The Mineral Nutrition of Citrus



      Any comprehensive discussion of citrus nutrition should include sections dealing with the mechanism of nutrient absorption, the influence of one ion on the absorption of another, translocation and mobility of nutrients within the plant, the effects of light, temperature, and humidity on nutrition, the relation of soil to nutrition, internal physiology, and other topics.   Two pertinent recent works are Nutrition of Fruit Crops edited by Childers (1966) and Diagnostic Criteria for Plants and Soils edited by Chapman (1966).
      This chapter is confined to a discussion of the mineral composition of citrus, the effects of nutrient deficiencies and excesses, salinity, pH, aeration, and some consideration of both the causes of nutrient disorders and methods of control.   A detailed discussion of fertilization is presented in Volume III [text version, Revised Ed.].


      Many analyses of the fruit and the various parts of citrus trees have been made.   Among the early analyses, the most complete are found in publications by Thorpe (1868), Wolff (1871-80), Ricciardi (1880a, 1880b), Boschi (1895), and Oliveri and Guerrieri (1895).
     Less complete, but more inclusive as regards the number of parts of the tree included, is an analysis of a nineteen-year-old Marsh Seedless grapefruit tree by Barnette et al. (1931).   This analysis is especially valuable since the tree was mature and total weights of the various parts of the tree were recorded.   Data from the analysis are reproduced in part in table 3-1.
      Representative analyses of the important vegetative parts of citrus trees are assembled in table 3-2.   In table 3-3, analyses of citrus fruits are brought together.   Considering the number of factors affecting plant composition and their variability, the reported data are in fair agreement.
      With respect to the quantities of ash constituents in various parts of the tree, it is evident that the leaves rank first.   The more woody parts rank lowest; such determinations as have been made on bark and wood show that in ash content the bark is more nearly comparable with the leaves.
      The percentage distribution of various elements in the ash is given in table 3-4.   In all vegetative parts of citrus trees, calcium is the dominant constituent, comprising, in most parts, more than 34 per cent of the total weight of the ash.   Next in abundance is potassium, comprising from 7 to 15 per cent.   Lower, and all roughly of the same magnitude, are magnesium, phosphorus, sulphur, and silicon.   Chlorine is the least abundant element.   Sodium is variable.   (Other constituents will be considered under trace elements.)   In the fruit, potassium rather than calcium is the dominant base.   The other elements are present in fruit ash in about the same proportion as in the ash of vegetative parts, phosphate being perhaps somewhat higher in the fruit ash.
      The analysis of a nineteen-year-old grapefruit tree by Barnette et al. (1931) provided interesting information on the distribution of various elements in different parts of the tree (table 3-1).   Most of the calcium and magnesium was found in the branches, the trunk, and the large roots; relatively more of the phosphate was present in the roots.   Nitrogen and potassium were predominantly localized in the leaves and roots, with lesser amounts in the limbs and trunk.   The analyses are assembled here to provide a generalized picture of amounts, proportions, and distribution of mineral constituents in citrus trees.   They are suggestive of trends only.   The proportions of total dry matter and ash in various parts of mature citrus trees often vary considerably.   This is well brought out by comparing data for the nineteen-year-old grapefruit tree given above with data for a ten-year-old Valencia tree (Cameron and Appleman, 1934).   For both trees, the entire weights of leaves, branches, trunk, and roots were recorded, and nitrogen analyses were made.   Since data for the roots may be somewhat in error, results are given only for leaves, branches, and trunks (table 3-5).   For the Valencia tree, it will be noted that foliage represented 28.35 per cent of the total dry weight and contained 59.39 per cent of the total nitrogen in the above-ground part of the tree, as compared with 8.85 per cent and 34.76 per cent, respectively, for the grapefruit tree.
     The amounts and proportions of ash constituents in any part of a tree are profoundly influenced by age.   This was strikingly shown in leaf analyses reported by Kelley and Cummins (1920), whose data are reproduced in part in table 3-6.   As may be seen from the table, the leaves became progressively richer in ash and calcium with age.   Increases in calcium were accompanied by decreases in potassium and phosphate.
      Because of current interest in trace elements, miscellaneous analyses by various investigators are assembled in table 3-7.   Of the various elements reported, iron is the most abundant, ranging roughly from 100 to 200 ppm in the dry leaf.   Boron, aluminum, and fluorine range from 30 to 50 ppm.   Manganese and zinc are less abundant, and copper is least abundant.
      In addition to these analyses, data published by Chapman (1960)—much of which was derived from unpublished information supplied by Vanselow—are shown in table 3-8.   Roberts and Gaddum (1937) determined by spectrographic means the minor elements in various citrus fruit juices.   Their results are recorded in table 3-9.
      More recent data, showing the inorganic composition of leaves and fibrous roots of young Valencia orange trees fertilized with various amounts of nitrogen, potassium, and magnesium, have been published by Smith et al. (1954).   These data are shown in table 3-10.   In addition, Smith and Reuther (1953) published data relating whole fruit composition to variable fertilization with nitrogen, potassium, and phosphorus.   These data are shown in table 3-11.
      Some pertinent data on the variable composition of the rootlets of Valencia oranges (on rough lemon rootstock), showing marked differences in rootlet composition with soil depth, were published by Reuther, Embleton, and Jones (1958).   These data from Florida orchard experiments are given in table 3-12.
      Based on analyses recorded in the foregoing tables, calculations have been made showing the amounts of the various elements removed from an acre of soil by an average crop of oranges (20,000 pounds).   These results (table 3-13) show that potassium, calcium, nitrogen, and phosphorus are the elements removed from the soil in the greatest quantity.


      There is no evidence as yet that aluminum is essential for the growth of citrus or other higher plants.   Much work, as is well known, has been done on aluminum toxicity.
      In water-culture experiments with lemon cuttings, Liebig, Vanselow, and Chapman (1942) produced healthy plants with aluminum at concentrations as low as 0.005 ppm in the culture solution.   Leaves from these plants showed 28 ppm in the dry matter; whole stems, 5.6 ppm; root wood, 4.5 ppm; root bark, 92 ppm; and fine roots, 280 ppm.   Comparable plants in solutions containing 0.5 ppm of aluminum showed the following concentrations in the dry matter: leaves, 22; stems, 4.5; root wood, 9.2; root bark, 700; and fine roots, 1,000 ppm.   Aluminum at rates of 0.1, 0.5, 2.5, and 5 ppm in the culture solution offset the toxic effect of copper when it was present in concentrations of 0.1 ppm, but at 0.02 ppm of copper, as little as 0.1 ppm of aluminum somewhat decreased top growth, and 2.5 and 5 ppm of aluminum exerted a substantial depressing effect on top growth but did not affect root growth.   In fact, levels of 2.5 and 5 ppm of aluminum tended to stimulate root growth.   The reasons for this behavior were not ascertained.   Blevins and Massey (1959) and Hiatt, Amos, and Massey (1963), working with wheat and excised wheat roots, respectively, confirmed that aluminum can decrease copper uptake.   The latter investigators believe that aluminum competes with copper for binding sites on plant roots.   They also allude to further research which suggests that aluminum can markedly reduce manganese uptake by tobacco.   Rees and Sidrak (1961) found that aluminum reduced manganese toxicity in Atriplex hastata var. deltoidea, but not in spinach and barley.   Whatever the mechanism involved, it is probable that low concentrations of aluminum in the 0.1- to 1-ppm range may (under conditions of excess copper, manganese, and probably other metal ions) exert a beneficial effect on plant nutrition.
     The toxicity of aluminum to plants is well known.   In a recent summary of this subject, Pratt (1966) reviewed the results of controlled culture work by many investigators with quite a range of plant types.   He noted, in general, growth reduction from nutrient concentrations ranging from 0.1 to over 5 ppm, depending on plant and nutrient conditions.
      Because of its relatively high content in soils and probable low plant requirement (should aluminum prove indispensable), it seems unlikely that lack of aluminum will ever become a problem with citrus or other plants.   On the other hand, aluminum salt or chelate additions to soils might prove beneficial in situations where copper or manganese toxicity is a problem.   Aluminum may, in fact, be a natural factor in soils in offsetting to some extent the toxicity of copper, manganese, and other constituents made soluble under acid soil conditions.   It may also pose toxicity problems.
      Aluminum does not appear to be transported significantly to citrus tops and does not produce specific toxicity symptoms.   The seat of toxicity appears to be in the roots, exerting simply a depressing effect on top growth.   Aluminum's effect on copper uptake and perhaps manganese and other essential nutrients might, under some conditions, produce deficiencies of these elements.


      Special interest attaches to arsenic in relation to citrus nutrition because of early findings that arsenical sprays reduce fruit acid and thus make for earlier maturity.   While there is no evidence that arsenic is essential for citrus, it is toxic to plants.   In so-called "normal" soils, arsenic ranges only from <1 to about 50 ppm (according to a survey of the literature made by Swaine, 1955), but in some soils, where it has been used as a pesticide or herbicide, much higher amounts are found.
      Arsenic is still used some in snail baits and also in sprays to hasten citrus fruit maturity.   Some well and spring waters contain abnormal amounts of arsenic, and Fergus (1955) found several soils near Brisbane, Australia, where the natural levels of arsenic were sufficient to cause disorders in bananas.   These soils contained from 100 to 200 ppm of total arsenic.   The source of the arsenic was traced to veined quartzite rock outcrops and fragments, apparently coming from mispickel in the rock veins.
      Liebig, Bradford, and Vanselow (1959) reported that lemon cuttings grown in solution cultures were injured by 5 ppm of arsenite arsenic and 10 ppm of arsenate form.   Root tips showed injury first, and leaves wilted as the injury progressed; at 10 ppm of arsenic as arsenite, the lemon cuttings died after eleven days.   Arsenate was less toxic than arsenite at equivalent concentrations.   The roots of plants grown at the 10-ppm arsenic level gradually became stubby, and root analyses showed up to 1,200 ppm in the dry matter of plants grown in solutions containing 10 ppm, as compared to 0.15 ppm in solutions to which no arsenic was added.   A little root toxicity eventually showed up at 2 ppm of arsenite and 5 ppm of arsenate arsenic.   Tender, new roots were more susceptible to injury than older roots.   Because of root injury, arsenic does not appear to accumulate appreciably in the tops of plants, though Liebig et al. (1959) found up to 11 ppm of arsenic in the old leaves of affected plants, as compared to 0.10 to 0.15 ppm in comparable leaves from arsenic-free cultures.   Old leaves contained more arsenic than young leaves, and (from Liebig's experiments) values for total arsenic in excess of 1 ppm in old, lower leaves of citrus suggest the need for further exploratory work to determine the possibility of toxicity.
      An interesting observation in Liebig's experiment was that 1 ppm of arsenic produced a stimulating effect on roots but not on top growth.
      The only abnormal leaf symptom to develop in lemon cuttings receiving excess arsenic was vein clearing—"vein chlorosis" or "yellow vein."   This condition results from root rotting, breaks in the vascular system, or (particularly in lemons) from nitrogen deficiency.   Impaired root activity and absorption of nutrients was probably the cause of this condition in these experiments with excess arsenic.
      Working with Pineapple orange seedlings, Rasmussen and Henry (1964) found that as little as 0.1 ppm of arsenic as potassium arsenate in the nutrient solution at pH 5 to 6 reduced growth significantly, whereas 0.25 ppm reduced growth to about one-half of the controls.   These levels were associated with about 0.4 to 0.5 ppm and 0.5 to 0.6 ppm, respectively, in leaves and 40 to 50 ppm and 50 to 70 ppm in roots.   Most seedlings died after a few weeks at 4 ppm.   In sand cultures, about 1 ppm in the substrate reduced growth about the same as 0.1 ppm in solution cultures.
      A type of symptom seen on grapefruit in Florida, resulting from arsenical sprays, is shown in figure 3-1.
      It has long been known that arsenical sprays or dusts will decrease the acidity of citrus fruit.   This fact was first reported in the Florida State Horticultural Society Proceedings of 1894 (vol. 7, p. 81), and subsequent studies in Florida, California, South Africa, and Japan confirmed these early findings.   Miller, Bassett, and Yothers (1933) reviewed the early work on this subject and made a thorough study of the effects of lead arsenate sprays on the arsenic content of leaves, wood, fruit, and blossoms; they also studied its persistence on or in the tree, in the soil, and its effect on fruit composition, foliage, and fruit metabolism.
      They found that arsenical sprays increased respiration rate in leaves and that catalase activity was increased by dilute arsenical sprays, but markedly decreased where more concentrated sprays were used.   These metabolic changes were related to the fruit acid decreases.   Fruit acid in oranges was decreased in unsprayed trees from 1.5 per cent to 0.5 per cent where the arsenous oxide of the leaves showed values of 0.1 mgm in 10 gm of fresh leaves.   When arsenic reached concentrations of 1 mgm of arsenous oxide in 10 gm of fresh leaves, respiration increases ranged from 50 to 75 per cent, and leaves usually dropped.   Catalase activity was increased when arsenous oxide was around 0.01 mgm per 10 gm of fresh leaves, but almost stopped when the arsenous-oxide levels reached 1 mgm.   Very little arsenic found its way into the fruit juice.   These authors reported that even after seventeen sprayings at 5 gallons per spray, only 0.16 mgm of arsenous oxide was found in a liter of juice.
      Repeated lead arsenate sprays applied to the soil under citrus trees led to the accumulation of 2,000 ppm of arsenous oxide in the first 2 inches of soil, and 6 ppm in the 8- to 10-inch depth; comparable untreated horizons showed only 0.04 ppm.   However, there was very little arsenic in the leaves of trees growing on the arsenic-contaminated soil, and there were no effects on growth, foliage color, fruit acid, and solids.   This work was carried out on a Norfolk sand in an area where the annual rainfall is about 55 inches.
      Roy (1943) and Gardner (1943) reported a sand-culture study with Duncan grapefruit trees on sour orange rootstock.   Increasing amounts of sodium arsenate were added to the sand culture at rates of 10, 50, 200, and 800 ppm of arsenic.   At 800 ppm, the trees began to defoliate after two months, and the few leaves remaining showed arsenic levels of 116 ppm.   Leaves from trees growing in cultures receiving 200 ppm of arsenic contained 83 ppm of arsenic.   The fruits from these trees were yellowish-green in color and showed brown stains in the albedo similar to the effects produced by boron deficiency.   A theory was advanced that boron may function in the plant as part of an enzyme molecule and that arsenic interferes with this enzyme, producing the same effects as lack of boron.


Boron Deficiency
      The first researcher to produce and describe symptoms of boron deficiency on citrus was Haas (1929b, 1930).   Later, Haas and Klotz (1931) described some of the anatomical and chemical changes produced in young citrus plants when this element is lacking.   Morris (1937, 1938), in Rhodesia, South Africa, was the first to diagnose boron deficiency of citrus in the field and describe the symptoms.
      The most conspicuous symptom produced under greenhouse conditions, as described by Haas (1929b), was a spectacular enlargement, splitting, and corking of leaf veins (see fig. 3-2, D).   However, in the field, Morris (1938) found none of these symptoms, but described a fruit condition to which he gave the name "hard fruit."   Subsequently, Chapman and Kelley (1943), in a fertilizer experiment begun in 1934 with navel orange trees in 55-gallon containers (using a granite-derived soil), produced boron deficiency and found both the "hard fruit" condition described by Morris (1938) and some of the vein splitting of leaves and bark splitting and gumming described by Haas (1929b), together with the abortion of young growth and dieback described by both investigators.   The conditions were cured by either foliar sprays or soil additions of boric acid.   The boron content of leaves (Chapman and Kelley, 1943) ranged from 5 to 15 ppm in the dry matter.   Subsequent to Morris's findings, Camp (1939) noted symptoms in trees in Florida akin to those reported from Rhodesia; later still, Smith and Reuther (1950a) described symptoms on field trees in Florida, noting all of the fruit and foliage symptoms previously described.   The boron content of the foliage showing strong deficiency symptoms averaged 14 ppm.   These trees responded to sprays containing one pound of borax to 100 gallons of water, as well as to soil applications of three ounces of borax per tree.   Meanwhile, Roy (1943), in Florida, and Haas (1945a), in California, published the results of controlled culture experiments, confirming the range of symptoms as described by earlier investigators and supplying further data on the boron content of deficient citrus plants.   In further work, Smith and Reuther (1951) determined the effects on citrus tree size, foliar symptoms, leaf analysis, and vitamin C in fruit from applications of 0.01, 0.5, and 2 ppm of boron to sand cultures.   At concentrations between those producing near deficiency and slight excess, no special effects on growth, fruiting, or mineral content were found.   In Japan, lshihara et al. (1965) found soil applications of borax increased fruit size and the proportion of well developed seed in satsuma.   As a result of all this work, a fairly complete understanding of the foliage, growth, and fruit conditions produced by moderate to acute boron deficiency is at hand, including information on the range of boron levels in leaves which show typical hunger symptoms.
      Table 3-14 presents a tabulation of the foliar, growth, and fruit characteristics which may be produced when boron is lacking, together with leaf analysis values associated with boron deficiency.   Also shown for each abnormality are other known conditions which can produce identical or somewhat similar symptoms.
      It is immediately apparent that no one of the visual foliar, growth, or fruit symptoms is of itself a sure sign of boron deficiency; thus, leaf curling, lack of luster, dieback, vein enlargement, splitting and gumming of twigs and branches, and the fruit abnormalities listed are common to other nutritional disorders, or some of these could be damage effects from insects, insecticides, hail, ice, and other causes.
      When observations of visual effects are supplemented by leaf analyses, however, diagnosis can be almost certain.   In other words, trees showing the complete syndrome of visual effects, together with leaf values showing 5 to 15 ppm of total boron, are almost certainly deficient in boron.
      DeVilliers and Beyers (1961) got boron responses in citrus trees which showed up to 50 ppm of total boron in the dry matter of leaves, and they found evidence that water-soluble boron in leaves may be a better index of adequacy or inadequacy than total boron.   Values of less than 15 ppm of water-soluble boron in the dry matter would be considered low; values well above this would be considered sufficient.
      Boron deficiency effects are shown in figures 3-2 and 3-3.   The fruit symptom on citrus most indicative of boron deficiency is the presence of a darkish, irregular ring or various localized darkish-colored spots in the white albedo of young orange fruits.   The only other conditions known to produce such symptoms are (1) excess arsenic (likely to be encountered only where arsenical sprays are used), and (2) a virus disease reported in Sicily.   The fruit symptoms of boron deficiency (or arsenic toxicity) are shown in figure 3-2.   Where such symptoms are encountered and if leaf analysis shows total boron in the dry matter in the range of 3 to 15 ppm, it is almost certain that boron is lacking.   Roy (1943) has shown that excess arsenic (which produces similar symptoms) does not impede boron uptake, and leaves from such plants will show normal boron levels (i.e., 30 ppm or more of boron.)
      Early Symptoms of Boron Deficiency.—As with many other deficiencies, no information is available that would afford a positive diagnosis of the early stages of boron deficiency.   However, low yields, abnormal abortion of young fruit, dieback of new growth, young fruit with albedo discoloration, and a foliage content of 25 ppm of boron or less would be suggestive and warrant trial applications of borax sprays or soil applications.
      Causes of Boron Deficiency.—Boron deficiency commonly occurs in soils basically low in boron content (such as those derived from igneous rocks), or where crop removal has depleted the supply or liming has decreased boron availability.
      Unfortunately, insufficient soil analysis data from soil areas where boron deficiency of citrus has been found in the field are at hand to be of use; but the boron content of the granitic soil in which boron deficiency ultimately developed in the author's oil-drum experiment (Chapman and Kelley, 1943), showed, at the outset, 10 ppm of total boron in the soil.   Many soils derived from igneous-type rocks run much lower, as has been shown by Whetstone, Robinson, and Byers (1942), and the acid-soluble portion (which is probably the only portion available to plants) runs less than half the total, on the average.   It has been suggested by Whetstone et al. (1942) that soils containing less than 10 ppm of boron (as determined by digestion in 85 per cent phosphoric acid followed by distillation with methyl alcohol) are suspect with regard to boron adequacy.   Unless the fertilizer program or irrigation water adds enough to meet plant needs, such soils are likely to become boron-deficient.
      Whetstone et al. (1942) analyzed some 300 soils representative of most of the major soil types in the United States and got values for total boron ranging from 4 to 133 ppm.   The lowest values generally were obtained on soils derived from basic igneous rocks, such as granite, gneiss, and schist.   Soils derived from shales, limestone, and sedimentary rocks generally are higher in boron.   Swaine (1955) reported values for total boron in soils ranging from 2 to 100 ppm.
      From the evidence available, boron deficiency is to be looked for in the following kinds of soils: (1) those naturally low in boron, e.g., those derived from igneous and metamorphic or sandstone rocks; (2) those which are acid regardless of parent material, where the more soluble fractions have been leached out; (3) those which, other factors being equal, are low in colloid content; (4) those where laterization has taken place, i.e., soils which become progressively lower in silica compared with iron and aluminum; (5) acid muck and peat soils; and (6) alkaline soils which, while commonly containing ample boron, are low in available supply.   From this it is understandable why liming of acid soils has often induced boron deficiency; conversely, the acidification and leaching of alkaline soils will accelerate the outgo of boron from soils and can bring on boron deficiency.   Bingham, Martin, and Chastain (1958) have noted that high phosphate decreases boron absorption.
      In addition to soil factors, boron deficiency in citrus is more likely to occur with citrus varieties grown on sour orange rootstock, since it appears that the feeding power or permeability of cells for boron in sour orange is less than that of other common rootstocks used for citrus.   (This subject will be reviewed in greater detail in the discussion below of boron excess.)
      Control of Boron Deficiency.—Correction of boron deficiency is a comparatively simple matter.   Nutritional sprays, using either boric acid or borax at rates of one pound per 100 gallons of water, or soil applications of either of these compounds at rates of 1/4 to 1/2 pound per tree, give control.   Because of the narrow range between deficiency and toxicity, low application rates are definitely safer.   Smith and Reuther (1950a) reported that three ounces of borax broadcast per tree produced correction on the sandy soils of Florida.   Morris (1938), in South Africa, used rates varying from 1 to 35 ounces per tree.   The heavier application produced no injury, but the soil was heavy and probably alkaline, and under these conditions the danger of injury is less.

Boron Excess
      Boron excess produces a characteristic apical leaf mottling which is easily recognizable when fully developed and can usually be distinguished by leaf and soil analyses from other conditions producing similar symptoms.   The principal effects of boron excess are outlined in table 3-15 (see fig. 3-4 for illustrations).
      Excess sulfate produces a very similar type of apical leaf mottling, and somewhat similar patterns are produced by excess biuret and perchlorate (see discussion under Nitrogen).   Also, the early stages of molybdenum deficiency, as produced in controlled cultures by Vanselow and Datta (1949), were characterized by an apical mottling.
      In soil areas where these complications are likely or suspected, leaf analyses should be undertaken to confirm the visual diagnosis.   If the symptoms are linked to boron, the total boron of the leaf will probably run higher than 250 to 300 ppm on a dry matter basis, but not infrequently both boron and sulfate may be involved; in some instances, where Chilean nitrate has been used, this also leads to combination symptoms.
      In Greece, Demetriades, Holevas, and Gavalas (1962) reported values up to 874 ppm of boron in dry matter of leaves affected by excess boron.   Nonchlorotic leaves from injured trees showed up to 505 ppm.   Hence, it is possible to find rather a high boron content in leaves which do not show symptoms.
      On older leaves, particularly orange, grapefruit, and mandarin, there are brown, resinous spots or irregular areas on the undersurface of leaves, where the mottling is caused by excess boron.   This latter effect is not seen when sulfate, biuret, or perchlorate is the cause; hence, when the underleaf resinous spotting appears, it helps to confirm the boron excess diagnosis.
      Boron accumulates progressively as citrus leaves age; hence, the apical mottling is usually more pronounced in the late summer and fall than in the spring and early summer.   As injury progresses, there is necrosis of the leaf tip and margins, and the mottling may affect nearly all of the leaf blade.   The boron content of the injured leaf area is higher than in the green portion of the leaves, according to analyses made by Eaton (1935).   Eaton found the following concentrations of boron in various parts of injured lemon leaves:

                                                                       Ppm Boron in
Part of Leaf                                                     Dry Matter  

Midvein and petioles                                                    47
Green portions of blade                                              438
Yellowed portion                                                     1,060
Dead portion of apices and margin                           1,722

Oertli (1960) has published confirmatory evidence and found up to 1,180 ppm of boron on a fresh-weight basis in necrotic areas of boron-injured rough lemon leaves.
      The very earliest stage of boron injury usually starts as a tip yellowing or mottling, but injury from many other sources can also induce this condition; hence, this symptom of itself is not too diagnostic, unless supplemented by analysis of the yellowed apices for boron.
      In severe cases of boron injury, leaves fall prematurely.   Successive flushes of growth usually follow; these new leaves are at first green, but subsequently develop typical boron-excess patterns.
      Severe leaf injury and abscission eventually result in twig dieback, and, in very severe cases, the entire tree may become defoliated.   Death of the tree can occur, but only where the boron present in the soil is very high; otherwise, the tree will survive, but year after year show the typical leaf-injury symptoms and pronounced defoliation.
      As shown in table 3-15, the boron of leaves from trees deficient in the element may be as low as 3 ppm and, in cases of severe toxicity, as high as 2,400 ppm.   Leaf symptoms due exclusively to boron accumulation usually (but not always) begin to show when the concentration approaches 300 ppm in the dry matter and become progressively acute as the boron increases.
      Smith and Reuther (1951) grew young Valencia trees in sand cultures for three years.   Excess boron was used in amounts sufficient to produce only mild leaf symptoms, and no evidence of decreased tree growth or yield was obtained.   Bradford and Chapman (1964, unpublished) have comparable data.   Scofield and Wilcox (1931) state that at the stage of mild leaf symptoms there may be some increase in yield.   However, there is insufficient evidence to substantiate this latter observation.
      The author is acquainted with many citrus orchards in California where mild symptoms of boron excess appear not to have affected either yield, size or quality of fruit, or tree vigor.
      Where boron is present in amounts sufficient to cause pronounced leaf symptoms and twig dieback, tree vigor and yield are reduced, but no data are available as to the effects on size or quality of fruit or the exact point of injury at which there is impairment in the vigor and yield.
      Causes of Boron Excess.—Boron excesses in irrigated regions commonly result when irrigation waters contain more than 0.5 ppm of boron.   Where the concentration is greater than 1 ppm, boron toxicity is apt to be fairly pronounced.   Waters containing more than 2 ppm of boron are generally regarded as unfit for citrus irrigation.   The nature of the soil and the climate, of course, exert considerable effect, as do rootstock and scion, and hence these figures can be regarded as approximate only.
      Some soil areas are naturally high in boron, and citrus trees growing in these soils will show boron-excess symptoms.   Penman and McAlpin (1949) report naturally occurring boron toxicity on citrus in some Australian soils of marine origin.   Similar soil areas occur on the west side of the San Joaquin Valley of California, and no doubt elsewhere.
      Boron toxicity also has been noted in soils receiving borax wash waters from citrus packinghouses.   Some fertilizers and soil amendments contain enough boron to produce severe boron injury of citrus.
      Data on the concentrations of boron in irrigation waters and substrate which can produce boron toxicity of citrus are tabulated in table 3-16.   With soils where the soil solution contains in excess of 0.5 ppm of boron, leaf patterns of excess boron will commonly appear.   Hence, analysis of soil extracts for water-soluble boron provides supplemental data useful in diagnosing boron excess.
      Effect of Variety and Rootstock.—There is extreme variability among plants in the amount of boron absorbed from the same substrate, great differences in the amounts accumulating in different parts of the same plant, and pronounced variety and rootstock effects.   All of these factors were clearly demonstrated by Eaton and Blair (1935), who grew a variety of plants in sand cultures supplied with a nutrient solution containing 5 ppm of boron.
      Analysis of the entire plant revealed the following concentrations of boron in the dry plant tissue: beans, 648; wheat, 453; tomatoes, l59; cotton, 123; and carrots, 69.   Eaton and Blair also showed that lemon plants on their own roots accumulated 1,065 ppm of boron in their leaves (dry matter) and only 283 ppm when grown on box orange (Severinia buxifolia [Poir.] Tenore) roots.   Box orange, grown on lemon rootstocks, accumulated 877 ppm of boron in the leaves (dry matter).   On its own roots, it accumulated only 390 ppm in the leaves.
      Eaton and Blair concluded that boron absorption is determined in part by the character of absorbing root cells, the nature of boron compounds accumulating in leaf cells, and the equilibrium between mobile and nonmobile forms of boron within the plant.
      In Florida, Roy (1943) found that grapefruit absorbed less boron when grown on sour rootstock than when grown on trifoliate.
      Haas (1945b) analyzed the leaves of citrus growing on various rootstocks in a rootstock experiment conducted on a uniform soil; he found least accumulation where the stock was sour orange and greatest accumulation on trifoliate orange root.   Values of 43 to 48 ppm of boron were found when the stock was sour orange and 119 ppm when it was trifoliate orange.   Rough lemon roots produced 75 ppm in the leaf, and boron in the leaves of trees on sweet orange root ranged from 58 to 60 ppm.   Smith, Reuther, and Specht (1949) reported results on five-year-old Valencia trees in which the least boron uptake was on sour orange rootstock and the maximum was on Rusk citrange.   The boron levels in six-month-old leaves were respectively 67.3 and 94.6 ppm in the dry matter.
      Penman and McAlpin (1949), in Australia, found that Washington navel and late Valencia oranges were more affected by boron toxicity when grown on rough lemon (Citronelle) than when grown on sour orange.   They were least affected on sweet orange stock.   Trifoliate orange stock had about the same rating as sweet orange stock.   Cooper and Edwards (1950), working with Valencia orange and grapefruit, found no difference in boron absorption when trees were grown on either sour orange or Cleopatra mandarin rootstock.   In later work, Cooper and Peynado (1959b) found that Citrus macrophylla Hook. rootstock tended to exclude boron, but accumulated chloride; Cleopatra, Tonkan, and Sunki mandarin rootstocks accumulated boron but tended to exclude chloride.   No rootstocks excluded both boron and chloride.   Embleton, Labanauskas, and Bitters (1962) found that young, mature leaves of Frost Nucellar Eureka lemons, growing on various rootstocks, showed the following amounts of boron in the dry matter of leaves: Citrus macrophylla, 101; C. moi, 164; Rangpur lime, 175; sweet orange, 178; C. taiwanica, 191; Yuzu, 196; and grapefruit, 215 ppm.   From these data, it is apparent that boron toxicity (as well as deficiency) can be greatly influenced by rootstock.
      As regards scion varieties, it generally is observed that the lemon is more susceptible to excess boron than the orange; the order of decreasing susceptibility, as found by Penman and McAlpin (1949) in Australia, is lemon, sour orange, mandarin, kumquat, grapefruit, and sweet orange.
      Control of Boron Excess.—Control measures depend to some degree on the cause of boron excess.   Where excessive boron is present in irrigation water, the best solution, of course, is to find another source of water.   If this is not feasible, then the growth of some more tolerant crop is the only alternative, if the injury is severe enough to prohibit economical citrus growing.
      Where injury is only moderate, it can be lessened by using water in sufficient amounts to promote frequent root-zone flushing.   Heavier than usual applications of nitrogen fertilizer, especially calcium nitrate (Cooper, Peynado, and Olson, 1958), and moderate liming of acid soils are beneficial.
      Where high boron is due to natural soil contamination, irrigation practices which promote root-zone leaching, heavier than usual nitrogen applications, and moderate soil liming if the soil is noncalcareous are helpful.


Calcium Deficiency
      Until recently, calcium deficiency of citrus had been produced only under controlled sand- and water-culture conditions.   Various descriptions have been published by Reed and Haas (1923a, 1923b, 1924); Camp et al. (1941); Chapman and Kelley (1943); Camp, Chapman, and Parker (1949); Bryan (1940); Brusca and Haas (1956b); Chapman (1961, 1966), and Chapman, Joseph and Rayner (1965).
      Responses to calcium sprays (2 per cent slaked lime) were reported by DeVilliers et al. (1958) on field-grown oranges in South Africa.   Mature leaves contained as little as 2 per cent total calcium; the soil contained 0.5 meq calcium per 100 gm; and trees showed sparse foliation, twig dieback, chlorosis symptoms, profuse blossoming, and excessive drop of young fruit.   Fruit which matured was often misshapen, acid, and on the flat side as regards total solids.   DeVilliers et al. (1958) noted that there were several areas in South Africa where calcium was low, and thus there was the possibility of calcium deficiency showing up.   In Florida, Spencer (1960) and Spencer and Koo (1962) uncovered calcium deficiency on Ruby Red grapefruit (rough lemon root) and on Pineapple orange growing in a Lakeland fine sand.   The trees which developed calcium deficiency were on acid check plots of a fertilizer experiment involving limestone and superphosphate.   No calcium deficiency developed on either the limestone or superphosphate plots.   The calcium-deficient trees were stunted and produced much less fruit than those receiving calcium.   A leaf chlorosis developed, characterized by irregular marginal yellowing, with the midribs and veins and areas surrounding the veins retaining their green color.   These symptoms were much the same as those illustrated by Chapman and Kelley (1943, fig. 160, p. 731) and Bryan (1940) (see fig. 3-5).   The calcium content of six-month-old leaves from nonfruiting terminals, as reported by Spencer and Koo (1962), ranged from 0.80 to 1.23 per cent in the dry matter, as compared to 3.17 per cent for comparable leaves from non-calcium-deficient trees.
      As reported by Spencer and Koo (1962), analyses of late-summer-flush grapefruit leaves sampled in January from trees lacking calcium compared with similar leaves from healthy trees were as shown below.

Treatment and           Percentage in Dry Matter of Leaves            Ppm in Dry Matter of Leaves
                 Ca         N           P           K           Mg           Fe          Mn         Zn           Al 
+Ca—green            2.47      2.26       0.14      2.06         0.28          60           27         19           46
-Ca—chlorotic        0.24      1.97       0.12      2.96         0.30          47           15         21           53

      Accompanying the marked decrease in calcium was a decrease in nitrogen, iron, and manganese.   Potassium and magnesium, on the other hand, were increased.
      In the Florida trees, premature leaf drop, multiple buds, and dieback of the trees occurred.   Some of the leaves were heart-shaped due to shortened midribs.   Some of the fruit was misshapen from shrunken juice vesicles.   The juice was more acid.
      These symptoms correspond well with those produced under controlled conditions by Chapman and Kelley (1943) and Bryan (1940), and with those described by DeVilliers et al. (1958) on oranges in South Africa.   A summary of the various growth, foliar, fruit, and leaf analyses conditions which characterize calcium deficiency on citrus, as produced both under controlled conditions and in the field, is presented in table 3-17.   Illustrations of tree, foliar, and fruit symptoms are shown in figs. 3-5 and 3-6.
      Causes of Calcium Deficiency.—The two field cases of calcium deficiency reported thus far—in South Africa and Florida—were on soils of very low calcium content.   The South African soil had 0.5 meq of calcium per 100 gm (or 100 ppm), while the Florida soil showed 8 to 30 ppm on soils of pH 4.3 to 4.5.   Recently, Martin and Page (1965) produced calcium deficiency of sweet orange seedlings in potted soils in which the exchangeable calcium percentage was reduced progressively.   When the percentage calcium saturation of the soil was at about 15, low leaf levels of 0.5 to 0.7 per cent were produced in the dry matter.   No consistent leaf patterns developed, however.
      The author and his colleagues (Chapman, Joseph, and Rayner, 1965) recently concluded an eight-year experiment on the calcium nutrition of bearing orange trees (Troyer citrange rootstock) growing outdoors.   Moderately acute calcium deficiency occurred in cultures where the calcium concentration ranged from 10 to 20 ppm in the nutrient medium.   Only slight deficiency developed on trees in cultures containing 40 to 50 ppm of calcium.   While the calcium deficiency was not acute enough to produce marginal yellowing of leaves, some vein clearing was noted on the foliage, trees were stunted, and growth flushes were shortened.   Manganese deficiency occurred in the winter, and some root rotting was noted in container areas remote from the aerator tube.   In similar areas of the culture container where calcium was higher, no root rotting occurred.   This strongly suggests that roots under subnormal calcium conditions are more susceptible to fungi invasion than those whose calcium nutrition is good.   (Fusarium spp. were the chief fungi.)   This may explain the rather widespread, vein-chlorosis condition which the author has seen in parts of Australia, India, and elsewhere.   Poor aeration (high moisture), low calcium, low nitrogen, cold soil temperatures, and viruses may all be involved in vein chlorosis also.
      The results suggest that while acute calcium deficiency of citrus is not likely to be encountered except in sandy, acid soils of low-calcium content, low-calcium conditions may be more widespread than previously supposed.   Trials with calcium salts or limestone should be initiated wherever the calcium content of mature, four- to six-month-old leaves is less than 3 per cent.
      Low-calcium conditions, while most common in acid soils, may possibly occur under conditions of high soluble sodium or pH conditions resulting from exchangeable sodium percentages well above usual levels.   Calcium is important in counteracting the effects of high soluble sodium levels and may be of such low solubility in soils dominated by exchangeable sodium as to bring on calcium deficiency.
      Low-calcium conditions have been reported on serpentine-derived soils and strongly acid peats.   Calcium absorbed on kaolinite appears more available to plants than that absorbed on montmorillonite.
      Control of Calcium Deficiency.—The Florida field case of calcium deficiency, as already mentioned, was corrected by soil treatment with limestone and/or superphosphate.   In South Africa, correction was obtained by a foliar spray of 2 per cent lime hydrate.   The worldwide successful use of limestone and/or gypsum for correcting low calcium and calcium deficiency under a broad range of conditions indicates that no serious correction problems are encountered with citrus when deficiency or low-calcium conditions occur.   Sand-culture studies have shown that trees almost dead from lack of calcium recovered rapidly when this element was restored.

Calcium Excess
      Calcium excess effects are due more to the anion with which it is associated than to the calcium ion per se.   In the case of soluble salts, such as calcium chloride or sulfate, it appears to be the excessive absorption of chloride and/or sulfate that produces trouble rather than the accumulation of calcium in tissues.   However, more research needs to be carried out on possible specific effects of high-calcium concentration.       In the case of calcium carbonate, the alkalinity and buffering effects it produces in soils and the effects of these on the availability of manganese, iron, zinc, copper, boron, and phosphorus are of principal importance.


