should be recommended until it can be obtained to a definite specification, and the behaviour of its separate constituents is known.

3 and 4. — A somewhat different point of view is entertained by Stone and his pupils, of the Massachusetts Agricultural Experiment Station. They recognise the stimulant effects of soil sterilisation on plant growth, but they maintain that the benefits resulting from soil sterilisation depend upon the chemical and physical conditions rather than upon the number of protozoa.

Lodge and Smith have carried on three series of experiments with two types of soils, a loam rich in organic matter and a subsoil deficient in organic matter. 400 gr. of each soil were placed in a percolation tube and lukewarm distilled water was allowed to percolate several times through the soil; the percolated water was then placed in flasks, each flask containing 100 c. c. of percolate; then the decoction was subjected to steam pressure of 15 lb. for 45 minutes and at 121o C.

In series No. I a sterilised and unsterilised loam were used, and the sterilised decoctions inoculated with ordinary soil bacteria; the sterilised loam decoctions were found to contain a far greater number of bacteria than the unsterilised decoction. In series No. 2 a sterilised and unsterilised loam, and in addition a sterilised and unsterilised subsoil, were used, and the sterilised decoctions inoculated with ordinary soil bacteria; when a sterilized loam was used a greater number of bacteria were also found present than in the unsterilised loam decoction; but when a sterilised and unsterilised subsoil were used in the decoctions, a greater number of bacteria were found in the unsterilised decoction; this fact is assumed by the experimenters to prove that the sterilising of this particular soil resulted in adverse conditions for bacterial increase, and that decoctions made from different soils affect the growth of bacteria in a decidedly different manner. The remaining contents of the soil culture used in inoculating decoctions in the experiments of series Nos. I and 2 were subjected to a careful microscopic examination; no protozoa were found; neither were they observed to be abundant in a number of samples of the loam and subsoil. To avoid any possibility of introducing protozoa at the time of inoculating the decoctions, in the experiments of series No. 3 a sterilised and unsterilised loam and subsoil were used, as in series No. 2, but inoculations were made from a pure culture of Bacillus subtilis; the data given show that Bacillus subtilis multiplies in great numbers in all decoctions; a great number were found in the sterilised loam decoction as compared with the unsterilised and also a greater number were found in the unsterilised subsoil decoctions than in the sterilised decoctions.

The stimulating or retarding effects on the development of bacteria of the two types of sterilised soil used are similar to those produced in a previous experiment upon the growth of crops in these soils.

Growth of Soy Beans in Sterilised and Unsterilised Loam and Subsoil.

These experiments, made with a different soil, would not confirm those of Russell and Hutchinson, who maintain that protozoa influence the number of bacteria in soils, since the development of bacteria differs in soil decoctions according to the composition of the soil used; that is, the number of bacteria which develop in a soil would depend upon the chemical and physical conditions rather than upon the biological conditions. These experiments, however, do not necessarily preclude the idea that protozoa might play a much more important rôle in soils other than those experimented at the Massachusetts Experiment Station, Amherst, Mass.

PROVISIONAL CONCLUSION.

I. It can be fairly stated, that partial sterilisation, especially of sick soils, may prove an efficient method of stimulating the development of bacteria and therefore of plant growth.

II. It is not equally sure if the limiting factor of bacterial growth be a biological, or a chemical and physical property of the soil; perhaps this factor may differ according to circumstances.

15 - Researches on Alkaline Meadows and Pastures.

FLODERER, A. (Chemist at the Agricultural Experiment Station of Magyaróvár, Hungary). A békéscsabai szikes öntözött réten és legelön folytatott szik tanulmányokról. Országos m. kir növenytermelési Kisérleti állomás, Magyaróvárott. Kisérletügyi Közleményk, Vol. XV, No. 3, pp. 390-418. Budapest, July-September, 1912.

The Agricultural Experiment Station of Magyaróvár has made at Békés-esaba during the last 10 years a series of experiments on irrigated alkaline meadows and pastures (« Szik ») to determine whether irrigation diminishes the amount of alkaline salts, or in other words, if the alkaline soils can be thus modified and improved. To obtain detailed data respecting the distribution of the salts in the soil, the writer made several borings on the same plot and studied, not only the superficial layer, but also the strata at depths of from 30-120 cm. (12 to 48 inches) and estimated the salinity of these. The summary of his results was as follows:

Irrigation produces a certain amount of lixiviation of the harmful salts; the diminution in soda, the alkali which is the most injurious to plants, being especially noticeable.

