UNIVERSITY OF CALIFORNIA COLLEGE OF AGRICULTURE AGRICULTURAL EXPERIMENT STATION BERKELEY, CALIFORNIA EFFECT OF FERTILIZERS ON ORANGE YIELDS E. R. PARKER and L. D. BATCHELOR BULLETIN 673 October, 1942 UNIVERSITY OF CALIFORNIA BERKELEY, CALIFORNIA CONTENTS PAGE Introduction 3 Progress report on fertilizer trial at the Citrus Experiment Station, Riverside, California 4 Conditions of the experiment 4 Relative effectiveness of different nitrogenous materials 5 Effect of time and frequency of application of nitrate fertilizers 7 The effect of winter covercrops and other sources of organic matter on crop production 9 Effect of supplementing nitrogenous fertilizer with phosphate and potash .... 13 Effect of various kinds and amounts of bulky organic fertilizer alone and in combination with concentrated sources of nitrogen 14 Results of different methods of using dairy manure 17 The effect of using agricultural minerals in conjunction with adequate fertilizer 18 The effect of applying various amounts of nitrogen 20 Fertilizer experiment at Arlington, Riverside County 25 Kern County and Tulare County fertilizer experiments 26 Kern County experiment 28 Tulare County experiment 30 Discussion of results of Kern County and Tulare County experiments 32 Orange County experiments 33 Experiment at Fullerton 34 Experiment at Orange 35 Discussion of results of Orange County experiments 36 Discussion and summary 37 Acknowledgments 39 EFFECT OF FERTILIZERS ON ORANGE YIELDS E. E. PAEKER 3 and L. D. BATCHELOR 4 1> 2 INTRODUCTION During the past thirty-five years, the University of California College of Agriculture has conducted a total of thirteen field experiments, inde- pendently or cooperatively, for the purpose of learning more about the relative effectiveness of different fertilizer materials in the production of citrus fruit. Two of these experiments have been complex, long-term fertilizer trials on property of the University of California Citrus Ex- periment Station at Riverside. One trial was begun in 1907 and con- tinued for twenty -three years. The other was begun in 1917 and is still in effect. The rest have been short-term trials, carried out for periods ranging from three to twelve years, on properties of cooperating or- chardists, in locations scattered from Tulare County to San Diego County. Nine different soil types, eight of which belong to entirely different soil series contrasting as widely as Porterville adobe and Hanf ord grav- elly loam, have been represented in these experiments. All of the soils are widely used in California as orchard sites for oranges and lemons. Results of some of the early experiments were reported by Vaile 5 in 1922 and by Surr and Batchelor 6 in 1926. An extensive survey of orchard practices, including fertilization, was reported by Vaile 7 in 1924. The present paper reports results of the six most recent trials, con- ducted from 1921 to 1940. Results of the current experiment at Riverside are given in the form of a progress report. Results of the five other, co- operative experiments are briefly summarized. While some of the coop- erative trials have been maintained for relatively brief periods of time, the data obtained from these trials are pertinent to the general problem and are significant for short periods of treatment, such as the probable duration of a war, during which restrictions may be placed on the sale 1 Eeceived for publication July 22, 1942. 2 Paper No. 466, University of California Citrus Experiment Station, Kiverside, California. 3 Associate Horticulturist in the Experiment Station. 4 Professor of Horticulture in the Citrus Experiment Station, Horticulturist in the Experiment Station, and Director of the Citrus Experiment Station. 5 Vaile, R. S. Fertilizer experiments with citrus trees. California Agr. Exp. Sta. Bui. 345:465-512. 1922. (Out of print.) 6 Surr, Gordon J., and L. D. Batchelor. Citrus culture in central California. Cali- fornia Agr. Exp. Sta. Bui. 405:1-23. 1926. (Out of print.) 7 Vaile, Roland S. A survey of orchard practices in the citrus industry of south- ern California. California Agr. Exp. Sta. Bui. 374:1-40. 1924. (Reprinted June, 1929. Out of print.) [3] 4 University of California — Experiment Station of certain fertilizer materials. These data serve, also, to provide infor- mation regarding the responses of citrus trees to fertilization under a wide range of soil and cultural conditions and thus permit more general conclusions regarding the value of various fertilizers. PROGRESS REPORT ON FERTILIZER TRIAL AT THE CITRUS EXPERIMENT STATION, RIVERSIDE, CALIFORNIA Conditions of the Experiment. — The Washington Navel orange trees in this experiment were planted in 1917 (in rows 24 feet apart, with trees 20 feet apart in the rows) on soil classified as Ramona loam. The site had not previously been irrigated. The orchard was first fertilized in the spring of 1927. Prior to that time, winter covercrops of yellow bitter clover (Melilotus indica) or purple vetch (Vicia atropurpurea) had been grown annually throughout the orchard; during the first six years of the period, summer covercrops of black-eye beans or cowpeas (Vigna sinensis) had also been grown. The cultural practice during the preliminary period and the outline of the experiment were reported in 1928. 8 A full description of the method of designing the plot distribution of this experiment was published in 1932. 9 Those who may be especially interested in the details of the planning are referred to the complete reports. It should suffice to state here that in each unit plot there were 8 orange trees. Each fertilizer treatment was repeated in 4 unit plots in the orchard. Thirty-two trees were thus included in each treatment. The total area fertilized in each treatment, including the area adjacent to the guard trees of each plot, was slightly more than three fourths of an acre. The treatments herein reported were carried out through a twelve- year period, 1928-1939. Most of the fertilizer applications were made in rather small amounts, in comparison with applications customary in comparable commercial orchards. This was considered desirable, especially in making compari- sons of the relative effectiveness of different nitrogenous fertilizers. Only 1 pound of actual nitrogen per tree was therefore applied annually in most of the treatments. Very heavy applications of nitrogen might have been in excess of actual needs ; thus the least effective material might still have provided ample nitrogen. In that case the relative effectiveness of the different methods of fertilization would have been masked. The 8 Batchelor, L. D., E. B. Parker, and Eobert McBride. Studies preliminary to the establishment of a series of fertilizer trials in a bearing citrus grove. California Agr. Exp. Sta. Bui. 451:1-49. 1928. (Out of print.) 9 Parker, E. E., and L. D. Batchelor. Variation in the yields of fruit trees in rela- tion to the planning of future experiments. Hilgardia 7(2) :81-161. 1932. Bul. 673] Effect of Fertilizers on Orange Yields 5 production of the orchard has, in general, been light, owing, in part at least, to the small amount of fertilizer applied. The rate of fertilizer dosage was increased threefold in the fall of 1939, after the conclusion of the experimental period under discussion, and a very decided increase in yield had already followed by the time of the 1941 harvest. Cultural operations and pest control in the experimental orchard have been equivalent to good practice in average commercial groves. No at- tempt has been made to exercise extraordinary care in these respects. Where the penetration of water into the soil has been aifected by the treatments, special care has been used at critical times to insure an ade- quate supply of water to the tree roots; otherwise, the irrigation has been equivalent to the usual orchard practice in this district. Of the 44 fertilizer treatments used in this orchard, there are a num- ber which may logically be grouped for easy comparison. For the pur- pose of this report, the pertinent data of small groups of treatments will be presented briefly by tabular means. It has seemed to the writers that a statement of the relative yields of oranges produced by the several fertilizer treatments is the clearest method of recording the data at this time. A more complete, technical presentation may be made at a later date. It seems probable that the conclusions presented herein will, for practical purposes, be about the same as those based upon more extensive analyses. The present report has been greatly facilitated by the design of the experiments and, particularly, by the precautions taken in the choosing and combining of the plots for the treatments. This was done in such a manner that the average yields of the 4 combined plots devoted to each treatment were approximately equal for a period prior to the starting of the fertilizer trials. During this preliminary period all the plots had received the same culture. Relative Effectiveness of Different Nitrogenous Materials. — The fact that applications of nitrogen-containing fertilizers are of paramount importance in the fertilization of citrus trees in California, has been established by experiments previously mentioned. It becomes relevant to the general problem, therefore, to determine whether some materials are more effective than others as sources of nitrogen and thus possibly more desirable for use. By "desirable" is meant generally effective and, all things considered, economical and thus profitable. The prices of the different nitrogen-containing materials change from time to time and vary in different localities. In comparing prices of a list of commonly used materials, the actual cost of a unit of nitrogen may be found to vary over a range of 100 per cent. No attempt is made here to describe the trends in prices of such materials ; these are usually 6 University of California — Experiment Station known locally. If the relative effectiveness of nitrogen from the several sources procurable is known, each orchardist may decide for himself the most profitable fertilizer to buy. The comparative effectiveness of various fertilizers is shown by means of relative adjusted 10 orange yields in table 1. The treatments include six different nitrogenous concentrates, nitrate of soda in combination with gypsum, and a mixture of three different fertilizer materials (treat- TABLE 1 Eelative Orange Yields Showing Effectiveness of Various Concentrated Nitrogenous Fertilizers* Treatment 1928-1931 1932-1935 1936-1939 1928-1939 Mean annual yield per tree, pounds 27. Nitrate of soda 27. Nitrate of soda 28. Nitrate of soda and gypsum 21. Nitrate of lime 15. Sulfate of ammonia 18. Urea 17. Dried blood 43. Cottonseed meal 12. Mixed sources of nitrogenf. . 