Copper Deficiency
      In Florida, a citrus condition first described as "dieback" or "red rust" was reported by Fowler (1875).   Floyd (1914) found that copper Bordeaux sprays corrected the condition, and Grossenbacher (1916) found that soil applications of copper sulfate restored affected trees.   However, it was not until 1925 and after that copper was fully accepted as an essential element for plants.   This acceptance was preceded by the research of McHargue (1925) on copper distribution in plants and animals, Bortels' work (1927) with Aspergillus niger, the work of Felix (1927) and Allison, Bryan, and Hunter (1927) on responses of various crops to copper additions to peat soils, and the more exact work of Sommer (1931) and Lipman and MacKinney (1931).   Haas and Quayle (1935) were probably the first to produce copper deficiency symptoms on citrus.   Valencia orange trees on sour rootstock were grown in large outdoor sand cultures from 1920 to 1928.   By 1926 the trees which had been supplied with nutrient solutions lacking copper developed the symptoms then described as exanthema.   Additions of copper sulfate corrected the condition on all but one of the trees.   During this same period, somewhat comparable observations and research led to the conclusion that zinc, boron, and manganese were also essential for plant growth.   Thus, much of the previous evidence fell into place.   Wade (1944) later produced copper deficiency of citrus in controlled water cultures, and Arnon and Stout (1939a) produced copper deficiency in tomato.   Copper deficiency in plants has been widely reported.   Good reviews of copper in relation to plant and animal nutrition are given by Gilbert (1952), McElroy and Glass (1950), Reuther (1957), Reuther and Labanauskas (1966), and others.
      Symptoms of Copper Deficiency.—The growth, foliar, and fruit symptoms characterizing copper deficiency of citrus have been well described by Camp, Chapman, and Parker (1948, p. 314) as follows:

       The first evidence of mild copper deficiency is the occurrence of unusually large, dark-green leaves on long, soft, angular shoots, the leaves being commonly of irregular contour, usually with a "bowing up" of the midrib.   The soft twigs sag at the tips or become S-shaped.   In this stage the tree appears to the casual observer as unusually vigorous, although in California this excessive growth is not so prominent.   When the deficiency is more acute, very small leaves may develop and quickly shed on twigs that are going to die back, but on the older wood the leaves will be large, dark green, and somewhat twisted or malformed.   This peculiar twisting and malformation of foliage is particularly common on copper-deficient grapefruit trees.   In very acute cases the leaves may be greatly distorted, the margins irregular, and the color light green with a fine network of darker green veins.   In such cases the growth of the twigs is restricted, fine, and very angular.
       Following the appearance of the initial symptoms, the affected twigs usually show a development of multiple buds.   These produce a dense, somewhat bushy growth, particularly in lemon trees of moderate vigor.   Occasionally gum pockets develop between the bark and the wood.   These may rupture the bark and permit the gum to exude.   In dry weather the gum collects and dries on the surface of the bark.   However, it is readily soluble in rain water and frequently is overlooked for that reason.   As the deficiency becomes more acute, new growth develops and then dies back for several inches.   In very acute cases, heavy twigs having multiple buds put out a profusion of young soft shoots with small leaves, which quickly die back from the tips.   In this stage the twigs have a reddish excrescence over a large portion of the bark.   Neither the dying back nor the reddish excrescence is so pronounced in grapefruit as it is in oranges, and it is still less obvious in tangerines.   As the acute stage is reached, there is a pronounced loss of growth due to dying back, and in very serious cases the tree may be almost killed.   The dying starts on the outer shoots of the tree; soon the characteristic large soft shoots appear as water sprouts on large branches in the center of the tree, and these in turn develop the characteristic symptoms.   In acute stages gumming has been found in the roots also, and considerable loss of roots takes place.
       In cases of severe copper deficiency, the fruit is marked irregularly with reddish-brown excrescences, which are light-colored on young fruit but progressively darken until they may be black on mature fruit.   Young fruits are sometimes bumpy and generally have an unusually light green color and a very smooth skin with or without the light reddish-brown excrescences.   By June some of the fruits will be almost covered with these excrescences and drop from the tree.   Such fruits as are left usually have juice low in acid and very insipid, and the pulp dries out early in the season.   In such acute cases there are gum pockets in the rind and gum at the axes of the segments; splitting of young fruits is common and includes both the ordinary longitudinal splitting with starts at the stylar end and transverse splitting which starts in an excrescence and extends part way around the fruit.
      The symptoms described are typical of oranges.   In grapefruit, the excrescences on the peel are less common, and though gum pockets are numerous in the rind, gum is seldom found around the seeds.   Fruits from copper-deficient grapefruit trees are commonly yellower than normal and are frequently lopsided, and as they mature brown pits develop in the rind that are similar to storage pits but smaller.   Acute copper deficiency is rather uncommon in grapefruit and seldom observed in tangerines.
       Acute deficiency of copper may put trees entirely out of production, while in the intermediate stage there is a reduction in yield and also considerable loss due to lowered grade of the fruit.
       Copper deficiency tends to reduce the soluble-solids content and also the acid content of orange juice.   Since the reduction in acid tends to be disproportionate to the reduction in solids, the fruit from deficient trees tends to have a higher ratio of solids to acid.   The vitamin C content of the juice is also reduced, more or less in proportion to the reduction in acid.   It is interesting to note that this reduction in solids and acid and also vitamin C varies with different varieties as it relates to the occurrence of other symptoms.   The Hamlin variety of orange may show a very high reduction in solids and acid with very few of the visible morphological symptoms, whereas in other varieties the extent of the symptoms more or less parallels the reduction in the soluble constituents of the juice.
       Zinc deficiency is commonly associated with copper deficiency in Florida.   This results in some modification of the growth symptoms owing to the growth-restraining characteristics of zinc deficiency.   There is also considerable evidence to indicate that copper deficiency restrains absorption of zinc by the roots, so that in acute cases of copper deficiency, symptoms of zinc deficiency almost always develop even though the amount of available zinc in the soil is adequate for healthy trees.

      Table 3-18 summarizes some of the foliar, growth, and fruit symptoms of copper deficiency and includes tissue analysis values.
      Illustrations of various copper deficiency effects on fruits, shoots, and foliage are shown in figures 3-7, 3-8, and 3-9.
      Causes of Copper Deficiency.—In Florida, copper deficiency occurs prominently on leached, acid sandy soils, on peat soils, or even on those soils high in organic matter.   It is common in young plantings on virgin soils.
      In California, the condition (while much less prevalent than in Florida) has been noted on alkaline sandy soils and on heavy soils rather high in organic matter; also, along with severe zinc deficiency, it has been noted in old corral sites and Indian burial grounds.   This may be due to the accumulation of phosphorus, since Bingham et al. (1958) showed that heavy phosphate fertilization can induce copper deficiency.   In field phosphate experiments, Reuther et al. (1949) found that foliage levels of copper were reduced by heavy phosphate treatments.
      In Australia, copper deficiency has been noted on sands, sandy lateritic and sandy calcareous soils, and acid muck soils (Teakle and Stewart, 1939; Teakle, 1942).
      Excessive nitrogen fertilization has been considered as a cause of copper deficiency, especially in Florida.   Insufficient copper in the soil or a supply which is quickly exhausted by cropping is considered a prime cause of the formerly widespread copper deficiency of citrus on the acid sands of Florida.   Bryan (1958) reported that, in Florida sandy soils, from 3 to 5 ppm of total copper are sufficient for citrus where phosphate is high and that 1 ppm is ample in low phosphate soils.   He further stated that 10 ppm of copper are excessive.   Dr. P. F. Smith of Florida, in private correspondence with the author, stated that virgin Florida soils vary from 2 to 15 ppm total copper, and that toxicity begins around 100 ppm and over.   Copper forms highly insoluble compounds with organic matter and is tenaciously fixed by clays in forms not readily replaceable.
      Liebeg et al. (1942) found, in controlled culture work, that aluminum reduced copper toxicity, and, by the same token, soil conditions could obtain (e.g., acid soils) where copper deficiency might be produced by soluble aluminum.   No doubt many other specific interrelations will come to light as knowledge in this field advances.

Copper Excess
      The toxicity of copper to microorganisms has been known for many years, and copper has been used in various formulations as a fungicide.   Bordeaux spray originated in 1882 near Bordeaux, France.   The toxic effects of copper on higher plants became known before the turn of the century.   (For references on much of this early work, the reader is referred to the Bibliography of the Literature on the Minor Elements, published by the Chilean Nitrate Educational Bureau, Inc., New York, and to Forbes, 1917).
      In general, copper excess problems have arisen chiefly from the excessive use of Bordeaux sprays or through soils, streams, and irrigation waters becoming enriched or contaminated by copper mining operations.
      In Chile, the author saw some areas where copper injury had developed from mining operations; also, some streams were pointed out that contained elevated copper concentrations as a result of water percolation through copper-bearing rocks.
      As far as citrus is concerned, toxic effects to the tops from repeated applications of Bordeaux sprays have been commented upon by Fawcett (1936, pp. 358-59).   He noted that injury was more likely to occur from repeated sprayings in summer, especially in interior valleys, and that affected trees gradually lost many leaves, though no spotting or burning took place.   Yields were decreased and many small twigs died.   However, no permanent injury resulted.   Scale-insect infestations increased, probably as a result of the killing of fungus parasites.
      The use of Bordeaux sprays, followed too closely by hydrocyanic acid fumigation, has often been noted to result in injury.   In these cases, severe defoliation may take place.   Various kinds of leaf burn may occur, such as burned spots or death of the entire blade; there also may be only a spotting, sometimes referred to as "stellate melanose" (Fawcett, 1936, p. 552).   During the period when hydrocyanic acid fumigation was widely used, it was always recommended that at least six months elapse between spraying with Bordeaux and fumigation.
      There has been some reason to believe that smog components can solubilize the copper in Bordeaux residues on foliage and bark and cause injury and that moisture condensation on leaves, perhaps coupled with carbon dioxide formation from honeydew, may solubilize copper sufficiently to cause injury.   Discussion of these and other possibilities has been presented by Klotz, Calavan, and DeWolfe (1956).   These investigators recommend the incorporation of zinc sulfate to offset copper spray injury in areas subject to such trouble.   A formula consisting of 3 pounds of zinc sulfate monohydrate (ZnSO4 · H2O), 2 pounds of copper sulfate (CuSO4 · 5H2O), and 6 pounds of hydrated lime (Ca(OH)2) is satisfactory.
      Another type of injury resulting from copper accumulations in the sandy acid soils (pH 4 to 5.5) of Florida—after repeated use of Bordeaux sprays and soil applications of copper sulfate—has been described by Reuther and Smith (1953, 1954) and Reuther (1957).   Chlorotic foliage of low-iron content, marked reduction in vigor and yield, and dieback of twigs were some of the effects on the above-ground parts of the tree.   Fibrous roots growing in the zone of high-copper content were sparse, dark, and stubby.   Such soils commonly contain from 100 to 400 ppm of total copper in surface layers, as compared with values of less than 5 ppm in soils where Bordeaux sprays or soil applications have never been used (Reuther et al., 1951; Reuther, Smith, and Scudder, 1953; Wander, 1954).
      Iron chelates (e.g., FeEDTA) at rates of 1/4 to 1/2 pound per tree, coupled with lime to increase alkalinity to about pH 7, will correct the condition.   The lime reduces copper solubility, and the chelate furnishes iron to the nonaffected roots in lower soil horizons.   Phosphate or lime will also decrease copper toxicity.
      In culture work, Forbes (1917) noted that citrus seedlings placed in distilled water to which 2.5 ppm of copper were added wilted within 48 hours.   Chapman, Liebig, and Vanselow (1940) and Liebig et al. (1942) noted that excessive copper in nutrient solutions sometimes produced iron chlorosis.   The latter investigators found that as little as 0.1 ppm of copper in the nutrient solutions markedly affected the roots and tops.   Swollen, short-root laterals and brown color characterized the roots of the copper-injured plants.   The injured fine roots showed up to 890 ppm of copper, as compared to less than 100 ppm in healthy roots.   It was found that aluminum at concentrations comparable to the copper would offset the toxic effect of the copper (see fig. 3-10).
      Smith and Specht (1952, 1953) produced iron chlorosis in citrus seedlings from additions of copper (0.1 ppm) to plants grown in vermiculite.   Higher concentrations of zinc (3 ppm) and manganese (5 ppm) also produced iron chlorosis.   Root injury occurred in all cases.
      In short-term, solution-culture experiments with wheat, Blevins and Massey (1959) found that 0.4 to 1 ppm of aluminum decreased copper uptake, but that 0.1 ppm of aluminum appeared to increase copper absorption.   Copper concentration in the nutrient solution was 0.1 ppm.   In pot-culture experiments with soil, an inverse correlation was noted between soil aluminum, extracted by 0.1N sodium chloride, and copper uptake by millet.
      In work with corn, Willis and Piland (1936) found that copper sulfate added to a peat soil produced iron chlorosis.
      Neither plant appearance nor leaf analysis can be relied upon as infallible guides to copper excess effects.   Table 3-19 summarizes some of the principal effects of excess copper on citrus trees and roots.   Unlike boron and some other elements, copper does not always accumulate in sufficient amounts in leaves to afford a reliable indication of copper toxicity; but in cases of suspected injury, leaf analyses should be made.
      Where soil contamination is suspected, analyses of both rigorously cleaned fine roots and soils should be made.   Total copper concentrations of 50 to 100 ppm or over in the soil should be regarded with suspicion.   Copper levels of over 15 to 20 ppm in leaves should also be regarded as indicating possible copper toxicity.   In pot-culture work with citrus seedlings, Reitz and Shimp (1954) noted that decreased growth from additions of either copper oxide or copper sulfate was associated with over 20 ppm of copper in the dry matter of the leaves.
      Roots injured from excess copper may show from 400 to more than 900 ppm of copper in the dry matter (Liebig et al., 1942; Smith and Specht, 1952, 1953).
      In recent experiments by the author (unpublished) with young navel orange trees on Troyer rootstock, copper in leaves up to 18 to 20 ppm in the dry matter was associated with marked yield reductions though tree appearance and growth were not appreciably affected.   High calcium concentration (from gypsum) reduced copper intake, and at leaf levels of 10 to 12 ppm of copper, good crops of fruit were matured.


      There is no evidence that fluorine is essential for citrus or any other green plant, but its wide distribution in soils, plants, and waters, and its effects on plants and animals have been the subject of special study over many years.1

Fluorine Excess
      Possible fluorine toxicity to citrus first came to the author's attention in 1951, when leaf-scorch of apricot trees, comparable to that described by DeOng (1946) from fluoride deposits on leaves, was found near a steel plant in San Bernardino County, California (McCornack et al., 1952).   Citrus growers in the area became alarmed over excessive leaf drop and poor yields.   Preliminary analyses of citrus leaves from trees near the steel plant showed much higher fluorine than did leaves from trees remote from industrial contaminants.   This led to an extensive leaf analysis survey of citrus by Kaudy et al. (1954, 1955).   Values up to 211 ppm of fluorine in the dry matter of citrus leaves were found in orchards near the steel plant, as compared with values of less than 10 ppm in areas away from industrial centers.   Investigations by Brewer et al. (1959, 1960a, 1960b) under controlled conditions were immediately begun to assess the importance of fluorine to citrus performance.   Meanwhile, in Florida, the investigation of a type of leaf chlorosis occurring on citrus trees in the vicinity of phosphate manufacturing plants by Wander and McBride (1956) disclosed that the cause was fluorine air pollution, and leaf values up to 370 ppm of fluorine in the dry matter of the leaves were found.   In areas away from these plants, the fluorine content ranged from 12 to 30 ppm.   Spray applications of 0.1N hydrogen fluoride or fluosilicic acid produced a similar type of chlorosis.
      Actual field injury of citrus from excess fluorine has been reported only in Florida and California, and in both areas it is from atmospheric pollution.   This is true for plants generally.   Soil fluorine (occurring naturally or resulting from contamination by waters, fertilizers, or the atmosphere) is relatively insoluble and therefore not appreciably absorbed by plants.
      The main sources of fluorine air contamination are: (1) aluminum reduction plants, (2) iron and nonferrous smelting plants, (3) ceramic industries, and (4) phosphate fertilizer plants.   Naturally occurring fluorine or the fluorine in cryolite added in smelting and reduction operations is driven off by high temperatures; in the case of phosphate rock, treatment with acid releases fluorine contaminants.   The chief volatilization products are hydrogen fluoride and fluosilicic acid.
      The amounts of fluorine in the air, based on limited numbers of air analyses, vary in industrial areas from 0.1 up to 147 parts per billion (ppb), depending on the industry in question and proximity to the industry.
      Plants vary enormously in their sensitivity to fluorine.   Some sensitive plants (such as gladiolus and apricot) may be injured when the atmosphere contains less than 1 ppb of this element.   Plant foliage generally tends to accumulate fluorine both as a surface contaminant and internally, but there are great differences in tolerance for reasons not yet understood.
      Citrus is quite tolerant to fluorine, as compared with sensitive plants such as gladiolus, apricot, peach, some grape varieties, and Ponderosa pine.   But there are also differences in citrus species and variety susceptibility.
      Symptoms of Fluorine Excess.—Aside from a type of chlorosis that closely resembles the excess boron mottling described by Wander and McBride (1956) and shown in fig. 3-11, all that is known of the growth and symptomatology effects of excess fluorine has emerged from controlled nutritional and fumigation experiments.
      In a water-culture experiment with four-year-old navel orange trees growing outdoors, Brewer et al. (1959) found that the addition of 100 ppm of fluorine to the aerated solutions resulted in leaf wilt within 24 hours.   This was followed by complete defoliation.   Another group of trees, given 25 ppm of fluorine in the nutrient solution, was grown along with control trees for 18 months.   No specific leaf symptoms developed, but, compared with the controls, growth was reduced, spring leaf drop was more severe, leaf sizes were smaller, and the trees eventually became quite weak and sparsely foliated.   Leaf fluorine in spring-cycle leaves collected over a period of months showed values up to about 60 ppm (dry-weight basis), compared with less than 10 ppm in the leaves from control trees.   At the conclusion of the experiment, samples of leaves, branches, trunk, roots, and fruit were analyzed from both sets of trees.   The results are reproduced in table 3-20.   Some elevation of fluorine occurred in all parts of the tree, but that in the leaves and small roots was greatest.   No doubt the very high values in fine roots were due, in part, to surface contamination of insoluble calcium fluoride.
      In a sand-culture experiment with lemon cuttings (Haas and Brusca, 1955a), 400 ppm of fluoride produced tip burn, chlorotic leaf patterns resembling manganese and iron deficiencies, reduced leaf sizes, and reduced growth.   No data on plant composition were given.
      Fumigation of various citrus varieties under controlled greenhouse conditions by Brewer et al. (1960a, 1960b), using hydrogen fluoride gas at concentrations in the air varying from 2 to 12 ppb, reduced growth rate and leaf sizes and produced various types of leaf injury.   The injury often began with marginal yellowing, tip and marginal burning, and chlorotic patterns resembling manganese deficiency and boron excess (see fig. 3-11).   Injured leaves dropped prematurely, and this was frequently followed by multiple new weak growth emerging from axillary buds.   Leaves showing the burn and chlorosis (fig. 3-11) usually contained more than 200 ppm of fluorine in the dry matter.   Thomas (1961) noted that with many plants that fluoride absorbed by leaves is translocated to the margins and tips of leaves with gradients of 2:1 to 100:1 between the concentrations in the tips and margins as compared to that of the rest of the leaf.   He stated that appreciable translocation from leaves to other parts of the plant does not occur.
      Brewer et al. (1960a) noted species and variety differences in tolerance to fluoride, lemon and grapefruit species being somewhat more sensitive than orange species.
      A summary of foliage, growth, fruit, and composition effects of fluorine on citrus is presented in table 3-21.
      Control of Fluorine Excess.—From the available evidence, it appears that fluorine injury to citrus (and most plants, for that matter) is likely to occur only from air pollution and only when citrus is located within a radius of ten to twenty-five miles from air pollution sources.
      With some plants, lime dusts and sprays have given protection against damage from fluorine air pollution.   In the case of citrus, insufficient information is available as yet as to the effectiveness of lime.
      In soils, the only likely circumstance where excess soluble fluorine would occur is under acid conditions.   Therefore, the simple solution would be to neutralize the acidity with limestone.


Iron Deficiency
      The essentiality of iron for plants has been known for over one hundred years.   Gris (1843) is said to have been the first to identify lime-induced chlorosis as an iron problem.   He was able to green grape leaves by the use of iron sprays.
      A number of reviews have been written dealing with one or another aspect of the occurrence, nature, cause, and control of iron deficiency.   The reader is referred to the following works for a broader discussion of the subject than is possible here: Brown (1956, 1961); Brown, Holmes, and Tiffin (1959); Bear (1957); Thorne, Wann, and Robinson (1951); Wallace and Lunt (1960); Wallace (1962); Nicholas (1961); and Evans (1959).
      The distinction between various types of citrus leaf chlorosis caused by lack of manganese, zinc, iron, and magnesium did not become clear until the range of patterns of these disorders was delineated by Camp and Fudge (1939) and Chapman and Kelley (1943).   The visual range of foliage, growth, and fruit effects of both acute and mild deficiencies is now well known, and much has been learned about the multiple causes of this disorder.
      While iron chlorosis is less prevalent and more spotty in its incidence than zinc deficiency, it is reasonably widespread throughout the world and a problem of great commercial importance.
      Symptoms of Iron Deficiency.—The predominant leaf pattern produced by the lack of iron (most pronounced on terminal shoots) can best be described as a network of fine green veins against a lighter-colored background; symptoms range in severity from a barely distinguishable venation to one in which the leaf becomes almost totally yellow or ivory-colored, with perhaps just a tinge of green in the basal part of the midrib.   At some stages (in young growth particularly), it is difficult to differentiate the symptoms from manganese deficiency, and often both disorders occur in the same tree simultaneously.   In such cases, one can best judge the situation by looking only at mature leaves.   With iron deficiency, the network of green veins continues to be sharply delineated from the lighter-colored areas between the veins as the leaf matures, whereas with manganese and zinc deficiencies a fringe or band of green develops along the midrib and main veins, producing a more mottled appearance.   As stated, in severe deficiency the leaves may appear almost completely cream-, ivory-, or yellow-colored, and sometimes they become tinged with red or brown; acute zinc deficiency can produce comparable symptoms, though it is fairly common in this case to see small green dots in mesophyll areas.
      Photographs illustrating typical leaf patterns on lemon and orange leaves are reproduced in figures 3-12, 3-13, and 3-14.   When one is in doubt about the diagnosis, the easiest way to confirm it is to paint or dip individual leaves with a solution of iron sulfate and then tag them for identification.   A 6.5 per cent ferrous sulfate (FeSO4 · 7H2O) solution can be used.   This solution contains about 1,000 ppm of iron.   If the condition is due to a lack of iron, partial greening of the leaf or a speckling of green spots will appear.   It usually requires one to several weeks for this to occur, depending on the time of year and the leaf age (see figs. 3-13 and 3-14).
      Leaf analyses for total iron may not always be conclusive, though leaf levels of 10 to 35 ppm of iron in the dry matter, associated with iron chlorosis, would constitute fairly conclusive evidence that the condition was due to lack of iron.   According to Wallihan (1966), the very earliest stages of iron deficiency may merely reduce leaf size and not produce a chlorosis.
      Acute iron deficiency results in a progressive dieback of trees, giving them a brushy, open appearance.   Often one limb is affected more than others, and the tops are more affected than the lower parts.   Leaves do not reach normal size, and there is premature abscission of affected foliage.   Fruit production is curtailed, and, even where the condition is mild, less mature fruits are to be found on mildly iron-deficient terminals than on healthy green ones.
      Fruit quality does not appear to be severely affected.   Generally, the fruit on trees affected with this disorder is smaller and tends to be smoother than fruit from healthy green trees.   Total solids are reduced, but acidity increases somewhat.   Mature oranges may be less highly colored.
      Leaf Analysis Values.—As far as leaf composition is concerned, there is good evidence with citrus, if leaves are thoroughly cleaned with detergent and/or weak acid (e.g., 0.3N HCl), that a fairly good correlation exists between iron content and degree of chlorosis.   Smith, Reuther, and Specht (1950) found that mature citrus leaves, thoroughly cleaned with a detergent solution (sodium lauryl sulfate), showed the following values on a dry matter basis:

Degree of Chlorosis                                              Total Iron (Ppm)
Nonchlorotic leaves                                                    35 to 100
Mildly chlorotic leaves                                                18 to   30
Severely chlorotic leaves                                             10 to   15

      Wallihan (1955) got a fairly good correlation between degree of chlorosis and iron content on orange, lemon, and grapefruit leaves.   He washed each leaf carefully in a soap solution and then thoroughly rinsed it.   The data were as follows:

                                                                             Iron in Dry Matter
      Degree of Chlorosis                                                (Ppm)           
  No chlorosis                                                              42 to 137
  Moderate chlorosis                                                    32 to   68
  Severe chlorosis                                                         24 to   59
  Extreme chlorosis                                                       16 to   33

      Kuykendall (1955), in a study of the iron content of chlorotic and nonchlorotic citrus leaves, found less iron in chlorotic leaves than in green leaves.
      Jacobson (1945) got a good correlation between the degree of iron chlorosis and both total and acid-soluble iron in pear, corn, and tobacco leaves when he thoroughly cleaned the leaves in 0.3N HCl.   Iron extracted from dried, ground-leaf material by 1.0N HCl showed a good correlation with chlorophyll extracted from the fresh leaves.   In iron-chlorotic pear leaves, the iron content was always under 40 ppm on a dry matter basis.   Deep-green pear leaves showed total iron values ranging from 39.2 to 79.5 ppm.   For corn and tobacco, the iron values in both chlorotic and green leaves ran higher than in pear, but there was still a good correlation between iron content and degree of chlorosis.
      As stated, deficiency levels of iron in citrus leaves usually range from 10 to 40 ppm iron in the dry matter, though there is some overlapping.   The leaves of some top-producing, nonchlorotic orchards were found by Bradford and Harding (1957) to be in the 30- to 40-ppm range; conversely, some iron-deficient citrus foliage was found by Wallihan (1955) to show as high as 68 ppm of iron.
      It is well known that iron is much less mobile in green plants than any of the other essential elements, and there is considerable evidence to indicate that a variety of internal conditions, such as high phosphorus, low potassium, low magnesium, and probably other nutritional imbalances, can reduce iron availability.   Thus, close correlations between iron levels in the leaf and degree of chlorosis are not to be expected.
      A summary of the effects produced by iron deficiency on citrus is given in table 3-22.
      Causes of Iron Deficiency.—Iron deficiency in citrus (and in plants generally) can result from many causes.   This appears to stem, in part, from the very low solubility of iron in soils, its relative immobility in plants, and the fact that luxury consumption and/or translocation and reuse does not go on to the same degree as with many of the other nutrients.   Iron deficiency may also result from the multiple roles of iron in plant metabolism and the probability that unbalanced nutritional conditions of many kinds can interfere with one or another of these functions.
      Under slightly acid, neutral, and alkaline soil conditions, it is certain that a part of the plant's iron supply comes from numerous localized root-soil particle zone contacts.
      At pH 4, Wallace and Lunt (1960) estimated the solubility of ferrous hydroxide as 0.003 ppm of iron.   Data supplied to the author by Dr. H. Bohn, Postdoctoral Fellow, Department of Soils and Plant Nutrition, University of California at Riverside, are of interest.   This investigator analyzed soil solutions which had been equilibrated with 0.01M calcium chloride for four months.   Twenty grams of soil were placed in one liter of 0.01M calcium chloride and saturated with air for four months prior to analysis.   The iron content of the solution phase at various soil suspension pH values was as follows:

  Soil Suspension (pH)                             Iron in Solution (Ppm)
              3.88                                                         0.04
              4.43                                                         0.007
              4.70                                                         0.002
              4.78                                                         0.002
              4.78                                                         0.003
              5.14                                                         0.003

      Oertli and Jacobson (1960) found that plants could not get enough iron from maintained solution concentrations comparable to those existing in soils.
      The low solubility of many iron compounds may also be a factor in the immobility of iron within leaves and the ease with which its transport in the vascular system is upset.   The extent to which naturally occurring organic chelates of iron are present in soils and in plants and their relative importance in iron nutrition, generally, is little known at present.
      The ability of plants to extract iron from highly insoluble sources was first demonstrated by Eaton (1936), who found that 0.1 per cent of ground magnetite (incorporated in the acid sand cultures) made the addition of soluble iron compounds unnecessary.   In a continuing study using magnetite, Chapman (1939) showed that orange seedlings growing in a sand-magnetite mixture were able to extract iron from magnetite through a root-iron particle contact mechanism, and that additions of calcium carbonate to the sand-magnetite culture markedly reduced iron extraction.   By increasing the magnetite supply, the effect of calcium carbonate could be overcome.   The sand cultures in this experiment were periodically flushed with a nutrient solution from a common reservoir, using a time-clock-controlled, compressed-air pumping arrangement.   It was found that frequent or continuous flushing of the calcium carbonate-magnetite-containing sand cultures brought on iron chlorosis and an increase in alkalinity of the circulating nutrient solution (from pH 7.3 to 8.2).   This was the result of the continuous aeration of the circulating nutrient solution by the compressed-air-ejector-type pumping arrangement, and the resulting removal of carbon dioxide produced by the plant roots.   Iron solubility in the nutrient medium was reduced from 0.01 ppm iron at pH 5.9 to less than 0.005 ppm at pH 8.1.   The pH change in the periodically circulating culture solution no doubt influenced the pH in the intimate zones of root-iron particle contact and prevented the roots from securing adequate iron from the magnetite.   Guest (1944) confirmed some of these findings.
      In somewhat parallel experiments, using alfalfa as a test plant, Jenny (1961) and his coworkers, Glauser and Jenny (1960a, 1960b), Grunes and Jenny (1960), and Charley and Jenny (1961), confirmed and extended some of the aforementioned findings and showed that:
      1. In calcareous, alkaline media the rate of iron uptake by young alfalfa plants is directly proportional to the number of solid, iron oxide grains which touch the root surface.
      2. Hydrogen-ion exchangers (e.g., amberlite and amberplex membranes and roots) readily decompose iron oxide particles by contact interaction.
      3. The iron ions acquired at the outer surface of amberplex membranes and root plugs migrate through these porous bodies by exchange diffusion.
      Wynd (1951) found that wheat plants could extract iron through a root-contact mechanism from a glass frit containing 5 per cent ferric oxide.
      The presence of considerable magnetite and other iron compounds in soils and their low solubility, on the one hand, together with the prevalence of iron chlorosis on some calcareous soils and not on others, the seasonal fluctuations in iron chlorosis, effects of excess moisture, low temperatures, and lack of oxygen, all fit in with the idea that in many situations root-soil particle contact exchange is one of the important mechanisms by which plants secure needed iron from soils.
      A brief discussion of the factors known to produce or aggravate iron deficiency in citrus and other plants follows:
      Calcareous soils (lime-induced chlorosis).—Excess lime often produces severe iron chlorosis in citrus and most other plants.   While many explanations have been advanced, the most reasonable is the low solubility of iron under the pH conditions which calcium carbonate produces in soils and the buffering effect of calcium carbonate.   The calcium bicarbonate produced in such soils may possibly have some effect on both the iron absorption and its mobility within the plant.   But this explanation does not appear entirely plausible, since iron-deficient leaves usually show lower-than-normal calcium levels (and higher potassium).   Also, Chapman (1939) and Glauser and Jenny (1960a) found that the effects of calcium carbonate could be overcome by the addition of more magnetite or iron-coated sand grain to the rooting medium.   The fact that not all soils containing free lime produce iron chlorosis can be explained by any one or a combination of conditions, including the amount, form, and particle sizes of the lime and iron compounds present; variables in the organic iron content of the soil; and aeration, temperature, moisture, and biological-activity fluctuations.
      Overmoist soil conditions and aeration.—There is abundant field and pot-culture evidence that persistent overmoist soil conditions will produce iron deficiency in citrus and many other plants.   In fact, this is one of the most common causes of, or contributing factors to, iron chlorosis of citrus in California.   Until the advent of chelates, the most effective way to correct iron chlorosis of citrus was to modify irrigation practices by various means, such as alternate middle irrigation, changing irrigation frequency, and the guidance of irrigation by the use of tensiometers.   All of these means were designed to reduce or obviate persistent overmoist conditions.   The simplest explanation of the effects of too much moisture is lack of adequate aeration.   Wallihan et al. (1961) have recently shown in pot experiments with porous soil that, by reducing the oxygen content of the atmosphere over the surface of the soil from 18 to 21 per cent to 2 to 6.2 per cent, the growth and iron content of leaves was reduced from 64 to 37 ppm of iron, using a soil of pH 6.2, and from 48 to 27 ppm of iron where calcium carbonate had been added to the soil.   In the presence of calcium carbonate and low oxygen, the leaves showed iron chlorosis.   Low oxygen, of course, will reduce respiration rate and production of carbon dioxide, and thus also reduce the degree of acidity and its solubilizing influence on iron in solid-phase forms.
      Higher moisture in soils also will slightly increase hydrolysis of base-carrying minerals and thus slightly increase the alkalinity of the soil.   This increase may be just enough in some cases to make the difference between iron sufficiency and insufficiency.
      Under high moisture conditions, more manganese is solubilized, and this may be a further contributing cause to the prevention of iron absorption by, or translocation out of, the root.   Changes in soil organism populations and activities may also be brought about by high moisture conditions.
      Bicarbonate.—There is substantial evidence that the bicarbonate ion, if present in sufficient amounts, can bring on or aggravate iron deficiency.   In sand cultures containing 1 per cent powdered magnetite as a source of iron and periodically flushed with a complete nutrient solution from large reservoirs, Pearson and Chapman (1964, unpublished data) found that 30 meq of sodium bicarbonate added to the nutrient solution produced severe iron chlorosis of citrus plants, whereas comparable amounts of sodium chloride, sodium sulfate, and sodium nitrate did not cause iron chlorosis.   The pH of the high sodium chloride, sulfate, and nitrate cultures varied between pH 5.5 and 6, but those containing the sodium bicarbonate ranged from pH 8.2 to 8.5.   No doubt, the higher pH and buffering capacity of this solution made it impossible for the citrus roots to create sufficiently acid root-iron particle contact zones to extract needed iron.   The bicarbonate, of course, as will be mentioned later, may have exerted other effects.
      In five-hour absorption period experiments with radioactive iron (Fe59), Wallihan (1961) showed that 10 meq of sodium bicarbonate added to a nutrient solution (complete except for phosphate), prevented radioactive iron from being taken up into the stems and leaves of sweet orange seedlings and greatly reduced the radioactive iron found in the roots.   The control cultures contained 10 meq of sodium sulfate.
      All irrigation waters carry some bicarbonate.   Depending on the relative amounts of sodium, calcium, and magnesium in such waters, there may be deposition of both sodium bicarbonate and calcium and magnesium carbonate in irrigated soils.   Acid or neutral soils will become alkaline, and iron-bearing minerals may be coated with insoluble carbonate.   Broadbent and Chapman (1950) found in a 15-year outdoor lysimeter experiment that the application of a rather low-salt irrigation water (347 ppm of total solids), of which 2.94 meq (181 ppm) were bicarbonate, changed soil pH (paste) from a value of about 6.8 to 8.5 and deposited several tons of calcium carbonate per acre.   The rate of irrigation water usage was about 26 acre-inches per year.   Harley and Lindner (1945) found that the use of well waters high in bicarbonate (210 to 360 ppm) led to lime deposition in soils and on plant roots and produced iron chlorosis in apple and pear orchards.
      Bicarbonate may raise the pH of plant sap (Wallace, 1962) and thus decrease iron solubility.   Miller and Evans (1956) showed that the cytochrome oxidase activity of root preparations was decreased more by bicarbonate than by chloride, sulfate, nitrate, or phosphate ions.   Many other investigators, working with a variety of plants, have noted that bicarbonate in some manner adversely affects iron absorption and/or translocation or utilization in the plant.   (For a more complete review of other investigations on this subject, the reader is referred to the citations mentioned [above]).
      Nutrient imbalance.—Excesses and deficiencies of both macroelements and microelements can cause iron chlorosis in citrus and other plants.   A brief discussion, element by element, follows:
      Potassium.—In outdoor water cultures with bearing citrus trees, the author and his colleagues have repeatedly noted persistent iron chlorosis patterns on trees acutely deficient in potassium.   Iron sprays applied to these leaves caused the familiar partial greening, and leaf analyses (Chapman, Brown, and Rayner, 1947) showed an iron content of 35 ppm in affected leaves, as compared with 90 ppm in similar healthy trees.   We have also noted iron chlorosis in trees produced under low calcium and high potassium conditions.
      Phosphorus.—The author has found, under neutral or alkaline nutrient-culture conditions, that high phosphorus can sometimes produce iron chlorosis.   Many investigators (Olsen, 1935; Bolle-Jones, 1955; Biddulph and Woodbridge, 1952; and Brown, Holmes, and Specht, 1955), working with other plants, have obtained similar results.   It appears that this effect is more likely due to iron precipitation within the plant than to interference with absorption by the root and may depend upon associated conditions within the plant; e.g., calcium, copper, pH, and the respective levels of iron and phosphate.
      There is considerable doubt about the field importance of excess phosphorus, however, as copper and zinc absorption or utilization becomes limiting first.   In pot-culture studies with nine California soils, Bingham et al. (1958) found that, in soils varying from pH 4.4 to 7.4, application of monocalcium phosphate (up to 1,800 pounds of phosphorus per acre) did not reduce iron absorption, but markedly reduced copper uptake.
      Calcium.—Aside from the lime effect mentioned, high concentrations of calcium ions are frequently reported as producing iron chlorosis in some plants.
      Magnesium.—In an outdoor sand-culture experiment, where prolonged acute magnesium deficiency was produced in Valencia orange trees, the author noted that secondary iron chlorosis began to appear on new foliage.   This condition could be corrected by iron sprays.   In other experiments (Chapman, 1944, unpublished data) with high magnesium, the addition of 100 meq of magnesium nitrate to lemon cuttings in nutrient cultures produced severe iron chlorosis.   The roots of these cuttings became thickened and stubby.
      Zinc.—In the course of nutritional work with citrus, Chapman et al. (1940) tried to correct zinc deficiency in Valencia orange cuttings grown in sand cultures by the addition of a few grams of zinc dust or zinc oxide.   Instead of correcting the zinc deficiency, marked iron chlorosis developed.   Further research confirmed that excess zinc in the nutrient medium would cause iron chlorosis.   Analyses of leaves, stems, and roots showed that the chlorosis was due to failure of sufficient iron to reach the leaves and a probable tie-up of iron in the roots or the cortex.   Hewitt (1948) produced iron chlorosis in sugar beets when 0.5 to 1 meq of zinc was added to the nutrient medium.   Also, Hunter and Vergnano (1953), Millikan (1947b), and others have produced iron deficiency from the addition of zinc under controlled culture conditions.
      Copper.—In our early work with zinc (Chapman et al., 1940), it was found that excess copper would sometimes produce iron chlorosis in sand cultures.   In Florida, as mentioned earlier in the section on copper, Reuther and Smith (1953) and Smith and Specht (1953) found that copper accumulations in soil from the use of Bordeaux sprays and soil applications of bluestone were responsible for iron deficiency under acid soil conditions.   This subject was extensively investigated.   The findings were that affected citrus showed low iron content in chlorotic foliage, marked reduction in vigor and yields, dieback of trees, and sparse, stubby roots in the zone where copper had accumulated.   Compared with total copper levels of 5 ppm or less in soils where Bordeaux had never been used, values of 50 to 200 ppm copper were found in the soils where such citrus troubles occurred.
      Other workers (Willis and Piland, 1936; Hunter and Vergnano, 1953; and Hewitt, 1951) have shown with other plants that excess copper can produce iron chlorosis.   Brown et al. (1955) found that various copper-phosphorus combinations would cause iron chlorosis in some varieties of soybean.
      Other heavy metals.—Other metals—notably nickel, cadmium, cobalt, chromium, molybdenum, and manganese—have been shown capable of producing iron chlorosis in various plants (Millikan, 1947b; Hewitt, 1948, 1951, 1953; DeKock, 1956; Hunter and Vergnano, 1953; and Brown, 1956).   The effectiveness of these metals varies, and there are great differences between plants in their susceptibility to metal-induced iron chlorosis.   Smith and Specht (1953) found with citrus that manganese could induce iron chlorosis.
      While the mechanisms of these effects are not exactly known, the idea has been put forward that certain heavy metals displace or compete with iron in chelate-protein centers within the roots, and the order of effectiveness in producing iron deficiency is related to the relative stability of the various metal chelates (see Hewitt, 1951; Brown, 1956,1963; and Wallace, 1962).
      Low soil temperatures.—In California, it is a common observation that iron chlorosis of citrus (especially lemons) is aggravated in the wintertime when soil temperatures under 55°F occur and persist for considerable periods.   In an outdoor water-culture experiment with bearing lemon trees currently under way, the author has noted that in spite of the presence of ample iron in the culture solution, iron chlorosis developed on the winter growth cycle.   Leaves painted with iron sulfate became green, thus demonstrating that the chlorosis was due to lack of iron and failure of the roots to absorb or translocate sufficient iron to meet the needs of the leaves.   In such cases, both in the field and in controlled cultures, the chlorosis cleared when the root zone warmed up.
      Similar observations have been made with gardenias (Baudenistel, 1957).   With this plant, FeEDTA soil applications greatly reduced the chlorosis.
      Soil organisms.—It has been a common observation that young citrus plantings on prefumigated soil often suffer less iron, zinc, and manganese deficiencies than comparable plantings on nonfumigated soil.   This may be due to better root growth and feeding power if root-injuring nematodes and soil fungi are not present.   Whatever the mechanism, it is clear that root-attacking soil organisms can cause or aggravate iron chlorosis.   Conversely, soil organisms undoubtedly are important in making iron available to plants through the production of carbon dioxide and organic chelating agents.
      Other factors.—In addition to the foregoing, other soil and environmental factors (such as high light intensity, high soil temperatures, viruses, and high nitrate nitrogen) have been mentioned as involved in the iron chlorosis problem with some plants under certain conditions.
      Effect of Variety and Rootstock.—As with many other plants, there are differences in susceptibility of various citrus varieties to iron chlorosis, and rootstocks also play an important role.
      Lemons, for example, are more prone to chlorosis (particularly that due to low soil temperatures) than are oranges and grapefruit.   Oranges axe more prone to iron chlorosis on trifoliate orange rootstock than on rough lemon rootstock.   Many other plant species and varieties show great differences in ability to extract adequate iron from the soil.
      Control of Iron Deficiency.—Until the spectacular correction of iron deficiency, reported by Stewart and Leonard (1952a), with an iron chelate FeEDTA (iron ethylenediamine tetraacetic acid) in Florida citrus, the use of iron sprays, tree-injection procedures, and massive soil applications of iron sulfate had been only partially successful.   In irrigated areas, as previously mentioned, better control of soil moisture so as to avoid persistent overmoist conditions (achieved by decreased irrigation frequency, alternate middle irrigation, use of green manure crops, better surface drainage, and use of tensiometers to guide irrigation practices) had often been the most practical and effective means of reducing iron chlorosis in citrus.
      The aforementioned results with iron chelates opened up a whole new chapter in iron chlorosis control, not only with citrus, but with plants in general.
      It was soon found that though FeEDTA worked miraculously on the acid soils of Florida, it was not as effective on calcareous soils (Stewart and Leonard, 1957).   This discovery has led to much research to develop more effective compounds for these latter soil types.
      Of the many compounds tried, one described as EHPG or EDDHA-Fe3, and chemically as sodium ferric ethylenediamine di-(o-hydroxyphenylacetic acid) or Chel-138, has proven to be among the best for neutral and calcareous soils.   This compound contains 6 per cent metallic iron, and it is generally used at rates of 1/3 to 1/2 pound per mature tree on light sandy soils, and 1/2 to 1 pound per tree on heavy soils.   The material is uniformly distributed over the soil surface beneath the canopy of the tree.   Following application, the material is thoroughly watered into the soil.   The treatment is applied in the spring, with another application in the summer, if required.   EHPG has given good results at rates of 12 to 24 gm of iron per tree in both the sandy and heavier calcareous soils of Arizona.   Kuykendall, Hilgeman, and Van Horn (1957) and Cooper (1957) found that 5 gm of iron in the form of EDDHA-FE3 and another iron compound RA-157-Fe were highly effective on a calcareous fine sandy loam (pH 7.9) with three-year-old Dancy tangerines on Cleopatra mandarin rootstock.
      With twelve-year-old Redblush grapefruit trees, suffering from iron chlorosis associated with the virus disease cachexia and high soil salinity, 1 pound of EDDHA-Fe3 per tree cleared up the iron chlorosis, and the trees remained green for ten months (Cooper and Peynado, 1959a).
      Kochan (1962), working with severely chlorotic peach trees, secured good correction with EDDHA-Fe3 on calcareous soils when using an injection method.   He devised an injector from a conventional spray gun.   A solution was prepared containing 1 pound of EDDHA-Fe3 per 25 gallons of water.   A spray rig developing 400 pounds of pressure was used to make twelve injections 24 inches deep in a radius 4.5 feet from the tree trunks.   This treatment, applied on May 25, 1961, produced 93 per cent correction of iron chlorosis by June 20, 1961.
      Under the acid sandy soil conditions of Florida, use of FeEDTA at rates of 20 gm of iron per tree has been very successful (Stewart and Leonard, 1957).   Since this compound is water-soluble, leaches away, and in due course decomposes, soil applications need to be made every two to three years.   The material has not proven especially toxic, and the aforementioned investigators found that it required up to 1,000 gm of iron as FeEDTA to produce toxic leaf symptoms.   This material can be mixed with other fertilizers, though its effectiveness may be decreased some by phosphates and lime.
      In comparisons with other chelates for acid soils, FeEDTA has given good performance.   Root growth is increased by FeEDTA (Ford, Stewart, and Leonard, 1954).
      In summary, control of iron chlorosis in citrus can now be achieved in most cases by one or more means, the particular method used depending on the cause or causes.
      Better control of soil moisture will result in improvement where overirrigation is implicated; soil fumigation will often improve iron chlorosis by making for better root growth; leaching of soluble salts will help where salinity buildup, particularly soluble bicarbonate, is a part of the salinity picture; correction of other deficiencies where they occur will help; where excess copper, zinc, manganese, or other metals occur (especially under acid soil conditions), controlled liming to bring the contaminated soil up to pH 6 to 6.5 will help; and, finally, the various chelates already discussed are effective.

