Hawaiian Planter's Record 15(3):155-160. September, 1916
THE FIXATION OF ATMOSPHERIC NITROGEN BY SOIL BACTERIA WHICH LIVE INDEPENDENT OF A HOST PLANT.*
*Sugar. June, 1916.
By WM. L. OWEN
Bacteriologist Sugar Experiment Station of the Louisiana State University.

CHAPTER 1.

HISTORICAL.

In the previous series of, articles we discussed some of the various phases of the fixation of atmospheric nitrogen by legumes, or what is known as "Symbiotic Fixation." We learned that the fixation of nitrogen in that case is the result of the joint activities of the legume and the bacteria that inhabit the nodules upon its roots. In the non-symbiotic type of fixation which we are now to discuss the bacteria live and carry on their beneficent activities entirely independent of a host plant.

It seems strange that the discovery of the non-symbiotic type of fixation actually antedated by several years that of the nitrogen-fixing power of legumes. The role played by the legume in enriching the soil is now so much more generally understood than the non-symbiotic type that it would be naturally supposed that the scientific facts in connection with the former had been known to the public for a much longer time. Almost everyone today thinks of the legume in terms of its ability to add nitrogen to the soil while only a few ever give any thought to the operations of that highly important group of soil bacteria which add atmospheric nitrogen to the soil without the cooperation of higher plants. These bacteria are of just as much importance to the soil as the legume bacteria are and their activities in the soil were understood before the nodule bacteria were known and yet in spite of this fact their importance seems to pale into insignificance in the opinion of the average farmer as compared to the much more esteemed "symbionts" of the legumes.

Although there is even today but scant attention paid by the farmer to the operations of these invisible microscopic friends in his soil, yet the benign effect of their activities has been noted from the earliest of times. In 1858 Bousingault, the famous French chemist, wrote as follows:

Vegetable earth contains not only dead organic matter but living organisms, germs, the vitality of which is suspended by drying and reëstablished under favorable conditions of moisture and temperature.

About twenty-five years later another celebrated French chemist, Berthelot working at the experiment station at Meudon, France, undertook a thorough study of soils with particular reference to their increase in nitrogen. These studies resulted in the first authoritative data on the nitrogen increase in bare unsterilized soils. The results obtained by Berthelot were soon confirmed by a host of other investigators among whom may be mentioned Koch, Schloesing and Laurent. Berthelot (lid not rest content with having proven that soils when uncropped do increase in nitrogen but endeavored to isolate the causative agency in the process. He succeeded in isolating a number of species of bacteria. some of which he showed to be capable of fixing considerable quantities of nitrogen in culture media. Unfortunately Berthelot did not describe the species of bacteria which he had isolated and with which his experiments were conducted. Their identity therefore can only be inferred, although it appears very probable that they were members of the group of nitrogen-fixing bacteria that are now well known. To Berthelot, however, belongs the distinction of having first proved that certain soil bacteria are capable of adding nitrogen to uncropped soils and his classic researches undoubtedly paved the way for the more complete work which shortly followed.

The work of Berthelot's was taken up in a more systematic manner by Winogradski who began his experiments about 1893. The investigations of the latter began with a study of the butyric acid bacteria commonly found in the soil. He found that this species could fix appreciable quantities of nitrogen in culture solutions deficient in nitrogenous substances. By employing culture solutions made up with distilled water and certain nutritive salts together with small quantities of dextrose and adding a small particle of soil as inoculating material he obtained considerable nitrogen fixation. In the culture solutions inoculated in this manner he discovered a large spore bearing rod-shaped bacteria which formed butyric acid in considerable quantities. The butyric acid bacteria seemed always to be accompanied by other species which seemed to have no influence upon nitrogen fixation in the solutions. Winogradski later succeeded in isolating this species in pure cultures and, incidentally, discovered that it was anaerobic in nature (that is thriving best in the absence of oxygen). On account of this property the growth of this species in aerated soils or in solutions, is favored by the presence of other species which demand oxygen for their development. In fact when air is present the butyric acid bacteria can only live when other oxygen consuming species are at hand to utilize this element. Winogradski, in his very thorough study of this species. found that its nitrogen fixing ability was influenced by the amount of combined nitrogen already existing in the nutrient solution. He found the maximum nitrogen fixation was attained when there was no combined nitrogen in the culture solution. In this case there were 3.1 milligrams of nitrogen fixed for every gram of dextrose originally added to the solution. When the solution contained 1.8 milligrams of combined nitrogen the nitrogen fixation was 1.7 milligrams per gram of dextrose, while with 4 grams of combined nitrogen present. there was only 0.6 of a milligram of atmospheric nitrogen added to the solution by the species of bacteria in question. Not only was the fixation of atmospheric nitrogen found to be directly influenced by the amount of combined nitrogen already in the solution but it was further shown by Winogradski that the nitrogen increase was always in a definite ratio to the amount of dextrose in the solution. When the ratio between combined nitrogen and dextrose is high, an actual loss may take place, as for example when there were 21.2 milligrams of nitrogen per gram of dextrose. In order for there to be any gain in nitrogen the ratio between combined nitrogen and dextrose should not be higher than 6 to 1,000.

