UNIVERSITY OF CALIFORNIA COLLEGE OF AGRICULTURE AGRICULTURAL EXPERIMENT STATION BERKELEY, CALIFORNIA Reclamation of the Fresno Type of Black-Alkali Soil W. P. KELLEY and E. E. THOMAS BULLETIN 455 June, 1928 UNIVERSITY OF CALIFORNIA PRINTING OFFICE BERKELEY, CALIFORNIA 1928 Digitized by the Internet Archive in 2012 with funding from University of California, Davis Libraries http://www.archive.org/details/reclamationoffre455kell FOREWORD It has been known for many years that the excess of alkali in some of the soils of California constitutes an important agricultural prob- lem. The late Dr. Eugene W. Ililgard, for many years (1875 to 1905) Professor of Agriculture and Director of the California Agricultural Experiment Station, prosecuted extensive studies on alkali soils and added greatly to our knowledge and understanding of them. He was a pioneer in this field of research and his investigations were classical for their time. However, much remained to be determined as regards a thorough understanding both of the fundamental nature of alkali soils and of the best practical methods of reclamation. The rapid extension of irrigated agriculture has caused the alkali problem to become more acute in various places. The importance of this phase of the question has made it imperative to conduct further investigations. The Kearney Vineyard, owned by the University of California, experienced considerable damage from the rise of the water table and the accumulation of alkali on or near the surface of the soil. Experiments were therefore undertaken there. Several years ago an attempt was made to reclaim a quarter section of this soil by tile drainage and flooding, but the results were only partially successful. It became evident that further investi- gation of the fundamental aspects of the alkali problem was needed. Accordingly a detailed laboratory study and an extensive series of field experiments were begun in 1919 by the College of Agriculture and placed under the direction of Dr. W. P. Kelley, Professor of Agricultural Chemistry in the Citrus Experiment Station. The present bulletin by W. P. Kelley and E. E. Thomas presents the first popular report of these field experiments. In addition to the papers listed in "Literature Cited" (p. 37), the following technical papers have already been published on this work or are in the course of preparation : 1. Kelley, W. P., and A. B. Cummins. 1921. Chemical effect of salts on soils. Soil Sci. 11:139-159. 2. Kelley, W. P. 1922. Variability of alkali soil. Soil Sci. 14:177-189. 3. Cummins, A. B., and W. P. Kelley. 1923. The formation of sodium carbonate in soils. California Agr. Exp. Sta. Tech. Paper 3:1-35. 4. Kelley, W. P., and S. M. Brown. 1924. Eeplaceable bases in soils. California Agr. Exp. Sta. Tech. Paper 15:1-39. 5. Kelley, W. P., and S. M. Brown. 1925. Base exchange in relation to alkali soils. Soil Sci. 20:477-495. 6. Cummins, A. B. [1926.] The solubility relationships of calcium carbonate with special reference to the formation of sodium carbonate in soils. Un- published thesis submitted in 1926. As a result of these investigations and similar studies made in Europe, some of the more puzzling phases of the alkali problem have already been elucidated. Probably the most important point that has come out of these technical studies is the thorough establishment of the fact that alkali soils can no longer be regarded as soils which merely contain an excess of soluble salts. The soluble salts upon accumulation react with the clay constituents of the soil, thus chang- ing their fundamental chemical nature and the changes thus wrought must be overcome in the treatment of the soil before it can be said to be reclaimed. These and other investigations are being continued, and further papers will be issued as the work progresses. H. J. Webber, Director, Citrus Experiment Station. RECLAMATION OF THE FRESNO TYPE OF BLACK-ALKALI SOIL 1 W. P. KELLEY- and E. E. THOMASa INTRODUCTION This bulletin reports the results of a series of alkali-reclamation experiments that have been made at Kearney Park, near Fresno, California. Within a few years after irrigated agriculture was introduced into the section southwest of Fresno, the water table began to rise as a result of seepage and excessive irrigation. This caused the soluble salts that were present in the deeper subsoil layers to move upward by capillarity and finally to accumulate on or near the surface of the soil. The consequence has been that a comparatively large area of formerly alkali-free, productive soil has become severely affected with alkali. A similar condition has developed in other parts of the San Joaquin Valley. The soil conditions of these areas closely resemble those of a much larger area still further south and west of Fresno, where an excess of soluble salts has accumulated as a result of purely natural causes. An attempt was made in 1914 and 1915 (4) to reclaim a quarter section (160 acres) of alkali soil on the Kearney Vineyard by means of tile drainage and flooding. The experiment was only partially successful. Comparatively large spots scattered here and there over the drained area remained unproductive, and in 1919 a large part of this quarter section failed to produce a profitable crop of barley. The soluble salts of this soil consist chiefly of the carbonate, bicarbonate, chlorid, and sulfate of sodium. Special investigations have shown that as the soluble salts accumulated in the soil, the sodium reacted with and consequently altered the chemical nature of an important component of the clay constituents of the soil. We now know that this change is more important chemically and much more difficult to overcome practically than the excess of soluble salts. i Paper No. 183, University of California, Graduate School of Tropical Agri- culture and Citrus Experiment Station, Riverside, California. -Professor of Agricultural Chemistry in the Citrus Experiment Station and Graduate School of Tropical Agriculture and Agricultural Chemist in the Experi ment Station. 3 Assistant chemist in the Experiment Station. 6 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION Ordinary leaching does not bring about the needed changes in these constituents. It is partly because of this fact that the previous drain- age and flooding experiment failed to produce satisfactory results. The proper objective in the treatment of this soil, and the same is true of black-alkali soils in general, is, therefore, not merely to remove the excess of soluble salts, but also to convert the clay con- stituents back into the state that existed before the excess of soluble salts accumulated in the soil. In normal soils these constituents are predominantly calcium compounds, whereas in the Fresno alkali soil they have been partially converted into sodium compounds through the action of soluble salts. In order to remove the soluble salts and to prevent their return into the soil by capillarity, the water table must be kept down perma- nently. It now seems certain that unless the water table is kept well below the zone of root development, permanent reclamation cannot be accomplished. The temporary nature of the benefit that has characterized many previous alkali-reclamation efforts has been due chiefly to faults in the drainage conditions. Good drainage is, there- fore, an important phase of alkali reclamation. In fact its importance can scarcely be over-emphasized. Soon after the experiments reported herein were begun it became evident that the tile-drainage system, which was then in operation on the experimental area, was not adequate for the best results. Accordingly, in June, 1924, a 16-inch well was bored to a depth of about 70 feet at a point near the experimental plots. Subsequently we have been able to control the ground water by pumping from this well throughout each irrigation season. The water is discharged into a canal through an underground pipe line. As was pointed out by Weir, (5J this scheme has been entirely successful from a drainage standpoint. Rarely since this well was first put into operation has the level of the ground water beneath the experimental plots been nearer the surface than 8 feet. A special advantage of this method of drainage is afforded by the water itself. As will be discussed in greater detail elsewhere, the well water, owing to its calcium content, is more suitable for irrigat- ing alkali soil than pure water or the supply that is furnished by the main canal of this section. Moreover, it is available at all seasons of the year, whereas the canal supply is ^independable after midsummer. The irrigation supply used on these experiments is now drawn ex- clusively from this well. BUL. 45o] RECLAMATION OF FRESNO TYPE OF BLACK ALKALI SOIL EXPERIMENTS WITH GYPSUM Experiments with Different Quantities of Gypsum. — In the sum- mer of 1920 a series of nine plots, each 100 by 330 feet in size, was staked out in a badly affected part of Section 6 of the Kearney Vine- yard. Soil samples representing each foot down to a depth of 4 feet were drawn at intervals of 10 feet throughout the greatest