174 
THE TROPICAL 
AGRICULTUEIST. [Sept. 1, 1899. 
From 
1-8 
to 
14-3 
8-650 
do 
3-046 
10-993 
do 
3-910 
)) 
18-672 
do 
2038 
17-140 
do 
•127 
)) 
•435 
do 
•356 
!J 
') 
4-981 
1-695 
do 
•217 
3-367 
1-276 
do 
•118 
)> 
•619 
•300 
do 
-044 
)) 
•293 
•118 
do 
nil. 
JJ 
•007 
do 
■100 
-309 
•236 
the huge quantities of nianurial matter returued 
to it in the form of prunings, leaves, fallen and 
broken pods. It muisfc, in addition, be one, in 
which the course of nitrification readily takes 
place ; in other words, a fairly rich, .friable and 
well-drained soil," 
Instead of quoting in full the analyses of the 
eight samples of fertile soils, I shall simply indicate 
the range of some of the significant constituents 
of the soil, and state the average amounts of the 
more important of these present : — 
FERTILE CACAO .SOILS. 
per cent, per cent. Avge. 
p. c 
Water retained in air- 
dried soil 
Organic matter and 
combined water 
Oxide of Iron 
Alumina 
Oxide of Manganese- 
Lime 
Magnesia 
Potash 
Phosphoric acid 
Chlorine 
Nitrogen 
These figures show that these soils are character- 
ised by a high percentage of lime, magnesia, 
potash and I may add, nitrogen; as only one of 
them has less than '2 per cent of this substance 
The amount of phospheric acid in most of them 
is rather low. In five out of the eight samples it 
is under •! per cent. The one, however, which 
contained the largest amount of phosphoric acid 
was a sample from Nicaragua, and is described by 
. Mr. J. H. Hart F. L. S., who supplied the sample, 
as a very fertile soil. Mr. Hart also vouched for 
the high fertility of a Trinidad sample, which con- 
tained only ^084 per cent of phosphoric acid. 
The water retained by the air-dried samples 
varied to a great extent, viz., from IS to 14^3 
per cent. As some of these difi'erences could not be 
explained by chemical composition only, they 
must be due to a considerable extent, to a different 
state of division of the soil-particles. 
The amounts of chlorine in the samples are so 
low, that in two cases tiie chlorine is recorded as 
"nil,'' and in five cases as " traces " only, and 
only in one sample is there a numerical expres- 
sion of the amount. All the samples contained 
appreciable amounts of oxide of manganese. 
Professor Carmody, tlie Government analyst, 
considerinsf that the data collected with regard 
to cacao-soil analysis was insufficient for anything 
like a classification of these soils according to their 
chemical constituents, had an offer, or rather the 
second offer, made through the Agricultural Society 
to estate owners, to the effect that samples of 
cacao-soils sent to the Government Laboratory 
would be analysed free of charge. Professor Car- 
mody was anxious not only to be furnished with 
fertile -soils; but v.-ith soils in which the cultiva- 
tion of cacao had been unsuccessfully tried. The 
report of the analyses you have just sent me is 
that on the samples sent in response to this second 
invitation. There are 15 analyses in all, one of 
which is specially referred to and marked 
" Venezuela good soil." the others are regarded as 
.soils of average fertility. 
Professor Carmody's report on these analyses is 
fco brief, that I may q-viote it in full. It shows that 
he still requires more data before he can draw 
general conclu.sions fro)n the analyses : — 
Government Laboratory, Trinidad, April 28, 1899. 
The Secretary, Agricultural Society. 
Sib, — I have the honour to forward the results of the 
analyses of samples of cacao soils sent by members of 
your Society, in respoDBe to an invitation from toia 
department. 
In requesting your members to forward samples, 
the main object in view has been to obtain not only 
good soils, but also those in which cacao h-ds refused 
to grow, or to give satisfactory resnita. The Trinidad 
soils sent are ordinary cacao-soils of average quality, 
but I hope that now a larger number and variety of 
soils will be sent, and will include some bad cacao soils. 
General conclusions drawn from the results of the 
analyses of a few samples are not safe ; u,ud, at 
present, I would simply direct atteution to the 
comparatively large proportions of lime, magnesia, 
potash chlorine and nitrates in the Venezuelan soil, 
which is said to yield cacao of excellent quality. The 
small proportion of phosphates and total nitrogen it 
contains are also worthy of notice. 
My long absence in England during last year- 
contributed to the delay in completing these analyses, 
which have been made under my supervision by Dr. 
Ince, Assistant Government Analyst. — I am Sir, 
your obedient servant, P, Carmody, f.i.c, i c.s.. 
Government Analyst. 
Here again, instead of quoting the 15 analyses 
in full, I shall indicate the range and average 
amounts of severpl of the constituents of the 14 
soils of average quality, and I shall quote in full 
the analysis of the Venezuelan soil specially 
referred to as yielding cacao of excellent quality: — 
CACAO BOILS OF AVEBAGE FEBTII.ITY. 
Average 
per cent, per cent, per cent 
Water retained by 
air-dried soil . . From 1-040 to 9 48 5^048 
Loss on ignition .. „ 2-870 „ 9 00 
Oxide of iron and 
Alumina .. „ 4-410 „ 14^920 
Lime .. „ -084 ,, ,700 309 
Magnesia .. „ -076 „ -498 -264 
Potash .. „ -114 „ -267 ^217 
Phosphoric acid .. „ ^096 „ -237 •156 
Chlorine .. „ -004 „ •QOg 
Nitrogen _ .. „ '067 „ -165 .111 
Nitrogen as Nitric 
Acid .. ,i •OOOg „ -00335 -00175 
Readily available potash and phosphates — 
Potash .. From -059 to 1698 -115 
Phosphoric acid .. ,, -049 ,, "104 -0767 
If the foregoing two sets of figures are gone 
over, it will be found that, in all important in- 
gredients save phosphoric acid, the fertile soils are 
distinctly the richer, whether we take the mini- 
mum, maximum, or averase figures ; whereas in the 
case of the soils of only average quality, the 
minimum and average amounts of phosphoric acid 
are higher to a marked extent than in the more 
fertile soils. This is not a little puzzling, and, 
indeed, before accepting this result, one would 
fain have the assurance of the analyst or analysts 
that the same analytical process was followed in 
both cases as otherwise the results are not 
strictly comparable. In the second set of ana- 
lyses, the analyst has gone to the trouble of 
making two determinations of potash and phos- 
phoric acid, with a view to form some idea of 
what proportion of these two substances is readily 
available. Roughly speaking, the readily avail- 
able is about half of the total, according to these 
results, but this does not help us to account for 
the apparent anomaly of the more fertile soil.i 
being poorer in pho.sphoric acid thm those of 
average quality, seeing that the minimum readily 
available phosphoric acid in the average soil is 
higher than the minimum total in the more 
fertile soils. 
Supposing the samples to be fairly representa- 
tive of the soils, and the results of the two .sets at 