The removal of the salts, which is effected chiefly at the commencement of irrigation, together with the increased water supply, explains the maximum yields obtained during the first three years. The plants

not only grew more vigorously, but the good species became predominant. The improvement takes place chiefly in the superficial soil. The richer the first stratum of 30 cm. (12 in.) is in salts, the greater is the improvement, for as the roots do not penetrate very far into stiff alkaline soils, the part played by the sub-soil is negligible.

In the case of a more intensive cultivation of the superficial layer, recourse must be had after some time to a systematic use of fertilizers. The manuring of alkaline soils only attains its ends if the former are improved to the point of growing the most useful plants in a satisfactory manner. Where the amount of alkali present is large in clay soils, only phosphatic and nitrogenous fertilizers are of advantage, potassic fertilizers are of no effect.

With regard to the powers of resistance possessed by various plants, sugar beets can grow in a soil containing from 0.15 to 0.2 per cent. of salts; but mangolds are able to develop in the presence of even larger quantities. Lucerne, on account of its long roots, can only be cultivated where the salts content is from 0.1 to 0.15 per cent., or at most, from 0.15 to 0.2 per cent. The presence of soda, however, greatly hinders its development. Lotus corniculatus, on the other hand, still grows in soils containing 0.2 to 0.3 per cent. of salts and 0.14 per cent. of soda; thus it resists alkalinity in the soil better than lucerne or red clover, the latter being a more delicate plant than lucerne.

Der Tro

16- The Composition of Soil Suitable for Rubber Cultivation. WOHLTMANN, F.: Südamerikanische und Ostafrikanische Kautschuk Böden. penpflanzer, Year 16, No. 11, pp. 571-581. Berlin, November, 1912. The writer, Director of the Agricultural Institute of the University of Halle, has analysed samples of the soils which were most suitable for growing rubber and which had been sent from Brazil ("Seringal”, Saõ Francisco, near Mt. Alegro), from Bolivia ("Seringal", Philadelphia, near Cobijia), from the higher basin of the Amazon and from German East Africa.

The following table gives some of the detailed results obtained by the writer.

In comparing the fertility of the soils of South America with those of German East Africa, it is necessary not to lose sight of the fact, that in the former country the annual rainfall is 120 inches, while in the latter it only amounts to 80 in. near Tanga and in East Usambara, and 48 in. in Morogoro. Thus, in the soils of South America the nutritive substances are more transportable and more easily assimilated than in the soils of German East Africa, which would only be as fertile as the former if they contained about 33 per cent. more of nutritive substances. The species of rubber trees cultivated are different in the two cases: Hevea brasiliensis, which requires an abundant rainfall, is cultivated in South America, and Manihot Glaziovii, which fares best in a dry climate, is grown in German East Africa. But the products obtained are chemically the same, whence it is believed that both species require the same, or nearly the same, soil.

This idea seems to be confirmed by the analytic results of the writer. He considers himself justified in deducing the following conclusions regarding the selection of a good soil for rubber trees :

I. Such a soil should be fine, of medium coherence, rather loose than heavy, and deep. The combined fineness and depth of the soil bring about a dampness, which appears to be indispensable to the formation of latex. Loams or clay loams with from 95 to 100% of fine particles are the soils to be preferred, those rich in laterite, or iron compounds, are apt to be deceptive, while compact clays, and especially soils with extreme characters, should be rejected.

2. With regard to the nutritive substances:

a) It is not necessary that the nitrogen content should be high; a large percentage of humus is perhaps downright injurious; 0.1 per cent. of nitrogen is sufficient, especially in districts with a large rainfall.

b) As for lime and magnesia, rubber trees only need very limited amounts of these substances. It is not yet known whether a high percentage of lime and magnesia hinders latex formation, or if the trees suffer in calcareous soils.

c) The rubber tree appears to have no special requirements as regards content in phosphoric acid.

d) It seems, on the other hand, that a large amount of potash in the soil promotes growth and the formation of latex; it is therefore ad visable to use fertilizers containing potassic salts.

17- Soils from the East Africa Protectorate.

Bulletin of the Imperial Institute, Vol. X, No. 3, pp. 405-422. London, Oct. 1912. Analyses of a certain number of soils from the East Africa Protectorate have already been published in the above Bulletin (1907, Vol. 5, p. 243), and a further series are now given

The results are tabulated in the accompanying table. The lime is deficient in almost all cases, except in the Jubaland soils. Phosphoric acid is also most consistently deficient whilst the nitrogen content varies considerably. The potash is abundant everywhere and an analysis of rock struck while sinking a well indicates that the source of these large supplies is andesite. In the Jubaland soils the soluble alkali salts are very high and would have to be taken into consideration in subsequent treatment. Applications of ground rock phosphate, ground limestone, and green manure are recommended to supply deficiencies of phosphates, lime and nitrogen respectively.