Iron Excess
      Except for the leaf-injury effects noted by Stewart and Leonard (1957), when they added up to 1,000 gm of iron as FeEDTA to soils, and injury to leaves and fruit from sprays of iron sulfate and other iron compounds, there has been little investigation of excess iron effects on citrus.
      In plant nutritional work with citrus, the author occasionally has seen a fine speckling of necrotic spots on leaves after the accidental addition of too much iron sulfate to the nutrient solution (see fig. 3-15).   He has also noted an occasional leaf drop on young orange trees in large 700-liter water cultures maintained at pH 4 to which 0.1 ppm of iron as iron sulfate had been added.   Apparently when small plants with an actively absorbing root system are grown in large containers, enough iron can be absorbed from fairly dilute solutions to produce iron toxicity.   Wallihan (1962, unpublished data) also noted this effect, and his analysis of injured leaves showed up to 239 ppm of iron in the dry matter.
      Because of the generally low solubility of iron in soils under the pH ranges and the conditions of culture and soils in which citrus is grown, it is unlikely that iron toxicity will ever prove a practical problem except where excessive amounts of iron chelates are used.
      In submerged rice soils, Ponnamperuma, Bradfield, and Peech (1955) found from 35.8 to 525 ppm of ferrous iron in percolates.   Rice plants showed brown spots on older leaves, and the tips gradually took on a reddish-brown color, which spread toward the leaf base, especially along the edges.   In time, young leaves began to show a brown speckling, and the older leaves dried up.   Roots were coarse and brown.   Symptoms were absent where drainage was good, submergence delayed, or there was a high initial concentration of nitrate in the soil.   This is the only case known to the author of iron toxicity under field conditions.


      There is no evidence as yet that lithium is essential for citrus or green plants generally, but it is widely distributed in small amounts in rocks, soils, plants, and waters.   The result of a recent survey of the lithium in 400 California irrigation waters by Bradford (1963) has shown that 25 per cent of them contain enough lithium (> 0.05 ppm) to be potentially toxic to citrus and other lithium-sensitive plants.

Lithium Excess
      Both stimulating and toxic effects of lithium have been noted by a number of workers in various crops.   Some investigators have found that lithium in sprays and nutrient solutions has produced beneficial effects on certain plant diseases.
      Bradford (1966) reviewed some of the literature on lithium distribution in soils and plants.   Bradford's paper and the Chilean Nitrate Educational Bureau's Bibliography of the Literature on the Minor Elements (1948-55) provide a key to much of the lithium literature.
      Haas (1929a) was the first to produce lithium-toxicity symptoms on citrus in controlled cultures.   Later, Guest and Chapman (1946, unpublished data) applied various lithium salts to lemon trees in the field and produced marked leaf-injury symptoms (see fig. 3-16) and curiously a slight stimulation of tree growth.   To explore further the possible stimulating and toxic effects of lithium under more controlled conditions, Aldrich, Vanselow, and Bradford (1951) set up pot-culture studies with citrus seedlings and produced the same leaf patterns as those illustrated in figure 3-16.   It required only 2 ppm of lithium as lithium sulfate, added to the soil, to produce injury.   In the field only 12.5 gm of lithium as lithium chloride per tree were required to produce toxicity symptoms in ten-year-old lemon trees.   No growth-stimulating effects were noted.
      Analyses of lemon leaves and fruits after lithium chloride had been applied in the field showed lithium concentrations as follows:

     to Soil in                                        Lithium in Dry Matter
18-foot Square                                              (Ppm)                    
  Around Tree                          Mature          Fruit             Fruit
        (Gm)                                Leaves           Peel             Juice         
         12.5                                  30.0              ….                ….
         25.0                                  13.5              ….                ….
         50.0                                  50.0              2.0                0.5

All of the leaves for which analyses are reported showed toxicity symptoms.   Uninjured leaves from areas where lithium is not a problem usually show less than 1 ppm in the dry matter.   Lithium, like boron, accumulates largely in the leaves of citrus trees and increases with leaf age.
      Based on the experiments of Aldrich et al. (1951), lithium-toxicity symptoms were noted by the same authors on field citrus trees in Santa Barbara County, California.   Samples of affected leaves showed the following values for lithium:

                                                                              Lithium in Leaf
              Kind of Citrus                                       Dry Matter (Ppm)
              Lemon                                                            12
              Grapefruit                                                       14
              Navel orange                                                  23
              Valencia orange                                              14

      In the pot work by Aldrich et al., analyses of various parts of sweet orange seedlings showed the following lithium concentrations.

                                                                                        Lithium in Dry Matter(Ppm)                                      
          Lithium Applied              Young               Mature              Top              Top              Bark              Root
            to Soil (Ppm)                Leaves              Leaves              Bark           Wood             Root              Wood                    
                     2.0                         25.0                220.0               25.0             15.0              20.0                5.0
                     5.0                         35.0                140.0               10.0               5.0                8.5                4.5

      The toxicity was traced to lithium in the well waters used for irrigation.   Four wells in the vicinity showed 0.075, 0.045, 0.055, and 0.08 ppm of lithium, respectively.
      From the evidence available, it appears that soils derived from rocks and sediments high in ferro-magnesium minerals are likely to be higher in lithium than soils derived from granites and quartzite.   According to Swaine (1955), total lithium in soils varies from about 8 to 400 ppm.
      In the light of Bradford's (1963) water analysis survey, the possibility of lithium toxicity should be considered in evaluating citrus performance in irrigated areas, especially where waters and soils contain a higher-than-average sodium and boron content.


Magnesium Deficiency
      The essentiality of magnesium for green plants has been known since about 1860.   Magnesium was demonstrated to be a constituent of chlorophyll by Willstätter in 1906.
      Reed and Haas (1924) were the first to produce and describe the effects of magnesium deficiency on young orange trees.   Parbery (1935), working in New South Wales, Australia, described a type of leaf chlorosis on citrus with symptoms corresponding to those described by Reed and Haas (1924).   Subsequently, Chapman and Kelley (1943) produced and described leaf symptoms of magnesium deficiency.   In Florida, Bahrt (1934) reported favorable citrus response to soil applications of magnesium, and further Florida work was soon reported by Tait (1936) and Bryan and DeBusk (1936).   Later, Fudge (1938, 1939) and Camp and Fudge (1939) reported in more detail on magnesium deficiency in Florida.   Good reviews dealing with magnesium nutrition in general are given by Embleton (1966), Uexküll and Kämpfer (1963), Hewitt (1963), and Jacob (1958).
      Perhaps the most severe and widespread deficiency of magnesium in citrus occurred in Florida before the deficiency was recognized as such.
      After identification and correction of widespread magnesium deficiency in Florida, this disorder was noticed in citrus areas all over the world.   In California, mild magnesium deficiency was found rather widely, owing mainly to the accumulation of potassium in soils from the use of fertilizers and manures.   The potassium accumulation in soils resulted in potassium-magnesium ratios unfavorable to magnesium absorption.   Much the same pattern of occurrence has been noted in other citrus areas of the world (Oberholzer, 1941; Averna-Sacca, 1912; Raciti, 1956).
      Symptoms of Magnesium Deficiency.—Leaf patterns and fruit effects of magnesium deficiency are shown in figures 3-17 and 3-18.   A good description of the effects of magnesium deficiency was given by Camp et al. (1949) and is paraphrased in the following paragraphs.
      Symptoms of magnesium deficiency occur in leaves at any season, but they commonly develop in the late summer or fall when the crop is maturing.   The leaves are usually mature and normal in color up to the appearance of symptoms.   Irregular yellow blotches start along the midribs of leaves that are close to fruit and eventually coalesce to form an irregular yellow band on each side of the midrib (see fig. 3-17).   This area enlarges until only the tip and base of the leaf are green, with the base often showing a more or less inverted V-shaped area pointed on the midrib.   This fading from green to yellow does not follow a fixed pattern, apparently being influenced by other nutritional factors and light, but in acute deficiencies the leaves eventually become entirely yellow.   As soon as any considerable portion of the leaf becomes yellow, it may fall if conditions become unfavorable as a result of cold weather, toxic sprays, or other shocks to the tree.   If conditions remain favorable, the yellowed leaf may stay on the tree for a long time.   The defoliated twigs are weak and subject to fungus infection and may die by the following spring, necessitating a severe pruning of trees with a heavy crop.
      Fudge (1939) showed that magnesium deficiency symptoms develop as a result of translocation of magnesium from the leaves to the growing fruit, although there may also be magnesium translocation from older leaves to young developing leaves on the same shoot.   The mobility of magnesium in the tree differentiates it from iron, zinc, manganese, and copper.   The symptoms due to deficiencies of these elements develop coincidentally with the new growth and not on older leaves which previously have been normal in appearance.
      Varieties producing seedy fruit are more severely affected than varieties producing seedless fruit.   In adjoining plots with the same treatment, seedy grapefruit may be almost defoliated while Marsh Seedless trees with an even heavier crop of fruit are only slightly yellow.   A similar phenomenon is common in oranges, seedy varieties like the Pineapple being severely affected, while seedless Hamlins show only very slight yellowing.   The yellowing is also related to the extent of fruiting.   Heavily fruited limbs develop extreme deficiency symptoms.   They may even become completely defoliated, while adjoining limbs with little or no fruit show no symptoms.   Again, this characteristic is most pronounced with seedy varieties.   It may be aggravated by the tendency of seedy varieties to set fruit more unevenly on the tree than do seedless varieties.   The relationships of seediness to magnesium nutrition have been discussed in some detail by both Bahrt (1934) and Fudge (1938).
      Alternation in bearing is common in seedy varieties growing under magnesium-deficient conditions.   This results from the fact that when trees are bearing a heavy crop, the loss of wood in the summer and fall as a result of defoliation reduces the fruit-bearing wood for the next year.   The tree then goes through a season of recovery in which the leaves are green and little or no fruit is produced.   The following year the tree may produce a good crop of fruit on the twigs of the previous year and again defoliate and die back.   This alternation is a function of individual trees; some may bear fruit and others may not in the same year unless the cycle is set up by freeze damage or drought.   Alternation in condition and yield may even pertain to parts of the same tree, certain branches bearing in one year and others in the alternate year.   This mixture of green, healthy-looking trees and yellow trees is very striking, and the occurrence of occasional trees that have one or two branches with yellow leaves and the rest with green leaves, or vice versa, aids greatly in diagnosis of the deficiency.
      Magnesium deficiency results in a reduction in total crop, though in some years there may be large crops which are usually followed by complete crop failures.   In years when there is a crop of commercial size, there is also a reduction in the size of individual fruits, and the fruit is weak and likely to break down in transit.   Sites (1948) and others also have shown that fruit from magnesium-deficient trees is low in soluble solids, total acid, and vitamin C.   In oranges, there is a very marked paleness of color of both pulp and peel.   Magnesium appears to have more influence than any other element on the development of a rich orange color in the peel and pulp.
      Data of Smith et al. (1950) and Sato, Ishihara, and Hase (1963) showed that in common with other nutrients the severity of magnesium deficiency may be influenced appreciably by rootstock.
      Magnesium deficiency, like other deficiencies, makes trees more susceptible to cold injury than normal trees (Lawless and Camp, 1940).   This applies primarily to trees bearing a crop at the time of the freeze, since on the off-year trees actually may not be in a deficient condition.   This susceptibility to cold damage applies to both tree and fruit, although it has not been determined whether damage to the fruit is caused by an actual susceptibility or whether it results from the fact that as a result of defoliation fruits are exposed to the sky and consequently are colder than protected fruit because of radiation losses.   This difference in cold resistance is of great importance to growers; on numerous occasions magnesium-deficient trees in Florida have been observed to lose all leaves, twigs, and fruit, while adjoining trees on the same fertility program plus magnesium lose only a few leaves.
      Some of the major effects of magnesium deficiency in citrus are summarized in table 3-23.
      Cause and Treatment of Magnesium Deficiency.—Magnesium deficiency in citrus in Florida is caused primarily by the low native level of magnesium in the soil and may be aggravated when other bases are added as fertilizers.   The deficiency is particularly acute on the light sandy soils from which magnesium readily leaches.   Leaching of added magnesium is especially rapid as the acidity increases (low pH).   Formerly, most of the citrus soils of this type had a pH of 4.5 to 5, and generally the fertilizer used was both physiologically acid and practically devoid of magnesium.   The present program on such soils includes the use of dolomite to bring the pH up somewhere in the range of 5.5 to 6, which at the same time also furnishes some magnesium.   Soluble magnesium, usually as the sulfate, also is applied to furnish quickly available magnesium.   Occasionally, magnesium oxide is used as a pH corrective and to furnish magnesium, but it has not been entirely satisfactory.   On soils with a pH higher than 5.5, soluble magnesium only is recommended and sometimes it must be used in large amounts on calcareous soils.
      Some adjustment of magnesium fertilizer practices must be made to compensate for the level of potassium available to the trees.   Fudge (1946) showed that the usual antagonism of potassium and calcium to magnesium exists in the field in a pronounced way.   This has complicated most of the interpretation of experimental results.   In practice, it means that levels for magnesium fertilization can be determined only in relation to potassium levels.   At high levels of potassium availability, magnesium must be increased.   To a lesser degree, the same situation holds with reference to calcium.   To get results, the amounts of magnesium applied on calcareous soils should be greater than those applied on soils that are low in calcium.
      When the percentage of magnesium in the exchange complex decreases, with a corresponding increase in exchangeable hydrogen or potassium, magnesium is less readily absorbed by the plant.
      The ammonium ion interferes with magnesium absorption, but probably this is not important under most field conditions.
      Accumulation of potassium in soils from the use of manures or other potassium-containing fertilizers is prevalent in California.   In many cases, the degree of magnesium deficiency is mild, although occasionally severe deficiency occurs.   Where potassium accumulates, it often requires rather large and repeated soil applications of magnesium sulfate to bring about correction.   Bingham, McColloch, and Aldrich (1956) found that the K:Mg ratio of exchangeable bases in magnesium-deficient soils ranged from 0.3 to 0.9, whereas little magnesium deficiency occurred with ratios of 0.1 to 0.3 (e.g., 10 to 3.3 times as much exchangeable magnesium as potassium).
      Under irrigation agriculture with waters of low-magnesium content, leaching and loss from crop removal may exceed the amounts added in irrigation water and manures and gradually deplete the magnesium supply of soils (Broadbent and Chapman, 1950; Pratt and Chapman, 1961; Pratt et al., 1959).   The addition of fertilizer and soil amendments accelerates magnesium loss.   In view of these findings, regular use of some magnesium in the citrus fertilizer program, especially where potassium is included, appears to be a sound practice.
      Because of the slowness with which magnesium deficiency is corrected under conditions in which soil potassium buildup has taken place, spray applications of magnesium nitrate have proved beneficial.   Embleton and Jones (1959) recommended a spray mixture containing 7.5 to 10 pounds of magnesium nitrate per 100 gallons of water.   This mixture produces correction in six months, and meanwhile soil applications can be initiated.   Mixtures of calcium nitrate and magnesium sulfate are effective.
      In California, nitrogen fertilizers often bring about partial or complete correction of magnesium deficiency.   This could be due to the fact that the nitrate added or produced mobilizes calcium and magnesium, which helps to suppress potassium absorption and permits more ready uptake of magnesium.
      The effect of nitrogen in ameliorating magnesium deficiency has been noted by many researchers.   Embleton (1966) reviewed this subject.

Magnesium Excess
      The author is not aware of specific cases of magnesium toxicity under field conditions.   However, in controlled cultures the author has produced magnesium toxicity.   In sand-culture experiments with sweet orange seedlings (incorporating magnetite as a source of iron), Guest (1944) found that additions of basic magnesium carbonate at a rate of 0.1 per cent of the weight of the sand produced severe iron chlorosis and root stunting.   Up to 0.5 per cent of calcium carbonate had no effect under comparable conditions.   The effect of the magnesium was ascribed to a specific toxic effect of magnesium on the roots.
      In subsequent experiments with solution cultures, Chapman (1944, unpublished data) found that 30 meq/l of magnesium chloride added to a complete culture solution severely reduced root growth and produced iron chlorosis of the tops.   Comparable amounts of calcium chloride produced no effect (fig. 3-19).   In further experiments, 50 to 100 meq/l of magnesium nitrate produced not only iron chlorosis but a marginal leaf burn (fig. 3-19).   Comparable amounts of calcium nitrate were without effect.   From these limited observations, it is evident that excesses of both solid-phase and soluble magnesium are much more toxic than equivalent amounts of calcium.
      Under field conditions, one might look for magnesium toxicity in serpentine-derived soils and also in soils where considerable magnesium carbonate is present.   In South Africa, Bathurst (private correspondence) reported a condition called "yellow branch."   It occurs in soils of low calcium and high magnesium content.


Manganese Deficiency
      As with a number of other trace elements, the wide distribution of manganese in soils and plants was known to early workers.   Research by Bertrand (1897a, 1897b, 1905, 1912) and Mazé (1914), followed by the work of McHargue (1919, 1921, 1923, 1926) and others, showed beyond much doubt that manganese was essential for green plants.   Good literature reviews dealing with manganese are given by Mulder and Gerretsen (1952), Scharrer (1955), and Labanauskas (1966).
      The first controlled culture manganese work with citrus was by Haas (1932a, 1932b), who found that lack of this element caused chlorosis of terminal growth.   In Florida, field responses to manganese were reported by Skinner and Bahrt (1931) and Skinner, Bahrt, and Hughes (1934).   Later, Chapman, Liebig, and Parker (1939) produced manganese deficiency both in young orange trees growing outdoors in sand cultures and in lemon cuttings growing in the greenhouse.   The leaf patterns closely resembled those noted in Florida by Camp and Reuther (1937), and analyses of affected leaves showed them to have a much lower manganese content than green leaves.   Also, the leaves became green when treated with manganese solutions, and additions of manganese to the nutrient medium cleared up the condition.   Primarily on the basis of clear identification of leaf patterns of manganese deficiency in lemon and oranges produced under controlled cultures, chlorotic leaf conditions thought to be caused by lack of zinc (Parker, Chapman, and Southwick, 1940) were identified in the field as manganese deficiency.   Widespread deficiencies of this element were then found in California citrus orchards and plantings of other fruit trees.   After this early work in Florida and California, manganese deficiency was found and identified in a great many citrus-growing regions of the world (Koto and Takeshita, 1957; Carpena, Guillen, and Costa, 1959; Taylor and Burns, 1938).
      Symptoms of Manganese Deficiency.—The effects of manganese deficiency on citrus have been well described by Camp, Chapman, and Parker (1949).   The following account is taken essentially verbatim from their description.
      Both young and old leaves show symptomatic leaf patterns.   Young leaves have a fine network of green veins on a lighter green background, but the pattern is not as distinct as in zinc or iron deficiencies and the leaf is greener.   As the leaf matures, the pattern resolves itself, in the milder forms of manganese deficiency, into dark green, irregular bands along the midrib and main lateral veins, with lighter green-colored areas between the veins (fig. 3-20).   With increasing severity, several gradations in color occur.   Similar gradations for lemons are shown in figure 3-21.   These range from light green to dull, pale green splotches between the main lateral veins.   Although the pattern approaches that of zinc deficiency, it never develops the extreme contrast which characterizes the latter.   If the deficiency is still more severe, the leaf assumes a dull-green or yellowish-green color along the midrib and main lateral veins, and the bands of these colors become narrower.   At the same time, the interveinal areas become still paler and duller.   In extreme cases in California, the interveinal leaf areas of oranges, lemons, and grapefruit develop many whitish spots.   Sometimes lemon leaves show a whitish or gray appearance.   Parker, Southwick, and Chapman (1940) also observed that in California the leaf symptoms are generally more pronounced on the shaded side of the trees.   Such leaves are not noticeably reduced in size or changed in form by manganese deficiency.   Severely affected leaves acquire a prematurely senescent (aged) appearance and fall from the tree.   Some limbs lose more foliage than others, with the result that the trees are unevenly foliated.   When trees are only slightly affected, the young leaves show a mild pattern which disappears after the flush of growth is completed and the leaves mature.   This is especially apparent on the spring growth flush.
      In addition to the foregoing symptoms, Chapman et al. (1939) noted in severe cases a speckling of brown spots over the leaf.   These symptoms are illustrated in figure 3-22.   A somewhat similar speckling has been noted on other plants, notably sugar beet, oats, barley, corn, and sugar cane.
      No particular twig symptoms have been reported, but in acute cases there is considerable dying of small twigs.   In Florida, this is more noticeable in tangerines and Temple oranges than in other varieties.   Dying of twigs is more noticeable in trees affected with combinations of zinc and manganese deficiencies than in trees deficient only in manganese.
      No striking fruit symptoms have been noted, such as those characteristic of zinc and copper deficiencies.   Skinner and Bahrt (1931), Skinner et al. (1934), and Bahrt and Hughes (1935) reported an increase in yield in some orchards, and an intensification in skin color and juice color as well as increased firmness after applications of manganese to deficient trees.   Roy (1937) reported an increase in sugar in the juice.   Later work in Florida showed the effect of manganese deficiency on fruit color to be much less pronounced than that of magnesium deficiency.   In California, fruit color effects of manganese deficiency have not been observed even in fruit from orchards very deficient in manganese.   The addition of this element resulted in no improvement of commercial grade or increase in fruit size (Parker, 1945, unpublished data).   The yields of citrus trees apparently are not reduced appreciably by very mild forms of manganese deficiency in which the leaf symptoms are transitory.   Yield reduction may occur, however, when the leaves are chronically affected.   Orchard studies in California by Labanauskas, Jones, and Embleton (1963) and Labanauskas and Puffer (1964) have shown yield increases ranging between 7 and 19 per cent as a result of foliar applications of manganese applied annually.   A slight increase in total soluble solids was also noted.   A summary of the principal effects of manganese deficiency is presented in table 3-24, and additional illustrations of manganese deficiency effects are shown in figures 3-23 and 3-24.
      Causes of Manganese Deficiency.—Manganese deficiency occurs on both acid and alkaline soils; it is probably due to leaching losses and consequent low levels in acid soils and to very low solubility in alkaline soils.   On very acid sands, it is commonly associated with deficiencies of zinc, copper, and magnesium, as evidenced by both tree responses and soil analyses.   The most severe cases in California however, have not necessarily been associated with marked zinc deficiency, though the two occur together frequently.   The artificial acidification of naturally neutral or alkaline soils in California has been observed by Parker and others (unpublished data) to decrease the severity of the deficiency symptoms.   Culture studies by Chapman et al. (1939) showed that manganese deficiency occurred in a neutral or alkaline medium (pH maintained at 7 or above) even though manganese was regularly supplied.
      Manganese in soils occurs as oxides and hydrated oxides in exchangeable form and in organic combination.   Below pH 5.5 it is predominantly in exchangeable and water-soluble forms, but above this pH it occurs at trivalent and tetravalent oxides.   It is well known that above pH 6.5 the various manganese oxides are highly insoluble.   Manganese solubility (and valence) also fluctuates with the oxygen (aeration) status of soils.
      Manganese deficiency of many plants commonly occurs on the following kinds of soils:
      1. Thin, peaty soils overlying calcareous subsoils.
      2. Calcareous silts and clays.
      3. Acid, sandy soils where manganese was low or has disappeared by leaching.
      4. Old, black, granular soils where manure and lime have been applied.
      5. Phosphate-deficient soils.
      6. Sandy soils containing considerable organic matter.
      Practices or conditions that commonly produce and aggravate manganese deficiency are liming of soils sufficient to increase the pH to above 6.5, burning of organic soils, irrigation with waters that increase pH and deposit bicarbonates and carbonates, drought conditions, high potassium fertilization, low or deficient calcium, and soil pathogens.
      Practices and conditions tending to increase manganese availability are soil acidification, phosphate fertilization, and decreased soil aeration.
      Control of Manganese Deficiency.—Soil applications of manganese compounds have not been very successful under neutral or alkaline soil conditions, owing to their rapid conversion to insoluble forms.   However, Leonard and Stewart (1959), in recent work on a calcareous soil, found that applications of a mixture of 5 pounds of manganese sulfate, 5 pounds of calcium chloride, and an additional 10 pounds of wettable sulfur, applied in a series of five to ten piles of about 1 foot square, resulted in high manganese uptake and correction that lasted over eight months; however, the trees became manganese deficient later.   Under acid soil conditions, especially on the acid sands of Florida (where the deficiency is caused primarily by low manganese content of the soil), regular additions of manganese compounds to fertilizer mixes have been effective.   Chelates of manganese have not been extensively tried as yet on citrus.
      Foliar sprays containing 3 to 5 pounds of manganese sulfate and half as much hydrated lime will temporarily correct manganese deficiency, but the treatment must be repeated yearly or more often.   To reduce spray residue on leaves, Labanauskas (unpublished data) is now advocating the use of sprays containing manganese sulfate alone at a rate of 1 pound per 100 gallons of water.

Manganese Excess
      Although leaf symptoms of excess manganese commonly have been produced on citrus from either excessive soil applications or sprays, or in controlled cultures (Haas, 1932b; Smith and Specht, 1953), naturally occurring cases of manganese toxicity on citrus have been rather rare.   Perhaps this is because citrus is fairly tolerant to manganese excess.
      On the other hand, cases of manganese toxicity, either caused by high soil content of manganese (e.g., pineapples in Hawaii) or soil acidity, have been reported for many other crops by Johnson (1924), Fergus (1954), Sherman and Fujimoto (1947), Parbery (1943), Berger and Gerloff (1948), Schmehl, Peech, and Bradfield (1950), Hewitt (1947), and Löhnis (1951).   In recent years manganese toxicity to citrus has occurred in Japan as a result of the heavy use of ammonium sulfate fertilizer.
      Some of the leaf symptoms of manganese excess on citrus are shown in figure 3-25.   The most characteristic effect is a marginal yellowing of the leaf, with a central green area remaining.   Sometimes the veins in the yellowed marginal areas retain their green color.   In addition, it is not uncommon to see necrotic spots on various parts of the leaf.   Necrotic spotting of leaves and stems has been noted on other plants suffering from manganese excess.   (See Hewitt, 1963, for descriptions on other plants.)
      Control of manganese excess, where it results from soil acidity, is accomplished by soil liming.
      Löhnis (1960) has found that both magnesium and calcium markedly decrease manganese uptake with some crops.


Molybdenum Deficiency
      The essentiality of molybdenum for green plants was established by Arnon and Stout (1939b) and Piper (1940).
      Vanselow and Datta (1949) were the first to produce molybdenum deficiency of citrus under controlled culture conditions.   Stewart and Leonard (1952b, 1952c, 1953a, 1953b) showed that "yellow spot" of citrus in Florida, a condition long recognized (Floyd, 1908), was due to lack of molybdenum.
      The leaf symptoms of molybdenum deficiency, as produced on lemon leaves (and later oranges) under greenhouse conditions by Vanselow and Datta (1949), are different from those seen in the field in Florida, and yet there can be little doubt about the authenticity of these two different effects.   Vanselow and Datta found that the growth and foliar effects of molybdenum deficiency could be prevented by addition of as little as 0.0001 to 0.001 ppm of molybdenum to the water cultures.   Dilute foliar molybdenum sprays were also effective in producing recovery.   The molybdenum content of affected leaves was 0.01 to 0.04 ppm of molybdenum in the dry matter as compared with 0.2 to 0.3 ppm in control cultures showing normal growth.   Analyses for other trace elements in these leaves were not enough out of line to account for the leaf burn noted, but it could be that some constituent such as nitrate accumulated sufficiently to produce the symptoms noted by Vanselow and Datta (1949).   The foliar effects produced in lemon and orange plants in water cultures are very similar to those produced on tomatoes in nutrient cultures by Arnon and Stout (1939b).
      In Florida, Stewart and Leonard (1952c) found that as little as 1 ounce of sodium molybdate in 100 gallons of a dilute lime-sulfur spray (at 10 gallons per tree, equivalent to 0.1 ounce) would cause the yellow spots to become green within three to four weeks after application.   The molybdenum content of "yellow spot" leaves was less than 0.05 ppm in the dry matter.
      As far as the author is aware, the field occurrence of molybdenum deficiency in citrus has been noted only in Florida and Japan (Otsuka and Takahashi, 1962).   However, reports of low molybdenum values in citrus leaves have come from DeVilliers (private correspondence) in South Africa.   In California, Lucie-Smith, Vanselow, and Chapman (1950, unpublished data) made leaf analyses for molybdenum on samples from fifty-five commercial citrus orchards (oranges and lemons), representing all of the major California soils in which citrus is grown.   The leaves were spring-cycle from fruit-bearing terminals ranging in age from four to twelve months.   Values ranged from 0.013 to 0.16 ppm of molybdenum in the dry matter, the majority falling into the range of 0.02 to 0.04 ppm.   In no case were symptoms of molybdenum deficiency found, but since seven of the fifty-five orchards showed less than 0.02 ppm of molybdenum, it is probable that sooner or later molybdenum deficiency will be found, not only in California citrus orchards, but in many other areas of the world.   In pot-culture trials with a granite-derived California soil, Brusca and Haas (1956a) got slight growth increases from the addition of small amounts of sodium molybdate.
      Effects of Molybdenum Deficiency.—The most characteristic field symptom, as seen in Florida by Stewart and Leonard (1952c), is yellow spots on the leaves.   They appear first as water-soaked areas in early spring and then develop into larger interveinal yellow spots, with gumming on the lower leaf surfaces in late summer.   These gum spots may in time become nearly black (fig. 3-26).   In severe cases, complete defoliation of the tree occurs.   Trees on grapefruit rootstock appear more susceptible than those on other stocks.   Molybdenum in leaves tends to be concentrated in interveinal areas.   When it becomes deficient, these areas are the first to be affected.   The yellow spots may coalesce as they increase in size.
      Fruit is not affected except when the deficiency is severe, and then irregular brown spots with a yellow halo may be found on the rind.   The discoloration does not affect the albedo.
      The spots on the leaves and fruits are more common on the sunny side of the tree.
      According to Stewart and Leonard (1952c), the following changes occur in the formation of the yellow spots.   First, gum and oil are deposited in the cell cytoplasm, causing the cells to swell and crowd out the air spaces within the leaves.   Large deposits of gum are formed in the intercellular spaces.   Second, cork cells are formed on the undersides of the affected leaves.   Third, there is decomposition of chlorophyll, causing the characteristic yellow spots.   Spraying with molybdenum initiates renewed production of chlorophyll and the yellow spots largely disappear.
      As produced by Vanselow and Datta (1949) in controlled culture solutions, none of the typical yellow spotting (as seen in the field) occurred.   Instead, the effects were similar to those seen in tomato and many other plants, with first an early mottling of the leaf margins, followed shortly by a necrosis and upward curling of the edges of the leaves (figs. 3-27 and 3-28).   The necrotic leaves soon fell, and new growth emerged from the axils of the abscised leaves.   Only the terminal growth on these cuttings showed the acute stages of molybdenum deficiency; old leaves did not display any symptoms.   Total growth of the molybdenum-deficient plants was less than that of controls (fig. 3-29).   Vanselow and Datta (1949) found that molybdenum-deficient plants could absorb enough molybdenum from 0.0001 ppm of molybdenum in the nutrient solution to produce normal plants.   They also found that the entire plant recovered when the foliage of just one branch of a typically molybdenum-deficient lemon plant was sprayed with 4 mg of molybdenum, as a potassium molybdate (KMoO4) solution.   Burned leaves do not recover, of course, but new growth is normal.
      Causes of Molybdenum Deficiency.—Molybdenum deficiency is more prevalent on acid soils than on neutral or alkaline soils.   Acid sands and acid-residual soils are more apt to be deficient in this element than acid-alluvial soils.   Red soils of low pH, derived from basalt and granite, are often lower in molybdenum than those derived from sedimentary rocks.   Extensive weathering and leaching reduce the available molybdenum of soils.   Molybdenum forms highly insoluble compounds with various secondary clay minerals under acid soil conditions.   Liming increases molybdenum availability.   Florida soils in which molybdenum deficiency of citrus was found contained from 0.2 to 0.4 ppm of molybdenum (Stewart and Leonard, 1953b).
      In addition to its direct essentiality for plants, molybdenum is necessary for nitrogen-fixing organisms and other soil organisms, including those concerned with nitrification.   Response of legumes to molybdenum is often due to its effect on the nodule bacteria, hence the appearance of nitrogen deficiency symptoms when this element is lacking.   It is essential for nitrate reduction within the plant and also performs other functions.
      Since phosphorus and sulfur deficiencies commonly occur on acid soils, molybdenum deficiency frequently develops after these have been corrected.   In addition to increasing the crop need for molybdenum, there is evidence that sulfate may decrease molybdenum availability by competing for absorption sites on plant roots (Stout et al., 1951).   Phosphate, on the other hand, is considered to enhance molybdenum uptake.
      Manganese excess has been reported by Mulder (1954) to induce molybdenum deficiency.   Manganese solubility increases with increasing soil acidity, whereas molybdenum solubility decreases.   Millikan (1948, 1949) has produced evidence of a manganese-molybdenum relation in flax.
      Soil-acidifying fertilizers, such as ammonium sulfate, will aggravate molybdenum deficiency by converting molybdenum to less available forms in soils.
      Low calcium and high nitrogen may increase the need for molybdenum, according to results obtained with cauliflower by Alexander and Stark (1959).
      Control of Molybdenum Deficiency.—Molybdenum deficiency is easily corrected by foliar sprays and, in many soils, simply by liming the soil.   Stewart and Leonard (1952b, 1953b) found that 1 ounce of sodium or ammonium molybdate in 100 gallons of water, applied as a spray at the rate of 10 gallons per tree, corrected the deficiency in three to four weeks.   One spraying a year was sufficient.   This element, unlike iron and some others, is mobile in plants, and, though it does not appear to move from young to older leaves, it does move readily in the opposite direction.
      Soil applications of molybdate salts on citrus in Florida have not proved as effective as sprays, though with most crops soil applications are effective.
      In general, rates of 1 to 2 ounces of sodium molybdate per acre are ample, and, in many cases, the molybdate is incorporated in phosphate or other fertilizer mixes to insure more uniform distribution when applied to the soil.   The amounts of molybdenum needed are increased where sulfate, iron, and manganese are high or phosphate is low.   For these and other reasons, acid soils require higher rates.   Frequency of application will depend on the crop, soil, and associated soil conditions.   In Australia, for example, it was found that 2 ounces of molybdenum trioxide (MoO3) per acre, applied in 1946 for subterranean clover, was still providing correction in 1955 (Anderson, 1956).

Molybdenum Excess
      Molybdenum is not nearly as toxic as boron, copper, zinc, and other trace elements.   There are not much data on molybdenum excess for citrus.   P. F. Smith of Florida (private correspondence) has grown orange seedlings with 50 ppm of molybdenum in a culture solution without harmful effects.
      Stewart and Leonard (1953b) found that sprays containing as much as 1 ounce of sodium molybdate per gallon (0.01 ounce is sufficient for correction) did not cause injury, though the injection of 1 ounce per gallon into the trunks of trees caused many branches to die.   Gerloff, Stout, and Jones (1959), working with tomatoes, found that excesses of molybdenum induced iron chlorosis.   They suggest that molybdenum and iron form a precipitate of low solubility in root tissues.
      For information on other crops, the reader is again referred to the bibliography on molybdenum by Borys and Childers (1961), Albrigo, Szafranek and Childers (1965) and also to reviews by Anderson (1956) and Johnson (1966).


