EFFECT OF POTASSIUM, CULTIVAR, AND SEASON ON TOMATO BLOTCHY RIPENING, GRAYWALL, AND FRUIT QUALITY By David Harry Picha A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1980 TO MY PARENTS AND TO RON AND TERRY ACKNOWLEDGEMENTS The author expresses his sincere thanks to Dr. C. B. Hall, Chairman of his committee, for his support, encouragement, and counsel during the course of his research. He extends appreciation to Dr. T. E. Humphreys, Dr. R. E. Stall, and Dr. D. J. Cantliffe for also serving on his committee Special thanks go to Dr. R. C. Smith and Dr. S. J. Locascio for both their professional counsel and personal friendships. Appreciation is extended to Dr. D. D. Gull and Dr. D. N. Maynard for their interest and support. The author is grateful to Mrs. Wanell Cheshire for typing this manu- script. Most thanks of all go to the author's family for their understanding and support . iii TABLE OF CONTENTS Page ACKNOWLEDGEMENTS iii LIST OF TABLES vi LIST OF FIGURES ix ABSTRACT x INTRODUCTION 1 LITERATURE REVIEW 3 Blotchy Ripening 4 Graywall 10 Fruit Quality of Blotchy Versus Normal Fruit- Compositional and Biochemical Differences 21 MATERIALS AND METHODS 23 Blotchy Ripening — Greenhouse 23 Field Experiments 27 Blotchy Ripening 28 Graywall 28 Fruit Quality Measurements 30 Graywall — Greenhouse 33 Ethylene Measurements 35 Oxygen Consumption 36 Statistical Analyses 37 RESULTS 38 Influence of Potassium and Cultivar on Blotchy Ripening of Greenhouse Grown Tomatoes 38 Influence of Potassium, Cultivar, and Season on External Blotchy Ripening of Field Grown Tomatoes 38 Influence of Potassium, Cultivar, and Season on Internal Blotchy Ripening of Field Grown Tomatoes 46 Influence of Potassium, Cultivar, and Season on Leaf Potassium Content 51 Influence of Potassium, Cultivar, and Season on Pericarp Potassium Content 53 Influence of Potassium, Cultivar, and Season on Pericarp Calcium, Magnesium, and Phosphorus Content 53 iv Page Influence of Potassium, Cultivar, and Season on Bacterially Induced Graywall Development 56 Influence of Potassium and Cultivar on Bacterially Induced Graywall Development of Greenhouse Grown Tomatoes 60 Influence of Storage Temperature on Bacterially Induced Graywall Development 60 Influence of Potassium and Cultivar on Natural Gray- wall Development 62 Influence of Potassium, Cultivar, and Season on Pericarp Soluble Solids 62 Influence of Potassium, Cu'ltivar, and Season on Pericarp Acidity 66 Influence of Potassium, Cultivar, and Season on Pericarp pH 69 Influence of Potassium, Cultivar, and Season on Pericarp Reducing Sugars 69 Influence of Potassium, Cultivar, and Season on Pericarp Dry Weight 72 Ethylene Evolution and Oxygen Consumption Rates of Blotchy Versus Red Tissue 72 DISCUSSION 77 Blotchy Ripening 77 Tissue Analyses 79 Graywall 81 Fruit Quality 83 Ethylene and Respiration 86 SUMMARY AND CONCLUSION 88 APPENDIX 92 LITERATURE CITED 109 BIOGRAPHICAL SKETCH 117 v LIST OF TABLES Table Page 1 Standard Johnson's solution. 24 2 Monthly temperature averages during the 1979 spring and fall growning seasons at the Horticultural Unit, Gainesville, Florida. 34 3 Effect of reduced potassium levels after first cluster anthesis on external blotch of three tomato culti- vars grown in greenhouse sand cultures. Spring, 1978. 39 4 Effect of reduced potassium levels after first cluster anthesis on internal blotch of three tomato culti- vars grown in greenhouse sand cultures. Spring, 1978. 40 5 Ratings of external blotch of three different harvest dates of fruit from four tomato cultivars field grown under five potassium levels during the fall 1979 season. 41 6 Ratings of external blotch at three different harvest dates of fruit from four tomato cultivars field grown under five potassium levels during the fall 1979 season. 42 7 Rating of external blotch of fruit from four tomato cultivars field grown under five potassium levels during the spring and fall 1979 seasons. 44 8 Ratings of internal blotch at three different harvest dates of fruit from four tomato cultivars field grown under five potassium levels during the spring 1979 season. 47 9 Ratings of internal blotch at three different harvest dates of fruit from four tomato cultivars field grown under five potassium levels during the fall 1979 season. 48 vi Table Page 10 Rating of internal blotch of fruit from four tomato cultivars field grown under five potassium levels during the spring and fall 1979 seasons. 49 11 Potassium content (% dry weight) of leaves of spring and fall 1979 field grown tomatoes from four cultivars grown under five soil potassium levels. 52 12 Potassium content (% dry weight) of fruit pericarps of spring and fall 1979 field grown tomatoes from four cultivars grown under five soil potassium levels. 54 13 Calcium content (% dry weight) of fruit pericarps of spring and fall 1979 field grown tomatoes from four cultivars grown under five soil potassium levels. 55 14 Magnesium content (% dry weight) of fruit pericarps of spring and fall 1979 field grown tomatoes from four cultivars grown under five soil potassium levels. 57 15 Phosphorus content (% dry weight) of fruit pericarps of spring and fall 1979 field grown tomatoes from four cultivars grown under five soil potas- sium levels. 58 16 Rating of bacterially induced graywall of spring and fall field grown tomatoes under high and low potassium treatments. 59 17 Rating of bacterially induced graywall in fall green- house grown tomatoes under high and low potas- sium treatments. 61 18 Rating of bacterially induced graywall resulting from one day of chilling at 4°C followed by six days of storage at 20°C. Fall field experiment. 63 19 Rating of bacterially induced graywall after five days at 20°C without any previous chilling exposure. Fall greenhouse experiment. 64 20 Percentage of fruits having natural graywall from winter greenhouse tomatoes grown under four different potassium treatments. 65 vii Table Page 21 Pericarp % soluble solids of spring and fall field grown tomatoes from four cultivars grown under five soil potassium levels. 67 22 Pericarp % acidity of spring and fall field grown tomatoes from four cultivars grown under five soil potassium levels. 68 23 Pericarp pH of spring and fall field grown tomatoes from four cultivars grown under five soil potas- sium levels. 70 24 Pericarp % reducing sugars of spring and fall field grown tomatoes from four cultivars grown under five soil potassium levels. 71 25 Pericarp % dry weight of spring and fall field grown tomatoes from four cultivars grown under five soil potassium levels. 73 26 Ethylene evolution rates of yellow (blotchy) versus red outer pericarp tissue. 74 27 Oxygen consumption rates of yellow (blotchy) versus red outer pericarp tissue. 76 viii LIST OF FIGURES Figure Page 1 Illustration of fruit with an external blotchy ripening rating of 1, 3, and 5. 26 2 Illustration of fruit with an internal blotchy ripening rating of 1 and 5. 26 3 Illustration of fruit with a graywall rating of 1, 2, 3, and 4. 31 ix Abstract of Dissertation Presented to the Graduate Council of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy EFFECT OF POTASSIUM, CULTIVAR, AND SEASON ON TOMATO BLOTCHY RIPENING, GRAYWALL, AND FRUIT QUALITY By David Harry Picha December 1980 Chairman: Dr. C. B. Hall Major Department: Horticultural Science Tomatoes (Lycopersicon esculentum Mill.) from four different cultivars, 'Healani', 'Homestead 24', 'Walter' and 'Flora-Dade' were grown under five different applied soil K rates (0, 93, 186, 372, and 745 kg K/ha) during spring and fall seasons to determine the influence of K rate, cultivar, and season on blotchy ripening, graywall, and fruit quality. Under both spring and fall field conditions, fruits from all four cultivars had more blotchy ripening with decreasing K fertility. At all K rates and with each cultivar both external blotch and internal blotch were more severe in the spring than in the fall. Yellow shoulder, the primary external blotch symptom, was severe in the spring with the cultivars 'Homestead-24', 'Walter', and 'Flora-Dade'. 'Healani' showed resistance to yellow shoulder blotch. During x the fall, yellow shoulder blotch was not a severe problem in any cultivar. 