A spontaneous tomato (Lycopersicon esculentum Mill.) triploid was studied with a view to its commercialization. Fruits induced by auxin contained 50% more DNA and 30% more protein than their diploid counterpart. The fruits were 50% larger than those of the diploid counterpart and were juicy but seedless. All fruit quality characteristics checked (polygalacturonase activity, reducing sugars content, electrical conductivity, pH, titratable acidity, pigment content, and shelf life) were comparable to the diploid except for ethylene evolution rate, which was lower than that of the diploid counterpart, and flavor, which was superior. The line seems suitable for agricultural cultivation. Triploid tomato plants are not readily obtainable through crosses between tetraploid and diploid tomato cultivars (Cooper and Brink, 1945; Jorgensen, 1928; Nilsson, 1950). However, they do arise occasionally in tomato plots and are readily identified by their vigor and fruitlessness (Rick, 1945). In a study aimed at estimating the commercial potential of fruits from triploid tomatoes, we searched for triploids among the commercial greenhouses of the Negev (southern area of Israel). A triploid plant discovered in a greenhouse of the tomato hybrid cultivar FC121 was verified by chromosome count, and was found to descend from a haploid female and a diploid male parent (Kagan-Zur et al., 1991). Triploid tomato plants often do not bear fruits in nature (Jorgensen, 1928; Nilsson, 1950; Rick, 1945). If induced by artificial means, fruits have virtually no seeds due to chromosomal imbalance of the gametes (Strickberger, 1976). Tetraploid tomato plants bear smaller fruits that contain about one-tenth as many seeds as isogenic diploid lines (Kagan-Zur and Mizrahi, 1987; Nilsson, 1950). Fruit size has been correlated with the number of seeds developing in the fruit (Varga and Bruinsma, 1976) and seed number correlated with auxin level (Mapelli et al., 1978). In diploid plants, it is possible to meet the auxin demands of the ovary and produce seedless parthenocarpic fruit by exogenous application of an auxin to emasculated flowers (BungerKibler and Bangerth, 1982/3; Mapelli et al., 1978). We assumed that the same treatment would induce fruit set and development on triploid plants. As polyploidy usually causes gigas appearance of plant organs (Jorgensen, 1928; Rick and Butler, 1956; Saimbhi and Brar, 1978; Tarn and Hawkes, 1986), and the smaller fruit size of tetraploids is correlated with lack of auxin, we surmised that auxin treatment would cause triploid fruits to surpass normal diploid fruits in size. We undertook this study to assess the possibility of obtaining triploid fruits, and to study their physiology and potential as a crop. Materials and Methods Plant material. A triploid of the tomato hybrid cultivar FC121 was identified in a commercial greenhouse in the Negev area for publication 18 Dec. 1989. We thank the Israel Endowment Fund the Isreael Ministry of Agriculture for partial financial support of The cost of publishing this paper was defrayed in part by the payment arges. Under postal regulations, this paper therefore must be hereby vertisement solely to indicate this fact. he Dept. of Biology. and propagated through cuttings together with the diploid hybrid obtained from the same source. Parent lines-of ‘FC121’ were grown from seeds (kindly provided by its breeder, N. Kedar). The parent plants were initially grown from germinated seeds, then propagated from cuttings as needed. Culture. Seeds were sown in vermiculite in a greenhouse, maintained under natural light conditions, and irrigated with Hoagland’s solution (Hoagland and Arnon, 1950). Triploids as well as ‘FC121’ diploids were propagated through cuttings, which were allowed to root in half-strength Hoagland’s solution. Threeweek-old seedlings and rooted cuttings were transferred to dark, opaque 10-liter buckets containing 1 vermiculite : 1 perlite : 1 C1 Finnish peat (by volume). Plants were irrigated with halfstrength Hoagland’s solution every other day. Ten plants of each cultivar were cultivated and trained to a single stem. Only the first four fruits on each inflorescence were allowed to develop. Flowers were either manually pollinated, tagged, and allowed to develop, or emasculated and treated with auxin. A commercial diluted solution of auxin [“NO Seed”, Bruinsma Ltd. Holland, consisting of 0.1 g of b -naphthoxy-acetic acid/liter in water and a drop of the surfactant Tween 20] was applied manually to emasculated diploid flowers and to nonemasculated triploid flowers at full anthesis. Plants were grown in four consecutive seasons—winter, spring, summer, and fall. Cell size. Thin slices of fresh pericarp sectioned with a razor blade