Nitrogen Deficiency
      Nitrogen fertilizer is generally required in greater amounts by citrus than any other fertilizer.   Since nitrogen disappears from soil through leaching, volatilization, crop removal, and runoff, sooner or later supplemental nitrogen from one source or another invariably is required for commercial citrus production.
      Nitrogen plays a dominant role in citrus nutrition; deficiency, when acute, arrests vegetative growth and greatly reduces fruit production.   Excesses, on the other hand, may adversely affect quality, especially with oranges, grapefruit, and mandarins.   Also, nitrogen can affect nutrition with respect to phosphorus, potassium, calcium, copper, zinc, and other elements, either directly or through secondary effects of various nitrogen carriers on soil reaction, structure, or nutritional balance.
      It is easy to understand, therefore, why so much research has centered around the nitrogen nutrition of citrus.   In spite of this, many questions remain unresolved.
      Although there are great differences from soil to soil in the time required for native soil supplies to become deficient, the author is aware of no citrus-growing area where the use of some form of nitrogen fertilizer has not ultimately become necessary.   Some virgin soils are sufficiently well supplied, however, that no need becomes apparent for the first ten to fifteen years of orchard life.   When nitrogen fertilization was discontinued for two to six or more years in heavily fertilized, mature orchards, Jones and Embleton (1959) and Taylor, Pratt, and Goodall (1960) found no impairment of fruit production.   Judging by leaf analysis and foliage color, enough residual supply had accumulated to amply supply tree needs for a lengthy period.   There are some situations in which irrigation water contains enough nitrogen to meet or reduce nitrogen needs.   In low rainfall areas, requirements for nitrogen are usually lower than in areas of leaching rainfall.
      Effects of Nitrogen Deficiency.—Many investigators have described nitrogen deficiency of citrus.   Among them are Camp and Fudge (1939); Chapman and Kelley (1943); Bryan (1940); Camp et al. (1941, 1949); Chapman, Brown, and Rayner (1945); Smith and Reuther (1954); Malavolta et al. (1962); and Uexküll and Kämpfer (1963).
      Since nitrogen is a constituent of proteins, chlorophyll, and other essential components of plants, acute lack or shortage of this element brings vegetative growth to a halt and results in a general bronzing and yellowing of foliage, followed by dieback of twigs and decreased and finally almost complete cessation of flowering and fruiting.   When trees are well supplied with nitrogen and then gradually become deficient, various transient leaf symptoms appear.   In lemons particularly, green leaves may develop yellowish, irregular, blotchy areas before becoming entirely yellow (fig. 3-30).   In addition, older leaves may develop vein chlorosis or "yellow vein," as it is sometimes called (fig. 3-31).   This condition can also be caused by root rotting, branch injury, calcium deficiency, and other disorders.   Sometimes a yellow-vein condition of old leaves will develop when a particularly vigorous new shoot has emerged or much fruit has been borne on the shoot.   This symptom is somewhat less conspicuous in oranges and other species than on lemons, but it does occur.   If the leaves are not too old, restoration of nitrogen sometimes will cause such leaves to regreen fully or partially.   Completely yellow leaves, if not too old, will regreen when nitrogen becomes available; if temperature conditions are favorable, abundant new growth will commence.   The author has noted the yellow-vein condition rather widely on oranges and mandarins during the winter in Australia and India.   The condition commonly clears up in summer.   Leonard and Stewart (1961) have noted it on red and pink grapefruit in Florida.   Analyses of affected leaves showed them to be low in nitrogen and calcium, but normal in phosphorus, magnesium, and boron.   Field experiments by Leonard and Stewart (1961) have demonstrated that the condition can be markedly improved by extra nitrogen fertilization, but not by adding calcium chloride or calcium sulfate.   Nitrogen fertilization not only decreased the yellow-vein symptom, but increased both the nitrogen and calcium of the leaves.   Sand-culture trials have confirmed the aforementioned findings.   Although this condition can arise from other causes (e.g., rotted roots, calcium deficiency, and in weak trees afflicted with virus disease), it is apparently often no more than another manifestation of seasonal nitrogen deficiency.
      New leaves of trees moderately or acutely deficient in nitrogen are apt to be thin and fragile when young, and the angle between stem and leaf may be narrower than when nitrogen is ample (figs. 3-32 and 3-33).
      Trees continuously lacking in nitrogen are stunted, unsymmetrical in shape and twig growth, short, and irregular.   Such trees seldom die and remain in this condition with some seasonal, weak new growth appearing from time to time.   Fruit production is greatly decreased, but quality is not seriously impaired.   The fruit rind is smooth, may be somewhat pale in color, and fruit size may be decreased.   Juiciness may actually increase, and total solids are not much affected.
      Nitrogen deficiency at certain stages resembles phosphorus or sulfur deficiency, but an analysis of mature leaves will establish a clear-cut differential diagnosis.   If the trouble is lack of sulfur or phosphorus, nitrogen content will be equal to or higher than in trees well supplied with nitrogen (ranging from 2.5 to over 3 per cent of nitrogen in the dry matter).   If, on the other hand, the disorder results from lack of nitrogen, values of this element will usually be less than 2 per cent.
      Many other conditions produce a tree appearance similar to acute nitrogen deficiency—for example, salinity, high water table, insect and insecticide damage, neglect, and drought.   These conditions are not always easy to evaluate and diagnose, but leaf analyses, additions of nitrogen fertilizer, or the use of foliar urea sprays will make it possible to determine whether the difficulty is lack of nitrogen or something else.   If the problem is nitrogen starvation, new growth will soon start if air and soil temperature conditions are favorable.   Under favorable temperature conditions, citrus trees will begin to show leaf greening and initiation of new growth in ten days to two or three weeks from the time additional nitrogen is available to the roots.   Hilgeman (1941), using leaf analyses to check nitrogen uptake, got significant increases in leaf nitrogen content within fifteen days of soil applications of calcium nitrate in late summer.   Urea foliage sprays will bring about even more rapid regreening of foliage.
      Mild nitrogen deficiency in oranges may become apparent only in the spring at the time of blossom and new growth (fig. 3-34) or in the late summer and fall when the blossoms and fruit are drawing heavily on nitrogen supply.
      For spring blossom and new foliage, nitrogen stored in the older leaves, bark, and wood is heavily drawn upon, and, if insufficient, smaller-than-usual leaves are produced and the new foliage is yellowish and weak.   Cameron and Appleman (1934), Cameron et al. (1952), and Smith and Reuther (1950b) reported on this problem.   Data on the distribution of nitrogen in a ten-year-old Valencia tree, as reported by Cameron and Appleman (1934), are shown in table 3-25.   It is of interest that the leaves alone contained 41 per cent of the tree's total nitrogen, and the rest of the top, 28.1 per cent.   Roots accounted for 10.4 per cent, and the fruit for 20.5 per cent.   Wallace, Cameron, and Mueller (1951) showed that fruit, flower, and leaf production required large amounts of nitrogen, calcium, and potassium.
      The early stages or trends toward nitrogen deficiency in oranges can usually be diagnosed by the analysis of four- to seven-month-old spring-flush leaves.   Orange leaves from fruiting terminals should contain more than 2.30 per cent of total nitrogen in the dry matter, and leaves of comparable age from nonfruiting terminals should run 2.50 per cent of nitrogen or higher.
      Not enough data are available for grapefruit, lemons, and mandarins to indicate critical nitrogen levels, nor is it known at just what stage of nitrogen deficiency fruit set and final yield begin to be impaired.   There are some indications that mild nitrogen deficiency (a stage at which there may be just a little yellow or bronze cast to the tree as a whole, in contrast to the deep green of adequate nitrogen) does not materially impair either fruit set or yield (Chapman, Joseph, and Rayner, 1955, unpublished data on oranges).   Smith and Rasmussen (1961) in nitrogen rate trials with grapefruit found that trees which showed seasonal nitrogen deficiency in spring and in the fall bore just as well as those receiving ample and high nitrogen rates.   However, more information on various varieties under a range of soil and climatic conditions is needed to throw further light on this aspect.
      There is some evidence that high nitrogen can offset the ill effects of high sulfate and boron (Foote and McElhiney, 1937; Haas and Thomas, 1928).   A recent paper by Jones et al. (1963) suggests that under either high sulfate or boron or both of these conditions, higher leaf levels of nitrogen should be maintained.   Smith and Rasmussen (1961) reported that toxic effects of lead arsenate sprays on grapefruit were offset or diminished on trees receiving plenty of nitrogen.
      When orange trees are mildly nitrogen-deficient, the only noticeable effect may be a slightly reduced growth rate.   Otherwise, the trees appear healthy and green.   However, they produce less blossoms and fruit and production may in some instances be impaired.
      A summary of nitrogen deficiency effects is presented in table 3-26, and various foliage and growth characteristics are illustrated in figures 3-30 through 3-36 [figure 3-30, figure 3-31, figure 3-32, figure 3-33, figure 3-34, figure 3-35, figure 3-36].
      Forms of Assimilable Nitrogen.—Although nitrate is the chief form in which nitrogen is absorbed by citrus under normal arable soil conditions, there is abundant evidence from controlled culture experiments that the ammonia form can be used.   In work with orange seedlings and lemon cuttings in water and sand cultures, Chapman (1951) summarized several years of research on this subject as follows:

       (1)   At concentrations of 5 meq/l of ammonium, growth was as good in most cases as with the same concentration of nitrate.   Equal mixtures of nitrate and ammonium at the same total concentration gave comparable results.   At higher concentrations of ammonium as the sole source of nitrogen, various troubles appeared; in some experiments, manganese deficiency symptoms were produced; in nearly all instances, roots showed signs of injury.
       (2)   With plants which had been previously growing in cultures receiving nitrate, a sudden change to ammonium halted growth temporarily; sometimes iron chlorosis developed.   This, however, disappeared when the plants became adjusted to the change.
       (3)   Severe pruning of healthy plants growing in ammonium cultures often was followed by root rotting, whereas in nitrate cultures maintained at the same concentrations, no root rotting occurred.

      Wander and Sites (1956) found that rough lemon seedlings could absorb and use the ammonium ion, and Wallace and Mueller (1957) likewise found that the ammonium ion could be used by rough lemon cuttings.   Van der Merwe (1952-53) reported on extensive field and controlled sand-culture experiments dealing with nitrate and ammonium absorption by citrus, noting that where the latter ion predominated, higher leaf levels of nitrogen were found, potassium absorption was decreased, and the acidity of the nutrient medium was increased.   Also, the ammonia form seemed to adversely affect fruit quality.   In sand cultures, leaf symptoms of zinc, manganese, and magnesium deficiencies developed if the ammonium ion predominated.
      Smith (1957) grew Pineapple sweet orange seedlings in aerated water cultures supplied respectively with nitrate, nitrate-ammonium, and all-ammonium nitrogen sources.   At pH 6, growth was as good in the all-ammonium cultures as with all nitrate.   At pH 4, growth in the ammonium cultures was somewhat less than with nitrate.
      Seasonal Absorption of Nitrogen.—Chapman and Parker (1942), working with young bearing orange trees growing outdoors in water cultures, measured weekly nitrogen absorption over a continuous period of forty months.   They showed that absorption rate was largely determined by solution temperatures.   Data obtained are shown graphically in figure 3-37.   Roy and Gardner (1946), working in Florida, obtained similar results.   Bitcover and Wander (1950), using gravel cultures, showed that under certain conditions an appreciable amount of nitrogen can be accumulated very rapidly (within minutes) by citrus roots.   While absorption is much less in winter months when soil temperatures are lower, there is some nitrate absorption at all times.
      Causes of Nitrogen Deficiency.—The nitrogen cycle in soils has been so widely described that there is no need to repeat it here.   As has been stated, leaching and volatilization losses plus substantial crop uptake (heavy crops of citrus fruit, e.g., 40,000 pounds, will remove between 45 and 55 pounds of nitrogen per acre per year) are the basic reasons why supplemental nitrogen is sooner or later needed for citrus.   Also, the organic nitrogen in most soils of the subtropical and tropical zones where citrus is grown reaches equilibrium levels below those found in colder areas.   This is another reason for the universal need of supplemental nitrogen in the citrus-growing areas of the world.
      There are other factors which influence nitrogen need and availability.   Bulky organic materials of high carbon-nitrogen ratio cause a temporary tie-up of available soil nitrogen by soil organisms.   High sodium, chloride, sulfate, and boron in soils require heavier nitrogen usage.   The beneficial effects of nitrogen may be indirect, since most nitrogen added to soil, irrespective of form, increases calcium in the soil solution.   Calcium, in turn, reduces sodium and boron absorption by roots and tends to limit the solubility of sulfate.
      Under waterlogged (anaerobic) conditions, nitrogen is lost by denitrification.   Additions of elemental sulfur have been shown by Martin and Ervin (1953) to cause biological reduction of fixed nitrogen compounds.
      There is evidence that high phosphorus increases nitrogen need.   On the other hand, high potassium may decrease nitrogen need (Chapman and Liebig, 1940).
      From this brief account, it is not difficult to understand why the nitrogen requirements of citrus vary enormously from soil to soil and why the old rule-of-thumb recommendations about annual nitrogen fertilizer needs are now being abandoned in favor of the use of leaf and soil analyses for guiding fertilizer practices.
      Control of Nitrogen Deficiency.—This topic is treated more extensively in a later chapter, and no attempt will be made to discuss it here, except to point out a few guiding principles.   Questions concerning amounts, timing, form, and manner of application continue to be of interest and importance.
      Many investigators have shown that with some citrus (especially grapefruit, oranges, and mandarins) excessive nitrogen can impair fruit quality, especially during the period of fruit maturation and under climatic conditions favorable for vegetative growth.   However, because of the heavy draft on nitrogen during the spring bloom and growth flush, it is generally held that nitrogen supplies must be adequate then and that excesses during this period are not apt to be harmful.   Hence, there is emphasis on early spring fertilization and the use of foliar urea sprays if foliar appearance and/or leaf analyses indicate a nitrogen need at that time.   While more data are needed for different citrus varieties, it appears that five- to ten-month-old spring-cycle leaves from fruit-bearing terminals of oranges should contain better than 2.3 to 2.4 per cent of nitrogen in the dry matter.   Similar cycle leaves from nonfruiting terminals should show at least 2.4 to 2.6 per cent of nitrogen.   The use of nitrogen in general should be decreased later in the season, although this will depend on species, variety, and climatic factors.
      While much emphasis was formerly placed on the annual requirements of nitrogen per tree, it is now clear that because of soil, variety, age, yield, and climatic variables such recommendations are almost meaningless.   About all that can be said is that there are very few situations where less than 1 pound of nitrogen per tree per year for mature trees will suffice.   Conversely, there are few situations where more than 3 pounds of nitrogen are needed.   The form, time, and manner of application appears to be largely conditioned by prevailing soil, tree, and climatic circumstances.   If a quick buildup of nitrogen in the foliage is needed in the early spring at or following bloom, urea foliar sprays are useful (Impey and Jones, 1960a, 1960b).   If soils are acid (e.g., pH 5.0 or lower), ammonium sulfate and other soil-acidifying fertilizers should be avoided and calcium, sodium, or potassium forms employed.   If there is already a sodium problem, sodium nitrate should be avoided.   If there is a salinity problem, urea or ammonium nitrate should be used; if more calcium is needed, calcium nitrate can be employed; if soil structure needs improving, organic manures and (under some conditions) calcium nitrate can be employed.   If excessive leaching is a problem, slowly available organic forms should be used.   Resin- or plastic-coated fertilizer offers interesting possibilities.
      The discussion above is sufficient to illustrate that control methods must be tailored to the soil, plant species, plant age, and climatic situations.   No universally applicable system can be established.
      Jones and Parker (1949b), Jones and Steinacker (1953), Jones, Embleton, and Goodall (1955), Haas (1949a), Kuykendall and Wallace (1953), and others have investigated the absorption and use of urea as a foliar spray and demonstrated that this method of supplying nitrogen is practical.   It is especially useful when a quick buildup in the tree is needed.   Several sprayings a year may be required if this is the only method used.

Nitrogen Excess
      In addition to the effects of excess nitrogen on growth, fruit yields, and quality, excessive nitrogen in some cases directly or indirectly influences the availability of copper, zinc, manganese, molybdenum, phosphorus, and other elements.   In many cases, the effect is due not to nitrogen excesses per se but to effects on soil reaction, or to the forms, manner, timing, and quantities applied.   Also, in some soils high rates of fertilization can cause soil structure deterioration and a whole chain of consequential changes.
      Ford, Reuther, and Smith (1957) found marked reduction of feeder roots in Valencias on Florida soils receiving high nitrogen rates.
      Under special circumstances there may be injury from impurities carried by various nitrogen salts, such as biuret from urea and potassium perchlorate in Chilean sodium nitrate.   Nitrite may be formed under some conditions, and various side effects from ammonium ion absorption may occur.
      With regard to what might be termed "direct" effects, nitrogen promotes vegetative growth, delays fruit maturity somewhat, and, under some conditions (particularly where climate is favorable for vegetative growth), stimulates tender late-fall growth flushes which are vulnerable to winter frosts.
      Also, with Valencia orange fruit, excess nitrogen sometimes promotes a certain amount of regreening (Jones and Embleton, 1959).   Regreening decreases sales appeal and, if the fruit is sweated by packinghouse treatments, may lower keeping and shipping qualities.   Under some conditions, excess nitrogen will produce fruits with somewhat coarser rinds, thicker skins, less juice, and poorer keeping and shipping qualities.
      However, these effects are not universal, and some of them may result from an unbalancing effect on tree nutrition.   These nutritional imbalances can be corrected by appropriate treatments.   Rootstock also can make a big difference.   Excess nitrogen is often worse on oranges budded to rough lemon than on sweet orange (Bouma and McIntyre, 1963).
      There are reports from many places of slight to moderate adverse effects of excess nitrogen on quality and yield (Jones, Embleton, and Steinacker, 1957; Wallace et al., 1952; Jones and Embleton, 1959; Oppenheimer and Heymann-Herschberg, 1954; Singh and Agrawal, 1959, 1960; Bain, 1949; Jones, Van Horn, and Finch, 1945; Bouma, 1959; and Sites, Wander, and Deszyck, 1962).   Most of the aforementioned investigations are not complete enough to determine the degree to which the adverse effects reported might have been offset by the use of a different rootstock or altered nutrition (e.g., more phosphorus), and especially the ameliorating or aggravating effects of climatic variables.   Often the experimental comparison is with fruit from plots where nitrogen is deficient.   As is well known, fruit produced on low or nitrogen-deficient trees is generally smooth and of good size and quality, but yields are low.   To assure good yields, there must be adequate nitrogen.   Under these conditions, sizes may sometimes be smaller (largely owing to increased numbers of fruit set), and fruits may be a little coarser in rind texture, have slightly thicker peel, and have somewhat lower juice percentage.   However, these effects are sometimes small and of little commercial consequence.   Hence, experimental results with nitrogen variables have to be evaluated in terms of the conditions under which they were obtained, and what might have been the result under other conditions of usage.
      There are many conflicting data in the literature relating nitrogen tests to yield, fruit sizes, and quality.   A brief review of data is presented here to document some of the preceding statements.
      Samuels (1931), in a study of soil nitrate as related to navel and Valencia orange fruit sizes in California, found that in many orchards fruit growth and sizes were greater when soil nitrate was high than when it was low.   This may have resulted from the greater use of manure on the high nitrate orchards and hence more potassium rather than nitrate per se.   In any case, the data are of interest even though not suspectible [sic] to critical analysis.
      With respect to yield, increased nitrogen, up to a point, commonly results in increased yields.   There are some cases on record where yields have been materially decreased by high or excessive nitrogen usage.   Where yields have been decreased, it has often been due to adverse secondary effects on soil phosphorus availability or other causes.
      After twelve years of a field trial with navel oranges on sweet orange stock involving nitrogen at rates of 1, 2, and 3 pounds of nitrogen per tree per year, Parker and Batchelor (1942) obtained highest yields at the 3-pound rate.   One half of the nitrogen in each treatment came from manure.   Winter cover crops were also grown and turned under.   It seems likely that the results obtained largely reflected a direct nitrogen effect and not, as has been suggested (Jones et al., 1961), that the increased yields were due to improved soil permeability resulting from the greater organic matter additions from the manure and from the increased growth of winter cover crops.   No yield responses in this experiment were obtained from phosphate or potash, and zinc deficiency was controlled by sprays.   Where 3 pounds of nitrogen came from manure alone, the yields were no greater than where only half as much was used and the remaining nitrogen came from urea.
      In both sand- and solution-culture experiments, carried on over a four- to ten-year period with young navel orange trees, the author (Chapman and Liebig, 1942, unpublished data; and Chapman et al., 1955, unpublished data) found no evidence of yield decreases from excessive amounts of nitrogen (as nitrate).   Burkhart (1959), in field experiments with young Lisbon lemons on rough lemon rootstock on a sandy soil, got best yields from the use of about 3 1/2 pounds of nitrogen per tree per year.   Part of the nitrogen came from manure, and phosphate fertilizer was added.   Less nitrogen decreased yields, and more nitrogen (about 5 pounds) produced close to the same yield as the 3 1/2-pound rate.
      Embleton et al. (1959), comparing the yields where nitrogen was omitted for five years with a "ranch treatment" (which provided about 1 pound nitrogen per tree per year) on an old Valencia orchard planted on a deep Yolo loam soil, found that the residual nitrogen from previous years of fertilization was sufficient to meet tree needs.   The additional nitrogen (over the 1 pound of "ranch treatment") did not decrease yields, neither did it increase yields.   Leaf analyses made for three of the years of the experiment showed 2.22, 2.32, and 2.32 per cent nitrogen in the dry matter of the "no nitrogen" plots as compared to 2.51, 2.37, and 2.36 per cent nitrogen in the "ranch-treatment" trees.
      Various California surveys (summarized by Platt, 1960) showed that where yield was related to nitrogen usage, greatest yields were on orchards receiving the highest nitrogen rate.   For example, in the early 1920's a survey of 1,000 orchards showed highest yields where 350 pounds of nitrogen per acre per year were used.
      Reitz and Hunziker (1962) conducted a nitrogen-rate experiment on native Marsh grapefruit budded to sour orange rootstock on a hammock soil in the Indian River area of Florida.   Best yields (six-year average) and size were obtained by adding 2 pounds of nitrogen per tree per year.   Four pounds of nitrogen per tree slightly decreased yields and sizes and coarsened the fruit somewhat.   Analyses of leaves from spring-flush, nonfruiting terminals picked in July showed an average of 2.13 per cent of nitrogen at the 2-pound nitrogen rate, 1.70 per cent with no nitrogen, and 2.25 per cent at the 4-pound nitrogen rate.   Stewart, Leonard, and Wander (1962), in a five-year experiment with Pineapple oranges on rough lemon rootstock found that best yields with ammonium nitrate were at rates of 100 to 200 pounds of nitrogen per acre per year, while 550 pounds markedly decreased the yield.   They suggested that the high rate of fertilizer might have solubilized enough copper to be toxic.   With sodium nitrate, best yields were at a rate of 350 pounds nitrogen per acre per year.   Smith and Rasmussen (1961) found with Marsh grapefruit that 540 pounds of nitrogen per acre annually did not decrease yield, as compared with rates of 120 and 240 pounds of nitrogen.
      Quoting from a general discussion on fertilization of citrus in Florida, Smith (1959) says: "In Florida, nitrogen has little direct effect on fruit quality over a fairly wide range of rates and sources.   Indirectly it has greatly improved fruit quality, as ample use of this element is the most important nutritional factor in gaining a good set of fruit.   As just indicated, a good set of fruit is essential for high quality fruit.   The average grove should receive 150 to 200 pounds of nitrogen annually per acre.   The consensus of technical opinion indicates that 0.4 pounds of nitrogen per box of fruit is still an appropriate guide for orange production in most cases, but experimental work now shows that this is excessive for grapefruit and the recommendation has been lowered to 0.3 pounds for this variety."
      Bouma and McIntyre (1963), reporting on orange yields from 1950 to 1960 in a field experiment in the Murrumbidgee irrigation area of New South Wales, Australia, indicated that maximum yields were obtained in noncultivation plots (weeds controlled by kerosene sprays) at a nitrogen rate of 1.6 pounds per tree.   Yields were about as good with 0.8 pound of nitrogen and were depressed at 3.2 pounds of nitrogen.   The source of nitrogen was ammonium sulfate, and at the highest nitrogen rate soil acidity was reduced to pH 4.5.   Reduced phosphorus availability may have accounted for poorer yields and fruit quality at the high nitrogen levels.   The low pH of the soil may also have produced other detrimental effects.
      A six-year experiment with fourteen-year-old Shamouti oranges (on sweet lime stock) on a loamy sand at Rehovot, Israel, carried on by Oppenheimer and Heymann-Herschberg (1954), produced maximum yields at rates of about 200 pounds of nitrogen per acre, though yields were only slightly less with half this amount and but slightly decreased at the 300-pound rate.   The six-year average per tree for three nitrogen rates of 100, 200, and 300 pounds of nitrogen per acre were 97.8, 104.4, and 101.3 kgm of fruit per tree, respectively.   Ammonium sulfate was used as a nitrogen source and all trees received, in addition, both superphosphate and potash.   In spite of this, available soil phosphate was reduced by the heaviest ammonium sulfate application, but leaf phosphate was not affected.   During the last year of the experiment, nitrogen in spring-flush leaves from fruiting terminals showed 2.59, 2.87, and 2.82 per cent of total nitrogen in the dry matter of the leaves for the three nitrogen rates, respectively.   Fruit size was not affected, but commercial quality was decreased by the high nitrogen treatment.   This may have been due to lack of sufficient phosphate.
      Singh and Agrawal (1959, 1960) carried on a nitrogen-rate experiment with several varieties of oranges and mandarins (all on Karna Khatta rootstock).   The soil was a loam to clay loam.   Differential nitrogen applications using ammonium sulfate were begun when the trees were only two years old, and four rates were used.   At the outset, the respective rates per tree were 2.3, 3.6, 6.0, and 9.6 ounces of actual nitrogen for the N1, N2, N3, and N4 treatments.   When the trees were six years old, the respective rates were 12, 18, 30, and 48 ounces of nitrogen per tree.   The average cumulative yields on the four orange varieties for the last four years of the experiment were 36.8, 44.3, 81.0, and 79.5 kgm of fresh fruit per tree for the N1, N2, N3, and N4 treatments, respectively.   The percentages of juice, total solids, citrus acid, and vitamin C were only very slightly affected.   The average last four-year yields of the mandarin varieties were 57.8, 72.1, 78.9, and 86.1 kgm per tree for the N1, N2, N3, and N4 treatments, respectively.   The N3 treatment had the highest yield of juice.
      Reuther and Smith (1950, 1952) conducted a Florida field experiment involving three rates of nitrogen, potassium, and magnesium with five-year-old, bearing Valencia oranges on rough lemon stock on a sandy soil.   They obtained best yields at the highest nitrogen rate (3 pounds of nitrogen per tree).   Fruit size decreased slightly, as did total soluble solids and vitamin C at the high nitrogen rate; total acid increased slightly, and degreening of the rind and maturity of the fruit was slightly delayed.   Spring-flush (March) leaves from fruiting terminals picked in December showed 2.30, 2.48, and 2.65 per cent nitrogen, respectively, in the dry matter.
      In an outdoor, sand-culture experiment with Valencia oranges on rough lemon rootstock, involving three levels of nitrogen (30, 80, and 210 ppm of nitrogen in the nutrient solution), potassium, and magnesium, Smith and Reuther (1953) reported best tree growth and yields at the highest nitrogen levels.   Fruit size was somewhat decreased, as was vitamin C, but total juice, soluble solids, and acid were increased.   Rind thickness was not appreciably different at the three nitrogen rates.
      Maxwell and Dacus (1958) applied ammonium nitrate, along with other nutrients, to red grapefruit under sod culture in Texas.   They found that only nitrogen influenced yields and that fruit quality was more affected by season than by treatment.
      Rosselet, Helff, and Langenegger (1962), reporting on a 20-year experiment with Valencias on rough lemon rootstock in South Africa, found that highest yields resulted from the use of ammonium nitrate plus manure.   During the last five years of the experiment, this treatment, which supplied 4.25 pounds of nitrogen per tree per year, produced an average of 651 pounds of fruit per tree as against 579 pounds from ammonium nitrate alone at a rate of 2.3 pounds of nitrogen.   A control treatment receiving no nitrogen produced an average of 196 pounds of fruit.   All plots received a uniform application of phosphorus, potassium, and magnesium fertilizers and were regularly treated with a zinc-copper spray.   Fruit quality was not materially affected by the higher nitrogen rate.   Rind thickness for the control, ammonium nitrate (NH4NO3), and NH4NO3 plus manure, respectively, was 0.39, 0.53, and 0.50 cm; total solids 10.7, 11.1, and 11.2; percentage juice 55.0, 52.0, and 54.0; and per cent acid 1.24, 1.22, and 1.41.
      No leaf analysis values were reported.   The authors ascribed the high yields on the NH4NO3 plus manure plot over NH4NO3 alone to the extra nitrogen.   The soil in the experiment was a reddish, lateritic-type sandy loam, well drained and permeable.   This experiment showed that in the absence of other limiting factors high rates of nitrogen fertilizer produced high yields with no impairment of fruit quality.
      In contrast to these results, a seven-year South African fertilizer experiment with young navels on rough lemon rootstock planted in 1952 (differential fertilizer treatments started in 1956) on a deep-reddish sandy loam, which had had several croppings to cowpeas preceding the citrus, gave no response to nitrogen over a basic 1 pound of nitrogen per tree per year rate (Bester, 1965).   This experiment consisted of sixty-four treatments involving nitrogen, phosphorus, potassium, calcium, magnesium, and manure in all combinations.   Average yields per tree for the last five years of the experiment (1959-63) and leaf analysis values on nine- to ten-month-old spring-cycle leaves from fruiting terminals were as follows:
                                                                 Yield in
                                                               Pounds of
                                                               Fruit per                    Per Cent
                                                            Tree, 5-Year                  Total N
                                                               Average,                   in Leaves
Treatment                                              1959-63                      in 1963     
Control (no nitrogen)                                 198.0                          2.20
Nitrogen                                                    257.6                          2.51
Nitrogen and phosphorus                           317.3                          2.61
Nitrogen, phosphorus, and manure             307.5                          3.20
Nitrogen, phosphorus, and
     potassium                                             350.1                          2.50
Nitrogen, phosphorus, magnesium,
     and manure                                           366.9                          2.97

      There were significant yield responses to phosphorus as well as nitrogen in the experiment, but none to potassium.   The manure added 1.72 pounds of nitrogen in addition to the 1 pound from the NH4NO3.
      Lack of yield responses to extra nitrogen in the experiment is probably due (1) to the trees being young—eleven years from planting at the end of the experimental period; and (2) very low leaching losses of nitrogen owing to the careful water-sparing irrigation practices used and low effective rainfall in the area.   No fruit quality information was reported in Bester's thesis, although quality data was obtained.
      In long-term experiments with Washington navel oranges in the San Joaquin Valley of California, Jones and Embleton (1967) found no yield response to nitrogen beyond an annual application of one pound per tree.   They noted also that deleterious effects on fruit quality were obtained with increasing rates of nitrogen application, and that Washington navel oranges have about the same minimum adequate level of nitrogen in leaves as in Valencia oranges.
      The review of literature above is sufficient to indicate that in many, though not all, cases increased nitrogen has resulted in increased yield.   Yield decreases, where they have occurred, are often related to adverse effects on the availability of other nutrients (notably phosphorus), or the release of toxic amounts of elements such as copper.   On the other hand, fruit sizes and external and internal quality factors may be slightly impaired under some conditions, but this is not universal.
      As for the secondary or side effects of nitrogen on soil reaction, nutrient availability, soil structure deterioration, nitrite accumulation, biuret, and potassium perchlorate toxicity, these may stem from the forms, rate, timing, and manner of application; soil characteristics, rootstocks, and varieties also are involved.
      In addition to the well-known effects of different nitrogen fertilizers on soil reaction and the whole chain of changes these may evoke in nutrient availability, there are the more or less immediate changes in the soil solution resulting from the cation or anion with which the nitrogen is associated.   The nitrification of organic and ammonium forms of nitrogen usually makes for temporary increases in the calcium (and to a lesser extent magnesium) content of the soil solution, and large increases in these cations may decrease the phosphorus, boron, zinc, manganese, and copper intake by plant roots and, in some soils, actually aggravate deficiencies of these elements.   Where excess boron is a problem, nitrogen fertilization sometimes decreases boron toxicity (Cooper, Peynado, and Olson, 1958; Jones et al., 1963).
      The effects of various nitrogen fertilizers on the micronutrients of citrus leaves have been studied by Labanauskas et al. (1958b).   Where soils had become more acid from the use of ammonium nitrate and sulfate, manganese absorption was increased.   On the other hand, calcium nitrate decreased manganese absorption.   All three fertilizers reduced boron absorption, but none of them affected zinc and iron absorption.
      In special circumstances (e.g., acid, sandy soils), heavy applications of ammonium fertilizer may result in considerable temporary accumulation and uptake of the ammonium ion and depression of manganese, magnesium, and potassium absorption.
      Soil structure and related changes (water penetration, salt accumulation, and tree deterioration), resulting from the continued use of sodium nitrate and ammonium sulfate at rates giving 3 pounds of nitrogen per tree per year, have been described by Aldrich, Chapman, and Parker (1944); Aldrich, Parker, and Chapman (1945); Jones, Cree, and Embleton (1961); and Lombard et al. (1962).   Smith (1962) found that Valencia orange trees on rough lemon roots were injured by sodium nitrate applied at rates of 285 and 330 pounds of nitrogen per acre per year.   The injury was due to excessive sodium absorption.
      Aldrich et al. (1944) found that the sodium from sodium nitrate partially displaced calcium from the exchange complex in the surface of the soil, decreased permeability to water, and resulted in salt accumulation, moisture deficiency, and tree deterioration.   The ammonium sulfate treatments made the soil so acid that ammonium, instead of nitrifying, accumulated in the exchange complex, causing decreased permeability to water and increased salt accumulation.   This treatment thus produced the same effects on the trees as sodium nitrate.
      The accumulation of nitrite in soils was thoroughly explored by Chapman and Liebig (1952).   In alkaline soils, they found especially high accumulations of nitrite from the use of urea.   In winter periods, when the soil is cold, this constituent sometimes persisted in soils for several months before it was converted to nitrate.   Under acid soil conditions, no nitrite accumulated.   The apparent explanation is that under alkaline soil conditions nitrobacteria convert nitrite to nitrate less rapidly than ammonium is converted to nitrite, and thus the latter accumulates.   No damage to the citrus trees were observed in plots where up to 90 ppm of nitrogen as nitrite accumulated.   Bingham, Chapman, and Pugh (1954) found that nitrite was much less toxic under alkaline than under acid soil conditions.   Since no nitrite has been encountered under the latter conditions in arable soil, it appears that nitrite is not likely to be a problem except under such special circumstances as anaerobic conditions.   In a study of silica-gravel cultures to which urea was added, Bitcover and Wander (1950) found that nitrite accumulated in the warm period of the year.   The trees wilted in cultures where up to 23 ppm of nitrogen as nitrite was found.
      The toxicity of biuret (NH2CONHCONH2 · H2O) to citrus has been investigated by Jones (1954), Impey and Jones (1960b), Haas and Brusca (1954a), Iwasaki and Shichijo (1962), and others.   Biuret is formed when urea is heated.   Its toxicity to pineapple plants was reported by Sanford et al. (1954).
      The yellow-tip mottling effects produced on citrus leaves by biuret are illustrated in figure 3-38.   Biuret causes permanent injury and the leaves fall prematurely.   Injury can result from either foliar or soil application of urea containing biuret in amounts varying from 0.5 to 2.5 per cent or more.   This compound accumulates, as such, in the apex of citrus leaves and appears not to be metabolized, according to Impey and Jones (1960b).   They found somewhat less protein in the injured leaf tips and an increase in several amino acids, especially glycine and serine.       The biuret problem has been largely overcome by the production of urea containing < 0.25 per cent of this compound and by using urea (in spray formulations) at rates not exceeding 7.5 pounds to 100 gallons of water.
      Potassium perchlorate injury is illustrated in figure 3-39.   It was widely observed in citrus in California and elsewhere during the World War II years, when Chilean sodium nitrate was exclusively used.   It occurs as an impurity in this salt.   In the early stages of leaf injury, it appears not to adversely affect tree health or production, but severely injured leaves drop prematurely and there are thought to be some adverse effects on production.
      This brief review of nitrogen excess problems is sufficient to indicate the diversity of troubles which may ensue from the overuse of nitrogen and some of the difficulties for which to watch.   Leaf and soil analyses are powerful tools to help guide fertilizer practices.   It is not enough, however, to merely check citrus leaves for nitrogen content alone.   Periodically, complete leaf analyses are needed and adjustments of fertilizer and management programs must be made.   As the secondary ill effects of nitrogen fertilization are corrected, higher yields can often be obtained from extra nitrogen fertilizer additions with increased profits to the grower.
      Control of Nitrogen Excess.—Fortunately, the ill effects of overuse can be avoided or easily corrected.   Decreases in nitrogen use, or changes to other forms, correction of soil acidity or alkalinity, use of gypsum or organic materials to improve soil structure, and foliar sprays to correct nutritional deficiencies are some of the self-evident measures which can be put into use when it is determined that the form, amount, timing, or manner of application of nitrogen fertilizer is causing harmful effects.