'Healani' was generally the most blotch resistant cultivar and 'Flora-Dade' was the most blotch susceptible. There was more K in both the leaf and pericarp tissue of all cultivars with increasing soil K rate in both seasons. The differential susceptibility between cultivars to blotchy ripening was not associated with cultivar differences in leaf K or pericarp K content during either season. A graywall inducing bacteria, Erwinia herbicola (Dye), was inoculated in the outer pericarp of immature green tomato fruit to determine the effect of cultivar and K fertility on tissue susceptibility to graywall. All four tomato cultivars developed more graywall with low K (0 kg K/ha) compared to high K (745 kg R/ha) during the spring season. Only 'Flora-Dade' developed less graywall with increasing K rate during the fall season. During both seasons, 'Flora-Dade' and 'Homestead-24' were the two most graywall resistant cultivars while 'Healani' and 'Walter' were the most susceptible cultivars. Natural graywall (contrasted to bacterially induced graywall) appeared in 'Healani' and 'Homestead-24' fruits grown with low K treatments in a sand culture greenhouse experiment. Both cultivars were free of natural graywall with the high K treatment. 'Flora-Dade' was completely resistant to natural graywall with all K treatments. Environment, cultivar, and applied soil K all influenced the quality of field grown tomato fruit. During the spring each cultivar except 'Flora-Dade' had a higher pericarp soluble solids and reducing sugar content under 745 kg K/ha than 0 kg K/ha. During the fall only xi 'Healani' and 'Flora-Dade' had higher soluble solids with 745 kg K/ha, but no difference in reducing sugar content was found between applied soil K rates. 'Healani' was the only cultivar during both seasons to increase in pericarp dry weight with increasing K rate. During both seasons and with each cultivar there was an increase in pericarp acidity with increasing K rate. Fruit from all cultivars decreased in pericarp pH with increasing soil K rate during the spring season. 'Healani' had the highest and 'Flora-Dade' had the lowest pericarp soluble solids, reducing sugars, dry weight, and pH each season. 'Healani' and ' Horaestead-24 ' had the lowest pericarp acidity while 'Walter' had the highest. Reducing sugars, dry weight, and pH were all higher during the spring than fall. Acidity was greater during the fall. Only 'Healani' and 'Flora-Dade' had a higher soluble solids content during the spring than fall. 'Healani' generally had the most desirable quality characteristics while 'Flora-Dade' had the least. Fruit grown without added K generally had the poorest quality. xii INTRODUCTION Tomato, Lycopersicon esculentum Mill., Is one of the most widely cultivated vegetable crops in the U.S. and throughout the world. In terms of crop value, it is the number one vegetable grown in Florida, with an annual income of 220 million dollars during 1978-79 (Florida Agricultural Statistics, 1980). Slightly over 450,000 acres were grown for fresh market and processing in the U.S. during 1979, with a crop value of over one billion dollars (U.S.D.A., 1979). Tomatoes may be affected by a number of physiological and patho- logical disorders during the maturation and ripening processes. Ulti- mately these disorders result in reduced yields and/ or quality of the fruit. Two of the most troublesome and economically important dis- orders affecting fruit quality are blotchy ripening (BR) and graywall (GW). Fruits afflicted with either disorder generally are unmarket- able. Blotchy ripening appears externally on the fruit as areas of yel- low or orange discoloration intermixed with the normal red areas. Blotchy areas may develop some red color, but never to the extent of normally colored areas. The amount of discoloration may range from very small patches to large areas in which more than half the fruit is discolored. The location of the discolorations may be random through- out the fruit or semi-ordered in which only the shoulder region is 1 2 yellow-orange. Internal blotchy ripening symptoms range from slight to severe whitish discoloration of the placenta and pericarp tissue. In severe cases the outer pericarp appears brownish due to lignifica- tion of the vascular strands. Externally the fruit may be perfectly red, but internally it may contain white tissue. Graywall symptoms appear externally as grayish-brown discolora- tions in the outer fruit wall. In contrast to BR, GW symptoms appear on immature green fruit. The grayish-brown areas may become slightly depressed and roughened. Internally, severe browning usually appears in the outer pericarp, especially in the areas next to the vascular bundles. The author's objectives were to determirae the effect of applied soil K rate, cultivar, and season on the incidence and severity of BR, bacterially induced GW, and fruit quality. Physiological differences and/or similarities between blotchy and normal tomato tissue were also studied with respect to ethylene evolution and respiration rate. By meeting the above objectives, a better understanding regarding the control of these disorders may result. Numerous factors may Influence the severity of BR and GW without being the primary cause. A study of the above factors should be of benefit because the knowledge gained may contribute to the control of these disorders. LITERATURE REVIEW Much confusion exists in the literature regarding terminology of uneven tomato ripening disorders. The range of terras used to describe such ripening abnormalities includes blotchy ripening, graywall, vascular browning, internal browning, cloud, waxy patch, and green- back. Inadequate description of the symptoms of these disorders has often prevented comparison and accurate interpretation of research re- sults. Using specific names for what may be different disorders with very similar symptoms or which may be different stages of development of the same disorder has confused the picture. It is the author's contention that all these terms have been used to describe two separate disorders: BR and GW. A close examination of the literature reveals that BR and GW are unique and distinct dis- orders. The terms vascular browning, internal browning, and cloud are probably forms of GW and the term greenback is a form of BR. Graywall occurs in green fruit whereas BR by definition can only be identified in fruit which has begun to color. Only when one compares a mild form of GW to a severe case of BR (brownish discoloration of vascular bundle area due to lignif ication) is there justification for confusion regarding what term should be applied to the disorder. Researchers in this area of uneven tomato ripening should confine themselves to the terms BR and GW in describing their disorders. Use of additional synonyms to describe these two disorders can only result in further confusion. Hobson et al. (1977) provided an excellent 3 4 description of tomato ripening disorders and included illustrations on the separate disorders of BR and GW. Blotchy Ripening The first mention of BR in the scientific literature occurred in 1922 in the Annual Report of the Cheshunt Experiment Station (Anon, 1921). Blotchy ripening affected fruits were described to have hard green areas around the calyx and elsewhere, in which the vascular bundles were or were not discolored, while the remainder of the fruit was red. Since 1922 much work on BR has been done throughout the world. The term has come to mean little more than non-unif ormly colored fruits. Bewley and White (1926) described BR as hard green areas arising in any position on the fruit without an abrupt line of demarcation be- tween the green and red areas. In severe cases, the hard areas were waxy or glassy in appearance and the bundles lying beneath these clear glossy areas were brown and necrotic. It is possible that these authors were describing both GW and BR but had classified them all as BR. "Greenback” or "hardback” was placed in a separate category from true BR and was used to describe fruits that had a hard green or orange area around the stem end while the rest of the fruit turned red. The authors found the amount of BR to vary with season, culti- var, and fertility. Blotchy ripening was thought to be the result of primarily K and also N deficiencies. Sunlight or other factors were postulated to influence BR since a small amount of BR occurred on heavily manured plots. Susceptibility to greenback was increased by K and N deficiency and exposure of the deficient fruits to excessive sunlight 5 A histological study of BR was made by Seaton and Gray (1936). They also grouped GW and BR symptoms together and they stated that BR is characterized by a failure of areas of the outer fruit wall to color normally. As the