UNIVERSITY OF CALIFORNIA COLLEGE OF AGRICULTURE AGRICULTURAL EXPERIMENT STATION BERKELEY, CALIFORNIA THE COMPOSITION OF CANNING TOMATOES L. G. SAYWELL AND W. V. CRUESS BULLETIN 545 December, 1932 UNIVERSITY OF CALIFORNIA PRINTING OFFICE BERKELEY, CALIFORNIA CONTENTS PAGE Introduction 3 Review of the literature 3 Methods of analysis 5 Effect of locality on composition of tomatoes 8 Effect of locality on total solids and acidity of commercially grown tomatoes 8 Effect of locality on the composition of tomatoes from a single parent 13 Effect of time of picking on total solids and total acid 14 Annual variation in total solids and total acid 15 Effect of water supply on total solids and total acid 16 Effect of variety on total solids and total acid 17 Total solids and total-acid content of tomatoes from different plants 20 Total solids of fruit from various locations on the plant 21 Total-solids content of different portions of the fruit 21 Protein content of tomatoes 22 Sucrose, reducing sugars, and dextrose-levulose ratio 24 Starch content of tomatoes '. 27 Acid-hydrolyzable material 27 Pectin and protopectin content of tomatoes 27 "Insoluble solids by difference" 29 pH value of juice of tomatoes 30 Summary 30 Acknowledgments 31 Literature cited 31 The Composition of Canning Tomatoes 1 L. G. SAYWELL2 and W. V. CRUESSs INTRODUCTION The yield and quality of tomato products depend in great measure upon the composition of the raw material. Nevertheless, tomatoes are purchased by canners and tomato-products manufacturers without re- gard to this fact. The price paid for sugar beets, on the other hand, usually varies according to their sugar content. The investigation re- ported upon in this publication was undertaken in order to determine whether the variation in composition of canning tomatoes is sufficient to warrant differences in price in accordance with the concentrations of the desirable constituents, and to ascertain to what extent locality, variety, seed selection, water supply, date of picking, and other factors might affect these constituents. Owing to the present rapidly increasing production of canned tomato juice, information on the concentration of the more important constituents of tomatoes is particularly desirable at this time. REVIEW OF THE LITERATURE Atwater and Bryant (1906) 4 published proximate analyses of a num- ber of vegetables, including analyses of tomatoes, but gave no compara- tive results and did not give the source of the samples reported upon. In a recent publication Chatfield and Adams (1931) have given the follow- ing average, maximum, and minimum values for several of the con- stituents of tomatoes : sugar, 3.37, 4.06, and 2.30 per cent ; starch, less than 0.1 per cent ; protein, 1.00, 1.80, and 0.70 per cent ; and total solids, 5.90, 9.4, and 3.3 per cent. Meyers and Croll (1921) reported on the carbohydrate content of two samples of tomatoes. Bell, Long, and Hill (1926), in studies on human metabolism, found the available carbohydrates (soluble sugars) of the tomatoes used in their studies to range from 2.6 to 3.0 per cent. Law- rence and McCance (1929) found an average of 2.4 per cent soluble sugars in tomatoes. i Received for publication May 12, 1932. 2 Research Assistant in Fruit Products. s Associate Professor of Fruit Products and Chemist in the Experiment Station. 4 See "Literature Cited," pages 31 to 32, for complete data on citations, which are referred to in the text by author and date of publication. 4 University op California — Experiment Station All of the investigations cited above were made by those chiefly inter- ested in the nutritional value of tomatoes. The following citations are of papers upon the physiology of the tomato plant, and upon the compo- sition of the fruit in relation to its value for canning or manufacturing purposes. Borntrager (1925), in studies of changes occurring during ripening, reported that citric and malic acids are the only organic acids present in the ripe fruit. Rosa (1925, 1926), in studies made at the University Farm, Davis, California, upon the changes occurring during ripening of several varieties of tomatoes used for the fresh market, found the reducing sugar of mature, sound tomatoes of the Earliana, Globe, and Stone varieties to range from 2.63 to 3.80 per cent, sucrose less than 0.10 per cent, starch 0.012 to 0.052 per cent and total acid-hydrolyzable material, 0.180 to 0.212 per cent. That is, invert sugar made up most of the carbohydrate substance. Benoy and Webster (1930) found no sucrose and stated that all sugar present in tomatoes is invert. They gave the average sugar content as 2.43 per cent and starch as 0.069 per cent. Bigelow and Fitzgerald (1915) made an extensive study of the com- position of tomato pulps, both the raw and the partially concentrated products, from many factories, principally in the eastern and middle western United States. The raw pulp reported upon by them is the mixed unconcentrated juice and screened flesh free of skins and seeds. The average composition of 11 samples reported by them was : total solids, 4.37 per cent ; total acid as citric, 0.31 per cent ; and sugar as invert, 2.19 per cent. Bigelow and Stevenson (1923) elaborated upon the previous work of Bigelow and Fitzgerald and prepared tables showing the relation be- tween specific gravity of filtrate and total-solids content of tomato pulp ; and the relation of refractive index to total solids of tomato catsups. They also devised a formula expressing the relation of refractive index to per cent total solids in tomato pulp. Huelsen (1925) prepared tables giving the total solids and refractive index of tomato pulps of various concentrations made from tomatoes grown in the eastern United States. Cruess and Say well (1928) and Cruess, Say well, and Mark (1929) reported that locality and climate are important factors affecting the composition of tomatoes. Cruess and Say well (1929) have indicated that the relation of refractive index to total solids for California toma- toes is not identical with that given by Huelsen (1925) for eastern tomatoes. Bul. 545] Composition of Canning Tomatoes MacGillivray (1929) has published comparative analyses of different portions of the fruit of three varieties of tomatoes grown in Indiana. In each variety the jelly-like pulp had the highest acidity and the inner locule pulp the highest total-solids content. 30 O "» 15 O 1.3400 1.3440 1.3480 1.3520 1.3560 1.3600 1.3640 1.3660 1.3720 1.37 REFRACTIVE INDEX or PULP Fig. 1. — Graph showing the relation of refractive index to total solids of tomato pulp. METHODS OF ANALYSIS Most of the analyses were made upon the pulp, which consists of the screened whole fruit free of skin, seeds, and coarse fiber, but containing in finely divided form practically all of the red pulpy portions of the fruit. However, in a number of instances the entire fruit was used for analysis. Many of the pulp samples were secured from the pulping ma- chines in canneries. Where whole tomatoes were used in preparing pulp for analysis the fruits were quartered through the polar diameter and two alternate quarters of each fruit were retained. These were then ground coarsely in the laboratory by means of a kitchen-sized food grinder and were rubbed through a monel-metal screen of approxi- mately % 6 -inch mesh. The samples of pulp taken in canneries were preserved by heating in sealed cans. Those prepared in the laboratory were preserved by storage at 0° to 10° F in the frozen condition. Com- parative analyses showed no detectable change in composition by either method of preservation. 