CHAPTER 34

Relation of Calcium to the Nodulation of Soybeans on Acid and Neutral Soils

AS A RESULT OF RECENT TESTS on the effects of calcium-bearing soil treatments, such as lime, acid phosphate, and calcium salts, upon nodulation of soybeans, there is an indication that the beneficial effect of these materials is essentially one of calcium stimulation. Frequent and repeated reports of failures to obtain nodulation of legumes by the pure culture method on certain acid soil types of northeastern Missouri and southeastern Illinois led to a search for the responsible factors. Soybeans, as an acid tolerant legume, have taken a prominent place in this territory within recent years, but thorough inoculation of even this crop has been difficult unless the soil was well limed. A study of the beneficial effects of lime on soybean inoculation, reported herein, leads to the belief that much of this is due to the element calcium.

Historical

Harper and Murphy, have recently given a review of the literature dealing with the factors affecting nodule formation by soybean plants. Other recent papers summarize the literature similarly, including specific phases of inoculation. Wilson, Scanlan, and Karraker, have each reported an increase in nodulation in consequence of the application of calcium as the carbonate or as a soluble salt. Alway, who was comparing the effectiveness of inoculation of alfalfa by soil transfer with that by pure cultures on lime-deficient sandy soils, found these two methods of equal efficacy when the land had been limed well in advance of seeding. But when the land had not been limed, the soil transfer method was far more effective. An increase in the amount of culture, many times beyond the usual rate, did not make it as effective as soil transfer. This seems to indicate an adaptation of the organism to a lime-deficient soil habitat in consequence of several previously grown, host crops. Bryan, in a study of the effect of acid soil reactions on nodulation of soybeans, found that, in general, the hydrogen-ion relations for the organisms tend to be the same as those for the host plant. He secured a maximum nodulation at pH 6.5 and none below 4.9, although the critical hydrogen-ion concentration for the organisms was found to be pH 3.5 to 3.9. Scanlan concluded that hydrogen-ion concentration must have no direct effect upon inoculation of soybeans by Bacillus radicicola. Although both calcium carbonate and calcium acetate stimulated inoculation tremendously, the carbonate neutralized the hydrogen-ion concentration and the acetate had no effect upon it. Fellers noted that the bacterial infection of roots did not take place readily on acid soils even when a good supply of bacteria was present.

It has been pointed out by Lohnis and Smith that bacteria undergo a fairly definite cyclic change. Bewley and Hutchinson found similar cyclic changes and that these are even specific with reference to cultural conditions. They stated that as long as there was sufficient available carbohydrate to support growth, the organism remained in the motile rod form with no changes. Their “pre-swarmers” (one of their cyclic forms); could be induced by the addition of calcium or magnesium carbonate to the medium. Of a considerable number of compounds other than carbohydrates, calcium phosphate alone was capable of bringing about the change from “pre-swarmers” to rods. The response to the reaction of the soil, in the main, was a rapid change from the normal rod form to “pre-swarmers” in calcareous soils; a production of a highly vacuolated form and the eventual death of the organisms in acid soil; and a continuous growth without significant change of form in a slightly alkaline soil. Thornton and Gangulee, in an even more recent investigation along this line, also noted a similar, regular cycle of change, in which unbanded rods, cocci, and banded rods successively followed each other. They stated that an increase in the percentage of cocci was associated with increased bacterial numbers and with the appearance of motile forms. They found that by modifying the liquid used to suspend the inoculum added to the soil, the time of appearance of cocci in predominance could be altered, and on this basis they recommend the inoculation of the soil with a bacterial suspension in milk to which is added 0.1 per cent of calcium acid phosphate.

Increased nodulation of soybeans, upon an already well-inoculated soil, resulted from the use of calcium treatment on such soil in the greenhouse. This was an acid soil (pH 5.5), taken from the experimental field, and already well inhabited with the symbiotic organism in consequence of inoculated crops of soybeans of the three preceding seasons. Limestone, equivalent to 4 tons per 2 million pounds, was added to part of the soil. The seed from a single mother plant was used for a stand of 5 plants per pot on 30 pots from both the limed and the unlimed soil. They were grown for 5 weeks, after which the uniform and healthy plants were taken up and the nodules counted. Although Erdman, has presented an argument for the importance of the size as well as the number of nodules in determining the effectiveness of inoculation, the numbers only were taken. The data are summarized in the first half of table 1.

The results show an increase of 336 per cent in numbers of nodules formed as a result of liming, even though the soil was already well inoculated with the organism. In spite of the fact that these data show a correlation between inoculation and neutralization of the soil acidity, this does not necessarily establish a causal relationship.