fruit approached maturity, these areas remained hard and green and eventually assumed a waxy or glassy appearance (GW?). The vascular bundles lying beneath the blotchy tissue appeared brown and necrotic and in severe cases cavities appeared adjacent to the vascular bundles. First evidence of blotchiness was observed only after the fruit began to develop color. Histological examination revealed that the discolored tissues were parenchyma cells of the fleshy layer and the vascular bundles were not involved. Bands of discolored material were located between the epidermis and bundles which made it appear that the bundles were affected when viewed through the epidermis. The authors stated that the breakdown of cells adjacent to the bundles in the blotchy areas severed the connection to outlying cells for transfer of assimilates and water, and normal ripening was inhibited. It was also thought that BR was caused by withdrawal of water from the fruits during periods of excessive transpiration, occurring two to five days before the fruit ripen. White (1938) hypothesized that BR may be the result of abnormal carbohydrate metabolism in which translocation of carbohydrates from leaves to fruits was impaired and abnormal water relations may be involved. The failure of blotchy and greenback areas to ripen was thought to be associated with a lack of starch to sugar conversion. Increased K and increased sunlight reduced the amount of BR. 6 Selman (1943) found that high K fertilization reduced BR in tobacco mosaic virus (TMV) infected and non-inf ected plants. The per- centage of fruit that devloped BR increased with TMV infection. Jones and Alexander (1956) reported high K fertility reduced BR and the effect of TMV infection on BR was inconclusive. Closs (1958) reported that low light intensities 1-2 weeks before coloration caused abnormal ripening and' this condition was associated with reduced sugar content. However, he made no separation between GW and BR. Geraldson (1960) also did not clearly distinguish between GW and BR, but reported the incidence and severity of BR was associated with low nutrient concentrations and low nitrate to chloride ratios in the nutrient solution. Cooper (1960) found no correlation between the amount of BR and number of fruits in an inflorescence or position of fruits within the inflorescence. Berry et al. (1964) noted that blotchy outer wall tissue was associated with underlying locular areas of low seed density and associated low auxin activity. The authors hypothesized that these areas of low auxin activity impaired normal metabolism and resulted in BR. In addition to reports previously cited, many workers have noted that increasing K rate reduces BR. Winsor et al. (1961) reported a reduction in BR and greenback with increasing potash to nitrogen ratio from 1.0 to 2.5. Winsor and Long (1967) and Winsor et al. (1965) found all forms of BR to be markedly reduced by increasing K fertil- ity. Hayslip and Iley (1964, 1965, 1967) studied the influence of N to K2° ratios on "graywall," even though they were actually looking at a combination of GW and BR as judged by their symptom descriptions. Based on their study, it is impossible to determine the influence of 7 K on the incidence of the separate disorders. They found much more "graywall" in 'Homestead-24' fruits fertilized with 50 than 400 and 800 lbs K20 per acre. High N/K20 fertilizer ratios resulted in much more "graywall" than low N/K20 ratios. They suggested this disorder might be reduced by using fertilizer with at least twice as much K20 as N, but also noted that additional climatic factors were necessary for "graywall" development. Jones and Alexander (1962) reported an inverse correlation between K nutrition and BR. Collin and Cline (1966) found that BR resulted from replacement of K in the nutrient solution with Ca or Na, but not NH^. Shoulder blotch (greenback) and white wall tissue were attributed to a low K supply and excess light. The premise that K deficiency causes BR was not completely supported because substituting NH^ for K did not result in increased BR. Van der Boon (1973) concluded that BR incidence was lower at the high K/Ca fertility ratio (5:1) than the low K/ Ca ratio (1:5). Van Lune and Van Goor (1977) also reported that BR and greenback lessened with increasing K/Ca ratio. Minges and Boutonnet (1966) and Ozbun et al. (1967) found low K treatments to result in significantly more BR with respect to internal white tissue than high K treatments. A highly significant correlation occurred between white tissue and low K content of the petiole. Turnipseed and Ozbun (1974) reported that a BR resistant breeding line showed a higher K uptake rate than susceptible lines and hypothesized that this may be a factor in BR resistance. In contrast to the above reports on decreased incidence of BR with increasing K, Clarke (1944) found no clear relationship between K and BR and suggested that sunlight was more important than K in 8 reducing BR. In several seasons. Cotter (1961) was generally unable to find an effect of varying levels of K, N, B, or TMV infection on BR. Woods (1964a) reported that addition of K2SO4 did not reduce the incidence of greenback and total blotch in three out of four years. Woods (1964b) found that varying the N/K ratios from 2:1 to 0.33:1 did not influence the occurrence of any form of blotch in any of four years. Many environmental factors have been shown to influence BR. Cooper et al. (1964) observed that shading the greenhouse reduced BR. Woods (1963) reported that shading reduced BR only when the plants were grown against the outside greenhouse walls, with most of the re- duction being in the amount of greenback. Defoliation resulted in a marked rise in greenback. The incidence of yellow blotch and waxy patch (GW?) was increased by shading. Woods (1965) concluded that there was no consistent relationship between 16°C and 21°C tempera- tures and either green or yellow blotch. Greenback was least in the lower temperature regime (16°C) in three out of four years. Venter (1965) suggested that greenback was caused by variations in terapera- ature on the fruit pericarp during maturation. Higher fruit tempera- tures were associated with a greater incidence of greenback. Collin and Cline (1966) stated that shade and high humidity did not increase BR, but exposing greenhouse grown plants to supplementary light pro- duced severe blotch regardless of nutrient treatment. The effect of supplementary light could have been due to the heat given off by the lamps. Lipton (1970) stated that yellow shoulder development was not primarily a heat effect, but mainly a result of short wave radiation. He found more yellow shoulder under high relative humidity (70%-100%) 9 than low humidity ( 50%— 90% ) and attributed this to a higher effective radiation load with the higher relative humidity. Jones and Alexander (1962) reported that BR was increased by subjecting greenhouse grown plants to low temperature followed by high temperatures, or to a regime of high soil moisture, high N, and shade. High levels of BR occurred on plants infected with or free of TMV. Unfortunately, no distinction was made between BR and GW. Matsumoto and Hornby (1974) and Hornby and Matsumoto (1974) concluded that alternating weeks of sunny and cloudy days produced the highest amount of BR. Alternating weeks of different temperature levels with consistent sunny-day light also produced high levels of BR. They hypothesized that varying tem- perature or light can produce a stress on plants resulting in blotchy fruit. An extensive histological study of the various external and in- ternal symptoms of BR was conducted by Sadik and Minges (1966). They stated that white and brown tissue are the two types of abnormal tis- sues present in blotchy fruits. White tissue appeared to be the basic abnormality associated with internal and external symptoms. The pre- sence and location of white tissue in the pericarp determine the type of external symptoms, i.e. blotch, yellow shoulder, yellow streaks, subsurface yellowing, or yellow ring. White tissue may also be found in the septa and placental areas. When the white tissue is adjacent to the fruit surface this externally results in blotch; when the white tissue is separated from the fruit surface by a layer of red tissue the fruit may look perfect externally or have slight subsurface yel- lowing; when the white tissue is deeply embedded in the pericarp the fruit is not externally