November j, 1888.] THE TROPICAL AGRICULTURIST. 
par Is of tlic province where they were tried, and the 
straw proved to be rich in saccharine juices. Average 
.samples of the juice of the saccharatiwm was found 
to contain 10'59 per cent of crystallisable, aud l - 79 
per cont of non-cystallisablo sugar. Ordinary sugar 
cane, I may remind yon, contains about 15 per cent 
of crystallisable sugar and about half as much of 
Die uon-crystallisablc. 
Comparing ordinary sugar-cane with sugar sorghum, 
the former nerds a rich deep soil which necessitates 
a large expenditure in preparing the land and pro- 
viding the cane sets, &e., demands regular irrigation 
during its entire growth and occupies the land over 
twelve months. The sugar sorghum will yield fair 
crops on vory ordinary soils, the seed is cheap and 
the land needs no expensive; preparation, irrigation 
is unnecessary in ordinary seasons and the crop never 
occupies the land longer than from three to four 
months ; while unlike the sugarcane, the sorghum 
will yield grain, a crude sugar, and a feediog straw. 
The sugar of the juice of the ordinary S. sac- 
charatum (the Chinese sugar-cane) always crystalised 
with difficulty, but that of the juice of a variety 
known as the Early Amber, and that of tho juice of 
the S. kaffra/riwin we introduced from the United 
States usually crystallised freely, though as far as 
our experiments went in, neither case was this so 
readily or perfectly as in the case of the sugar of tho 
ordinary sugar-cane juice. However, as far as regards 
the manufacture of jaggery, as large an outturn was 
obtained from sorghum juice as from sugar-cane juice. 
All the ordinary cereals of South India received 
more or less attention, but I have not now time to 
refer to other cereals. I may just mention that near- 
ly 8ll these cereals yield small seeds with a consider- 
able percentage of husk and usually contain but a 
small percentage of albuminous matter. The pulses 
are better as a rule but the area of land occupied 
by the pulses is small. This is much to be regretted, 
both on agricultural and physiological grounds, for 
cereals occupy far too much of the cultivated land, 
and pulso is so especially needful in the diet of a 
non-tlesh-oatiug people such as live in South India. 
Cotton. 
In many respects the soil and the climate of the 
province are suited for cotton culture; indeed, there 
nro vast areas of black soil specially fitted for the 
crop. T.ut, while the average yield of cotton in 
Madras is only. 70 pounds per acre, the average yield 
in the cotton States of Ameiica is over 300 pounds 
per acre, while American cotton usually sells here for 
trom J."> to ol» per cent more money than Indian cotton. 
The desirability of improving the quality and the 
yield per aero of tho cotton crop has long been re- 
cognised; and efforts have beon mado to introduce 
better rarieties of the plant and improve the system 
0l its cultivation. Much attention was given to this 
matter more than forty years ago, under the direc- 
tion of an cmiuent botanist, Dr. Wight, but unfortu- 
nately he failed to recognise how intini'te'y tho 
question of cotton improvement is conuccl.d with 
that of agricultural improvement. 
At Stidapet attention has been giveu chielly to 
the introduction and distribution of good seed of the 
New Oilcan ;, Yea Vallov, ptiftn and one or two 
other well-known and appreciated varieties of cotton. 
New Orleins has been p irt icularly successful, and by 
carefully selecting tho seed for several years, tho pro- 
portion ( ,f lint to seed has been greatly increased. 
I'bi New Orleans has proved a heavy oroppor nnd a 
hardy plant. Tho Yes Valley, a Peruvian variety 
obtained through the agency of Mr. Clements Mark- 
ham, c. B., was found to grow and thrive under 
good cultivation near the coast and to yield a long 
and valuable lint. A sample of this lint was valued 
in Liverpool at a higher price than any other kind of 
cotton on snlo at the tune from either India or 
Egypt, and as high as the host American cotton. Tho 
plant I- unfortunately very liable to the attacks of 
i Bnib ol .. cob-opt inns in K,'y|>tUn cotton grows 
wi ll ui M.idra ., and s.imc may be said regarding tho 
11'iinititfhaut, ami one or two other kiuds introduced 
from Uotubuy, 
It has been clearly shown that there is no necessity 
to confine the culture of cotton to the black soils 
aud better loams as is now the custom, for under good 
treatment poor sandy soils produce twice as heavy crops 
as the average crop yielded by much of the black soil 
region. 