length of the plots. Gypsum, as recommended by Hilgard many years ago, was applied at the rate of six tons per acre to six of the plots and plowed under, after which all of the plots were flooded with water to a depth of about 10 inches for a period of three weeks. When sufficiently dry the soil was again plowed and sampled. Analysis showed that these plots still contained considerable sodium carbonate. Therefore, three additional tons of gypsum per acre were applied to three of these plots and all of them were again flooded for a period of two weeks. Later the soil was plowed and prepared for planting. Barley was sown in December and harvested the following May. Although the growth of the barley showed that the gypsum treatment had produced considerable effect, the yields were not satisfactory. The soil still contained considerable sodium carbonate. Hence in June, 1921, an additional application of gypsum was made at the rate of six tons per acre, followed by heavy flooding. Thus it is seen that a total of 15 tons per acre of gypsum was applied to three of these plots and 12 tons per acre to three others, while three plots were merely flooded, plowed, and cultivated without the application of gypsum. Since these applications of gypsum were made in 1920 and 1921 no further materials have been applied to these plots. They have merely been cultivated, irrigated and cropped as shown in the tabular statement of the results (table 1). TABLE 1 Experiments with Gypsum. Crop Yields in Pounds per Acre Plot Treatment* Tons per acre 1920, Barley hay 1921, Barley hay 1922, Barley hay 1923, Melilotus indica and cowpeas 1924, Melilotus alba 1925, Alfalfa 1926 1927, Alfalfa 3 4 5 Gypsum Gypsum Untreated 12 15 875 696 1,428 2,154 2,865 1,770 2,438 3,216 2,584 Plowed under as green manure. Plowed under as green manure. 5,422 5,955 3,585 T3 i - a 10,728 11,742 6,255 * The first application of gypsum was made after the barley was harvested in 8 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION When this series of experiments was begun the importance of the extreme variability of the Fresno alkali soil was not adequately appreciated. It was found later that certain of these plots containued spots of considerable size in which there was originally very little alkali (fig. 1). Because of the great variability of the soil, the crop records will be given for only three of the plots of this series, i.e., plot 3, treated with 12 tons per acre of gypsum ; plot 4, treated with ]5 tons per acre of gypsum, and plot 5,. untreated. These plots, although not free from variation, were more uniform as regards the alkali distribution than the remaining plots of this series. Fig. 1. — General view of the area on which the gypsum plots 1 to 9 were located. The photograph was taken in May, 1920, just before the experiments were begun. Note the spotted condition of the soil. Analysis showed that certain spots on which barley was growing were almost free from alkali, whereas the barren areas contained much alkali. Barley was sown over this entire area in November, 1919. The yields reported for June, 1920, represent the amounts of barley hay (grain and straw) that were obtained immediately preceding the time when the first application of gypsum was made. It will be seen that the yield was light, the greatest amount being then obtained from plot 5, which was used subsequently as a check plot. Barley was again grown in 1921 and 1922, and, although increases were obtained, the yield from each plot was still unsatisfactory. The appearance of the growing barlej^ on the gypsum plots indicated that there was an inadequate supply of available nitrogen in the soil. Accordingly, Melilotus indica and cowpeas were grown in 1923 (figs. 2, 3, and 4) and Hubam clover (Melilotus alba) in 1924. These crops Bul. 4^5 RECLAMATION OF FRESNO TYPE OF BLACK ALKALI SOIL !) were plowed under as green manures, and alfalfa was sown in Feb- ruary, 1925. A fair stand was secured over practically all of the gypsum plots and also on certain parts of the untreated plots. On Fig. 2. — CoAvpeas on plot 3, treated with 12 tons of gypsum per acre, graphed in October, 1922. Photo- Fig. 3. — Cowpeas on plot 4, treated with 15 tons of gypsum per acre. Photo graphed in October, 1922. Comparison of this figure with figures 2 and 13 shows that 15 tons of gypsum gave better results than either 10 or 12 tons. the other parts of the untreated plots the alfalfa seed failed to germinate. Owing to a severe infestation of Bermuda grass {Cynodon dactylon) these plots were plowed up in December, 1925, and kept dry with frequent cultivation throughout the following summer. By 10 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION this means Bermuda grass was practically eradicated. Alfalfa was sown again in October, 1926. The yield in 1927 was heavy on each of the gypsum plots, being almost twice that of the untreated plot. Fig. 4.— Cowpeas on check plot 5. The spots which show good growth were practically free from alkali at the outset. Although this plot was subjected to prolonged leaching in 1920 and 1921, the badly affected spots have remained unproductive. The photograph was taken on the same day as those shown in figures 2 and 3. Fig. 5. — Alfalfa on plot 3, treated with 12 tons of gypsum per acre, photograph was taken in September, 1927. The The general appearance and growth of the alfalfa throughout 1927 (figs. 5 and 6), together with the results obtained in laboratory studies, lead us to believe that the choice of crops in connection with BUL. 455] RECLAMATION OF FRESNO TYPE OF BLACK ALKALI SOIL 11 this series of experiments was not the best for practical results. There is good reason to believe that had alfalfa been planted in 1921 after the last application of gypsum was made, or perhaps better still, had Melilotus alba been grown and plowed under and this followed with alfalfa, satisfactory yields of alfalfa would have been obtained. Fig. 6. — Alfalfa on plot 4, treated with' 15 tons of gypsum per acre. Photo- graphed in September, 1927. Fig. 7. — Alfalfa on check plot September, 1927. Note the bare spots. itograplied in There is no reason to doubt the possibility of completely reclaiming this soil by means of gypsum coupled with flooding and drainage. It is merely a matter of using an adequate amount of gypsum and leaching out the soluble salts. However, the more seriously affected 12 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION parts of this soil will require too large an amount of gypsum to justify its practical use at the present prices. In localities where gypsum can be obtained more cheaply or where the amount required is not so large, gypsum will be found effective and profitable as a treatment for black-alkali soils. EXPERIMENTS WITH SULFUR Comparison of Untreated Sulfur and Gypsum. — In 1921 an experi- ment was begun with the use of sulfur. The area chosen (plot 10, 100 by 260 feet) was almost entirely bare of vegetation. Barley had been sown the previous November, but the seed failed to germinate ;-S--lk Fig. 8. — This photograph was taken in April, 1921, and shows the general con- dition of plot 10 before sulfur was applied. The plant shown in the foreground is chiefly an alkali weed, Tissa salina. With the exception of a few small spots this plot was entirely unproductive before the experiment was begun. over practically all of this area (fig. 8). The alkali conditions were as extreme as could be found anywhere in this field. Ordinary finely ground crude sulfur, unmixed with any other material and untreated in any way, was applied at the rate of 3,600 pounds per acre. For purpose of comparison, gypsum was applied to an adjacent area (plot 11) (fig. 9) at the rate of 10 tons per acre. These materials were applied and plowed under in June, 1921, after which a light irrigation was given, this was followed by shallow cultivation, the immediate purpose being to promote conditions that BUL. 455] RECLAMATION OF FRESNO TYPE OF BLACK ALKALI SOIL 13 are favorable for biological oxidation. The efficiency of sulfur de- pends upon its being oxidized to sulfate, and the oxidation is a process involving the action of certain species of bacteria. Since the oxidation takes place gradually, a certain interval must elapse before sulfur can have any effect ; the full effect where large amounts are required may be delayed for several years. For this reason the sulfur-treated Fig. 9.- — The condition of plot 11 before gypsum was applied. Although severely affected with alkali, this plot was not as uniformly unproductive as was plot 10 (fig. 8). Much of the growth shown here is an alkali weed (tixsa salina). Photographed in April, 1921. plot was allowed to stand until late summer. Then both this and the adjacent gypsum plot were flooded twice, after which they were plowed and seeded to barley. The crops grown since that time and the yields are reported in table 2. TABLE 2 Comparative Effect of Sulfur and Gypsum. Crop Yields in Pounds per Acre Plot Treat- ment* Amount per acre 1921, Barley Hay 1922, Barley Hay 1923, Melilotus indica and cowpeas 1924, Alfalfa 1925, Alfalfa 1926, Alfalfa 1927, Alfalfa 10 11 Sulfur Gypsum 3,600 pounds 10 tons 145 430 300 1,815 Plowed under as green manure. 