114 145 120 12G Relative yield (treatment 27 = 100) 100 105 108 98 104 100 94 103 100 100 105 103 110 99 97 95 98 100 103 100 101 94 102 96 100 105 106 97 100 101 97 100 * Fertilizer material applied broadcast in the spring at a rate to supply 1 pound of nitrogen per tree annually. Used with winter covercrop in all cases. t Nitrate of soda, dried blood, and sulfate of ammonia, supplying equal amounts of nitrogen. ment 12) as sources of nitrogen. The twelve-year period represented by these yield data is, for convenience, divided into three four-year periods. This segregation makes any change in trend readily apparent. It also smooths out the annual variations in the yield comparisons. The mean relative yields for the entire twelve-year period, 1928-1939, are included in table 1. With the mean yield of nitrate of soda (treatment 27) taken as 100, comparisons indicate that there probably has not been any con- sistent superiority in the effectiveness of these different nitrogen car- riers. Differences shown here are considered within the realm of chance. 10 The design of the experiment has made possible the reduction of certain sources of variation inherent in all field experiments. The actual yields of all plots, with the exception of those serving as check plots (treatment C) and of those receiving no fertilizer (treatments 1 and 6), have therefore been adjusted by covariance on check- plot yields. Variation correlated with the yields of trees before the start of the ferti- lizer experiment has also been reduced by elimination of "yield-group" effects based on 1927 yields. The relative yields reported are calculated from the adjusted yields. Bul. 673] Effect of Fertilizers on Orange Yields 7 It is not practical to present the annual yields in detail for compari- son. Reference to those data shows, however, that cottonseed meal (treat- ment 43), which was lowest in production, relative to nitrate of soda (treatment 27), actually produced more fruit than nitrate of soda five years out of the twelve. Urea (treatment 18) shows, in the annual comparisons, a greater production than nitrate of soda (treatment 27) four years out of twelve and an equal amount one year. Comparisons of yields on the annual basis show, also, that nitrate of soda plus gypsum (treatment 28), as well as nitrate of lime (treatment 21), produced more fruit than nitrate of soda alone (treatment 27) eight years out of twelve ; while for the remaining four years, the converse was true. The ranges of these annual comparisons were, for nitrate of soda plus gyp- sum (treatment 28), from 19 per cent more to 5 per cent less than for nitrate of soda (treatment 27) ; and, for nitrate of lime (treatment 21), from 14 per cent more to 10 per cent less than for nitrate of soda (treat- ment 27 ) . It seems probable, therefore, that the small, relative differences ex- pressed in table 1 as four-year averages, and finally as twelve-year aver- ages, are chance variations due to factors which are wholly or partially beyond control, such as soil, weather, and cultural practice. As far as this trial is concerned, these several nitrogenous fertilizer materials have been, essentially, equally effective. Effect of Time and Frequency of Application of Nitrate Fertilizers. — Probably the most important single factor affecting the response of citrus trees to applications of soluble nitrate fertilizers is leaching of such materials from the soil before their absorption by the trees. It is conceiv- able, therefore, that it might be advisable to apply nitrates with due consideration of seasonal rainfall and applications of irrigation water. Leaching varies greatly with different soils, especially in relation to the slope of the land, the permeability of the surface soil, and the natural drainage conditions of the subsoil. The absorption of nitrogen by cov- ercrops reduces losses by leaching, and the decomposition of covercrop growth in the spring results in the formation of nitrates and thus in- creases the availability of nitrogen to the trees at that season. The average annual rainfall at Riverside is 11.1 inches; but during the twelve-year period under consideration (1928-1939) the annual rainfall ranged from 5.1 to 21.5 inches and averaged 11.7 inches. The quantity of irrigation water applied to the test plots annually has been 30 to 36 acre-inches per acre. The Ramona loam on which the experimental orchard is growing is well drained, although percolation is slow. This soil is not so readily 8 University of California — Experiment Station permeable to rainfall and irrigation water as several other soil types on which citrus is grown in California, and there is considerable runoff of rain water in the winter period even where the land is covercropped. This is especially true after the 3 or 4 feet of surface soil have been filled with water to field capacity. The plots have an average slope of 1.5 per cent. It seems probable that the loss of nitrates as a result of leaching in this orchard would be somewhat less than that under conditions in many California citrus orchards, but there are some other areas where leaching would be even less than in this experimental orchard. TABLE 2 Eelative Orange Yields Kesulting from the Application of Nitrate of Lime at Different Times of Year* Treatment 21. Single application in February 21. Single application in February 23. Single application in October 24. Equal applications in February, June, and October 6. Covercrop onlyf 1928-1931 1932-1935 193G-1939 1928-1939 Mean annual yield per tree, pounds 123 160 118 134 Relative yield (treatment 21 = 100) 100 98 92 67 100 100 98 122 84 108 42 17 100 105 94 42 * Spring applications of nitrate were broadcast; summer and fall applications were made in irrigation furrows. Fertilizer applied in amounts to supply 1 pound of nitrogen per tree annually and used with winter coverrrops. t Yields unadjusted. These experiments involved the same total amount of nitrogen ap- plied annually, as nitrate of lime, in three comparable treatments. In the first treatment, all the nitrate was applied in February ; in the sec- ond, all was applied in October; and in the third, equal applications were made in February, June, and October. No other fertilizer was applied, although winter covercrops were grown. Only 1 pound of nitrogen, a relatively small amount, was applied annually per tree in each treat- ment. The spring applications were broadcast ; the summer and fall applications were made in the irrigation furrows prior to an irrigation. The relative yields resulting from the applications of nitrate at the different seasons are shown in table 2, in comparison with yields of trees receiving no fertilizer. During the second four-year period there was apparently a slight disadvantage in applying the fertilizer in three ap- plications, as compared with the single applications. During the last Bul. 673] Effect of Fertilizers on Orange Yields 9 four-year period there was apparently an advantage in the single ap- plication of nitrate in October. It was noticeable, especially during the last four years covered by the data in table 2, that the plots fertilized only in October produced better covercrops than the plots fertilized at other times of year. The results indicate that, when relatively small amounts of nitrogen are used, the superior growth of winter covercrops may be important. Under conditions of nitrogen deficiency, the fall fer- tilizer application is apparently more favorable to the growth of cover- crops than application at other times 11 and may thereby favorably affect tree yields. There has, however, so far as orange yields during the twelve- year period of these experiments are concerned, been no striking and consistent advantage in applying nitrate of lime at any particular time of year. The range of variation in the averages for the twelve-year period are believed to be within the realm of chance. The Effect of Winter Covercrops and Other Sources of Organic Mat- ter on Crop Production. — There are two general classes of organic mat- ter which the citrus grower in California may find of practical use, namely, covercrops grown during the rainy season, and manures or other bulky organic materials such as cotton hulls and the various hays and straws. The merit of these will vary widely in different localities and will depend upon the rainfall, or upon the cost of irrigation water necessary to grow a covercrop, and upon the proximity and prices of supplies of bulky material which may be purchased. The value of winter covercropping in California, as a factor in the prevention of soil erosion, has long been known. The cost of growing a covercrop and the possible inconvenience in connection with it, if the orchard has to be heated, have deterred its use as a universal practice. The advantages may greatly outweigh the disadvantages, however, if the beneficial effect of the covercrop on the trees and on the yields of fruit is great enough. The increased fertility of the soil, as measured by the greater produc- tion of annual crops following the use of covercrops, was show T n by one of the early plot experiments 12 conducted at the Citrus Experiment Station. The beneficial effect that the incorporation of organic matter has upon the physical condition of agricultural soils has been recognized by farmers for many generations. The favorable effect of such applica- tions upon the readiness with which soils absorb water has been generally observed in California orchards for several decades. This is especially 11 With the larger amounts of fertilizer used annually since the fall of 1939, the winter covercrops have grown better than heretofore on all plots of these treatments. n Mertz, W. M. Green manure crops in southern California. California Agr. Exp. Sta. Bul. 292:1-31. 1918. (Out of print.) 