NATURE 
541 
THURSDAY, OCTOBER 5, 1899. 
BERTHELOT’S AGRICULTURAL CHEMISTRY. 
Chimie végétale et agricole. Par M. Berthelot. Four 
volumes. Pp. xvi + 511, vit 441, vi +517, vi + 528. 
(Paris: Masson, 1899.) 
HE ancient Chateau de Meudon, Seine-et-Oise, which 
was left in ruins at the end of the war of 1870, was 
thirteen years later converted into an agricultural experi- 
ment station by the French Government, and permanently 
attached to the Professorship of Organic Chemistry in 
the Collége de France. In the four bulky volumes now 
before us, the professor, M. Berthelot, has brought 
together an account of the various investigations carried 
out at the station between 1883 and 1899, under his 
direction, with the assistance in many cases of M. G. 
André. Besides these reports, the volumes contain an 
accomnt of several earlier investigations by M. Berthelot. 
We have in fact brought before us the whole of his 
investigations on plants, soils and various cognate 
matters, carried out during the last forty years. 
Agricultural chemists will heartily welcome the public- 
ation in sucha convenient form of this great mass of 
original investigation. M. Berthelot is well known as a 
first-class man of science, and as one of the most prolific 
and versatile workers of the present age. The new ideas 
he has brought forward concerning many obscure points 
in agricultural chemistry will be highly valued. Never- 
theless, those acquainted with the peculiarities of M. 
Berthelot’s work will not be surprised that a cautious 
critic is unable always to accept the conclusions to which 
he has apparently too easily arrived. The more startling 
and novel are the conclusions brought before us, the 
more thorough and unmistakable ought surely to be the 
basis of fact on which these conclusions are built. A few 
experiments, relating to only a part of the facts in ques- 
tion, must fail to carry conviction when new laws are 
propounded, or we are asked to surrender as a mistake 
views previously arrived at after much patient research. 
It will be gathered from what has just been said that 
the papers in the volumes before us are of unequal value. 
All the investigations are indeed highly suggestive, and 
no experimental investigator would desire one of them to 
be omitted ; but students of agricultural chemistry will 
not unfrequently find it advisable to examine with much 
care the evidence brought forward before accepting all 
the conclusions of the author. 
That we may do no injustice to the book we will, in the 
first place, call attention to the very valuable investigation 
upon the nature and properties of humus, which occupies 
more than one hundred pages in the fourth volume. The 
elucidation of the chemical nature of humus has been re- 
garded as an almost hopeless problem by the ordinary 
agricultural chemist. Berthelot has brought to bear 
upon the subject the methods and conceptions of modern 
organic chemistry, and his work has resulted in a 
considerable increase to our knowledge. 
Berthelot has carefully studied the composition and 
properties of the simple nitrogen-free humus obtained by 
boiling sugar with hydrochloric acid. It appears to be 
a mixture of a condensed anhydride and hydrate, the 
NO. 1562, VOL. 60] 
| simplest expression for the former being C,gH,,O,. It 
swells up in water, forming a colloid body. It absorbs 
| a considerable quantity of alkali from an aqueous solu- 
tion. One-third of the potash or soda thus absorbed is 
permanently retained in a practically insoluble condition 
after long washing with water. Placed in contact with 
ammonia an insoluble amido-compound is produced, 
from which ammonia is not recovered by boiling with 
magnesia. The oxidation of humus under the influence 
of light, and its more rapid oxidation in the presence of 
alkali are also studied. The heat relations of the prin- 
cipal reactions have also been ascertained. All this is 
fundamental work of very great importance, and throws 
much light upon the behaviour and functions of humus 
in a soil. 
The natural humus in soil is also studied, and the 
action of acids and alkalis upon it investigated. 
The gradual formation of ammonia when the nitro- 
genous humus of soils is boiled with weak acids, 
soluble nitrogenous compounds being simultaneously 
produced, is pointed out as in full agreement with the 
assumed amido nature of the humic matter. The humus 
of soils is, however, a very complex substance ; it may 
contain a very distinct amount of sulphur, and even 
phosphorus, in a state of organic combination. It will 
certainly be a novel fact for most agricultural 
chemists to hear that a soil may yield 0'183 per cent. of 
phosphoric acid when boiled with strong hydrochloric 
acid, 0°222 per cent. when the silica has been entirely 
removed by hydrofluoric acid and 0292 per cent. when 
the soil is burnt in oxygen gas and the products retained 
by sodium carbonate. The excess obtained by com- 
bustion in oxygen is regarded by Berthelot as represent- 
ing the phosphorus in organic combination. This part 
of the subject clearly requires much further investigation. 
Phosphorus, if present, is possibly a survival of the 
nuclein occurring both in the animal and vegetable 
kingdom. 
We take our next example from one of the less satis- 
factory of M. Berthelot’s investigations, in which the 
evidence brought forward seems quite insufficient to 
warrant the conclusions which he seeks to establish. 
He has determined the quantity of nitrates present in 
certain plants, and has conceived the idea that plants 
have the power of producing nitrates abundantly in their 
own tissues. This assumption, if proved, would clearly 
furnish an entirely new departure in vegetable physiology 
One would have thought that to establish such an 
hypothesis the plant would have been grown in a medium 
supplying no nitrates ; any appearing in the plant would 
then clearly be due to the work of the plant itself. M. 
Berthelot makes no such experiment. To establish his 
position, he grows the plant (borage or Amaranthus) in 
the open field, without any knowledge of the quantity of 
nitrates produced in the soil during the season of growth, 
and without taking into account the upward movement of 
subsoil water containing nitrates during the dry summer 
of his experiment. He is satisfied by ascertaining that 
on September 25 a square foot of soil contained only 
about 1/20 of the quantity of nitrate contained in the plant 
pulled up from it, and that a similar bulk of soil taken at 
the beginning of the season, from another part of the 
AA 