Part II

In consequence of the fact that an application of calcium as the carbonate, produced an important increase of nodulation on an already well-inoculated soil, and of the belief that this stimulation was not necessarily the result of a change in hydrogen-ion concentration, or at least not entirely so, a test was made of the effect of calcium, as the chloride, upon the inoculation of soybeans by the pure culture method on an acid soil which was sterile with regard to B. radicicola of soybeans. To each of 30 pots of a rather heavy Union silt loam (pH 5.4), there were added at planting time 25 cc. of a solution of calcium chloride supplying calcium equivalent to that of 200 pounds of calcium carbonate per 2 million of soil. Thirty pots of the untreated soil were planted also. Liberal quantities of an inoculating suspension were supplied directly to the beans at the time of planting. Eight pots of untreated and uninoculated soil were planted as checks. When, at the end of 5 weeks, the examination for nodules revealed none formed, 16 pots each of the inoculated, calcium treated and of the inoculated, untreated soil were immediately replanted with sprouted beans in the same pots. After 5 weeks of growth, these were likewise carefully taken up, washed, and examined for nodules. The complete data are summarized in the latter half of table 1.

The nodulation of the beans of the second planting indicates that the calcium as a chloride has a stimulating effect upon the longevity and viability of the organism in the acid soil. This agrees with Scanlan’s results of increased viability of the organism in water cultures.

Part III

Since the addition of a small amount of a neutral calcium salt to an acid soil (pH 5.4) had kept B. radicicola viable within the soil from the time it was applied by pure cultures on the first crop until the second planting, and since an addition of large amounts of calcium carbonate produced important increases in nodulation of soybeans on an already well-inoculated soil, it was thought possible to determine whether this stimulating effect was due, (a) to calcium within the plant, (b) to an effect of calcium upon the organism in the soil, or (c) to an effect of calcium upon the soil as the habitat of the organisms, by growing plants so arranged that one part of the root system of each was growing in an acid soil and the other part in the calcium-treated soil.

Soybean seedlings were grown in sterile sand from 10 to 14 days, or until the lateral roots about an inch long were sufficient to support the plant. These seedlings were taken up, washed, and the tap roots cut off just below the longest lateral roots. They were then planted with half of their roots on one side and half on the other side of the water-tight partition of a two-compartment pan. The moist acid soil was filled in around the roots on one side, and moist calcium-treated soil on the other. Liberal quantities of the inoculating suspension were supplied directly to the roots. The soil used was a Putnam silt loam (pH 5.14). Five plants were planted with their roots divided by the partition and five more plants with their tap roots similarly pruned were planted wholly on each side of the partition as checks. Although there was a high mortality of plants, those that lived grew satisfactorily for the five weeks, after which a count of the nodules was made. The results are given in table 2.

The total yield of beans from a single plant was sterilized, germinated for 24 hours, and planted into pots of these sterile substrates, half into the calcium-free and half into the calcium-bearing sand. The pots were set to their shoulders into moist soil, which was sterile with regard to B. radicicola, and after 10 or 11 days the plants were taken up, washed, and replanted to the inoculated soil which had been prepared and silted at a suitable moisture content into ordinary greenhouse flats. Both the calcium-bearing and calcium-starved seedlings were grown simultaneously on their respective halves of the same flat and within the same soil. Seedlings so treated were transplanted and grown on two different soils, one an acid, lime-deficient soil, and the other, a neutral, fertile, garden soil. No inoculation was added, since both soils had grown well-inoculated crops of soybeans during two consecutive seasons just previous to this test. After a growth of 5 weeks, the plants were taken up readily without injury to the roots and the nodules counted. The data from the count are summarized in table 3. Included in the table are also the analytical data giving, (a) the calcium content of the soybean seeds from a single, similar plant, (b) the calcium content of 10-day-old calcium-bearing and calcium-starved seedlings, and (c) the total electrodialyzable calcium of the soils as determined by Bradfield’s method of measuring the total electrodialyzable base. These analyses were made in order to correlate the nodulation with the calcium content of the plants as influenced by the treatments.

The increased nodulation of the calcium-bearing seedlings on the calcium-deficient, acid soil demonstrates that the presence of calcium within the plant increases nodulation of soybeans on such soil. On the other hand, the lack of difference in nodulation of the seedlings on the neutral, calcium-laden soil indicates that the presence of this element within the plant on a calcium-sufficient soil does not affect nodulation, or that if it does, the calcium-starved plants are able to take calcium from the soil rapidly enough to offset the measurable differences in nodulation.