From this point, where Winogradski's epoch-making investigations paved the way for a more general conception of non-symbiotic fixation in soils, the attention given to this phase of soil fertility was naturally very great. In connection with the contemporaneous observations of that period regarding the increase of nitrogen in uncropped soils, the experiments of Caron in Germany are deserving of special mention. Caron, who was the owner of the large Ellenbach estate in Germany and who was much interested in soil bacteriology, observed that samples of soils taken from lands on which leafy crops were grown contained a greater number of bacteria than samples taken from other soils, especially from those on which grain was grown. He also observed that soils contained more bacteria in the spring than in the autumn, and found the greatest increase in numbers in soil samples from a summer fallow. Caron isolated a number of species of soil bacteria, some of which were employed in pure cultures for the inoculation of pot experiments. The results of his inoculation experiments, although quite variable, showed in many cases a very large increase in favor of the inoculation. In some cases this gain was as much as 195 to 100 in favor of the inoculation while in others it was only 139. Caron named the most efficient of these species of nitrogen fixers, Bacillus ellenbachensis, after his estate. His investigations led to the exploitation of the species of bacteria which he had isolated, by a commercial firm in Elberfeld, which gave the preparation the name of "Alinit." The most extravagant claims were made for this preparation by its manufacturers and in fact it was expected to revolutionize over night the whole method of supplying nitrogenous fertilizers to crops. The manufacturers claimed that by the use of their "Alinit" non-legumes could fix atmospheric nitrogen as well as legumes, and that this preparation would enable the farmer to entirely dispense with expensive nitrogenous fertilizers. "Alinit" was expected to mark the beginning of an entirely new era of crop fertilization but unfortunately for the agricultural world it was found to be without any special merit and was recognized as an incipient fraud. It is true that some investigators did publish results tending to confirm the claims of the manufacturers of this preparation but these were coincidences rather than proof of the merit of the article for the consensus of opinion among those who investigated "Alinit" was, that it possessed no particular merit as a fertilizer. A most important contribution to the subject of non-symbiotic nitrogen fixation was made in 1901 by Beijerinck, whose name will be remembered in connection with his priority in the discovery of the nodule bacteria of legumes. Several very important species of nitrogen-fixing bacteria were first isolated by that investigator and indeed those species of bacteria are still regarded as of first importance in the fixation of atmospheric nitrogen. From his very extensive investigations of the species which he had isolated, Beijerinck concluded that they could not fix nitrogen when grown in pure cultures but that it was essential for this purpose that other species be associated with them. From the results obtained in his experiments Beijerinck was led to conclude that this association between the active and inert species could result in either one or both species acquiring a greatly increased nitrogen fixing power. On the other hand the work of Gerlach and Vogel proved conclusively that the nitrogen fixing species of Beijerinck could fix large quantities of nitrogen even when growing in pure cultures. The truth of this assertion has been amply confirmed by the investigations of many other bacteriologists.

The conception of the fixation in the soil of atmospheric nitrogen without the aid of legumes was also strengthened by the cumulative evidence of field observations. Among these earlier observations of field experience those of Kuhn are of great interest. He reported in 1901 that he had grown non-leguminous crops for twenty years upon a soil to which no nitrogenous fertilizer had been added without any apparently diminished crop returns. Similar observations were reported at the celebrated Rothampsted Station where fields abandoned between 1882 and 1904 showed an annual gain of more than 100 pounds of nitrogen per acre. Regarding these results Hall, the director of the Rothampsted Station, inquires as follows:

How comes it that the Geescroft soils with no plants growing on them which are capable of fixing free nitrogen have yet gained an enormous quantity of nitrogen during the twenty years under review, a quantity which at the lowest reckoning amounts to about twenty-five pounds per acre per year? The nitrogen brought down in the rain would account for perhaps five pounds per acre per year, a little more would come in the form of dust, bird droppings and other casual increments while some may be due to fixation of atmospheric nitrogen by bacteria in the soil not associated with leguminous plants like the Azotobacter chroococum of Beijerinck and Winogradski's Clostridium pasteurianum. The Azotobacter has been found abundantly in the Rothampsted soils and as in the ease of grass lands like the present, the decaying vegetable matter would supply the carbohydrate which the bacterium must oxidize in order to fix nitrogen, it is quite possible that it may have effected considerable gains in nitrogen.