4,000t l,500t 18,467 12,838 23,658 14,368 20,138 15,103 * These materials were applied after the barley was harvested in 1921. t Yield obtained from a single cutting in September. It will be noted that the effects at the end of one year, as reflected by the yield of barley in 1922, were not encouraging. Gypsum in fact 14 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION was more effective at that stage of the experiment than sulfur, although neither produced a satisfactory yield of barley (figs. 10 and 11). Cowpeas, grown as a green manure in 1922 soon after the barley was harvested, gave the first indication of benefit from the ■ '*:''''*. &§*$£$! Fig. 10. — Unsatisfactory growth of barley on plot 10 one year after sulfur had been applied at the rate of 3600 pounds per acre. Photographed in May, 1922. Fig. 11. — Barley on plot 11 one year after 10 tons of gypsum per acre had been applied. Comparison of this figure with figure 10 shows that better effects were produced by gypsum at this stage of the experiment than by sulfur. The growth of barley was spotted. In the foreground the maximum height of the barley plants was about 18 inches, and much of the seed failed to germinate. sulfur treatment. As is shown in the illustrations (compare figs. 12 and 13), the growth of cowpeas was somewhat greater on the sulfur plot than on the gypsum plot. The beneficial effect of sulfur and Bul. 4f>; RECLAMATION OF FRESNO TYPE OF BLACK ALKALI SOIL 15 its superiority over gypsum as a treatment for this soil became definite in the growth of Mclilotus indica in the winter and spring of 1923. The Melilotus seed germinated over practically all of the sulfur plot ■ Fig. 12. — Effect of sulfur on the growth of cowpeas 18 months after an application of 3600 pounds per acre was made. Photographed in October, 1922. The germination and growth of the cowpeas gave the first indication of definite effect of sulfur on this plot. Fig. 13. — Effect of 10 tons of gypsum per acre on the growth of cowpeas. Comparison of this photograph (taken in October, 1922) with figure 12, shows that at this stage of the experiment the effect of gypsum was less favorable than that of sulfur, although the opposite was the case with the barley that was grown previously (compare figs. 10 and 11). 16 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION and the growth was normal in appearance (fig.. 14). On the other hand, the germination of Melilotus was poor on the gypsum plot and the stand was. too thin to justify its retention as a green-manure crop. The results that have been obtained each year since alfalfa was sown in February, 1924, show clearly that sulfur is remarkably effective on this soil. The yields from the sulfur-treated plot have been extraordinarily heavy, being fully twice the average yield of alfalfa for the San Joaquin Valley as a whole. The tabulated results Fig. 14. — Effect of sulfur on the growth of Melilotus indica two years after an application of 3600 pounds per acre was made. Note that with the exception of one small spot the growth was vigorous throughout this plot. Photographed May, 1923. shown in table 2, together with the series of photographs of this plot (figs. 8, 10, 12, 14, and 15) are self-explanatory. The latter presents a striking illustration of the effects of sulfur. It will be noted that the sulfur plot has yielded approximately 50 per cent more alfalfa hay than the gypsum plot of this experiment (compare figures 15 and 16). However, it is probable that a heavier application of gypsum (more than 10 tons per acre) would have given better results. This point will be more fully discussed in a later section of this bulletin. In this connection it seems desirable to point out that the reduced yield of alfalfa obtained in 1927, as compared with that of 1926, was probably not due to a return of alkali into the soil, but was caused chiefly by the spread of Bermuda grass over this plot. In this and other sections of California the rapid spread of Bermuda grass makes it difficult to maintain a good stand of alfalfa longer than Bul. 455] RECLAMATION OF FRESNO TYPE OF BLACK ALKALI SOIL 17 three or four years. However, the growth of this grass is not confined to alkali soil, although the effect may be more serious on alkali soil for the reason that alfalfa is more sensitive than Bermuda grass to alkali. Fig. 15. — Effect of 3600 pounds of sulfur per acre on the growth of alfalfa. The corresponding effect of gypsum at this stage of the experiment is shown in figure 16. Photographed in May, 1925. Fig. 16. — The effect of 10 tons of gypsum per acre on the growth of alfalfa. Note the spotted appearance. On the lighter-colored areas the alfalfa seed failed to germinate. Later Bermuda grass spread over these spots. Comparison with figure 15 shows that sulfur was much more effective than gypsum. Photo- graph taken in May, 1925. Experiments with Inoculated Sulfur. — A second sulfur experi- ment (plots 21 to 25) was begun in May, 1923, in cooperation with Charles D. Samuels. (3) As in the sulfur experiment discussed above, 18 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION the area chosen for this experiment was also badly affected with alkali. A series of five plots, each 40 by 135 feet in size, was staked out, sampled, and prepared for irrigation. Sulfur was applied to four of these plots, the fifth being left untreated as a check. The sulfur used in this series was finely ground and had been artificially inoculated with a culture of sulfur-oxidizing bacteria. This material was applied at the rate of approximately two tons per acre. In addi- tion to sulfur, gypsum was also applied to plot 21 at the rate of 2.5 tons per acre, and ground limestone to plot 22 at the rate of two tons per acre. These materials were plowed under, after which the plots were allowed to lie fallow until February, 1925, with only an occasional light irrigation and shallow cultivation. They were then flooded twice, plowed, and sown to Melilotus alba. Fairly good growth was obtained on each of the sulfur-treated plots but the seed failed to germinate on the check plot. The Melilotus alba was plowed under as a green manure in September, 1925, and alfalfa was sown in February, 1926. A good stand was secured on each of the sulfur- treated plots, but the alfalfa seed failed to germinate on the check plot (compare figs. 17 and 18). As shown in table 3, heavy yields of alfalfa have since been harvested from these plots. TABLE 3 Effect of Inoculated Sulfur. Crop Yields in Pounds per Acre Plot Materials* Tons per acre 1925, Hubam clover 1926, Alfalfa 1927, Alfalfa 21 22 23 f Sulfur ) Gypsum f Sulfur \ Lime Untreated Sulfur Sulfur 2 \ VA \ 2 \ 2 J Plowed under as green manure 10,916 11,500 83 12,166 9,583 17,549 18,897 615 24 25 2 2 18,848 17,031 * Applied in May, 1923. It will be seen that practically the same yields of alfalfa have been obtained from each of the sulfur-treated plots of this series. Since, in addition to sulfur, gypsum was applied to plot 21 and ground limestone to plot 22, whereas sulfur alone was applied to plots 24 and 25, it is clear that the effects produced in each case have been due chiefly to sulfur. The chemical studies necessary to determine whether there is a detectable chemical difference in the soil due to the gypsum have not yet been made. BUL. 455] RECLAMATION OF FRESNO TYPE OF BLACK ALKALI SOIL 19 Fig. 17. — Ineffectiveness of leaching when unaccompanied by other treatment. The plots to the right and left of plot 23 were treated with sulfur in May, 1923. Photograph taken in October, 1926. Fig. 18. — Effect of inoculated sulfur on alfalfa. The material was applied in May, 1923, and the photograph was taken in October, 1926. Comparison of this figure with figures 15 and 19 shows that the effect of uninoculated sulfur was similar to that of inoculated sulfur. The same is shown in the tabular state- ment of yields. 