XI B RAR.Y OF THE UNIVERSITY Of ILLINOIS 30.7 UGb *o.354-3G3 cop. 2. EGHICUtTURE NOTICE: Return or renew all Library Materials! The Minimum Fee tor each Lost Book is $50.00. The person charging this material is responsible for its return to the library from which it was withdrawn on or before the Latest Date stamped below. Theft, mutilation, and underlining of books are reasons for discipli- nary action and may result in dismissal from the University. To renew call Telephone Center, 333-8400 UNIVERSITY OF ILLINOIS LIBRARY AT URBANA-CHAMPAIGN APR 2 8 2DI15 LI61 O-I096 Breeding Two New Varieties of Greenhouse Tomatoes Resistant to Fusarium Wilt By WALTER A. HUELSEN and MERI. C. GILLIS THE NEW VARIETIES READY FOR MARKET UNIVERSITY OF ILLINOIS AGRICULTURAL EXPERIMENT STATION BULLETIN 361 SUMMARY MOST important disease affecting tomatoes grown in Illinois JL greenhouses is Fusarium wilt. In an effort to overcome this difficulty growers have resorted to the use of resistant field varieties, but these have not proved entirely successful under glass. For this purpose Marglobe is more widely used than any other resistant va- riety and seems to be the one best adapted. In the cloudy weather and short days of late fall and early winter, Marglobe, however, fails to pollinate well and the yields are often seriously reduced. The authors have attempted to overcome these defects by breed- ing wilt-resistant varieties that are especially adapted to greenhouse conditions. Good results have been obtained by combining certain varieties that are wilt-resistant with the Grand Rapids Forcing va- riety, which is noted for its ability to blossom freely and to set fruits under adverse conditions. The outcome of these investigations is the development of two distinct varieties that are ready for commercial use and have been named BLAIR FORCING and LLOYD FORCING by the investi- gators. Evidence has been presented herein which shows that both these varieties are wilt-resistant and have the ability to fertilize and mature fruits in late fall and in winter under conditions where yields of resistant field varieties are seriously impaired. The yields of spring and early summer crops have also compared very favorably with those of the best field varieties. In addition to their yielding ability Blair Forcing and Lloyd Forcing have a desirable type and size of fruit which has proved acceptable in several markets. These varieties are also characterized by earliness of maturity and a picking season that is relatively short as compared with Marglobe. The authors conclude from these investigations and from obser- vations on several commercially grown crops that Blair Forcing and Lloyd Forcing tomatoes may be used very profitably by greenhouse growers. Urbana, Illinois December, 1930 Publications in the Bulletin series report the results of investigations made or sponsored by the Experiment Station Breeding Two New Varieties of Greenhouse Tomatoes Resistant to Fusarium Wilt By WALTER A. HUELSEN, Associate Chief in Olericulture, and MERL C. GILLIS, Associate "^OMATOES, lettuce, and cucumbers are undoubtedly the three major vegetable crops grown in Illinois greenhouses. Many growers consider tomatoes of greater importance than the two other crops and at the same time considerably more profitable owing to the fact that they seem able to compete much more successfully with the southern-grown product. Many growers are now raising both fall and spring crops of tomatoes in the same house. Such in- tensification has naturally brought to the fore a number of problems which those interested in the business wish to have solved. The increasing occurrence of tomato wilt (Fusarium lycopersici Sacc.) has caused considerable concern. Such favorite varieties as Globe, Bonny Best, Grand Rapids Forcing, Comet, and others, to- gether with a host of strains selected for years by individual green- housemen, succumb more or less rapidly to the wilt disease, especially in the spring on account of higher soil temperature then. Growers have attacked this problem of their own accord in two general ways: they have sterilized their soils with live steam by a number of different methods of varying efficiency; and they have attempted to use resistant varieties, hoping to avoid the necessity of steam sterilization. Sterilization usually affords protection for only a single crop. While theoretically the organism should be eliminated entirely by this method, the fact is that sterilization is scarcely ever complete, and reinfestation is therefore only a matter of time. The chief objections to steam sterilization are usually the lack of equipment necessary to do it thoroly and the expense involved. Scarcely any growers have been satisfied with the results they have had in trying to use resistant varieties of tomatoes. Marglobe, de- veloped by Dr. F. J. Pritchard of the U. S. Department of Agriculture, has been used most frequently, but it often sets fruits rather poorly under greenhouse conditions, especially in late fall, and ripens slowly. Altho meeting with marked success outdoors, Marglobe is not entirely suitable for greenhouse use under conditions in Illinois. 409 410 BUU.KTIN No. 361 [December, PURPOSE OF EXPERIMENTS The investigations reported herein were confined to an attempt' to breed a variety of tomato that would be resistant to wilt and at the same time pollinate readily under the conditions prevailing in the greenhouse. There have always been a large number of good green- house varieties which set readily, but all of them are more or less sus- ceptible to wilt. Before embarking upon a program of this kind, pre- liminary experiments were planned in order to determine the yielding ability under glass of some of the more promising wilt-resistant varie- ties. These data were used as a guide in subsequent crossing and selec- tion. A great deal of excellent tomato breeding work in the direction of wilt-resistance has been accomplished, but this is not particularly per- tinent here because it was directed toward the breeding of field varie- ties. Only one wilt-resistant variety especially bred for greenhouse pur- poses is known to the authors. This variety is the Invincible, bred by McWhorter and Parker 11 ' 12 * from a cross between Marvel and Bonny Best. PRELIMINARY TESTS OF PROMISING VARIETIES Materials. The early investigations (1926) consisted of a com- parison in the greenhouse of selections from the more promising wilt- resistant varieties. These selections had been made during the sum- mer of 1925 out of the best-yielding varieties in a field test.