6 University of California — Experiment Station During the seasons of 1925 and 1926 the total-solids content was determined by drying 5-gram samples to constant weight in a vacuum oven at 70° C and 28 to 29 inches vacuum, as recommended by the Association of Official Agricultural Chemists (1925). In subsequent seasons many of the values for total solids were determined from the refractive index. Readings were made upon the juice by means of an Abbe refractometer, and the corresponding total solids were obtained from a graph constructed by plotting the refractive indexes (obtained in 1925 and 1926) against the total solids determined by drying at 70° C. In establishing the curve shown in figure 1, which gives the relation of total solids of the pulp to refractive index of the pulp, more than 100 samples of natural and of partly concentrated pulps were examined. Determinations of total solids were made by drying samples in a vacuum oven at 70° C, and the refractive index was determined upon a few drops of the juice at 20° C. Samples from twenty-one different localities, representing all important tomato-producing sections of the state, were used in this standardization. None of the samples contained any added salt or sugar. Trial calculations show that the total-solids content of the pulp of California tomatoes is given with reasonable accuracy by using the following formula : TS = 781 (n D — 1.3332) — 123 (n D — 1.3396) where TS is the total solids in percentage by weight and n D the refractive index at 20° C. If the reading is made at any other temperature than 20° C, add to the observed refractive index 0.0001 for each degree Centi- grade above 20° C and subtract 0.0001 for each degree below 20° C. In the formula, 1.3332 is the refractive index of pure water at 20° C and 1.3396 is the refractive index of tomato pulp containing 5 per cent total solids. At 5 per cent solids the expression 123 (n D — 1.3396) becomes 123 (1.3396 — 1.3396) =0. Below 5 per cent solids the value of this expression is negative and must be added to the expression 781 (n D — 1.3332), i.e., a double negative is a positive. For example, an observed refractive index is 1.3382. TS = 181 (1.3382 — 1.3332)— 123 (1.3382 — 1.3396) =-3.91 — (—0.172) = 3.91 + 0.172 = 4.08. Above 5 per cent total solids this expression is subtracted. Thus the observed refractive index of a pulp is 1.3472. TS = 781 (1.3472 — 1.3332) -- 123 (1.3472 — 1.3396) =10.934 —0.934 = 10.00 per cent total solids. Total acid was determined by titration of 10 cc of the clear nitrate from the pulp with 0.1 normal sodium hydroxide using phenolphthalein indicator. The acidity is expressed as citric acid in grams per 100 cc. Bul. 545] Composition of Canning Tomatoes For the sugar determination a weighed sample of the pulp was diluted to mark in a volumetric flask and clarified by addition of a small amount of finely powdered Home's lead subacetate and filtered; the excess lead was removed by precipitation with powdered potassium oxalate, and a second filtration. Reducing sugars in this filtrate were determined by the Shaffer-Hartmann (1921) iodometric method. Tomatoes are known to contain only a very small proportion of sucrose. The sucrose content of a few samples was determined by the Shaffer-Hartmann method after hydrolysis of the sucrose with dilute hydrochloric acid. However, in order to determine the accuracy of this method, a representative sample of freshly prepared pulp was divided TABLE 1 Comparison of Various Methods of Inversion" of Sucrose in Tomato Pulp Sample Per cent total sugar Per cent reducing sugar No. 10 per cent citric acid 5 cc cone. HC1 per25cci'orl0min. 5 cc cone. HC1 per 25 cc for 7 niin. Invertase 1 hour at 55° C Untreated control 1 2 3 Average 3.792 3.792 3.785 3.790 3.708 3.708 3.708 3.708 3.792 3.780 3.788 3.787 3.792 3.788 3.790 3.790 3.760 3.756 3.764 3.760 into five portions. The first portion was hydrolyzed by heating on a steam bath at 100° C for 10 minutes with 10 per cent of citric acid ; the second was similarly heated with 5 cc of concentrated HC1 per 25 cc aliquot for 10 minutes ; the third portion was heated for 7 minutes with 5 cc of concentrated HC1 per 25 cc aliquot on the steam bath ; the fourth portion was treated with invertase at 55° C for 1 hour, a method that inverts only the sucrose and not hemicelluloses, etc. ; and a fifth portion was used as an untreated control. All hydrolyses were conducted in triplicate. The acidified samples were neutralized. All were clarified as previously described and reducing sugars were determined in the filtrates by the Shaffer-Hartmann method. Table 1 gives the data on the fresh, or wet, basis. It is apparent that 10 minutes heating with HC1 was too severe and resulted in loss of some sugar. Heating with HC1 for 7 minutes, how- ever, gave results agreeing closely with those obtained by inversion with invertase. Therefore, the method of heating 7 minutes with HC1 was adopted, since it is more rapid and convenient than inversion with invertase. 8 University of California — Experiment Station It was thought that some inversion of sucrose by invertase possibly present in the tomatoes might occur during preparation and storage of the samples. Accordingly 0.2 gram of sucrose per 100 cc was added to freshly prepared pulp. Another portion with no sucrose added served as a control. Both were held at 55° C, the optimum temperature for in- vertase action, for 1 hour. Analyses of both samples were made in triplicate. There was no evidence of invertase action and it was con- cluded, therefore, that no detectable inversion would occur during preparation of the samples, storage at a freezing temperature, and subsequent thawing. Acid-hydrolyzable material, which includes starch, was determined on a few samples by hydrolyzing the unfiltered sample for 2% hours at 100° C on a steam bath with 20 per cent by volume of concentrated HC1. Other substances hydrolyzable by HC], such as pentosans, as well as starch are determined by this method. Starch in several samples was determined by hydrolysis with diastase by a modification of the method of Olmsted (1920) . To a 25 cc aliquot was added 0.1 gram of Taka diastase. The sample was held at 55° C for 1 hour. It was then cooled, clarified, and the sugar determined by the Shaffer-Hartmann method (1921). A blank determination was made with 0.1 gram of diastase dissolved in 25 cc of water. Nitrogen was determined on weighed portions by the Kjeldahl method, using an oxidizing mixture of H 2 S0 4 , CuS0 4 , and K 2 S0 4 ; protein was taken as 6.25 times the weight or percentage of nitrogen. EFFECT OF LOCALITY ON COMPOSITION OF TOMATOES In the investigation on the effect of locality on composition, the total solids and total-acid content of samples taken from the various impor- tant tomato-growing sections of the state during the five seasons, 1925, 1926, 1927, 1928, and 1929 were determined. In order to minimize the effect due to seed variation, plants of the same stock were grown at Davis and Berkeley and the composition of the fruit compared. Effect of Locality on Total Solids and Acidity of Commercially Grown Tomatoes. — Since only total solids and total acid were determined on the majority of the commercially grown samples these data are con- sidered separately from the data on sugar content, protein, starch, and pH value presented later in this report. For greater ease of comparison, the values for total solids and total acid given in table 2 are presented as averages, maximums, and mini- mums only. From 4 to 30 samples were used in calculating most of the averages ; in three cases only one sample was analyzed. Bul. 545] Composition of Canning Tomatoes table 2 Total Solids and Total Acid of Tomatoes from Different Localities Variety Locality Year Num- ber of samples Total solids, per cent Aver- Maxi- mum Mini- mum Total acid, grams per 100 cc Aver- age Maxi- mum Mini- mum San Jose Canner . Stone Norton.. Norton Selection Norton.. Alameda Count 1 Sacramento • Alameda County. San Fernando Valley Davis SantaClaraCounty San Mateo County Riverside / Fullerton San Fernando Valley El Monte. 