The increase in calcium content of the calcium-starved seedlings over that in the seed, as shown by the analytical data, was due to the calcium that was carried back into the acid-extracted sand by the tap water with which it was washed. An elimination of this factor might have served to intensify further the differences obtained.

Part V

Since calcium exerted an intimate effect upon inoculation by the organism B. radicicola, this effect was deemed possible through an inter-relation with the soil colloids, the main chemically reactive part of the soil. It is known that the colloids are highly absorptive, and that minute quantities of calcium are effective in flocculating them, hence this phase of the experiment was undertaken to detect such possible relation.

A suspension in distilled water of the organisms from several agar cultures was added to a 0.4 per cent solution of colloidal clay in a ratio of four parts of the bacterial suspension to five parts colloidal solution. To this mixture was added one part of water containing the desired amount of flocculating agent.

Those mixtures left unflocculated received the equivalent of distilled water. Thus the resulting solutions contained 0.2 per cent colloidal clay and equal numbers of organisms throughout. These were made up in units of 100 cc. in test tubes. The chlorides of potassium and calcium were used as comparative flocculating agents at the rate of 0.5 milliequivalents, or the minimum requirement of potassium chloride as electrolyte at this concentration of colloidal clay. Mixtures of the organism at the same concentrations in the natural colloid, in distilled water, and in tap water were set up as checks.

* Duplicate pots were grown with 5 plants each.

The tubes were incubated for 7 days, after which the liquid supernatant to the flocculated clay, the flocculated clay itself, the natural colloidal clay, and the suspensions in water, were tested for the presence of the viable organism by applying specific quantities to sterile, germinated soybeans as they were planted into sterile sand. At the end of 5 weeks the plants were taken up and the nodules per plant counted. The data are presented in table 4.

The results obtained in this experiment were duplicated almost identically in a repetition of the experiment 6 weeks later. The nodulation obtained indicated that the colloidal clay absorbed the bacteria but did not destroy their viability. Flocculation of the clay with calcium chloride carried the organisms down and retained them within the flocculant. This was not the case for the potassium chloride. In the potassium chloride the supernatant was as effective for inoculation as the flocculant.

Table 5

Plate Counts of Bacillus radicicola as Influenced by Colloidal Clay Treatments

Portion of treatment sampled	Average count per cubic centimeter
Supernatant to calcium flocculant……………….…	8
Supernatant to potassium flocculant……………..…	13,100
Upper half calcium-bacteria suspension………..…	200
Upper half potassium-bacteria suspension…….…	7,000
Lower half calcium-bacteria suspension….……….	4,300
Lower half potassium-bacteria suspension…….…	1,000
Supernatant to centrifuged inoculated colloid……	22,100
Inoculated colloid—not centrifuged…………………..	9,550,000
Tap water suspension…………………………………..…	3,600

In order to verify the accuracy of this test and to determine whether the calcium chloride or the colloid is the active factor in carrying the organisms out of suspension, this experiment was repeated. Platings were made from the solutions into sterile petri dishes at the time of planting. Also, the pure bacterial suspensions were flocculated by potassium chloride and calcium chloride and then plated. A sterile colloidal clay inoculated 7 days previously was also plated. The effectiveness of the absorption of the bacteria by the colloid was tested by centrifuging the colloidal material out of an inoculated colloidal clay and then plating the centrifuged solution. The counts are given in table 5.

The relation of the calcium to the retention of the bacteria by the flocculated clay, as previously found, was substantiated in this trial. The liquid, supernatant to the potassium chloride flocculant contained over 13,000, whereas that over the calcium chloride flocculant contained but 8 bacteria per cubic centimeter, showing that the calcium flocculated clay carried the bacteria out almost completely whereas the potassium flocculated clay did not. Calcium used independently of the clay, carried out the bacteria, since the water suspension given potassium chloride contained a count of 7,000, whereas the treatment with calcium chloride reduced this to 200 per cubic centimeter. In comparing the inoculated colloid suspension with the same after centrifuging, the number of about ten million in the former was reduced to about 22,000 in the supernatant in consequence of centrifuging.

These data suggest that though the clay carries the bacteria out of suspension, certainly the calcium does likewise, whether used alone or whether combined with the clay colloid. When the calcium is used in conjunction with the clay, however, a more nearly complete removal of the organisms is obtained. This is no doubt due to the simultaneous coagulation of the bacteria and to the flocculation of the colloid. This does not hold true for the potassium chloride.

