Sept. 1, 1898.] Supplement to the " Tropical Agriculturist.*' 
5,23 
specially well on it. The beans ground up make an 
excellent fertilizer for fruit trees, but are aho fit 
for table use. The vine is however specially re- 
commen'iedns a renovater of the soil when turned 
in ns greeii manures. 
Prof. Person, Chemist of the Florida Station, 
gives the following analysis of the bean : ash, 2 2d ; 
albuminoids, 21-.36; fat, 7'14; fibre, 8'46; carbo- 
hydrates, 60'7o. 
The florida bean is said to be specially adapted 
to sandy soils as has been repeatedly tested in 
Florida. Plots in which the beau has been grown 
and the vines allowed to wither on the land ai-e 
described as being covered with a mulch fully 4 
inches deep, leaving the soil on their decay of a 
rich alluvial character. One writer graphically 
describes his experience of the velvet bean as " the 
boss for making humus, and gathering the most 
expensive of fertilizers — ammonia. I would ad- 
vice every one to shade their laud from the hot 
sun and provide a fertilizer and humus for future 
use by planting their beans wherever possible." 
It is pointed out in the Queen-'^land Ar/ricultural 
Journal that the velvet beans has been wrongly 
named Dolichos multijioreif, aud f'at ilr, F. M, 
Bailey identities it as Mucuna pvuriens var. utihs, a 
variety of Cowitch. Other names for it are " Pea 
banana," " field pea,"' " Banana stock pea." 
COCONUT OIL CAKE. 
To judge from the pamphlet issued by Lever Bros, 
of Sunlight Soap fame, tiie refuse cake in the 
extraction of coconut oil for making Sunlight 
Sonp is bearing an exceedingly popular food for 
dairy stock. There are of course many kinds of 
oil cakes of which linseed, groundnut, giugelly, 
coconut, and kekuna are used as cattle food. In 
Ceylon the two best known are giugelly and 
coconut oil cake or " poonac " as they are gener- 
ally called. The former is a by-product m the 
manufacture of gingelly (sesamum) oil, and is of 
far higher value than coconut poonac as a food 
for milch cows. In fact it is the exception to use 
coconut poonac for milkiog stock, though it is 
the common food for working bulls. Horses, too, 
are never fed on coconut cake in Ceylon, the usual 
diet being CI ushed paddy (rice in the husk) and 
gram (the legumes of Cicer arieiimmi). The 
recognised foods for dairy stock are gingelly poo- 
nac and crushed cotton seed, which are often 
supplemented with bran and sometimes with 
"black gram" {Phaseolus mungo, var. radiatus), 
a very rich milch -producing food. A full daily 
diet for a good milker might thus be made up 
of 5 lbs. gingelly poonac, 2 lbs- cotton seed, 2 lbs. 
bran and 1 lb. black gram. Both the cotton seed 
and gram, which have to be ufed in moderation, 
and mixed with poonac and bran, are considered 
to " improve the quality" of milk, while again 
rice "conjee" or gruel is sometimes giveu to 
" increase the quantity " as it is said. 
Coconut, cake or poonac is generally classified 
as "chekku" and "mill" poonac, the former 
being the product of the JN'ative oil mill worked 
by cattle, and containing a larger percentage of 
oil than ihe latter, whicli is the by-product from 
machine-made oil, and is considered to be consider- 
ably inferior to t1\e other as cattle food. 
For pigs, poonac is considered to be an excel- 
lent food for fattening purposes. 
Though coconut cake is looked down on locally 
as food for milch cows, we are glad to see that it 
is very popular abroad, as tiiis fact should go 
to still further ensure the stabili(;y of what is always 
considered one of the safest agricultural enterprises 
in the Island, viz,, coconut ciikivation. 
THE ANALYSIS OF SOIL AS A GCIDE TO 
ITS FERTILITY. 
By Bebnard Dyeh, csc, f.i.c, Londox, 
The request that I should contribute this Daner 
is no doubt aue to the fact that, in 1894 I i d 
the honour of contributi„g to the TranUctious 
of the Cliemical Society, a paper of some c^n-ider 
able dimensions, " On the Analytical Dete-m'ination 
of probably available Mineral Plant-Food in Soils" 
which paper embodied the results of some vearV 
work on the subject, and involved an attemnt to 
approach it on somewhat new Hues The w k 
then recorded has been extended since its DubHo. 
tiou ; but although some reference will be mad« 
to uewly-obtaiiieJ and hitherto unpublished rP 
suits, It is nevertheless difEcult to render th« 
present article much more than a popular sum- 
mary of the more extensive paper to which refer- 
ence has been made. 
it is exactly five and twenty years a^o tha^ I 
began to analyse soils, under the guidnnce of t'l,« 
late Dr. Augustus Vcelcker, whose pupU ? w 
pMvilegea to be. The late Dr. Yoelcker's nrme 
IS not .specially connected with any brilliant 
scientific discovery, but it is a name that wil 
last lu the history of agricultural chemistry for 
the many essentially practical investigations with 
which It IS associated. He did not introduce 
agricii tural chemistry into England, but hp came 
from Germany to Edinburgh, as assistant 'to t^ 
late Professor Johnston, at a time when the va ue 
of agricultural chemistry appealed to compara- 
tively few farmers; and it is not too much to 
say, without loss cf respect to the memory of 
Johnston, or even of Liebig, whose lecture., trans- 
lated by Playfa.r, had already attracted atten- 
tion, that it was Yoelcker who acted as the chief 
pioneer of agricultural chemistry in the United 
Kingdom, dealing as he did, one after ano er 
with so many interesting problems of agriculture 
aud always in so practical a fashion thnV v, 
atmosphere of the laboratory never se^m d whh 
him to vitiate that of the open country Onl of 
the subjects I well recollect ou which he was 
never satished, although dogmatic scientific ortho- 
doxy claimed to have said its la.t word there.in 
was that 0 the worlds maintenance of it" stoek 
of vege able nitrogen. It wa, imu..s ble ^ 
gainsay the negatr. e experiment, of Boussincault 
and of Lawes aud Gilbert as to the -:on-a>1„n 
lability of atmo.pl.eric nitrogen by certain 
plants under certain conditions. Too iZ v 
caemists held these xperimeius as a general a d 
complete demonstration of the non-assimilabi?i y 
of atmospheric nitrogen ; and no little effort w,.^ 
made to adopt the theory of the non-assimih - 
bihty of atmospheric nitrogen to the existing facts 
of soil condition and productiveness, Whenever 