Phosphorus Deficiency
      Before World War II, phosphorus deficiency of citrus had not been positively identified in the field.   Knowledge of its effects on citrus was based on the symptomatology produced on young citrus plants and trees grown in sand, soil, and solution cultures.   Haas (1936b) reported experiments with lemon cuttings grown in solution cultures lacking phosphorus and also experiments with young Valencia oranges grown in minus-phosphorus sand cultures.   He found that a deficiency of this element caused the older leaves of both species to turn a brownish-green color and fall prematurely.   Many leaves developed water-soaked and, later, necrotic areas on the margins or tips (fig. 3-40).   Phosphorus-deficient leaves showed values ranging between 0.04 to 0.07 per cent of phosphorus in the dry matter.
      Subsequently, Chapman and Brown (1941a) and Chapman, Brown, and Liebig (1943) described the effects of acute phosphorus deficiency on the fruit, foliage, and composition of navel orange trees growing in soil.   The foliage symptoms were similar to those reported by Haas.   The fruit produced on the phosphorus-deficient trees was rather coarse, had thicker rinds, and had a lower juice content than fruit from trees well supplied with this element.   The acid was higher, but total solids in the juice were not measurably affected.   Analyses of various parts of healthy and phosphorus-deficient trees showed very low levels of phosphorus in all parts of the plant except the young leaves.   The old leaves of the phosphorus-deficient trees contained 0.05 per cent total phosphorus as compared with 0.11 per cent in comparable leaves from healthy trees.   Nitrogen, potassium, and magnesium were higher in the phosphorus-deficient leaves, but calcium was lower.
      Still later, Chapman and Rayner (1951) published the results of a nine-year experiment with orange trees growing outdoors in nutrient solutions of graded phosphorus content.   The effect of phosphorus lack on growth, behavior, yield, fruit size, fruit quality, and inorganic composition confirmed all of the previously mentioned observations and added much additional information.   Of special interest was the finding that the very earliest effect of low phosphorus was on fruit quality.   A greater percentage of the fruit from such trees lacked firmness, showed pith separation, and had thicker rinds and less juice than the fruit from trees getting more phosphorus.   These findings coincided with studies of fruit quality emerging from field experiments reported from various parts of the world: Takahashi (1931), Morris (1937), van der Plank and Turner (1936), Anderssen (1937), Anderssen and Bathurst (1938), Crous (1937), Allwright (1938), Esselsen and Oberholzer (1939), Bathurst (1945), Innes (1946), Forsee and Neller (1944), Hilgeman (1941), Finch and McGeorge (1945), Jones and Parker (1949a), Smith, Reuther, and Gardner (1949), Young and Forsee (1949), and Winnik (1950).   Much of the literature cited above was summarized by Chapman and Rayner (1951) and need not be repeated here.
      Up to 1949, the only field experiment in which yield increases from phosphorus had been reported was that of Forsee and Neller (1944).   They found that Valencia oranges growing on an acid peat soil in Florida benefited from phosphorus.   Trees in a plot not receiving phosphorus were smaller, achieved less vegetative growth, and had smaller and somewhat narrower leaves.   The fruit from phosphorus-deficient trees was coarse, soft, bright orange in color, thicker in rind, and frequently misshapen; the interior structure of the fruit was coarse, and the juice was more acid.   There was a greater preharvest drop of fruit from the trees lacking phosphorus than from those receiving phosphorus fertilizer.   The high-phosphate treatment produced a greater percentage of ammoniation (copper deficiency in the fruit).   Total phosphorus in the leaves of the phosphorus-deficient trees was 0.07 as compared with 0.098 per cent in the dry matter in trees receiving a 3-12-12 fertilizer.   The phosphorus content of the fruit juice was 0.89 compared with 3.46 ppm in the deficient and amply supplied treatments, respectively.   In the soil, an acid extract showed 8 versus 94 pounds of phosphorus per acre for the two plots.   A more recent account of this work was given by Young and Forsee (1949).
      In 1949, a lemon tree condition noted in California, characterized by a brown-leaf-spot splotching on older leaves fig. 3-41 and often accompanied by poor tree condition, was diagnosed by Aldrich and Haas (1949) and Aldrich and Coony (1951) as being due to phosphorus deficiency.   Leaves thus affected showed total phosphorus levels in the deficiency range, as established by Chapman and Brown (1941a) and Chapman (1949).   Responses to phosphate injection into the tree or applications to the soil were dramatic.   Affected lemon trees were found in several parts of California, and a few orange orchards were also discovered which showed phosphorus deficiency symptoms.   Subsequently, Embleton, Kirkpatrick, and Parker (1952) and Embleton et al. (1956) found an orange orchard with many of the same fruit and foliar symptoms, and leaf analysis values were in the phosphorus deficiency range.   Marked responses to phosphorus fertilization were obtained.   Phosphorus deficiency in this case had been brought about by the continued use of ammonium sulfate on a soil already low in total phosphorus.
      Heymann-Herschberg (1952) in Israel, reporting on phosphate trials of several Shamouti orange groves, found indications of a yield response in one grove and noted fruit size and quality effects consistent with those reported by others; in several groves, phosphate decreased yield, apparently because of an antagonistic depression of nitrogen absorption.
      In a Valencia orange (on rough lemon rootstock) field experiment at Nelspruit, South Africa, carried on from 1942 to 1956 with trees planted in 1934, Rosselet et al. (1962) got marked yield responses to phosphate fertilizer.   Fruit rind thickness was decreased, percentage of juice increased, juice acidity decreased, and total soluble solids slightly decreased in the fruit from phosphate-fertilized trees.   Phosphorus deficiency has also been noted in a few other places in South Africa (DeVilliers et al., 1958).   Bester (1965) has recently reported yield increases from phosphate in the soil of the Zebediela estate plantings in South Africa.   Bouma (1959) and Bouma and McIntyre (1963) reported marked yield and fruit quality improvements from phosphate in the nontillage plots of a Valencia and navel orange fertilizer experiment carried on in the Murrumbidgee irrigation of New South Wales, Australia.   The plots had been given variable irrigation and nitrogen treatments (as ammonium sulfate) beginning in 1947.   By 1950, evidence of phosphorus lack became apparent.   From that time on, yearly applications of superphosphate were made.   There was a marked improvement in yield and fruit quality.
      In Jamaica, Innes (1946), reported yield responses to phosphate.   Workers in Brazil also have stated that phosphorus is needed on many of their soils.
      Based on an extensive series of soil analyses, pot-culture tests, a review of the earlier California phosphate experiments (Chapman, 1934), and a subsequent leaf-analysis survey by Chapman and Fullmer (1951a), it was concluded that very few commercial citrus orchards in California were in need of supplemental phosphorus.   This is due in part to large accumulations of phosphate from the past use of manure and mixed fertilizers, coupled with the fact that the natural supply of phosphorus in many California citrus soils is good.   Also, the phosphorus requirement of citrus is not high as compared with that of many quick-growing field and vegetable crops.   However, it is necessary to be on the alert for phosphorus deficiency in all citrus-growing areas.   Periodic leaf analyses should be made and fruit quality checked for deficiency symptoms.
      Effects of Phosphorus Deficiency.—As a result of the aforementioned nutritional work and field fertilizer trials, a good understanding of the effects of phosphorus lack on tree appearance, growth behavior, leaf symptoms, and fruit quality is now at hand.   Moreover, the leaf tissue content of phosphorus and other inorganic nutrients which characterize phosphorus deficiency have been established.   In addition, reasonably good soil analysis criteria for detecting phosphorus deficiency of citrus are available.
      As with boron, potassium, nitrogen, and other deficiencies, there is no one specific leaf pattern or condition which of itself signifies clear and unmistakable evidence of phosphorus deficiency.   However, the combination of foliar, growth, and fruit characteristics (supplemented by leaf and soil analyses) makes possible an almost certain diagnosis when the deficiency is moderate to acute.   The range of foliar growth, yield, and fruit effects produced by lack of phosphorus is recorded in table 3-27.
      In the moderate to acute stages of phosphorus deficiency, there is premature abscission of old leaves, especially heavy in the spring following bloom; many of these old abscising leaves may show burned areas, either at the tip or elsewhere on the leaf (fig. 3-40).   New growth is sparse and weak, and bloom is commonly markedly reduced (fig. 3-42).   Weakened shoots die back, and the net effect of the reduced growth is to produce a thinly foliated tree, with much dead wood in evidence (fig. 3-43).   The foliage lacks luster, and many leaves take on a faded-bronze appearance similar to that characterizing nitrogen deficiency.   In the case of lemon trees, it is common to find many old leaves which show irregularly rounded, brown spots on the upper surface (fig. 3-41).   Such leaves abscise prematurely.
      Not only is fruit set markedly curtailed, but there is commonly a marked preharvest drop in cases where the deficiency has not yet become acute.
      The maturing fruit lacks firmness, has somewhat thickened rinds, and some of the fruits are misshapen.   There is a center separation of pith, and sometimes the segments are separated a little.   The fruit is rather soft and spongy and of low juice content (figs. 3-44 and 3-45).   Maturity is delayed and comparison at any given time shows a higher acid and vitamin C content than with fruit not lacking in phosphorus.   The color of phosphorus-deficient fruit is often a deeper orange than that of fruit not lacking phosphorus.   Fruit size is usually somewhat increased where phosphorus is lacking; this is probably due in part to less fruit on the tree.
      Early Symptoms of Phosphorus Deficiency.—In the carefully controlled water-culture experiment reported by Chapman and Rayner (1951), the earliest evidence of phosphate lack was denoted by fruit symptoms only.   The trees themselves showed no obvious foliar or growth symptoms (fig. 3-46).   Foliage was dark green and abundant, new cycle growth was abundant, bloom was normal, and yields of fruit were good.   However, some of the fruits from these trees were spongy, lacked firmness, and had hollow centers.   These conditions characterized a high percentage of fruit from trees moderately or acutely deficient in phosphorus.   Where such conditions are encountered, supplemental leaf and soil analyses should be made, and, if these values are on the low side, trials with phosphate fertilizer should follow.
      Leaf Analysis Values.—Although there are seasonal trends in the per cent of total phosphorus in the dry matter of leaves, the phosphorus content of leaves from trees not lacking this element usually show 0.10 per cent or more of total phosphorus at all seasons.   On the other hand, three- to seven-month-old leaves from phosphorus-deficient trees usually show values of less than 0.10 per cent of phosphorus.   Table 3-28 shows the monthly values for total phosphorus in spring-cycle leaves picked from bearing trees growing in solutions of graded and maintained phosphorus content (Chapman and Rayner, 1951).   It will be noted that the leaves from trees acutely deficient in phosphorus were always under 0.10 per cent; the leaves from trees showing the very earliest stage of phosphorus deficiency showed values under 0.10 per cent only in the spring months; whereas, trees amply supplied rarely showed values under 0.10 per cent.   Leaf analyses reported by Aldrich and Coony (1952) and Embleton et al. (1952) show total phosphorus values in orange leaves from phosphorus-deficient trees below 0.10 per cent except in young April-picked leaves.   With the May sampling, however, the orange leaf values had dropped to 0.08 per cent of phosphorus, and they remained below the 0.10 per cent value for the remainder of the year.   Associated with the low phosphorus content of leaves from phosphorus-deficient trees is a lower-than-average calcium content and higher nitrogen and potassium values (table 3-29).   Other workers have reported similar findings.
      In addition to the analyses already reported here, Haas (1936a, 1947) has reported many analyses for phosphorus in fruit, flowers, and other plant parts.
      Soil Analysis Values.—Aldrich and Buchanan (1954) reported that the water-soluble phosphate (PO4), as determined by the Bingham (1949) method, usually had less than 0.3 ppm of phosphate in the citrus orchard soils showing phosphorus deficiency symptoms.   However, they found some orchards low in phosphorus where no signs of phosphorus lack were apparent, and, conversely, one phosphorus-deficient orchard where the water-soluble phosphate was 3.21 ppm.   In Florida, Reuther et al. (1949) got no yield response to phosphate in a soil showing 2.8 ppm of water-soluble phosphorus.
      Acid-soluble phosphorus, as determined by the Truog (1930) method, was found by Aldrich and Buchanan (1954) to be less than 30 ppm of phosphorus in orchards where phosphorus deficiency was evident.   However, they found one phosphorus-deficient lemon orchard on a soil showing 226 ppm of acid-soluble phosphorus.   In Florida, Forsee and Neller (1944) reported 8 pounds per acre of acid-soluble phosphorus (method of extraction not stated) in an orchard responding to phosphate.   Reuther et al. (1949), using the method of Truog (1930), reported acid-extract values of 85 ppm of phosphorus in an orchard soil where no yield response was obtained.
      As regards total phosphorus, Aldrich and Buchanan (1954) usually found less than 300 ppm of phosphorus in soils where phosphorus deficiency of citrus was present.   But again there was an exception: one phosphorus-deficient orchard soil contained 1,045 ppm of total phosphorus.
      From the limited data available, it is evident that soil analysis alone is not an infallible guide to citrus phosphorus adequacy.   However, where tree condition and tissue analyses suggest the possibility of phosphorus deficiency, determinations of total, water-soluble, acid-soluble, and 0.5M-sodium bicarbonate-soluble phosphate in soils may prove useful in helping to confirm the diagnosis.
      Causes of Phosphorus Deficiency.—The causes of phosphorus deficiency have been exhaustively investigated, and only a few comments will be made to indicate something about the multiple conditions under which this deficiency occurs.   Low total phosphorus supply, the forms in which the phosphorus is held, presence of excess lime, competition by soil organisms for available supplies of phosphorus, too much nitrogen fertilizer (perhaps associated with high-soluble calcium), scion-rootstock combinations, climatic factors, insufficient magnesium, and soil moisture deficits are all possible factors.
      Because of the variability of soils and plants, and of plant requirement and feeding power, no one of these conditions automatically implies phosphorus deficiency.   The total phosphorus of soil varies widely.   In some soils values as low as 0.004 per cent of phosphorus are found, and, in others, values over 0.5 per cent occasionally occur, but the majority of soils range from 0.01 to 0.18 per cent.   Low total phosphorus does not necessarily mean that responses to phosphate will be obtained, nor does low water- or acid-soluble supply in itself always imply a need for supplemental phosphorus.
      However, in the majority of soils requiring supplemental phosphorus, it is found that either water-soluble phosphorus is low, the supply of weak acid- or sodium bicarbonate-soluble phosphate is low, or the total supply of phosphorus is low.
      It has long been known that the phosphorus supply of soils consists of both inorganic and organic forms, the latter from plant residues and microorganisms.   Although extensive research on the mineralogical and inorganic chemical aspects of soil phosphorus has been made and it is usually possible to isolate and identify a few phosphate-bearing minerals from the coarser fractions of a soil (such as apatite), the actual isolation and specific identification of most of the inorganic phosphorus compounds of soil are still impossible.   Therefore, most of what we know about the inorganic fraction has resulted from studies of the differential solubility of soil phosphorus in water, acid, and alkali extracts; studies of fixation by hydrous iron and aluminum compounds and by various clay minerals; and, of course, studies of the calcium chemistry of phosphorus.
      It is generally agreed that in acid soils inorganic phosphate is predominantly tied up in iron and aluminum compounds of variable composition, whereas, in neutral or alkaline soils in which the prevailing constituent is calcium, most of the inorganic phosphate is in various calcium forms.   In both acid and alkaline soils, some phosphorus is absorbed by the clay minerals, and some investigators believe the iron, aluminum, and calcium forms may also be tied to clay minerals.
      The nutritional researches of Parker (1927), Tidmore (1930), and Parker and Pierre (1928) established that some fast-growing plants (such as corn, barley, sorghum, and tomato) can achieve maximum growth if solution concentrations are maintained at from 0.10 to 0.50 ppm of phosphate.   However, it was noted by Parker (1927), Parker and Pierre (1928), and Tidmore (1930) that the displaced soil solution of some productive soils had less than these concentrations of phosphate in the soil solution, implying that root-soil contact feeding also is important in phosphorus nutrition.   Working with citrus, Chapman and Rayner (1951) found that maintained concentrations of 2.5 to 3.5 ppm of phosphate in solution cultures were insufficient for most citrus needs, and yet citrus could secure adequate phosphorus from soil where quick-growing field crops failed, and where soil solution values were much lower than the 2.5 to 3.5 ppm range in the water cultures.   This suggests that root-soil contact phenomena are important in the phosphorus nutrition of citrus.   Additional support for this belief is found in the total and water-soluble phosphate found in the soils of citrus orchards which do and do not respond to phosphorus fertilization.   The soils in which responses to phosphate occur usually show less than 0.3 ppm of phosphate in water extracts, though some exceptions have been found.   In these soils, possibly special rootstock-scion interrelations or some associated soil condition limiting phosphate uptake by the plant may have intervened.
      While not a great deal of specific information is available on rootstock effects in relation to phosphorus nutrition, Aldrich and Buchanan (1954), studying a soil containing 3.2 ppm of water-soluble phosphate by the Bingham method, 226 ppm of acid-soluble phosphorus, and 1,045 ppm of total phosphorus, found that lemons on rough lemon rootstock showed phosphorus deficiency symptoms, while in adjacent rows the same scion on sweet orange roots showed no phosphorus deficiency symptoms.   In a phosphorus fertilized orchard in Florida, Smith et al. (1949) showed small, but significant effects of various rootstocks on the phosphorus concentration in leaves of Valencia oranges.
      A number of field cases of phosphorus deficiency, e.g., Bouma (1959) and Embleton et al. (1952), have developed where soils had become acid due to the continued use of ammonium sulfate.
      Phosphorus is less available in many plants where moisture stresses periodically occur.   Hibbard and Nour (1959), for example, found that increasing moisture stresses decreased the leaf level of phosphorus in peach, apple, mustard, and strawberry.   Olsen, Watanabe, and Danielson (1961), working with corn, found that phosphorus uptake by corn seedlings on a relative basis was 100, 94, 50, 50, and 35 for 1/3, 1/2, 1, 3, and 9 bars of soil moisture tension, respectively.   Power, Reichman, and Grunes (1961), working with wheat in field trials, found similar moisture tension-phosphorus uptake relations.
      From what is known about phosphorus deficiency causes generally, one would look for phosphorus deficiency of citrus on the following types of soils: (1) those which are acid and in which water or weak acid extract show low values; (2) soils which have low total supplies, i.e., less than 0.03 to 0.05 per cent of phosphorus; (3) soils which contain free lime; and (4) soils which have a long cropping history, irrespective of soil type.   Modifying influences would be rootstock, climate, moisture, biological activity, and chemical factors of the soil such as soluble calcium, nitrogen, and zinc.
      Control of Phosphorus Deficiency.—Although nearly all soils fix phosphorus, it has not proven difficult to correct phosphorus deficiency of citrus by broadcast soil applications of soluble phosphate fertilizers.   The bulk of the feeder roots of citrus are found in the surface 18 inches of soil; in orchards where noncultivation is practiced, many feeder roots are commonly found two to four inches under the soil surface; and in cultivated orchards, many roots are found immediately below the cultivated zone.   Except with soils of unusual fixing power, there is some slow migration of phosphorus downward, especially in neutral or alkaline soils.   In a detailed study of phosphorus status of citrus orchards made by the author some thirty years ago (Chapman, 1934), it was found that substantial movement of phosphorus had taken place into the second 6 inches of most orchard soils and, in most, some movement into the second and even third foot.   These were in orchards which had had repeated applications of manures and/or other phosphate-carrying fertilizers.   Organic matter seemed to aid in the rate of phosphorus movement, as heavily manured soils showed deeper penetration than those not receiving organic matter.
      Pratt, Jones, and Chapman (1956), studying phosphorus distribution in an irrigated orchard soil after twenty-eight years of using superphosphate and/or manure, found 80 per cent of the applied phosphate in the top 12 inches of soil, but there was evidence of some movement beyond the 24-inch depth.   They also noted a buildup under the trees due to leaf drop.   There is usually root development in the area beneath the trees, even though under irrigation agriculture it is drier than the irrigated areas.   However, there is always some lateral movement of moisture from furrows at the drip of the tree, and, when rains occur, good moisture conditions then prevail under the tree.
      Because of the known fixing power and slow rate of phosphorus movement in soils, Aldrich and Coony (1952), in seeking to quickly correct phosphorus deficiency, applied 20 pounds of ammonium phosphate (11-48) per tree, broadcast around the dripline (skirt) of the trees.   (This amounted to 9.6 pounds of P2O5 per tree.)   Within four months marked responses to the phosphate occurred.   The soils of these experiments were moderately acid sandy loam and loamy sand types.   Leaf analyses made six months to a year following treatment showed increased phosphorus absorption by the tree.   Embleton et al. (1952) found that phosphoric acid (75 per cent), applied at rates of 16 pounds per tree per year, corrected phosphorus deficiency, as did treble superphosphate.   The probabilities are that lower amounts of phosphate would suffice, probably 5 pounds of P2O5 per tree or less.   Fuller and Hilgeman (1955) applied radioactive phosphoric acid at rates of 5 pounds of P2O5 per tree to a circular basin 6 feet in diameter around navel orange trees and noted absorption of 19 to 24 per cent of the radioactive phosphate in a few months.   The soil was a calcareous clay loam.   There was no significant difference in absorption rate by rough lemon and sour orange rootstocks.
      Nakama et al. (1962) applied 79.7 gm of superphosphate tagged with 470 μc of P32 to a nineteen-year-old satsuma tree growing on a terraced gravelly sandy soil.   The application was in July, and by November 20, 0.5 per cent of the P32 applied had been absorbed by the tree.   Some 27 per cent of the P32 was retained in the upper 9 cm of soil.   There had been 1,085 mm of rain during this period.   The rest of the phosphorus must have migrated to lower soil depths, but the P32 was so diluted as not to be detectable.
      It is also possible to secure some correction of phosphorus deficiency by sprays containing soluble phosphorus (Sato, Ishihara, and Harada, 1954).   But the correction is temporary, necessitating repeated sprayings, and spray residue encourages insect buildup and involves additional machine traffic in orchards.   Therefore, except in unusual circumstances, reliance on phosphate sprays to correct deficiency of this element is not recommended.
      As to the form of phosphorus best suited for soil applications, soluble forms such as treble superphosphate or ammonium phosphate are recommended.   From special penetration trials, it appears that the former may move a little more readily than the latter (Chapman, unpublished data).   Applications of P2O5 at the rate of 5 or more pounds per tree around the drip of the tree will usually result in correction for several years.
      In calcareous soils where phosphorus deficiency is due to the low solubility of phosphorus, Chapman (1936) found that the use of soil-acidifying agents, such as sulfur and ammonium sulfate, would increase the availability of phosphorus, provided these materials were present in the area where absorbing roots are present.
      DeRemer and Smith (1961) have suggested from work with various chelates in sand cultures that these compounds can react with insoluble iron and aluminum phosphates to release soluble phosphate.   In acid soils, when phosphorus is tied up in iron and aluminum forms and not readily available, chelates might prove very useful.
      Excessive applications of phosphate should be avoided because of the effects on the availability of other nutrients.

Phosphorus Excess
      Unlike potassium, boron, and some other elements, phosphorus is not excessively absorbed to the point of producing direct injury symptoms.   It does accumulate in the plant to some degree under conditions of nitrogen and zinc deficiencies.   In early work on inorganic phosphate in citrus, Chapman (1935) found especially high concentrations of PO4 in both leaf and woody tissues of zinc-deficient trees.
      The chief ill effects of high phosphorus are on the availability and/or absorption or utilization of other elements.   There is now abundant evidence that phosphate in soils can decrease the availability of copper, zinc, boron, iron, and other micronutrients and may affect nitrogen nutrition unfavorably.
      With respect to nitrogen, Chapman and Brown (1941a), in a fertilizer experiment with orange trees in 55-gallon oil drums, found that heavy phosphate applications produced nitrogen-starvation symptoms.   Lilleland (1933) noted similar effects on deciduous fruit trees.   Anderssen (1937) and Anderssen and Bathurst (1938) noted reciprocal relations between nitrogen and phosphorus, high nitrogen trees having low phosphorus and low nitrogen trees having high phosphorus.   There is much evidence to support the view that the ill effects of too much nitrogen, as it affects fruit quality, can be offset by increased phosphorus.   In field experiments, Heymann-Herschberg (1952) found less nitrogen in leaves where phosphorus had been applied.
      With respect to micronutrients, Chapman, Vanselow, and Liebig (1937) found that in controlled cultures zinc deficiency was accentuated by high phosphorus, and West (1938) found in field trials involving superphosphate that zinc deficiency was worse in plots receiving this fertilizer than in plots not getting phosphate.   Results with many crops, both in the field and in pots or controlled cultures, are in harmony with these findings and have been noted by others: Bathurst (1945); Reuther and Crawford (1946); Mowry and Camp (1934); Chapman et al. (1940); Thorne and Wann (1950); Labanauskas, Embleton, and Jones (1958a); Loneragan (1951); Millikan (1947a); Rogers and Wu (1948); Bingham and Martin (1956); Bingham et al. (1958); Burleson, Dacus, and Gerard (1961); and Langin et al. (1962).   However, some workers have found no significant interaction: Boawn, Viets, and Crawford (1954); Thorne (1957); Seatz, Sterges, and Kramer (1959); Bingham (1963) and Smith, Scudder, and Hrnciar (1963).   Some believe that the phosphate effect on zinc is due more to reaction or effects within the plant or plant roots than in the soil.   Boawn and Leggett (1964), working with potatoes, found that the phosphorus-zinc ratios in the plant determined whether or not the plant was zinc-deficient.   Healthy plants had phosphorus-zinc concentration ratios of < 400, and zinc-deficient plants had ratios of >400.   It seems likely that this same effect may take place in the soil, and that zinc uptake from the soil as well as utilization within the plant depends upon phosphorus-zinc ratios in which not only zinc solubility, but competition for absorption sites on the root and carriers (chelates) in the plant and associated cation-anion relations are involved.
      Even more striking effects of phosphate on copper availability have been noted.   Forsee and Neller (1944) and Forsee and Allison (1944) found on an acid peat that phosphate increased the amount of ammoniation (copper deficiency).
      Reuther et al. (1949) found that phosphate fertilization decreased the copper content of citrus leaves, but increased the zinc and manganese contents.   In a series of experiments with soil in pots, Bingham and Martin (1956), Bingham et al. (1958), Bingham and Garber (1960), and Bingham (1963) found with many soils and several forms of phosphate that copper intake by citrus was markedly reduced by increased phosphate addition.
      In recent work with four different soils in pot cultures, Martin and Van Gundy (1963) observed that additions of calcium phosphate markedly decreased the copper content of the leaves.
      With respect to other micronutrients, Bingham and Garber (1960) reported that high applications of phosphate reduced boron intake, but increased manganese absorption.   Reuther et al. (1949) also determined that leaf manganese was increased by phosphate, and Chapman and Brown (1941a) found that manganese deficiency developed when phosphorus was deficient and disappeared when phosphate was added to soils.
      In nutrient cultures of sweet orange seedlings, Chapman et al. (1940) found that iron chlorosis was produced by high phosphorus concentrations.   Iron-phosphate interactions have been noted by others, e.g., Olsen (1935).
      Phosphate applications commonly increase molybdenum absorption (Anderson, 1956).
      Creasing of Fruit.—There is some evidence that excessive phosphorus or unbalanced nutritional conditions (especially low potassium and nitrogen) may lead to creasing, or "crinkle rind," as it is sometimes referred to.
      In the Citrusdal region of South Africa, Fourie and Joubert (1957) and Cooper (1958) reported that phosphorus fertilization seemed to increase the amount of this trouble, especially if potassium was low.   On the other hand, the latter element decreased creasing, though in all cases the amount of this disorder varied from year to year, indicating a strong climatic influence.
      Embleton et al. (1952) noted that an increase in creasing occurred on the plots of a phosphorus-deficient soil receiving phosphate.
      Haas (1950) found that high phosphorus in his cultures increased the amount of creasing.
      Anderssen (1937) noted that greater waste of fruit in storage occurred on the low nitrogen (high phosphorus) fruit.   In their study of variable phosphate supply on navel oranges under outdoor water culture, Chapman and Rayner (1951) noted in one of the years of their experiment a greater amount of creasing in some of the trees of the high phosphate group.   It is apparent that size of crop, climatic factors, and nitrogen-phosphorus-potassium nutritional relations are all involved in the creasing problem.
      Biological Effects.—In outdoor water-culture experiments with citrus, Chapman and Brown (1942) found that high phosphate encouraged a marked root infection with Thielaviopsis basicola fungus; very little infection was found in the low phosphate cultures.   By decreasing the acidity of the nutrient cultures to pH 4, it was possible to suppress the growth of this fungus even in the presence of high phosphate.
      In a study with the citrus nematode (Tylenchulus semipenetrans), Martin and Van Gundy (1963) found that high phosphate additions decreased the nematode population and that sweet orange seedlings, though depressed in growth by the nematodes, were less depressed in growth by the high phosphate than by low phosphate.   Phosphate rates were at 0, 76, 360, 900, and 1,800 pounds of phosphorus per acre.   The reduction in nematodes was quite marked at the two latter rates on three of the four soils used.
      From this brief review, it is evident that tree nutrition can be profoundly influenced by excess phosphorus, acting mostly through indirect effects on the chemistry and biology of the soil.


Potassium Deficiency
      Most of the detailed information about the effects of potassium deficiency and excess on citrus has come from controlled experiments in sand and solution cultures.   Results obtained under these conditions harmonize well with various descriptions which have emerged from field experiments.   Unlike many crops where the effects of potassium deficiency produce specific and easily recognizable symptoms (such as marginal and interveinal leaf scorch on apple and peripheral spotting on alfalfa and clovers), those on citrus are so varied and multiple that no one symptom taken by itself is sufficient to positively identify it as due to lack of potassium.   (See table 3-30 for a summary of foliage and fruit effects.)
      Reed and Haas (1923a) grew young orange trees in sand culture without potassium for seventeen months.   Only the earliest stages of potassium deficiency were produced.   Leaf size was reduced; there was some loss of chlorophyll in the leaves; and some iron chlorosis was seen.   Later, Haas (1936c) grew young citrus trees in sand cultures lacking potassium and noted some necrosis on leaves, twig dieback, and gum exudation on wood.   Some of the leaves showed resinous spots.   Bryan (1935) grew grapefruit seedlings in potassium-deficient cultures and noted retardation of growth, leaf puckering, and a fading of chlorophyll in small, irregular spots on the leaves.   Pustular brown spots also appeared in the leaves.   The branches drooped and showed a lack of rigidity.   Eckstein, Bruno, and Turrentine (1937) cite leaf puckering along the midribs and brown necrotic spots on leaves as indicative of potassium deficiency.
      Oppenheimer and Mendel (1938) grew sweet lime cuttings in sand cultures receiving variable potassium supplies and noted that where potassium was lacking there were brown spots on the tips and margins of leaves, a yellow coloration followed by necrosis of the youngest leaves and buds, premature defoliation, and bronze discoloration of branches.   The potassium levels of leaves from the plants of deficient cultures ranged from 0.42 to 0.79 per cent of potassium in the dry matter, compared with values of 1.65 to 4.14 per cent in normal plants.
      Chapman and Brown (1943) grew sweet orange, grapefruit seedlings, and lemon cuttings in sand cultures for twelve months under potash-deficient and nondeficient conditions, varying the nutrition of the two sets of cultures so as to produce potassium deficiency under widely variable nutrition as regards nitrate, sulfate, phosphate, sodium, calcium, and magnesium.   Though the onset of potassium deficiency was delayed by high sodium and hastened by high calcium and magnesium, the same fundamental growth and leaf characteristics ultimately developed in all cultures.   Reduced growth rate, weak lateral shoots, decreased leaf size, leaf puckering and curling, with a series of patterns and markings characterized by some vein clearing, yellow spots or stippling, resinous spots, and some necrosis were among the main effects noted.   Some of these symptoms are illustrated in figures 3-47 through 3-51[figure 3-47, figure 48, figure 3-49, figure 3-50, figure 3-51].
      Some of the weak new shoots showed sunken or pitted areas in the bark and wood and abscised prematurely (fig. 3-52).   There were no irregular marginal and interveinal leaf-scorch patterns and effects, as noted for so many other plants.   Leaf burn from excess sodium accumulation in the high-sodium cultures was more pronounced where potassium was deficient.   Also, boron tended to accumulate more in the leaves of potassium-deficient plants than in those receiving ample potassium.   Depriving the potassium-deficient plants of water by allowing them to wilt did not bring on any additional symptoms such as leaf scorch.   There was some transient iron chlorosis in most of the cultures.
      Analyses of leaves from the potassium-deficient cultures showed total potassium in the dry matter ranging from 0.06 to 0.25 per cent, and calcium, magnesium, and nitrogen were always higher in these than in the controls.
      To gain a more complete understanding of the growth and visual symptoms produced on bearing trees growing outdoors, as well as effects on fruit yield and quality, four-year-old navel and Valencia orange trees on sour orange rootstock were grown for six years in solutions of various potassium content by Chapman et al. (1947).   The trees were ten years old at the termination of the experiment.
      Trees originally well supplied with potassium when deprived of this element gradually showed less new growth; some of the leaves on the sunny side showed a greater tendency to bronze and sunburn, and curled a little.   In the spring at blossom time, there was a greater drop of old leaves than on the trees receiving ample potassium, but the amount of blossoms and fruit set at this stage were not seriously affected.   As a result of less new growth and greater leaf drop, the foliage became thinner, dieback of twigs occurred, and there was pronounced curling and puckering of many leaves during the second year of potassium starvation.   Such leaves became thickened, and many of them became yellowish or partly yellow, but with no one fixed or particular pattern predominating as in iron, zinc, manganese, and magnesium deficiencies.   The new growth that emerged from time to time was weak, and the leaves were undersized.   Some of the stems of the new growth were at first yellowish at the point of attachment to the longer twigs, and because of stem weakening, there was a tendency to sag and develop an S-curvature.
      In subsequent years, as the deficiency became very acute, dieback, leaf drop, sparse bloom, and yellowed, crinkled foliage became very pronounced.   There was also some iron chlorosis on some leaves.   If iron sulfate solution was painted on the leaves, green spots appeared, as occurs with typical iron chlorosis.
      Trees held at potassium levels producing only a very slight deficiency showed nothing in the way of foliage color, density, or leaf size symptoms, as contrasted to trees receiving ample potassium.   However, in the spring there was a somewhat greater leaf drop, and many of the abscising leaves were more yellowish than those that normally characterized the leaf fall at this time of year from control trees.   Photographs of trees at the very earliest stage and at the acute stage of potassium deficiency are presented as figures 3-53 and 3-54 [figure 3-53, figure 3-54].
      The chief distinguishing feature on the trees only slightly lacking in potassium was the production of smaller-sized fruit.   This feature was very consistent and, of course, even more marked on the moderate and acutely potassium-deficient trees.   Yield on the slightly deficient trees did not appear to be affected.   The fruit on these and the more potassium-deficient trees colored earlier than those on trees receiving ample potassium.
      The effects of varying potassium supplies on fruit size, as found in this six-year experiment, are shown in figures 3-55 and 3-56 [figure 3-55, figure 3-56].   Data taken from Chapman et al. (1947), relating potassium status of trees to fruit size, are given in table 3-31.
      The fruit produced on trees slightly and extremely deficient in potassium was smooth in rind texture, thin-skinned, juicy, and of good acid and total solids content (though generally lower in acid than the high potash-treatment fruit).   Excess potassium made for large fruit, coarse rinds, thick skins, coarse flesh, and poor eating quality.   Fruits and twigs from potassium-deficient and nondeficient orange trees in this experiment are shown in figure 3-57.
      The size-reducing effect of potassium lack or the size-increasing effect of ample potassium has been noted in field trials by Benton and Stokes (1925) and Benton and Stokes (1931), in Australia; Bahrt and Roy (1940), Neetles and Forsee (1941), and Reuther and Smith (1952), in Florida; Parker and Jones (1950), in California; Winnik (1950), in Israel; Deszyck and Koo (1957), in Florida on grapefruit; on Hamlin and Valencia oranges in Florida by Deszyck, Koo, and Ting (1958); and on Ponkan mandarin in Taiwan by Chu (1963).   Samson (1960) got increased sizes of oranges after applying potash in a fertilizer experiment in Surinam.
      In pot cultures of soil using lemons (on sour orange rootstock), Haas and Brusca (1955b) found that periodic additions of potassium increased lemon fruit sizes and the potassium content of the flowers, peel, and pulp.   In pot cultures in Japan, Ishihara, Hase, and Sato (1965) found increased potassium fertilization increased yield, fruit size, and keeping quality of satsuma in storage when nitrogen fertilization was adequate.   At a low nitrogen level, heavy potassium reduced yield and fruit size.
      In a sand-culture trial in Florida, Smith and Rasmussen (1959) found that adding potassium to young Valencia orange trees (on rough lemon rootstock) increased fruit sizes and rind coarseness.   Reciprocation with calcium or magnesium made no difference in the potassium response.
      Embleton and Jones (1956) have reported on field experiments in which potash application reduced orange fruit sizes.   Magnesium uptake was reduced by the potassium, and it is known that lack of this latter element will result in smaller fruit sizes.
      Many other conditions can reduce fruit sizes, such as periodic water deficits, lack of sufficient heat units, desiccating winds, lack of other nutrients, as well as virus diseases, nematodes, root-attacking fungi, and insect infestations.   Hence, it cannot be expected that increased potassium supplies will necessarily improve fruit size under all circumstances.
      However, barring the foregoing complications, the evidence is clear that low potassium (before growth or other visual symptoms appear) will reduce fruit size.   Conversely, even though the total number of fruits borne may not increase, increasing potassium beyond sufficiency levels will increase fruit size.   Excess conditions decrease quality by producing fruits of coarse, thick rinds and poor internal character and quality.
      The size-increasing effects on citrus produced by increased uptake of potassium have also been noted on many other fruits, such as apples, pears, peaches, grapes, and berries.
      Since potassium alone accounts for over 40 per cent of the total ash of citrus fruit, this may be one reason why, more than any other element, lack of potassium causes reduction in fruit size.
      In addition to the decreased fruit-size effects noted, lack of potassium has been found by Koo (1961) to be involved in the fruit-splitting problem.   Since lack of this element produces smoother, thinner rinds and is involved in the water economy of the plant, it is reasonable to suppose that splitting might well be involved.   Climatic factors are doubtless involved as well.
      There is also evidence that low potassium (and high phosphorus) may account for creasing in oranges and poor keeping quality.   Fourie and Joubert (1957), in South Africa, and Sites and Deszyck (1953), in Florida, have noted such relations.
      Phosphorus accumulation in the tree, due in part to lack of potassium, may also play a part.   Creasing is much more prevalent in some years than others and, like the splitting problem, suggests that there are important interrelations with climate.
      Lack of potassium is thought to make citrus trees more susceptible to drought, damage from desiccating winds, and frost.   Reports from Russia by Marchaniya (1957) indicate that cold resistance is increased by potassium fertilization.
      Although the size and quality effects mentioned above appear prior to yield reductions, it is now apparent that there are many situations where if the nutrient disorders brought on by potassium fertilization are taken care of yield responses to potassium will be obtained (e.g., magnesium deficiency in particular and, in some cases, manganese and zinc deficiencies).
      Aldrich and Coony (1952) got yield responses to potassium on lemons in six different California locations.   In all of these, the potassium leaf content was below 1 per cent in the dry matter.
      Leaf-analysis surveys by Chapman and Fullmer (1951a, 1951b) and by Haas (1949b) suggested that beneficial effects from potassium on both yields and fruit size may occur in many California citrus orchards.   Subsequent field trials by Embleton, Jones and Page (1966) have confirmed these indications.
      In an experiment with grapefruit in Florida, Sites (1951) reported yield responses from potash fertilizer and found that fruit drop was decreased and that trees were more resistant to stormy winds.   Sites (1951) and Deszyck and Koo (1957), reporting on fruit quality in their experiments, found that potassium increased soluble solids, acid, and vitamin C.   Fruit sizes and rind coarseness were increased by potassium.   In other experiments with Valencia and Hamlin oranges, Sites and Deszyck (1953) and Deszyck et al. (1958) reported yield responses to potassium.
      In the latter years of a field experiment at Nelspruit, South Africa, Naude (1954) obtained yield responses with Valencia oranges from the application of potash fertilizer.   Fruit drop in autumn was quite marked in plots not receiving potassium.
      In Australia, Arnot (1946, 1947) reported potassium deficiency symptoms of citrus on a low-lying, gray, sandy soil.   Reduced yields, weak new growth, pale small leaves, dieback, pitting and necrosis of bark on young shoots, defoliation, and a low potassium content of foliage were the chief effects.   In the Sydney area of Australia, Levitt (1947) reported yield increases in young citrus trees from the application of 5 pounds of potassium sulfate per tree.
      Bahrt and Roy (1940) reported yield increases from potassium on a Florida soil.
      Uexküll and Kämpfer (1963), reporting on a fertilizer experiment in Surinam (Dutch Guiana) by Samson (1960), indicated that substantial yield increases were obtained from both potassium and magnesium applications to a sandy soil of the coastal plain.   In addition, fruit sizes and keeping quality in storage were increased by potassium.
      In Taiwan, Chiu, Lian, and Hsu (1960) reported yield increases from potassium on three soil types—a yellowish-brown, clay loam, an alluvial, silty loam, and a latosol clay loam.   These experiments were on trees nine to twenty years old, and they received potash (K2O) at rates of 500 to 600 gm per tree.   A similar yield response was noted in Taiwan by Chu (1963) with six-year-old Ponkan trees in the field.
      Leaf Analysis Values.—The first comprehensive study of citrus leaf analysis as a diagnostic tool with special reference to potassium was made by Chapman and Brown (1950).   On the basis of controlled cultures, field experiments, and a survey of the literature, the data of table 3-32 were prepared.   These data, based on the analysis of three- to seven-month-old bloom-cycle leaves from fruit-bearing branches, require little or no modification today.   But it is emphasized that if similar-aged leaves (bloom-cycle leaves) are taken from nonfruiting terminals, then the following values for per cent potassium in the dry matter of Valencia orange leaves would be more definitive.   These are data from Reuther et al. (1958):

                                                                        Percentage of
                                                                         Potassium in
  Condition                                                         Dry Matter   
   Deficient                                                          <0.6
   Low                                                                   0.7 to 1.10
   Optimum                                                            1.2 to 1.70
   High                                                                   1.8 to 2.30
   Excess                                                             >2.40