1925 1927 1928 1929 1925 1926 1927 1928 1929 1925 1927 1928 1929 1925 1926 1927 1928 1929 1925 1927 1928 1929 1925 1925 1925 1927 1925 1925 1927 1926 1926 1929 1926 1927 1928 1929 1925 1926 1927 1928 7.02 6.61 6.71 6 20 6.86 6.82 6.21 6.97 6.34 6.26 7.02 5.02 5.23 5 20 5.97 6.20 6.10 7.60 6.30 4.89 4 97 6.39 6.00 8.60 6.20 6.00 6.63 7.30 5.87 6.25 7.89 7.70 8.50 5.75 6.40 7.60 6.70 7.50 7.85 8.00 8.50 7.10 6.75 7.20 7.00 7.70 7.70 5.41 5.60 5.60 6.40 55 54 4.99 5.60 7.50 6 50 9.20 6.80 6.70 7.00 8.50 6.27 6.50 6.00 6.80 5.86 5.30 6.50 6.20 6.03 4.50 5.10 6.30 6.70 5 30 5.20 6.50 4.56 4.00 5.20 5.60 7.91 .06 4.80 4.20 5.80 5.60 7.70 5.40 5.00 5.90 6.50 6.06 5.70 5.90 5.70 0.48 39 0.30 0.34 0.33 0.31 0.29 0.33 0.27 0.49 0.36 0.34 0.31 0.43 0.31 39 0.27 0.34 0.31 0.29 0.38 0.26 0.72 0.42 0.54 0.60 0.47 0.42 0.36 0.47 0.41 0.43 0.43 0.45 0.40 0.47 0.41 0.45 0.45 0.45 0.77 0.50 0.41 0.42 0.41 0.33 0.33 0.36 0.32 0.49 0.50 0.44 0.35 0.52 0.35 50 0.36 0.37 0.38 0.40 0.51 0.33 0.81 0.59 0.49 0.40 0.58 0.47 0.47 0.49 0.60 0.45 0.58 0.42 0.48 0.46 0.52 0.29 0.33 0.23 0.26 0.27 0.22 0.26 0.30 0.24 0.33 0.28 0.24 0.26 0.35 0.27 0.33 0.24 0.27 0.26 0.21 0.28 0.22 0.63 0.39 35 0.27 0.43 0.36 40 39 0.36 0.33 0.38 0.39 0.43 0.44 0.38 10 University of California — Experiment Station In table 2 the data have been arranged according to variety as well as locality in order that comparison of different localities may be made upon a single variety. The northern California samples are presented first in each series. Four fairly distinct kinds of climate are represented ; first, the cool, foggy climate of the San Francisco Bay region and Salinas Valley; second, the moderately warm climate of the Santa Clara Valley where TABLE 3 Data of 1925 Season Arranged to Show Effect of Climate on Total Solids and Total Acid Locality Summer climate Total solids, per cent Total acid, grams per 100 cc Variety Aver- age Maxi- mum Mini- mum Aver- age Maxi- mum Mini- mum Alameda County Santa Clara County Cool, foggy... Warm 7.02 6.71 4.99 5.02 8.66 6 20 6.10 6.30 6.16 4.89 7.89 7.85 5 75 5 41 9.55 6 54 6.27 4.99 5.86 6.03 4.65 4.56 7.91 6 06 6.06 4.80 48 49 33 31 72 42 0.54 47 41 42 0.77 49 0.41 0.38 0.81 59 42 0.49 29 0.33 27 annef [ Hot, arid Cool, foggy- 0.26 f 63 San Fernando Valley 39 ^ 1 c San Mateo County*. El Monte Warm Warm Hot, arid 0.39 35 One sample only. early morning high fogs are frequent during the summer ; third, the moderately hot summer climate of the Sacramento and San Fernando Valleys ; and fourth, the hot, arid summer climate of Riverside. The tomatoes grown in the cool, foggy climate of the San Francisco Bay district were considerably higher in total solids than those of the same variety grown in warmer localities, and in general, the higher the average summer temperature, the lower was the total-solids content. Total acid did not follow a similar trend (table 4). Unfortunately for comparative purposes, different varieties are grown commercially in southern California than in northern California. The San Jose Canner and its derivatives predominate in the north, while the Norton, Globe, and Stone predominate in the south. The Canner varieties cannot be grown successfully in southern California because of their suscepti- bility to certain diseases prevalent in that region. Bul. 545] Composition of Canning Tomatoes 11 In order to demonstrate more clearly the effect of climate on total solids and total-acid content, some of the data of table 2 have been rearranged in table 3. The San Jose Canner had 7.02 per cent total solids in Alameda County, only 4.99 per cent at Sacramento, and 5.02 per cent at River- side. In other words, in the cool climate of the San Francisco Bay district this variety developed nearly one-third greater total-solids content than in the relatively hot climates of the Sacramento Valley and Riverside. The acidity of the tomatoes from Alameda County was much higher than that of tomatoes from Sacramento and Riverside counties. TABLE 4 Averages of Total Solids and Total Acid of All Samples from Northern California and Southern California for 1925, 1926, 1927, 1928, and 1929 Average total solids, per cent Average total acid, grams per 100 cc Year Northern California Southern California Northern California Southern California 1925 6.57 5 79 48 40 1926 6 01 6 16 29 44 1927 6 53 5.86 41 0.42 1928 6.47 6 40 30 41 1929 7.27 6.82 31 39 Average 6.53 6.08 0.37 43 The effect of climate on the Stone and Norton varieties was similar to that on the San Jose Canner. Compare data in table 3. Kassab (1931) determined the total-solids content of tomatoes from several localities. The Santa Clara variety at Hayward had 9.00 per cent total solids ; at Davis, 6.55 per cent ; at Salinas, 6.92 per cent ; and at Centerville, 9.80 per cent. The cool San Francisco Bay district climate favored development of high total-solids content in this variety also. In table 4 the average total solids and total acid of all samples from northern California and from southern California are given for each of the years 1925, 1926, 1927, 1928, and 1929. The averages for the five years for the two regions were obtained by taking the average of 277 samples from northern California and 236 from southern California. Most of the northern California samples were obtained in Alameda and Santa Clara counties, while most of the southern California samples came from the lower San Fernando Valley, El Monte, Fullerton, Buena Park, and Arlington. 12 University of California — Experiment Station Most of the northern California samples were of the San Jose Canner variety and its derivatives ; most of those from southern California were of the Norton and Stone varieties. As will be shown later, these latter two varieties are usually much higher in total solids and total acid than the San Jose Canner variety when all are grown in the same locality. If this fact is taken into account, the advantage of the northern Cali- fornia tomatoes in total-solids content would be considerably greater than indicated in the table. It would be of interest to compare the composition of tomatoes grown in California with those grown in eastern states ; however, very little published data could be found on eastern varieties that are also grown TABLE 5 Total-Solids Content of the Stone and Marglobe Varieties Grown in Maryland and in California Stone, total solids, per cent Marglobe, total solids, per cent Sample No. Maryland* California Maryland* California 1 7.22 8 29 6 00 6.14 2 6.96 9.55 5 40 6.51 3 5.86 7.91 5 51 7.00 4 5.72 6.08 5 56 7 12 5 5 20 6.35 t 7.20 Average 6.19 7.64 5.62 6.89 * Data for Maryland samples from Appleman and Conrad (1927). t Analysis not reported. in California. Appleman and Conrad (1927) have published some data on total-solids content of the Stone and Marglobe grown in Maryland. We have also analyzed these varieties grown under California condi- tions. Comparative data are given in table 5. The first 3 of the California samples of the Stone variety given in table 5 were grown at Centerville and the 2 remaining samples at Van Nuys in southern California. The samples of California-grown Mar- globe tomatoes came from Van Nuys, Salinas, and Davis. Bigelow and Fitzgerald (1915) published a number of analyses of pulps of tomatoes grown in eastern states, but did not give variety or locality where grown. Their average values were : total solids, 4.37 per cent, total acid as citric, 0.31 per cent, and