Lipman and Voorhees called attention to the fact that the gain in soil nitrogen as reported by Hall must actually have been much greater than the soil analyses indicated for, as they point out, considerable amounts of nitrogen must necessarily have been lost in the course of the decay of the vegetable matter in the soil. These same authors showed that in their numerous experiments along this line it was observed that there was a gain of nitrogen equivalent to one-third of the nitrogen originally present or supplied, and that this increase often took place within the time included in two short growing seasons. This increase in soil nitrogen was attributed by Lipman and Voorhees to the activities of the non-symbiotic type of soil bacteria. From that time to the present day the confidence in the ability of those species of soil bacteria to appreciably increase the nitrogen in the soil has gained a firmer hold in the minds of the agriculturists and indeed the time seems not far distant when the farmer will so fully appreciate the beneficial activities of these bacteria that he will concern himself with providing them with more favorable conditions for their development. [H. P. A.] (To be continued.)

Hawaiian Planter's Record 15(4): 212-225. October, 1916
THE FIXATION OF ATMOSPHERIC NITROGEN BY SOIL BACTERIA WHICH LIVE INDEPENDENT OF A HOST PLANT.*
*Sugar, July and August, 1916.
By WM. L. OWEN
Bacteriologist Sugar Experiment Station of the Louisiana State University.
 (Continued from the September issue.)

CHAPTER II.

THE SPECIES INVOLVED.

Although there are quite a large number of species of soil bacteria which have the power of fixing small quantities of nitrogen in the soil, there are only two groups now known to be of any great economic importance in this connection. The first species to have its nitrogen-fixing power demonstrated was the butyric acid bacillus which is widely distributed in the soil. As we learned in the previous article, it was Winogradski who first determined the nitrogen-fixing power of this organism when grown in pure cultures. He found that the amount of nitrogen that could be fixed by this species was dependent upon the proportion between the dextrose and combined nitrogen already in the solution. The most favorable ratio between dextrose and combined nitrogen was found to be 1000:6, while the average amount of nitrogen fixed by this species was 1.5 to 1.8 milligrams per two grams of sugar used by the organism as a source of energy. It is an interesting and significant fact that the nitrogen-fixing power of this species is influenced by the amounts of combined nitrogen already available. Thus Winogradski showed that the amount of nitrogen that was added to the solution by the activities of this species decreased in proportion to the amount of combined nitrogen already in the solution. He showed that when the solution contained as much as 21.2 grams of nitrogen per gram of dextrose the bacteria not only did not fix any nitrogen but an actual loss of nitrogen occurred. The fact that the amount of nitrogen taken from the atmosphere by these bacteria is limited by the amount of this substance already in the soil is of great practical importance. It is this influence that limits the utilization of atmospheric nitrogen to only those soils which are deficient in nitrogen. In other words, it is one of Nature's economies similar in operation to what we have learned of the effect of nitrogen upon the activities of the nodule bacteria.

In that case we learned that where there was much assimilable nitrogen in the soil the fixation of atmospheric nitrogen by the legume was greatly decreased.

As regards the efficiency of the various nitrogen-fixing bacteria in the soil, the butyric acid bacillus which we have already described is greatly inferior to the second group which we shall next consider. It was Beijerinck, again, who contributed another most important chapter to soil bacteriology when in 1901 he isolated the group of bacteria known as Azotobacter. Two of the species of this group, Azotobacter chroococum and agilis, he studied and described. Lipman of the New Jersey Station isolated two more species of this group in 1903, viz., Azotobacter vinelandii and beyerinckii, and to this list he made another contribution in the following year, when he isolated the fifth species, viz., Azotobacter woodstownii. Beijerinck and Van Delden, who first studied the nitrogen-fixing power of these species, were of the opinion that they could only fix very small quantities of nitrogen when growing alone, but were able to fix their maximum amounts when associated with other soil bacteria. Gerlach and Vogel, however, found that Azotobacter chroococum is capable of fixing large quantities of free nitrogen when grown in pure culture. The nitrogen fixed by the Azotobacter group is, as we have already noted, much greater than that fixed by the butyric acid bacillus. As we have already stated, Winogradski got an increase of 1.5 milligrams of nitrogen for every two grams of sugar originally present in the solution. Gerlach and Vogel, on the other hand, found the following amounts of nitrogen fixed by the Azotobacter species:

Grams glucose per liter Milligrams of N. increase N. increase per gram, glucose
1 7.4 7.4
2 13.5 6.75
3 17.3 5.76
4 31.4 7.85
5 39.4 7.88
6 45.9 7.65
7 59.9 8.57
10 91.4 9.14
12 127.9 10.66
15 62.9 4.1

In the series containing 12 grams of glucose per liter the greatest nitrogen fixation took place, and when more than this amount of glucose was used the nitrogen fixation was relatively much smaller per gram of sugar. It appears, therefore, that glucose to the amount of 1.2 per cent is most conducive to the greatest nitrogen-fixing power of these species.