TonACCo. 
Tobacco is s-nother crop to which much attention 
has been given; it is grown in all parts of the province 
but only in small plots. Tho total area of land usual- 
ly cropped with it is only about 80,000 in a cultivated 
area of over 23,000,000 acres. The plant grows well 
in most parts of the province, and fair crops are 
raised, but generally tho leaf is very inferior, from 
its proverty in mineral constituents, and the curing 
to which it is subjected is always unsatisfactory. The 
tobacco is almost unsaleable in this country, though 
the cheroots made at Trichiuopoly meet with a cer- 
tain amount of appreciation. 
At my suggestion samples of tobacco from all 
the chief tobacco-growing districts were chemically 
examined. More than 50 samples wero analysed. The 
investigation was confined to a determination 
of the percentages of nicotine and ash, the 
percentage of carbonato of potash in the ash 
leaf. Tne results of these investigations wero valu- 
able and suggestive. Nicotiue varied from 1J to 7 
per cent; tho average was about 3 per cent. The 
percentage of nicotine determines what is usually 
called strength in tobacco ; when there is over 4 
per ceut the tobacco is a strongly intoxicating one. 
Havana tobacco contaios usually about 2 per cent. 
The percentage of ash averages about '20 ; iu a few 
cases it was over 25, but there were others in which 
it was under 15. The greatest variations were iu the 
percentages of carbonate of potash iu the ash. In 
39 of the 50 analyses made, the percentages of car- 
bonate of potash was under 7 ; the majority being 
under 5; iu some instances it was as low as 0'78. 
One sample, grown on the Nilgherry Hills, contained 
nearly oO per cent, of carbonate of potash in the 
ash, and fine samples, grown in other thinly popu- 
lated parts of the province, contained from 10 to 17 
per ceut. The potash in a tobacco influences greitly 
its burning, the larger the percentage the bettor it 
burns, and tho whiter is the ash. Samples of tobacco 
produced in the United States contain frequently as 
much as 35 per ceut of carbonate of potash in the 
ash. The low percentage of carbonate of potash 
generally fouud in the ashes of the tobacco of the 
older cultivated districts of Madras is a sure indica- 
tion of the proverty of the soil in potash and this 
mineral is, it must bo remembered, a food of the 
greatest value for nearly all kinds of cultivated plants. 
The close connection between tho percentage of car- 
bonato of potash in the soil and iu the ash of the 
tobacco grown on it, was shown in a remarkable 
manner in some tobacco experiment'' conducted at the 
Saidapet farm, the soil of which in 1808 contained but 
i trace of this mineral in 1871, the tobacco proiuced 
gave an ash containing on the avorag«_ only 0-6 per 
ceut. of carbonate of potash ; in 1879 after seven 
more years of improved cultivation niue samples of 
tobacco produced iu the same year were analysed. Of 
these, tho ash of four samp'cs was found to contain 
nearly five per cent of carbonate of potash and the 
ash of all the remainder contaiuod each more than 3 
per cent. In other words, the percentage of carbonate 
of potash was fully 12 times greater. 
Tobacco seed of good varieties was obtained from 
the United States and other countrios aud distributed 
to cultivators along with suggestions for the more 
careful selection of soils and treatment of the crop. 
Special attention was devoted to the curing of the 
tobacco : the native system being a highly unsatis- 
factory one, chielly in the excessive lermeutation to 
which the leaf is exposod. 
Selected Seep. 
While efforts were made to introduce seed from 
other countries suited to the want-, ot the pra?inOO, 
tho fact was not overlooked that in some parts of 
the country superior varieties of indigenous crops arc 
grown, the seeds of which mirfbt with a lvantigubu 