* Owing to their good yields and other qualities the resistant varieties, Louis- iana Red, Louisiana Pink, Marglobe, and Marvana, were chosen as being most suitable for the purpose of the experiment. Seed of the two Louisiana varieties had been contributed originally by Dr. C. W. Edgerton of the Louisiana Experiment Station. Marglobe and Mar- vana were furnished by Dr. Fred J. Pritchard of the U. S. Department of Agriculture. Grand Rapids Forcing and Bonny Best were pur- chased from commercial sources. Location of the Experiments. Some of the preliminary experi- ments, together with the breeding experiments proper, were conducted at Urbana in a small greenhouse equipped with raised benches beneath which were located the heating coils. This arrangement provided a high soil temperature. Consequently the wilt became almost always extremely virulent after the experiments had progressed a year. Previ- ous to this the soil had been sterilized before planting each crop. Discussed in 111. Agr. Exp. Sta. 39th Ann. Rpt., 162. 1926. 1930] Two NEW VARIETIES OK GREENHOUSE TOMATOES 411 Physiological Strain of Fusarium. The strain of Fusarium lyco- pcrsici Sacc. came from Union county, Illinois, having been introduced in soil samples used in the course of thesis work by a graduate student. Having once gained a foothold it spread rapidly thruout the green- houses and the Urbana farm. Varieties such as Grand Rapids Forcing, Bonny Best, and Comet nearly always were heavily infected. FIG. 1. LEAF TYPE AND CHARACTERISTIC CLUSTER OF BOTH BLAIR FORCING AND LLOYD FORCING Growing Conditions. The soil was manured and fertilized previous to planting each crop but was not sterilized. In fact the aim was to create conditions as favorable as possible for the organism. Rather careful plot technic was required thruout the experiments in order to determine the resistance of each progeny. The bench green- house mentioned previously is built with its axis running east and west. Therefore, in winter the east half of the house is always considerably warmer than the west half because of prevailing westerly winds. Soil temperatures are known to exert an important influence on the growth of Fusarium and its toxicity to the plant. Clayton 3 * has shown a temperature of 28 C. to be most favorable for the development of the organism, while Haymaker"* states that the excretory products from cultures grown at 28 C. are more toxic to the plant than those 412 BULLETIN No. 361 [December, produced at other temperatures. The maintenance of a soil tempera- ture of 28 C. was desirable, therefore, from the standpoint of deter- mining relative wilt-resistance. This proved to be impossible except at the eastern end of the greenhouse. Accordingly, the plots were always distributed with this point in view. Unfortunately, however, during warm, sunshiny days, the soil temperatures at the eastern end of the FIG. 2. LEAF TYPE AND CHARACTERISTIC CLUSTER OF MARGLOBE This photograph was taken on the same date as Fig. 1. Note difference in maturity. greenhouses often rose above 30 C., thus causing an unbalanced physiological condition in the plants, which rendered them more sus- ceptible to the action of the toxic products mentioned by Haymaker. 6 * Apparently this was the reason most of the resistant plants mentioned later succumbed to wilt at the eastern ends of the benches but proved entirely resistant everywhere else. However, special attention was always given plants surviving at the eastern ends of the benches. Arrangement of Plots and Technic. Duplicate plots containing 10 plants each were set in the greenhouse benches in March, 1926. The plants in this bench test were spaced 18 by 20 inches. The soil was infested with wilt, as already described. A check plot of Bonny Best, a variety very susceptible to wilt, was planted adjacent to each resist- ant selection. 1930] Two NEW VARIETIES OF GREENHOUSE TOMATOES 413 The plants in this and in all the later bench experiments were trained to a single stem and headed back at 70 inches, the maximum permissible owing to the structure of the house. Plants as short as these account for the somewhat low yields. Method of Pollination. Bouquet 1 ' 2 * and Schneck 14 * have shown that hand methods of pollination are greatly superior to any other, but of course involve considerably greater expense. In considering the method of pollination to use in these experiments, the authors be- lieved that simply jarring the plants would demonstrate much better than any other method the inherent fertilizing ability of each strain. Accordingly, the plants were simply tapped lightly each day with a stick. Method of Grading. The fruits were picked when ripe and di- vided into three grades. "Firsts" consisted of the largest and smooth- est tomatoes free from cracks and blemishes. "Seconds" were usu- ally somewhat smaller than firsts but there were also included in this grade large-sized fruits which were blemished, cracked, or irregular. "Culls" consisted of unmarketable fruits, chiefly so because of small size. In some of the later experiments "seconds" and "culls" were combined. Thruout this bulletin, where yields are mentioned the weights of "firsts" are meant unless other grades are stated. Results of First Preliminary Experiment The data for the crop grown in the spring of 1926 are given in Table 1. The susceptible variety, Bonny Best, outyielded all the re- sistant varieties except Marglobe in spite of a