There are many other species of soil bacteria besides the two groups which we have mentioned that have some nitrogen-fixing power, but their power of increasing the soil nitrogen is comparatively small and need hardly to be considered from an economic standpoint. It is of interest, however, to note that the bacteria causing the deterioration of sugars. and which constitute a group of soil bacteria, are likewise credited with the ability of utilizing atmospheric nitrogen. Beijerinck experimented with the nitrogen-fixing power of this group of bacteria and noted a considerable amount of nitrogen increase. It is perhaps owing to their ability to utilize atmospheric nitrogen that they can subsist in substances containing so little nitrogen as is true of white sugars.

Lohnis has likewise demonstrated the ability of the Bacterium pneumoniae to fix small quantities of nitrogen in the solution in which it grows, and we referred in a previous article to the fact that the nodule bacteria have the power of fixing small quantities of free nitrogen even when living independent of a host plant.

We next come to the consideration of the extent of the distribution of these beneficial species of bacteria in the soil. Do they occur in all soils or are only a few types of soil favored by their presence? Investigations have shown that these species are very widely distributed in the soil, and especially is this true of the butyric acid bacillus, which appears to be much more widely distributed than the Azotobacter. The writer has found in a few preliminary experiments that the former seems to predominate in the soil at the Sugar Experiment Station. An interesting and highly significant fact in connection with the natural occurrence of the nitrogen-fixing bacteria is that the application of lime to soils seems to help to establish these bacteria in soils, in which they are previously inactive. Fischer has shown that the application of lime to soils which were very low in nitrogen-fixing power resulted in a marked decrease in this power. Christiansen claims in this connection that the extent of the occurrence of Azotobacter in a soil may be used as a relatively reliable criterion of its lime requirements. A very thorough and comprehensive investigation of this question convinced Christiansen that one of the greatest benefits of the application of lime to soil, was its influence upon the development of nitrogen-fixing bacteria in the soil, especially upon the Azotobacter group.

Regarding the factors influencing the distribution of the Azotobacter in the soil, it is interesting to note that their occurrence in large numbers in a soil is often associated with the presence of certain forms of algae or the lower forms of plant life. This association is readily understood, when we remember that the Azotobacter are dependent upon some form of carbohydrate for their source of energy, and the algae seems to offer a suitable source of this material. The presence of some forms of algae in the soil to an extent where they could be observed with the unaided eye has long been known as a sign of fertile soil by the European farmers. Hugo Fischer has shown that Azotobacter chroococum and ozcillaria live in a symbiotic relationship. The fact that these species of bacteria derive an apparent advantage by association with other forms of plant life makes it necessary for its to speak of them as non-symbiotic nitrogen-fixers only with the greatest reservation. Investigations tend to confirm the theory that the relation between the higher and lower form of plant life in this case is mutually advantageous and, therefore, just as essentially symbiotic as the legume and its nodule bacteria. The algae appears to furnish the carbohydrate required as a source of energy for the bacteria and the bacteria in return share the supply of nitrogen taken from the air. There is this difference, however, between the association between the legume and the bacteria in its nodules and the bacteria and algae in the role of nitrogen fixers. The Azotobacter species of bacteria are less dependent upon the algae for nitrogen fixation, than the nodule bacteria are upon the legume as their host. The nodule bacteria can fix only very small amounts of nitrogen when grown independent of its host, while the Azotobacter in the presence of a suitable amount of energy material can fix as much if not more nitrogen than in association with the algae.

1 Colorado Exp. Station Bulletin No. 160.

An interesting case has recently been observed concerning this symbiotic relationship between the nitrogen-fixing species of bacteria in the soil, and the algae. In Colorado there is a soil formation which is known as "Brown Spots," upon which no vegetation exists. These spots are very toxic to plants, and the nature of the injurious action appears very similar to the action of alkalies. Sackett and Headden1 of the Colorado Station have made a very extensive study of this phenomenon, and have found the composition of the soil composing these spots to be excessively high in nitrates. While the average soil contains only from five to eight parts per million of nitrates, the brown spots of Colorado were found to contain as much as 6.54% of nitrate nitrogen, or an equivalent of 11 tons of nitrate per acre for the surface inch of soil.