20 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION Sulfur Experiment on a Larger Area. — A third experiment with sulfur was begun in December, 1924, on an area of about seven acres (plots 26-44). Finely ground uninoculated sulfur was applied at the rate of one ton per acre and disked in. Melilotus alba was sown in March, 1925, and a fair stand and good growth were secured except on the most seriously affected spots. This crop was plowed under as a green manure in September, 1925, and barley was planted the following December. The barley crop was not weighed, but the growth was vigorous on all parts of the treated area with the excep- tion of a few small spots. After the barley crop had been harvested, the soil was plowed and allowed to remain dry during the summer Fig. 19. — Alfalfa on the seven-acre area treated with sulfur in December, 1924, at the rate of 1 ton per acre. Alfalfa was sown in October, 1926, and photographed in September, 1927. This plot has not been leached at any time since the experiment was begun. The irrigation water was applied at two-week intervals throughout the hotter months of 1927. of 1926, and in October alfalfa was sown. The seed germinated well, and vigorous growth has resulted. The first cuttings made in 1927 were not weighed. The average yield per cutting after June 1, 1927, was 2,445 pounds per acre, which is usually considered to be a satisfactory yield. In this experiment the soil has not been as heavily leached as was the case with the previous experiments. Since alfalfa is usually irrigated by the flooding system, the attempt is being made- to leach out the soluble salts and wash down the sulfur-oxidation products by means of ordinary irrigation. The results to date (February, 1928) indicate that the experiment will be successful (fig. 19). BUL. 455] RECLAMATION OF FRESNO TYPE OF BLACK ALKALI SOIL 21 Sulfur with Stable Manure and with Ground Limestone. — A fourth sulfur experiment was begun on plots 16 and 19 in October, 1925, using ordinary finely ground untreated sulfur, similar to that which was applied to plot 10 in 1921. The application was made at the rate of 1000 pounds per acre. In 1922 plot 19 had been treated with stable manure and plot 16 with ground limestone, the applica- tion of each being followed by heavy leaching. The effects of the manure and ground limestone had not been discernible either in the growth of barley in 1923 and 1924 or in that of cowpeas sown in the summer of 1923 or in that of Melilotus alba sown in May, 1925. The sulfur was applied and plowed under in October, 1925, and alfalfa was sown in February, 1926. The germination was good on plot 19 and fair on plot 16. TABLE 4 The Effect of Sulfur on Soil Previously Treated with Ground Limestone and Stable Manure. Yields in Pounds per Acre Plot Material applied* Amount per acre 1923, Barley 1924, Barley 1925, Hubam clover 1926, Alfalfa 1927, Alfalfa ..I ■ 20 Sulfur Ground limestone Sulfur Manure 1,000 lbs. 9 tons 1,000 lbs. 18 tons ) 108 I o J 30 24 15 Plowed under as green manure 1,800 10,440 312 11,634 20,022 2,502 * The ground limestone and manure were applied in June, 1922; the sulfur was applied in November, 1925. In certain respects this sulfur experiment, particularly the plot that was treated with stable manure three and one-half years pre- viously, has given the most striking results of all in this field (fig. 20). The rate of sulfur oxidation, as indicated by the evidence of sulfate formation in the soil, has been especially pronounced. As is shown in table 4, heavy yields of alfalfa were obtained from this plot in 1926 and 1927. The untreated check plot of this series has been a practical failure so far as the growth of alfalfa is concerned (fig. 21). At no time since the sulfur was applied in October, 1925, have these plots been heavily flooded. They were irrigated twice each month during the spring and summer of 1926 and approximately once every three weeks subsequently. As is shown by the chemical analyses, the effect of the sulfur has not yet penetrated as deeply on plot 19 as it has on plot 10. This is probably due both to the absence of heavy flooding and to the briefer period that has elapsed since the sulfur was applied. 22 UNIVERSITY OF CALIFORNIA — EXPERIMENT STATION Fig. 20. — Alfalfa as affected by the combined use of stable manure applied at the rate of 18 tons per acre in May, 1922, and of sulfur applied at the rate of 1000 pounds per acre in October, 1925. The alfalfa was sown in February, 1926, and was photographed in October, 1926. J^J*4#$SRSf:- Fig. 21. — Failure of leaching as a practical method of reclamation. Alfalfa was sown in February, 1926, and photographed in September, 1927. BUL. 455] RECLAMATION OF FRESNO TYPE OF BLACK ALKALI SOIL 23 EXPERIMENTS WITH IRON SULFATE AND ALUM Previous laboratory experiments (2) gave results which indicated that iron sulfate and alum might prove effective on this soil. Accordingly plot 17 was treated in August, 1922, with iron sulfate at the rate of approximately nine tons per acre, and plot 18 with alum at the rate of 11 tons per acre. Plot 14 was untreated. Immedi- ately after these materials were applied, the soil was heavily flooded. Barley was planted the following December. The seed germinated well and the barley seedlings grew vigorously for a few weeks on . : . ," ' * "' '• ; Fig. 22. — Failure of leaching as a practical method of reclamation. Alfalfa was sown in February, 1926, and photographed in September, 1927. each of the treated plots. Later the plants became pale in color and the yield was light. Cowpeas planted after the barley was harvested in May, 1923, also germinated well on both of the treated plots, but later the plants became pale and unthrifty on the alum plot. Barley was again grown in 1924 but the yield was inferior. The plants presented an appearance indicative of a deficiency of available nitro- gen in the soil. Alfalfa was sown in February, 1926. The seed germinated well and the subsequent growth has been normal in appearance. As shown in table 5 the untreated check plot has remained almost entirely unproductive, whereas the iron sulfate and alum plots have given good yields of alfalfa (compare figs. 22, 23, and 24). 24 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION TABLE 5 Effect of Iron Sulfate and Alum. Yields in Pounds per Acre Plot Material applied Tons per acre 1923, Barley 1924, Barley 1925, Hubam clover 1926, Alfalfa 1927, Alfalfa 14 528 2,406 1,014 192 1,101 801 ] Plowed under | f as green < J manure ( 60 11,880 10,320 12,739 618 17 18 Iron sulfate* 9 11 5 19,176 18,762 45 Iron sulfatet 20,504 * Applied in June, 1922. t Applied in October, 1925. Fig. 23. — Effect on alfalfa of iron sulfate applied in August, 1922, at the rate of 9 tons per acre. Photographed in September, 1927. «^~ . Fig. 24. — Effect on alfalfa of alum applied in August, 1922, at the rate of 11 tons per acre. Photographed in September, 1927. BUL. 455] RECLAMATION OF FRESNO TYPE OF BLACK ALKALI SOIL 25 Chemical studies made on samples of soil from these plots prove, as will be reported elsewhere/ 1 } that both the iron sulfate and alum have brought about important chemical changes in the soil. In each case the chemical changes have been similar to those in the sulfur- treated plots referred to above. A second experiment with iron sulfate (plot 45) was begun in October, 1925, the material being applied at the rate of five tons per acre. After one flooding this plot was seeded to alfalfa in the follow- ing February. The subsequent yields of alfalfa have been heavy, the extraordinary yield of over six tons per acre having been obtained in 1926, the first year after the alfalfa was sown (see fig. 25). In 1927 good yields were again obtained. Fig. 25. — Effect on alfalfa of iron sulfate applied in October, 1925, at the rate of 5 tons per acre. Photographed in September, 1927. EFFECT OF LEACHING WITHOUT OTHER TREATMENT As stated already, attempts made in 1911 and 1915 to reclaim this soil by flooding and drainage were unsuccessful. In order to test the leaching method further, three plots in the gypsum series were left untreated. They were flooded and otherwise subjected to the same cultural treatment as the gypsum plots. The heavy floodings to which these plots were subjected in 1920 and 1921 afforded a practical test of the leaching method of reclamation. One plot in connection with the experiment with iron sulfate and alum and one in each of the sulfur series of 1923 and 1925 were also left untreated. 26 UNIVERSITY OF CALIFORNIA — EXPERIMENT STATION No material has been applied to these plots, but otherwise they too have been plowed, flooded, cultivated, and irrigated just as the other plots of these series. Parts at least of one of the check plots in the gypsum series have been markedly improved, as is shown by the crop records and the chemical analyses. On the other hand, several spots of considerable size on the other check plots of this series have been benefited very little, if at all. Neither the check plot of the iron sulfate and alum series (plot 14), nor those of the sulfur series (plots 20 and 23) have shown more than the most meager indication of improvement. Although these experiments show that it is possible to improve the crop-producing power of a part at least of this soil by leaching without adding other materials, and, as will be shown in another paper, it is even possible by this means to effect its complete con- version into normal soil, the difficulty of so doing is considerable. The superior effects that may readily be obtained by the use of sulfur and other materials, together with the fact that certain fundamentally important losses are sustained by the soil when it is subjected to prolonged leaching, strongly argue against placing sole reliance on the leaching method of reclamation. Where sulfur and possibly other materials are applied it is probably best to limit the leaching to the amount that is required to remove the soluble salts. As stated already, the failure of leaching as a practical method of reclamation is due to the fact that, unless some special material is applied in addition to