10 University of California — Experiment Station true on the older granitic soils, such as those of the Ramona and Pla- centia series. In the experiments reported in the present paper, three concentrated nitrogenous fertilizers, namely, nitrate of soda, nitrate of lime, and Fig. 1. — Orange tree showing mottle-leaf condition. This tree was fertilized with nitrogen, but without organic matter. Compare with figure 2. urea, were used separately with winter covercrops of mustard (Brassica nigra, cult. ) and bitter clover. The same materials were also used with- out covercrops and in the same number of unit plots. The relative yields resulting from these treatments are presented in table 3. Results of two other experiments, in which dairy manure, a source of organic matter and nitrogen, was used with covercrops, have also been included in this Bul. 673 J Effect of Fertilizers on OrxVnge Yields 11 table for comparison. The last two four-year periods show a clear-cut benefit resulting from the practice of growing winter covercrops. It seems probable that the lack of greater effectiveness from the growing of covercrops during the first four-year period may have been due to the Fig. 2. — Same tree as that shown in figure 1, ten months after zinc treatment. residual benefits of ten years of covercropping of all plots prior to the earliest period under consideration in table 3. During the last eight years, however, there was an average decline in production of about 20 per cent when the covercrops were omitted. We are of the opinion that this difference would probably have been much greater had not all plots been periodically sprayed with zinc compounds to prevent mottle-leaf. 12 University of California — Experiment Station Before the spraying with zinc compounds was begun, in 1934, the trees in the clean-culture plots were much more seriously affected with mottle-leaf than those in comparable covercropped plots. During the period 1932-1935, there were individual years when the clean-culture plots fertilized with nitrate of soda and those fertilized with nitrate of lime declined in yield 35 and 38 per cent, respectively, as compared with comparable covercropped plots. Spraying applications cured practically all the mottle-leaf by the summer of 1935 and eliminated the depressing TABLE 3 Eelative Orange Yields Showing Effect of Organic Matter from Covercrops and Other Sources, When Used with Nitrogenous Fertilizers Treatment* 18 (urea), (21 nitrate of lime), and 27 (nitrate of soda) averaged ; with covercrop 18 (urea), 21 (nitrate of lime), and 27 (nitrate of soda) averaged ; with covercrop 2 (urea), 20 (nitrate of lime), and 26 (nitrate of soda) averaged ; without covercrop C. Urea and manure, t with covercropt 29. Nitrate of soda and manure, t with covercrop 1928-1931 1932-1935 1936-1939 1928-1939 Mean annual yield per tree, pounds 118 149 119 129 Relative yield (average of treatments 18, 21, and 27 = 100) 100 94 96 97 100 100 83 81 107 102 107 113 100 86 102 106 * Nitrogen supplied at the annual rate of 1 pound per tree, from all sources. t The concentrated fertilizer and manure supplyingjequal amounts of nitrogen. t Yields unadjusted. effects of mottle-leaf on yields of the two sets of treatments. 13 The effect of the zinc treatment on one of the trees in a clean-culture plot is shown in figures 1 and 2. This treatment did not entirely equalize the yields of plots with and without covercrops, however, perhaps because of the poor physical condition of the soil in plots which were not covercropped or which had no applications of organic matter from bulky fertilizers. The lack of organic matter in the nitrogen-fertilizer plots resulted in more compact soil, which absorbed water slowly. The penetration and movement of irrigation water was therefore restricted in the soil of these plots. The effect of applying additional organic matter, in manure, to plots that were covercropped was of little consequence, provided the total amount of nitrogen applied was the same in all plots. In other words, 13 Parker, E. E. Experiments on the treatment of mottle-leaf of citrus trees. III. Amer. Soc. Hort. Sci. Proc. 34:213-15. 1937. Bul. 673] Effect of Fertilizers on Orange Yields 13 where covercrops were grown, only a very small, and perhaps nonsignifi- cant, increase in yield resulted from the use of additional organic matter. Effect of Supplementing Nitrogenous Fertilizer with Phosphate and Potash. — The response of many crops in humid regions to phosphate and potash fertilizers has caused citrus growers to become interested in the possible usefulness of these materials as fertilizers for citrus trees. There are places in California where vegetable crops, forage, and cereal crops have shown a response to phosphate; and on several soil types in this state, response of deciduous fruit trees and vines to potash has been noted. It therefore seemed important to determine the effect on citrus of supplementing nitrogenous fertilizer with phosphate and potash used separately and in combination. The total nitrogen was applied in all cases at the rate of 1 pound per tree annually. Relative orange yields where phosphate and potash were thus used are shown in table 4. Results of treatment C (urea and manure) are used as a standard for comparison for treatments in which urea was the only source of nitrogen. Results of treatment 12 are the standard for compari- son where mixed concentrated sources of nitrogen were used. There are some indications from these data, especially for the four-year period 1936-1939, that the use of phosphate without potash (treatment 8) had a depressing effect upon yields, as compared with the effects of using nitrogen alone (treatment 18). Phosphate and potash used together, as in treatments 9 and 10, apparently had no significant influence on yields, when compared with the effects of application of nitrogen alone (treat- ment 18). The same may be said of the use of potash without the phos- phate (treatment 11). The small differences observed (usually less than 10 per cent) may be within the realm of chance. Similar differences may be noted in comparisons of results of treat- ments 12 and 13 (table 4) . Treatment 13, which included the use of both phosphate and potash, produced an average of 6 per cent more fruit during 1928-1939 than treatment 12, in which mixed sources of nitro- gen, only, were used. This difference was not consistent, however, and in comparing annual yields, which are not presented here, no definite trend during the twelve years is in evidence. During the succeeding three-year period, 1940-1942, the average total yield per tree for treat- ments 12 and 13 has been 635 and 639 pounds, respectively. 14 The differ- ence in yields is obviously within the limits of casual error. 14 After 1939, the nitrogen applications in all the treatments shown in table 4 -were increased to 3 pounds per tree annually ; applications of potash were increased to 3 pounds of K 2 in treatments 10 and 13 only; phosphate was increased to 3 pounds of P 2 5 in treatment 13 only. 14 University of California — Experiment Station Analysis of the foliage of the trees shows that both phosphate and potash were taken up by the trees in greater amounts where these ele- ments were applied as fertilizers; hence there is no doubt that these added elements were available to the trees. TABLE 4 Effect on Orange Yields of Supplementing Nitrogenous Fertilizers with Phosphate and Potash, Used Singly or in Combination Treatment* 1928-1931 1932-1935 1936-1939 1928-1939 Nitrogen from urea C. Urea and manure! . C. Urea and manure! 18. Urea 8. Urea and phosphate 1 1 . Urea and potash 9. Urea, phosphate, and sulfate of potash . . 10. Urea, phosphate, and muriate of potash . Mean annual yield per tree, pounds 114 160 122 132 Relative yield (treatment C = 100) 100 104 94 104 104 98 100 100 89 98 87 79 96 100 99 96 95 88 100 96 87 99 100 94 Nitrogen from mixed sources! 12. Nitrogen. 12. Nitrogen 13. Nitrogen with phosphate and potash in mixed fertilizer (8-8-8) Mean annual yield per tree, pounds 117 149 115 127 Relative yield (treatment 12 = 100) * Annual applications were at a rate to supply 1 pound of nitrogen per tree, in all cases, and 1 pound each of phosphate (P2O5, from treble superphosphate) and potash (K2O) per tree, where these materials were used. Winter covercrops were grown in all cases. t Yields unadjusted. t Nitrate of soda, dried blood, and sulfate of ammonia, supplying equal amounts of nitrogen. Annual observations indicate that phosphate and potash did not cause an increase in growth of covercrops or weeds, in comparison with that resulting from the use of nitrogen alone. Effect of Various Kinds and Amounts of Bulky Organic Fertilizer Alone and in Combination with a Concentrated Source of Nitrogen. — Several bulky organic fertilizers besides dairy manure are readily ob- Bul. 673] Effect of Fertilizers on Orange Yields 15 tainable by citrus growers. Of these, bean straw, alfalfa hay, and cereal straw are possibly the best known and the most widely used, ordinarily in conjunction with some concentrated form of nitrogen. The relative yields of oranges produced by the use of these materials with urea and of dairy manure with and without urea and with and without covercrop are shown in table 5. TABLE 5 Belative Orange Yields Resulting from Use of Various Kinds and Amounts of Bulky Organic Fertilizers, Alone and in Combination with Concentrated Sources of Nitrogen Treatment* 18 (urea), 21 (nitrate of lime), and 27 (nitrate of soda) averaged ; with covercrop 18 (urea,) 21 (nitrate of lime), and 27 (nitrate of soda) averaged ; with covercrop C. Urea and manure,! with covercropt 39. Urea and cereal straw (same amount of organic matter as in C) 38. Urea and cereal straw (three times as much organic matter as in C) 37. Urea and bean straw, t with covercrop 36. Urea and alfalfa hay.t with covercrop 31. Dairy manure, with covercrop , 30. Dairy manure 32. Dairy manure applied in spring 1928-1931 1932-1935 1936-1939 1928-1939 Mean annual yield per tree, pounds Relative yield (average of treatments 18, 21, and 27 = 100) 100 96 95 79 97 96 93 100 89 100 100 107 102 97 96 91 87 101 101 91 100 89 69 92 101 85 80 100 102 96 100 95 84 97 84 * All materials were broadcast. Unless otherwise stated, the organic materials were applied in the fall, and the inorganic materials in the spring, in quantities to supply a total amount of 1 pound of nitrogen per tree annually. t Urea and bulky fertilizer supplying equal amounts of nitrogen. t Yields unadjusted. In table 5, average yields from treatments with the more commonly used concentrated fertilizers (treatments 18, 21, and 27, averaged) are shown for basic comparison with yields from treatments involving bulky fertilizers such as manure (treatment C) and other less commonly used materials (treatments 36, 37, and 39). From a comparison of results of treatments 18, 21, and 27, averaged, and treatment C, it is apparent that manure and urea, used together as equal sources of nitrogen, were as effective as the three concentrates. There is nothing to indicate, however, that nitrogen from manure is more valuable than nitrogen from the concentrates, provided a covercrop is grown to supply organic matter in all cases. 