2 Bulletin No. 184, Part II, Algae in Some Colorado Soil,. Cob. Exp. Station.

The question that immediately arose was the source of this incredible amount of soil nitrogen. Some claimed that it was due to the surface evaporation of underground waters which in Colorado are high in salts. It seemed past belief that this amount of nitrates could have been formed by bacterial action. Sackett and Headden claimed that the nitrates were formed by nitrifying bacteria, and that the original source of this nitrogen was the atmosphere, from which the Azotobacter in the soil had taken it and changed it into a combined state. But this theory was opposed upon the grounds that there was insufficient carbohydrate material in the Colorado soils to furnish the necessary amount of energy for those nitrogen-fixing bacteria to have added such large amounts of nitrogen to the soil. An investigation of the source of carbohydrate in those "Brown Spots" of Colorado was undertaken by Robbins of the Colorado Station,2 who found that they were particularly rich in algae. In concluding his report on that investigation he states as follows: "It is well known that many different kinds of algae inhabit the soil. As a rule it is generally understood that such a soil is necessarily muddy or very moist. In such cases the algae growth is visible to the naked eye, forming on the soil a characteristic plant mass. The soils from which the foregoing twenty-six samples were taken, were, with the exception of No. 17, just ordinary cultivated soil with a varying water content. The samples were representative of soils in rather widely separate localities in Colorado. At the time of collection during October, 1911, no algae were noticeable on the soil surface; furthermore, one would not ordinarily think of such soils being moist enough to support an algal flora. And yet cultures from these soils, with but two exceptions, samples Nos. 30 and 17, revealed the presence in them, of a considerable number of species of algae, and a healthy development of these. It is unquestionably true that during favorable seasons of the year there is developed in certain of our soils a rich growth of algae.

"This is probably confined to the surface layer. To what depth algae extend will depend largely upon the texture of the soil, its ventilation and methods of cultivation. It is probably true, however, that the top crust of soil, the first inch or less, is usually too dry to favor algae. Irrigation may play a part in determining the distribution of soil algae. Whether or not an unirrigated soil possesses an algal flora remains to be found out. But it can be readily understood that the turning of water onto an unirrigated area would introduce from the streams an abundance of algae. Although evidence is still insufficient, it is within the bounds of reason to believe from the preliminary investigation that all of our ordinary cultivated soils, especially those under irrigation, are far richer in algae than is usually supposed to be the case. More than this, we venture to assert that soil algae play a far more important part in soil fertility than is generally believed. Unquestionably the organic matter furnished by soil algae must be reckoned with as an important source of energy for the nitrogen-fixing organisms."

It seems very probable that the conclusions drawn by the investigator from whose report we have above quoted, are fully justified by actual conditions. That the species of nitrogen-gathering bacteria, with which the present series of articles deals, are greatly benefited by an intimate association with soil algae. there is small reason to doubt. Moreover, it appears highly probable, in the absence of definite experimental data either pro or con, that, in the Louisiana soils, moisture temperature and the presence of organic matter all combine to favor a vigorous development of these beneficial forms of plant life. and no doubt such an advantageous association exists between algae and nitrogen-gathering bacteria in our soils.

CHAPTER III.

THE CONDITIONS AFFECTING NITROGEN FIXATION BY NONSYMBIOTIC BACTERIA.

The farmer, as a rule, believes the legume to be the only connecting link between his land and the enormous supplies of free nitrogen in the atmosphere above it. As a matter of fact, the gain in nitrogen in bare uncropped soil is sometimes greater than that resulting from the growth of legumes. This gain in nitrogen in uncultivated soils is due to the action of the species of bacteria with which the present series of articles are concerned. The gain in nitrogen resulting from the activities of these bacteria is very variable and, indeed, much more so than is true of the legume group of nitrogen-fixing bacteria. In one of the previous articles of this series we learned that in some cases the increase of nitrogen in soils amounts to as much as 100 pounds per acre per annum. Evidently the importance of an agency capable of adding such large quantities of nitrogen to the soil cannot be easily overestimated, and the farmer should endeavor to learn how this process may be stimulated in his soils. In looking into the conditions that are essential to the operation of this process in the soils we have the following factors, which, if not all the conditions essential to the maximum activities of these species, nevertheless represent a highly satisfactory environment for the species of bacteria concerned in this highly valuable process. These factors are: (1) the presence of a suitable quantity of energy material for the bacteria, such as cellulose, dextrose, etc.; (2) the presence of sufficient moisture and conditions that admit of a free interchange of air beneath the surface of the soil; (3) proper temperature. and (4) the presence of sufficient lime.

We will now consider these factors individually in order to determine their relative importance in maintaining ideal conditions for the beneficent activities of the nitrogen-fixing bacteria in the soil.