discolored. An unknown gas contained in the 10 white tissue was found to cause the whitish appearance. The hardness of this tissue was caused by cell wall lignif ication. Starch content of this tissue did not decrease with ripening which indicated a lack of starch hydrolysis. Early stages of BR were microscopically observed in immature green fruits about 25 mm in diameter. Sadik and Minges stated that this early disorganization starts in a single or small group of parenchyma cells in the pericarp located around vascular bundles. Lignif ication of the parenchyma cell walls ultimately leads to cell breakdown and associated ripening abnormalities. In severe cases of BR, brown tissue occurs as strands or bands in tissues around and be- tween the vascular bundles. Brown tissue may extend into the abscis- sion zone and into the septal and placental tissues. Brown tissue is much less common than white tissue, with which it is always asso- ciated. This BR brown tissue described by Sadik and Minges differs from that caused by TMV induced internal browning (Wharton and Boyle, 1957) and GW associated browning (Stoner and Hogan, 1950; Hall and Stall, 1967). Immature internally browned or GW fruit, if allowed to ripen, will be "blotchy-ripened. ” Therefore it is understandable how GW and its associated browning can be confused with true BR containing brown tissue. Even Minges and Sadik (1964) in an earlier report clas- sified typical GW symptoms as a type of BR. Graywall One of the first published reports concerning GW was made by Conover in 1949 (Conover, 1949). He called the disorder vascular browning and noted that this malady was first recorded in the South Florida area in 1927 and had been seen annually in Dade county since 1937. In describing the symptoms of vascular browning, Conover stated 11 vascular browning appears externally on the fruit as a grayishbrown discoloration seen through the somewhat translucent outer wall of the tomato. The margin is usually indistinct and the discolored areas vary from nearly circular to longitudinal sectors in that portion of the fruit wall covering the locules. This discoloration is more noticeable on the sides of the fruit, but has been seen extending from the stem scar to the blossom end. In severely affected fruit, the entire tomato may be discolored and the skin somewhat corrugated but not broken. In cross section the affected area is dark reddish brown and is centered in and around the vascular bundles of the fruit wall . . . the septa and placental tissue are not involved. Fruit of all sizes larger than one-half inch may be affected, pg 283-284. Conover also noted that vascular browning may show up at any picking and vines bearing affected fruit may subsequently set fruit that mature normally without vascular browning. The disorder was usually more severe where vines were luxuriant and the fruit heavily shaded. Young (1946) reported on a similar occurring tomato disorder in Texas called internal browning and core rot. This disorder was correlated with periods of abundant rain. A disorder almost identical to that described by Conover was reported in New Jersey as internal browning (Haenseler, 1949). Symptoms first appeared on 1/4 to 3/4 grown fruit. The most conspicuous symptom was the brown tissue inside the fruit, and this was most prominent near the stem end. The corky stem scar was often more prominent on infected fruits and the septa were severely browned in advanced cases. No consistent foliage symptoms were associated with internal browning and plants which yielded severely diseased fruits early in the season bore normal fruits thereafter. A tomato disorder characterized by internal browning was also found in tomatoes grown in New York, Pennsylvania, Florida, and New Jersey (Friedman, 1949). 12 Stoner and Hogan (1950), working in South Florida, used the term GW to describe an almost identical disorder to Conover's vascular browning. The local name GW was derived from the appearance of the condition in green and pink fruits. Internal browning appeared as pale to dark gray areas in the outer fruit wall. As the fruit ripened the gray areas became brownish and were slightly depressed and roughened in the fully ripe fruits. Disintegration and browning of one or many vascular bundles was observed and under severe conditions the core and placenta were discolored. The browning was observed to stop at the stem scar or extend into the pedicel. Based on mechanical inoculations, grafting, and sexual propagation experiments, the authors concluded that GW or internal browning of tomatoes in Florida was not due to a transmissible pathogen. There was no regular pattern in the location on the plant of the affected fruits and there was no concomitant foliage symptom. In California a condition called GW or thinwall was described by Lorenz and Knott (1942) and was shown to be due to a combination of intense light and heat. The symptoms described for thinwall indicate it is a form of sunburn and entirely different from Florida GW. No mention of internal browning was made. Studies on the influence of certain environmental factors assoc- iated with vascular browning or GW in Florida revealed that shade , mist, cool night temperatures, and to some extent soil compaction all contributed to the incidence of this disorder (Dennison and Hall, 1954; Hall and Dennison, 1955). Affected fruits were usually found on vigorous plants beneath the foliage where they were shaded. Graywall 13 was frequently found following consecutive days of rain and cloudy weather. The largest amount of GW occurred during the cool seasons of the year. Cloud was the term used to describe a form of abnormal ripening of tomatoes grown in unheated greenhouses in New Zealand (Kidson and Stanton, 1953a). The authors description of the symptoms of cloud closely resemble those used to describe GW. They reported that fruit affected with cloud had brownish-green areas against the normal red areas and the necrotic browning affected the vascular system and sur- rounding parenchyma cells. Rapidly growing, heavy yielding plants were very susceptible. Cloud was less severe in heated houses and was also noted to affect outdoor fruit in a wet season. In the green- house, treatments which increased vegetative growth and fruit shading all increased the incidence of cloud. The opposite extreme of heavy defoliation also increased cloud incidence. Cloud was decreased by heavy potash fertilization during the winter, and light watering, N, potash, or CaCl2 treatments throughout the season. The beneficial treatments produced a thinner-stemmed, less rank type of growth. No consistent differences in mineral composition between cloud fruit and healthy fruit were found (Kidson and Stanton, 1953b). No correlation was found between mineral content of the foliage and cloud incidence. Cloud was associated with a low dry matter content of the leaves and fruit. A hypothesis was proposed in which cloud was the result of abnormally low organic matter content in the fruit which might have been caused by excessive water uptake and/or reduced photosynthesis under less illumination. High K levels in the soil reduced cloud and imbalances between K and other elements increased cloud incidence 14 (Kidson, 1963). No conclusive evidence was obtained on the relation- ship between trace elements and cloud. The browning near the vascular system characteristic of cloud was polyphenoloxidase browning (Kidson, 1958). A number of workers have studied the relationship of TMV to internal browning (IBr). The symptoms of IBr and GW are identical and therefore the two terms are probably synonymous. Most workers study- ing this relationship between TMV and abnormal fruit ripening have only used the term IBr. The identical symptoms between GW and IBr are better appreciated when one compares the following IBr description to that previously presented for GW. Wharton (1956) and Wharton and Boyle (1957) stated that IBr was usually confined to the outer fleshy pericarp, but in severe cases many extend into the septa and central column. The large brown to brownish-gray areas radiated from the stem end and were restricted to the fleshy parenchyma tissue. These areas were most prominent near the fruit shoulder and no sharp line of demarcation existed between discolored and normal green areas. Both immature green and mature green fruit were affected. Symptoms on the red ripe fruits were yellow to brownish streaks with associated sunken areas near the shoulder. Cellular collapse of the large thin-walled parenchyma cells resulted in a furrowing of the fruit surface. The vascular tissue did not appear affected. Holmes (1949, 1950) was the first to report on the association of IBr with strains of TMV in the affected fruits. TMV resistant plants did not develop IBr. Holmes was unable to reproduce characteristic IBr symptoms in TMV innoculated plants. 