16 University of California — Experiment Station When supplemented with urea, cereal straw in moderate amounts (treatment 39, table 5) and dairy manure (treatment C) were equally effective as sources of organic matter and nitrogen. The ratio of organic carbon to nitrogen 15 in the respective fertilizing mixtures used, in these two treatments, was the same. When the amount of straw was increased threefold in treatment 38, but with the same total amount of nitrogen per tree as in treatment 39, the yields were consistently lower. Other trials than those shown in table 5 involved the application of as much organic matter as in treatment 38 (table 5) but with an in- creased amount of nitrogen from concentrated sources. The yields of such trials were satisfactory. This indicates that cereal straw, and prob- ably strawy manure, may be used as bulky fertilizers, but that care should be taken to reinforce these fertilizers with additional amounts of nitrogen from concentrated materials; this is especially important on soils low in nitrates. The total mixture should have a carbon : nitrogen ratio approaching more closely that of the soil on which it is to be used than was the case with treatment 38. The carbon : nitrogen ratio of the soil in these experimental plots is very close to 10 :1, which is typical of much of the soil in the citrus areas of California. It seems apparent from the results in table 5 that bean straw or alfalfa hay may be used as a substitute for manure. In fact, when reinforced with a concentrate so as to provide against too wide a carbon : nitrogen ratio, all the bulky materials used, including cereal straw, gave essen- tially the same results as manure. Where alfalfa was the source of nitro- gen and organic matter, the carbon : nitrogen ratio of the total fertilizer application averaged approximately 10 : 1 ; where bean straw was used, it was 18 : 1, over the twelve-year period. There is nothing in these ex- periments to indicate that the nitrogen from manure is preferable to the nitrogen from other bulky organic fertilizers. The use of dairy manure alone, as in treatments 31, 30, and 32, table 5, as a source of nitrogen was not so consistently productive of fruit as treatment C, where half the nitrogen came from manure and half from a concentrate. All nitrogen applications referred to in table 5 were in the relatively small amount of but 1 pound of nitrogen per tree annually. At this rate of application, the use of manure with a winter covercrop (treatment 31) resulted in decreased yields, particularly in the 1936- 1939 period. The decrease in yield in this treatment has been associated with a reduction in available nitrate during the blooming and fruit- 15 The carbon: nitrogen ratio for treatments C and 39, for the twelve-year period, averaged approximately 11:1; that for treatment 38 was nearer 32:1. The carbon content of fertilizer materials can be estimated from the analysis of organic matter by multiplying by 0.58. Bul. 673] Effect of Fertilizers on Orange Yields 17 setting period. When the same quantity of manure was applied in the spring, without a winter covercrop (treatment 32), the yields were like- wise decreased, and for the same reason. The use of manure alone, ap- plied broadcast in the fall without a winter covercrop, was satisfactory. In other cases (not presented in table 5), where 3 pounds of nitrogen was applied per tree annually, it made no difference in yield whether all the nitrogen came from manure, whether half came from manure and half from a concentrate, or whether all the nitrogen was derived from manure and covercrops were grown. The growth and production of the trees and the growth of covercrops were all greater where the larger amounts of nitrogen were applied, and this increased response to fer- tilizer was correlated with increased amounts of available nitrogen in the soil. It seems clear that the value of manure, when used as the sole source of nitrogen, is dependent upon the nitrogen supply available in the soil upon which it is used. If available nitrogen is high, then manure alone can safely be used; if low, manure applied alone may have a depressing effect on yields, as compared with the effect of manure plus a nitrogenous concentrate. The experiments reported in table 5 indicate that the practicability of using manure as a partial or sole source of nitrogen will depend upon the relative cost of the manure and other sources of nitrogen and upon the feasibility of growing covercrops. The application of manure at a rate to supply 3 pounds of nitrogen per tree annually was probably uneconomical. A similar amount of nitro- gen, derived half from manure and half from the concentrate urea, gave essentially the same yields at a considerably lower cost. Apparently, a moderate application of manure is as effective as heavier applications, provided the total nitrogen supply is maintained by the addition of a concentrate. Urea happens to be the concentrate used in this comparison. Equally good results were obtained by using nitrate of soda with ma- nure, as shown by a comparison of treatments C and 29 in table 3. We are of the opinion that, if used with manure, any of the other concen- trates listed in table 1 would have proved similarly effective. Results of Different Methods of Using Dairy Manure. — In some of the citrus-growing districts of California, the price of dairy manure is so low that orchardists largely depend upon it as a source of nitrogen. The organic-matter content of manure may also often be of vital importance, especially in old, shady orchards, where covercrops grow poorly and the use of manure is the most practical means of supplying organic matter. The best season for applying manure and the best method of applica- tion are therefore questions frequently raised. The practical value of a 18 University of California — Experiment Station covercrop, if notable amounts of organic matter are supplied by manure, is also a consideration. Results of using manure in three different ways, as the sole fertilizer material, are shown in table 6, in which treatment 25, manure and winter covercrop, is considered as the basis for com- parison. It is plain that manure applied in trenches (treatment 33) is inferior in effectiveness to manure applied broadcast and disked into the soil (treatment 19). In treatments 25 and 19 (table 6), the manure was applied broadcast and disked into the soil each fall, usually in September, to a depth of 4 TABLE 6 Relative Yields of Oranges Resulting from the Use of Dairy Manure Applied by Different Methods Treatment 25. Manure disked into soil*t 25. Manure disked into soil*t 19. Manure disked into soil*t 33. Manure applied in trenchest§ 1928-1931 1932-1935 1936-1939 1928-1939 Mean annual yield per tree, pounds 124 IK) ' 163 155 Relative yield (treatment 25 = 100") * In amounts to supply 1 pound of nitrogen per tree annually, 1927-1929; 3 pounds, 1930-1939. t Used with winter covercrop. J Used without winter covercrop. § In amounts to supply 1 pound of nitrogen per tree annually, 1927-1931; 3 pounds, 1932-1939. to 6 inches. In treatment 33, the trenches for the application of manure were plowed 12 inches deep on opposite sides of each tree, in line with the irrigation furrows. The first j^ear (1927), the trenches were at the outer drip of the tree branches. Each successive year, they were moved 1 foot farther away from the tree, until it finally became necessary to go back to approximately the original location and again move progres- sively farther away from the tree. It is clear from the data in table 6 that this method of using manure was not so effective as that of applying it broadcast and disking it under. The trenching operations cut a notable number of roots, particularly when the first trenches were plowed, and it is believed that this injury to the trees is responsible for the relatively poor production when manure was applied in this manner. The Effect of Using Agricultural Minerals™ in Conjunction with Ade- quate Fertilizer. — The use of supplementary applications of agricul- Commonly spoken of, heretofore, as "soil amendments." Bul. 673] Effect of Fertilizers on Orange Yields 19 tural minerals when adequate fertilizer materials have already been applied, is not a common practice in California. Orchardists are occa- sionally induced to apply ground limestone, gypsum, or sulfur, how- ever, in the anticipation that such materials may be found to be beneficial. If applied in large enough quantities, some of these minerals may affect the acidity of the soil throughout part of the citrus-tree root system, for at least a short period. TABLE 7 Belative Orange Yields Resulting from Use of Urea and Manure with Bhosphate and Botash and with Various Agricultural Minerals Treatment* 1928-1931 1932-1935 1 1936-1939 1928-1939 Mean annual yield per tree, pounds 14. Urea, manure, phosphate, and potashf 121 165 117 134 Relative yield (treatment 14 = 100) 14. Urea, manure, phosphate, and potashf 100 93 93 93 100 97 101 93 100 107 104 114 100 34. Urea, manure, and gypsumj 35. Urea, manure, and limestone§ 99 100 7. Urea, manure, phosphate, potash,! and sulfurl 99 * Winter covercrops used in all cases. Urea and manure, each furnishing Yi pound nitrogen per tree annually 1927-1931 and \Yi pounds per tree annually 1932-1939, were used in all treatments. t Phosphate (P2O5) and potash (K2O) applied annually at the rate of 1 pound per tree 1927-1931; 3 pounds, 1932-1939. 1 Gypsum applied at the rate of 22 pounds per tree annually (2,000 pounds per acre). § Ground limestone applied at an average rate of 22 pounds per tree annually. \ Sulfur applied in small amounts annually, at a rate to supply a total of 2 tons per acre for the period 1927-1940. The relative yields of oranges from plots which were given consistent and relatively liberal amounts of agricultural minerals, in conjunction with liberal applications of fertilizer, are presented in table 7. It is clear from this table that limestone, gypsum, or sulfur did not affect the yields to any consistent, significant degree. Nor is there any evidence from the data of table 7 that phosphate and potash were beneficial when used with a urea-manure fertilizer program (the mean annual yield with urea and manure alone, with covercrop, is given in table 4, p. 14) . Annual observations also showed that these materials failed to increase the amount of covercrop grown in the plots. In fact, the first application of sulfur, which was at a rate of 20 pounds per tree, prevented the growth of covercrop the following winter and greatly reduced covercrop growth the following year. This effect was temporary, however. Certain agricultural minerals appreciably affected the pH of the soil throughout a large portion of the root zone of orange trees in these 20 University of California — Experiment Station experiments. Soil samples were taken in the plots periodically, from 1929 to 1940. In table 8, determinations made during May, 1940, are presented to show the differences in pH of the soil at that time. The mean annual yields of oranges for the twelve-year period are also shown in this table. The appearance of the trees in the several plots was essentially the same throughout the experimental period. It is clear from the data in table 8 that the addition of ground lime- stone (treatment 35) made the soil in the top 36 inches notably more alkaline than did the other fertilizer treatments. Soil in the top 36 inches TABLE 8 Effect of Certain Agricultural Minerals on pH of Soil (Determined at End of Thirteen- Year Period, 1928-1940) and on Average Orange Yields, 1928-1939 Treatment* The pH at depths statedf Average annual yield of 0-6 inches 6-12 inches 12-24 inches 24-36 inches fruit per tree, pounds 14. Urea, manure, phosphate, and potash . 