First, we will consider the factor of an energy supply for these organisms and in what forms this source of energy can be supplied to them. We noted in a previous article the relation between nitrogen fixation by non-symbiotic bacteria and the amount of suitable carbohydrates available to them as a source of energy. Of the substances mentioned in that connection the most important was dextrose, which would not ordinarily occur under field conditions, so the bacteria would have to depend upon some other course of energy. Cellulose, although not directly assimilable by the nitrogen-fixing bacteria, and hence unsuitable for immediate use by them, can be so changed in its composition by other soil bacteria that it may be converted into an excellent source of energy. In this way the large amounts of cellulose in crop residues when left on the land may increase the nitrogen of the soil by indirect means, that is, by furnishing the source of energy necessary for the nitrogen-fixing bacteria to perform their work. For example, the following figures show the amounts of nitrogen fixed in the presence of cellulose and dextrose by an impure culture of azotobacter and cellulose bacteria:

Cellulose   Dextrose Amounts of Nitrogen
Fixed
2.5 + 0.2 14.8—25.6 mg.
5.0 + 0.2 56.3 mg

In addition to cellulose many pectin substances are suitable sources of energy for the nitrogen-gathering bacteria of the soil. Among these xylose and arabinose are especially deserving of mention. The salts of organic acids are also assimilable sources of energy for these bacteria. The fact that cellulose-dissolving bacteria are very widely distributed in the soil would indicate that cellulose would indirectly furnish the nitrogen-gathering bacteria with the supply of energy necessary for the fixation of large quantities of nitrogen.

In connection with the utilization of cellulose by the azotobacter, as a source of energy for their nitrogen fixation, it is interesting to note that exceptionally large quantities of this material are incorporated in the soil under cane cultivation. In plant cane there is added to the soil about 800 pounds of cellulose, which, together with the following crop from the stubble, would amount to approximately 1600 pounds of cellulose added to the soil in three years, or equivalent to more than 500 pounds per year. These are relatively enormous amounts of cellulose to be incorporated in the soil, but it might appear from what has already been said about the influence of this material upon nitrogen fixation, that very beneficial results should be obtained. Undoubtedly under proper conditions of drainage and aeration of the soil, this would result, but where drainage is so difficult as it is on most of the cane plantations, the large amounts of cellulose may be harmful rather than beneficial. Cellulose is only indirectly available to the nitrogen-fixing bacteria as a source of energy, but is directly utilized by the group of nitrate-destroying bacteria in the soil. This latter group, however, are only active in the soils when there is a lack of air, which is a condition often resulting from poor drainage. So, whether cellulose will act as a soil stimulant or a soil impoverisher depends largely in the last analysis on the effectiveness of drainage and cultivation of the soil.

The most interesting example of the utilization of this principle of stimulating the nitrogen-fixing bacteria by the addition of cellulose in crop residues is offered in the case of the Alabama negro farmer, Sam McCall. This illiterate negro, whose farm yields have astonished, not only his neighbors, but the whole country, and whose methods have commanded the closest scrutiny from our greatest agricultural authorities in this country, uses no fertilizer but returns all of the corn stalks and oat straw back to the soil. It is not at all probable that Sam McCall knows even the first principles of soil fertility or of plant nutrition, and even more improbable it is that he would attribute the beneficial effects of his corn stalks to their action upon the nitrogen-fixing bacteria in the soil. However, in spite of the fact that he does not understand the exact means by which the added cellulose stimulates the beneficial agencies in his soil, he seems to have accidentally developed a highly successful method for the conservation of soil nitrogen.

In connection with the stimulative action of straw, corn stalks and other crop residues upon the nitrogen-fixing bacteria of the soil, it should be noted that this class of substances is very inferior to leguminous hays as a source of energy for these beneficial bacteria in the soil. In one of the previous articles on this subject it was stated that legumes are an excellent foodstuff for the azotobacter in the soil, and as compared to corn stover are from three to four times more valuable. This fact would indicate that green manuring with legumes is a valuable practice even when considered entirely apart from the actual amounts of plant food it adds directly to the soil. Undoubtedly the secondary influence of turning under legumes is a very important one.

We will next consider the influence of moisture and aeration upon nitrogen fixation. The optimum moisture conditions for these species of soil bacteria varies, naturally, with the type of soil in which they occur. The degree of moisture in a soil not only exercises a very great direct influence upon the nitrogen fixation of soil bacteria, but it also has a very pronounced indirect action since it affects the degree of aeration obtainable in any soil. We know that one of the ways in which an excess of moisture in a soil renders it unsuited to the growth of most crops is that under this condition the transformation of plant food does not take place in a normal way. In a water-logged soil, for example, the lack of air promotes the activities of the nitrate-destroying bacteria and retards the fixation of atmospheric nitrogen. The accompanying table illustrates the influence of aeration upon nitrogen fixation in the soil:

TABLE SHOWING INFLUENCE OF AERATION UPON NITROGEN FIXATION.