water, the necessary changes in the clay constituents of this soil take place very slowly. Where the irrigation supply contains calcium salts, the required amount of gypsum, sulfur, or other materials will be correspondingly reduced, and if the water contains considerable calcium salts, mere leaching may suffice to bring about complete reclamation. This appears to be the case with a certain black-alkali soil near Salt Lake City, Utah. CHEMICAL CHANGES IN THE SOIL The soil samples that were drawn from these plots before the various materials were applied and the corresponding samples drawn subsequently, have been analyzed for water-soluble salts. Although the chemical changes that are requisite to the complete reclamation of this soil must involve the clay constituents as well as the soluble salts, a determination of the latter has distinct value. Such determination not only affords proof concerning the changes BUL. 455] RECLAMATION OF FRESNO TYPE OF BLACK ALKALI SOIL 27 in the soluble salts and the effectiveness of the leaching per se, but also gives valuable indication as to the progress of the changes in the clay constituents. A separate paper (1) will be devoted to a more theoretical discussion of the chemical changes that have taken place in these plots. At present it may be pointed out that a decrease in the sodium carbonate content takes place concurrently with the needed changes in the clay. The changes found in soluble carbonate and calcium tell us at once whether a given treatment is proving effective. The chlorin deter- minations give the necessary evidence as to the effectiveness of the leaching. The averages of the analyses of the samples from certain plots are reported in tables 6, 7, and 8. It will be noted that each of these treatments has brought about a reduction in the content of chlorin and soluble carbonate. There is, however, a marked difference in the extent to which the changes have taken place in the different plots. TABLE 6 Chemical Effect of Gypsum Depth of samples in inches Parts per million C0 3 CI Ca May 1920* Oct. 1921 Dec. 1927 May 1920* Oct. 1921 Dec. 1927 May 1920* Oct. 1921 Dec. 1927 Plot 3, f gypsum, j 12 tons per acre [ 0-12 12-24 24-36 36-48 253 116 143 110 5 33 96 84 45 135 248 256 227 250 30 42 17 24 18 27 27 35 Trace Trace Trace Trace 101 22 Trace Trace 45 25 Trace Trace Plot 4, f gypsum, 1 15 tons per acre ( 0-12 12-24 24-36 36-48 283 147 128 44 17 73 105 60 75 105 116 162 137 114 16 18 22 32 18 18 27 35 Trace Trace Trace Trace 76 13 Trace Trace 27 Trace Trace Trace Plot 5, 1 untreated 0-12 12-24 24-36 36-48 315 146 127 44 99 89 146 86 45 90 240 210 220 220 205 120 13 35 145 194 27 53 177 204 Trace Trace Trace Trace Trace Trace Trace Trace Trace Trace Trace Trace * The analyses for 1920 represent the original soil before the experiment was begun. Although the determination of the average composition of the different plots before and after the materials were applied brings out the relative effect of the different treatments, this method fails to present a true picture of the chemical variation within a given plot. For example, there are several spots of considerable size on plot 11, which was treated with gypsum at the rate of 10 tons per acre, that still contain considerable sodium carbonate and only traces of soluble 28 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION calcium. This fact is not indicated by the analyses of this plot reported in table 7. On these spots alfalfa has not made satisfactory growth. On the other hand, the improved chemical conditions and the growth of alfalfa have been much more uniform where either 12 or 15 tons of gypsum per acre were applied. A similar uniformity of effect has resulted from the application of sulfur, iron sulfate, and alum. TABLE 7 Comparative Chemical Effect of Sulfur and Gypsum Depth of samples in inches Parts per million Treatment C0 3 CI Ca Apr. 1921* Oct. 1921 Dec. 1925 Dec. 1926 Apr. 1921* Oct. 1921 Dec. 1925 Dec. 1926 Apr. 1921* Oct. 1921 Dec. 1925 Dec. 1926 Plot 10, f 0-12 346 144 3 544 48 50 36 Trace Trace 32 28 sulfur, 1 12-24 179 141 35 16 440 76 50 30 Trace Trace 13 18 3,600 lbs. 24-36 158 181 109 109 248 235 75 40 Trace Trace Trace Trace per acre ( 36-48 127 134 154 204 174 368 106 55 Trace Trace Trace Trace Plot 11, f 0-12 221 37 32 16 491 35 50 32 Trace 27 15 21 gypsum, j 12-24 162 92 124 106 376 41 67 36 Trace Trace Trace Trace 10 tons 1 24-36 127 88 221 235 264 105 128 53 Trace Trace Trace Trace per acre [ 36-48 64 79 178 273 201 209 244 93 Trace Trace Trace Trace The analyses for April, 1921, represent the original soil before the experiment was begun. TABLE 8 Chemical Effect of Iron Sulfate and Alum Depth of samples in inches Par s per mi lion COs CI Ca May 1922* Nov. 1922 Dec. 1926f May 1922* Nov. 1922 Dec. 1926t May 1922* Nov. 1922 Dec. 1926f f 0-12 356 185 210 598 40 70 Trace Trace Trace Plot 14, 1 12-24 132 72 225 246 260 177 Trace Trace Trace untreated 24-36 50 56 90 285 161 301 Trace Trace Trace I 36-48 37 56 60 348 77 230 Trace Trace Trace Plot 17, ( 0-12 313 20 47 349 69 32 Trace 32 Trace iron sulfate, 1 12-24 95 21 78 204 78 47 Trace Trace Trace 9 tons 24-36 49 51 71 134 117 134 Trace Trace Trace per acre [ 36-48 27 16 27 86 138 128 Trace Trace .19 Plot 18, f 0-12 298 2 41 314 109 65 Trace 256 Trace alum, ) 12-24 113 39 58 160 104 93 Trace Trace 12 11 tons 24-36 48 17 57 138 173 159 Trace Trace 32 per acre { 36-48 24 28 26 121 98 132 Trace Trace 50 * The analyses for May, 1922, represent the original soil before the experiment was begun, t These samples of plot 14 were drawn in December, 1927. BUL.455] RECLAMATION OF FRESNO TYPE OF BLACK ALKALI SOIL 29 Satisfactory growth of crops in this soil is conditioned upon the reduction of soluble carbonate (C0 3 ), at least in the upper layers of the soil, to less than 50 parts per million and the presence of an appreciable amount of soluble calcium. The analyses show that the chemical changes brought about by gypsum, sulfur, iron sulfate, and alum were in these directions. Several years ago theoretical consideration of this problem con- vinced us that each of these materials, if applied in sufficient amounts, would inevitably produce similar changes in the soluble carbonate and the clay constituents of the soil. A special study of these plots has fully verified this deduction/ 1 ' In the case where gypsum was applied, the soluble carbonate has been converted into insoluble calcium carbonate, and obviously the soluble calcium was furnished by the gypsum. On the other hand, an important part of the bene- ficial action of sulfur, iron sulfate, and alum depends on the presence of calcium carbonate in the soil. It is true each of these materials may neutralize the soluble carbonate, but the fundamentally important transformation of the clay constituents necessitates either that insoluble calcium minerals of the soil be made soluble or else that soluble calcium be added to the soil. As pointed out already, the oxidation products of sulfur will neutralize the soluble carbonate and dissolve calcium carbonate, and the calcium thus made soluble will in turn bring about the needed changes in the clay constituents. In the case of iron sulfate and alum the reactions are essentially the same. These materials are acidic owing to hydrolysis, the important hydrolytic product being sulfuric acid in both cases. It is this latter compound that brings about the important chemical transformations in the soil. Thus it will be seen that the success of treatments with sulfur, iron sulfate, and alum depend on the presence of calcium carbonate or some other readily decomposable calcium mineral in the soil. Other- wise the essential changes in the clay cannot be produced by these materials. Fortunately, the Fresno soil contains calcium carbonate in amounts sufficient for the purpose. In fact, American alkali soil usually contains more or less calcium carbonate. As Samuels pointed out, (3) sulfur, alum, and iron sulfate should ultimately produce better effects than a chemically equivalent amount of gypsum, because these materials are able to bring more calcium into solution from the carbonate form than is contained in the gypsum. 