severe infection of wilt. In this instance Bonny Best matured the major part of its crop before the wilt disease curtailed the yield very severely. This also occurs outdoors, as shown by Edgerton 4 * in the case of very early varieties such as Earliana and June Pink. Bonny Best, however, is usually somewhat later than these varieties and the grower cannot always be certain of securing a good yield before the plant succumbs to the disease. This hazard 'makes Bonny Best an unreliable variety for the grower to use in wilt-infested greenhouses. With the exception of Marvana the fruits of the different varie- ties did not differ very much in size (Table 2). Marglobe produced the largest percentage of firsts and the smallest percentage of seconds and culls. Louisiana Red and Marvana produced the smallest per- centage of firsts and the largest of seconds and culls. The most de- sirable varieties, according to the results shown in Table 2, were Mar- globe, Louisiana Pink, and Bonny Best. 414 BULLETIN No. 361 [December, TABLE 1. MEAN YIELDS OF FOUR WILT-RESISTANT VARIETIES OF TOMATOES COM- PARED WITH ADJACENT CHECK PLOTS OF SUSCEPTIBLE BONNY BEST: SPRING OF 1926 (Calculated to a basis of 10 plants) Variety Firsts Seconds Culls Louisiana Pink No. 70 93 -23 68 92 -24 101 66 35 82 90 -8 oz. 376 503 -127 352 504 -152 580 378 202 392 538 -146 No. 24 42 -18 34 34 30 25 5 45 36 9 OS. 85 156 -71 112 128 -16 93 102 -9 157 151 6 No. 12 30 -18 24 13 11 10 10 31 20 11 oz. 36 64 -28 58 34 24 20 26 -6 68 60 8 Bonny Best check .... Increase over check. Bonny Best check. . . . Increase over check. Marglobe Bonny Best check .... I ncrease over check . Bonny Best check. . . . Increase over check. TABLE 2. MEAN SIZE OF FRUITS AND PERCENTAGE OF EACH GRADE OBTAINED FROM FOUR WILT-RESISTANT VARIETIES OF TOMATOES: SPRING OF 1926 Variety Mean size of fruits Percentage of total yield in each grade Firsts Seconds Culls Firsts Seconds Culls oz. 5.4 5.2 5.7 4.8 5.5 oz. 3.5 3.3 3.2 3.5 3.9 02. 3.2 2.4 1.9 2.2 2.6 75.6 67.4 83.6 63.5 73.9 17.1 21.4 13.4 25.4 19.4 7.3 11.2 2.9 11.1 6.7 Marvana Bonny Best checks. . . . TABLE 3. PERCENTAGE OF TOTAL CROP OF FIRST-GRADE FRUITS MATURED BY FOUR WILT-RESISTANT VARIETIES OF TOMATOES DURING HARVEST OF SPRING CROP OF 1926 1 Variety May 11-20 May 21-30 May 31- June9 June 10-19 June 20-29 June 30- July9 July 10-19 Louisiana Pink. . . Louisiana Red . . . Marglobe 2.7 8.2 2.0 17.5 23.0 8.6 40.8 32.6 36.0 22.6 20.3 21.9 12.3 13.6 20.2 2.6 1.6 3.8 1.5 .7 7.5 Marvana 1.4 23.9 44.1 15.5 13.1 1.0 1.0 Bonny Bestchecks 2.9 20.4 30.7 20.7 14.4 7.0 3.9 'Total percentage harvested =100. Marvana was the earliest variety, having produced 84.9 percent of the total crop up to June 19 (Table 3). Louisiana Red and Pink were both on a par and practically as early as Marvana. Bonny Best was somewhat earlier than Marglobe. Results of Second Preliminary Experiment A second and more extensive experiment was started in the summer of 1926. Seed was sown July 20 and transplanted to the houses Sep- 1930] Two NEW VARIETIES OF GREENHOUSE TOMATOES 415 tember 1. One series was planted in a ground-bed house used for fruit-breeding work, where the soil was free from wilt. A second series was planted in the same benches as the first experiment, the soil not being sterilized. Each series was replicated four times and planted in such a way that a Grand Rapids Forcing check was ad- jacent to one side and a Bonny Best check to the other. The series on infested soil was planted 22 by 25 inches and the wilt-free series 24 by 26 inches. Each plant was pruned to a single stem. TABLE 4. MEAN YIELDS OF THREE WILT-RESISTANT VARIETIES OF TOMATOES AND Two SUSCEPTIBLE VARIETIES, ON INFESTED AND ON WILT-FREE SOILS Variety Weight of fruit per plant Number of fruits per plant Firsts J Seconds Culls | Total Firsts Seconds Culls | Total Fall crop of 1926 on wilt-infested soil oz. ot. OS. OS. Louisiana Pink .... 38 8 4 6 1 44 4 9 6 2 1 12 6 Marvana 31 8 4 1.4 37 2 9 2 1 6 g 11 6 Marglobe 46.6 3.3 .8 50 7 10 6 1 4 6 12 6 Bonny Best checks .... 23.5 4.0 1.7 29.2 6.4 1.6 .9 8.9 Grand Rapids checks. . . 31.0 10.1 5.3 46.4 11.7 5.2 4.5 21.4 Fall crop of 1926 on wilt-free soil Louisiana Pink 48 5 9 9 50 3 11 4 4 5 12 3 Marvana 32 6.7 3 8 42 5 9 2 2 3 2 6 14 1 Marglobe 41.1 2.8 .9 44.8 10 1.4 .8 12 2 Bonny Best checks .... 41.8 5.2 2.1 49.1 11.0 2.0 1.2 14.2 Grand Rapids checks. . . 35.0 11.6 7.2 53.8 14.2 6.4 6.4 27.0 On infested soil Bonny Best gave the lowest and Marglobe the highest yield (Table 4). On wilt-free soil Marglobe and Bonny Best yields were practically the same, and the yield of Lousiana Pink was considerably higher. This increase in the yield of Louisiana Pink was due to the fact that in the wilt-free series ground beds were used and the plants were headed at a height of 7% feet. Louisiana Pink, with a long, straggly vine, accordingly yielded much better than on the benches where it was headed at 70 inches. Marvana, a short-growing variety, was not affected at all, and gave practically equal yields in both houses. CHARACTERISTICS OF VARIETIES SELECTED FOR BREEDING The 1926 tests enabled the authors to make a fairly comprehensive study of the characteristics of each variety under greenhouse condi- tions. 416 BULLETIN No. 361 [December, Grand Rapids Forcing was outstanding in the number of tomatoes matured. This variety on wilt-infested soil matured a total of 21.4 fruits per plant, compared with 12.6 per plant in the case of Louisiana Pink and Marglobe, the two next highest. On wilt-free soil the differ- ences were just as large (Table 4). In size of fruits Grand Rapids Forcing was much smaller than any other variety (Table 6). Louisi- ana Pink produced the largest fruits in the wilt- free series and Mar- globe ranked first in the wilt-infested series. Bonny Best and Marvana were somewhat mediocre in performance. TABLE 5. PERCENTAGE OF TOTAL CROP MATURED BY THREE WILT-RESISTANT VARIETIES OF TOMATOES IN EACH FIFTEEN-DAY PERIOD DURING HARVEST OF FALL CROP OF 1926 1 Variety Oct. 15-30 Nov. 1-15 Nov. 16-30 Dec. 1-15 Dec. 16-31 Jan. 1-15 Fall crop of 1926 on wilt-infested soil Louisiana Pink 5 12 31 29 21 2 Marvana 12 24 26 23 15 4 23 26 13 28 6 Bonny Best checks 26 26 27 9 11 1 Grand Rapids checks 7 26 30 20 13 4 Fall crop of 1926 on wilt-free soil Louisiana Pink 1 8 21 41 28 1 Marvana 3 21 19 22 33 2 Marglobe 2 15 25 23 31 4 Bonny Best checks 6 19 32 19 23 1 Grand Rapids checks 1 16 33 20 25 5 l Total percentage harvested = 100. TABLE 6. MEAN SIZE OF FRUITS OF EACH GRADE HARVESTED FROM EACH VARIETY OF TOMATO GROWN IN FALL OF 1926 Variety On wilt-infested soil On wilt-free soil Firsts Seconds Culls Firsts Seconds Culls Louisiana Pink 02. 