Depth of Soil
Layer, Inches
Nitrogen in
Aerated Soil
Nitrogen in
Non-aerated Soil
Nitrogen
Increase
1/2 to 8 0.132 0.113 0.019
8 to 16 0.109 0.074 0.035
16 to 24 0.076 0.059 0.017
24 to 32 0.069 0.046 0.023

It will be noted from these figures that the greatest increase in nitrogen that resulted from aeration occurred at the depth of from 8 to 16 inches. This fact tends to explain one of the great benefits of thorough cultivation of the soil, and one which is but little appreciated by the average farmer. A study of the above table shows that although the greatest increase in nitrogen of the aerated over the non-aerated soil layer took place at the depth of from 8 to 16 inches, yet the largest amounts of nitrogen were fixed in the top layer of soil. This is exactly what would be expected in view of the fact that the tipper layer of soil is in more intimate contact with the atmosphere and the nitrogen-fixing bacteria are provided with more atmospheric nitrogen upon which to work. As a certain amount of exchange of air takes place in the top layers of soil, the increase due to cultivation and the aeration incidental thereto is less noticeable here than at the lower depths.

We will next consider the influence of temperature upon nitrogen fixation by these species of bacteria. Fortunately for agriculture and particularly for the conservation of soil fertility, the range of temperature in which these species are active is very wide and their ability to adapt themselves to extreme temperatures is rather pronounced. Hence we find them active in soils of the tropics as well as in the far northern climes. As an example of the activities of these species at various temperatures the following table is given:

50—53.6 F

68—71.6 F.

86—89.6 F

3.15 mg. N. fixed.

455 mg. N.

4.27 mg. N. fixed.

According to the experiments of Berthelot, the range of temperature for the nitrogen-fixing species of bacteria is from 50 to 113° F. The seasonal variation in the nitrogen fixation of soil bacteria has been extensively studied by A. Koch, from whose published results the following table is taken:

Increase in nitrogen in 1000 grams of soil:

May 29—Oct. 10

0.0709—0.0933 = .0224

Oct. 10—April 30

0.0983—0.0910

April 30—Oct

0.0910—0.1119 = .0209

2 Handbuch der landwirtschaftliche
Bakteriologie, p. 742.

In addition to the moisture, aeration and temperature of the soil, the presence of lime is another very essential factor in determining the degree of activity of nitrogen fixation in any soil. Christiansen2 in a recent very thorough investigation of the influence of lime upon the fixation of atmospheric nitrogen by soil bacteria, has found that a very definite relation between the activities of these species and the lime requirements of the soil exists. He found that these species rarely ever occur in any number in a soil that is acid, and he regards the condition of a paucity of these species in any soil as a reliable criterion of the need of lime. Undoubtedly the beneficial effect of time upon many soils is due chiefly to its favorable action upon the nitrogen-fixing bacteria in the soil.

1 Lohnis Handbuch der Landwirtsehaftliche Bakteriologie, p. 68.

It would be expected that the application of various kinds of fertilizers would have a varying influence upon the nitrogen-fixing species in the soil. The accompanying table from Lohnis1 shows that there is a great difference in the action of various combinations of fertilizing elements upon nitrogen fixation.

TABLE SHOWING EFFECT OF FERTILIZERS UPON NITROGEN FIXATION BY SOIL BACTERIA.

Fertilizer. Nitrogen Fixed.
Unfertilized 100
Lime 78
Lime and Potash 116
Lime and Phosphoric Acid 160
Potash and Phosphoric Acid 129
Lime, Phosphoric Acid and Potash 146
Phosphoric Acid, Potash and Nitrate of Soda 142
Lime, Phosphoric Acid, Potash and Nitrate of Soda 173
Manure 145

It is interesting to note from this table that the greatest fixation of nitrogen took place in the soil receiving the complete fertilizer with an additional application of lime. The value of lime as a stimulative influence upon nitrogen fixation is strikingly shown when we notice the increase in all the combinations containing lime over the check combination having the same kind and amounts of fertilizer without lime.

3 A study of the bacterial activities of virgin and cultivated soils, Cent. für Bakt., Bd. 41.

We have already referred to the fact that cultivation tends to increase the nitrogen-fixing power of the soil. Investigations in recent years have shown that the nitrogen-fixing power of cultivated soils is much greater than in virgin soils. Greaves3 of the Utah Experiment Station recently showed that more than twice as much nitrogen was fixed in the cultivated soils of Utah as in the virgin soils of that state. The relative amounts of nitrogen fixation in the two types of soil is shown in the accompanying table, where the results are given in milligrams of nitrogen fixed per hundred grams of soil.

TABLE SHOWING THE RELATIVE NITROGEN FIXATIONS IN CULTIVATED
AND VIRGIN SOILS. (CENT. FUR BAKT. BD. 41, GREAVES.)