35 Urea, manure, and limestonej 6.89 7.89 4.42 7.70 8.26 5.84 8.10 8.31 6.60 8.19 8.47 7.38 134 133 7. Urea, manure, phosphate, potash, and 134 * Winter covercrops used in all cases. f Sampling and determinations made in May, 1940. + Ground limestone applied at the rate of 1 ton per acre annually. § Sulfur applied in small amounts annually, at a rate to supply a total of 2 tons per acre for the period 1927-1940. of the sulfur-treated plots (treatment 7) was, on the other hand, more acid than that given the other fertilizer treatments. In the area where citrus-tree roots are most concentrated (the top 24 inches), the soil was decidedly acid where sulfur was applied. It seems evident from this table that, when other conditions are favorable, orange trees may be adapted to a relatively wide range of soil pH without showing any difference in appearance or productivity. This has been generally believed to be the case by many observers familiar with the growth of citrus trees in Cali- fornia. The Effect of Applying Various Amounts of Nitrogen. — In the ma- jority of the experiments heretofore mentioned, the nitrogen applica- tions were held to a relatively low level. A series of experiments was therefore undertaken to determine the effects of various amounts of nitrogen. Applications of nitrogen were made at the rates of %, 1, 2, and 3 pounds per tree annually. The results indicate that nitrogen is clearly deficient in this soil. Marked differences in yield responses Bul. 673] Effect of Fertilizers on Orange Yields 21 were also obtained. Comparative yields are shown in table 9, the produc- tion of trees having no fertilization except a covercrop (treatment 6) being- taken as the basis for the calculation of relative yields. An unfer- tilized tree and a tree fertilized with liberal quantities of nitrogen are illustrated in figures 3 and 4, respectively. The contrast in yields caused by different amounts of fertilizer be- came clearly apparent even during the first four-year period shown in table 9 ; this effect increased with time. The remarkable increase in pro- TABLE 9 Eelative Orange Yields Kesulting from Use of Different Amounts of Nitrogen* Treatment! 1928-1931 1932-1935 1936-1939 1928-1939 Mean annual yield per tree, pounds 6. Covercrop only J 82 68 20 57 Relative yield (treatment 6 = 100) 6. Covercrop only! 100 136 139 150 159 100 202 235 258 292 100 412 610 634 723 100 40. Nitrogen, H pound per tree annually C. Nitrogen, 1 pound per tree annually! 195 233 41. Nitrogen, 2 pounds per tree annually 42. Nitrogen, 3 pounds per tree annually 250 278 * Nitrogen from urea (one half) and from manure (one half). t Winter covercrops used on all plots. t Yields unadjusted. duction brought about by the application of only % pound of nitrogen per tree annually (treatment 40) is one of the outstanding features of this table, although the yields thus produced would not be considered commercially satisfactory. The differences between the yields resulting from the use of 2, and also 3, pounds of nitrogen, and those from 1 pound, are proportionately somewhat less than the response to the use of only % pound of nitrogen. This illustrates a point which has frequently been shown by other ex- periments, especially with annual crops : that the increase in crop pro- duction due to fertilization is not in direct proportion to the amount of fertilizer used. Under these circumstances, it may be more profitable to use a moderate amount of fertilizer than to use a large amount. It is apparent from these and other comparable data that as the rate of fertilization increases, a point is finally reached where the small in- crease in yield may not pay for the extra fertilizer. 22 University of California — Experiment Station The actual effect on yield of variation in the amount of fertilizer applied will doubtless vary in different orchards. The results will de- pend upon the age of the trees, upon the soil, upon previous fertilizer Fig. 3. — Orange tree not fertilized. practice in the orchard, and probably upon other factors. The data pre- sented in table 9 are included primarily to illustrate a principle com- monly observed in many fertilizer experiments. The data should not be interpreted as applying literally and in the same degree to all orchards. Casual observations indicate that the annual application of 2 to 3 pounds of nitrogen per tree in mature orchards is a common practice Bul. 673] Effect of Fertilizers on Orange Yields 23 among commercial producers of citrus. The profitableness of such appli- cations will vary from time to time and will depend upon the cost of fer- tilizer and the price of fruit. If fertilizer costs are relatively high and Fig. 4. — Orange tree fertilized annually with 3 pounds of nitrogen. Compare with tree shown in figure 3. fruit is relatively cheap, the use of only moderate amounts of fertilizer may be the most profitable. The effects of the application of various amounts of nitrogen on cover- crop growth are also of interest. Although the amount of growth was good in all plots at the beginning of the experiment, the plots receiving 1 pound of nitrogen per tree annually produced very poor growth in the 24 University of California — Experiment Station Fig. 5. — Growth of winter covercrop in plot at Riverside, fertilized with 1 pound of nitrogen per tree annually. Compare with figure 6. Fig. 6. — Growth of winter covercrop in plot at Eiverside, fertilized with 3 pounds of nitrogen per tree annually. later years of the twelve-year period. The growth in plots receiving larger amounts of nitrogen was good each year (see figures 5 and 6). It was therefore evident that the annual application of only 1 pound of nitrogen per tree was not sufficient to maintain indefinitely the nitro- gen supply in this orchard soil. Bul. 673] Effect of Fertilizers on Orange Yields 25 FERTILIZER EXPERIMENT AT ARLINGTON, RIVERSIDE COUNTY An experiment in an orchard at Arlington, Riverside County, Cali- fornia, was started in 1930. Four years' yield records were obtained dur- ing the period 1931-1934, while the plots were receiving differential fertilizer treatment. The orchard is planted on a deep Hanford sandy TABLE 10 Yields of Oranges Showing Effect of Various Fertilizer Treatments, Arlington Experiment Treatment 101. Manure 102. Manure and am- monium sulfate! . . 103. Manure and mixed fertilizer! 104. Cereal straw (2 tons per acre) 105. Manure 106. Ammonium sulfate 107. Mixed fertilizer. . . . 108. Manure and mixed fertilizer! plus 10 pounds sulfur per tree annually 109. Ammonium phos- phate Amount of nitrogen, phosphate, and pot- ash applied annually per tree, pounds Average tree, annual yield per in field boxes N P2O5* K 2 0* 1931 1932 1933 1934 1931- 1934 4 2.0 8.0 5.11 6.08 7.38 5.34 5.98 4.0 10 4.0 4.74 5.68 7.05 5.09 5.64 4 3.0 5.0 4.47 5.46 7.03 4.83 5.45 3 0.1 0.9 5.11 4.94 6.95 3.82 5.21 2.0 1.0 4.0 4.87 5.12 6.67 4.96 5.40 4 0.0 5.66 5.54 7.57 5.32 6.02 4 4 2.0 4.89 4.46 7.12 4.57 5.26 4 3.0 5 5.26 5.82 7.43 5.68 6.05 4.0 5.0 5.02 5.61 6.92 4.19 5.44 Relative yield, 1931-1934 (treat- ment 101 = 100) 100 94 91 87 90 101 101 91 * Manure and straw contain appreciable amounts of phosphate (P2O6) and potash (K2O) ; amounts of these constituents reported here are estimates based on analyses made at the Citrus Experiment Station, Riverside. t Manure and the concentrate supplying equal amounts of nitrogen. loam and, as far as natural and cultural conditions are concerned, is favorably located for Washington Navel oranges. The foliage of the trees in this orchard showed moderately severe symptoms of zinc defi- ciency during the experiment. At its conclusion, a general application of a spray containing zinc, improved the appearance of the trees. Complete details of the plan of the experiment will not be presented here. The treatments were applied in plots of 8 to 12 trees each, separated by guard rows. Most of the treatments were replicated four times, and annual tree growth and yields were recorded. Descriptions of treatments applied and a summary of results of the Arlington experiments are presented in table 10. It should be borne in 26 University of California — Experiment Station mind that this orchard had been maintained under very good cultural conditions and had been liberally fertilized with organic matter and nitrogen for some time before the trials herein described were begun. The results of nine different methods of fertilization are shown. All the treatments described in table 10, except 104 and 105, included very liberal amounts of nitrogen. Treatments 101, 103, and 108 included large applications of potash, and treatments 107 and 109 included rela- tively large applications of phosphate, in comparison with those of nitro- gen. If an exception is made of treatment 104, the table shows that, during the period of the experiment, there were eight different ways of fertilizing this orchard with approximately equal effectiveness. The low nitrogen applications in treatment 104 resulted in a reduction in yield in the fourth year of the trial. Such a short-term trial, carried out in a soil which had been heavily fertilized, would not be expected to show any effect of applying organic matter in amounts varying from none, in treatment 106, to relatively large amounts, in treatment 101. The results of this trial, summarized in the last column of table 10, show the relative yields of the various plots when the manure-fertilized plot is given a basic value of 100. The indications are that, for at least a short period, the most economical