15 Internal browning was reported in Maryland (Cox and Weaver, 1950) and was associated with virus containing plants. Raychaudhuri (1952) was unable to reproduce IBr in tomato fruits by grafting diseased scions onto healthy stocks, even though mottling of the foliage re- sulted. The leaf sap of IBr affected plants contained strains of virus slightly different from the ordinary TMV strain. The relation- ship between virus and IBr was further questioned when Raychaudhuri (1953) reported that TMV free plants produced typical IBr fruits. He noted that excess water applied to the fruit resulted in IBr of some virus infested plants and commercial growers also reported IBr to be associated with high soil moisture content. Further evidence concerning the role of virus in inducing IBr was provided by Boyle (1956). Results from field and greenhouse experi- ments revealed that: 1) plants inoculated with certain strains of TMV did not produce IBr; 2) cuttings from plants that had produced IBr fruits once do not do so again; 3) strains of TMV originating from plants that have produced IBr fruits caused IBr when introduced into healthy plants as the fruits begin to ripen. The expression of IBr was thought to be a "shock" reaction coming as a result of virus in- vasion and a hypersensitive response of the host. Boyle and Wharton (1956, 1957b) consistently reproduced IBr by injecting TMV into healthy plants just when the fruits began to ripen. Symptoms were ex- pressed within ten to twenty days when the peduncle of a ripening fruit hand was inoculated hypodermically. Inoculating tomato plants soon after transplanting resulted in foliage symptoms but no fruit IBr. Those who would doubt Boyle and Wharton's hypothesis that IBr is caused by a shock reaction point out that TMV infection approaches 16 100% in tomato fields near the end of the season and TMV presence in IBr fruits may be of no etiological significance. They would also point out that Boyle and Wharton (1957a, 1957b) were unable to produce IBr in all virus inoculated plants and in several cases virus free plants produced IBr fruits. Results from four different TMV isolates showed only 15-38% of the plants to have IBr fruits after hypoderm- ically inoculating into the peduncle of the first hand. Less than 2% of the fruit from these inoculated plants showed IBr. Even the most potent TMV isolate was only able to produce 14% IBr from over 5000 fruit from inoculated peduncles. Rich (1958) provided further evi- dence against Boyle and Wharton's theory. In one year he found no IBr symptoms in the fruit after inoculation with one of the more potent IBr producing strains of TMV. However, in the next year all virus inoculated treatments had more IBr than non-inoculated treatments. High K fertility reduced the amount of IBr in inoculated plants, but not the to the level of uninoculated plants. Tompkins and Stark (1964) reported that K did not influence IBr, but heavy shade resulted in more IBr. Another theory was proposed in which IBr may result from restric- tions of the movement of sugars into the developing fruit (Taylor, 1956). Fruit from plants grown under high moisture plus mist had a greater percentage of IBr and were 31% lower in reducing sugars com- pared to low moisture treatments. Plants producing few IBr fruit and having a relatively high reducing sugar content were also character- ized by a higher boron (B) content. Inadequate B nutrition was one proposed factor resulting in sugar deficiency. Maynard et al. 17 (1959) supported Taylor's theory that reduced sugar movement into the fruit caused by B deficiency results in more IBr. Boyle and Bergman (1967) also found IBr to be markedly affected by soil moisture and cultivar. Years of low rainfall resulted in much less IBr. Gilbert (1960) reported that vascular browning was more severe during wet weather with cool nights. Windy conditions which resulted in air movement around the plants reduced vascular browning during these wet cloudy periods. The first worker to differentiate GW from IBr was Murakishi (1960a, 1960b). He stated that it was not possible to distinguish be- tween GW and IBr symptoms and both disorders occur simultaneously in the same field. Differentiation was based on whether the affected fruit came from TMV infected plants (IBr) or TMV free plants (GW). This was probably an erroneous separation of the same disease. The severity of both GW and IBr in several cultivars was greater under low light intensity and several TMV-resistant breeding lines were resis- tant to both GW and IBr. However, two GW resistant varieties were IBr susceptible. The most recent hypothesis concerning the cause of GW involves a pathogenic bacterial association in the fruit. Beraha and Smith (1964) reported a dry, firm decay of mature green tomatoes to result after bacterial inoculations into the pericarp. Symptoms were typical of GW. Hall and Stall (1967) observed that a GW-like condition on chilled, detached tomato fruits was associated with a bacterial ooze in the stem scar. A pure culture of a bacterium similar to Erwinia ananus Serrano was isolated from the ooze and from brown GW tissue. Injection of the bacterium into the outer pericarp of green fruits 18 before or after chilling at 3°C induced GW-like symptoms. The severity of browning increased with the length of chilling. Thompson et al. (1960) and more recent reports by Stall and Hall (1968, 1969) also noted that chilling predisposes the fruit to GW. The external symptoms produced by the bacterial inoculations consisted of a brown or grayish-brown discoloration of the outer wall generally limited to the inoculation area. The discolored areas remained firm as the fruit ripened and the symptoms closely resembled those described by Conover (1949). Differential cultivar susceptibility to GW was also noted. Stall and Hall (1967) reproduced GW symptoms by injecting crude extracts from GW affected tissue into healthy fruits. Treatments that removed or killed the bacteria in these extracts also resulted in a loss of the GW producing principle. It was unlikely that the symptoms produced were the result of TMV contamination because symptoms were not produced with bacteriologically filtered extracts, which would not have removed virus particles. Toxic metabolites produced in and extracted from GW affected tissue were probably not important in GW development since filtration would probably not have removed them. Graywall was not specific to one bacteria (Stall and Hall, 1968, 1969). Thirty bacterial isolates from GW-diseased fruits incited GW symptoms after inoculation into susceptible fruits. Most of the iso- lates were species of Bacillus , Erwinia , or Aerobacter. An associa- tion was established between numbers of bacteria in the tomato tissue and visible browning. Bacteria occurred in healthy fruit as well as diseased, but were found in higher numbers in diseased fruit. The regular occurrence of bacteria in healthy green and red tomato fruit was previously reported (Samish et al. 1961, 1963). Stall and 19 Hall concluded that GW expression after bacterial multiplication was associated with hypersensitivity of the tomato tissue. Endophytic bacteria incited GW, but the possibility existed that GW could occur in the absence of bacteria since constant association of bacteria with natural GW was not established. Studying the interaction between GW and TMV, Stall et al. (1969) reported that differences in GW incidence could not be attributed to presence of TMV. No TMV was found in GW affected fruit from plants not inoculated with virus, but TMV was detected in fruits from virus inoculated plants. Graywall occurred on TMV susceptible as well as TMV resistant plants. They stated the possibility exists that bacteria, like TMV, are a predisposing factor and not the primary causal agent of GW. Two types of GW resistance mechanisms were postulated (Hall et al. 1970). One would be resistance to entrance of the bacteria and the other would be some factor in the tissue which prevents bacterial development after invasion of the tissue. Using a vacuum