4.05 3.46 4.40 3.67 2.65 oz. 2.30 2.50 2.36 2.50 1.96 oz. 1.00 1.75 1.33 1.89 1.18 oz. 4.25 3.48 4.11 3.80 2.46 Of. 2.25 2.91 2.00 2.60 1.81 ot. 1.80 1.46 1.12 1.75 1.12 Marvana Marglobe Bonny Best checks Grand Rapids checks In addition to producing small fruits, Grand Rapids Forcing proved to have a rather small, slow-growing vine. It was exceeded in earliness only by Bonny Best in both series (Table 5). Marvana equaled Grand Rapids Forcing in earliness in one series (Table 5) and was slightly later in the other. In yield the two vari- eties were about the same (Table 4), but in mean size of fruits Mar- vana was larger (Table 6). 1930] Two NEW VARIETIES OF GREENHOUSE TOMATOES 417 Marglobe is too well known to require much comment. It yielded well under all conditions and was somewhat earlier than Louisiana Pink. In size of fruit it was about the same as Louisiana Pink. Mar- globe set an average number of fruits. Grand Rapids Forcing, Marvana, and Marglobe are orange-red fruited while Louisiana Pink has a colorless epidermis of the type described by Lindstrom 8 ' 9 * and red flesh. For that reason it gives the illusion of being pink, while in reality this difference in color is due to the absence in the epidermis of the carotinoid pigment described by Howard. 7 * Louisiana Pink is a rapid-growing variety with a long, straggly vine, the clusters being rather far apart. It is always the first variety to require heading. It is slightly later than Marglobe and has good-sized fruit. It is only average in the number of fruits matured. Louisiana Pink, Marvana, and Marglobe are all wilt-resistant and Grand Rapids Forcing partially so. Bonny Best is susceptible, and for that reason was discarded as being unsuitable for the breeding tests. Louisiana Pink, Marvana, and Grand Rapids Forcing are all more or less oblate in shape. On the other hand, Marglobe is deeply oblate to spherical. In the crosses which were made as described later, the aim was to self in the F 2 generation plants with spherical fruits similar to Marglobe, as this shape is very popular. This proved to be rather difficult owing to the fact that in crosses of oblate by spherical types, the spherical shows a lack of dominance, according to Lindstrom. 10 * CROSSES MADE Six crosses were made on the crop grown in infested soil during the fall of 1926 as follows: No. 1000 Marglobe Strain 1-2 X Grand Rapids Forcing No. 1001 Marglobe Strain 1-2 X Grand Rapids Forcing No. 1002 Marvana Strain 3-17 X Grand Rapids Forcing No. 1003 Marvana Strain 3-17 X Grand Rapids Forcing No. 1004 Louisiana Pink Strain 10-63 X Grand Rapids Forcing No. 1005 Louisiana Pink Strain 10-63 X Grand Rapids Forcing Grand Rapids Forcing was used as the male parent in all the crosses. The three resistant varieties were not intercrossed, as none of them possessed the desirable character of fertilizing well in the greenhouse. The F! progenies were grown on the benches in replicated plots during the spring of 1927. A large number of plants were self-polli- nated but seeds were saved only from those which appeared to be most resistant to wilt. 418 BULLETIN No. 361 [December, A large number of F 2 progenies were grown in the fall of 1927. Yield records consisting of number and weight of fruits and the per- centage of defective fruits were taken. As expected, there was con- siderable segregation both as to resistance and as to horicultural char- acters. Two of the most promising individuals were selected out of each cross on the basis of yield, small percentage of defective fruits, size of fruits, uniformity, and general appearance. The F 3 and F 4 generations were planted in the same greenhouse in the spring and fall, respectively, of 1928, and the procedure fol- lowed was in all respects identical with that described for the F 2 progeny. The F 5 generation planted in the spring of 1929 was tested more elaborately, and copious notes were kept. This test was supplemented by a strain test on a much larger scale, planted in ground beds in an- other greenhouse where the soil infestation was considerably milder than in the bench greenhouse. The data from these tests are not given here, being of value mainly in the work of selection. It will be suffi- cient to say that the F 5 inbred lines performed very creditably. The F 6 generation was planted in the fall of 1929 on the same benches and selected as described above. Selection work was continued with the F 7 generation in the spring of 1930 and in addition, crosses were made between the more promising lines. TWO PROMISING VARIETIES DEVELOPED Lloyd Forcing The first strain to show considerable promise was a selection out of cross No. 1004 (Louisiana Pink and Grand Rapids Forcing). This cross was still segregating for pink fruits in the F 3 generation, but a red-fruited selection continued to breed true for color in the F 4 , F 5 , F 6 , and F 7 generations. Since this selection is quite different in its characters from other known varieties, the authors have named it LLOYD FORCING in honor of Dr. J. W. Lloyd, of the Department of Horticulture at this Station. Lloyd Forcing is a rapid and vigorous grower having a large vine of the Louisiana Pink type. It sets blossoms very freely and has the pollination characteristics of Grand Rapids Forcing. The fruit is ob- late, smooth, and solid. It keeps very well and has not