Type of Soil Milligrams of
Nitrogen.
Cultivated 14.28
Virgin 6.99
Main crop was wheat 11.83
Crop when sampled wheat 9.84
Crop when sampled alfalfa 12.01
Crop when sampled fallow 22.81

The above table shows that there was more than twice as much nitrogen fixed in the cultivated as in the virgin soils, and the alfalfa soil showed a greater increase of nitrogen, due to fixation, than the wheat soil. The explanation of the latter fact is the superiority of alfalfa straw over wheat straw as a source of energy for the nitrogen-fixing bacteria in the soil.

4 Reed & Williams: Va. Exp. Station, Bulletin No. 3.

The observations of Greaves have, in a large measure, been confirmed by Reed and Williams4 of the Virginia Experiment Station. These investigators undertook a thorough comparison of the nitrogen-fixing power of cultivated and virgin soils of Virginia. The accompanying table, taken from their bulletin, shows the results of their comparative studies.

TABLE GIVING THE COMPARATIVE RESULTS OF NITROGEN FIXATION
BY 44 VIRGIN SOILS AND 44 CULTIVATED SOILS.

Number of pairs in which cultivated soils excelled 26
Number of pairs in which virgin soils excelled 18
Average increase where cultivated soils excelled 3.98
Average increase where virgin soils excelled 2.08
Total nitrogen fixed by cultivated soils 389.74
Total nitrogen fixed by virgin soils 331.66
Original nitrogen in cultivated soils 455.25
Original nitrogen In virgin soils 513.88

These figures conclusively demonstrate the beneficial influence of soil cultivation upon the activities of the nitrogen-fixing bacteria. The explanation of this fact is not difficult to find and we have, in fact, already alluded to it in the discussion of the benefits of soil aeration. We noted in a previous paragraph that nitrogen fixation was much greater in the upper layers than in the deeper layers of the soil, for the reason, no doubt, that the upper layers of soil are better aerated than the lower. Indeed, it is very probable that the greater nitrogen fixation in cultivated soils is due almost entirely to the better aeration resulting from cultivation, for in many other respects the virgin soil offers more favorable conditions for nitrogen fixation. Undoubtedly a thorough cultivation of the soil enables more air and its contained nitrogen to reach the nitrogen-fixing bacteria, and as a result these organisms have an abundance of material upon which to work.

5 Cent, für Bakt., 1912.
6 Soil Science vol. 1, No. 2.

The question of the benefits of aeration upon nitrogen fixation in soils takes on an added significance in connection with cane cultivation. Here we have relatively enormous quantities of material that, under proper conditions of aeration, would serve as a stimulus to the fixation of atmospheric nitrogen. On the other hand, this material serves as a stimulant to nitrogen-destroying bacteria under conditions of poor drainage and lack of aeration incident thereto. It is quite true that this destruction of nitrogen, or the process of denitrification, is not regarded now as the menace that it was some years ago. The able work of Von Caron's5 on this subject showed that only in the presence of large amounts of carbohydrates and under rather exaggerated field conditions was this bacterial action a menace to the fertility of the soil. Brown and Allison6 have recently made a very thorough investigation of the influence of various humus-forming materials upon the activities of soil bacteria. They found that the applications of substances with a wide nitrogen carbon ratio like straw, corn stover and non-leguminous hays caused a great increase in the activities of the nitrogen-fixing bacteria in the soil. In one case they report an increase of 3.27 mg. of nitrogen in a soil to which corn stover was added, while an equal quantity of cow pea hay only gave an increase of 2.55 mg. The corn stover was applied at the rate of 3 tons per acre and the cow pea hay at the rate of 4 tons. The crop yields from the soil treated with corn stover was 181 grams the first year as compared with a 250-grain yield from the soil which had received the cow pea hay. The authors attributed this difference in crop yields the first year partially to the fact that the soil was deficient in nitrogen which the nitrogen-fixing bacteria in the soil were unable to correct the first year in spite of the fact that more nitrogen was fixed in the corn stover than in the cow pea plot. It was also suspected that the decomposition of the cellulose had not been sufficiently completed in the first year for the soil to have derived the full benefit therefrom as a source of energy for the nitrogen-fixing bacteria. The yields from the two plots the second year confirmed this assumption, for the corn stover plot was 49 against 44 for the cow pea plot, making a total yield for the former of 230 against 294 for the cow pea plot. From these results it appears that turning under substances with a wide nitrogen carbon ratio has a tendency to increase nitrogen fixation in a soil, and this influence does not attain its maximum the first year. These and other similar experiments tend to throw more light on the phenomenal crop yields obtained by the Alabama negro to which we have already referred. If the incorporation of cellulose into the soil is capable of producing such astonishing benefits, when the soil is well cultivated, then improved cultivation of sugar soils should be especially desired in order that the full advantage may be obtained from the large amounts of cellulose occurring therein. [H. P. A.]