30 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION EFFECTS ON THE PHYSICAL CONDITION OF THE SOIL It is well known that the physical condition of black-alkali soil is almost always poor. The clay particles are cleflocculated, or readily become so, and water penetrates very slowly. After the major part of the soluble salts has been leached out, much of the irrigation water may stand on the surface until it evaporates. When dry the soil becomes extremely hard. Under the plow it breaks into large clods and a good seed bed is difficult to prepare. When the experiments were begun this was the general condition of all of these plots ; it is still the condition of check plots 14, 20, and 23 and parts of check plots 2, 5, and 8. On the other hand, all of the treated plots, except a few small spots in plot 11 (treated with gypsum) and a small part of plot 16 (treated with CaC0 3 and sulfur), have become friable, and water now penetrates them freely. The improved physical condition was noted soon after the gypsum was applied and was most pronounced where the largest amount was applied. Still more striking physical effects were produced by iron sulfate and alum. Where these materials were added the water that has been applied has been absorbed rapidly. On the other hand, the beneficial effect of sulfur was not immediate, in some cases the lapse of a year or more being required. Later, however, the physical condition of the soil was markedly improved by sulfur. With each of these materials the physical effect is dependent upon the chemical reactions that have already been discussed, and until these reactions proceed beyond a certain point the physical condition of the soil may not show any improvement whatever. When the chemical conditions are made favorable for plant growth the physical factors in this type of soil will probably take care of themselves. GENERAL DISCUSSION From the preceding discussion it is apparent that each of several different materials may be effective as a treatment for the Fresno type of black-alkali soil. The cost, however, is not the same. Sulfur is by far the most economical material that we have used in these experiments. An average application of not more than one ton per acre will probably suffice. At this rate the cost of the sulfur should not exceed $45.00 an acre. It is possible that a less expensive form of sulfur than that used in these experiments may give equally good results. By combining the use of sulfur with barnyard manure or BUL. 455] RECLAMATION OF FRESNO TYPE OF BLACK ALKALI SOIL 31 with the growing of an alkali-resistant legume, such as Melilotus alba, it is possible that an application of 1000 pounds of sulfur per acre will give good results. It seems probable that black-alkali soils of other types can also be successfully treated with sulfur, but it is certain that the amount of sulfur required for the best results will vary in different localities depending on the content of sodium carbonate and the amount of replaceable sodium present in the clay materials of the soil. More- over, there is no direct proportionality between the content of sodium carbonate and that of replaceable sodium. Generally speaking, heavy types of black-alkali soil will require more sulfur than light types which contain similar amounts of soluble carbonate, because the former are likely to contain the greater amounts of replaceable sodium, but this is not necessarily the case. In any event, it is im- portant for those engaged in black-alkali soil reclamation to realize that the required amount of sulfur is not necessarily determined solely by the content of sodium carbonate. In connection with the use of sulfur it is important to bear in mind that a period of several months, perhaps a year or more, will be required before the full effects will become manifest. Sulfur must undergo oxidation before it can produce any beneficial effect, and the oxidation is a relatively slow process. The results thus far obtained lend no support to the view that artificial inoculation of sulfur with oxidizing bacteria is either neces- sary or beneficial on this soil. It is evident that one or more of the species of sulfur-oxidizing bacteria occur in an active form in this soil, and nothing appears to be gained by special inoculation. This question is now being further investigated and the results will be reported later. It seems desirable to point out that sulfur is not recommended for any and all types of alkali soil. At present it can be safely recom- mended only for black-alkali types and alkali soils which contain relatively large amounts of replaceable sodium. The amount to be applied will depend on the content of sodium carbonate and replace- able sodium. We do not recommend sulfur where the soil contains a considerable amount of soluble calcium salt. Moreover, it is doubtful whether the benefits from sulfur treatment will be commensurate with the cost on any alkali soil unless the drainage conditions are favorable and unless the ground water level can be kept down continuously to a depth of six or more feet below the surface of the soil. Although excellent crops of alfalfa have been grown on several of these experimetnal plots, it would be premature and certainly 32 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION decidedly incorrect to conclude that the soil has as yet been completely reclaimed. The subsoil still contains an excess of sodium carbonate, and, as will be discussed in greater detail elsewhere, more thorough chemical study shows that, in other respects, the chemical changes requisite to the conversion of this alkali soil into normal soil have been only partially brought about. Nevertheless, it seems safe to say that it will not be necessary to apply an additional amount of any of these materials, with the possible exception of gypsum in the case of plot 11, and of sulfur in the case of plot 16, provided, however, that careful attention is given to the drainage conditions and to the general management and irrigation of the soil. As was pointed out above, the changes required to convert this soil into a state of normality necessitate the transformation of the clay constituents into calcium compounds. Thus far this change in the treated plots seems to have taken place to the extent of not more than 50 per cent of the theoretical possibility. This chemical effect, together with the removal of the excess of soluble salts from the upper layers, has greatly improved the conditions for crop growth, and the soil has been made reasonably porous and permeable to water. The remaining part of the chemical transformations can probably be brought about b/v growing deep-rooted crops, together with the liberal use of irrigation water and the plowing under of green manures. By these means the native calcium carbonate will slowly be made effective in promoting a continuation of the needed chemical changes in the soil. The length of time required to complete these changes cannot now be predicted. It seems reasonably certain that many years will be necessary for their completion. In the meantime successful crops can probably be grown. Whatever material is applied, it will probably be desirable to grow some leguminous crop, especially after a period of heavy leach- ing, since the available nitrate supply will be reduced to a low level. Nitrates are known to be easily leached out of soil. It is of the utmost importance for the practical farmer to realize that in the management and reclamation of alkali soil under the climatic conditions of California and other southwestern states, the abundant use of irrigation water is very desirable. It is well known that black-alkali soils absorb water very slowly, and unless they are irrigated at frequent intervals the crop will suffer because of excessive salt concentration and may even perish before it has time to become established. This is especially important in the early stages of the growth of alfalfa. In our experimental work we have found it neces- sary to irrigate young alfalfa about once every two weeks throughout BUL.45.1J RECLAMATION OF FRESNO TYPE OF BLACK ALKALI SOIL 33 the greater part of the first season. During the second year the irrigation in hot weather has been approximately once every three weeks. Moreover, it is important to apply the water in excess of the requirements of the crop in order to affect the subsoil favorably, but this should be done in such a way as to prevent injury to the crop. 11 is only as a result of the deep penetration of water applied to the surface that the effects of artificial treatments can be carried deep into the subsoil. The objective should be to transform the soil to a depth of at least four feet and possibly deeper, and the needed changes in the subsoil cannot be brought about quickly with the use of sulfur or any other material except at prohibitive cost. While these plots have not as yet been completely reclaimed, there is good reason to believe that the conditions for crop growth will continue to improve, provided the ground water is kept down below the zone of root development. In this connection it should be clearly understood that soluble salts are the cause either directly or indirectly of the adverse conditions of all alkali soils. Once the salts have been removed there is no possibility of their return unless the ground water approaches the surface, or unless saline irrigation water is applied. If the water table is not kept down, it is certain that the beneficial effects of the treatments will be merely temporary. Finally, we believe that the successful outcome of these reclama- tion experiments has not been due solely to the fact that good drainage conditions were established and the right kind of materials have been applied. The care that has been given to the details of soil prepara- tion and irrigation have probably also played an important part in the results. Before the various materials were applied the soil was carefully leveled so that each check could be irrigated uniformly. Subsequently it was necessary to relevel the soil from time to time owing to inequalities that developed in consequence of settling or plowing. It is especially important to provide a good seed bed and suitable moisture conditions before sowing alfalfa. If the soil is too dry, it may be necessary to irrigate before the seed will germinate, and, as is well known, this should be avoided on any soil ; its avoidance is even more important with alkali soils because of their pronounced tendency to bake. We have secured equally good results from seeding in October and February, but in neither case was it necessary to irrigate before the alfalfa seed germinated. As stated already, light irriga- tions were applied at approximately two-week intervals during the first summer after alfalfa was sown. Subsequently water has been applied freely and at intervals of approximately three weeks. 