infiltration technique, suspensions of bacteria were placed around the stem scar with the stem attached and were infiltrated into the fruit. Typical GW symptoms resulted. Previously reported field resistant cultivars developed less GW than susceptible cultivars. Resistance may have been due to some structural characteristic of the stem scar area. If the stem was removed before infiltration, no cultivar difference was found. The cultivars which showed GW resistance with the infiltration technique had narrower corky rings and smoother shoulders with a general absence of deep creases. The injection technique of hypoderm- ically inserting a bacterial suspension into the pericarp gave 20 different results compared to the infiltration method. The cultivar most susceptible to GW in the infiltration method (' Homestead-24’ ) was less susceptible to GW with the injection method than many naturally GW resistant lines. Ideally a combination of resistance to entrance and tissue resistance is preferred. It was presumed that environmental changes or virus infection could reduce tissue resistance. The bacteria occurring in healthy tomato fruit probably enter through the connective tissue at the stem end. They can persist in a latent state because of the inhibiting action of healthy cells, but these harmless commensals may be activated by some physiological dis- turbance in the tissue. Samish et al. (1961, 1963) reported that bacterial populations within tomato fruit showed a distinct gradient, being largest in the connective tissue at the stem end and decreasing through the central core towards the peripheral and distal tissue. Such a gradient agrees with the location of GW symptoms in the fruit. Samish et al. 1963 also found that fruits harvested from fields using overhead irrigation contained far more bacteria than those harvested from furrow irrigated fields. The greater incidence of GW on fruits harvested from the interior of the canopy or from moist environments supports a bacterial related cause of this disorder. In order for GW to develop, the bacteria must enter the fruit and multiply to a high enough level. The entrance and growth are dependent on the interaction of physiological and biochemical changes in the fruit. In summary, basically four causes of GW (IBr) have been proposed and are due to the following agents: 1) TMV, 2) environment and/or 21 nutrition, 3) bacteria, 4) a combination and interaction of two or more of the above. The causes of both disorders, BR and GW, are elusive and not truly known. Both disorders may occur under varied conditions and may be caused by a complex of factors rather than a single one. Nutritional, environmental, and pathogenic factors may all have their influence and whether they act in primary or secondary roles is not completely understood. Symptoms produced in both disorders may be due to some toxic substance entering the fruit. The marked variability in amount of BR and GW produced within plants, between plants, between cultivars, and between environments implies that there is no simple cause and effect solution. Fruit Quality of Blotchy Versus Normal Fruit — Compostional and Biochemical Differences Several workers have determined compositional differences between blotchy fruit (or blotchy tissue within a red fruit) and normal red fruit. Compared to red tissue, blotchy tissue was found to contain significantly lower amounts of total solids, dry matter, titratable acidity, soluble solids, and reducing sugars (Winsor and Massey, 1958; Ells, 1961; Winsor et.al. 1962; Tompkins and Stark, 1964; Davies, 1966). Winsor and Massey (1958) concluded that there was no evidence for a critical level of any of the constituents below which the fruit would always ripen unevenly. Malic acid was higher and citric acid lower in blotchy green areas compared to corresponding red areas (Davies, 1966). The malic 22 acid to citric acid ratio was about double in the green than red areas. Normal tissue was found to have a lower pH than blotchy tissue (Winsor and Massey, 1958). Several workers found the K content of blotchy fruit walls to be lower than that of normal walls (Winsor and Massey, 1958; Van der Boon, 1973; Hobson and Davies, 1976). Others found no significant difference in K or other mineral 'concentrations in blotchy versus red tissue (Davies, 1966; Ells, 1961). A higher K content in internally browned tissue compared to normal tissue was reported by Tompkins and Stark (1964). Biochemical differences exist between blotchy and normal red tissue. The normal reduction in polyphenoloxidase activity that accompanies uniform ripening does not occur in blotchy tissue (Tompkins and Stark, 1964; Davies, 1966; Hobson, 1967). Polygalacturonase activity was much lower in blotchy tissue than red tissue (Hobson, 1964). Protein components and isoenzyme differences between blotchy and normal tissue were of a random nature (Hobson and Davies, 1976). Blotchy tissue had about double the content of pectic substances compared to red tissue (Hobson, 1963), but failure of the blotchy areas to ripen was not thought to be primarily due to reduced pectinesterase activity. Davies (1966) stated that blotchy areas should be considered abnormal and not merely delayed in ripening. MATERIALS AND METHODS Blotchy Ripening — Greenhouse A sand culture experiment was conducted in the greenhouse during the spring-summer of 1978 to determine the influence of cultivar and K treatments on BR. Three tomato cultivars, 'Healani', 'Homestead- 24', and 'Flora-Dade' were seeded on April 28 in thirty cm diameter plastic pots filled with sterilized #2 grade sand. Johnson's (Hewitt, 1966) solution (Table 1) was applied daily until first cluster anthe- sis (mid-June) whereafter four K treatments were applied daily until the conclusion of the experiment. The four K treatments were: 1) IK = Johnson's solution. 2) 1/10K = Johnson's solution for all elements except K which was 1/10 the normal level. 3) -K = Johnson's solution for all elements except K which was omitted. 4) H2O = Water containing four milliequivalents per liter of CaC^* The IK solution was applied to treatments 3) and 4) for seven day periods in early July and early August to maintain plant growth and fruit set. In the 1/10K treatment, NaN03 was added to maintain the NO^ concentration equal to the IK treatment. In the -K treatment, KNO3 was replaced with NaN03 to maintain the NO3 concentration equal to the IK treatment. 23 24 Table 1. Standard Johnson's solution (IK). Macroelements 1 Molar solution Ml of 1 Molar solution needed per liter of total solution KN03 6 Ca(N03) 2* 4 H20 4 NH4H2P04 2 MgS04‘7 H20 1 Microelements g per liter KC1 h3bo ZnS04*7 H20 MnS04*H20 CuS04* 5 H20 H2Mo04 Iron FeCl3*6 H20 Na2H2EDTA 3.728 (One ml of all micro- 1.546 elements needed ! per 0.575 liter of total solu- 0.338 tion) 0.125 0.081 g_ per liter 8.71 (One ml of iron needed 11.99 per liter of total so- lution) Element Final concentration of Final concentration nutrient solution (uM) in ppm N K Ca P S Mg Cl Fe B Mn Zn Cu Mo 16000 6000 4000 2000 1000 1000 50 32.2 25 2 2 0.5 0.5 224 235 160 62 32 24 1.77 1.80 .27 .11 .131 .032 .05 25 A completely randomized block design was used with four re- plications per treatment making a total of forty-eight plants (four replications x four K treatments x three cultivars). Twelve fruit were picked from the first three clusters of each plant, so data analyses were based on an average of forty-eight fruit per treatment (twelve fruit per treatment x four replications). The plants, pruned and trained to a single stem, were grown under approximately a 21°C night temperature and ambient day temperature regime. Day temperatures generally ranged between 32°C-36°C. Plants were irrigated to saturation each day with their respective K treatments to prevent both salt accumulation and wilting. Fruits were harvested at the red ripe stage over a four week period from late July through August and rated for external blotchy ripening (EB) and internal blotchy ripening (IB) symptoms. To deter- mine IB, each fruit was cut cross-sectionally in half. The rating scale used in assessing the amount of BR ranged from 1 to 5. Each number was associated with the following amount of BR, in which the percentages relate to the amount of either external surface area or internal pericarp surface area affected with BR: 1 = completely blotch free 2 = 0-5% blotch 3 = 5-25% blotch 4 = 25-50% blotch 5 = greater than 50% blotch Pictorial illustrations of EB and IB are provided in Figures I and 2. 