cracked in any of the greenhouse tests. The color is a solid red without green streaks at the stem end, nor is it subject to yellow blotches. It clings to the vine unusually well and is rather difficult to pick, probably owing to 1930] Two NEW VARIETIES OF GREENHOUSE TOMATOES 419 the fact that the abscission layer in the pedicel does not develop suffi- ciently early. This character prevents the premature dropping of partly ripened fruit, which is so annoying to the grower, but on the other hand makes it difficult to preserve the calyx still attached to the fruit. To some growers this might be a serious objection if they desire to market the fruit with calyx attached, as demanded by several markets. FIG. 3. RELATIVE GROWTH OF BLAIR FORCING AND LLOYD FORCING BEFORE HEADING AS COMPARED WITH MARGLOBE The two smaller rows in the center are Marglobe. The taller rows on the left and right are respectively Blair and Lloyd Forcing. Lloyd Forcing is as wilt-resistant as Marglobe under the green- house conditions of these experiments. It is not particularly suscep- tible to the physiological condition known as blossom end rot. Lloyd Forcing was tested in a mildly infested ground-bed green- house in the fall of 1929. The results are given in Table 7. No at- tempts were made to pollinate other than to tap the plants once a day. Lloyd Forcing yielded more than twice as much as Grand Rapids Forcing and 50 percent more than Marglobe. In average size of fruit it was slightly larger than Marglobe. According to the data in Table 8, Lloyd Forcing is a trifle earlier than Marglobe and somewhat later than Grand Rapids Forcing. A statistical study in some respects like Myers' 13 * was made of the Lloyd Forcing variety from the crop planted in the fall of 1929. The means, differences, and coefficients of variability are shown in Table 9. 420 BULLETIN No. 361 {December, TABLE 7. YIELD OF LLOYD FORCING IN FALL OF 1929 COMPARED WITH YIELDS OF Two OTHER VARIETIES 1 Picking season in 10-day periods Yield of marketable tomatoes per plant Grand Rapids Forcing Lloyd Forcing Livingston's Marglobe Lloyd Forcing Nov. 5-14 OS. 3.15 11.42 12.56 .86 3.73 3.83 2.79 38.34 2.10 or. 1.80 14.74 31.28 4.31 7.96 11.49 6.44 78.02 3.07 01. 2.35 10.42 12.44 3.97 7.20 6.94 6.94 50.26 2.94 01. 1.74 15.96 23.64 10.92 9.64 11.41 6.29 79.60 3.26 Nov. 15-24 Nov. 25-Dec. 4 Dec. 5-14 Dec. 15-24 Dec. 25-Jan. 3 Jan. 4-13 Total yield per plant Average size of fruits 'A plot of Lloyd Forcing was planted on each side of the Grand Rapids Forcing and the Living- ston's Marglobe plots. Each plot contained 66 plants trained to single stems and planted 32 by 16 inches on soil infested with Fusarium wilt. The means of the Lloyd Forcing plots include 132 plants and those of the other two varieties 66 plants each. TABLE 8. RELATIVE MATURITY OF DIFFERENT VARIETIES OF TOMATOES IN YIELD TEST DURING FALL OF 1929 (Percentage of total crop picked in each 10-day period) Picking season in 10-day periods Grand Rapids Forcing Lloyd Forcing Livingston's Marglobe Lloyd Forcing Nov. 5-14 8.22 2 31 4 68 2 19 Nov. 15-24 29.79 18.89 20.73 20.05 Nov. 25-Dec. 4 32 76 40 09 24 75 29 70 Dec. 5-14 2.24 5 52 7 90 13 72 Dec. 15-24 9.73 10.20 14 33 12.11 Dec. 25-Jan. 3 9.99 14.73 13.81 14.33 Jan. 4-13 7.28 8.25 13.81 7.90 Lloyd Forcing is taller and grows more vigorously and rapidly than either Grand Rapids Forcing or Marglobe, as shown by comparing the respective mean heights before pruning. Lloyd Forcing is also some- what more variable in height than the other two varieties, as shown by the slightly larger coefficient of variability. Grand Rapids Forcing sets the first cluster nearest the ground, as may be noted by comparing the means of the distances between the ground and the first cluster. Marglobe is intermediate in this respect, and Lloyd Forcing sets the first cluster much higher. At the same time, Lloyd Forcing is much more variable than either of the other varieties. A comparison of the stem diameters just above the second cluster shows Grand Rapids to be smallest, Lloyd Forcing intermediate, and Marglobe largest. The extreme mean difference, however, is less than 2 mm. In variability Lloyd Forcing is again the highest. In Grand Rapids the clusters were much closer together. Marglobe and Lloyd Forcing were practically the same. Lloyd Forcing, however, 1930] Two NEW VARIETIES OF GREENHOUSE TOMATOES 421 o i o (x* II U o gfc O tn <: Q 0, oo^io o to v) o i/) o & o 10 o o* to I <*> OOirO OW(N OOO O"5* lOf^ IO W <> WOv w ' t) (NfSOO f)OOO rr> ~* wj IO1OO OOiOOO OtOOv 00 - -H+I-4H HH+I-H -H-H-H -H+I-H -H-H-H -H-H-H (SfN'f OvOOt^ WIO^O O O> O> >O"tO >5 > 'M :-oc-~ i-oc-^ .-oc -'a'O J -'a'0.2 -'a'O 422 BULLETIN No. 361 [December, was much less variable in this respect, as shown by the small coeffi- cient of variability. Grand Rapids had a mean of 6.71 fruits per cluster and Lloyd Forcing, 5.48. Marglobe set fewer fruits, having a mean of only 3.53 per cluster. The coefficients of variability are essentially the same. FIG. 4. FRUIT OF BLAIR FORCING AND LLOYD FORCING Both varieties are smooth and slightly flattened in shape. In cross-section they are solid, with rinds sufficiently thick to withstand shipment. The longi- tudinal sections show their solidity and small cores. The ruler shows the com- parative size of the varieties. Grand Rapids averaged 44.24 fruits per plant, Lloyd Forcing 36.00 and Marglobe only 21.59. The coefficient of variability for Lloyd Forcing is somewhat higher than for the other two varieties. In general Lloyd Forcing proved to be somewhat more variable than either Marglobe or Grand Rapids. Marglobe was more variable 1930] Two NEW VARIETIES OF GREENHOUSE TOMATOES 423 than Grand Rapids but the differences lacked significance. The greater variation in Lloyd Forcing and Marglobe is most likely due to the hy- brid origin of the two varieties. Lloyd Forcing was more variable than Marglobe in only two out of five observations, but as compared with Grand Rapids it proved to have a larger variation four times