      A more recent tabulation by Embleton, Jones, and Platt (1967) is shown in figure 3-58.   This summarizes in a general way the combined experience and experimental results of many who have been working on potassium nutrition.   The direction of changes are for the most part valid, but one needs to avoid a too close or literal interpretation of leaf analysis values at which changes in yield, size, quality, or other characters may occur.
      Deficiencies of nitrogen, phosphorus, calcium, magnesium, and zinc are known to cause substantial accumulations of potassium in leaves.   In determining potassium status and the need for potassium fertilization, leaf analyses for these and other elements need to be made so that the nutritional balance and status can be correctly evaluated.
      Koo (1963) suggested that on July-picked, spring-cycle leaves from nonfruiting terminals levels of 1.2 per cent of potassium in the dry matter for Hamlin and 1.5 per cent of potassium for Valencia oranges are best for maximum total solids in the fruit.   However, too much potassium can decrease solids, as has been shown by Willson and Arey (1959).   Also to be borne in mind is the fact that there are significant year-to-year variations in trees receiving the same treatment.   This was noted by Chapman and Brown (1950), Deszyck and Koo (1957), and Willson and Arey (1959).
      Potassium in Other Plant Parts.—Chapman and Brown (1950) reported analyses of wood, bark, roots, and fruit in relation to potassium status.   Although all plant parts showed differences in relation to supply, they found that leaves were probably best used for diagnosis and fertilizer guidance work.   Haas (1948a, 1948b, 1949b) has reported much data on flowers, leaves, rootlets, and fruit composition.   Koo (1963) reported that the potassium of fruit reflected well the varying potassium status of trees.
      Causes of Potassium Deficiency.—Potassium apparently performs many functions in the plant.   It is an activator of enzymes, promotes growth in meristem tissues, neutralizes organic acids, and activates stomata opening and closing.   When potassium is lacking, respiration increases; carbon dioxide assimilation decreases; carbohydrates first increase, then decrease; proteins decrease; amino acids increase; and mineral nutrients such as calcium, magnesium, iron, phosphorus, and nitrogen increase.
      As with other elements, it is apparent that deficiency can result from unbalanced nutrient conditions within the plant and failure of plant roots to secure enough potassium from the soil or at a pace (in periods of maximum need) sufficient to produce best performance.
      A complete account of what is known on the subject cannot be given here, and only a few soil factors will be mentioned.
      Potassium exists in soil primarily in silicate mineral forms: (1) a part is easily exchangeable, being located on the platy surfaces and broken edges of a number of clay minerals, and (2) a part is not readily exchangeable or in "fixed" form, being located in less accessible parts of clay minerals.   The mineralogical makeup of soils is an important factor in the distribution of potassium between exchangeable and nonexchangeable forms.
      While few cases of a marked field deficiency of potassium on citrus have come to light, experience with many other plants shows generally that the most important cause of potassium deficiency in plants is a low content of easily exchangeable potassium and/or a very low total content of potassium in the soil.   Plants can get some potassium from "nonexchangeable or difficultly exchangeable forms" (and this varies, too, depending on the kind of mineral clay which holds the potassium in "difficultly exchangeable" forms), but in many soils the difficultly exchangeable or fixed potassium cannot be obtained at a rapid enough rate to meet plant needs.   It is, therefore, the more readily obtainable, exchangeable potassium which constitutes the important immediate source for plants.   Exchangeable and nonexchangeable potassium in soils is in equilibrium, which is influenced by moisture content, plant withdrawal rates, temperature, and other factors.   A good review of the subject has been given by Reitemeier (1951), and more recent work by Mortland and Ellis (1959), Mortland (1961), and Lai and Mortland (1961), has contributed to a better understanding of the mechanisms involved in potassium-diffusion processes in soils.
      The calcium and magnesium status of soils is of undoubted importance in the potassium nutrition of citrus.   Reuther, Smith, and Specht (1949), in a leaf analysis survey of citrus in Florida, Texas, Arizona, and California, noted that the potassium leaf levels were prevailingly low in California and Arizona compared with Florida.   Among the soil differences which might account for this are the much higher calcium and magnesium contents of California and Arizona soils.
      Fudge (1946) and many others have noted that increased calcium and magnesium in soils will decrease potassium uptake to some extent.   Welch and Scott (1961) have shown that the ammonium ion can reduce the release of difficultly exchangeable potassium to plants.
      Potassium has long been associated with the water economy of plants.   Its availability in soils is decreased by lack of soil moisture, and a good supply of potassium in plants tends to offset to some extent the effects of drought.   Thus, lack of moisture in the soil increases potassium need, both through effects in the soil and in the plant.   Increasing potassium levels in tissues of citrus have been shown to be associated with increasing succulence (Smith et al., 1953).
      Prolonged cropping will gradually deplete the available potassium of soils originally well supplied.   In a fifteen-year lysimeter experiment, where the only fertilizer was calcium nitrate to provide 200 pounds of nitrogen per acre per year, Broadbent and Chapman (1950) found that there had been a net loss of 3,552 pounds of potassium and a net gain of 6,189 pounds of calcium on a per acre basis.   Under these conditions, it is plain that potassium would sooner or later be needed.
      Effect of Rootstock.—There is insufficient data from field experiments and controlled cultures to assess the rootstock effect in relation to potassium adequacy, but differences in potassium composition of flowers from Eureka lemon and Valencia orange grown on various rootstocks were reported by Haas (1948b).   He found that high potassium in flowers of both species occurred where tangelo and grapefruit stock were used, as compared with lower values with sour orange and rough lemon.   Values ranging from 1.1 to 1.8 per cent of potassium in the dry matter were found.   Further analyses of fruit peel, leaves, and rootlets of oranges, grapefruit, and lemon for potassium, calcium, magnesium, and phosphorus were reported by Haas (1948a).   In peel, potassium was commonly high when trees were grown on trifoliate stock, and low on rough lemon.   There were marked differences in composition generally in relation to rootstock, indicating an important effect on nutrition.   However, the relation of these findings to soil potassium levels and deficiency and excess effects requires further exploration.
      Smith, Reuther, and Specht (1949) found that grapefruit and sweet orange roots induced higher leaf potassium than citrange roots.   Gorton, Cooper, and Peynado (1954) found that Webb Red Blush grapefruit absorbed less potassium (and more calcium) when grown on Cleopatra mandarin rootstock than when grown on sour orange rootstock.
      Relation of Potassium Deficiency to Disease Resistance.—Although the author is unaware of specific data for citrus, there are many diseases that are aggravated by a lack of potassium.   A review and discussion of these diseases and other soil fertility conditions affecting plant diseases is given by McNew (1953) and Chapman (1964).   Some of the diseases aggravated by lack of potassium are angular leaf spot of tobacco (Pseudomonas tabaci), "take-all" disease of wheat (Ophiobolus graminis), cotton wilt (Fusarium vasinfectum), powdery mildew of cereals (Erysiphe graminis), cereal rusts (Puccinia glumarum, P. graminis, and others), tobacco mosaic virus, and cabbage yellows (Fusarium oxysporum, F. conglutinans).
      Control of Potassium Deficiency.—While it is possible to easily correct moderate to acute potassium deficiency with applications of potassium sulfate or potassium chloride (1 to 2 pounds of K2O equivalent per tree), it requires time on many soils to substantially increase leaf levels from soil applications.   The author and his colleagues made yearly leaf analyses of a large number of California orchards where 5 to 10 pounds of potassium sulfate was broadcast yearly per tree.   A summary of these data are presented in table 3-33.   They show that in many cases rather large amounts of potassium fertilizer must be applied over a period of years to substantially increase potassium levels in the tree.
      In a citrus fertilizer experiment at Riverside, California, analyses of leaves from plots receiving 1 pound of K2O per tree from potassium sulfate from 1927 to 1940 showed an increase in leaf potassium from values of around 0.62 to 0.65 to 1.07 to 1.18 per cent in dry matter.   The soil was a rather compacted, sandy loam receiving about 30 to 36 inches of irrigation water a year and an annual average rainfall of about 11 inches.   Trees in plots receiving dairy manure at rates supplying 3 pounds of nitrogen per tree showed potassium values in leaves up to 1.56 per cent.   Because of the ameliorating effects of organic matter on soil structure, it appears likely that with improved permeability to water, potassium penetrates into dense citrus root areas more rapidly when organic matter is used as a source of potassium or along with inorganic sources.   The author has repeatedly noted marked citrus feeder root development in and around organic manure masses in soils.
      Recommendations of desirable annual application of potassium on mature orchards vary from 1 to 4 pounds of K2O per tree per year or from about 100 to 400 pounds per acre.   The latter rate is higher than needed for maintenance, once a low potassium condition has been corrected.
      Table 3-13 indicates that 20,000 pounds of fruit will remove about 63 pounds of K2O.   Heavier average yields would require proportionally more potassium.
      As stated earlier, the clay fraction of soils fixes potassium both in exchangeable and nonexchangeable forms; potassium movement downward is usually slow, and leaching losses are negligible.   In a loam soil of 7.0 meq/100 gm exchange capacity, Broadbent and Chapman (1950) found that yearly losses by leaching of potassium from a 4-foot depth of soil over a fifteen-year period ranged from less than 1 pound to not much over 2 pounds of potassium per acre per year.
      In a sandy Florida citrus soil of very low clay content, Kime (1944) found that potassium moved quite readily.   In this kind of soil, it would be very easy to correct potassium deficiency.
      In a recent study of the possible use of foliar sprays, Page, Martin, and Ganje (1963) found that the potassium of citrus leaves could be substantially increased by the use of 0.5M potassium nitrate (42 pounds per 100 gallons).   They used Triton B-1956 (0.25 per cent v/v) spreader.   They applied approximately 3 gallons of the potassium nitrate spray per tree per treatment to ten-year-old Valencia oranges in May and June and were able, with one spray treatment, to increase the potassium content of the leaves by 0.32 per cent in the dry matter.   Only very minor leaf injury occurred.   The calcium, magnesium, and sodium contents of the leaves were not affected.
      More work is under way to determine the safest and most effective method of using potassium sprays, but the foregoing results show that potassium deficiency and low potassium status can be quickly remedied by the use of one or more foliar spray treatments.   Treatments can then be followed by maintenance soil applications.

Potassium Excess
      Potassium is readily absorbed by citrus and other plants, and characteristic toxicity symptoms have been produced under controlled culture conditions.   However, under field conditions, because of the high fixing power of soils for this element, high concentrations of potassium seldom are encountered in soil solutions; hence, absorption of this element in amounts sufficient to decrease growth and produce leaf injury has not been reported.   However, as noted previously, adverse effects on fruit quality (such as fruit coarseness, delayed maturity, elevated acid, and lowered total solids) are possibilities and have been noted in Florida (Koo, 1963; and others).   Under these conditions, tree growth and fruit yields are not impaired, but fruit quality is reduced.
      Another more common effect of increasing potassium accumulation in soils is the production of magnesium deficiency.   This has occurred rather widely to a moderate degree on citrus in California and has been noted with a very large number of other crops in many parts of the world.
      There is also evidence that in some soils excess potassium interferes with manganese and zinc absorption.
      In controlled culture work, Chapman and Brown (1942) found that orange trees grown in solutions of high-potassium content were more susceptible to brown-rot gummosis (Phytophthora parasitica).
      Effects of Potassium Excess.—In water- and sand-culture experiments, the author produced a characteristic type of leaf burn from excessive absorption of potassium (fig. 3-59).
      In some cases, water-soaked areas developed in old leaves, and these lesions soon became necrotic.   In other experiments, a marginal yellowing followed by necrosis took place.   The potassium in these burned leaves was very high—nearly 5 per cent of potassium in the dry matter.
      Growing trifoliate citrus seedlings in soils containing increased amounts of exchangeable potassium, Martin, Bitters, and Ervin (1959) found that growth was decreased when the percentage of potassium as of total bases exceeded 23 per cent.   The potassium in the leaves of these plants exceeded 3 per cent of potassium in the dry matter.   In similar experiments with young lemon trees, Jones, Martin, and Bitters (1957) produced depressed growth in a loam soil when exchangeable potassium was at 23 per cent.
      From the evidence available, the likelihood of direct foliage and tree injury from excess potassium accumulation in the soil (except under some alkali soil conditions) is remote.   However, unbalanced nutrition (due to effects on magnesium, calcium, manganese, and zinc absorption) does occur widely, and the possibilities of significant fruit quality impairment (rough, thick rind, lower juice, and lower total solids) are very real.   Leaf analyses for potassium, calcium, and magnesium plus visual symptoms (supplemented if need be by soil analysis and evaluation of soil-management practices) offer the best means of determining whether potassium is excessive or moving in that direction.   The leaf analysis values for potassium reported in table 3-32 are a useful guide, and reference to leaf analysis standards (Chapman, 1949, 1960; Reuther et al., 1958) will prove helpful in evaluating the total nutritional situation.


Sulfur Deficiency
      Effects of Sulfur Deficiency.—
This section deals only with sulfur as an essential plant nutrient and not with its use in ameliorating soil alkali or alkaline soil conditions.
      Because of the widespread use of sulfur-containing materials (such as ammonium sulfate, potassium sulfate, heavy metal sulfates, superphosphate, gypsum, animal manures, insecticides, and fungicides), the presence of sulfur as sulfate in irrigation waters, and sulfur's occurrence in some areas as an air pollutant, sulfur deficiency of citrus has never been encountered under field conditions.
      In a study of the sulfur balance problem in plots at the Woburn Experimental Station in England, Mann (1955) concluded that under normal rural conditions sulfur from rain and other atmospheric sources is sufficient to maintain the sulfur content of the soil.   This study was conducted in an area where barley had been grown for fifty years.   However, in nonirrigated areas of both the semiarid and humid regions, sulfur deficiency has been uncovered by investigators in many crops and soil types through the world, especially in the Northern Hemisphere: Powers (1923); Alway (1940); Conrad (1950); Coic and Lefebvre (1960); Harper (1959); Kramp (1958); McClung and de Freitas (1959); Storey and Leach (1933); Jordan and Bardsley (1958); Bardsley and Jordan (1957); Harward, Chao, and Fang (1962); Chittenden and Dodson (1961); Kamprath, Nelson, and Fitts (1957); Ollagnier and Prevot (1957); and Walker (1957).   Freney, Barrow, and Spencer (1962) recently reviewed various aspects of the sulfur problem, such as the forms of sulfur in soil, accession from the air and rain, losses from soils, transformations in soils, and methods of diagnosing sulfur status.
      The chief reservoir of soil sulfur is organic matter.   Relatively lesser amounts are found as sulfides in rocks.   In some soils, especially in the arid and semiarid regions, sulfates may be deposited by water and accumulated from the oxidation of organic forms and sulfides, and thus provide an additional supply.   These sulfate deposits are especially likely to be found in subsoil horizons.   Also, there is some evidence that calcium sulfate may be cocrystallized as an impurity in calcium carbonate (Williams and Steinbergs, 1962), and there may be insoluble barium and strontium sulfates in some soils.   Sulfate also may be absorbed by some clay minerals in very acid soils (Ensminger, 1954; Kamprath et al., 1957).
      The air contains various sulfur impurities.   Even in rural areas, rainfall may bring down annually from 1 to 12 or more pounds of sulfur per acre, depending principally upon location with respect to industrial centers and the seashore.
      As stated earlier, sulfur deficiency of citrus has never been reported in the field.   Haas (1936d) described a type of leaf chlorosis on young Valencia orange trees growing in sand culture.   Later, Chapman and Brown (1941b) produced sulfur deficiency symptoms on young, bearing, navel orange trees growing outdoors in soil cultures.   A pronounced yellowing of the new growth (the leaves were uniformly yellow as in nitrogen deficiency) was the first symptom.   Subsequent growth cycles were yellower, the leaves were smaller, and premature abscission of many affected leaves occurred.   A tendency for the production of an abundant but weak bloom was noted.   Weakened shoots died back, but no gumming of the bark took place.   The fruits produced were often somewhat undersized and misshapen, and they failed to develop the orange color characteristic of navel oranges grown in the area of Riverside, California.   The rinds were abnormally thick, and in severely affected fruits the juice sacs were shriveled and dry.   Where the symptoms were less pronounced, granulation of some sections of the fruit was noted.   All parts of the tree were abnormally low in sulfur content, with nitrogen above average.   Guest and Chapman (1946, unpublished data) found that leaf microinjections of calcium, sodium, or potassium sulfates caused the leaf to become green in an area surrounding the point of treatment (fig. 3-60).   Some effects of sulfur deficiency are shown in figures 3-61 through 3-64 [figure 3-61, figure 3-62, figure 3-63, figure 3-64].   Recovery of these plants was rapid when sulfate was added to the nutrient medium.
      Table 3-34 presents data showing the inorganic composition of various parts of sulfur-deficient and healthy trees.   The comparative sulfur and nitrogen contents of leaves from trees deficient in these respective elements are shown in table 3-35.
      Leaves from sulfur-deficient trees are low in sulfur, ranging from 0.05 to 0.13 per cent in the dry matter, whereas the nitrogen content is high.   Commonly, leaves from nitrogen-deficient plants are low in nitrogen and high in sulfur.   Since there is considerable similarity in the appearance of trees lacking sulfur and nitrogen, leaf analysis is an effective means of differentially diagnosing the two conditions.
      Rasmussen and Smith (1958) have reported the effects of leaf age on the sulfur content of Valencia orange leaves, and their data are reproduced in table 3-36.   The data show remarkably little change with leaf age.   One-week-old leaves are slightly higher than those three weeks to eleven months old, and those at thirteen to fifteen months of age are only slightly lower.
      Diagnosis of Sulfur Status.—Since sulfur deficiency of citrus has not been encountered in the field, no special effort has been made to develop diagnostic techniques.   However, the good correlation noted by Chapman and Brown (1941b), Aldrich, Buchanan, and Bradford (1955), and others between the total sulfur of leaves and sulfur status indicates that leaf analysis and visual symptomatology are satisfactory means of determining current sulfur status.   There is quite a little evidence with other plants that sulfate-sulfur is a better index of sulfur status than total sulfur (Eaton, 1966), and this may prove true for citrus.
      The increasing incidence of sulfur deficiency in soils has led to considerable research on the development of soil test methods.   Phosphate-extraction methods of Ensminger (1954), Fox Olson, and Rhoades (1964), and Bardsley and Lancaster (1960) appear promising.   While leaf analysis and other plant criteria can reveal current sulfur status, soil availability tests make possible predictions of sufficient reliability to warrant field testing.
      Causes of Sulfur Deficiency.—As stated earlier, the main source of sulfur in soils is in the organic fractions,3 with lesser amounts present as sulfides and sulfates.   The nonsulfate forms gradually become converted to soluble sulfates, and, with crop removal on the one hand and leaching losses on the other, sulfur deficiency would soon develop but for the addition of sulfate-containing fertilizers, organic manures, pesticides, fungicides, irrigation waters, or additions from the atmosphere.
      Though specific cases of sulfur deficiency of citrus in the field have not yet been reported, they should be looked for in nonirrigated citrus areas and especially on soils low in organic matter, such as many of those found in the tropics and high rainfall regions.
      As to factors affecting sulfate absorption and utilization, some data with citrus and other plants have been reported.   These data are briefly reviewed in the following subsection.
      Effects of Other Nutrients on Sulfate Absorption.—In carefully controlled nutritional experiments, Chapman and Liebig (1940) found that increasing nitrate supplies did not depress sulfate absorption.   Rasmussen and Smith (1958) got similar results with Valencia orange on rough lemon rootstock.   The sulfur content of leaves of trees fertilized at three rates with ammonium nitrate (0.75, 1.5, and 3 pounds of nitrogen per tree) showed, respectively, 0.33, 0.34, and 0.34 per cent of sulfur in the dry matter.   However, under conditions of sulfate excess, Foote and McElhiney (1937) found that extra nitrogen fertilizer seemed to decrease sulfur-excess effects.   Haas and Thomas (1928) noted similar effects in pot tests.
      Jones et al. (1963) are of the opinion that under conditions of high sulfate extra nitrogen is helpful.   Perhaps the true explanation of these somewhat conflicting results is that when nitrogen is somewhat deficient, sulfate tends to accumulate in leaves and produce the typical, excess-sulfur leaf pattern.   Chapman and Liebig (1940) noted a higher leaf sulfur content in nitrogen-starved sweet orange seedlings than in those receiving ample nitrate.
      With regard to phosphorus, Rasmussen and Smith (1958) likewise found that rates of phosphate fertilizer, ranging from 0 to 4.5 pounds of phosphate (P2O5) per tree and under conditions of equal sulfate supply, did not have any effect on sulfate uptake.   Chapman and Brown (1941a) got essentially the same results with phosphate in their fertilizer tests with navel oranges growing under phosphatic fertilization.
      Effects of Soil Moisture.—Dutt (1962b) found with sugar cane that rain and irrigation produced beneficial effects on sulfur-deficient plants and drew the inference that drought reduces soil-sulfur absorption.   This appears plausible, since similar findings have been noted with phosphorus, potassium, and other nutrients.   Among other things, the nutrient diffusion rate in soil films is reduced as moisture decreases.
      Effect of Rootstock.—Though not much data with citrus are available on sulfur uptake and leaf composition as affected by rootstock, Rasmussen and Smith (1958) found with Valencia and Parson Brown scions on grapefruit and rough lemon rootstocks, respectively, that under the same culture conditions the leaf sulfur of the two scions was 1.14 and 0.91 per cent of sulfur in the dry matter, as compared with 0.43 and 0.41 per cent of sulfur for scions grown on rough lemon stock.   Since marked effects of rootstock on leaf content of boron, potassium, nitrogen, chlorine, and other elements occur, it is reasonable to assume that sulfate uptake would also be affected.
      However, much more work is needed in this respect to determine which rootstocks would be most effective, especially where sulfate is in excess.   (See the salinity section for a more complete account of rootstock effects on salt uptake.)
      Control of Sulfur Deficiency.—With other crops, control has been easily accomplished by soil application of gypsum and other sulfate-containing fertilizers (McKell and Williams, 1960).   Elemental sulfur can also be used.   All sulfate forms are sufficiently mobile that no problem of movement to roots is involved, and gypsum is of low enough solubility that large applications can be made without much danger of toxicity to citrus.   Correction lasting for a period of years can be obtained with gypsum.
      Leaves will absorb sulfate from foliar applications, and though no field data with citrus are available, Dutt (1962a, 1962b) has reported that foliar correction of sulfur deficiency on jute and sugar cane was achieved with sprays of 0.08N sulfuric acid and 0.3 per cent solutions of copper or zinc sulfates.
      Rates for soil application on sandy soils, which have proved sufficient for correction of sulfur deficiency of crops such as clover, are as low as 16 pounds of sulfur per acre as gypsum (Neller and Bartlett, 1959).

Sulfur Excess
      Only the nutritional aspects of sulfate toxicity will be dealt with in this section.   The toxicity of elemental sulfur to foliage and fruit when used as a fungicide and insecticide will not be covered.   Turrell (1950), Turrell and Chervenak (1949), and Turrell and Boyce (1953) have investigated and reported on this subject in great detail.
      Sulfate and chloride, and to a lesser extent bicarbonate and nitrate, are the principal anions which make up the salt constituents of alkali and saline soils.   The subject of alkali and salinity is treated in a later section.   Only brief mention of the specific effects of excess sulfate will be taken up here.
      An element of uncertainty exists with respect to the specific effects of the sulfate ion per se on citrus.   The associated cation may be of greater importance in determining toxicity than sulfate.   However, it can be stated in general that the sulfate ion is much less toxic and not taken up as readily by citrus roots as the chloride ion, and with respect to the associated cation, sodium sulfate is much more toxic than calcium or magnesium sulfates.   In fact, because of its limited solubility, excess gypsum (CaSO4 · 2H2O) appears not to materially limit citrus growth or production where supplies of nitrogen and other elements are adequate.   Many of the Imperial Valley soils of California contain excess solid-phase gypsum, and (where not complicated by a high water table or other soluble salts) excellent, high-producing orange and grapefruit orchards can be found.   However, under some conditions, excess calcium sulfate can be injurious.
      Haas and Thomas (1928) grew budded lemon trees in 10-gallon crocks of quartz, to which 1 pound of calcium sulfate was added.   Nutrient solutions of increasing calcium nitrate content were added to various cultures.   Under low-nitrogen conditions, the leaves became mottled and yellowish and showed 1.57 per cent of total sulfur in the dry matter.   Addition of extra nitrate decreased the amount of leaf yellowing and mottling, but did not completely prevent it.   Injured leaves dropped prematurely.   Leaf patterns characteristic of excess sulfate are shown in figure 3-65.
      In carefully controlled sand-culture experiments with sweet orange seedlings involving nitrate and sulfate variables, no growth depression occurred from the presence of 20 meq/l of sulfate (derived from calcium, magnesium, and potassium) even at low-nitrogen concentrations (Chapman and Liebig, 1940).
      Aldrich et al. (1955) grew Eureka lemons (on sweet orange rootstock) in 3-gallon pots of an alkaline soil treated with sufficient sulfur to increase soil acidity to values of approximately pH 7, 6, 5, and 4, respectively.   In practically all cases, increased sulfur additions and soil acidity decreased growth.   The chief cause appeared to be excessive sulfate absorption by the plant and resulting leaf injury.   Leaf patterns identical to those produced by Haas and Thomas (1928) were noted, and leaf values for sulfur varied from 0.20 to 1.47 per cent of sulfur in the dry matter.   First signs of leaf patterns appeared when the sulfur content of the leaf reached 0.5 per cent.   The sulfur treatments produced marked increases in the soluble calcium and sulfate content of saturation extracts, the latter reaching values up to 68.6 meq/l in one soil treatment.   In this case, the soluble calcium, magnesium, and potassium values were 21.5, 23, and 6.8 meq/l, respectively.   The total soluble salt content in some of these soils reached conductivity values of over EC 2.0 mmho/cm, and decreased growth might have been the result not only of excess sulfate uptake, but because of the salinity of the soil solution.
      In a long-time fertility experiment at Riverside, sulfur was applied regularly to certain plots.   From 1927 to 1949, 80 pounds of sulfur per tree (or an equivalent of 7,200 pounds per acre) were applied.   Soil pH was markedly reduced in the first two feet of soil, with values somewhat less than pH 5 having been produced.   In comparable treatments with less sulfur, the reaction varied from pH 6.4 to 7.2.   Soluble salts (mostly calcium sulfate) were produced in the sulfur-treated plots, but yields were not affected by the sulfur treatment.
      Zusman (1956) grew sweet lime and sour orange seedlings in pots of sandy soil treated with increasing amounts of sulfuric acid and produced thereby sulfates in amounts varying from 381 to 2,293 ppm sulfate (SO4) in the soil.   Top growth (rated at 100) at the 381-ppm sulfate level was reduced by 50 or more per cent at the highest sulfate level, and a leaf yellowing and browning starting at the periphery was noted similar to that produced by Haas and Thomas (1928).   Total sulfur ranged from 0.35 to 0.64 per cent in sour orange leaves from trees growing in soils of increasing sulfate content, while total calcium in the leaves decreased as sulfur increased.
      Control of Sulfur Excess.—Excesses of sulfate resulting from salinity increases in soils must be dealt with by better control of drainage and irrigation practices, as outlined under the section on salinity and alkali.
      As stated earlier, the use of extra nitrogen appears to help overcome excess sulfate in soils, especially under conditions where nitrogen may be deficient.   Not enough is known about rootstock effects as yet, but the data of Rasmussen and Smith (1958) clearly indicate that orange trees on rough lemon stock may tolerate high sulfate better than when on grapefruit stock.


Zinc Deficiency
      As with manganese and boron, early workers noted plant stimulation and responses to zinc and suggested that this element might be essential: Raulin (1863, 1869, 1870); Javillier (1908, 1912, 1914a, 1914b); Steinberg (1918, 1919); Mazé (1914, 1915); Sommer and Lipman (1926); and Sommer (1927, 1928).   Final proof that zinc was an essential and indispensable element for citrus and other plants emerged from several related investigations in the 1930's.   First there was marked correction of "little-leaf " of peaches and citrus "mottle-leaf" from massive soil applications of zinc-contaminated iron sulfate.   When various zinc compounds alone were applied to the soil and as foliar sprays, there was spectacular recovery from these disorders (Chandler, Hoagland, and Hibbard, 1932, 1933, 1934, 1935; Johnston, 1933; and Parker, 1934, 1936, 1937a, 1938).   This field work was followed by controlled water-culture experiments on apricot by Hoagland, Chandler, and Hibbard (1936), and on citrus by Chapman et al. (1937).   In this controlled work, typical little-leaf of apricots and mottle-leaf of citrus were produced by reducing zinc contamination in culture solutions.
      After the California field work, Mowry and Camp (1934) were able to correct "frenching" of citrus with zinc sprays in Florida.   Thus ended a long period of investigational work into the cause of citrus mottle-leaf.   For years this disorder had plagued citrus growers in California.   While it was known to be worse on sandy and loamy soils (especially those derived from igneous rocks) than on heavier types, to be aggravated by fertilizer (such as sodium nitrate), and to be ameliorated by organic matter, still the disorder was a serious problem, and the basic cause was unknown until the aforementioned pioneer work of Chandler and others in California.
      The discovery of its true cause and practical means of control is one of the classic examples of benefit to an industry and agriculture through the persistent efforts and curiosity of research scientists.
      The striking and characteristic visual symptoms produced on citrus by lack of zinc have made it one of the easiest nutritional disorders to identify and this, plus the fact that it occurs under a wide range of soil conditions, is the chief reason why zinc deficiency has been noted in every citrus-growing region of the world.   Next to nitrogen, it is the most widespread of all nutritional disorders of citrus.
      Symptoms of Zinc Deficiency.—The most striking effects of zinc deficiency on citrus are: (1) characteristic, irregular, and chlorotic leaf spots (thus the name "mottle-leaf"); (2) small leaves on terminal growth; and (3) severe dieback of twigs.   As with manganese deficiency, the first leaf effects are a fading of chlorophyll in mesophyll areas between the veins, with the main veins, midribs, and bands or fringes adjacent to these remaining green except when the deficiency becomes very acute.   Thus, a series of irregular, chlorotic blotches and patterns appear on leaves, as illustrated in figures 3-66 through 3-70 [figure 3-66, figure 3-67, figure 3-68, figure 3-69, figure 3-70].   There is usually more complete fading of chlorophyll in the interveinal areas, with sharp yellow and cream colors appearing, than is the case with manganese deficiency.   Also, marked reduction in leaf size and a tendency for leaves to be narrow and elongated is another characteristic.   Decreased stem elongation and shortened internodes resulting in small leaf clusters produce a rosetted appearance of twigs.   In addition, the mottling with zinc deficiency is more acute on the sunny side of the tree, whereas with manganese deficiency, leaf patterns are generally more pronounced on the shady side of the tree.   Weakened terminal growth (due to the poor, weak, chlorotic, and small leaves) result in marked twig dieback, and trees severely deficient become open and bushy in appearance.
      Another characteristic of zinc-deficient leaves is the frequent presence of small green dots in the yellowed areas.   This is a further distinguishing feature not seen in manganese-deficient leaves.   The small, narrow, and pointed leaves often stand more erect (the angle made between the leaf and stem is more acute than is the case with iron- and manganese-deficient leaves).
      In many areas, zinc deficiency is more prevalent in the fall than in the spring, but this tendency varies some.   In India, zinc deficiency commonly comes on strongly during the late spring and summer; in the fall, at the cessation of monsoon rains, there is often considerable new growth, which is normal in appearance.
      Fruit production is markedly curtailed when zinc deficiency reaches an acute stage and may be a little depressed even in the early stages.   This is the reason why a fertilizer program designed to avoid the first appearance of mottled leaves is desirable.   In a study of grapefruit yields in relation to mottle-leaf severity, Parker (1937b) found little yield reduction when only a small amount of foliage was visually affected.   However, Embleton, Wallihan, and Goodall (1965) obtained evidence that some yield decrease of lemons may occur even at earliest stages of deficiency, when there are no visible symptoms and only low-zinc levels in lemon leaves.
      Fruit from zinc-deficient trees is small, and, if the deficiency is severe, oranges and grapefruit may be thick-skinned and misshapen.   In such cases, the pulp is woody, dry, insipid, and low in acid and vitamin C.   Lemon fruit is smaller, elongated, thin-skinned, and juicy.   The rind color of oranges is lighter.
      Orange trees are somewhat more prone to zinc deficiency than grapefruit or tangerine, but all types of citrus are affected.
      Budded citrus appears to be somewhat more susceptible to zinc deficiency than seedlings, but there are exceptions.   One of the worst cases of zinc deficiency the author has ever seen was on seedling mandarins in the Coorg area of India; budded orange trees in the same soil were much less affected.   The principal zinc deficiency effects on growth, foliage, and fruit are summarized in table 3-37.
      Leaf Analysis Values.—Almost without exception, leaves with zinc deficiency patterns are low in zinc.   When responses of citrus trees to zinc sprays were first noticed, Thomas, Vanselow, and Laurance (1935, unpublished data) made an extensive collection of leaves from mottle-leaf-affected and normal citrus orchards in California.   Leaves with various degrees of mottle usually showed less than 15 ppm zinc in the dry matter, whereas green leaves were usually higher.   Others have since reported analyses of leaves from zinc-deficient and nondeficient orchards (Gaddum, Camp, and Reuther, 1936; McGeorge, 1939; Healy, 1952; Sato et al., 1952; Khanduja, 1955; and Kanwar and Dhingra, 1962).   With few exceptions the results of these investigations confirm that leaves from zinc-deficient trees range from about 4 to 20 ppm of zinc in the dry matter, whereas leaves from healthy trees showing no signs of zinc deficiency usually contain more than 20 ppm.   However, there is some overlapping, as is true with most other nutritional deficiencies, and the author knows of orchards showing 10 to 12 ppm of zinc in the dry matter of nine- to ten-month-old spring-flush leaves which show no mottle.
      Zinc deficiency markedly affects some of the other inorganic constituents of leaves.   Kelley and Cummins (1920) reported analyses of normal and mottled orange, lemon, and grapefruit leaves.   Some of their data are reproduced in table 3-38.   The data show clearly that calcium is lower and potassium, phosphorus, and nitrogen are markedly higher in mottled (zinc-deficient) leaves than in healthy, green leaves.   These symptoms are usually regarded as effects of zinc deficiency rather than causes, but it is of interest that high phosphorus can also bring on or aggravate zinc deficiency.   Although the evidence is less extensive, high nitrogen and potassium are thought to aggravate this disorder (Reuther and Smith, 1950).
      Causes of Zinc Deficiency.—Zinc deficiency of citrus and other plants occurs on a wide range of soils.   Many conditions and factors can cause this disorder.   According to Swaine (1955), total zinc in soils normally ranges from <10 to 300 ppm, but much higher levels can be found in soils contaminated by industrial sources, mine wastes, and ores.
      Low total-zinc content, soil alkalinity, and high phosphorus (the latter either naturally occurring or resulting from fertilizer applications), nitrogen fertilizers (especially alkaline-forming types), organic matter, soil moisture, excess potassium, excess copper, and imbalances from other elements have all been noted as variously involved or implicated in zinc deficiency.
      Zinc deficiency is especially prevalent in acid-leached, sandy-type soils where the cause is commonly low-zinc content.   It also occurs widely in alkaline soils, and in these low-zinc solubility is the chief cause.   Both low- and high-zinc values have been reported in various peat and muck soils.   In the acid sands of Florida and other areas, liming has frequently brought on zinc deficiency (Floyd, 1917; Lott, 1939; Wear, 1956; Seatz et al., 1959).   Jurinak and Thorne (1955) studied zinc solubility in bentonite systems.   Minimum zinc solubility in acid-bentonite suspensions treated with increasing amounts of sodium and potassium hydroxide was in the pH range of 5.5 to 7.   The amounts of zinc in solution per gram of clay in this range amounts to around 1 μgm of zinc per gram of bentonite, or 1 ppm.   When treated with calcium hydroxide, minimum solubility was at pH 7.5 and above, and the amount of zinc in solution was around 0.5 ppm.   In systems treated with sodium and potassium hydroxide, zinc solubility increased at pH values above 7, probably because of the formation of zincate ions.   Zinc solubility did not increase in the calcium systems as pH increased.
      Soils derived from granites and gneiss are more prone to zinc deficiency, and, as already noted, mottle-leaf of citrus in California is much more prevalent on sandy, loamy, and gravelly soils than on heavier clay types.
      Already discussed at some length in the section on phosphate excess ([above]) are the many investigations and observations which show that zinc deficiency in some soils can be induced or aggravated by phosphate fertilization.   In field and pot work not hitherto cited, Fortini and Morani (1960) demonstrated that monocalcium phosphate (at soil application rates of 330 to 1,000 ppm of P2O5) depressed zinc uptake by tomato and spinach on soils low in zinc.   Applications of zinc sulfate at rates providing 20 ppm of zinc eliminated the phosphate-induced zinc deficiency.   Many researchers are now of the opinion that the phosphorus effect is due more to reactions within the plant than solubility effects in the soil.
      Nitrogen fertilization often increases the severity of zinc deficiency (Ozanne, 1955; Chapman et al., 1937; Camp, 1945; Reuther and Smith, 1950; Haas, 1936d).   This may be related more to associated soil pH changes or to cation increases in the soil solution rather than to the specific effects of nitrogen per se.   For example, Viets, Boawn, and Crawford (1957) found that sodium nitrate decreased zinc uptake, whereas ammonium nitrate and ammonium sulfate increased zinc uptake.   In later work, Boawn et al. (1960) found that uptake was not as good with calcium nitrate as with the ammonium forms.   However, Wear (1956) found that substantial increases in the calcium of soils (through gypsum application) did not decrease zinc uptake.
      Organic materials may either increase or decrease zinc availability.   Miller and Ohlrogge (1958a, 1958b) found that aqueous extracts of ground alfalfa and manure could complex zinc and decrease zinc uptake by corn and soybean from nutrient solutions.   On the other hand, it has been a common observation in California citrus orchards that winter cover crops (turned under in the spring) and organic manures usually decrease the severity of zinc deficiency.
      Soil Analysis Values.—Though very little soil analysis data relating the zinc content of orchard soils to citrus performance have been developed, considerable data have accumulated with other crops.   A few investigations will be mentioned here.
      Brown, Krantz, and Martin (1962) found with corn growing in potted soils that 0.55 ppm of dithizone-extractable zinc was a level below which this crop responded to zinc applications and above which there was little response.
      Shaw and Dean (1952) found for many crops that a value of 1 ppm of zinc, and less in soils of pH 7 and higher, indicated a need for zinc.   Their extractant was a solution of ammonium acetate and dithizone-carbon tetrachloride.   With soils of pH 6 and lower, 2.5 ppm of zinc was ample.
      Massey (1957), working with thirty-four silt-loam soils of Kentucky, found that zinc uptake by corn in pots was related to pH and to the amount of zinc extracted by Shaw and Dean's ammonium acetate-dithizone-carbon tetrachloride reagent.   He proposed the multiple regression formula:
                  Ŷ = 99.2 - l2.2X1 + l0.9X2,
where X1 is pH, X2 is the dithizone-extractable zinc in ppm of dry soil, and Ŷ is the calculated zinc uptake expressed in micrograms of zinc.   Where the value of Ŷ is greater than 40, no zinc deficiency can be looked for; and where it is lower, the chances are good that responses to zinc will be obtained.   A multiple correlation coefficient of 0.802 was found between zinc uptake and the pH and zinc analysis values.
      Wear and Sommer (1948), using 0.1N HCl and a 1-to-10 soil-extractant ratio, found that zinc deficiency occurred in general when soil values of zinc were less than 0.90 ppm.
      Other methods have been proposed but the aforementioned are sufficient to show that some indication of zinc status may be had from soil test methods.   As with all availability tests, the nature of the crop, associated soil conditions, climate, soil organisms, aeration, soil moisture, and other factors impose qualifying conditions of sufficient magnitude so that soil tests provide information for general guidance only.   Leaf analyses for zinc and visual symptoms will provide a more certain guide with citrus and many other crops.   However, soil analyses are especially useful in indicating potential supplies and trends toward zinc excesses.   Continued use of foliar sprays or injudicious applications of zinc salts to soils may eventually lead to iron chlorosis and other injury effects.
      Control of Zinc Deficiency.—Soil applications of zinc sulfate to correct zinc deficiency of citrus fell into disrepute in the early years of experimental work in California for the reason that while correction could often be achieved by spreading zinc sulfate in a circle of 4- to 5-foot radius from the trunk, tree injury frequently occurred.   In order to avoid injury and still get correction, the amounts applied had to be tailored to each soil type.
      When it was found that complete correction could be obtained from foliar sprays and correction would hold for one to several years, attempts to work out soil-treatment methods were abandoned.
      However, with the development of chelates, renewed research on soil application trials has begun.   Some of this work merits mention and will be discussed, along with other soil application techniques, after mention of foliar sprays in the section which follows.
      Foliar sprays.—A great deal of early work on the use of foliar sprays was carried out by Johnston (1933) and Parker (1934, 1936, 1937a, 1938), in which materials, rates, timing, lasting effect, and other matters were investigated.   It was found that zinc metal dusts, zinc sulfide, zinc oxide, zinc sulfate, and other zinc-containing materials all gave some correction.   Also, many formulations with copper and manganese sulfates, boric acid, sodium molybdate, urea, sulfur, lime sulfur, oil, and many other insecticides have been tried and are extensively used.
      Some of the many formulations currently in use or possible in California are tabulated in table 3-39.   In many cases, a single foliar application of a suitable zinc compound annually will suffice.   However, where the condition is severe or where special conditions prevail, two or more applications annually may be needed.   Best results are generally obtained by spraying when the spring-flush growth is one-third to two-thirds developed.   Poorest results are obtained in winter months or when sprays are applied to old, hardened leaves.   Timing is thus of considerable importance.
      Soil applications.—As stated earlier, soil applications of zinc sulfate in California applied in a circle around the trunks of citrus trees, while frequently effecting recovery of zinc-deficient trees, fell into disrepute owing to the difficulty of adjusting the rates so as to secure control without producing severe injury.   Also, the very clear evidence that slight excesses of soluble zinc (as with copper) can cause iron chlorosis is a further reason why soil applications have not been encouraged.   (In time, of course, residues from foliar spray may build up sufficient zinc in soils to prove toxic, especially if acid-forming fertilizers such as ammonium sulfate are used in sufficient amounts to increase soil acidity.)   In spite of this hazard, the possibility of finding suitable compounds and methods of soil applications should not be dismissed.
      In a recent experiment, Embleton et al. (1965), working with Eureka lemons on a fine sandy loam of about pH 7.3, increased foliage zinc content by applying zinc sulfate in two successive doses at a rate providing 3.6 pounds of metallic zinc per tree, the first time in a 3-inch band around the drip of the tree and five months later in six shallow holes around the drip line.   When used in combination with a foliar manganese sulfate spray, this treatment increased yields by about 125 pounds per tree over a 31-month period.   Recently matured leaves, sampled at five and seventeen months, respectively, after the last application of zinc sulfate, showed zinc levels of 23.5 to 28.7 ppm of zinc in the dry matter as contrasted with 14.9 to 20.4 ppm in the control trees.
      With respect to the use of zinc chelates, there is evidence that ZnEDTA and ZnHEEDTA are more effective sources of zinc for many plants and on many soils than zinc sulfate.   Stewart and Leonard (1955, 1957) and Leonard, Stewart, and Edwards (1957) found that the zinc of ZnEDTA would move more readily in soils than the zinc of zinc sulfate.   More zinc from ZnEDTA applications was absorbed by plants from soils of pH 6 to 7 than from soils of pH 4.5.   Soluble iron readily displaces the zinc in ZnEDTA.   This probably accounts for the poorer performance of ZnEDTA in acid soils.
      With avocados, Wallihan, Embleton, and Printy (1958) found that 1 pound of ZnEDTA per tree increased the zinc content of leaves from 15 to 50 ppm, and the effect lasted two and one-half years, but in subsequent experiments by Wallihan and Embleton, reported by Embleton and Jones (1964), not as good results were obtained.
      In comparison of various zinc carriers—ZnEDTA, zinc oxide, zinc phosphate, zinc carbonate, zinc frits, zinc granules, and two smelter byproducts, applied respectively to provide 2 ppm of zinc to a fine sandy loam in pots—Boawn, Viets, and Crawford (1957) found that zinc uptake was greatest from the chelate.   Except for the frits (from which no uptake occurred) and blast furnace slag (from which only a little zinc absorption occurred), the others were all about equal as zinc sources.
      Wallace and Mueller (1959) found that ZnHEEDTA was a better source of zinc for soybean growing on calcareous soils than zinc sulfate, but not as good as the latter on an acid soil.
      Benson, Batjer, and Chmelir (1957) found that spring applications of 1 and 2 pounds of ZnEDTA or ZnHEEDTA, applied under the spread of the branches on sandy loam soils of pH 5.6 to 9, were effective in supplying zinc to peach and cherry trees but not to apples.
      Leonard, Stewart, and Edwards (1959) and Stewart and Leonard (1963) found that a mixture of 5 pounds of zinc sulfate and 5 pounds of calcium chloride, applied in ten piles (1-foot square) around citrus trees in acid sandy soils, markedly increased the zinc uptake by the tree.   Embleton et al. (1964) tried out this technique on several citrus orchards in California.   A treatment consisting of a mixture of 2 pounds of zinc sulfate (36 per cent zinc) plus 2 pounds of calcium chloride, applied in five piles around the dripline of the tree on two-year-old Valencia oranges (on Troyer citrange root), increased the zinc content of leaves from 19 to 441 ppm in the dry matter.   The soil was a calcareous fine sand of pH 7.5 (paste), and the orchard was irrigated by a flooding system at frequent intervals.   The zinc level in the leaves from these trees was above that of the controls for two years, and, while there was some initial leaf burn, no permanent injury resulted.   Zinc sulfate alone applied to the soil, at the same rate, also produced some initial but lower zinc increases in foliage, but the effect did not last as long.
      In comparable experiments with mature Redbush grapefruit trees on the same type soil, 5- and 10-pound applications of the zinc sulfate-calcium chloride mixtures increased the zinc in leaves to values up to 327 ppm.   The effect was measurable for two years, but the zinc levels decreased to much lower amounts in time.   The 10-pound application rate produced some leaf injury and increased the chloride content of the leaves a little, but not enough to prove serious.
      On a heavier-type soil (Yolo sandy loam of pH 8), trials with 2 1/2, 5, and 10 pounds of the zinc sulfate-calcium chloride mixture, applied in six holes on two furrow sides of lemon trees, did not increase the zinc content of leaves nor produce injury.   Zinc sulfate alone at 20 pounds per tree did increase zinc levels in the trees.
      From this brief review, it is evident that with some soils and plants zinc chelates at rates of 1/2 to 2 pounds of ZnEDTA or ZnHEEDTA per tree will correct zinc deficiency.   Results with the zinc sulfate-calcium chloride mixtures have been successful in some cases and not in others.   Costs and possible toxicity suggest that soil applications are still not practical.   Foliar sprays remain the safest and cheapest means of correcting zinc deficiency of citrus trees.