were observed under a microscope with a calibrated ocular at 400 × magnification. The same cell layer was measured in each sample. The length and width were taken to the largest and smallest diameter of a cell, respectively. Cell volume was approximated as: (width) × (length) = volume. DNA content. Pericarp tissue (0.1 g) was frozen in liquid N and ground with methanol. After centrifugation for 10 min at 1350 × g in a bench centrifuge (Runne Heidelberg Mod 100-2; Heidelberg, Germany) the supernatant was discarded; this was repeated until a white pellet was obtained. The pellet was resuspended in 2.0 ml of 0.5 N perchloric acid and heated to 70C for 45 min. Following centrifugation as above, the supematant served for the DNA assay. DNA was assayed according to Richards (1974) using diphenylamine in glacial acetic acid and acetaldehyde. Protein content. For protein extraction, 2.0 ml of 1 M NaOH was added to ground pericarp after extraction of pigments (see below). The samples were then boiled for 20 rein, centrifuged at 1350 × g, and the supernatant set aside. This procedure was repeated twice at room temperature. All supernatants of a sample were combined and 0.4 ml was taken for protein determination. Protein content was determined using the method of Lowry et al. (1951), Na2C O3 and CuSO4 being added to the samples followed by Folin–Ciocalteus reagent. J. Amer. Soc. Hort. Sci. 116(2):228-231. 1991. Table 1. Effect of ploidy level and mode of fruit-set on fruit size and time from set to first color in hybrid tomatoes and their parent lines. zNOA = 0.1 g β -naphthoxy-acetic acid/liter. The averages include > 30 fruits per parental line and at least 60 fruits per hybrid line. yMean separation by Duncan’s multiple range test. Ethylene evolution. Ten fruits of each line were picked at 80% development (100% development being the time from anthesis to first red color in fruits) (Lyons and Pratt, 1964) and individually placed in the flowing air system described by Kopeliovitch et al. (1980). Ethylene evolution rate was measured daily by withdrawing a gas sample with a syringe and analyzing its contents by gas chromatography (Kopeliovitch et al., 1980). Each fruit was frozen 10 days after the onset of ethylene evolution, so that fruits all were at a comparable physiological stage when thawed for chemical analysis. Polygalacturonase activity. Polygalacturonase (PG) (EC 3.2. 1.15.) activity was measured as described by Mizrahi et al. (1976). Frozen tissue (40 g) was diced and homogenized in 1.0 M NaCl, then filtered and centrifuged (600 × g). The supernatant was, dialyzed against cold water. This served as a crude extract of the pectolytic enzyme. Ten milliliters of 4% sodium polypectate was added to 5 ml of crude extract. Substrate and extract were mixed in a viscosimeter (Cannon 200, United States), based on the principle that viscosity can be determined by recording the time needed for the solution to pass through a narrow calibrated glass column in the viscosimeter. Measurements were taken every 2 rein, and the activity was expressed as the time needed for a 50% decrease in the initial viscosity. Pigment analysis. Pigments were analyzed by weighing pericarp discs 11 mm in diameter and extracting each disk with 5 ml of 4 acetone : 5 hexane (v/v). After centrifugation, the optical density of the supernatant was read at several wavelengths with a Kontron UNIKON 810 spectrophotometer (Hager and Meyer-Bertenrath, 1967). Chemical analyses. Fruit tissue (10 g) was homogenized with 5 ml of water in a Virtis homogenizer, the homogenate centrifuged in a Sorvall RC2-B centrifuge (12000 × g, 10 rein), and the supernatant solution analyzed. Both pH and electrical conductivity were recorded using El-Hama Instruments PBS 710 and TH 250 (Tel Aviv), respectively. The amount of reducing sugars was measured according to the method of Sumner (1921) with dinitro-salycilic acid reagent. Acidity was estimated by titration with NaOH standardized against a KOH volumetric solution (BDH lg104 3u, Dorset, England). The analyses were performed on individual fruits from the ethylene evolution studies 10 days after the last of the tomatoes had entered the stage of ethylene evolution. The experiment was performed on three occasions several months apart, with essentially the same results, and all data were pooled. Sensory tests. Flavor was estimated using fresh fruits under red light (to avoid effect of color on taste estimation). Several fruits of each line were cut so that each slice contained pericarp and jelly and the slices were mixed together. Evaluations were requested to rate the samples as: 1 (worst), 2 (medium), or 3 (4 in one test) (best) (Kopeliovitch et al., 1982). Average flavor scores were calculated.