its great cheapness of course commends it. 
Chlorine compounds are not available. The caustic 
nature of " chloride of lime" brings about the destruc- 
tion of the leaf tissues, while the passage of gaseous 
chlorine, hydrochloric acid, and other alhed compounds 
into the stomata, is fatal to , the tree. 
Considering all the facts, I find sulphurous and alhed 
acids the best and most available reagents yet suggested 
for the destruction of the fungus on the coffee ; and of 
all the methods for procuring the destructive gases, there 
are noue to equal that proposed some time ago — the 
interactions of sulphur and freshly-burnt lime on the 
moist leaf — and for the following reasons : — 
(1.) The mixture is slowly continuous in action for 
long time. 
(2.) The slowly-evolved gases distribute themselves 
rapidly by diffusion. 
(3.) The chief gases are very readily dissolved in 
water. 
(4.) The solution formed is strong enough to kill the 
germinal tubes, but too weak to injure the leaf. 
(5.) The acid solution does not concentrate to a dan- 
gerous strength, since it is continually being absorbed 
by portions of the lime, and the compounds become 
further oxydi/,ed to neutral or nearly neutral salts. 
(6.) Not oidy are the final products harmless to the 
tree, but the chief one (sulphate of lime) is a valuable 
manure. 
(7.) The mixture is simply and readily applied by 
♦oolics' hands. 
* I omil here all mention of many other expensive 
bodies. 
It is therefore evident that sulphur and lime, evolv- 
ing sulphurous gases on the surface of the leaf, Ac, 
comply with most of the requirements above commented 
upon. Having decided upon the best reagent, and the time 
to apply it, it remains to examine another very im- 
portant point, viz., what is to be used as a measure 
of success in the experiments? 
In more or less preliminary trials with lime, or sul- 
phur and lime, I was much impressed with the fact 
that every one relied on the general impressions created 
by simply viewing the coffee from tune to time. This 
admits of too much chance of fallacious conclusions 
where a small difference may be of importance, and I 
accordingly tried to get rid of subjective sources of error 
by the following scheme. 
Since the reagent, by destroying germinating spores, 
should give the leaves on which they were germinating 
a longer lease of life, so to speak, if the trees are equal 
in other respects, the treated specimens should possess 
more leaves than those not treated : this being so, we 
can decide whether any resulting darker shade is due 
to an increase in the number of leaves or to some other 
change produced. If the leaves, however, remain longer 
on the tree, that tree should (hop less crop ; and hence 
the second test. Other checks to the work will be noticed 
below. 
A number of preliminary experiments with lime, sul- 
phur and lime, carbolic acid, etc., may be passed over 
as not yielding sufficiently definite results, or not having 
been performed at the proper time. Of my own numer- 
ous preliminary trials, I shall only quote those which 
throw direct light on the matter ; but I here take the 
opportunity of thanking several gentlemen for the aid 
afforded me, either by the practical application of my 
suggestions, or hi giving me valuable hints derived from 
then- former experiences. 
The first experiment on Harrow Estate. 
§ 15. Mr. Anton, of Pundaluoya, has been good enough 
to give me the following account of an experiment car- 
ried out by him on Harrow estate in 1879-80, and which 
from the results obtained, he felt encouraged to repeat 
during 1880-81. 
Sulphur and lime were applied on an area of coffee 
of a certain acreage from July 20th to August 12th, 
1879, and the results compared with what happened on 
another area of simhar coffee. Both the treated and 
non-treated fields were considered equal as regards soil, 
exposure, altitude, &c, and both had been well manured 
in May, 1878. At the time of application both areas 
were badly rusted with Hcmileia, and presented a good 
crop, which it was feared would not all be picked on 
account of the known tendency shown by diseased trees 
to drop crop. 
During October, 1879, to January, 1880, the crop was 
picked, and it was found that the sulphured coffee bore 
about 6 cwt. per acre, proportionally little light, while 
the non-treated area only gave 4 cwt. per acre and 
twice as much light. Mr. Anton suggested that the 
difference was due to more leaves having been kept on 
the trees by the treatment, and it will be seen that 
such was no doubt the case : the explanation of this 
may be deferred until the second experiment on Harrow 
has been described. 
The second experiment on Harrow Estate. 
% 16. After the results obtained in 1879-80 it ap- 
peared desirable to repeat the experiment of applying 
sulphur and lhne to the cofl'ee at a time when the rust 
had already made its appearance. I undertook the task, 
therefore, of watching a second attempt, and of examin- 
ing the results as it proceeded. 
A series of fields of coffee, covering 32 acres, and 
forming an irregular oblong block, were selected for the 
experiment : on 24 acres were applied sulphur and lime 
in the proportion of 1 to 3. The mixture was thrown 
in by hand at the rate of about 4 oz. per tree. The 
ether eight acres remained untouched. This field of 