out of Blair Forcing This variety originated from the same cross as Lloyd Forcing and is a pink selection made in the F 3 generation. It has bred true for pink fruit color in the F 4 , F 8 , F e , and F 7 generations. It is essenti- ally the same type as Lloyd Forcing in vine and general fruit charac- ters, but owing to the fact that its color is pink compared with the red of Lloyd Forcing, the authors have decided that it is sufficiently dis- tinct to be introduced as a separate variety. It has been named BLAIR FORCING in honor of Dr. J. C. Blair, head of the Department of Horticulture at this Station. Blair Forcing has the same tall-growing vine as Lloyd Forcing and sets fruits just as readily. A comparison of the performance of the two varieties in the bench greenhouse on soil heavily infested with Fusarium wilt is shown in Table 10. This crop was planted in the fall of 1929 and the plots were replicated in three series. In all three series Blair Forcing showed a slight tendency to pro- duce a greater number of fruits per plant. The odds, however, are too small to be of any significance. In total weight of fruit produced per plant, Blair Forcing outyielded Lloyd Forcing slightly in two of the three series. The average size of the fruits of Blair Forcing was slightly larger than Lloyd Forcing twice out of three times. In wilt-resistance Blair Forcing proved appreciably better than Lloyd Forcing. The results of the wilt counts w r ere as follows: Series 1 Blair Forcing, 16 percent wilted - Lloyd Forcing, 67 percent wilted (50 percent of these died) Series 2 Blair Forcing, 12 percent wilted Lloyd Forcing, 25 percent wilted (25 percent of these died) Series 3 Blair Forcing, 25 percent wilted Lloyd Forcing, 38 percent wilted None of the Blair Forcing plants, it should be noted, were infected heavily enough to cause the death of the plant, but in the case of Lloyd Forcing a considerable number died. These results should not lead to the conclusion that these varieties have a low resistance. As a matter 424 BULLETIN No. 361 [December, . |:J- : 2: 6 -** r. . . w a 0 ^0 000 5 00 fSCS t^ (N CN (N (M _ _ _ _ - o, 22 - - 52 2= 22 1 k f 1 Sfc e e f f 11 [U. ou, ou. u-g; u-g jj-o 'So 'a o 'So x_ ^_ SZj 1 ( 1 K- 00 fMO OOfS l^O 10 O uits per pla veight of frt ">o (MO\ ^"m (sr^ (Nt^ OOO weight of f **i **) oo cs o O 3 'c a a n Average (NW5 ^O 00 OMO f*5 CN O O (N1O 1~- M O ! ^fX OiO Blair Forcing Lloyd Forcing Blair Forcing Lloyd Forcing Blair Forcing Lloyd Forcing if II II fe o fe o b o U [z U< .* *^ (S *> 1930] Two NEW VARIETIES OF GREENHOUSE TOMATOES 425 of fact the authors have been unable to find a single variety that will withstand successfully the severe wilt conditions to be found in the bottom-heated benches. FIG. 5. A SEVERE ATTACK OF FUSARIUM WILT The center plant shows a typical infection on the greenhouse benches, tho the plants on each side are healthy. A comparison between Blair Forcing and Lloyd Forcing as to earliness of maturity is shown in Table 11. Blair Forcing is obviously somewhat earlier than Lloyd Forcing. The differences between the three series in time of maturity were due to the uneven distribution TABLE 11. COMPARATIVE EARLINESS OF BLAIR FORCING AND LLOYD FORCING TOMATOES: FALL OF 1929 (Percentage of total crop matured at each picking 1 ) Date of picking Series 1 Series 2 Series 3 Blair Lloyd Blair Lloyd Blair Lloyd Nov. 18. .. 9.78 13.25 11.43 18.18 23.91 15.18 8.27 ii!46 10 46 7.61 13.67 13.43 26.28 17.16 is! 58 6.52 17.17 17.74 23.99 19.00 n '<..' 1.81 15.83 16.60 26.77 27.38 '7!96 5.52 13.20 30.63 15.93 26.76 '&.3S 4.43 18.74 16.52 29.81 22.13 Nov. 23 Nov. 29 Dec. 3 Dec. 9 Dec. 16 Dec. 23 Dec. 30 l Total percentage harvested - 100. 426 BULLETIN No. 361 [December, Q H a So as o O ! Q O o 28 5 o o go SS3<* * 3$2oo = 1 - - PN -"< 1 1 U -ovrsto POOO-O-H Wr-JOCS C 8 < IT> 1 - -17 """2 5 sssjlrt- SS-o- SwSso- *3- - . ~sss" ^^S2? O ^* ^i 1 ^ ^ 52 2 "C B ^ '- SS=S ss^* 3S2S p^POOO - POlO = 5? m *} o> t- ~ f> 8 fc -J, (S -OTj- 1 1 oo 1 00 t^ SS^- oo u- OOiOPO t-^vO'* 1 Q O ^ f"* f> X ^l >C u". 255^ S25w5 '- i i - o - 33"^ SS*oo f 8 T 00 S 7f " S * sss SSS-^ Saoo^. I -HVOOO o oo OO-HOOO 000 0< 2 SS! l/> ^ ^^ u. o . . POCS fS sass 1 1 } f 1 Lloyd Forcing Grand Rapids Check . Increase over check Percent increase. . Odds . Lloyd Forcing Marglobe check Increase over check Percent increase . . . Odds . . Blair Forcing Grand Rapids Check. Increase over check Percent increase. . . Odds. . Blair Forcing Marglobe check Increase over check Percent increase . . . Odds 1930] Two NEW VARIETIES OF GREENHOUSE TOMATOES 427 of the heat, as explained before. Series 1, the earliest, was located at the eastern end of the house and Series 3 farthest west. In solidity of fruit, evenness of ripening, freedom from cracking and smoothness, Blair Forcing is very much like Lloyd Forcing. Blair Forcing, however, appears to develop an abscission zone somewhat earlier, thus facilitating picking, but this development is not so pre- mature that the fruits drop off before they are picked. TESTS OF NEW VARIETIES IN SPRINGBOK 1930 Blair Forcing and Lloyd Forcing were given a final test in com- parison with Grand Rapids Forcing and Marglobe in the spring of 1930. Another selection, not heretofore mentioned, was included in the test. This is designated as No. 1001 because it has been selected out of that cross number (Marglobe X Grand Rapids). Owing to its high yield and globe-shaped type No. 1001 may eventually prove very desirable, but it does not as yet reach the standard required. TABLE 13. BLAIR AND LLOYD FORCING COMPARED WITH MARGLOBE AND GRAND RAPIDS FORCING IN SIZE OF FRUITS CLASSED AS FIRSTS AND IN RATIO OF GREEN FRUITS LEFT ON VINES AFTER PICKING TO TOTAL YIELD OF RIPE FIRSTS (BY WEIGHT): SPRING CROP OF 1930 Series 1 Series 2 Series 3 Mean Variety Mean Ratio Mean Ratio Mean Ratio Mean Ratio size of fruits green fruits to ripe size of fruits green fruits to ripe size of fruits green fruits to ripe size of fruits green fruits to ripe OS. OS. oz. ot. Lloyd Forcing. Grand Rapids. 