Zinc Excess
      Leaf burn, defoliation, and twig dieback have been noted when too much zinc sulfate is applied to soil.   As previously stated, because of the difficulty of determining both effective and safe rates under different soil, climate, and tree conditions, soil applications of soluble forms have been largely discontinued.
      Foliar sprays formulated from soluble zinc compounds, if used at rates not exceeding 1 pound of zinc sulfate (ZnSO4 · 7H2O) per 100 gallons of water, are not apt to produce leaf burn.   In greater amounts, a precipitating agent (either sodium carbonate or lime hydrate) is needed.
      Excess zinc in the nutrient medium below the levels which cause severe root and top injury will often produce iron chlorosis.   This was first noted by Chapman et al. (1940) in a series of nutritional experiments.   Similar results have been obtained by Smith and Specht (1953) and with other crops by Hewitt (1948), and Hunter and Vergnano (1953), Millikan (1947b), and others.
      Insufficient data exist for a certain diagnosis of zinc excess in citrus, but leaf analyses values of 100 ppm of zinc in the dry matter and over, especially if accompanied by iron chlorosis, would be suggestive.   Soil analysis values well above those mentioned as critical for zinc deficiency would also be indicative.   More information on the subject is needed.
      Control of excess zinc in soils could probably be easily achieved by applications of lime and/or superphosphate.


      The sensitivity of citrus trees to excess soluble salts (salinity, alkali) and sodic salts and the effects thereof soon became apparent to early California workers.   Enough information now exists to diagnose those cases where serious injury occurs from total salt or from specific ions such as sodium, chloride, sulfate, carbonate, bicarbonate, boron, and lithium.4   However, much still remains to be learned about the influence of intermediate salinity levels as they affect citrus performance, the influence of variable nutritional factors on ion uptake, tissue and soil analyses criteria, rootstock and scion effects, moisture, climate, disease, and insect interrelations.
      The effects of boron, lithium, and sulfate excesses have been discussed in previous sections.   The information available on chloride, sodium, and total salt will be dealt with in this section.
      Saline Conditions.—At the turn of the century, Loughridge (1900) investigated certain orange orchards near Corona, California.   These orchards had been severely injured by the use of saline irrigation water from Lake Elsinore.   Analysis showed that the water of Lake Elsinore at that time contained 917 ppm of sodium chloride, 377 ppm of sodium sulfate, and 391 ppm of sodium carbonate and bicarbonate considered jointly.   Loughridge concluded that the injury was caused mainly by the carbonates.   A little later, when attention began to be focused on chloride in relation to citrus trees, Hilgard (1900) concluded that sodium chloride was the constituent of the Lake Elsinore water mainly responsible for injury to these orange trees.
      Subsequently, Loughridge (1901) directed attention to the extreme sensitiveness of the lemon tree to salt.   He said: "The lemon seems to be the least tolerant of all the fruit trees, for it was stunted by 1,440 pounds common salt (sodium chloride) per acre, distributed through 4 feet depth, and was killed by 1,900 pounds of sodium carbonate."   This would amount to 90 ppm of sodium chloride in a dry soil basis and about 120 ppm of sodium carbonate.
      Supporting the conclusions of Loughridge and Hilgard, Kelley and Thomas (1920) showed that various species of Citrus are extremely sensitive to chlorides.   Haas (1932b) showed that chlorine tends to accumulate in the leaves of citrus trees, and that the trees, when grown in culture media containing relatively high concentrations of chloride, absorb large quantities of this element.   He found that under such conditions the leaves became severely burned; moreover, an analysis showed that the injured leaves contained abnormally large amounts of chlorine.
      Since common salt (sodium chloride) is a constituent of many of the soils and irrigation supplies available in various semiarid regions where citrus is grown, the sensitivity of citrus trees to chloride is a matter of importance.   For example, Kelley and Thomas (1920) showed many citrus orchards, situated in several different counties in California, had been severely injured by the salts (chiefly chlorides) from irrigation waters.
      In discussing this problem, both Loughridge (1900) and Kelley and Thomas (1920) emphasized the fact that the injurious effects of saline irrigation water may vary greatly in different localities, and even in different parts of the same orchard.   This variability seems to be caused by differences in soil character.   Other conditions being similar, it is reasonably certain that the effects of a given supply of saline irrigation water will be determined by its effects on the concentration of the soil solution, on the one hand, and on soil permeability on the other.
      Figure 3-71 shows an orange orchard in the area of Riverside, California, completely defoliated by saline irrigation water high in chloride.   When the photograph was taken, this grove had been irrigated for about sixteen years with water containing 500 ppm of chlorine.   For several years before better water was obtained, these trees continued to show pronounced chloride injury; the smaller twigs died back each year, and some defoliation took place several different times.   The toxicity became apparent only after the orchard had been irrigated with saline water for several years.   Although the injury became severe and trees almost ceased to bear, they were not killed by the salts of the irrigation water.   After about twenty years of regular application, the saline irrigation water was replaced by nonsaline irrigation water.   The result was remarkable.   Within a year or two, the trees began to recover, and after a few years their recovery became essentially complete (fig, 3-72), though the trees never attained full size.
      Thus, it is evident that injury to citrus trees produced by saline irrigation water is not necessarily permanent, though more or less permanent stunting may result, especially when young trees are injured.   In fact, there are many citrus orchards in different citrus districts of California in which recovery from severe salt injury has been effected simply through applying nonsaline irrigation water by the usual method of furrow irrigation.   Studies on the soils of these orchards showed that the soluble salts which had accumulated in the soil from applying saline water were gradually leached out by the combined effects of rain and nonsaline irrigation water.   Conditions favorable for the growth of citrus trees were restored.   Under California desert to semidesert conditions, it is not usually safe to irrigate citrus trees regularly with water containing more than 150 to 200 ppm of chlorine.
      Kelley and Thomas (1920) found that in certain localities in California the regular use of irrigation water containing as little as 200 ppm of chlorine for a period of from fifteen to twenty years resulted in a slight injury to lemon trees.   In these orchards, they found that the chloride content of the soil had been materially increased.   The question naturally arises: what is the upper limit of chloride tolerance for citrus trees?   At present, no categorical answer can be given to this question.   In the first place, the effect of chloride and other soluble materials will depend upon the concentration of constituents in the soil solution.   It has been established that the application of a given amount of soluble salt may produce materially different effects on the concentrations of the soil solutions of different types of soil, depending on water-holding capacity, porosity of the soil, and the rate of evapotranspiration.   The effect of a given soluble constituent also is materially influenced by the amounts of other constituents present in the soil, especially the content of organic matter and nitrogen.   In addition, climatic conditions influence the results, and, finally, rootstock and variety are important.
      In several California localities the presence in the soil of considerably less than 500 ppm of chlorine, expressed on the basis of dry soil, has been associated with definite injury to citrus trees.   The soils in question were sandy loams.
      Effects of Saline-Alkali Conditions.—Alkali (high sodium) salts affect different varieties and species of Citrus somewhat differently.   Lemon trees show the effects by pronounced yellowing and burning of the margins and tips of leaves, often followed by unusually heavy leaf drop late in the dormant season; leaves also tend to curl somewhat.   Subsequent new growth may be vigorous and apparently normal for several months, but later the leaves of new shoots also turn yellow in irregularly-shaped areas along their margins and drop off excessively.   Where soil contains a high chloride concentration, complete defoliation may occur, and in extreme cases death of the trees may result.   The tendency toward mottle-leaf (zinc deficiency) and iron chlorosis is accentuated by moderately high concentrations of soluble salts.   Both the quality and quantity of the fruit produced may be impaired.   The size of the fruit will tend to be subnormal, and lemon fruits thus affected tend to become tree ripe before normal picking time.
      Orange and grapefruit trees also show the effects of alkali salts in different ways.   On certain soils, mottle-leaf is one of the first symptoms of injury.   In the hot interior valleys of California, a moderately high concentration of soluble salts in the soil may cause older leaves to develop a brownish hue and curl to some degree.   If soil contains excessive amounts of chloride, the margins and tips of leaves turn brown, and leaves may drop suddenly, especially if the soil is allowed to become unusually dry.   Orange trees, thus defoliated, may later develop a profuse growth of new shoots, but new leaves are likely to be subnormal in size and pale in color.
      Citrus trees injured by alkali salts are unusually susceptible to adverse climatic conditions.   Hot winds cause the young leaves of such trees to dry up and fall off; frosts likewise may be unusually severe on affected trees.   The relations between alkali injury and the severity of adverse climatic conditions are complex.   Among the various factors, it is probable that soluble salts affect the moisture relationships of growing tissues.   It has long been thought that increasing the concentration of the nutrient solution tends to cause a decrease in absorption of water by plants.   Another possible reason for the extreme susceptibility of alkali-injured trees to adverse climatic conditions is their generally weakened condition.
      Figures 3-73 and 3-74 show leaf-tip burning and yellowing caused by an excessive accumulation of chloride.   Such leaves usually show from 0.75 to over 1.5 per cent of chlorine in the dry matter.   The earliest symptom is a yellowing of the tip, followed by progressive yellowing and necrosis starting at the tip and proceeding downward.   Figure 3-75 shows typical necrotic burned areas on leaves caused by excessive sodium uptake.
      In carefully controlled sand cultures in the greenhouse, using sweet orange seedlings, maintained concentrations of 700 ppm of chloride (supplied by increasing calcium, magnesium, and potassium levels) produced leaf symptoms of chloride excess only where nitrogen (as nitrate) was deficient (Chapman and Liebig, 1940).   The chloride content of injured leaves in the nitrogen-deficient cultures attained values of 2.78 and 3.90 per cent in the dry matter, whereas in ample nitrogen cultures (no visual symptoms of excess chloride) the value was 0.53 per cent of chloride.
      Eaton (1942) grew rooted Eureka lemon cuttings for seven months in outdoor sand cultures receiving, respectively, 0.6, 50, and 150 meq/l of chloride (50 per cent as sodium and the remainder as calcium and magnesium).   Plants grown in the cultures containing 150 meq/l chloride (5,250 ppm) became yellow; the leaves fell, and the plants died.   At the 50-meq/l-level (1,750 ppm), growth was only 28 per cent of that in the low-chloride cultures.   No leaf analyses were made.   Seven other plants were grown under identical conditions in the sand beds.   Lemon cuttings were the most seriously affected by the chloride; the order of relative growth, compared with the control as 100, was lemon, 28; navy beans, 39; dwarf milo, 54; Chilean alfalfa, 73; Acala cotton, 75; Stone tomato, 78; and sugar beet (roots), 98.
      In the same publication, Eaton (1942) gives an analysis of the soil solution from two 13-year-old grapefruit orchards irrigated with Colorado River water, but growing on two different soils—one a sand, the other a silty clay.   The latter orchard (fig. 3-76), while of good appearance and well foliated, produced much less fruit and the trees were smaller than those of the orchard growing on sand (fig. 3-77).   Soil solution analyses from these two orchards are presented in table 3-40.   Though all of the difference in growth cannot be ascribed to the difference in salt, it seems certain that the high conductivity, sodium, and chloride values found in all horizons of the clay soil were important as regards the difference in tree size and yields.
      Hayward and Blair (1942) grew orange seedlings in nutrient solutions in the greenhouse for two months in Hoagland's solution with and without additions of 50 and 100 meq/l of mixed chlorides.   Growth was reduced by 73 and 47 per cent of the control at the 50 and 100 meq/l of chloride levels.   Nearly half of the plants at high-chloride levels died, and there was leaf-tip burn in the plants within two to three weeks.
      Pearson, Goss, and Hayward (1957) grew Ruby Red grapefruit on Cleopatra mandarin rootstock in outdoor lysimeters filled with a loam soil.   Trees were irrigated with (1) tap water of low salt content (300 ppm), (2) tap water plus 2,000 ppm of calcium chloride and sodium chloride, and (3) tap water plus 4,000 ppm of calcium chloride and sodium chloride.   Ammonium nitrate at a constant rate of 0.4 gm/l was added to all irrigation waters.   The trees were grown for 19 months.   The average electrical conductivity (EC) of the saturation extract of soil to a depth of 0 to 48 inches for the three waters was 1.5, 6.2, and 9 mmho/cm, where free drainage of the lysimeters had been provided.   Growth (fresh weight) resulting from the salt treatment, compared to controls at 100, was 46 and 38 for the two salt levels.   The chloride content of four- to six-month-old leaves for the three treatments was, respectively, 0.07, 1.28, and 1.35 per cent in the dry matter.
      The salt treatments resulted in a bronzing and burning of leaves.   Also involved in the experiment were water-table variables, and the chief effect of high water tables was to increase salt content of the soils and severity of tree injury.   A conductivity value of approximately 4 mmho/cm in the saturation extract of the soils was associated with a 50 per cent decrease in tree trunk area.
      Because of a reportedly high-salt tolerance of Siamese pummelos, Monselise (1961) grew seedlings of this species in plastic buckets containing a clay soil to which enough sodium chloride had been added to produce a salinity in soil extracts of approximately EC 3.1, 5.7, and 13.5 mmho/cm.   The control soil showed values of 2.55 mmho/cm.   The plants were grown for about ten months.   Growth was virtually unaffected, as compared to controls at the 3.1 mmho/cm salt level; at the highest level (13.5 mmho/cm), the plants died in a few months.   At the 5.7 mmho/cm level, total dry weight at the end of the experiment was 28.3 per cent of the control.   These results are quite consistent with those reported by Pearson et al. (1957).   Tissue analyses of leaves and roots were made but not reported.   Monselise states that sodium levels in two-month-old leaves were not in excess of what is considered normal, but that root levels were above 0.30 per cent of sodium in the dry matter.
      Chloride Accumulation and Movement.—There is much evidence that chloride in citrus (like lithium and boron) tends to accumulate primarily in leaves and this accumulation is a function of nutrient concentrations, time, and age.
      Groenewegen, Bouma, and Gates (1959) exposed Washington navel orange cuttings, grown first for 74 days in a base nutrient solution, to 200 meq/l of chloride for a three-day period, having approached this high level by increasing the salt gradually (25 meq/l of chlorine per day) for some eight days.   After the three-day exposure to high chloride, salt was flushed out and plants were then harvested at progressive periods of 0, 5, 25, and 82 days following chloride removal.   The investigators found very little movement of accumulated chloride out of the old leaves to younger leaves.   The chief movement was from accumulated chloride in roots and stems to the younger growth.   This is probably the reason there is premature abscission of salt-affected older leaves of citrus in the field, even though new growth continues to emerge.   Of the total chloride found in plants at harvest (82 days after salt removal), about two-thirds was in the leaf laminae, and most of the rest was in roots and stems.   At high-chloride levels, there was leaf burn and marked defoliation.
      Cassidy (1946), reporting on variable salt (presumably sodium chloride) treatments to citrus trees, noted that chloride is more readily absorbed into leaves than sodium, and that the earliest symptom is a tip yellowing followed by tip burn.   A dose of two hundredweight (cwt) salt per tree caused the tree to drop all leaves.   The author has noted similar drastic and rapid injury and leaf drop from the addition of very high sodium chloride applications to citrus trees in the field.
      Brusca and Haas (1958) grew Lisbon lemon on sour orange rootstock in large outdoor soil cultures which received complete nutrient solution additions periodically, to which were added variable amounts of chloride derived equally from calcium, magnesium, and potassium chloride.   Analysis of leaves and various other parts for chlorine were made, and part of the data is presented.

                                                                                          Per Cent Chlorine in Dry Matter     
                  Nutrient                              Mature           Bud            Bark of               Fine
                  Solution                              Leaves          Union         Main Root          Rootlets         Blossoms        Peel          Pulp  
Control                                                   0.016           0.013             0.020               0.221              0.007          0.015        0.015
Control plus 560 ppm of chlorine          0.188           0.252             0.142               0.387              0.038          0.243        0.026
Control plus 1,400 ppm of chlorine       1.613           0.413             0.267               0.443              0.169          0.466        0.188

      Very little data for citrus exist on the effects of other nutritional factors on chloride absorption, but from work with other plants there is some evidence that more chloride is absorbed when calcium is the dominant ion than when it is sodium (Elgabaly and Wiklander, 1961; Elgabaly, 1962; Brown, Wadleigh, and Hayward, 1953).   It is not known whether this applies to citrus.
      Sodium in Relation to Performance.—There is little evidence that sodium at any concentration or under any condition is beneficial to citrus.   With some other plants or groups of plants, sodium is either essential or beneficial under varying circumstances; for example, when potassium is low.   In a number of nutritional experiments, the author and his colleagues tried to find out whether sodium in either low or fairly high concentrations would partially replace or substitute for potassium.   The results were consistently negative.   In controlled sand cultures with orange and grapefruit seedlings and lemon cuttings, Chapman and Brown (1943) found that high-sodium levels somewhat delayed the onset of potassium deficiency but did not prevent it.   Ultimately these plants (potassium-deficient) developed the same deficiency symptoms as those grown without sodium.   The delaying effect of sodium was thought to be due to its effect on calcium; where this latter nutrient was high (under potassium deficiency), the plants showed acute potassium deficiency earlier than where calcium and magnesium were lower.   It appears, then, that high sodium merely offsets somewhat the effect of calcium and magnesium in hastening the onset of potassium deficiency.
      In studies by Pearson (1951), to be detailed later in this section, no benefits from sodium under low-potassium conditions were noted.   Page and Martin (1964) explored this problem and, in soils of low or deficient potash levels, found no evidence of a beneficial effect from increasing sodium additions.
      Sodium can be detrimental to citrus (and all plants) through: (1) its contribution to total salinity of soil (osmotic effect); (2) differential absorption and accumulation in roots and leaves (specific ion effect); (3) nutritional imbalancing effect (i.e., effect on absorption of other nutrients); (4) detrimental effect on soil structure and pH; and (5) leaf absorption from foliar sprays, etc.   In many cases, the first four effects are combined so that it is difficult to evaluate or separate them.
      This subsection will cite some specific cases in which sodium (due to one influence or another) has exerted a detrimental effect on citrus, describe leaf-injury symptoms due to excessive sodium absorption, and mention leaf values associated with excessive accumulation.   In a later section on rootstocks in relation to salt injury, reference to both sodium and chloride will be found.   Foliar absorption and injury from irrigation water and briny sprays will be dealt with in another section.
      In a lengthy fertilizer experiment with Washington navel orange trees on sweet orange root, Jones et al. (1952) reported a 25 per cent reduction in yield between 1940 and 1949 from the use of sodium nitrate (at 3 pounds of nitrogen per tree per year), compared with equivalent amounts of calcium nitrate.   Spring-flush leaves, picked from nonfruiting terminals in December, 1959, showed 0.106 versus 0.204 per cent of sodium in the dry matter.   Feeder roots showed 0.232 versus 0.666; root bark, 0.359 versus 0.555; and root xylem, 0.148 versus 0.324 per cent of sodium.   Earlier studies on these plots by Aldrich et al. (1945) demonstrated that the inferior tree performance was caused primarily by poor water penetration (and moisture deficits) resulting from an accumulation of sodium in the exchange complex.   The exchangeable calcium-to-sodium ratio in the calcium nitrate plots was 93:1, as compared with 8:1 in the sodium nitrate plots.   Base exchange capacity of this soil varied between 5.23 and 8.35 meq/100 gm of soil.   Exchangeable sodium in the sodium nitrate plots had built up to about 10 per cent of the exchange capacity.   Salts had also accumulated at the 2-, 3-, and 4-foot levels in this soil as a result of lack of water penetration.   Poor growth and yield of citrus in this orchard probably was not due as much to excessive sodium absorption by leaves and roots, as lack of adequate moisture for the trees.
      Huberty and Pearson (1949) and Pearson and Huberty (1959), working with Washington navel orange trees irrigated for fifteen years with softened and natural Colorado River water and a local well water in which sodium was respectively 78, 40, and 28 per cent, found that yields were reduced by 15 per cent when irrigated with the softened (calcium replaced by sodium) Colorado River water.   Valencia orange yields were only slightly reduced.   These differences are almost too small to be considered significant, yet there was a definite trend toward lower yields when the softened water was used.   The soil was a Ramona loam to clay loam, and the average annual rainfall was 14 inches.   Soluble salts, particularly sodium and chloride, increased in the various soil horizons of plots treated with the softened Colorado River water over those irrigated with natural Colorado River water or well water.   Exchangeable sodium as per cent total bases increased from values of 2.0 to 2.1 in plots irrigated with well water to values of 6.2 to 9.4 in plots irrigated with softened water.   Sodium and chloride increased somewhat in leaves, feeder roots, root bark, and xylem of both orange varieties, but the values hardly seem high enough to account for decreased yields.   It rather appears in this experiment that the increased total salt in the soil may have been a primary influence in decreasing navel orange yields.   The navel orange leaves showed 0.16 versus 0.054 per cent of sodium in the dry matter.   Saturation extract analyses made in the spring of 1949 (salt would be at minimal values in the spring after winter rains) showed average conductivity values of 1.9 and 3.04 mmho/cm for the upper 4 feet of soil.
      Martin, Harding, and Murphy (1953) grew sweet and sour orange seedlings in potted soils of various exchangeable sodium percentages.   The experimental soils were prepared as follows.   Large batches of calcium-, magnesium-, potassium-, sodium-, and hydrogen-saturated soils were first prepared.   Then samples of these were mixed to attain the proportions of exchangeable bases desired.   The sodium series was prepared to contain sodium varying from <1.0 to 28 per cent exchangeable sodium percentage (ESP).   Yields of tops in grams were 60, 45, 36, 22, and 7 in a second crop of sour orange seedlings on a Yolo loam, which had <1, 4, 7, 14, and 28 per cent ESP, respectively.   The sodium content of the leaves of these seedlings were respectively 0.05, 0.21, 0.26, 0.40, and 0.78 per cent in the dry matter.
      In further, similar experiments with Eureka lemon on several rootstocks, Jones et al. (1957) found on all rootstocks marked growth reductions at 13 per cent and drastic reductions at 28 per cent ESP.   There were differences in the degree of growth reduction on different rootstocks.   Greatest sodium uptake by leaves and leaf burn occurred in plants budded to Sampson tangelo and grapefruit.   Best growth and lowest sodium uptake was with Cleopatra mandarin root.   Best growth on all rootstocks and lowest sodium percentages in leaves were on the soils with an ESP of 1 or less.   Increasing sodium decreased calcium and magnesium contents of leaves and increased potassium somewhat.   All parts of the plant (leaves, stems, tap roots, and fibrous roots) reflected the varying sodium levels, but in general the spread between low and high sodium was greatest in the leaves.
      In similar studies with trifoliate orange seedlings on soils of varying exchangeable sodium (and potassium) percentages, Martin et al. (1959) got growth reduction at 7 per cent ESP in a clay loam soil of exchange capacity of 25 meq/100 gm, and marked reduction at 15 per cent ESP.   In a sandy loam of lower exchange capacity (10 meq/100 gm), growth reduction was much less at 7 and 15 ESP.   Increases in sodium decreased calcium and increased potassium percentages in the leaves.   These experiments indicate that trifoliate citrus seedlings may be more sensitive to sodium than some other citrus species.
      In still further experiments of a similar nature with soils compounded to contain varying exchangeable sodium percentages under three soil conditions—slightly acid, base-saturated, and containing 1 per cent of calcium carbonate—Martin, Ervin, and Shepherd (1961) got a substantial yield reduction on one soil, at 6 per cent ESP, under all three of the aforementioned conditions.   Where 0.1 per cent of VAMA was added to the soils, yield reductions were not as great.   In these trials, feeder roots showed somewhat greater sodium differences in response to the lower ranges of ESP levels than the leaves.   In general, the magnitude of growth reduction from increasing exchangeable sodium percentages was greater in base-saturated and limed soil than in slightly acid soil.
      Haas (1952a, 1952b) grew Lisbon lemons on Brazilian sour root in outdoor soil cultures supplied with nutrient solutions varying in sodium content from 0 to 348 ppm and in calcium from 318 to 0.   The highest sodium solution had the lowest calcium and vice versa.   These solutions were made from sodium and calcium nitrate.   Potassium sulfate and phosphate were also added to the nutrient solution.   Growth was generally less at increasing sodium and decreasing calcium levels.   Sodium in the leaves of plants depressed in growth showed values up to 0.17 per cent of sodium in the dry matter, and in these same high-sodium cultures rootlets showed between 0.56 and 0.75 per cent of sodium in the dry matter.
      In further experiments, Haas and Brusca (1954b) added increasing amounts of sodium nitrate (in a modified Hoagland nutrient solution) to soil cultures of various citrus seedlings (Troyer citrange, grapefruit, Pomeroy trifoliate orange, Cleopatra mandarin, Spanish sour orange, rough lemon, tangelo, and sweet orange).   The nutrient solution (of variable sodium nitrate content) was added periodically, and, in between, distilled water was added.   No sodium injury symptoms or growth depression was noted.   Leaf analyses for sodium, going from low to high sodium, showed values in the different seedlings from <0.10 to 0.35 per cent in the dry matter.   Highest sodium values in high-sodium cultures were found in Sampson tangelo seedlings and lowest in the sour orange.   Root values for sodium showed a greater range in sodium going from low to high than did leaves.   Lack of sodium injury in the cultures was probably due to the presence of ample calcium and magnesium and the diluting effect of distilled water additions to the soils.
      Zusman (1956) grew sour orange and sweet lime seedlings in fertilized pots of soil to which sodium carbonate and sulfuric acid were added to produce pH values ranging from 4.40 to 9.02 in the soils growing sweet lime, and from 4.16 to 8.88 where sour orange seedlings were grown.   All soils contained about 260 ppm of sodium.   After some fifty days of growth, the sweet lime seedlings showed leaf burn, which from the description corresponded to sodium injury.   Leaf analysis values ranged from 0.20 to 0.28 per cent of sodium in the dry matter.   On the other hand, the sour orange seedlings growing under comparable conditions showed no leaf necrosis, and sodium levels in the leaves varied from 0.07 to 0.13 per cent in the dry matter.
      Pearson (1951), working over a period of years in the author's laboratory and greenhouse, made an extensive investigation of sodium in relation to the behavior of lemon cuttings in carefully controlled sand and sand-bentonite mixtures.   He determined effects on growth, appearance, and composition of (1) variable sodium concentrations under differing potassium, calcium, and magnesium levels, (2) variable anion carriers of sodium (chloride, sulfate, nitrate, and bicarbonate) with sodium maintained at 30 meq/l, and (3) sand-bentonite mixtures in equilibrium with nutrient solutions of variable sodium concentrations.   In addition, he studied the usefulness of various plant parts as an indicator of sodium status.
      Since these data have never been published, it appears worthwhile to record some of the principal results here.
      All experiments were conducted in 3-gallon crocks of sand, flushed at frequent intervals with well-controlled nutrient solutions from large reservoirs.   Iron was supplied by incorporating finely ground magnetite in the sand.
      In the first experiment, increasing amounts of sodium chloride and sodium sulfate were added to a base nutrient solution to give sodium levels ranging from 0 to 20 meq/l.   After four months of growth, no growth depression from the 20-meq/l sodium treatment was noted, and three higher sodium levels were added to the experiment, utilizing the cultures which during the initial period had received only 1, 3, and 5 meq/l sodium.   At this time (December 16, 1944) two thirds of the plants were pruned back to the original cutting leaves and then grown for another year (with one additional severe pruning in April, 1945).   The severe pruning after four months and again after one year caused plants which had hitherto not been affected by sodium at the 7-, 10-, 15-, and 20-meq/l levels to decrease in growth rate.   The older leaves in the 20-meq/l treatment showed typical sodium burn injury and abscised.   The weights of leaves from plants pruned off after the first four months and of the tops in the subsequent year of growth are shown in table 3-41, together with leaf analysis data.
      The reduced growth and elevated sodium in leaves of pruned plants is no doubt a result of sodium being taken up by, or moved out of, roots into a much-reduced top, resulting in growth retardation and sodium buildup in the leaves.   These results indicate that trees which are growing well or reasonably well in salt-affected soils may be badly injured by severe pruning, especially if adverse weather conditions (heat and wind) should coincide or shortly follow.   Green, 4 1/2-month-old leaves with no sodium burn showed increasing sodium levels as growth decreased.
      In another experiment under similar conditions, comparisons were made between equivalent amounts (30 meq/l) of sodium salts (sodium nitrate, sodium chloride, sodium sulfate, and sodium bicarbonate).   The nature of the solutions used, growth, appearance, yield, and composition of leaves are shown in table 3-42.   Growth was severely depressed by sodium bicarbonate, the plants developed iron chlorosis, and the leaves showed much sodium burn.   The calcium content of leaves was greatly reduced.   While all of the other salts depressed growth and there was burning and some shedding of older leaves, the plants were only moderately affected.   Excluding the sodium bicarbonate culture, least growth depression occurred with sodium nitrate and greatest with sodium chloride.   In all cases, the necrosis on older leaves was typical of sodium injury, and leaf values on 5 1/2-month-old leaves were in the expected range of necrosis and growth depression.   As in the first experiment, there was decreased calcium and magnesium in leaves of the growth-depressed plants, and there was increased potassium absorption.
      In still another experiment carried out over thirteen months under comparable conditions, the effects of high levels of sodium at different calcium, magnesium, and total salt levels were studied.   Data from this experiment are shown in table 3-43.   Where calcium and magnesium were high, 30 meq/l sodium only slightly depressed total growth, and only a few older leaves showed sodium burn.   Sodium in four-month-old leaves increased only slightly.   The percentages of sodium as of total bases in the nutrient solution were 54 per cent (see data for tiles 2 and 4 in table 3-43).
      In tiles 8, 9, and 10 (table 3-43), sodium varied from 2.2 to 30 meq/l and as per cent of total bases, from 52 to 94.   Under the low calcium and magnesium conditions of these cultures, 15 and 30 meq/l of sodium markedly depressed growth, and there were corresponding increases in the sodium content of leaves and decreases in calcium and magnesium.   Potassium, nitrogen, and sulfur also increased.   Even though the exact physiological explanation of decreased growth is not possible, it is clear that under low calcium and magnesium conditions, increased sodium is much more disastrous in its effect than if the divalent cations are high.
      In still further experiments, bentonite of high exchange capacity (113 meq/100 gm) was introduced into the 3-gallon sand cultures, and lemon cuttings were grown in plain sand and sand-bentonite mixtures at low- and high-sodium levels.   It was found that the decreased permeability (and poor aeration) of the cultures to which 3 par cent bentonite was added (where the sodium saturation of the bentonite was 21 per cent) was more detrimental than the extra exchangeable sodium due to the presence of clay.   Neither leaves nor feeder roots showed appreciably more sodium in the high-sodium, sand-bentonite mixture than in the high-sodium, sand-alone medium.   These results are in harmony with those of other investigators working with other plants, namely, that the structural deterioration, pH increase, and related effects incidental to increasing sodium saturation are more adverse to plant growth than the direct effect on plant growth of the absorbed sodium per se.
      The many chemical analyses made and the nature of the experimental setup in these experiments afforded an excellent opportunity to secure information on the plant part best suited to indicate sodium status.   Analyses of leaves, stem, bark, stem wood, root back, root wood, and feeder roots of plants grown in control versus those receiving 30 meq/l sodium sulfate are shown in table 3-44.   The data clearly indicate that mature leaves and feeder roots show the widest range in sodium content.
      A summary showing the range of sodium in three- to six-month-old lemon leaves, as related to growth and presence or absence of sodium burn, is given in table 3-45.   From these and data of other experiments reported, it appears that sodium values of less than 0.10 per cent in the dry matter can be considered as normal in mature leaves.   Values in excess of this may indicate sodium troubles.   Leaves with typical sodium burn will generally show values in excess of 0.70 per cent.
      Estimation of the Exchangeable Sodium Percentage of Soils.—Since both exchangeable sodium and sodium in the soil solution influence plant growth and, in fact, are in equilibrium and because of difficulties in the exact determination of exchangeable sodium and ESP in some soils (particularly saline soils), Bower and Hatcher (1962) have suggested that the sodium absorption ratio (SAR) of an equilibrium extract may be more meaningful as a measure of the sodium status of a soil in its relation to plants than exchange sodium percentage (ESP).   Sodium absorption ratio in soil extracts and irrigation waters is given by the, following equation:
where the concentrations are expressed in milliequivalents per liter.   However, more data are needed relating these values to citrus performance than are currently available.
      Effect of Rootstock on Salt Absorption and Tolerance.—There is much evidence that the effects of salt on citrus are significantly affected by rootstock, variety, and other genetic factors.   Cooper and his associates, working in Texas, have produced considerable information on this subject, and because of its practical importance a fairly complete summary of his investigations is recorded here.
      Cooper (1948) grew Redblush grapefruit on sour orange and Cleopatra mandarin rootstocks in 15- by 15-foot basins irrigated with three waters.   The control treatment was Rio Grande water, which ranged in electrical conductivity from 0.6 to 1.4 mmho/cm or from approximately 380 to 840 ppm in total dissolved solids.   The other two waters were treated with equal amounts of calcium chloride and sodium chloride to raise the salt content to about 2,500 to 5,000 ppm, respectively.   Conductivity measurements made over the experimental period showed the waters varied from 3.0 to 4.8 mmho/cm for the medium salt content water and from 7 to 9 mmho/cm for the high salt content water.   Repeated rains caused fluctuation in the salinity of the soils, but conductivity determinations on repeated saturation extracts made of samples of the first foot of soil gave values for the three plots as follows: river water, 0.8 to 2.0 mmho/cm; medium salt, 0.7 to 4.3; high salt, 0.6 to 7.0.   Soil pH in all plots made on a paste ranged from 7.2 to 7.9.   There was no pH difference between the various salt plots.   The soil was a fine sandy loam.
      It soon became evident that the grapefruit was much more severely affected by salt when on sour root than on Cleopatra mandarin root.   Growth, as measured by increase in trunk circumference, and analyses of nine-month-old leaves are recorded in table 3-46.   These data clearly show a greater chloride intake by the sour orange root and the appearance of more salt injury symptoms on the foliage.
      In the same report, Cooper (1948) noted that Valencia orange on Cleopatra mandarin root was more tolerant to salt than Redblush grapefruit.   He further noted that Cleopatra mandarin seedlings were more sensitive to chlorosis in highly calcareous soils than sour orange seedlings.
      In a somewhat similar experiment with one-year-old Shary Red grapefruit on twenty different rootstocks, grown for 17 weeks in basins irrigated with water containing enough of a 50-50 mixture of calcium chloride and sodium chloride to produce a water of 4,000 ppm total salt, Cooper, Gorton, and Edwards (1951) reported the chloride content of a composite of spring-flush leaves picked on July 1 and September 22, 1950, and also noted the extent of injury.   The data reported are shown in table 3-47.
      It is evident that least injury occurred on Severinia buxifolia, Cleopatra mandarin, and Rangpur lime rootstocks, and greatest injury on stocks of trifoliate orange, citron, and Florida sweet orange.   Sour orange was intermediate, as were several of the tangelos and sweet oranges.
      Further data from similar experiments reported by Cooper (1962) are shown in table 3-48.   The results confirm other data in showing the salt tolerance of Rangpur lime and Cleopatra mandarin compared with the salt susceptibility of rough lemon, tangelos, sour orange, pummelo, citrange, and trifoliate orange.   Also good relations between leaf burn and chloride and sodium content of leaves are shown by these data.
      Continuing experiments with Shary Red grapefruit and Valencia oranges on sour orange and Cleopatra mandarin rootstocks were reported by Cooper, Gorton, and Olson (1952), and again salt injury and chloride uptake was much less with Cleopatra rootstock than with sour orange.   This experiment also involved boron variables.   This element (unlike chloride) was absorbed in significantly greater amounts by the trees on Cleopatra than on sour orange.   The relative absorption of salt by leaves, twigs, bark, trunk wood, and roots in the high- and low-salt treatments was reported.   The leaves were the best indicator of high chloride in the soil and also showed somewhat elevated sodium levels, but the taproot and lateral and fibrous roots were a little better indicators of sodium levels in the high-salt plots.   Leaves were the best indicators of varying boron status.
      In studies with young nucellar-line Redblush grapefruit trees, Cooper and Peynado (1959b) reported on the growth from April 24 to November 19, 1958, and on leaf composition and appearance with respect to chlorine and boron when trees were irrigated with a saline well water containing 3,300 ppm total solids and 5.5 ppm boron.   Thirteen rootstocks were used.   The sodium percentage of the water was 85.   The planting was irrigated seven times during the test period.   Data from the experiment are shown in table 3-49.
      These data show that much less chloride was accumulated from the mandarin stocks and Rangpur lime than from the citrange, citrumelo No. 4475, Citrus macrophylla, and Citrus moi stocks.   However, the order of accumulation for boron was different from that of chloride.
      In this same paper, Cooper and Peynado (1959b) reported on the growth, chloride accumulation, and severity of chloride-toxicity symptoms of young-line and old-line Valencia orange trees on sour orange rootstock, irrigated with waters of increasing sodium chloride content.   Each young tree was planted in rows 5 feet apart and irrigated in 36-inch-diameter circular plots, separated from surrounding soil by galvanized iron barriers inserted in the soil to a depth of 30 inches.   The single-tree plots were irrigated four times with 4 acre-inch equivalents of the various waters from June 24 to September 3, 1958.   Data on this experiment are reported in table 3-50 and show marked differences in chloride accumulation by young- and old-line scions.
      In further research with Webb Redblush grapefruit trees on sour orange and Cleopatra mandarin rootstocks, Cooper et al. (1958) grew trees for four years in small basins differentially irrigated with a low-salt-content river water and a highly saline, high-sodium well water.   (The soil was a fine sandy loam, well drained, with no water-table problem.)   The soil was further modified by additions of ammonium nitrate, calcium sulfate, and calcium nitrate.   Data on this experiment are condensed in table 3-51.
      The composition of the irrigation waters used and the effects of two years' application on the salt content of the first foot of soil show: (1) electrical conductivity values well above those thought safe for citrus, and (2) major increases in the sodium and chloride of the saturation extract.
      Growth of trees on both rootstocks was depressed when the well water was used, but in most cases more so on sour orange than on Cleopatra stock.   Also, visual symptoms of both boron and salt injury, in general, were worse where trees were on sour stock.   The addition of calcium nitrate to the high-boron-content well water markedly reduced the amount of boron injury symptoms and to some extent the boron uptake, as measured by leaf analysis.
      In spite of the rather high sodium in the first foot of soil (where most of the roots of these trees were growing), this element was not excessively absorbed.   No doubt this was due in part to the presence of considerable calcium, since calcium exerts a powerful blocking effect on sodium uptake.   As in previous experiments, chloride uptake was appreciably less on Cleopatra than on sour orange stock.
      The growth depression in these trees is probably due to the influence of (1) high-salt content of the soil solution (osmotic effects), and (2) excessive chloride and boron absorption.
      These data, aside from the rootstock effect, are of especial interest in showing that in spite of the use of a high-salt and boron-content water (approximately 3,300 ppm) reasonably good growth and not much salt or boron injury was evident during the four years of this experiment when Cleopatra mandarin root was used along with generous applications of calcium nitrate.   The latter definitely lowered boron uptake and probably had a beneficial balancing effect on sodium, though this is not reflected in the leaf analyses reported.   Annual rainfall during the four years of this experiment ranged from 14 to 22 inches.   This undoubtedly reduced salt accumulation considerably.   Total irrigation water applied was approximately 144 acre-inch equivalents for the four-year period.   Rain plus irrigation for the four years was as follows: 1953, 49 inches; 1954, 54 inches; 1955, 58 inches; and 1956, 54 inches.   With the type soil and prevailing climate, considerable root-zone leaching no doubt occurred and thus reduced salt buildup in the root zone.
      A paper published by Cooper et al. (1956) summarizes observations for a number of years on disease, salt, boron, lime, and cold tolerance of grapefruit on numerous rootstocks.   These observations are reproduced in table 3-52 [in three parts: table 3-52a, table 3-52b, table 3-52c (includes legend)].   Although they refer to conditions in Texas, the data have much wider application.   Very little other field data of the scope and character of Cooper's work have come to the author's attention.
      In Australia, Smith (1963) compared the chloride content of Washington navel oranges grown on Citronelle (rough lemon) and trifoliate roots.   Much more chlorine accumulated in the leaves of plants grown on trifoliate roots.
      Salt Injury From Sprays.—Foliar absorption of salt from overhead or underhead irrigation and ocean sprays and mists has been noted.   Citrus leaf injury from the use of underhead sprinkler irrigation systems, in which only the skirts of the trees were wetted, was noted in several areas of California by Harding, Miller, and Fireman (1958).   They found an appreciable accumulation of sodium and chlorine in the sprinkled lower leaves of grapefruit, orange, and mandarin orchards where waters had been used with total salt content ranging from 491 to 931 ppm of total soluble solids.   Data from this study are shown in table 3-53.
      In further studies of foliar absorption by citrus, Eaton and Harding (1959) found that Valencia orange trees, sprinkled intermittently with a water containing 969 ppm total salt (163 sodium, 135 chlorine, and 384 sulfate) for six 9-hour periods, absorbed substantial amounts of sodium and chlorine.   Absorption was much greater when sprinkling was done in the daytime than at night.   In addition to increased sodium and chloride absorption, there was a significant increase in sulfur in the leaves.   Absorption was much greater when the sprinkling was intermittent rather than continuous, the same total sprinkling time being used in both cases.
      Ehlig and Bernstein (1959) also conducted sprinkling trials with avocado, navel oranges, almond, apricot, plum, alfalfa, beans, cabbage, clover, lettuce, onion, spinach, tall fescue, and tomato, using various salt solutions of different concentrations.   All plants were grown in sand cultures.   The vegetable and fields crops showed little injury, even with waters containing up to 96 meq/l of sodium chloride.   Of the trees, all but avocados absorbed enough salt to be injurious to the leaves.   Orange leaves lost their gloss, and some showed tip burn when sprinkled intermittently with waters containing from 5 to 10 meq/l of calcium chloride, sodium chloride, or sodium sulfate for 42 hours of total sprinkling time.   Injury was worse from daytime than nighttime sprinkling.   Partial to complete defoliation occurred with almond, plum, and orange sprinkled with high sodium chloride and calcium chloride solutions (10 meq/l) intermittently for 59 hours in the daytime.   Injured orange leaves showed from 0.70 to 1.05 per cent of chlorine and 0.46 to 0.69 per cent of sodium.
      Miwa, Gomi, and Yamamoto (1957) found that defoliation of citrus trees from briny sprays could be reduced by a spray of 2,4-D at concentrations of 50 to 100 ppm.   Noro (1956) obtained similar results.
      Mungomery (1959), in Australia, has noted injury to citrus from overhead sprinklers where the application rate was so slow that leaves (in very hot weather) dried off between each rotation of the sprinkler.   Early morning, late evening, or night irrigation was recommended as a means of reducing salt injury.
      Smith (1963), working in Western Australia, conducted spraying experiments with Washington navel oranges somewhat similar to those reported by Eaton and Harding (1959).   Comparable results were obtained.
      By way of general summary on salinity as related to citrus performance, it is evident from present information that no exact limit can be placed on permissible salt levels in the soil.   For general guidance purposes, it would appear that when chloride in the root zone of any citrus soil exceeds 70 ppm (expressed on a dry soil basis) or exceeds 350 mgm/l (10 meq) in a saturation extract, the possibility exists for some chloride injury, though severe injury is not likely with values reaching several times these concentrations.
      With respect to sodium status, (1) exchangeable sodium percentage (ESP), (2) concentration of sodium in the soil solution, and (3) accompanying cations (especially calcium) and anions (especially bicarbonate and chloride) make it difficult to set limits.   However, the evidence suggests that when the ESP exceeds 6.0 in any part of the root zone, the possibility exists for decreased permeability, aeration, nutrient availability, and the chain of other consequences which follow.
      In the soil solution or saturation extract, soluble sodium in excess of 20 meq/l (particularly if calcium and magnesium make up less than 50 per cent of the soluble cations) may cause some growth depression, especially if the prevailing anion is chloride.   Serious trouble may be expected if sodium bicarbonate is found in amounts much under 20 meq/l.
      With respect to total soluble salt, the best estimate at present is to be had from a measure of the electrical conductivity of a saturation extract.   Values in excess of EC 2.0 mmho/cm should be regarded with suspicion.   As shown by Cooper's data, however, the temporary presence of higher salt levels may not be injurious, and rootstocks such as Cleopatra mandarin, Rangpur lime, and others may achieve reasonably good citrus performance under much higher salt conditions than the value now regarded as permissible.
      With respect to irrigation water quality, barring severe climatic extremes, such as sudden temperature peaks, desiccating winds, and frosts, and with the use of tolerant rootstock, adequate root-zone leaching, good care and fertilization, plus some leaching rainfall, it should be possible to secure reasonably good citrus performance in some cases with waters of 2,000 ppm of total solids (EC 3.0 mmho/cm, approximately) and fair performance with waters of 3,000 ppm.   It is generally considered that total salt levels of about 1,500 ppm and 200 to 250 ppm of chloride represent the upper limit for successful citrus production under most conditions.   Even with these, recurring climatic extremes, low leaching rainfall, and soils of poor permeability may rule out their use.
      Control Methods.—There is not space here to consider control and management in any detail, but the general principles of dealing with salt and alkali problems in a practical way are fairly well known.
      High or fluctuating water tables, especially when they are closer than 4 feet to the soil surface and when they are saline, must be dealt with by appropriate soil-drainage methods.   Open ditches, tile drains, or wells (coupled with control of the causes of the high water table, such as canal lining, intercepting drains, and more judicious use of water) are the traditional means of lowering and controlling high water tables.
      If a high water table is not a problem or is taken care of by drainage, then the use of sufficient irrigation water to promote periodic root-zone flushing, use of gypsum where too much exchangeable sodium has developed, and use of salt-tolerant varieties of rootstock plus heavier than usual rates of nitrogen fertilizer are the chief means of dealing with the problem of salt and alkali.
      Basin or flooding irrigation, use of tensiometers to achieve better control of soil moisture and prevent overirrigation, and periodic soil, leaf, and root analyses to check on salinity conditions and changes will all help in the management problem.