34 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION SUMMARY 1. Field experiments on Section 6 of the Kearney Vineyard have shown that the crop-producing power of the Fresno black-alkali soil can be greatly improved by the use of gypsum, sulfur, iron sulfate, or alum, provided these materials are applied in sufficient amounts. Yields of alfalfa ranging from 6 to 11 tons per acre per annum have been produced on land which at the beginning of the experiments was entirely unproductive. 2. The unproductivity of this soil is due to (a) an excess of soluble salts, especially sodium carbonate, and (&) the abnormal chemical composition of the clay-like constituents of the soil. The reclamation of the soil must involve the removal of the excess of soluble salts and the conversion of at least a part of the clay constituents into calcium compounds. The former may be leached out, but ordinary leaching fails to bring about the needed chemical changes in the latter. 3. Gypsum, sulfur, iron sulfate, and alum produce beneficial effects on black-alkali soils but at different rates. These materials act on the soluble carbonate and the clay constituents simultaneously. Gypsum brings about these changes because of its soluble calcium, while the effect of sulfur, iron sulfate, and alum is due to their acidic nature, in consequence of which soluble carbonate is decomposed and calcium minerals of the soil, especially calcium carbonate, are dis- solved. The calcium thus brought into solution reacts with the clay constituents. Iron sulfate and alum react with the soil most quickly because of their high solubility and acidic nature. Sulfur acts most slowly for the reason that this material must undergo oxidation before it can produce any important effect on the soil. 4. Gypsum produced uniformly successful results on this soil only when applied at the rate of more than 10 tons per acre. Relatively large amounts of iron sulfate and alum are also required. On the other hand, excellent results have been obtained by applying not more than one ton of sulfur per acre. 5. Sulfur has proven to be much more economical than the other materials. Large yields of alfalfa have been produced on soil that was badly affected with alkali and entirely unproductive at the outset by applying one ton of sulfur per acre; when used in conjunction with stable manure, 1000 pounds of sulfur per acre has given good results. BUL. 455] RECLAMATION OF FRESNO TYPE OF BLACK ALKALI SOIL 35 6. It seems safe to say that where the natural drainage conditions are satisfactory the Fresno black-alkali soil similar to that on the Kearney Vineyard can be reclaimed with sulfur at an average cost for the material itself of not more than $45.00 an acre. By combin- ing the use of sulfur with farm manures or the growing of alkali- resistant leguminous cover crops, it is probable that the cost will be somewhat less. This estimate is based on the current quotation of $45.00 a ton for practically pure, finely ground sulfur. Should it be found that some one or more of the less expensive forms of crude sulfur concentrates are effective, and it seems probable that such will be the case, then the cost would be materially reduced. 7. Since the Fresno black-alkali soil is extremely variable as regards the alkali conditions, we would recommend that sulfur be applied at the rate of 1000 pounds per acre. If after the lapse of two or more years, the growth of alfalfa or other crops should indicate the need for further treatment, an additional application should be made in accordance with that indication. It w r ill probably not be necessary to make a second application except on scattered spots. By following this plan the average cost for materials will probably be considerably less than $45.00 an acre. 8. Leaching experiments without the application of any material except water have thus far failed to bring about a satisfactory recla- mation of this soil. 9. Although the beneficial effect of sulfur is dependent on oxidation, and the oxidation is brought about by certain species of bacteria, these experiments have not shown any special advantage for artificially inoculated sulfur over that obtained from uninoculated sulfur. Active forms of one or more of the species of sulfur-oxidizing bacteria occur naturally in this soil and nothing appears to be gained by special inoculation of the sulfur. 10. It is important for the farmer to realize that several months and possibly a year or more must elapse after the sulfur is applied before the full effect will be exerted. During this period the soil should be kept moist and well tilled in order to promote aeration and hasten the rate of sulfur oxidation. It is probably best not to leach the soil during this stage of the work. If the content of soluble salts is high, the soil should be flooded prior to the time of planting a crop in order to leach out the excess of salts. If, on the other hand, the salt content is not especially high, water applied rather freely to the crops that are grown by the flooding system of irrigation will probably suffice to leach out the salts and wash down the sulfur-oxidation products and thus gradually ameliorate the subsoil. 36 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION 11. It is extremely important to irrigate young alfalfa on alkali soil at frequent intervals. In fact, success in the growth of crops during the early stages of the reclamation will probably depend largely on careful attention to the irrigation program, and especially so during the time when young alfalfa is becoming established. We have found it desirable to irrigate about twice a month during the first summer after alfalfa was planted. Alkali soil should never be allowed to dry out while crops are being grown, and success in alkali reclamation will depend to an unusual degree upon careful attention to the details of soil preparation and irrigation. Alkali soils are much more difficult to manage successfully than normal soils. 12. Unless the soil is well drained there is no reasonable prospect of permanently reclaiming any alkali soil. We believe that under the soil and climatic conditions that prevail in the Fresno section, the water table should never be less than 6 feet below the surface, and preferably deeper. Since 1924 the minimum depth to the ground water on the experimental area at Kearney Park has been about 8 feet, and during a large part of each year it has been 12 feet or more. The ground water has been kept down not by the tile-drainage system, but by pumping from a deep well located near the experimental area. This water is used on the experimental plot as the sole source of irrigation supply. When not needed for this purpose it is discharged into a near-by canal. The pump is kept in continuous operation about eight months each year. 13. The first foot of soil has been practically freed from soluble carbonate as a result of applying gypsum, sulfur, iron sulfate, or alum, and the second foot has been materially improved. The third and fourth feet of certain plots have also been affected. Nevertheless, none of the plots can yet be said to be completely reclaimed. It is probable, however, that by giving careful attention to the drainage, irrigation, and general management of the soil, the conditions will continue to improve, and that ultimately this soil will be completely restored to a condition of normality. BuL. 455] RECLAMATION OF FRESNO TYPE OF BLACK ALKALI SOIL 37 LITERATURE CITED i Kelley, W. P., and Alexander Arany. . The chemical effect of gypsum, sulfur, iron sulfate, and alum on alkali soil. Hilgardia. (In press.) a Kelley, Walter P., and Edward E. Thomas. 1923. The removal of sodium carbonate from soils. California Agr. Exp. Sta. Tech. Paper 1:1-24. 8 Samuels, Charles Danziger. 1927. The oxidation of sulfur in alkali soil and its effect on the replaceable bases. Hilgardia 3:1-26. 4 Weir, Walter W. 1916. Preliminary report on Kearney Vineyard experimental drain. Cali- fornia Agr. Exp. Sta. Bui. 273:103-123. s Weir, Walter W. 1927. Effect of pumping from deep wells on the ground-water table. Jour. Agr. Research 34:663-672. ' STATION PUBLICATIONS AVAILABLE FOR FREE DISTRIBUTION BULLETINS No. No. 253. Irrigation and Soil Conditions in the 386. Sierra Nevada Foothills, California. 