4.37 3.26 1:67.6 1:43.2 4.31 3.06 1:26.3 1:14.4 4.34 3.28 1:16.5 1:13.6 4.25 3.20 1.05 1:36.8 1:23.7 1:13.1 Lloyd Forcing. Marglobe 4.37 5.22 1:67.6 1:15.2 4.31 4.90 1:26.3 1: 9.5 4.34 5.35 1:16.5 1: 8.4 4.25 5.16 1:36.8 1:11.0 .91 1:25.8 Blair Forcing.. Grand Rapids. 3.72 3.34 1:79.1 1:19.3 4.03 3.28 1:62.7 1:13.6 3.74 3.04 1:39.8 1:12.2 3.83 3.22 1:60.5 1:15.0 .61 1:45.5 Blair Forcing.. Marglobe 3.72 4.90 1:79.1 1: 9.5 4.03 5.42 1:62.7 1: 8.1 3.74 5.16 1:39.8 1: 6.8 3.83 5.16 1:60.5 1: 8.1 Increase 1.3.J 1:52.4 The plot layout is shown diagrammatically in Table 14. There were 32 plants in each plot and three replications of plots, which are called series in Tables 12 and 13. Blair Forcing and Lloyd Forcing were planted so that a Marglobe check was adjacent to one side and a Grand Rapids check adjacent to the other. The yield data appear in Table 12. 428 BULLETIN No. 361 [December, Superiority of New Varieties Shown by High Spring Yields Both Blair Forcing and Lloyd Forcing outyielded Grand Rapids, the increases being significant according to the odds, which were cal- culated on the basis of Student's method (Table 12). Blair Forcing and Lloyd Forcing gave only a slight increase in yield over Marglobe. FIG. 6. GRAND RAPIDS FORCING ON THE LEFT AND LLOYD FORCING ON THE RIGHT These plants were adjacent to each other in the tests made during the spring of 1930. It is interesting to note that both these varieties also produced a larger number of fruits than either Grand Rapids or Marglobe. Both Blair Forcing and Lloyd Forcing produced fewer culls than Grand Rapids and more than Marglobe. The average weights of the fruits which were classed as "firsts" are shown in Table 13. The fruits of Blair Forcing proved to be .61 ounce larger than Grand Rapids and 1.33 ounces smaller than Marglobe. The Lloyd Forcing fruits were 1.05 ounces larger than 1930] Two NEW VARIETIES OF GREENHOUSE TOMATOES 429 Grand Rapids and .91 ounce smaller than Marglobe. Thus both Blair Forcing and Lloyd Forcing proved to be intermediate in size between Marglobe and Grand Rapids. However, Blair Forcing is somewhat smaller than Lloyd Forcing. New Varieties Superior for Fall Forcing The performance of a variety in the spring is by no means a cri- terion of its yields in the fall. Lloyd Forcing, however, proved to be considerably more consistent in its performance in the fall and spring than either Marglobe or Grand Rapids. The following results are taken from Tables 7, 12, and 13. Mean size of fruits Fall Spring oz. oz. Lloyd Forcing 3.16 4.25 Grand Rapids 2.10 3.21 Marglobe 2.94 5.16 Mean yield per plant Fall Spring oz. oz. Lloyd Forcing 78.81 171.39 Grand Rapids 38.34 110.32 Marglobe 50.26 144.42 Marglobe fruits in the fall proved to be approximately one-half as large as the fruits of the spring crop. Grand Rapids fruits were about two-thirds as large. On the other hand, the fall fruits of Lloyd Forcing were 75 percent as large as the spring fruits. In mean yield per plant Marglobe and Grand Rapids produced only about one-third as large a yield in the fall as in the spring. In contrast, the fall yield of Lloyd Forcing was nearly 50 percent as large as the yield obtained in the spring. On the basis of these comparisons it is evident that Lloyd Forcing is much better suited to year-round growth in the greenhouse than either Marglobe or Grand Rapids. Since Blair Forcing, both in the fall and in the spring, yielded about the same as Lloyd Forcing (Tables 10 and 12), it may be assumed that the two varieties are equally suitable for fall and spring forcing in the greenhouse. New Varieties Mature Early The 1930 spring yields of Blair Forcing have been plotted in Fig. 7 in terms of percentage of the total crop picked in each weekly period. Lloyd Forcing has been plotted in Fig. 8. Both Blair Forcing and Lloyd Forcing are strikingly similar to 430 BULLETIN No. 361 [December, Grand Rapids in earliness of maturity. Marglobe lagged behind con- sistently until the week of June 2 to 7, but between June 2 and 14 it gave a larger yield. This is somewhat different from its performance in the fall as shown in Table 8. In the spring, therefore, Blair Forcing PERCENT Et BLAIR FORCING GRAND RAPIDS FORCING 6 - APR. APR. APR.-MAY MAY MAY MAY MAY JUNE. JUNE JUNE JUNE JUKE-JUL H-19 21-26 83 5-10 12-17 19-^ 26-31 2-7 9-14 16*1 ZJ-26 305 DATE OF PICKING FIG. 7. COMPARISON OF PERCENTAGE OF TOTAL CROPS OF BLAIR FORCING, MARGLOBE, AND GRAND RAPIDS MATURING IN WEEKLY PERIODS Blair Forcing proved to be earlier than either Grand Rapids Forcing or Marglobe. and Lloyd Forcing may be expected to mature earlier than Marglobe, thus assuring the grower the advantage of higher prices prevailing early in the season. New Varieties Mature Large Proportion of Fruits Blair Forcing and Lloyd Forcing have the ability to mature a greater proportion of the crop than either Grand Rapids or Marglobe. This is shown by the ratios in Table 13. It is interesting to note that Blair Forcing is more efficient in ma- 1930] Two NEW VARIETIES OF GREENHOUSE TOMATOES 431 turing its crop than Lloyd Forcing. Both varieties, however, are much superior to Grand Rapids in this respect. Marglobe lags very far be- hind. In fact if Marglobe were able to mature its entire crop within a reasonable period, it would probably outyield the others. This lag- PERCENT 22 LLOYD FORCING GRAND RAPIDS FORCING APR. H-19 APR. APR-NAY HAY 21-26 28 J 5-10 MAY HAY MAY JUNE JUNE JUNE JUNE JUNE JULY IE-17 19-24 26-31 2-7 9-14 16-21 23-28 JO 5 DATE: OF PICKING FIG. 8. COMPARISON OF PERCENTAGE OF TOTAL CROPS OF LLOYD FORCING, MARGLOBE, AND GRAND RAPIDS MATURING IN WEEKLY PERIODS Lloyd Forcing proved to be earlier than either Grand Rapids Forcing or Marglobe. ging on the part of Marglobe is objectionable to the grower because field-grown tomatoes begin to compete early in July with those grown under glass, and prices fall very rapidly. Even tho no heat is neces- sary in July, nevertheless the cost of watering, constant pruning, etc., render it questionable whether a variety such as Marglobe would pay as well as one which matures more rapidly. Quick maturity mini- mizes the cost and assures an earlier market. Blair and Lloyd Forcing both approach this ideal much more closely than Marglobe. 