total sugar as invert, 2.19 per cent. While it would not be justifiable to draw definite conclusions, because of the small number of samples of eastern-grown tomatoes, nevertheless, it would appear that California-grown tomatoes are somewhat higher Bul. 545] Composition of Canning Tomatoes 13 than the eastern ones in total solids. There appears to be little difference in total acid. Effect of Locality on the Composition of Tomatoes from a Single Parent. — Analyses reported in table 10 show that there is considerable variation in composition of tomatoes from different plants grown from seed produced by a single parent. In order to eliminate this variable as far as possible, cuttings from a single plant of a Canner selection were grown at Davis by 0. H. Pearson, of the Division of Truck Crops. TABLE 6 Effect of Locality on" Composition of Tomatoes from Plants from a Single Parent Plant Davis Total solids, per cent Total acid, grams per 100 cc Reducing sugar, per cent Protein, per cent Berkeley Total solids, per cent Total icid, grams per 100 cc Reducing sugar, per cent Protein, per cent First Planting 1 2 3 4 Average 6.6 6.2 62 6.0 6.25 0.42 0.38 0.31 0.35 34 2.70 3.08 3.10 2.49 2.84 10 8 8.1 8.0 8.2 8.8 0.42 0.26 0.37 0.31 0.32 6.44 5.20 4.92 4.92 5.37 1 65 1.26 1.28 1 45 1.41 Second Planting 1 6.2 0.38 3.49 1.25 9.7 38 5.76 1.61 2 6.2 0.32 2.96 1.27 7.7 0.31 5.08 1.22 5.8 0.31 2.87 1.43 8.3 0.37 4.23 1.46 4 5.8 0.37 2.62 1.22 8.4 0.42 4.97 1.39 Average 6.0 0.36 2.98 1.29 8.5 0.37 5.01 1.42 Average of Both Plantings Average 0.35 2.91 1.29 8.6 0.34 5.19 1 41 Plantings of the young plants were then made simultaneously on each of two dates at Davis and Berkeley. Samples of tomatoes of apparently equal maturity were taken at both localities and were carefully analyzed. The plants in both plots were vigorous and normal in appearance. There was no evidence of lack of water in either planting. The analyses are given in table 6. The fruit grown at Berkeley contained 2.5 per cent more total solids, 2.28 per cent more reducing sugar, and 0.12 per cent more protein than the fruit grown at Davis. The total-acid content of the fruit from the two localities was about equal. These results confirm those obtained by analysis of commercially grown samples from the Sacramento Valley and the San Francisco Bay district which are reported in tables 2 and 3. 14 University of California — Experiment Station EFFECT OF TIME OF PICKING ON TOTAL SOLIDS AND TOTAL ACID The tomato-picking season usually lasts from late August to mid or late November in California. It is of interest to know the trend of change in composition with advance of the season in different localities. It is not practicable to give all the data obtained in this phase of the investigation ; however, typical data are given in table 7. TABLE 7 Effect of Time of Picking on Total Solids and Total Acid Locality Centerville.. Sacramento. Fullerton. San Fernando.. Variety San Jose Canner. Stone. Diener. San Jose Canner. Norton. Reference No. Date of picking Sept. 8, 1925 Oct. 3, 1925 Nov. 11, 1925 Nov. 25, 1925 Sept. 8, 1925 Oct. 3, 1925 Nov. 11, 1925 Nov. 25, 1925 Sept. 8, 1925 Oct. 3, 1925 Nov. 11, 1925 Nov. 25, 1925 Sept. 21, 1927 Oct. 4, 1927 Nov. 3, 1927 Oct. Oct. Nov. 3, 1927 10, 1927 2, 1927 Nov. 16, 1927 Sept. 4, 1928 2, 1928 8, 1928 Oct. Nov. Total solids, per cent 6 75 7.22 7.80 6.00 8.29 9.55 8.90 7.91 6.59 8.26 6.31 6 21 6.30 7.60 5 60 6 43 6 10 6 00 5.61 5 90 7.00 6.40 Total acid, grams per 100 cc 0.47 0.42 0.46 53 0.63 0.72 0.73 0.81 47 053 044 0.48 0.26 0.33 28 41 43 0.37 0.41 33 0.42 0.36 On the average, it will be seen that tomatoes picked during October were higher in total solids than those picked during September and November. Many other analyses not presented in the table support this finding. In November the days have become shorter and cooler, and cloudy skies are more frequent. These less favorable conditions prob- ably account for the observed decline in total solids toward the end of the season. Why the total solids should be lower in September than in October is not clear. However, there were several cases in which the Bul. 545] Composition of Canning Tomatoes 15 total-solids content was higher in September than in October. For example, during the 1927 season, Norton tomatoes at Burbank picked on September 7 contained 6.5 per cent total solids; on October 18, 6.2 per cent ; and November 12, 5.1 per cent. Kassab (1931) reports that Norton tomatoes grown at Milpitas con- tained, on September 7, 6.78 per cent total solids ; on October 7, 5.58 per cent ; and on November 7, 5.42 per cent total solids. Similar results were obtained with the Santa Clara, Alameda Trophy, and Marglobe varieties. Evidently local conditions and seasonal variations in some cases have a greater effect than date of picking on the total-solids content. Total-acid content did not appear to follow any definite trend with date of picking. Appleman and Conrad (1927) , reporting on tomatoes grown in Mary- land, give the following values for total-solids content of the Bonny Best variety at different dates : August 16, 7.04 per cent ; August 24, 6.75 per cent ; September 7, 4.93 per cent ; and September 14, 4.87 per cent. Similar values were obtained with the Greater Baltimore, Stone, and Marglobe. Tomatoes picked in Maryland in August were higher in total solids than those picked in September. The season is shorter and the weather conditions are totally different from those in California ; nevertheless, as is the case in California, early and midseason samples were higher in total solids than samples taken toward the end of the season. ANNUAL VARIATION IN TOTAL SOLIDS AND TOTAL ACID In table 2 are given the averages of total solids and total-acid content of tomatoes from several localities for three to five years. The average total-solids content of the samples from Sacramento for 1925, 1926, 1927, 1928, and 1929 were 4.99, 5.85, 6.50, 5.90, and 7.00 per cent respec- tively; and the total-acid values were 0.33, 0.31, 0.29, 0.33, and 0.27 grams per 100 cc. The greatest difference in average total solids was that between the 1925 and 1929 samples, a difference of 2.01 per cent. Samples from Alameda County for 1925, 1927, 1928, and 1929 con- tained 7.02, 6.61, 7.73, and 7.67 per cent average total solids, the greatest difference being between the 1927 and 1928 samples, which was 1.12 per cent. With the exception of the 1929 season, the samples of the Canner variety from Riverside showed relatively little variation in average total solids from year to year. The values for 1925, 1927, 1928, 16 University of California — Experiment Station and 1929 were 5.02, 5.23, 5.20, and 5.97 per cent. The total-acid values were 0.31, 0.29, 0.38, and 0.26 grams per 100 cc. Samples from the San Fernando Valley for 1926, 1927, 1928, and 1929 contained 6.20, 6.00, 6.63, and 7.30 per cent average total solids and 0.43, 0.45, 0.40, and 0.47 grams per 100 cc of total acid. Grand averages of all varieties are grouped in table 4 for the years 1925, 1926, 1927, 1928, and 1929 according to the two geographical regions, northern California and southern California. For northern California the maximum difference in total solids was between 1926 and 1929, a difference of 1.26 per cent, and for southern California between 1925 and 1929, 1.03 per cent. Northern California samples showed a maximum difference in total acid between 1925 and 1926 of 0.19 gram per 100 cc and southern California between 1926 and 1929 showed a difference of 0.05 gram per 100 cc. These variations are less than between samples from a single locality, and as might be expected where large numbers of samples are considered, the variations from locality to