Part VI.—Field Trials

After finding that applications of lime may stimulate nodulation on an acid soil already inhabited by the organisms, and that small amounts of calcium in the soil, as well as small amounts within the plant tissues, are important in stimulating nodulation, it seemed quite plausible that liming a soil may exert its influence not wholly as a secondary effect through the correlation of hydrogen-ion concentration, but also in consequence of its content in calcium. Work was done in the field to test whether small amounts of soluble calcium with no neutralizing capacity would improve nodulation by the pure culture method on acid soils that were difficult to inoculate without liming.

The work was done in cooperation with farmers experiencing difficulty in getting inoculation on unlimed land. The soybeans were inoculated at planting time with a tested strain of the organism, and applications of calcium chloride, calcium nitrate, calcium acid phosphate, and calcium hydroxide were made through fertilizer attachments on the seeding machinery. In addition to these, tests were made using inoculated soil, both with and without calcium salt treatments. The salts, including calcium chloride and calcium nitrate, were mixed, as a 2N solution, into the dry pulverized soil and the soil was then dried until it would operate through the fertilizer attachment. A determination of the nodulation was made when the beans were at full growth, just shortly before maturity. Samples of the soil were also taken then for hydrogen-ion measurements and for determinations of dialyzable base and calcium. Tables 6 and 7 give summaries of the data on nodulation in these field trials.

The data show that the culture inoculation was successful on the Marion Silt Loam in every case where it was supplemented by applications of small amounts of calcium salts, but failed wherever the calcium was omitted. These differences were very noticeable in the color of the plants. Soil inoculation was successful on this soil type without added calcium. The inclusion of calcium, however, increased the size of the nodules significantly. The lessened number of nodules per plant, when calcium was added to the inoculating soils, suggests possible death to the organisms by this salt treatment, though this is not significant enough to reduce the percentage of plants inoculated.

On fields other than the Marion Silt Loam, the culture used alone was successful without special treatment. However, in many cases the addition of the calcium, especially the hydroxide, which distributed itself more thoroughly on account of its fineness, gave increased nodules per plant, and increased the percentages of infection. On these fields the color differences were less pronounced than on the Marion Silt Loam, but yet significant differences in growth were evident.

Determinations of the hydrogen-ion concentration revealed a pH of 6.0 on the Marion Silt Loam where the calcium was beneficial to inoculation, and a much lower figure for the pH where calcium was less effective. Just what relations exist between the hydrogen-ion concentration and the effectiveness of the applied calcium, or between the effectiveness and the electrodialyzable base or calcium, is still a question. The total electrodialyzable base and electrodialyzable calcium content seem to decrease as the calcium additions were less effective.

In an attempt to determine the relation of dialyzable base or calcium of the soil to inoculation, the flat phase of the Putnam Silt Loam of the field trials was used in the greenhouse. Seedlings were started for 10 days in calcium-deficient and calcium-laden substrates and then transplanted to this soil given no treatment, given calcium carbonate equivalent to 5 tons per acre, and given calcium chloride at the rate of 1 part per 1500 parts of soil solution, considering the soil at 25 per cent moisture. Thorough inoculation was applied at planting and the plants were grown for 5 weeks, when they were examined for their nodulation. Analyses were made of the seedlings for their calcium content, and of the soil for the total electrolyzable base and calcium. The hydrogen-ion concentration was also determined. The complete data are given in table 8.

Though no statistical manipulation was undertaken to express the reliability of the data in the usual way, it is interesting to note that even though this soil gave no great improvements in its inoculation through calcium applied in the field trials, a significant increase occurred in the nodule numbers when the seedlings carried a liberal calcium supply. This difference was obliterated when the soil was given calcium, either as carbonate or as chloride. No correlation seemed to exist between the electrodialyzable calcium and the nodule numbers per plant. However, it is interesting to note that the insoluble calcium carbonate was less effective in increasing nodule numbers per plant, than was the soluble calcium chloride or calcium within the seedlings, for the short time of this trial.

Measurements of the electrodialyzable calcium within the soils in this study were not numerous enough to establish whether or not this quantity of the element might serve as a possible indication of the soil’s deficiency in calcium with reference to inoculation. Further data of this kind will be necessary to decide the question fully. However, the data thus far suggest that electrodialysis is scarcely a criterion as to whether or not the soil will yield sufficient calcium to guarantee thorough inoculation or whether added calcium mightimprove the establishment of the relation between nodule bacteria and their host plant. Under certain conditions, certainly, very small amounts of calcium are beneficial in establishing thorough inoculation and consequently the legume crop itself. This possibility might be inferred from the work of McCool and by the report of Jaeger. How small this amount may be in any case is still a question. Attention may need to be given to ionizable calcium or some other forms before soil analysis can contribute a simple chemical answer to this complex question of biological behavior.

Summary and Conclusions