Much work has been done in the past on the chemical composition of South African soils. The study of the physical properties of soils has, however, been somewhat neglected.

Transvaal soils differ from those of European countries in several important respects. In the first place, a large proportion of the soil is derived in situ from igneous rocks, and secondly, the country has never been glaciated in recent geological times. These two factors bear directly on the physical properties of the soils. Soil formation takes place chiefly by the chemical decomposition of rocks and only to a limited extent by mechanical disintegration. As a consequence we find in our soils very fine material, resulting from the chemical decomposition of easily weathered minerals, and rather coarse material which represents those rock constituents which are only subject in a limited degree to decomposition. In comparison with European soils particles of intermediate size do not play any dominant part in the mechanical make-up of Transvaal soils. Soils of alluvial origin are, of course, excepted. These latter occupy only a small area, but they are of considerable importance in the present stage of development of the country. Another factor which differentiates Transvaal soils from European ones is the presence in many of the former of finely divided hydrated ferric oxide. This feature will be discussed later.

The movement of water in soil, and the manner in which water is distributed over the particles and within the interstices, i.e. the dynamical and statical aspects respectively of moisture distribution, are of fundamental importance in soil science. The literature of the subject abounds with experimental determinations, but the difficulties of theoretical treatment are great, and it is only within recent years that any serious attempts have been made in this direction.

1. Certain colloidal properties of winter wheat plants have been found to be closely related to frost resistance, and provide indices of hardiness which may have practical application in the breeding and selection of hardy varieties.

2. The imbibition pressure of fresh leaves, in the winter-hardened condition, was found to be in most cases directly related to hardiness.

3. The volume of press-juice obtained per 100 gm. of hardened leaves was inversely proportional to the hardiness of a variety. This volume is largely determined by two hardiness factors, the dry matter content and imbibition pressure of the leaves.

4. When unhardened leaves were used, no relation could be found between imbibition pressure or volume of press-juice and hardiness.

5. The imbibition pressure of hardened leaves appears to depend on the physical state of the cell colloids characteristic of living tissues, since this property was lost when the tissues were killed.

6. The moisture content of hardened tissues tends to be inversely proportional to hardiness. There is some evidence that in hardy varieties it fluctuates less with changes in weather conditions, and a possible explanation of this as a colloidal property is suggested.

7. The quantity of hydrophilic colloids contained in the press-juice, as measured by an effect on the activity of water, was found to be directly proportional to hardiness.

1. The coeruleo-molybdate method of Denigès affords a rapid means of estimating phosphates in such dilute solutions as are given by aqueous extracts of soils, 1:5, even when the extract is diluted twenty-fold.

2. High phosphate values were found for bog peats, but the majority of the soils studied gave to the extract phosphate corresponding to under two parts per million of phosphorus pentoxide. Dunged soil gave 20 p.p.m. or more.

3. Extraction for 3–4 hours gives values as high as extraction for 4–7 days with ordinary soils of low phosphate content, but the phosphate of richer soils may undergo a reversion to an insoluble form during prolonged extraction.

1. The electrical conductivity of aqueous extracts of soils, in the proportion of one of soil to five of water, varies according to the time of extraction. In the more fertile soils extraction for three to four hours gives less than half as great a conductivity as is given in 4 to 11 days.

2. In peat the maximum value is reached almost at once, the extract having a high conductivity. The maximum value is also reached quickly in certain infertile soils, which give an extract of very low conductivity, closely similar to that of the purest upland streams.

3. A high electrical conductivity in the extract may only indicate the presence of an excess of salts, and does not necessarily indicate a good soil. It seems, however, that a rapid increase in conductivity as extraction is prolonged indicates increased solubility, partly through bacterial action and may be considered as a useful indication of fertility. A low conductivity, which remains low on continued extraction, denotes a soil so insoluble as to be unfertile.

It has been shown in earlier papers that the water present in soils is held in two ways: as capillary or interstitial water present as water wedges between the soil grains and as imbibitional or “gel” water which is associated in some way other than interstitially with the clay particles and with the colloidal coating of the soil grains. Imbibitional water is the cause of the swelling in water of soil and other colloidal (gel) systems such as cotton, wool, gelatin.

The nature of the imbibitional process is discussed and it is suggested that with clay, wool and soil imbibition may be due to the attainment of a Donnan equilibrium as is the case with gelatin. Experiments are quoted and discussed in support of this suggestion; and in particular it is pointed out that such swollen colloids appear to behave as perfectly elastic solids when under compression. This is not the case with sand and water nor with soil, wool or cotton in xylol, i.e. when no swelling occurs.

Imbibition may be a factor in the movement of soil water, in addition to capillarity, and it is shown that such a factor is quite consistent with the general mathematical treatment of the movement of soil water as developed by Gardner and others.

[It is realised that the point of view outlined above rests, as far as soil is concerned, on a very slender experimental basis, but it appears to be of sufficient interest to merit further investigation. The writer, however, has neither facilities nor opportunity for developing the matter further and this paper is published in its present form in the hope that it may prove of some interest to other workers more directly connected with soil research.]

An electrical method is suggested for reducing the friction between a metal surface and soil, or other moist substances; and its possible application to soil cultivation is considered.

Large reductions in friction were obtained in laboratory tests with a metallic slider moving over moist soil.

Preliminary tests in the field demonstrated that the draught of a plough was reduced by applying a current between the coulter and the mouldboard. The magnitude of the reduction obtained with this arrangement was too small to have immediate practical value, but the possibility of increasing the effect is discussed.