locality tend to counterbalance each other. Nevertheless, there is considerable variation in average composition from year to year, the 1929 average being considerably above the averages of the other years. EFFECT OF WATER SUPPLY ON TOTAL SOLIDS AND TOTAL ACID It is common knowledge that tomatoes from well-irrigated plants are larger, of brighter color, and less sweet in taste than from plants that are suffering because of lack of water. In table 8 are given several analyses of tomatoes from normal, well-irrigated plants and from non- irrigated plants grown in the same field. The irrigated plants were large and the leaves were of a healthy, dark-green color. The nonirrigated plants were stunted and the leaves were pale green in color. The tomatoes from the irrigated plants were TABLE 8 Effect of Water Supply on Total Solids and Total Acid of Tomatoes Locality Year Variety Total solids, per cent Total acid, grams per 100 cc Sample No. Well- irrigated plants Non- irrigated plants Well- irrigated plants Non- irrigated plants 1 2 Walnut Creek Walnut Creek 1925 1925 1930 1930 Alameda Trophy 6.72 6.88 9.80 8.95 8.30 8.11 10 30 11.00 0.55 074 0.58 0.71 3 Stone 4 Centerville San Jose Canner Bul. 545] Composition op Canning Tomatoes 17 large in size, juicy, and deep red in color ; those from the nonirrigated plants were small, rather mealy in texture and the color was rather pale. It is readily seen that the nonirrigated tomatoes were considerably higher in total solids. However, since the fruit from the nonirrigated plants was very small, poor in color, and of poor flavor, its higher total- solids content is of little practical importance. EFFECT OF VARIETY ON TOTAL SOLIDS AND TOTAL ACID A number of distinct varieties of tomatoes are grown commercially in California and there are also a number of selections or strains within each of several varieties, particularly the Canner and Stone. Analyses of several distinct varieties have been assembled in table 9. The samples from a given locality were grown in the same or adjoining plots and under similar conditions. In table 10 are given data illustrating the effect of seed selection on total solids and acidity. Many analyses have been omitted, but those included in table 10 are typical. TABLE 9 Effect of Variety on Total Solids and Total Acid Locality Date of picking Variety Total solids, per cent Total acid, grams per 100 cc { ( Sept. 6, 1925 681 133-5F 133-6A 10 11 12 7 70 5.90 7.70 0.28 0.31 0.46 Morse Canner. .< 133-6B 13 7.10 0.44 Sept. 12, 1929...: 133-6C 14 6.30 0.39 133-6D 15 7.10 45 133-5E 16 6.50 0.33 133-5F 17 5.90 0.31 Sept. 21, 1928.J San Jose 401-1 18 5.50 0.33 Canner 411-1 19 4.60 0.27 75 20 5.97 0.40 Davis ■ Oct. 3, 1925 < San Jose Canner Early Late 72-2-1-p 78-1-3 21 22 23 24 8.11 6.88 5.48 5.39 0.71 0.74 0.43 0.33 Nov. 4, 1925.... < San Jose 78-1-2 25 6.84 0.42 Canner 78-1-1 26 6.47 40 57-1-0 27 6.29 0.44 25041 28 5.16 38 Riverside Sept. 30, 1925. - San Jose Canner 25001 25042 25038 29 30 31 4.56 4.86 5.13 0.26 31 31 25043 32 5 41 030 Bul. 545] Composition of Canning Tomatoes 19 the Morse Seed Company, in some cases, also have resulted in a re- markable increase in total-solids content. The Early Canner is an out- standing example. Typical data are given in table 10 to illustrate the effect of seed selection. Fortunately, it was possible to secure from the experimental plots of the Morse Seed Company at Salinas, tomatoes grown in 1930 from seed selected in previous years. Composite samples of fruit from several plants of each selection were used for analysis in order to reduce the effect of variations from plant to plant. Some of the data are given in table 11. Samples 1, 2, 3, and 4 are of a single line ; similarly 5, 6, and 7 ; and 8 and 9. TABLE 11 Comparison of Composition of Tomatoes Grown in 1930 from Selected Seep (Morse Seed Company) Sample No. Variety Selection No. Plants grown from seed stored since year indicated below Total solids, per cent Total acid, grams per 100 cc Reducing sugars, per cent 1 f 9659 1926 6.4 32 2 Santa Clara Canner. J 9658 1927 6.5 33 3 9652 1928 6.6 32 4 1 9653 1928 6.9 33 5 f 1927 5.3 0.33 6 San Jose Canner i 9641 1929 7 0.37 3 28 7 9642 1929 6.8 42 2.84 8 Canner / 9651 1919 6.5 32 3 70 9 I 9650 1924 7.7 33 4.22 Evidently it is possible by seed selection to increase the total-solids content of a variety at least 2 per cent, or practically one-third. There- fore it would appear desirable and profitable for camiers to give more attention to this method of improving existing varieties for tomato- products manufacture. The data here given, however, are too meager to demonstrate to what extent observed variation is due to crossing, mutation, and heterogeneity. Naturally, the effect of selection is apt to be mixed; some of the progeny may be superior, some equally good, and some inferior to the parent stock. This is shown in samples 1 to 4 in table 11. Lesley and Rosa (1926), by three years' inbreeding and selection of the Santa Clara Canner variety, isolated lines differing greatly in smoothness, size, color, and time of maturity. Some of these lines were constant after the second generation, while others were variable even after three generations. 20 University op California — Experiment Station Composition of the fruit is, of course, not the only criterion by which tomato seed must be selected for propagation of tomatoes for canning and tomato-products manufacture. Size of the fruit, its color, flavor, and texture are all very important. The vigor of the plant, its resistance to disease, the yield of fruit per acre, and the time of ripening must also be taken into account. Therefore, the task of improving tomato varieties by crossing and seed selection is a complicated one. The point we wish to make is, that while chemical compostion of the fruit should not be unduly stressed, nevertheless, it should not be ignored in syste- matic attempts to improve existing varieties. TOTAL SOLIDS AND TOTAL-ACID CONTENT OF TOMATOES FROM DIFFERENT PLANTS Composite samples of the fruit from each of 10 plants of San Jose Canner selection 9641, grown in the same row in the Morse Seed Com- pany's plot at Salinas, were analyzed with the results as given in table 12. TABLE 12 Variations in Composition of Fruit from Consecutive Vines of a Single Selection Plant No. Total solids, per cent 1 7.7 2.... 6.9 3 ... 6 9 4..... 7.0 5 7 1 6 7 7 7 8 . 7 5 9 . . 6.9 10... 6.9 7 09 With the exception of fruit from plants Nos. 1 and 8, the various samples agree well in total-solids content. However, the fact that 2 plants out of 10 exhibited rather large deviation from the average shows the necessity of using composite samples from several plants if the representative composition of a given variety or selection is to be found. See also data in table 6. Bul. 545] Composition of Canning Tomatoes 21 TOTAL SOLIDS OF FRUIT FROM VARIOUS LOCATIONS ON THE PLANT Samples of fruit of apparently equal ripeness were selected from the north, south, east, and west quarters and from the top center of each plant of three selections in the experimental plot of the Morse Seed Company at Salinas. The specimens from each location were combined and the resulting composite samples were analyzed. The data are given in table 13. TABLE 13 Total Solids of Fruits from Various Locations on the Same Plant Total solids content in per cent Location San Jose Canner, Selection No. 9642 Early Santa Clara Canner, Selection No. 9644 Santa Clara Canner, Selection No. 9647 North 6 4 6 8 6.8 6 3 6.3 6.2 6.6 6.9 6.5 6.3 6.0 West 6 South 6 East 5.9 6 2 It would appear that fruits from the west and south quarters of the plant are somewhat higher in total-solids content. The total acid of the samples from San Jose Canner 9642 were north, 0.43 grams per 100 cc ; west, 0.44 ; south, 0.48 ; east, 0.40 ; and top center, 0.40. Fruits from the top center of the plant were not remarkably high in total solids, although since they receive more sunlight than fruit on other portions of the plant they might be expected to be richer in total solids. TOTAL-SOLIDS CONTENT OF DIFFERENT PORTIONS OF THE FRUIT The cores, the locules (the jelly-like material around the seeds), and the interlocular walls and outer walls of individual fruits were segre- gated and composite samples of several fruits were used for analysis. Some of the data are given in table 14. The average composition of the various portions and of the whole fruit of all the samples analyzed in this particular study are given in table 15. 