432 BULLETIN No. 361 [December, Blair Forcing More Resistant to Wilt A diagram of the occurrence of wilt in the spring-planted plots is shown in Table 14. These records were taken July 14, 1930, after the crop had been removed. Each plant was cut off at the ground level and inspected carefully for the appearance of the blackened vas- cular bundles characteristic in wilted plants. TABLE 14. PLOT LAYOUT OF YIELD TESTS IN SPRING OF 1930 SHOWING OCCURRENCE OF FUSARIUM WILT INFECTION Plot Variety Plot Nos. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 IS 16 Plants infected but not killed Plants dead Total plants infected 1 2 3 Grand Rapids... Lloyd Forcing. . . Marglobe (Condition of plants 1 ) iidiOOOidOOi i ion iOidiOdddO d d i i OiOOdOOOOO i OiOOOOOdi i i i OOOOOOOii i i perct. 37.5 21.9 Perct. 25.0 6.2 percl. 62.5 28.1 4 No. 1001 0000000000 iiOOOOiOOO 12.5 12.5 5 6 7 Grand Rapids... Blair Forcing. . . Marglobe OOOOiOOOOO OOOidOOOOO i i d iiiOiOOdOO i d 0000000000 0000000000 0000000000 i 12.5 25.0 12.5 12.5 37.5 8 9 10 Lloyd Forcing. . . Grand Rapids.. . No. 1001 OOOOOiiiOO d 0000000000 0000000000 OOOOiddiOO i d iOOOOOddOO i i OddiOOOOOO i i d i d 12.5 18.8 3.1 15.6 15.6 34.4 11 Marglobe 0000000000 iOOOiOOOOO 12.5 12.5 25.0 12 13 14 15 Blair Forcing . . . Grand Rapids. . . Lloyd Forcing. . . Marglobe iiiiOiOii OOOOiOOOOO 0000000000 i OiOOddOOOO d 0000000000 i i OOOiOOOOOO OOOOiOOOOO d i d OOdiiiOOOO 28.1 6.2 9.4 9.4 9.4 6.2 28.1 6.2 18.8 15.6 16 No. 1001 OOOOOiOiOO 0000000000 15.6 3.1 18.7 17 18 19 Grand Rapids... Blair Forcing . . . Marglobe 0000000000 idOiOOOOOO iiOOOOOOOO i 0000000000 OOiOiOOOOO iiOOOOOOOO 15.6 6.2 3.1 18.7 6.2 iOOOOOOOOO 9.4 9.4 l i = infected; d = dead; and = healthy. Plants termed "infected" included a rather wide range. Many plants showed no other evidence of wilt than a partially or wholly discolored ring. Others had stem lesions and there was a more or less severe killing of leaves, which apparently did not affect the yields materially. Dead plants were those which were so badly infected that prema- ture death occurred. 1930] Two NEW VARIETIES OF GREENHOUSE TOMATOES 433 The following results summarized from Table 14 indicate that Lloyd Forcing compares very favorably with Marglobe in respect to wilt-resistance and is very superior to Grand Rapids. Lloyd Grand Forcing Rapids Marglobe Percentage of plants infected but not killed 10.4 21.9 13.5 Percentage dead 4.1 16.7 2.1 Total percentage infected 14.5 38 . 6 15.6 Similar data for Blair Forcing are given below: Blair Grand Forcing Rapids Marglobe Percentage of plants infected but not killed 4.1 16.7 16.7 Percentage dead 8.3 1.0 Total percentage infected 4.1 25. 17.7 Marglobe proved much more susceptible to wilt than did Blair Forcing, and Grand Rapids had six times as much infection. It is evident that Blair Forcing is considerably more wilt-resistant than Lloyd Forcing. The data from the bench greenhouse comparing the two varieties already mentioned lead to the same conclusion. It should be understood that the authors do not claim that Blair Forcing and Lloyd Forcing are resistant to all strains of Fusarium wilt. Haymaker 5 * has shown that a considerable difference exists in the pathogenicity of various strains of Fusarium. He points out that there is also a certain selectiveness in plant resistance due to various soil temperatures. Accordingly, all wilt-resistant varieties may be expected to show considerable differences in resistance, depending upon temperatures and the virulence of the local strain of Fusarium. It has been explained previously that the strain of Fusarium with which the greenhouses were infested came from Union county, Illi- nois. Wilt has been present for many years in that section and from the field performance of susceptible strains there, one would infer that the virulence is very high. Accordingly the authors expect both Blair Forcing and Lloyd Forcing to perform well under a wide range of conditions, especially because varieties which are resistant to a more virulent strain will also prove resistant to a lesser one, as shown by Haymaker. 5 * The reverse is not true, however. 434 BULLETIN No. 361 LITERATURE CITED 1. BOUQUET, A. G. B. Pollination of tomatoes. Oreg. Agr. Col. Exp. Sta. Bui. 158. 1919. 2. - - Economic results in the pollination of greenhouse tomatoes. Oreg. Agr. Col. Exp. Sta. Circ. 55. 1924. 3. CLAYTON, E. E. The relation of temperature to the Fusarium wilt of the tomato. Amer. Jour. Bot. 10, 71-88. 1923. 4. EDGERTON, C. W. Tests of the wilt resistance of different tomato varieties. La. State Univ. Agr. Exp. Sta. Bui. 184. 1921. 5. HAYMAKER, H. H. Pathogenicity of two strains of the tomato wilt fungus Fusarium lycopersici, Sacc. Jour. Agr. Res. 36, 675-695. 1928. 6. - Relation of toxic excretory products from two strains of Fu- sarium lycopersici Sacc. to tomato wilt. Jour. Agr. Res. 36, 697-719. 1928. 7. HOWARD, GRACE E. Pigment studies with special reference to carotinoids in fruits. Ann. Missouri Bot. Card! 12, 145-212. 1925. 8. LINDSTROM, E. W. Inheritance in tomatoes. Genetics 10, 305-317. 1925. 9. - Hereditary correlation of size and color characters in tomatoes. Iowa Agr. Exp. Sta. Res. Bui. 93. 1926. 10. - The inheritance of ovate and related shapes of tomato fruits. Jour. Agr. Res. 34, 961-985. 1927. 11. MC\VHORTER, F. P., and PARKER, M. M. Fusarium wilt of tomato in Vir- ginia. Va. Truck Exp. Sta. Bui. 64. 1928. 12. A comparison of wilt resistant tomatoes in Virginia. Va. Truck Exp. Sta. Bui. 69. 1929. 13. MYERS, C. E. Statistical studies of inheritance in the tomato. Penn. State Col. Agr. Exp. Sta. Bui. 189. 1924. 14. SCHNECK, H. W. Pollination of greenhouse tomatoes. N.Y. (Cornell) Agr. Exp. Sta. Bui. 470. 1928. UNIVERSITY OF ILLINOIS-URBANA