26 Figure 1. Illustration of fruit with an external blotchy ripening rating of 1, 3, and 5. Figure 2. Illustration of fruit with an internal blotchy ripening rating of 1 (bottom) and 5 (top) . 27 Field Experiments Field experiments were conducted during the spring and fall 1979 growing seasons to determine the influence of K rate, cultivar, and season on BR, GW, and fruit quality. Tomatoes were grown at the Hort- icultural Unit in Gainesville on a Kanapaha fine sandy soil very low in organic matter. Spring and fall plots were planted on adjacent areas in a field that had been fallow for several years, thereby ensuring extremely low residual fertility. Soil K analyses measured before planting by acid extraction revealed residual K fertility to be between 30-40 ppm. Soil pH was 6.5. The plots were disked about five weeks before plant and fumigated with ethylene dibroraide three weeks before planting. Fertilizer N and P were broadcast at the commercially recommended rates of 270 kg N/ha and 118 kg P/ha. Ammonium nitrate and triple superphosphate were the sources of N and P. Raised beds were then prepared 240 cm apart to be sure that there was no cross feeding. Potassium sulfate was placed in two bands about 15-20cm from the center of the bed. Five different rates of K were used: 0, 93, 186, 372, and 745 kg K/ha (hereafter referred to as 0-K, 93-K, 186-K, 372-K and 745-K) . Black polyethylene mulch was applied over the top of the beds. The tomatoes were transplanted 45cm apart. The plants were pruned twice early in the growing season and then were staked and tied. Four tomato cultivars, 'Healani', ' Homestead-24* , 'Walter' and 'Flora-Dade' were grown at the five K rates during each season in a completely randomized block design. There were four replications of eight plants each. Fruits were harvested from the six interior plants. The two end plants served as guard plants. Tomatoes were 28 transplanted on March 30 for the spring season and on August 8 for the fall season. Commercially recommended fungicides and insecticides were applied twice a week. Supplemental overhead irrigation was ap- plied as needed to keep the soil moisture as near to field capacity as possible. Blotchy Ripening Three harvests were made during each season on the following dates: spring — June 10, June 17, and June 20; and, fall - October 25, November 5, and November 15. At each harvest all red ripe fruits were harvested from the six interior plants of each replication. Ten randomly selected fruits from each harvest lot were rated for EB and IB. A 1 to 5 rating scale, as previously described, was used in as- sessing the amount of each. Graywall Thirty-two randomly selected immature green fruits (6x6 size) were harvested from the six interior plants of each treatment. Four separate harvests (eight fruit per harvest) were made over a two week period in early June for the spring and over a two week period in late October for the fall season. The original objective was to determine the influence of K rate and cultivar on naturally occurring GW under field conditions. No natural GW appeared during either season, so it was decided to study the influence of K rate and cultivar on GW induced by injection of bacteria. The procedure of Hall and Stall (1967) was followed with minor modifications. Cells of the bacteria Erwinia herbicola (Dye) were suspended in sterile saline (adjusted to cellular osmotic concentration) and 29 inoculated into the outer fruit pericarp with an automatic syringe fitted with a 27-guage needle. Approximately 0.1 ml of inoculum con- taining 10^ cells per ml was injected into five different locations around the equator of each fruit. After inoculation, fruit were kept at 4°C for four to six days, returned to 20°C for another four to six days, and then rated for symptoms. The four to six days at 4°C served as a chilling treatment that insured the fruit to be in a GW susceptible condition. Fruits were also inoculated with 0.1 ml of sterile saline to act as controls. Bacterial suspensions were prepared by allowing the bacteria to multiply in nutrient broth and then the cells were centrifuged from the broth and resuspended in sterile saline. Inocula were standard- ized to 50% transmittance as determined by a Spectronic 20 at 600nm. Q A suspension of this turbidity contained approximately 10 viable cells per ml as determined by a dilution plate technique. The final concentration of 10^ cells per ml was obtained by dilution. The rating scale used in assessing the amount of GW development ranged from 0 to 5. Each of the five individual inoculation sites per fruit was rated separately and results are expressed as the average of these five GW ratings. Each treatment consisted of eight fruit x four replications. Each number in the individual inoculation site rating scale was associated with the following amount of GW development as seen through the outer fruit wall: 0 = no discoloration 1 = slight tissue discoloration 2 = light browning 30 3 = moderate browning 4 = deep browning 5 = extreme deep browning A pictorial illustration of different amounts of GW development is provided in Figure 3. Fruit Quality Measurements Fruit quality analyses were conducted on a total of forty fruit per treatment. Ten fruit from four different replications were ran- domly selected. For the spring analyses, all the fruit were picked in the red ripe stage on June 20 and for the fall fruits were picked in the red ripe stage on November 5. Locule plus placenta tissue was separated from the pericarp tissue and both were frozen at -20°C until analyzed. The following quality analyses were performed for both per- icarp and locule tissues: % soluble solids, pH, and % acidity. In addition, % reducing sugars, % dry weight, and mineral composition of the pericarps were analyzed. Samples were thawed and thoroughly homogenized at high speed in a Waring Blendor for one minute. A portion of the homogenate was cen- trifuged at ten thousand rpm for fifteen minutes in a Sorvall RC2-B centrifuge. Titratable acidity was determined by titrating ten ml of the supernatant and one hundred forty ml of pre-boiled CC>2 free dis- tilled water to pH 8.1 with 0.1N NaOH. Results are expressed as % citric acid [titratable acidity (in ml) x 0.064]. The pH was deter- mined from the ten ml supernatant-one hundred forty distilled water mixture with a Beckman Model 3500 digital pH meter. Percent soluble solids were determined from several drops of the supernatant with a Bausch and Lombe table top ref ractoraeter . 31 Figure 3. Illustration of fruit with a graywall rating of 1 (lower left), 2 (upper left), 3 (upper right), and 4 (lower right). 32 Percent dry weight was determined by drying fifty g of the homo- genate at 70°C for forty-eight hours. Percent reducing sugars were determined as follows: twenty-five g of the homogenate was blended with two hundred mis of 95% ethanol for one minute. The slurry was rinsed into a mason jar with 95% eth- anol and placed in a boiling water bath for thirty minutes. The sam- ples were then filtered through Whatman # four paper under suction us- ing a Buchner funnel. The final volume was adjusted to five hundred ml with 95% ethanol. A twenty ml aliquot was pipetted into a one hun- dred ml volumetric flask and brought to volume with distilled water. Reducing sugars were determined on a one ml aliquot of this dilution after the procedure of Ting (1956). Sugar concentrations were deter- mined on a Bausch and Lombe Spectronic 100 spectrophotometer set at 745nm. Pericarp K, Ca, Mg, and P concentrations were determined as fol- lows: oven-dried pericarp tissue was ground in a Wiley Mill to pass a #forty sieve and stored in air-tight bottles. One g of ground tissue was ashed in a muffle furnace at 475°C for eight hours. The ash was dissolved in IN HC1. Potassium was determined by flame photometry, Ca and Mg by atomic absorption spectrophotometry, and P by colorimetric spectrometry at the Soil Science Laboratories, University of Florida. Leaf mineral analyses were determined in the same manner. Several recently matured leaves and petioles near the top of each plant were sampled at the end of each growing season; late June for the spring and late November for the fall season. Results are based on four replications per treatment. By the end of both growing 33 seasons the plants in which K was not applied to were very chlorotic relative to plants grown at the higher K rates. Environmental