Nov. 1, 1898.] Supplement to the " Trojncal Agriculturi&t." 
371 
Roxburgh states that the kernels taste very much 
like fresh walnuts and are reckoned wholesomt!. 
Indeed, those who have tasted them will at once be 
struck with their pleasant nutty flavour. Again, 
the oil is recommended as a substitute for castor 
oil — over which it has a considerable advanrage 
in possessing an agreeable flavour and taste — its 
action being certain and unattended with either 
nausea, colic, or other ill effects. 
The roots afford a brown dye used for dyeing 
cloth, while as before stated, tlie wood has a 
marked value. 
^ 
THE ANALYSIS OF SOIL AS A GUIDE TO 
ITS FEETILITY. 
BY DR. BERNARD DYER, F.I.C. 
{Continued.) 
Altogether about one hundred acidity determin- 
ations were made on some si.tty species of plants 
belonging to twenty different natural orders. 
The details of the process used for determining tlie 
acidity are fully set forth in my original paper 
in tiie Chemical Society's Journal, and as they 
are interesting only to laboratory workers, I need 
not burden the present article by a repetition of 
them. It should he said, however, that no 
attempt was made to identify the actual organic 
acids to which the acidity was due; but as it was 
proposed ultimately to use citric acid as the 
actual solvent for soil analysis, the acidity was 
in each case calculated in terms of citric acid. 
The reason for preferring citric acid to any 
other organic acid was, firstly, that citric acid 
is a very convenient material to use ; and secondly, 
because it was the actual weak acid successfully, 
as was considered, employed by Stutzer in 
gauging the availalDility of phosphoric acid in 
fertilisers. 
The average acidity, in terms of citric acid, 
shown by the hundred plants examined, was 
about 0B6 per cent. If, however, the plants 
coming under each natural order were averaged, 
and these averages again avei'aged, the twenty 
natural orders showed an average acidity of O'SJl 
per cent in terms of citric acid. Further analysis 
of the results showed that in the orders Ranun- 
culacece, Crucifer?e, Caryophyllaceoe, LeguminosEe, 
AraliacefB and Boraginacete, the averages lie between 
the limits of 0'81 and 1'12 percent. In the orders 
Tropffiolaceffi, Primulacere, Umbelliferffi, Compo- 
sittB, Campauulacete, Chenopodiacete, and Grami- 
nese, the average acidity ranged from 0'53 to 0 68 
per cent. 
In the case of the Dipsacese and Solanacea^, 
(single species only) the figures were below the 
average, viz,, 0'44 per cent and 0 34 per cent., 
while the plants examined belonging to the Liliacere 
averaged (J'36 per cent, though one of the plants 
e.xamined in this order was as liigh as '56 per cent. 
The plants e.xamined belonging to the Rosaceae, 
Plumbaginea) and Onagracefc gave high results. 
Of the OnagraceiB only two plants were examined, 
both of the same species — namely, ffinothera, or 
Evening Primrose. These both gave nearly two 
per cent of acidity, while two specimens of 
Arraeria (Thrift! (Plumhaginere) thowed over 
two per cent. Of the Rosacea; lour specimens were 
exomiued, uamely, two of CTeum aud two of 
strawberry. The latter both showed nearly two per 
cenc, while the rootlets of the two plants of 
Geum showed respectively 4-24 and o-53 per cent. 
With the exception of these orders, however, the 
results are not very far rem-ved from the 
averages already mentioned, and appe.ir to indicate 
that the sap acidity of the rootlets probably 
generally falls within, and not very far from one 
per cent, calculated as crystallised" citric acid. 
It should perhaps, for the sake of clearness, be 
here stated that by " sap acidity " is meant the 
ratio of acidity to the total moisture contained 
in the rootlets of the plant ; so that the statement 
that a plant has a root acidity of one per cent means, 
in more precise language, that 100 parts of the 
moisture naturally contained in the roots have 
such an acidity as would be arrived at by dis- 
solving one part of citric acid in 100 parts of water. 
As was pointed out in my paper, the results 
arrived at in this inquiry were of but a crude 
kind, and were obviously open to much criticism 
from a physiological standpoint, but, on the 
whole, it was considered that, however wanting 
the results might be in a scientific sense, they 
seemed to lend confirmation to the wisdom of 
Stutzer in adopting a solution of one part of citric 
acid in 100 of water as a standard test of tho 
availability of phosphates in fertilisers, though 
he appeared to have fixed on that particular 
strength by experiments based on quite other 
grounds. 
Whether or not such a solution could be use- 
fully adopted for extracting the available mineral 
plant-food from soils, could of course only be 
ascertained by practical experiment ; and such 
experiment could only be carried out on soils 
the relative fertility of which, as related to the' 
various ingredients of plant-food, was already 
known. The obvious spot to turn to for such 
soils was Rothamsted, and Sir John Lawes and 
Sir Henry Gilbert were good enough to allow me 
to draw, for the purpose of testing ths powers of 
the solution, a series of samples of soil from 
the well-known Hoosfield, which had borne barley 
for forty years, and in which, during that period 
each plot had year after year been subjected to 
the same manurial treitraent. 
The manurial treatment adopted in the Hoos- 
field barley experiments was, moveover, such as 
to render these soils particularly well-adapted for 
the purpose in view. 
These soils samples, twenty-two in all, were first 
analysed in the ordinary way, bv determination 
of the total phosphoric acid, and by determination 
of the potash dissolved by strong hydrochloric 
acid, the total potash in all forms "being also 
determined. 
A weight of air-dried soil was then taken from 
each sample, corresponding to 200 grammes of com- 
pletely dry snil, and each of such portions was 
treated in a " Winchester quart botr le with two 
litres of distilled water, in which 20 grammes of 
pure citric acid had been dissolved (that is to say, 
with two litres of a one per cent solution of citric 
acid). The soil was allowed to remain in contact 
with the solution for a week, with frequent 
agitation. At the end of that time the solution 
was filtered, and in each case a portion of solution 
corresponding to 50 grammes of soil was taktn 
for the determination of dis.solved j)hosphoj^(j 