The measurement of the hydrogen-ion concentration of soils often presents great difficulties. When ordinary hydrogen electrodes are used, constant potentials are, in many cases, only obtained after hydrogen has been passed for several hours, while in many instances hydrogen electrodes cannot be used at all. The other important method of determining “pH,” the colorimetric method, can only be used in testing clear and almost colourless soil extracts, but not in testing soil mixtures.

Four years' additional experimental work on the application of a high tension discharge to the growth of field crops is described. Ten experiments with spring-sown oats and barley, two with winter-sown wheat, one with spring-sown wheat, and four with clover-hay have now been described.

The discharge was usually given at the rate of 0·5 to 1·0 milliamp. per acre from thin insulated wires stretched above the crop at a height of about 7 ft. and charged to a voltage of 40,000 to 80,000 (crest value). The discharge was usually given for 6 hours a day in two periods, 3 hours in the morning and 3 hours in the afternoon.

If the two experiments with spring-sown oats described in previous papers are included and the experiment of 1919 with spring-sown wheat (which yielded only 8 bushels to the acre) is excluded a series of 18 results is available spread over a period of 6 years. Of these 18 field experiments with various crops, 14 gave positive results in favour of the electrified plots, while 4 showed negative results, i.e. decreased yields compared with the controls. Of the 14 positive results only 3 show increases of less than 10 per cent, while 9 show increases of 30 per cent, and over, some reaching 50 per cent, and over. Of the 4 negative none shows a decrease of as much as 10 per cent.

Of the 12 experiments with spring-sown cereals 10 were positive and 2 negative. Of the positive results only 2 show increases of less than 10 per cent, while 6 show increases of 30 per cent, to 57 per cent.; on the other hand both the negative results are quite small, being 6 per cent. and 9 per cent. respectively.

The effect of electrification in increasing the yield of spring-grown oats and barley has thus been demonstrated. The mean increase in yield for such crops was 22 per cent.

A beneficial effect of the discharge on clover-hay is probable while the effect on winter-sown wheat is still uncertain.

Our knowledge of the proper conditions under which the discharge should be given is still so meagre that there is no reason to believe that the increased yields here described are the maximum obtainable as a result of electrification.

The mode of action of the current in producing increased growth and yield is still obscure. In several cases the electrified field crops showed a deeper green tint than that of the controls, and work already published has shown that in the case of the coleoptile (plumule sheath) of barley minute electric currents are able to bring about an increase in the rate of growth.

The effect of the discharge is of the nature of a stimulus. The additional energy available from the current is too small to have any direct effect since it is only about 1/1000th, or less, of the energy which the plant obtains from sunlight.

There is no evidence that gaseous products of the discharge play any part in the stimulation of growth. The additional supply of nitrogen brought to the soil from oxides of nitrogen produced as a result of electrification must be exceedingly small.

Pot-culture experiments carried out during a period of four years with Wheat, Barley and Maize show that these plants exhibit increase of dry weight when subjected to minute electric currents (of an intensity as low as 0·1 × 10–10 amp. per plant) from wire networks charged to a high voltage suspended above them. Of 28 sets in which the cultural conditions were satisfactory 23 showed an increased yield.

A percentage increase in dry weight of 27 ± 5·8 was shown by Maize plants grown under glass and little more than a month old. With Barley the largest percentage increase was 18 ± 2·4.

Direct currents were mostly studied, but increased growth was obtained with both direct and alternating current, the alternating current being apparently as effective or even more effective than the direct; the plants themselves, however, are able to bring about a slight rectification of the current.

Electrification of Barley for the first month of the growing season appears to be as effective as electrification during the whole growing season.

The discharging networks were usually charged positively, but a similar stimulating effect on dry weight production was obtained with a negative charge on the network.

Currents of the order of 1 × 10–8 amp. per plant and higher are injurious, causing a reduction of dry weight.

In one experiment plants screened from the normal atmospheric current by a series of parallel “earthed” wires (0·35 mm. diameter) showed, by comparison with the controls, a percentage decrease of 4·7 ± 2·3, which is hardly significant. It indicates however that though a favourable effect may be exerted by the normal atmospheric (air-earth) current, yet the effect can be but slight.

In one experiment with Barley indications of a differential effect of the discharge were obtained, for the increase in grain yield brought about by the discharge was greater than the increase in total dry weight.

The effect is to be.classed as of the nature of a stimulation since the energy-value of the additional plant material produced is out of all proportion to the energy supplied by the electric discharge.

These experiments with pot-cultures fall into line with those from the field (1) and from the laboratory(2); they leave no uncertainty as to the favourable action of the electric discharge.

Final yield of grain and straw reflect, or are functions of, the combination of processes conveniently designated “growth.” A successful analysis of “growth” would afford an analysis of “yield” and it is quite patent that “yield” must continue to be imperfectly understood so long as growth remains unanalysed. But although the full solution of the “yield problem” thus rests with plant physiology, it seems within the bounds of possibility that comparative studies upon varieties may bring to light distinctive (varietal) features of “growth” which are correlated with distinctive (varietal) features of “yielding power.” One or more of these may possibly serve as an “index” of yielding power, as the touchstone which is so urgently needed in plant breeding. The reliability of comparative studies of this kind rests fundamentally upon accuracy of “sampling.” For all the experimental attributes of freely tillering cereals, sampling is extremely difficult and the difficulties are most acute for “weight” attributes. Clearly then, a test of accuracy of sampling at all stages of growth is an indispensable preliminary. It was this consideration that determined the lines of the investigation to be recorded here and it is in the light of it that the results will be discussed.