262. Citrus Diseases of Florida and Cuba 387. Compared with those of California. 388. 263. Size Grades for Ripe Olives. 268. Growing and Grafting Olive Seedlings. 389. 273. Preliminary Report on Kearney Vine- 390. yard Experimental Drain, Fresno County, Calif. 391. 277. Sudan Grass. 278. Grain Sorghums. 392. 279. Irrigation of Rice in California. 393. 283. The Olive Insects of California. 394. 304. A Study of the Effects of Freezes on Citrus in California. 310. Plum Pollination. 395. 313. Pruning Young Deciduous Fruit Trees. 396. 324. Storage of Perishable Fruits at Freez- ing Temperatures. 397. 328. Prune Growing in California. 331. Phylloxera-resistant Stocks. 398. 335. Cocoanut Meal as a Feed for Dairy 400. Cows and Other Livestock. 402. 340. Control of the Pocket Gopher in 404. California. 405. 343. Cheese Pests and Their Control. 406. 344. Cold Storage as an Aid to the Mar- 407. keting of Plums, a Progress Report. 347. The Control of Red Spiders in Decid- uous Orchards, 408. 348. Pruning Young Olive Trees. 409. 349. A Study of Sidedraft and Tractor Hitches. 350. Agriculture in Cut-Over Redwood Lands. 410. 353, Bovine Infectious Abortion, and As- sociated Diseases of Cattle and New- born Calves. 411. 354. Results of Rice Experiments in 1922. 357. A Self-Mixing Dusting Machine for 412. Applying Dry Insecticides and Fun- gicides. 358. Black Measles, Water Berries, and 414. Related Vine Troubles. 361. Preliminary Yield Tables for Second- 415. Growth Redwood. 416. 362. Dust and the Tractor Engine. 363. The Pruning of Citrus Trees in Cali- 417. fornia. 364. Fungicidal Dusts for the Control of 418. Bunt. 366. Turkish Tobacco Culture, Curing, 419. and Marketing. 367. Methods of Harvesting and Irrigation 420. in Relation to Moldy Walnuts. 368. Bacterial Decomposition of Olives 421. During Pickling. 422. 369. Comparison of Woods for Butter Boxes. 423. 370. Factors Influencing the Development of Internal Browning of the Yellow 424. Newton Apple. 371. The Relative Cost of Yarding Small 425. and Large Timber. 426. 373. Pear Pollination. 374. A Survey of Orchard Practices in 427, the Citrus Industry of Southern California. 428. 375. Results of Rice Experiments at Cor- tena, 1923, and Progress in Experi- ments in Water Grass Control at the 429. Biggs Rice Field Station, 1922-23. 430. 377. The Cold Storage of Pears. 431. 379. Walnut Culture in California. 380. Growth of Eucalyptus in California 432. Plantations. 382. Pumping for Draininge in the San 433. Joaquin Valley, California. 385. Pollination of the Sweet Cherry. Pruning Bearing Deciduous Fruit Trees. Fig Smut. The Principles and Practice of Sun- Drying Fruit. Berseem or Egyptian Clover. Harvesting and Packing Grapes in California. Machines for Coating Seed Wheat with Copper Carbonate Dust. Fruit Juice Concentrates. Crop Sequences at Davis. I. Cereal Hay Production in Cali- fornia. II. Feeding Trials with Cereal Hays. Bark Diseases of Citrus Trees in Cali- fornia. The Mat Bean, Phaseolus Aconitifo- lius. Manufacture of Roquefort Type Cheese from Goat's Milk. Orchard Heating in California. The Utilization of Surplus Plums. The Codling Moth in Walnuts. The Dehydration of Prunes. Citrus Culture in Central California. Stationary Spray Plants in California. Yield, Stand, and Volume Tables for White Fir in the California Pine Region. Alternaria Rot of Lemons. The Digestibility of Certain Fruit By- products as Determined for Rumi- nants. Part I. Dried Orange Pulp and Raisin Pulp. Factors Influencing the Quality of Fresh Asparagus after it is Har- vested. Paradichlorobenzene as a Soil Fumi- gant. A Study of the Relative Value of Cer- tain Root Crops and Salmon Oil as Sources of Vitamin A for Poultry. Planting and Thinning Distances for Deciduous Fruit Trees. The Tractor on California Farms. Culture of the Oriental Persimmon in California. Poultry Feeding: Principles and Prac- tice. A Study of Various Rations for Fin- ishing Range Calves as Baby Beeves. Economic Aspects of the Cantaloupe Industry. Rice and Rice By-Products as Feeds for Fattening Swine. Beef Cattle Feeding Trials, 1921-24. Cost of Producing Almonds in Cali- fornia : a Progress Report. Apricots (Series on California Crops and Prices). The Relation of Rate of Maturity to Egg Production. Apple Growing in California. Apple Pollination Studies in Cali- fornia. The Value of Orange Pulp for Milk Production. The Relation of Maturity of Cali- fornia Plums to Shipping and Dessert Quality. Economic Status of the Grape Industry. Range Grasses of California. Raisin By-Products and Bean Screen- ings as Feeds for Fattening Lambs. Some Economic Problems Involved in the Pooling of Fruit. Power Requirements of Electrically Driven Manufacturing Equipment. No. 434. Investigations on the Use of Fruits in Ice Cream and Ices. 435. The Problem of Securing Closer Relationship Between Agricultural Development and Irrigation Con- struction. 436. I. The Kadota Fig. II. Kadota Fig Products. 437. Economic Aspects of the Dairy In- dustry. 438. Grafting Affinities with Special Refer- ence to Plums. 439. The Digestibility of Certain Fruit By- products as Determined for Rumi- nants. Part II. Dried Pineapple Pulp, Dried Lemon Pulp, and Dried Olive Pulp. BULLETINS— (Continued) No. 440. The Feeding Value of Raisins and Dairy By-Products for Growing and Fattening Swine. 441. The Electric Brooder. 442. Laboratory Tests of Orchard Heaters. 443. Standardization and Improvement of California Butter. 444. Series on California Crops and Prices: Beans. 445. Economic Aspects of the Apple In- dustry. CIRCULARS No. 257. No. 87. Alfalfa. 115. Grafting Vinifera Vineyards. 117. The selection and Cost of a Small 258. Pumping Plant. 259. 127. House Fumigation. 261. 129. The control of Citrus Insects. 264. 136. Melilotus Indica as a Green-Manure Crop for California. 265. 144. Oidium or Powdery Mildew of the 266. Vine. 157. Control of Pear Scab. 267. 164. Small Fruit Culture in California. 166. The County Farm Bureau. 269. 173. The Construction of the Wood-Hoop 270. Silo. 273. 178. The Packing of Apples in' California. 276. 179. Factors of Importance in Producing 277. Milk of Low Bacterial Count. 202. County Organization for Rural Fire 278. Control. 203. Peat as a Manure Substitute. 279. 209. The Function of the Farm Bureau. 212. Salvaging Rain-Damaged Prunes. 281. 215. Feeding Dairy Cows in California. 217. Methods for Marketing Vegetables in California. 282. 230. Testing Milk, Cream, and Skim Milk for Butterfat. 283. 231. The Home Vineyard. 284. 232. Harvesting and Handling California 286. Cherries for Eastern Shipment. 287. 234. Winter Injury to Young Walnut 288. Trees During 1921-1922. 289. 238. The Apricot in California. 290. 239. Harvesting and Handling Apricots 292. and Plums for Eastern Shipment. 293. 240. Harvesting and Handling California 294. Pears for Eastern Shipment. 296. 241. Harvesting and Handling California Peaches for Eastern Shipment. 298. 243. Marmalade Juice and Jelly Juice from Citrus Fruits. 300. 244. Central Wire Bracing for Fruit Trees. 301. 245. Vine Pruning Systems. 302. 248. Some Common Errors in Vine Prun- 304. ing and Their Remedies. 305. 249. Replacing Missing Vines. 306. 250. Measurement of Irrigation Water on the Farm. '307. 252. Support for Vines. 308. 253. Vineyard Plans. 309. 254. The Use of Artificial Light to In- 310. crease Winter Egg Production. 255. Leguminous Plants as Organic Per- 311. tilizers in California Agriculture. The publications listed above may be had by addressing College of Agriculture, University of California, Berkeley, California. 16to-6,'28 The Small-Seeded Horse Bean (Vicia faba var. minor). Thinning Deciduous Fruits. Pear By-Products. Sewing Grain Sacks. Preliminary Essentials to Bovine Tu- berculosis Control in California. Plant Disease and Pest Control. Analyzing the Citrus Orchard by Means of Simple Tree Records. The Tendency of Tractors to Rise in Front; Causes and Remedies. An Orchard Brush Burner. A Farm Septic Tank. Saving the Gophered Citrus Tree. Home Canning. Head, Cane and Cordon Pruning of Vines. Olive Pickling in Mediterranean Countries. The Preparation and Refining of Olive Oil in Southern Europe. The Results of a Survey to Deter- mine the Cost of Producing Beef in California. Prevention of Insect Attack on Stored Grain. Fertilizing Citrus Trees in California. The Almond in California. Milk Houses for California Dairies. Potato Production in California. Phylloxera Resistant Vineyards. Oak Fungus in Orchard Trees. The Tangier Pea. Alkali Soils. The Basis of Grape Standardization. Propagation of Deciduous Fruits. Control of the California Ground Squirrel. Possibilities and Limitations of Coop- erative Marketing. Coccidiosis of Chickens. Buckeye Poisoning of the Honey Bee. The Sugar Beet in California. Drainage on the Farm. Liming the Soil. A General Purpose Soil Auger and Its Use on the Farm. American Foulbrood and Its Control Cantaloupe Production in California. Fruit Tree and Orchard Judging. The Operation of the Bacteriological Laboratory for Dairy Plants. The Improvement of Quality in Figs.