conditions were different between seasons as indi- cated in Table 2. As the fruits were maturing during the spring, average minimum and maximum temperatures were increasing, daylengths were getting longer, and relative humidity was increasing. During the fall, average minimum and maximum temperatures were decreasing, day- lengths were getting shorter, and relative humidity was decreasing. Graywall — greenhouse Several additional greenhouse experiments were conducted to determine the influence of cultivar and K nutrition on GW develop- ment. The only experiment in which natural GW was studied, in con- trast to bacterially induced GW, was during the fall-winter of 1978- 79. Tomatoes were grown in thirty cm diameter plastic pots filled with sterilized #two grade sand. Three tomato cultivars, 'Healani', 'Homestead 24', and ' Flora-Dade' , were seeded on September 20 and watered daily with Johnson's solution (Table 1) until first cluster anthesis, which occurred in late November. After this, four different K treatments , as previously described for the greenhouse-BR experi- ment, were given daily until conclusion of the experiment. The four K treatments were IK, 1/10K, -K, and H2O. Two-seven day periods of the IK treatment were applied during mid-December and mid-January to the -K and ^0 treatments. All the fruit of marketable size on each plant were analyzed for natural GW symptoms and classified as either having GW or not having GW. No distinction was made between the severity of symptoms. Each treatment consisted of at least fifty to sixty fruit (four replications x thirteen to fifteen fruit per plant). Forty- eight total plants were grown in a completely randomized block design 34 Table 2. Month April May June September October November Monthly temperature averages during the 1979 Spring and Fall growing seasons at the Horticultural Unit, Gainesville, Florida. Temperature (°C) Season Maximum °C Minimum °C Spring 28.3 12.2 31.1 14.4 32.8 17.2 Fall 32.8 18.3 27.8 12.2 24.4 10.0 35 (four replications x four K treatments x three cultivars). Plants were grown under approximately 15°-21°C night temperatures and 26°C- 30° C day temperatures with the exception of a one week period in early January when the night temperatures were just above freezing (i.e. 2°C-5°C). This was due to an unusual cold spell accompanied by a failure of the greenhouse heating system. Another greenhouse experiment was conducted during the fall of 1979 to complement the fall 1979 GW field studies. Plants of four cultivars, 'Healani', 'Homestead-24', 'Walter', and ' Flora-Dade' ,were grown as above until first cluster anthesis when two K treatments were applied. The high K treatment consisted of daily applications of Johnson's solution (IK). The low K treatment consisted of daily ap- plications of Johnson's solution in which K was omitted. All other nutrients were applied at the same rate as in the high K treatment. For each cultivar a total of seventy-two fruit per treatment (high versus low K) were compared for bacterial induced GW development. The materials and methods for inoculating, storing, and rating the fruit were the same as for the field studies. A total of forty-eight plants were grown (six replications x two K treatments x four cultivars) in a completely randomized block design. Night temperatures were approxi- mately 21°C and day temperatures usually varied between 280C-32°C. Ethylene Measurements Ethylene evolution for yellow and red tissue disks taken from the same fruit was determined. Outer pericarp disks 1.0 cm in diameter were taken with a cork borer from field grown fruits exhibiting yellow shoulder. Each pair of yellow and red disks was taken from the same fruit at a similar longitudinal position. The disks were blotted on 36 tissue paper to remove any adhering cell sap, weighed, and placed in 9.0 ml volume glass vials capped with silicone stoppers at 25°C. After one hour, 0.5 ml gas samples were withdrawn from the vials with one ml gas-tight disposable syringes. Samples were injected into a Hewlett-Packard 5710A flame-ionization gas chromatograph fitted with a 1.83m x 0.32cm Cu tubing column packed with activated alumina. The flow rate of N carrier gas was forty ml/min and operating temperatures were 150°C, 100°C, and 150°C for the injection port, oven, and detec- tor respectively. The chromatograph was connected to a Soltec re- corder set at attenuation one and run at a chart speed of thirty cm/hr. A 0.1 ppm ethylene standard was regularly injected as a refer- ence gas for calibration. Differences in ethylene evolution (ul/kg/hr) between yellow and red disks were tested by the Student's t-test for paired samples at the 1% level. Oxygen Consumption Respiration rates for yellow and red tissue slices taken from the same fruit were determined by O2 consumption measurements using a Gilson Single Valve Differential Respirometer. Outer pericarp slices, approximately one g fresh weight, were taken with a scalpel from field grown Flora-Dade tomato fruits exhibiting yellow shoulder. Each pair of yellow and red tissues was taken from the same fruit at a similar longitudinal position. The tissue slices were blotted on tissue paper, weighed, and placed in respiratory flasks. The flask center well contained 0.4 ml of 10% NaOH to absorb CO2 during 0£ uptake mea- surements. The flasks and contents were equilibrated for thirty 37 minutes at 25° C with the stopcocks open before respiratory measure- ments were started. Results are expressed on a fresh weight basis. The O2 consumption rates represent the average of thirty paired tissue samples each taken from a different fruit. Differences in O2 consump- tion (ul/g/hr) were tested by the Student's t-test for paired samples at the 1% level. Satistical Analyses Analyses of variance, regression, and t-tests for the experimen- tal data were performed with the aid of the Northeast Regional Data Center in Gainesville. Appropriate square root transformations were made on rating data (i.e. EB, IB, GW) and are sin transformations were made on % data (i.e. soluble solids, acidity, reducing sugars, mineral content, dry weight). Statistics used in the data tables were based on the transformed analyses (Snedecor and Cochran, 1967). In the ana- lyses of variance tables (see Appendix) the cubic and quartic potas- sium sources of variation and their interactions were pooled into the error terra because in most cases they were not significant. RESULTS Influence of Potassium and Cultivar on Blotchy Ripening of Greenhouse Grown Tomatoes Fruit of each cultivar showed a significant increase in both EB and IB with the three deficient K treatments (1/10K, -K, 1^0) compared to the optimum IK treatment (Tables 3,4). There was no significant difference between the EB or IB ratings with the three deficient K treatments within any cultivar. With the 1/10K, -K, and H2O treatments, no difference in EB and IB was found between' Healani' and 'Homestead-24'. 'Flora-Dade' had more EB and IB with these deficient treatments than the other two cultivars. At the IK level, 'Healani' had the least amount of EB and IB and 'Flora-Dade' the most. Yellow shoulder was the primary form of EB affecting 'Flora-Dade' with the reduced K treatments, and to a limited extent with the IK treatment. Yellow shoulder was not observed in 'Healani' or 'Homestead-24'. There was no noticeable correlation between fruit • % size and BR or fruit cluster and BR in any cultivar. Blotchy ripening ratings were therefore averaged over all three clusters and expressed on a per plant basis. Influence of Potassium, Cultivar, and Season on External Blotchy Ripening of Field Grown Tomatoes There was no noticeable effect of time of harvest on EB during either season with any cultivar (Tables 5,6). Fruits were harvested 38 39 Table 3. Effect of reduced potassium levels after first cluster anthesis on external blotch of three tomato cultivars grown in greenhouse sand cultures. Spring, 1978. Cultivaryx IK K 1/10K treatment2 -K h2o 'Healani' 1. 9e 3. Obc 3. Obc 2.8c 'Homestead-24' 2. Id 2.9c 2.9c 2.9c 'Flora-Dade' 3.2b 4.4a 4.2a 4.3a 2 Mean separation by Duncan ' s multiple range test at the 5% level. ^Ratings were 1 to 5 with 1 being blotch free, treatments were applied after first cluster anthesis. 40 Table 4. Effect of reduced potassium levels after first cluster anthesis on internal blotch of three tomato cultivars grown in greenhouse sand cultures. Spring, 1978. Cultivaryx IK K 1/10K z treatment -K h2o 'Healani'