22 University of California — Experiment Station TABLE 14 Composition of Different Portions of the Tomato Fruit Variety Selection No. Total solids, per cent Total acid, per cent Reducing sugars, per cent Core Locule Walls Core Locule Walls Core Locule Walls 198-1-3 6.2 5.8 5.8 37 45 35 2 94 2.52 2.79 198-1-4 6 6 5 8 0.37 54 33 2.66 2.07 2 53 198-1-5 6 3 6 6.0 31 38 27 3.36 2.86 2.84 198-2-2 5 5 5 0.28 35 30 2.86 2 34 2 39 198-2-4 5 2 5 5 35 45 30 2.61 2.13 2.38 198-3-2 5 2 5 1 5.1 32 36 0.26 3.47 3.15 3.76 Canner 198-1-1 7 1 6.6 6.6 40 61 0.40 2 90 2.69 2.32 198-1-2 6 3 6 6.2 36 54 0.36 3.30 2 90 2.96 198-2-1 6 2 5 9 6.2 31 41 0.31 3.18 2.72 3.04 198-2-2 5 8 5.1 5.5 31 42 30 2.60 2.13 2.37 198-2-4* 6 3 5.8 5.9 36 59 35 1 89 1.51 1.50 198-1-1* 6 3 6.2 6.2 0.38 55 0.37 3 61 3 26 3.30 198-1-2* 6 3 5.8 6.2 30 41 19 2.77 2.30 2.85 Santa Clara Canner 5 1 5.0 5 38 36 0.27 3 30 2 98 3.07 * Fruit from second picking. TABLE 15 Average Composition of the Different Portions and of the Whole Fruit Portion Total solids, percent Total acid, grams per 100 cc Reducing sugars, per cent Protein,* per cent Acid hydrolyzable,* per cent 6.00 5 80 5 83 5.90 32 46 32 0.37 3.17 2 54 2 73 2.87 1.29 1 20 1.31 1.30 258 0.055 Walls 0.292 Whole fruit 0.255 * Protein and acid-hydrolyzable material were not determined on all samples and hence are not presented in table 14. It is apparent that the core is relatively high in total solids, reducing sugar, protein, and acid-hydrolyzable material. The locular material (the jelly-like substance in the seed cavities) is higher in total acid but lower in total solids, reducing sugar, protein, and acid-hydrolyzable material than the core and the walls. PROTEIN CONTENT OF TOMATOES The protein content of 110 samples about equally distributed between the 1927, 1928, and 1931 seasons was determined. Typical average re- sults and averages for each season together with the grand average of all samples are given in table 16. Considerable variation in protein content was found in samples of the same variety from a single location, in samples from different locali- Bul. 545] Composition of Canning Tomatoes 23 TABLE 16 Protein Content of California Tomatoes Variety Canner (198-1-1)* Canner (198-1-2)* Canner (198-1-4)*. Canner (198-1-3)*. Canner (198-2-4)*. Canner.. Santa Clara Canner. Trophy. Norton Canner. Locality Davis Berkeley Davis Salinas San Jose Salinas Hayward San Fernando J El Monte Van Nuys Burbank San Fernando . Fullerton. Burbank Miscellaneous. Averages Year Number of samples Protein, per cent Total solids, per cent ( 2 1.25 6 2 2 1 27 6.2 2 1.22 5.8 1931 \ 2 1 32 6.2 2 1 34 5.0 8 1.41 8.8 I 2 1.29 6.1 ( 2 0.81 7.5 1928 -j 2 131 7.1 1 2 0.90 7 1 f 2 67 5 5 1927 1 2 0.84 6.2 I 2 0.77 6.2 f 2 0.63 7.7 1928 \ 1 0.74 7.0 I 2 1.09 6.8 \ 2 85 6 1928 J 2 1 09 6 3 ) 2 83 6.8 1 2 0.82 6.9 f 3 1 04 5.8 2 1.08 6.4 1927 1 2 0.77 5.2 2 75 5 2 2 0.85 6.1 I 3 0.82 6.0 1928 / 2 1 00 6.8 I 2 0.85 5.6 1928 46 0.92 6.4 1927 21 0.83 5.8 1928 69 89 6.6 1931 20 1.34 5.8 1927 ] 1928 \ 110 0.97 6.2 1931 J *Selection No. ties, and in samples of different seasons. In general, high total-solids content was accompanied by high protein, although there was no close, direct relation. The lowest observed protein concentration was 0.55 per cent accompanied by 5.2 per cent total solids ; the sample was of a seed selection of the Santa Clara Canner (included in averaged data in table 16) grown at Salinas in 1928. The highest protein content was 24 University of California — Experiment Station that of a sample of Canner selection (table 6) grown in Berkeley in 1931. It contained 1.65 per cent protein and 10.8 per cent total solids. The grand average of 110 samples was 0.97 per cent protein and 6.2 per cent total solids. The 1931 samples were considerably higher in protein content than the 1927 and 1928 samples; possibly the drier season of 1931 was in part responsible for this difference. Sherman (1915) gives the average protein content of 19 samples of tomatoes as 1.20 per cent and total solids as 6.0 per cent. Chatfield and Adams (1931) report the average protein content of 34 samples as 1.0 per cent; maximum 1.8 per cent and minimum 0.7 per cent. The average total-solids content was 5.9 per cent. Neither Sherman nor Chatfield and Adams state the source of their samples. It would appear that tomatoes possess, on the average, about 1.00 per cent to 1.10 per cent of protein. From table 15 it may be seen that the cores and walls of the tomato are slightly higher in protein than the locules. SUCROSE, REDUCING SUGARS, AND DEXTROSE-LEVULOSE RATIO As shown in table 1, most of the sugar in tomatoes is present in the form of reducing sugar ; there is very little, usually less than 0.1 per cent, of sucrose. During the 1926 season 16 samples were analyzed for reducing sugars ; during 1928, 70 samples ; during 1929, 93 samples ; and during 1931, 33 samples. Typical data are given in table 17. In general, high total solids are accompanied by a high reducing- sugar content. The nonsugar-solids content was not a constant, but like the sugar tended to vary with the total solids. On the average, non- sugars by difference were about equal to the sugars, if slightly less than 0.1 per cent is allowed for sucrose. However, in individual samples the ratio varied considerably. The Norton did not differ materially from the San Jose Canner in reducing-sugar content. Tomatoes grown in Berkeley showed a considerably higher content of reducing sugars than of nonsugars; in those grown at Davis from the same parent as at Berkeley, the nonsugar solids exceeded the sugars. This finding would indicate that locality affects the sugar to nonsugars ratio materially. By inversion of a sample in 1931 with invertase there was found an average of 0.027 per cent sucrose and by inversion with HC1 a like amount in triplicate analyses. Rosa (1925, 1926) reported 0.03 to 0.04 per cent sucrose in Stone and Globe tomatoes grown at Davis. Benoy Bul. 545] Composition of Canning Tomatoes 25 TABLE 17 Beducing-Sugars Content of Tomatoes, on a Wet Basis Variety Locality Year Sample No. Total solids, per cent Reducing sugars, per cent Nonsugar solids by difference, per cent Sacramento / 1926 / 1 6 5 4 42 2 08 \ 1 2 6 3 3 46 2 84 f f 3 7 1 3.88 3 22 San Jose \ 1929 \ 4 6.9 3 50 3.40 { 1 5 6 5 3.54 2.96 6 8 1 3 90 4.20 Centerville 7 8 8 7 8 6' 4.16 4 00 4.64 3.60 San Jose Canned : 9 10 8 6 1 9 4.10 3 30 4 00 3 60 Sacramento < 1929 < 11 12 6 7 8 3 36 3 03 3 44 3 97 Hayward 13 14 8 8 5 4 05 4.57 3 95 3 93 Riverside 15 5 9 2.74 3.16 16 5 6 2.55 3 05 1 f 17 to 33 5 90 2 87 3.03 Davis \ 34 6.25 2 84 3.41 Canner 1931 < 35 6 00 2 98 3.02 (198-6-3)* < Berkeley / \ 36 37 8.80 8.50 5 37 5 01 3 43 3.49 { 1926 38 6.80 3.30 3 50 Burbank \ 1929 / 39 6.50 3.12 3.38 \ 40 7.00 3 33 3.67 I 1926 41 6.40 3 20 3 20 Norton • Fullerton \ 1929 / 42 7 70 2.96 4.74 { 1 43 8 50 3 60 4 90 Van Nuys c 44 7.10 3 56 3.54 El Monte 1926 1 45 6.40 3 38 3 02 