source of nitrogen is satisfactory as a fertilizer, and that the expense of applying phosphate, potash, or sulfur is not warranted in this orchard. The Hanford soil on which this experiment was conducted is one of the four soil types most extensively used for growing oranges in California. KERN COUNTY AND TULARE COUNTY FERTILIZER EXPERIMENTS Fertilizer experiments were begun in two commercial orchards in Kern and Tulare counties in 1929 to determine the effects of various fertilizer materials and methods of culture on yield, quality, and date of maturity of Washington Navel oranges. The orchards had previously been fertilized with inorganic nitrogen, manure, and covercrops, which is a common fertilizer practice in these localities. This fertilization was therefore continued in the two orchards as a basic treatment, and cer- tain experimental modifications were made on the different plots within each orchard. Although the duration of these two trials was short (three years and five years, respectively), the quantities of fertilizers applied were larger than those commonly used, and the results are probably in- dicative of what might have been expected had smaller applications been made over a longer period of time. Bul. 673] Effect of Fertilizers on Orange Yields 27 In each experiment the basic treatment given to all experimental plots consisted of a light application of medium-grade dairy manure at the rate of 5 tons per acre annually, each plot receiving, also, 1% pounds of nitrogen per tree annually from inorganic fertilizers. Winter covercrops TABLE 11 Annual Fertilizer, Applications,* Kern County and Tulare County Experiments (Treatments listed in this table are the basis for tables 12, 13, 14, and 15) Treatment 201 202 203 204 205 206 207 Materials applied and average number of pounds per tree for each annual application! Manure, 110 pounds; ammonium sulfate, 7.5 pounds Manure, 110 pounds; ammonium sulfate, 7.5 pounds; plus treble superphosphate, 15.5 pounds Manure, 110 pounds; ammonium sulfate, 7.5 pounds; treble superphosphate, 15.5 pounds ; plus sulfate of potash, 15.2 pounds Manure, 110 pounds; ammonium sulfate, 7.5 pounds; treble superphosphate, 15.5 pounds; sulfate of potash, 15.2 pounds; plus sulfur, 12 pounds Manure, 110 pounds; mixed fertilizer (8-8-8), 18.75 pounds Same as treatment 203, plus covercrop of Sudan grass grown in the fall Manure, 110 pounds; nitrate of lime (applied in the fall), 10 pounds; treble superphos- phate, 15.5 pounds; sulfate of potash 15.2 pounds Total amount of nitrogen, phosphate, potash, and sulfur applied annually per tree, from all sources,! pounds N P2O5 0.3 7.8 1.8 7.8 7.8 K2O 1.1 1.1 8.6 8.6 2.6 8.6 8.6 2.3 2.3 4.9 16.9 1.5 4.9 4.9 * Used with winter covercrops in all cases. t Manure was applied in the fall. Unless otherwise stated, applications of concentrates were made in the spring. t Manure contains appreciable amounts of phosphate (P2O:,) and potash (K2O) and some combined sulfur (S). Ammonium sulfate contains large quantities of sulfur. Amounts of such constituents reported here are estimates based on analyses made at the Citrus Experiment Station, Riverside. of mustard were grown on all plots. The individual fertilizer treatments are listed in table 11. In treatment 207, the inorganic nitrogen fertilizer was applied in the fall of the year in order to supply liberal amounts of available nitrogen during fruit maturation. In treatment 206, nitrogen availability was de- creased during the period of fruit maturity by the growing of Sudan grass during the late summer and early fall, in addition to the winter covercrop of mustard. To the basic treatment, mixed fertilizer contain- ing nitrogen, phosphate, and potash was added for treatment 205 ; for the other treatments, with the exception of treatment 201, large quanti- 28 University of California — Experiment Station ties of "simples" containing either phosphate, or phosphate and potash, were used. Treatment 204 included liberal use of sulfur. The fertilizers were applied annually — the manure in the fall or winter, and the inor- ganic nitrogenous fertilizers in the winter or spring, except in the case of treatment 207. Other materials were applied in the summer. All sum- mer and fall applications were made in the bottom of the irrigation furrows. Kern County Experiment. — The orchard of mature Washington Na- vel orange trees used in this experiment is located on San Joaquin sandy TABLE 12 Orange Yields Showing Effect of Various Fertilizer Treatments, Kern County Experiment Treatment* 1929 1930 1931 1929-1931 Average annual yield per tree, field boxes 201 1.54 3.58 2.95 2.69 Relative yield (treatment 201 = 100) 201 202 203 204 205 206 207 100 78 89 81 87 85 90 100 97 102 95 104 106 107 100 94 99 97 102 98 103 100 92 98 93 100 90 102 * For explanation of treatment, see table 11 (p. 27). loam soil at Jasmine. Previous to the start of the experiment, the orchard had usually received annual applications of dairy manure and either ammonium sulfate or nitrate of lime. The trees were large and, at the start of the experiment, showed moderate symptoms of mottle-leaf (zinc deficiency), a condition which improved somewhat during the course of the experiment. The experimental plots, consisting of 15 trees each, were systematically replicated four times. The plots were in single rows sep- arated by guard rows. There were thus 60 trees for each treatment. Three annual applications of the experimental materials were made (1929-1931) . In view of the large amounts of phosphate and potash ap- plied to many of the plots, the method of application, and the nature of the particular soil, it appears probable that these materials were made available to the trees. In the case of the sulfur treatment, a total of 35 pounds of sulfur was applied per tree during the three years. This quan- Bul. 673] Effect of Fertilizers on Orange Yields 29 tity of sulfur acidified the top foot of soil and was sufficient to prevent temporarily the growth of winter covercrops of mustard. None of the other differential treatments affected the growth of covercrops. Observations were made on the crops produced over the three-year period, 1929-1931. The relative yields of fruit produced in the plots during this period are given in table 12. It may be seen that none of the additions to the basic treatment of manure and inorganic nitrogenous fertilizers greatly influenced the yields of trees in the different treat- TABLE 13 Orange Yields Showing Effect of Various Fertilizer Treatments, Tulare County Experiment Treatment* 1925-19281 1929 1930 1931 1932 1933 1929-1933 Average annual yield per tree, field boxes 201 1.76 1.45 1.94 1.94 2.04 1.73 1.82 Relative yield (treatment 201 = 100) 201 202 .. . 203 204 205 .. . 206. 207 100 101 94 95 98 101 97 100 81 105 83 100 86 91 100 128 93 105 112 111 116 100 106 98 102 100 110 98 100 93 77 86 87 95 87 100 94 79 88 83 90 92 100 102 89 93 97 99 97 * For explanation of treatment, see table 11 (p. 27). t Before the beginning of the experiment ; ; data shown here for comparison. ments. Inspection of the individual plot yields indicates that the aver- age yields of the treatments are not significantly different. Observations on the maturity of fruit were also made each year in the early part of the harvest season. The observations consisted in compari- son of fruit color and tests of total soluble solids and total acidity. These were made upon repeated randomized samples. These measures of matur- ity were not affected by the variations in fertilizer practice. The grow- ing of a summer and fall crop of Sudan grass in the orchard (treatment 206) did, however, result in an increased and earlier development of orange color of the rind and in higher soluble solids and acidity in the juice. These effects were greatest in the years when the Sudan-grass crop was largest. The rind of the fruit on the Sudan-grass plots was also somewhat more coarse and "pebbly" than that of fruit in the other plots. The ratio of solids to acid was not affected. 30 University of California — Experiment Station Tulare County Experiment. — This orchard of Washington Navel or- ange trees is located near Porterville and is on Porterville clay loam adobe soil. At the beginning of the trial, the trees were small, but their foliage was of good color, and typical symptoms of mottle-leaf were of minor importance. Before the beginning of these experiments, the or- chard had been fertilized with liberal quantities of manure and some inorganic nitrogen fertilizers. The experimental treatments, listed in table 11, were initiated in 1929 and continued for five years. The treatments were applied in plots of 7 TABLE 14 Percentages* of Fancy Grades of Fruit Produced on Plots Receiving Differential Fertilizer Treatments, 1929-1933, Tulare County Experiment Treatment! 1929 1930 1931 1932 1933 1929-1933 per cent per cent per cent per cent per cent per cent 201 71 94 60 70 78 75 202 70 94 61 72 72 74 203 74 96 68 67 67 74 204 71 95 62 68 59 71 205 75 95 65 68 63 73 206 74 95 63 69 75 75 207 76 94 55 69 72 73 * Volume basis. t For explanation of treatment, see table 11 (p. 27). to 11 trees each, and were replicated in 4 blocks of the orchard. The unit plots were separated by guard rows. Individual tree-yield records which had been taken for three years prior to the beginning of the experiment indicated that the inherent average productivity of the plots allocated to each experimental treatment was very nearly equal. Chemical analyses of the fruits harvested at the conclusion of the five-year period indicated that the phosphate and potash fertilizers applied to the plots in the ex- periment had become available to the trees. The fruits from such plots were higher in these constituents than fruits from trees not given these materials. The total amount of sulfur applied during the five-year period was 60 pounds per tree. This treatment failed to have much effect on the pH of this particular soil. The average relative yields of the trees fertilized in the various ways over the five-year period are given in table 13. The average yields for the plots before the beginning of the experiment are also given. It is obvious that none of the treatments significantly increased the yield over that of the nitrogen-manure treatment (201). . Bul. 673] Effect of Fertilizers on Orange Yields 31 The fruit from each plot was graded separately in the packing-house each year. The percentages of Fancy grades of fruit are shown in table 14. Some years were more favorable than others for the production of good grades of fruit, and there are some small differences in the fruit from plots receiving different fertilizer treatments. These are not con- sistent, however, and the variations are no larger than might be ex- pected from an equal number of lots of fruit from a uniformly fertilized orchard. Apparently, the grade of fruit was not affected by the differen- tial fertilizers. TABLE 15 Percentages* of Fruits of Desirable Commercial Sizes (220 Per Packed Box and Larger) from Plots Beceiving Differential Fertilizer Treatments, 1929-1933, Tulare County Experiment Treatment t 1929 1930 1931 1932 1933 1929-1933 per cent per cent per cent per cent per cent per cent 201 69 44 44 47 56 52 202 ...... 