      No soil measurements are more meaningful, and yet more difficult to be highly specific about when it comes to relating them to citrus and plant performance generally, than pH and the reserve acidity or alkalinity of soils.   Repeated reference has been made in preceding sections to the effects of soil acidity, alkalinity, and pH on the availability, uptake, and toxicity of various elements.   There is little need to elaborate here, except to say that within the pH ranges generally encountered in citrus soils the indirect effects or basic soil characteristics (of which pH is an indicator) are important rather than direct effects of hydrogen or hydroxyl ions as such.
      Plants have sometimes been classified into acid-loving and nonacid-loving.   It is difficult to classify citrus in this regard, since high-performance orchards can be found on soils with pH (paste) values ranging from 4.8 to 7.5 (or on 1:1 suspensions from pH 4.5 to 8.5), and in reserve acidity and alkalinity, from low to high.   Citrus is grown on soils with even wider range of pH, but serious problems usually are manifest in soils more alkaline than pH 8.5 or more acid than pH 4.3 (pH on 1:1 soil:water suspensions).
      The only generalization possible is that less problems of nutrient deficiency or excess are encountered on soils in the slightly acid range—pH 6.0 to 6.5—than at either higher or lower pH values.   In a South African survey of pH in relation to citrus performance, Oberholzer (1944) noted that best growth and yields generally were found on soils of pH 6.0 to 8.5.   (He determined pH on a 1:1 suspension with a Beckman glass electrode.)
      As far as direct toxic effects of hydrogen and hydroxyl ions on citrus are concerned, these appear unimportant except at very high values.   A number of investigators have studied this question.   Guest and Chapman (1944) carried out experiments in sand and solution cultures.   They found that at pH 2.0 and 11.0 citrus seedlings were killed in a few days.   At pH 2.5 and 3.0 the plants did not die, but little growth occurred.   Good growth was obtained between pH 4.0 and 8.8 when special precautions were taken to provide ample but not excessive nutrients.   Quite a little healthy growth occurred at pH 9.7 at first, but gradually the plants at this pH became light green to yellow and showed both sodium burn and high-sodium levels in the leaves.   (Sodium hydroxide had been used to maintain the pH 9.7 level.)   Also, some manganese deficiency developed in plants in the pH 8.0, 8.8, and 9.7 range.   The fact that plants at pH 9.7 were at first quite healthy suggests that the hydroxyl ion concentration of this pH exerts little direct toxic effect on the plants.
      In many years of water-culture experiments with budded orange and lemon trees grown in outdoor cultures maintained at pH values ranging between pH 3.8 and 4.2 (the aim being to maintain a pH of 4.0), the writer has noted healthy root and top growth where sufficient (but not excessive) amounts of other nutrients were present.
      Rasmussen and Smith (1959), in water-culture investigations with Pineapple orange seedlings, established evidence of toxicity to roots at pH 4.0.   They reported that the roots were brown, stubby, and swollen, in comparison with smooth, white roots grown at pH 6.0.   The roots of our water-culture trees at pH 4.0 were generally white, well-branched, proliferated, and not swollen.   The writer has frequently noted brown, stubby, and swollen roots of citrus in acid-water cultures where there is a slight excess of copper (e.g., 0.1 ppm of copper).   Copper impurities at this concentration can come from bronze-hose bibs or bronze fittings used to dispense water, from distilled water (copper stills), or from chemicals.   This or some other toxic impurities or low-nutrient concentrations (direct hydrogen-ion toxicity may be greatly influenced by the amounts and proportions of other ions present, such as calcium, magnesium, and potassium) may account for the observations of Rasmussen and Smith (1959).   Girton (1927) noted some root injury and decreased elongation with sour orange seedlings in nutrient solutions of pH 4.0.   Best top growth was at pH 6.5, but most root hairs were produced at pH 5.0.   Again, the poor growth at pH 4.0 could have been due to copper contamination or low levels of calcium and other nutrients.
      In soils, on the other hand, serious problems begin to arise, e.g., excessive solubility of aluminum, manganese, copper, nickel, and other elements in soils of pH 5.0 or less, or deficiencies of phosphorus, calcium, magnesium, and molybdenum.   Conversely, as soils become more alkaline, there is decreased solubility of manganese, iron, zinc, boron, copper, and phosphorus.
      As stated, no precise limits under soil conditions for top citrus performance can be laid down except to say that, in general, one begins to look for and finds troubles at soil pH values of 4.8 and lower or 8.0 and higher.   This usually is due to deficiencies or excesses, imbalances, or soil structure and biological problems.
      Martin and Page (1962) grew sweet orange seedlings in treated soils from various citrus orchards.   The pH of large quantities of these soils was first adjusted downward by prolonged leaching with hydrochloric acid.   The soils became hydrogen-saturated and had pH values as low as 3.3.   Excess acid and reaction products were leached out with distilled water.   By appropriate additions of calcium carbonate, magnesium carbonate, and potassium bicarbonate in various ratios, soils of various pH and exchangeable base content (i.e., calcium, magnesium, and potassium) were produced.   These were then potted in 3-gallon jars, and sweet orange seedlings were grown for several months.   The data of one of the experiments, showing soil composition, dry weight of plants produced, and leaf and root composition, are reproduced in
table 3-54.
      Best plant growth was at pH 5.7 to 6.9, with only slightly less growth at pH 4.8.   At pH 4.2 and lower, growth was markedly reduced, primarily due to lack of calcium and perhaps aluminum toxicity.   It is probable that other potentially toxic metals, e.g., copper, zinc, manganese, and nickel, had been leached out by the prior acid leaching.
      Rasmussen and Smith (1957) grew Pineapple orange seedlings in the acid subsoils of three Florida sands.   Separate lots of soil were leached with 1M sulfuric acid and 1M calcium chloride to hydrogen saturate and to calcium saturate each soil, respectively.   Excess acid and salt were then leached out with rain water.   The hydrogen-saturated soil had a pH of near 4.0, and the calcium-saturated soil, near pH 7.0.   These soils were combined in proportions to give pH levels of 4.0, 4.3, 4.6, 5.0, and 6.0.   Seedling oranges were then planted.   The pots were periodically watered with a nutrient solution containing nitrate and ammonium, adjusted in pH to correspond to the various pH's of the potted soils.   After five months of growth, the highest dry weights were at pH 6.0, but in two of the soils almost as good growth was obtained at pH 4.3.   Growth was considerably poorer at pH 4.0.   It seems probable that the much poorer growth at this low pH was due to lack of sufficient calcium.
      In addition to the varying solubility of many inorganic elements in soils, soil pH modifies biological activities in important ways, and these plus the chemical changes, in turn, can influence soil structure and aeration.
      Martin, Harding, and Garber (1961) found that the citrus replant problem was somewhat less acute on acid than on neutral and alkaline soils.   The precise explanation is unknown, but there is considerable field evidence to support their findings.
      Aldrich et al. (1955) grew citrus seedlings in potted soils brought to various pH values by the addition of increasing amounts of sulfur.   Growth decreased with increasing acidity, but leaf and soil analyses showed that the cause was increased sulfate uptake by the plants (to the point of injury) and the increased soluble salt content of the soils.   This work is cited primarily to indicate that soil acidity, and in this instance soil acidification of a naturally neutral to alkaline soil (by means of sulfur additions), may produce byproducts in sufficient amounts to depress plant growth.
      Causes and Control of Soil Acidity and Alkalinity.—Soils become naturally acid through the prolonged action of rain in dissolving, replacing, and leaching calcium, magnesium, sodium, and to a lesser extent potassium from the silicate minerals and clays which in part came from the original soil-forming materials and form as secondary weathering products during soil formation.   These bases are replaced by hydrogen and aluminum ions.   Thus, naturally acid soils are generally low in calcium and magnesium, and poor plant growth is frequently due to lack of sufficient amounts of these elements for normal plant development.   Also, lack of available phosphorus, potassium, molybdenum, and other elements are frequently encountered in such soils.
      Natural soil alkalinity, on the other hand, arises from the weathering of rocks to form carbonates, bicarbonates, chlorides, and sulfates of sodium, calcium, and magnesium, followed by soil enrichment in these through periodic flooding, capillary rise from water tables, and irrigation.
      Soils may become acid through continued use of acid-forming fertilizers, such as ammonium sulfate, or applications of sulfur.   Conversely, soils may become alkaline through the deposition of salts and carbonates carried by irrigation waters and the application of fertilizers such as calcium nitrate, sodium nitrate, and lime.
      Various methods have been worked out to determine the approximate amounts of alkalinizing and acidifying chemicals needed to effect a given change in soils.   (A number of such methods are described by Chapman and Pratt, 1961.)   The amounts of alkalinizing or acidifying materials required vary greatly, depending primarily on the base-exchange capacity of soils.
      Under the intensive irrigation and fertilization program applied to most commercial citrus orchards, soil pH may change either up or down rather quickly and radically.   There are many instances, some already referred to in the sections on phosphorus and salinity, in which mineral deficiencies and excesses developed as a result of practices which altered soil pH.   Therefore, a periodic check of soil pH (using a 1:1 soil-to-water ratio) in various soil horizons of commercial citrus orchards is advised.   Practices should be altered if it appears that soils are becoming more acid than pH 5.5 or more alkaline than pH 8.5.


      Although not as sensitive to poor aeration as the avocado (Curtis, 1949), citrus is fairly intolerant to "wet feet."   Many troubles (such as root rotting, iron chlorosis, and salt injury) are brought on or aggravated by overmoist soil conditions or, perhaps more correctly, lack of sufficient oxygen for root respiration and the secondary soil conditions resulting therefrom.
      The classic work of Hoagland and Broyer (1936) and Steward (1935) established that oxygen is indispensable for vital activity in plant roots, and that the energy for nutrient and water absorption comes from the oxidation of carbohydrates in roots.
      Considering the importance of aeration to root health of citrus, nutrient and water absorption, resistance to root-decay organisms, and tree performance, this subject has not received deserved attention.
      After over thirty years of observations with citrus trees growing under controlled sand- and water-culture conditions and in many types of soils throughout the world, the writer has become convinced that a fair share of citrus disorders or substandard performance stems from inadequate aeration.
      In outdoor, continuously aerated water cultures with orange and lemon trees, accidental stoppage of aeration in summer for two days or less will often bring on root rotting.   Once started, root rotting is sometimes difficult to arrest.   When the drains of sand cultures sometimes become plugged by root growth, a telltale decaying root and peppery smell of the nutrient solution is a sign of trouble.   This is frequently followed by iron chlorosis and/or a lackluster appearance of foliage, followed by bronzing, vein chlorosis, and some abscission of leaves.
      Citrus trees are more likely to be adversely affected by lack of good aeration in summer than in winter.   This probably results from the following: (1) greater root and top growth (increased respiration rate and oxygen need); (2) greater microbiological activity and thus competition for oxygen; (3) less carbohydrate reserve in the plant root and a resulting lowered resistance to soil pathogens; and (4) other factors.   Reuther and Crawford (1947) noted much lower levels of oxygen in soil air after an irrigation in summer than in winter.   Undoubtedly, this was due to microbial activity.
      Prevatt (1959) grew seedlings of rough lemon, sour orange, sweet orange, and Cleopatra mandarin in flooded, sandy soils.   The flooding was started at different periods of the year.   When flooding was begun in January (at which time seedlings were making no growth), the plants showed no injury.   When flooding was begun in April (after they had started growth), leaf symptoms of water injury appeared after three weeks.   Seedlings flooded in June, July, and August showed injury symptoms in two weeks.   Seedlings grown in subsoil showed water-injury symptoms sooner than those grown in topsoil when flooding was started in summer months.   Carrizo citrange, Troyer citrange, sour orange, and rough lemon seedlings were more tolerant of free water in both limed and unlimed soils than sweet orange, Cleopatra mandarin, and Rusk citrange seedlings.   (Prevatt also noted that water extracts of incubated citrus roots and water extracts from waterlogged soil in metal cans contained toxic substances which caused citrus seedlings to wilt and become desiccated.)
      Poor soil aeration and resulting poor growth are not only a matter of insufficient oxygen for plant roots, but also carbon dioxide accumulation, the production of traces of other gases (such as hydrogen sulfide, methane, and hydrogen), accumulation of nitrite (in some instances), and increased solubilities of manganese, iron, and other nutrients.   It is likely that other secondary products of a toxic nature form.   Hence, this subject (like that of pH) requires consideration of both the primary effects of low oxygen as well as the adverse secondary conditions which develop in the wake of low-oxygen levels.
      A type of veinclearing in citrus leaves which results from the rotting of roots is shown in figure 3-78.   As pointed out in the sections on nitrogen and calcium, this condition can also arise from nitrogen and calcium deficiencies, old age, foot rot, and mechanical injury to the bark of trees, branches, and twigs; hence, it is not specific for waterlogging.
      Oxygen requirement and soil aeration have been studied by several techniques: (1) aerating nutrient cultures with gases of various oxygen partial pressures (e.g., oxygen and nitrogen mixtures), in which oxygen is varied from values approaching zero (100 per cent nitrogen) to over 20 per cent, and noting root and top growth, nutrient absorption, carbon dioxide excretion, and metabolic changes in plant roots and tops; (2) passing various oxygen-nitrogen gas mixtures over or into sealed pots of soil in which plants are growing; and (3) sampling and analysis of soil air (for oxygen, carbon dioxide, and other gases) at various depths and seasons of the year, in relation to soil temperature, moisture fluctuations, and soil fertilization.
      All of these methods are useful, but they do not measure actual oxygen conditions in the moisture films surrounding the plant root, for, as in all plant root-nutrient relations, adequacy is determined by both a concentration (intensity) and rate of renewal (capacity) factor.   It has been repeatedly shown that plants can obtain adequate nitrogen, phosphorus, potassium, and other nutrients from low concentrations provided these are maintained or renewed as fast as they are absorbed by plant roots.   The maintenance factors depend on supply, solubility, and diffusion rate.   With respect to oxygen, the following are important in determining both concentration in the water film and rate of its renewal: (1) the concentration of oxygen in the soil air; (2) the amount of soil air; (3) the effective diffusion coefficient through the water film surrounding the roots; and (4) the thickness of the water film; and (5) temperature (oxygen solubility decreases as temperature rises).
      A measurement of the amount of oxygen which can arrive at a given point per unit time (referred to as oxygen diffusion rate or ODR) is to some extent an integration of the aforementioned concentration, diffusion coefficient, and path-length factors.   The measurement can be made by inserting a platinum electrode and salt bridge leading to an Ag-AgCl cell into a soil, applying a certain electrical potential, and measuring the steady-state current.   Current flow at the particular potential chosen is produced by a two-step reduction of oxygen at the platinum-electrode surface involving the transfer of four electrons to each atom of oxygen (O2) reduced.   The current produced under specified conditions then is essentially a measure of the rate of oxygen diffusion (ODR) to the platinum electrode.
      Adopting the use of a 25-gauge platinum electrode with an applied voltage of 0.65 volt and readings of current flow after four minutes, Stolzy and Letey (1964) and their associates have made extensive studies.   Oxygen diffusion rate (ODR) is conveniently expressed as micrograms of oxygen reduced per square centimeter of electrode surface per minute or, in abbreviated form, μgm cm-2 min-1.   (Many results have been reported in units of gm cm-2 min-1 X 10-8.   These results can be converted to micrograms by dropping the 10-8 and moving the decimal point two places to the left.)
      In their preliminary research, plants were grown in sealed, double-walled cylinders of soil (the inside cylinder of plexiglass so as to observe root growth) supplied with various gas mixtures which flowed over the surface of the soil.   By inserting the electrode at appropriate ports in the sides of the cylinder and noting root growth through the transparent plexiglass walls in relation to the various gas mixtures, values for oxygen diffusion rate (ODR) at which growth practically ceased have been established for a variety of crops.   In the case of citrus seedlings, the critical limiting value for root growth at summer temperatures was about 0.20 μgm oxygen cm-2 min-1.   At values of 0.31, some root growth occurred.   In winter, a higher ODR was required, and it is hypothesized that perhaps in summer the lower ODR requirement is due to some internal movement of oxygen from top to roots.
      In studies correlating nutrient uptake with ODR, Stolzy et al. (1963) found that leaf concentrations of potassium, phosphorus, calcium, magnesium, iron, manganese, and boron were reduced at ODR's below 0.33 μgm cm-2 min-1, compared with an ODR above 0.62.   On the other hand, chloride values increased as the ODR dropped.
      In further studies of a similar nature, Labanauskas et al. (1965) found that uptake of eleven elements was decreased as oxygen supplied over the soil surface decreased.   They used sweet orange seedlings of the Bessie variety.   Sodium and chloride absorption increased in tops as ODR decreased, and similar results have been noted for manganese.
      Stolzy et al. (1963) studied the effect of ODR on the citrus nematode Tylenchulus semipenetrans and found that this organism was more sensitive to oxygen lack than sweet orange seedling roots.
      With respect to root-rotting fungi, Klotz, Stolzy, and DeWolfe (1963) found that Thielaviopsis basicola had a higher oxygen requirement (ODR) than Phytophthora parasitica and P. citrophthora.   The Phytophthora species were reduced in numbers when the ODR was 0.12 μgm cm-2 min-1 as compared with 0.22.   These preliminary results suggest that the citrus plant is more sensitive to oxygen lack than these organisms.
      Many other studies have been made, such as the effect of soil depth (ODR generally decreases with depth, as might be expected), effects of soil compaction, water table, temperature, soil moisture, and air pollutant damage in relation to ODR.
      These and many specific literature citations dealing with other crops are briefly mentioned in the summary published in Advances in Agronomy by Stolzy and Letey (1964).
      Preliminary standards for quite a few crops have already been worked out.   The comparative simplicity of the equipment and its portability has made it possible to make measurements in the field, as well as under controlled conditions in the greenhouse and laboratory.
      Studies by Stolzy and Letey (1964) and others indicate that with their technique, ODR values of 0.20 μgm cm-2 min-1 inhibit the root growth of many plants.   Values between 0.20 and 0.30 retard root growth.   With values greater than 0.40 μgm cm-2 min-1, top growth of many plants is not impaired.
      Combined with measurements of related or consequential effects in soils (such as other soil gases, solubility of soil constituents, and activities of microorganisms), a more realistic understanding of the aeration factor as it affects the performance of citrus and other plants is emerging from this newer approach.
      Previous data and findings with respect to oxygen requirement and soil aeration are of interest and are briefly covered here to round out this review.
      In short-term, solution-culture studies with sour orange seedlings in nutrient cultures aerated with various oxygen-nitrogen gas mixtures, Girton (1927) found that root growth was completely suppressed when the oxygen content of the gas was 1.2 to 1.5 per cent, as compared with controls receiving normal air.   At 5 to 8 per cent of oxygen, root growth was about one half that of the controls.   The latter experiment was with solutions held at 25°C, the former at 28°C.   The presence of carbon dioxide at 37 to 55 per cent (with oxygen at 16.5 to 20 per cent, the remainder nitrogen) almost completely suppressed root growth.
      In experiments with sweet orange seedlings, Cannon (1925) found some root elongation in cultures supplied with gas mixtures with as little as 0.8 per cent oxygen, 5.8 per cent carbon dioxide, and the rest nitrogen.   The temperatures were controlled and ranged from 20 to 25°C.   Where oxygen was excluded completely, no root growth occurred.   In another five-day test, root elongation was 10 mm where the gas mixture contained only 0.5 per cent oxygen, the rest nitrogen, as compared with 20 mm for the cultures receiving ordinary air.   In other experiments, the oxygen requirement appeared high.   Cannon (1925) concluded from his experiments that the minimum oxygen concentration for the sweet orange is above 1.4 per cent oxygen at a soil temperature of 20°C.   Brazilian sour orange seedlings appeared to have a somewhat lower oxygen requirement, ranging from 0.6 to 1.6 per cent.   Where the carbon dioxide content of the gas did not exceed 21 per cent, the roots of sweet orange were found to grow when the quantity of oxygen present was less than 2 per cent.
      Curtis (1949) grew sweet orange and avocado seedlings in 20-liter Pyrex bottles of nutrient solution supplied with nitrogen-oxygen gas mixtures providing solution concentrations ranging from 0.05 to 28.0 ppm of dissolved oxygen.   The work was done in a greenhouse in which solution temperatures ranged from 20 to 25°C.
      While root growth of citrus seedlings did not entirely cease during a 10-day aeration period with oxygen at 0.05 to 0.6 ppm, the growth was poor.   It was rated as fair at levels of 1 to 2.9 ppm, good at 6.4 to 6.8 ppm, and excellent at 8 to 8.6 ppm.   At 16 ppm, growth was only fair, and at 28 to 32 ppm, the roots became stubby.   Pure oxygen was used to secure the latter concentration.   The avocado was much more sensitive to low oxygen levels, and root growth stopped at 0.05 to 0.06 ppm.   The avocado leaves also wilted, whereas there was no leaf wilt of citrus at the lowest oxygen levels.   There was root-tip discoloration of citrus roots below 0.7 ppm of oxygen, and after several days the root tips became soft and sloughed off.   Injury to citrus roots was noted during the very first day of exposure to low oxygen.   Avocado roots showed even greater injury.   In spite of the sloughing of new root tips, new citrus roots pushed through the cortex of older roots even at the lowest oxygen concentration.
      Reuther and Crawford (1947) studied the fluctuation of oxygen and carbon dioxide in differentially-fertilized field plots with young grapefruit in relation to both frequent (wet) and infrequent irrigation treatments.   This work was carried out on a Coachella very fine sandy loam in the Coachella Valley of California.   The soil contained considerable free calcium carbonate.   Gas sampling wells were established at two soil depths—10 and 20 inches.   Determinations were made during all seasons.   Trees in the frequently irrigated plots (wet treatments) showed iron chlorosis in winter, but this could not be related to oxygen levels, for in general higher oxygen was found in the soil air in winter than in summer.   At no time did the oxygen levels of the soil air in winter fall below 12 per cent.   Of course, the oxygen diffusion rate could have been limiting in spite of a good oxygen supply in the pores.   One of the striking findings in this investigation was the rather sudden decrease in the oxygen content of soil air, especially at the 10-inch depth shortly after an irrigation.   Greatest decline was in a manure-treated plot, where on some occasions the soil air composition changed from 12 to 15 per cent of oxygen before irrigation to less than 1 per cent after irrigation.   This low oxygen state held for several days and did not return to the preirrigation levels for almost a week.   The sudden drop of oxygen was followed (with a lag period) by an increase in carbon dioxide.   These fluctuations after irrigation were much more marked in summer than in winter, reflecting greater plant and microbiological activity.
      Already mentioned under the section on iron deficiency is the work of Wallihan et al. (1961), who were able to lower the iron content of the leaves of citrus seedlings growing in potted soils, supplied with low-oxygen gas mixtures containing 2 to 6.2 per cent of oxygen.   The gas mixtures were passed over the surface of the potted (sealed) soils.
      Chapman et al. (1965) noted in outdoor water-culture studies with oranges growing under various calcium concentrations that roots under high-calcium conditions could withstand poorer aeration than those growing under low-calcium conditions.   Under moderate calcium deficiency, good roots were produced when aeration was at a maximum, but root rotting occurred if aeration was poor.
      Improvement of Soil Aeration.—Periodic deficiency of sufficient oxygen for optimum root growth and citrus performance probably occurs to some degree on all but the most sandy, open, or porous soils.   It is an especially serious problem where free and rapid downward movement of water is impaired by clay layers, hardpans, plowsoles, and/or perched water tables, where there are prolonged periods of heavy rainfall, or where surface soil structure has been so impaired that water stands on the surface for considerable periods.   On many irrigated soils, it is certain that during and immediately after irrigation there are temporary oxygen deficits in local soil zones or areas of marked root development, or in areas where there is considerable organic material and microbial activity is high.
      Improvement of soil aeration can be accomplished in many ways.   Some are quite simple and inexpensive, while others (such as the installation of the drains to facilitate movement of excess water out of the soil) are expensive.
      Improvement of soil structure, with some improvement in aeration, can be accomplished by the more or less regular use of organic matter (either grown as green manures and turned under or added as manures) or bulky organic materials (such as straws and hays of various kinds).   Organic mulches of wood shavings or cotton hulls and periodic mowing of permanent or semipermanent sod cover or weeds are all well-known techniques.
      Noncultivation and control of weeds by hand hoeing, mowing, or herbicides is another technique for improving or restoring soil structure in surface layers of soil.
      Plowsole which develops from overcultivation will gradually disappear or become less impervious under noncultivation.   Compacted layers can be broken up by deep chiseling, and while some roots are destroyed in the process, most experience indicates that no permanent harm to the orchard will result.
      Deeper clay layers and very heavy soil in which water moves slowly are harder to deal with, but the use of deep-rooted green manure crops, such as pigweed (Amaranthus retroflexus), alfalfa, or other crops, will prove helpful.
      Where clay layers or hardpans are not too deep, it is now a common practice before citrus is planted to break these up by deep plowing or chiseling.
      Rotational irrigation (in which alternate sections are watered, then allowed to dry to near the wilting point before wetting again) is a common practice in some soils and areas.   Tensiometers are especially useful in helping to guide irrigation practices.   Many citrus orchards, where periodic iron chlorosis has been a problem, have been strikingly improved by allowing lower soil horizons to dry out before irrigation.
      In some areas subject to heavy rainfall for prolonged periods, ridging of soils is a good practice.   For example, with certain Florida soils, tree rows are planted on the crests of rather good-sized ridges or "beds."   These are constructed so that the height between the ridge crest and bottom is as much as two to three feet.   Thus ridge valleys or large furrows separate the tree rows which are used for irrigation during the dry season and as drainage channels to take away excess water during heavy rainy periods.
      Vertical mulching with organic matter in trenches is an aid to aeration.   In the Ahmednagar district of Bombay presidency of India, Sahasrabuddhe (1927) reported that in a heavy-type soil where orange tree dieback (due probably to poor aeration) was severe, digging of trenches 3 feet wide and 3 feet deep and filling them with broken tile and stone in the bottom and soil plus manure on top greatly reduced the amount of dieback and improved tree color and fruiting as compared with trees which received the same amount of manure applied to the soil surface in a ring around the tree.   Best results were obtained with trenches on three sides of the tree, but even on one side of the tree a trench produced substantial improvement, The fertilizer per tree consisted of 20 pounds of farmyard manure, 10 pounds of bone meal, and 3 pounds of oil cake.
      Chemical soil treatments to improve aeration have been suggested.   Some researchers are of the opinion that ample nitrate fertilizer in the root zone and in roots will provide some oxygen for plant roots during periods of oxygen shortage in the soil.   Haas (1937) found in solution cultures that high nitrate could to some degree compensate for lack of aeration.   However, Bain and Chapman (1940) grew avocado and grapefruit seedlings in nitrate-treated, waterlogged soils and got no evidence of a beneficial effect.   In fact, with the avocado, injury from waterlogging was aggravated by the addition of calcium nitrate.   This was probably due to the immediate reduction of some of the nitrate to nitrite, and the latter was toxic to the avocado root.
      It is probable that some rootstocks and rootstock-scion combinations are much more tolerant of poor aeration than others.   The sour orange, for example, appears to be somewhat more tolerant of low-oxygen conditions than the sweet orange.   Much more information is needed in this field than now exists.
      From the preceding discussion of oxygen requirement and review of existing information, it is quite evident that citrus is reasonably tolerant to low-oxygen conditions and that complete cessation of root growth and actual root injury may not occur until levels of oxygen in soil moisture films are at levels below 0.5 ppm of oxygen.   However, best root growth occurs only when oxygen is somewhere near its maximum solubility (8 ppm at 25°C).
      The investigations of Stolzy and Letey (1964), using their oxygen diffusion rate (ODR) technique for determining oxygen status, have provided an excellent tool for studying this problem much more rapidly and thoroughly than has been possible hitherto.   From a practical point of view, it is especially important to find out whether, or the extent to which, top performance is impaired by temporary periods of poor aeration, the effects of variable nutrition on oxygen need, rootstock tolerance and effects, and relations to compacted subsoil layers, recurring climatic extremes, and other problems.
      Knowledge in these areas may point the way to better means of coping with soil aeration problems under a wide range of soil conditions.





1.   A key to the early literature on fluorine may be found in Volume 1 of the Bibliography of the Literature on the Minor Elements, published in 1948 by the Chilean Nitrate Educational Bureau, Inc., New York.   A more recent review, with special reference to effects of fluorine air pollutants on plants, has been given by Thomas (1961).

2.   A bibliography (with abstracts) of molybdenum literature, complete up to March, 1961, and containing over 1,200 citations, has been published by Borys and Childers (1961).   A supplemental bibliography (with abstracts) covering molybdenum literature from March, 1961, to June, 1965, prepared by L. G. Albrigo, Richard C. Szafranek and Norman F. Childers is available through Climax Molybdenum Company, 1270 Avenue of the Americas, New York 20, New York.

3.   The sulfur in the organic fraction may be tied up as sulfate covalently bound to polysaccharides, as sulfate esters of phenols, or in cystine and methionine forms (Freney, 1961).

4.   A good compilation by D. L. Carter entitled A Bibliography of Publications in the Field of Saline and Sodic Soil (Through 1961) was published in 1962 by the U.S. Department of Agriculture, Agricultural Research Service (Carter, 1962).