Long Beach 46 6.40 3.29 3.11 Arlington 47 6.40 3.34 3.06 1926 6.80 3.46 3 34 1928 6.45 3 45 3 00 1929 7.27 3.29 3.98 Averages ■ 1931 1926 1 6.60 3.46 3.14 1928 1 6.86 3 38 3.48 1929 f ' 1931 J Selection No. 26 University of California — Experiment Station and Webster (1930) were unable to find any sucrose in the samples analyzed by them. It is evident that the sucrose content of tomatoes is at most very low, probably less than 0.05 per cent in most cases. The dextrose-levulose ratio was determined on typical samples from the San Francisco Bay district and from southern California. Total TABLE 18 Dextrose-Levulose Content of California Tomatoes, 1932 Season Sample No. Dextrose, per cent Levulose, per cent Total reducing sugars, per cent Ratio of dextrose to levulose Southern California samples 9 10 11 Average 2.06 1 85 3.91 2.06 1.71 3.77 1.93 1.73 3.68 2.11 1.68 3.79 2.04 1 81 3.85 2.15 1.84 3.99 2.06 1.92 3.98 1.99 1 91 3.90 1 98 1.92 3.90 2.09 1.97 4.06 1.89 1.59 3.48 2.03 1.80 3.84 111 1.20 1.11 1 25 1.13 1.17 1 07 1.04 1.03 1.06 1.19 1.17 Northern California samples 12 2.50 2.37 2 35 2 24 2.39 2 12 2 33 2.04 1.97 1.81 1.73 1.96 1.78 1.88 4.54 4 34 4.16 3.97 4.35 3.90 4.21 1.22 13 1.20 14 1.30 15 . 1.29 16 1.22 17 1.19 1 24 reducing sugar was determined by the Shaffer-Hartmann method (1921), and dextrose by the Lothrop and Holmes method (1931). The latter method consists of selective oxidation of the dextrose in alkaline iodine solution. The data are presented in table 18. Apparently the dextrose content is definitely and consistently higher than the levulose content. The average ratio of dextrose to levulose for the 17 samples was 1.19. It is of interest that it was slightly lower for the southern California samples than for those from northern Cali- fornia. This may be due to effect of variety or climate. Bul. 545] Composition of Canning Tomatoes 27 STARCH CONTENT OF TOMATOES Tomatoes contain very little to no starch. Consequently only brief study was made of this constituent. It was determined after inversion by diastase as previously described (page 8). By diastasic inversion, an average of 0.045 per cent starch was found. This value is much less than that of the total acid-hydrolyzable material. This result compares favorably with those of Rosa (1925, 1926) who reported 0.044, 0.052, 0.034, 0.012, and 0.022 per cent starch in ripe tomatoes grown at Davis. These values are expressed upon the wet basis. ACID-HYDROLYZABLE MATERIAL Acid-hydrolyzable material was determined as previously described (page 8). Average results of analyses of different portions of the fruit are given in table 15. The acid-hydrolyzable content of the cores was 0.258 per cent ; of the locules, 0.055 per cent ; of the walls, 0.292 per cent ; while that of the entire fruit was 0.255 per cent. It will be seen that the content of acid-hydrolyzable material is several times that of starch as determined by diastase inversion. PECTIN AND PROTOPECTIN CONTENT OF TOMATOES The pectin content of tomatoes, as affected by variety, ripeness, local- ity, and water supply was determined by the Wichmann (1922, 1923) method during the 1930 season by M. A. Kassab, a graduate student in the Fruit Products Laboratory of the University of California. In a few cases protopectin was also determined. In order to conform to the usual practice and to permit comparison of Kassab 's results with those of Appleman and Conrad the data are presented on the dry basis, that is, pectin as percentage of total dry material present in the whole fruit. All other data in this publication are presented on the fresh, or wet basis. The whole fruit rather than merely the pulp was analyzed. The data of Kassab show that the pectin content varies considerably with the variety, although in most of the samples analyzed the pectin varies more or less directly according to the total-solids content. It would appear also that the Santa Clara Canner is, on the average, richer in pectin than the other varieties examined. 28 University of California — Experiment Station TABLE 19 Effect of Ripeness on Pectin and Protopectin Content of Tomatoes, Milpitas, 1930* Variety Stage of development Approx. diameter, inches Total solids, per cent of fresh fruit Protopectin, per cent on dry basis Pectin, per cent, on dry basis Early Canner ■ • { f Very green -j Moderately green 1 2 3 3 1 1 2 3 3 3 3 3 10.60 10.22 9.17 8.22 7.50 6.67 11 10 10.87 9.45 8.95 7.95 11.45 9.00 8.55 2.32 2.25 2.20 1.80 1.25 10 2.27 1 37 1.30 0.11 0.03 2.79 trace trace 87 0.87 0.87 1 25 2 28 Full red 2.58 Alameda Trophy f Very green «{ 0.77 94 0.98 2.20 Full red 2.70 f Canner i Full red 2.44 1.57 * From analysis reports of M. A. Kassab, graduate student. f Not determined. Kassab found that as the total solids decreased towards the end of the season, the pectin decreased also. Appleman and Conrad (1927), reporting on tomatoes grown in Maryland, found relatively little change in pectin concentration in the Bonny Best between August 14 and September 14 ; corresponding total-solids determinations were not given. They found greater change in pectin than in pectin plus proto- pectin. Average pectin plus protopectin (total pectic constituents) concentrations, expressed on dry basis, of all varieties studied by them varied as follows : August 16 2.35 per cent August 24 2.85 per cent August 30 3.18 per cent September 7 3.79 per cent September 14 3.79 per cent September 21 4.00 per cent Average 3.52 per cent Kassab reported that tomatoes grown at Davis were lower in pectin as well as in total solids than those grown in the cool, foggy climate of Hayward and Centerville. Lack of water, he found, caused an increase in pectin as well as in total solids of tomatoes grown near Centerville. Very green tomatoes contained more protopectin than pectin. As the fruit ripened the proto- pectin decreased and the pectin increased (see table 19). At full ma- Bul. 545] Composition of Canning Tomatoes 29 turity, the Wichmann (1922, 1923) method indicated that the proto- pectin had decreased to about 0.1 per cent. In passing, it may be noted that the total-solids content of partly grown, green tomatoes was higher than in the ripe fruit. Appleman and Conrad (1927) also found a change of protopectin to pectin during ripening, although their data show less change than that found by Kassab. The pectin constitutes only a relatively small portion of the total solids. The average total-solids content of the five varieties studied was 7.25 per cent of the fresh fruit and the pectin, 1.86 per cent on the dry basis. The fresh tomatoes would then contain only 0.13 per cent pectin. Appleman and Conrad found a similar concentration of pectin ; thus the average pectin content of the Bonny Best was 2.11 per cent on the dry basis and about 0.125 per cent on the wet basis. "INSOLUBLE SOLIDS BY DIFFERENCE" The proportion of suspended matter in tomato products affects their consistency and appearance. However, the content of suspended matter is affected materially by the thoroughness of the screening operation in pulping. Consequently, unless the history of given pulp samples is known, it is impossible to determine what proportion of any observed variation is due to manufacturing methods and what proportion is due to variety, weather conditions, date of picking, etc. Solids content of the unfiltered pulps and of the clear filtrates of most of the 1925 samples were determined by drying in vacuo. In- soluble-solids content was determined by difference, that is, by sub- tracting the solids content of the filtrate from that of the unfiltered pulps. Insoluble-solids content so determined is not necessarily the true insoluble solids. The latter can only be determined by filtering or centrifuging off the insoluble solids, repeatedly washing with water to remove soluble matter, drying, aud weighing the residue. However, "insoluble solids by difference" is of some interest as an index of variation in composition of tomato pulps. The average in- soluble solids by difference of 75 samples was 0.74 per cent and the average total solids was 6.50 per cent. The maximum insoluble-solids value was 1.11 per cent and the minimum 0.42 per cent. Only 10 samples out of 75 contained less than 0.60 per cent insoluble solids, while 23 samples were above 0.80 per cent. The remainder lay between 0.60 and 0.80 per cent. High total-solids content was usually accom- panied by high insoluble solids. 