70 38 40 34 50 46 203 66 38 46 44 56 50 204 . 67 41 36 37 48 46 205 66 50 48 16 58 48 206 65 36 48 32 53 47 207 66 37 36 35 51 45 * Volume basis. t For explanation of treatment, see table 11 (p. 27). The size of oranges is also an important factor in relation to their salability. Table 15 shows the percentages of fruits of the most desirable sizes produced annually by the trees given the different fertilizer treat- ments. The conclusion is similar to that made regarding grade of fruit ; that is, the differential fertilizer treatments were not effective in modi- fying the size of the fruit. It is apparent that other factors, including those of a seasonal nature, were more important than fertilizers in influ- encing fruit size. Observations of the fruit as it approached maturity were also made in the orchard and on large randomized samples picked for the purpose. These indicated no differences in color of fruits from the various plots, except in the case of fruit from the Sudan-grass plots (treatment 206). Such fruits developed orange color more rapidly than fruits from any other plots. This was especially true in years when the Sudan-grass crop was large. The soluble solids and acid content of the fruit juice were not consistently changed by growing Sudan grass. Additional experiments, involving the application of various mate- rials in the irrigation water for a two-j^ear period, were carried out in 32 University of California — Experiment Station this orchard in triplicated plots. Three or four applications of these materials were made each year, the rate of application being varied in such a manner as to supply the desired total amount of fertilizer ingredi- ent. The materials applied annually in addition to the basic treatment were: diammonium phosphate (3.2 pounds N and 7.5 pounds P 2 3 per tree) ; liquid phosphoric acid (7.5 pounds P,0 3 per tree) ; ammonium sulfate (3.2 pounds N per tree) ; and nitrate of lime (3.2 pounds N per tree) . None of these applications appeared to affect yield, color, grade or size of fruit, nor did they affect the growth of winter covercrops. Discussion of Results of Kern County and Tulare County Experi- ments. — The results of these two experiments, on widely different soil types, appear to be similar. Liberal applications of phosphate, or of phosphate and potash, applied as "simples," or in combination, as in mixed fertilizer, failed to affect the yields or commercial quality of the crop in orchards receiving a basic treatment of manure and inorganic nitrogenous fertilizer. The use of large quantities of sulfur likewise had no effect on yields, quality of fruit, or appearance of the trees. Nor did the season of application of the inorganic nitrogenous fertilizer have any apparent effect. It was noticeable, however, that the growing of a Sudan- grass covercrop in late summer and fall increased the intensity of orange coloring of the fruit before full maturity, although the solids : acid ratio was not changed. The surface of the fruit from the Sudan-grass plots was frequently more coarse and "pebbly" than that of fruit from other plots. This is the opposite of the effect obtained by Martin" with grapefruit on the Yuma Mesa in Arizona. The reasons for the effect of the Sudan grass in these trials are not known. It is possible that a drought effect may have been involved. An effort was made to maintain satisfactory soil moisture in all plots, but the growing of such a crop greatly increases the difficulty of soil-moisture control. It seems improbable that the effect of the Sudan grass on the fruit was due to any decrease in the amount of nitrate nitro- gen in the soil. Additional experiments involving the injection into the soil of large amounts of organic matter in solution, in order to cause the reduction of nitrates by another process, failed to affect the fruit. Con- versely, applications of nitrate of lime in the fall, failed to affect fruit quality. Further evidence that the application of phosphate and potash had no effect in these experiments was obtained from the treatment of cer- tain trees in both orchards with two applications of treble superphos- 17 Martin, William E. Some effects of cultural practices upon tree composition, yield and quality of Marsh grapefruit in Arizona. Amer. Soc. Hort. Sci. Proc. 37:68-75. 1939. Bul. 673] Effect of Fertilizers on Orange Yields 33 phate, in amounts up to 45 pounds annually per tree, and with two ap- plications of sulfate of potash, in amounts up to 30 pounds annually per tree, the applications being made just below the bottom of the irrigation furrows. These, also, failed to affect fruit color or the solids : acid ratio of the juice and caused no visible effect on the trees. Covercrops in both orchards failed to respond to the application of either phosphate or potash. The use of large amounts of sulfur temporarily reduced cover- crop growth in the Kern County experiment on sandy loam soil, but not in the Tulare County experiment on adobe soil. It appears from these Kern County and Tulare County trials with Washington Navel oranges grown on San Joaquin sandy loam and Por- terville clay loam adobe soils, respectively, that a satisfactory fertilizer practice consists in the application of manure and inorganic nitrogen, without the use of additional amounts of phosphate, potash, or sulfur. ORANGE COUNTY EXPERIMENTS Data from two experiments carried out in Orange County during the years 1921 to 1932, inclusive, have been given to the authors by the Uni- versity of California Agricultural Extension Service. 18 These trials were made in two orchards of Valencia oranges, one at Fullerton and one at Orange. Before the experiments were started, both orchards had been fertilized with various quantities of manure and inorganic nitrogenous fertilizers. Except for minor variations in treatment (see "Experiment at Fuller- ton," p. 34) , the trials in the two orchards were almost identical. Outlines of the treatments are presented in table 16. Since nitrogen applications were commonly regarded as commercially desirable, all the treatments involved the use of fertilizers containing this element. Regardless of its source, nitrogen was applied at a rate of 2 pounds per tree annually in all treatments. In treatment 302 (table 16) all the nitrogen was supplied by manure ; while in treatment 301, it was derived from concentrated materials, half of the nitrogen being sup- plied by ammonium sulfate and half by dried blood. To the manure treat- ment, superphosphate was added for treatment 303 ; while for treatment 304, both superphosphate and sulfate of potash were added. In treatment 305, the nitrogen was derived from the concentrated sources ammonium sulfate (one fourth) and blood (one fourth) and from manure (one half) . (This last-mentioned treatment was essentially the fertilizer pro- gram used uniformly on all plots of each orchard before the start of the 1S The fertilizer applications and the observations in these experiments were made by H. E. Wahlberg and W. K. Schoonover. 34 University of California — Experiment Station experiments.) Treatment 306 was like treatment 305, except that super- phosphate and sulfate of potash were added. Where manure was applied, it was spread broadcast in the fall of each year. Concentrated nitrogenous materials were applied each spring. The annual applications of superphosphate and sulfate of potash were di- vided, half being applied in the spring and the other half in the fall, both applications being made in the irrigation furrows. Winter cover- crops were grown each year in both orchards. TABLE 16 Annual Fertilizer Applications,* Orange County Experiments (Treatments listed in this table are the basis for tables 17 and 18) Treatment Materials applied and average number of pounds per tree for each annual application t Total amount of nitrogen, phosphate, and potash applied annually per tree from all sources, t pounds N P2O5 K2O 301 302 303 304 305 306 Ammonium sulfate, 5 pounds; dried blood, 8 pounds . . Manure to supply 2 pounds nitrogen Manure to supply 2 pounds nitrogen; superphosphate, 20 pounds Manure to supply 2 pounds nitrogen; superphosphate, 20 pounds; sulfate of potash, 4 pounds Ammonium sulfate, 2.5 pounds; dried blood, 4 pounds; manure to supply 1 pound nitrogen Ammonium sulfate, 2.5 pounds; dried blood, 4 pounds; manure to supply 1 pound nitrogen; superphos- phate, 20 pounds; sulfate of potash, 4 pounds 2 2 2 2 2 2 10 5 5.0 5 4 5 4 4 6 2 4 * Used with winter covercrops in all cases. t Manure was applied in the fall; concentrated nitrogen sources were applied in the spring. The appli- cations of superphosphate and of sulfate of potash were divided, half of each annual application being made in the spring and half in the fall. t Manure contains appreciable amounts of phosphate (P2O5) and potash (K;0); amounts of such con- stituents reported here are estimates based on anal yses made at the Citrus Experiment Station, Riverside. Experiment at Fullerton. — This experiment was conducted in a ma- ture Valencia-orange orchard on Yolo loam soil. Three plots were used for each treatment. Each plot consisted of a single row of 12 to 16 trees and was separated from the next plot by a guard row. The experiment was conducted from 1921 to 1932, a period of twelve years. During this period the fertilizer applications were made in accord- ance with the outline presented in table 16, with the following exceptions. In 1921 and 1922, the nitrogen applied on the plots of treatments 303 and 304 was derived equally from manure and concentrated sources (ammonium sulfate and blood). On plots of treatment 306, applications of agricultural lime, at a rate of 150 pounds per tree, were made in 1923 and 1927, in addition to the treatments indicated. It is improbable that Bul. 673] Effect of Fertilizers on Orange Yields 35 these variations in treatments complicated the interpretation of the experiment. At the start of the experiment the trees were pale green in color, a condition suggestive of nitrogen