30 University of California — Experiment Station Bigelow and Fitzgerald (1915) give a range of insoluble solids, in unconcentrated pulps determined directly, of 0.40 to 2.02 per cent ; their average being 0.701 per cent. pH VALUE OF JUICE OF TOMATOES The hydrogen-ion concentrations, expressed as pH values, of 76 filtrates from pulp samples were determined during the 1925 season by a hydrogen electrode and potentiometer. The average pH value was 4.1. The maximum was 4.4 and the minimum was 3.8. Most of the samples ranged in pH from 4.0 to 4.2 ; 53 samples out of 75 lay within this range. High acidity (titratable) was usually accompanied by low pH and low acidity by high pH value. SUMMARY The average total solids and sugar of tomatoes of a given variety grown in the cool, foggy climate of the San Francisco Bay region were higher than those of the same varieties grown in the Sacramento Valley. The average total-solids content of all samples from southern Cali- fornia for a five-year period was 0.55 per cent lower than that of the northern California samples. In general, the higher the summer tem- perature, the lower was the total-solids content. From the data at hand, tomatoes grown in California appear to be higher in total solids than those grown in the eastern United States. Tomatoes picked in November were considerably lower in total solids than those picked during October and September. In most seasons the tomatoes picked during October were somewhat higher in total solids than those picked in September. Tomatoes from plants suffering from lack of water were higher in total solids than those from well-irrigated plants. Of the important commercial varieties, the Stone was consistently highest in total solids and total acid. The Early Canner was also excep- tionally high in total solids. Analyses of many samples of tomatoes grown from seed selections indicated a wide range of variation in total solids and total acid. Seed selection has been found an effective means of altering the composition of a given variety of tomato. Tomatoes from adjacent plants and from different locations on a single plant were found to vary considerably in composition. Analyses of the cores, locules, and walls showed the locular material to be higher in total acid and slightly lower in total solids, reducing sugars, and protein than the cores and walls. The cores were richest Bul. 545] Composition of Canning Tomatoes 31 in total solids and reducing sugars. The locular material was much lower than the walls and cores in acid-hydrolyzable material. The average protein content of samples from three seasons' collec- tions was 0.97 per cent. Practically all of the sugars present in tomatoes were found to be reducing sugars. Sucrose concentration was found to be less than 0.05 per cent. Tomatoes do not contain a very active invertase. The starch content of tomato pulp was found to be less than 0.05 per cent. The pectin content of ripe tomatoes varied considerably with the variety, time of picking, ripeness of the fruit, and locality. The insoluble solids by difference averaged 0.74 per cent, but ranged in irregular manner from 0.42 to 1.11 per cent. The average pH value of the 1925 samples was 4.1 ; most of the pH values lay between 4.0 and 4.2. The investigation definitely shows that tomatoes vary sufficiently in composition to warrant purchasing them on the basis of total-solids content when they are to be used in the manufacture of tomato products. ACKNOWLEDGMENTS The writers wish to acknowledge gratefully the cooperation given by Dr. O. H. Pearson, of the Division of Truck Crops ; Dr. J. W. Lesley, of the Citrus Experiment Station; W. H. Nixon, of the Morse Seed Company at Salinas ; M. P. Duffy, Chief of the Cannery Inspection Service of the State Board of Health ; the California Packing Corpora- tion; the F. E. Booth Company; and F. A. Dixon, of the Canners League. LITERATURE CITED Appleman, C. O., and C. M. Conrad. 1927. The pectic constituents of tomatoes. Maryland Agr. Exp. Sta. Bul. 291:4. Association' of Official Agricultural Chemists. 1925. Official and tentative methods of analysis. 2nd ed., 535 p. Association of Official Agricultural Chemists, "Washington, D. C. Atwater, W. O. and A. P. Bryant. 1906. The chemical composition of American food materials. U. S. Dept. Agr. Office Exp. Sta. Bul. 28:69. Rev. ed. Bell, M., M. L. Long, and E. Hill. 1926. The available carbohydrate content of some fruits and vegetables. Jour. Metabolic Research 7-8:195-197. Benoy, M. P., and J. E. Webster. 1930. Chemical composition of fresh vegetables. Plant Physiol. 5:181-182. 32 University of California — Experiment Station Bigelow, W. D., and F. F. Fitzgerald. 1915. Examination of tomato pulp. Jour. Indus, and Engin. Chem. 7:602-606. Bigelow, W. D., and A. E. Stevenson. 1923. Tomato products. National Canners Association Bui. 21-L:2-60. BORNTRAGER, A. 1925. [The organic acids of tomatoes, besides citric acid, and their state of combination.] Z. Nahr. Genussm. 50:273-300. [Abstracted in: Chem. Abs. 20(1) :952. 1926.] Chatfield, Charlotte, and Georgian Adams. 1931. Proximate composition of fresh vegetables. U. S. Dept. Agr. Cir. 146:22. Cruess, W. V., and L. G. Saywell. 1928. Eefractometer tests for solids. Western Canner and Packer 20:12-13. 1929. Composition of California tomatoes. The Canner 67(18) -.17-18. Cruess, W. V., L. G. Saywell, and P. NT. Mark. 1929. Study of tomato composition. Western Canner and Packer 21:16-18, 30-32, 34. HUELSEN, W. A. 1925. The rapid determination of specific gravity of tomato pulp. Canning Age 6:631-635. Kassab, M. A. 1931. Observations on the pectin content of California tomatoes. Fruit Prod- ucts Jour. 10(10): 69-71. Lawrence, R. D., and R. A. McCance. 1929. Analyses of carbohydrate foods and their application to diabetic diets. Brit. Med. Jour. 1929(2) :241. Lesley, J. W., and J. T. Rosa. 1926. The improvement of tomatoes by selection. Hilgardia 2:25-45. Lothrop, R. E., and R. L. Holmes. 1931. Determination of dextrose and levulose in honey by use of iodine-oxidation method. Indus, and Engin. Chem. (Analytical ed.) 3:334-339. MacGillivray, J. H. 1929. Highest tomato quality. The Canner 67(10) :13-17. Myers, V. C, and H. M. Croll. 1921. The determination of carbohydrate in vegetable foods. Jour. Biol. Chem. 46:537-552. Olmsted, W. H. 1920. Availability of carbohydrate in certain vegetables. Jour. Biol. Chem. 41:45-58. Rosa, J. T. 1925. Ripening of tomatoes. Amer. Soc. Hort. Sci. Proc. 22:315-322. 1926. Ripening and storage of tomatoes. Amer. Soc. Hort. Sci. Proc. 23:324-329. Shaffer, P. A., and A. F. Hartman. 1921. The iodometric determination of copper and its use in sugar analysis. Jour. Biol. Chem. 45:365-390. Sherman, H. C. 1915. Food products. 529 p. Macmillan Publishing Company, New York. Wichmann, H. J. 1922. Report on the determination of pectins in fruit and fruit products. Jour. Assoc. Off. Agr. Chem. 6:36. 1923. Report on the determination of pectins in fruit and fruit products. Jour. Assoc. Off. Agr. Chem. 7:107. 12m-l,'33