deficiency. This condition gradually improved in the course of the experiment. The relative yields in the several fertilizer treatments, with the excep- tion of the yields for 1929, which were lost, are given in table 17. This table shows that the trees fertilized by the six methods produced, on the TABLE 17 Effect of Differential Fertilizer Treatments on Belative Yields of Valencia Oranges, Fullerton Experiment Treatment* 1921-1924 1925-1928 1930-1932 1921-19321 Mean annual yield per tree, field boxes 301 4.06 4.30 5.39 4.51 Relative yield (treatment 301 = 100) 301 302 303 100 105 93 90 87 97 100 100 100 102 95 100 100 94 97 96 93 96 100 100 97 304 305 306 96 92 98 * For explanation of treatment, see table 16 (p. 34). t 1929 omitted. average, about equal quantities of fruit. Covercrop growth, also, was unaffected by the differential fertilizer treatments. Observations made on the crop harvested in 1926 indicated no sig- nificant differences in quality of the juice of fruits from the different fertilizer plofs. Special fruit-storage and inoculation tests also failed to show any effects of the fertilizers on the resistance of fruits to decay organisms. Experiment at Orange. — The other experiment in a Valencia-orange orchard in Orange County was made near the town of Orange, on Han- ford sandy loam soil. The trees were approximately fifteen years old in 1923, when the experiment was started. Two plots, each containing 7 to 10 trees, were assigned to each treatment. The plots were separated by guard rows. The experiment was carried out until 1929, in accordance with the plan given in table 16, but the crop of 1925 was partially lost as a result of frost injury. Yield records were therefore obtained for but five of the six years of the experiment. 36 University of California — Experiment Station The yields of the trees receiving various fertilizer treatments are pre- sented in table 18. All the treatments resulted in about the same average tree yields, and no differential effect of the fertilizers was observed on covercrop growth. Discussion of Results of Orange County Experiments. — Although the types of soil differed very greatly in these two Orange County orchards, the results of the experiments are very similar. It is evident that in each TABLE 18 Effect of Various Fertilizer Treatments on Relative Yields of Valencia Oranges in Experiment at Orange Treatment* 1924 1926 1927 1928 1929 1924-1929t Mean annual yield per tree, field boxes 301 5.70 7.14 6.02 2.07 6.51 5.49 Relative yield (treatment 301 = 100) 301 100 100 100 100 100 100 302 94 91 85 98 96 92 303 94 94 96 115 97 97 304 86 90 89 89 91 97 134 109 93 103 93 305 96 306 91 81 92 111 98 92 * For explanation of treatment, see table 16 (p. 34). t 1925 omitted. orchard the application of equal amounts of nitrogen (2 pounds per tree annually) gave essentially the same results whether supplied by a combination of ammonium sulfate and dried blood, or by manure, or by a combination of all three of these materials. In both experiments the application of superphosphate or of super- phosphate plus sulfate of potash, when used with concentrated sources of nitrogen or with a combination of concentrated sources and manure, failed to increase yields. Although the fixing power for phosphate and potash was undoubtedly greater in the Yolo soil of the Fullerton experiment than in the Hanf ord soil of the Orange experiment, the longer duration of the former trial lends confidence to the results. The exceedingly heavy applications of lime (150 pounds per tree in 1923 and 1927) in one treatment of the Fullerton trial failed to affect the yields, growth of covercrops, apparent penetration of water into the soil, or appearance of the trees. Bul. 673] Effect of Fertilizers on Orange Yields 37 DISCUSSION AND SUMMARY The results of six fertilizer trials with orange trees are reported in the present paper. In each experiment the treatments were replicated on plots of 7 to 16 trees each, separated by guard rows. Reports on the six experiments are in the form of a progress report on one complex, long- term experiment which has already been conducted for twelve years at Riverside (Ramona loam) , and five final reports or summaries of experi- ments of three to twelve years* duration each at Arlington (Hanford sandy loam) , also in Riverside County, at Orange (Hanford sandy loam) and at Fuller ton (Yolo loam) in Orange County, at Jasmine (San Joa- quin sandy loam) in Kern County, and at Porterville (Porterville clay loam adobe) in Tulare County. The primary deficient elements in these trials were nitrogen and zinc. Organic-matter deficiency was also demonstrated in some soils. Other trials have shown that zinc is most satisfactorily applied to citrus trees by means of foliage sprays containing this element. The results at Riverside show large yield responses to nitrogenous fertilizers. Increased yields were obtained with nitrogen applications as high as 3 pounds per tree each year, but the greatest increase per unit of fertilizer resulted from smaller applications. Continuous an- nual applications of 1 pound of nitrogen per tree failed to maintain the original nitrogen supply in the soil, as indicated by covercrop growth. Several inorganic and organic materials were equally effective as sources of nitrogen. Time and frequency of application of nitrate of lime were found to be unimportant in these trials. Increased yield also resulted from application of organic matter in the form of covercrops or bulky organic fertilizers. Where organic matter was not applied or grown, tree responses to zinc treatment were especially marked. Cover- crops appeared to supply all the organic matter necessary. When the total amount of nitrogen applied was small, the spring application of manure as the sole fertilizer depressed yields. With a low rate of nitrogen fertilization, the growing of winter covercrops also depressed the yields of trees fertilized with manure. The depressing effect of covercrops was not observed, however, when the rate of nitrogen fertilization was high. These results indicate that the carbon : nitrogen ratio of the fertilizer materials is important when the supply of available nitrogen in the soil is low, and that extremely wide ratios should usually be prevented by supplementing bulky organic materials with concentrated nitrogenous fertilizers. No response was observed from applications of phosphate, potash, gypsum, limestone, or sulfur. 38 University of California — Experiment Station In the five cooperative trials, similar results were obtained from treat- ments which were comparable. Phosphate, potash, and sulfur had no effect on yields, or, so far as was observed, on time of maturity or com- mercial quality of fruit. These results are in agreement with those of comparable experiments reported previously. Time of application of inorganic nitrogen likewise had no effect. In two trials in which Sudan grass covercrops were grown in the plots just before and during the ripening of the Washington Navel oranges, the development of orange color was hastened, but the surface of the fruit was rougher than normal. This practice did not affect the soluble solids : acid ratio of the juice ; accordingly the practice does not appear commercially feasible. These experiments indicate that there are many ways in which nitro- gen and organic matter may be supplied to citrus trees in these six ex- perimental orchards. The following materials have proved to be satisfactory sources of ni- trogen : the common inorganic and organic fertilizers, nitrate of soda, nitrate of lime, sulfate of ammonia, urea, dried blood, cottonseed meal, and also bulky organics, such as manures, alfalfa hay, cereal straw, and bean straw. In cases of relatively low nitrogen content in the soil, appli- cations of bulky materials should be supplemented with nitrogenous concentrates; this is especially true if the amounts of organic matter applied are large, or if applications are made in the spring. Covercrops are satisfactory as sources of organic matter and, under favorable conditions of growth, may be able to supply all the organic matter needed. The trials suggest that, with the soils on which experiments were con- ducted, the time of application of fertilizers is not important, except in the case of low-grade, bulky organic materials, which may depress yields when applied in the spring in large amounts as the sole source of nitrogen. There is nothing in the results of these six trials to indicate that ap- plications of phosphorus or potassium are necessary. The soil types on which citrus is grown in California are initially relatively well sup- plied with these two elements, and the bulky organic fertilizers com- monly used contain notable amounts of phosphorus and potassium which have increased the supplies in many orchards. In numerous cases in such cultivated lands there are larger available supplies of these elements than in comparable adjacent uncultivated lands. The use of the agricultural minerals, limestone, gypsum, and sulfur, Bul. 673] Effect of Fertilizers on Orange Yields 39 although affecting the pH of some of the soils, has not produced a meas- urable effect on the orange trees or on their crops. Large applications of sulfur temporarily depressed the growth of covercrops. ACKNOWLEDGMENTS We wish to express our thanks and appreciation to all who have as- sisted in any way in this work. Mr. S. M. Brown, Dr. H. D. Chapman, and Mr. B. M. Laurance of the Division of Agricultural Chemistry, University of California Citrus Experiment Station, Riverside, were especially helpful in making chemical analyses of fertilizers, soils, and plant tissues. The greater part of the field operations in the Riverside experiment was supervised by Mr. Charles Wilson and Mr. D. C. Wylie. The experiments in Orange County were conducted by Mr. H. E. Wahlberg and Mr. W. R. Schoonover, and assistance in the Tulare County trials was rendered by Mr. R. W. Southwick, all of the California Agricultural Extension Service. We are particularly grateful for the assistance of the owners and operators of the ranches where the cooperative experiments were carried out. These include the following : The San Jacinto Land Co., Mr. A. S. Holden, and Mr. R. A. Weedon, Arlington, Riverside County ; the A. H. and R. W. Hills Ranch and Mr. Irving Barnes, Porterville, Tulare County; the 0. T. Stephens Ranch, Mr. Frank Daybell, and Mr. Charles Hart, Jasmine, Kern County; the Charles C. Chapman Ranch and Mr. G. W. Riehl, Fullerton, Orange County ; and the Orange County Farm, and Dr. H. E. Zaiser, Orange, Orange County. 207n-12,'42(3202)