Poysa, V. TBRT ReportImproved solids levels is a major objective of most paste tomato breeding programs. One of the most promising methods for increasing solids levels is transferring 'high solids' genes from the related species, Lycopersicon cheesmanii and L. chmielewskii to tomato. In this set of experiments seven accessions were used as sources of high solids, including two L. chmielewskii lines, LA 1306 and LA 1327, one L. cheesmanii form typicum, LA 1040, and four f. minor lines, LA 317, LA 483, LA 528, and LA 1508; which had soluble solids levels in California of from 12.4 to 15.9 % (Garvey and Hewitt, 1984). Purdue 812, a determinate, mid-season, relatively small fruited processing tomato cultivar was used as the adapted recurrent parent in most lines.
The early segregating generations were selected as single plants for a combination of fruit size (>20g), fruit yield potential (>50% of commercial), red fruit colour, determinate growth, earliness, and high soluble solids levels, as determined by hand held refractometer. A total of 118 experimental lines, involving from 0 to 4 backcrosses and ranging from 4 to 8 generations removed from the wild parent were grown in four replications. Solids levels were determined on a bulk sample of 20 ripe fruit per plot and yield was measured as all the fruit set on the five central plants in each plot.
The best experimental line derived from L. chmielewskii was a BC3-F3 line from LA 1327, which had 10% higher levels of total and 13% higher soluble solids than Purdue 812, combined with a 28% fruit yield advantage. Of the 35 lines derived from LA 1040, only two of had marginally higher total solids levels than Purdue 812, while four had slightly higher soluble solids levels. The majority (42 of 77) of lines derived from L. cheesmanii f. minor accessions had up to 20% higher total solids levels than the adapted parent, Purdue 812. Several of these lines also exceeded it in total fruit yield.
The early generation selection for solids level combined with fruit size and fruit yield was effective in breaking the negative association among these traits. In these experiments total solids levels were not significantly correlated with fruit size, fruit yield, nor earliness.
For total or soluble solids levels, one or two backcrosses followed by two to four cycles of selection resulted in a higher percentage of lines superior to the adapted parent than did four to seven cycles of selection in the segregating generations. For fruit yield and fruit size, two backcrosses were superior to one, which was superior to selection without backcrossing. There was little evidence that three or four backcrosses improved the frequency of either high yielding or high solids lines.
Literature cited:
Garvey, T.C. and J. Hewitt 1984 TGC 34:4-5.
Table 1. Mean total and soluble solids levels, fruit yield, fruit weight, and soluble solids yield per hectare for selected experimental lines and the recurrent parent, Purdue 812.
___________________________________________________________
GENOTYPE Source Total Soluble Fruit Fruit Brix
Solids Solids Yield Wt. Yield
(Brix) (Brix) (t/ha) (g) (t/ha)
___________________________________________________________
873311 LA 1327 8.3* 7.5* 33.0 39 2.48*
8617211 LA 1306 8.1* 7.3* 37.4 40 2.71*
4802822 LA 317 9.0* 8.5* 13.7# 10# 1.21#
4802762 LA 317 9.0* 8.1* 8.2# 11# 0.64#
4802722 LA 317 8.9* 8.2* 15.4# 10# 1.22#
3212 LA 317 8.7* 7.9* 17.2# 26# 1.33#
7463 LA 317 8.5 7.9* 25.6 16# 2.04
321 LA 528 7.7* 6.8* 31.2 20# 2.11
324 LA 528 7.6 6.8* 25.9 21# 1.73
244 LA 528 7.6 6.8* 25.7 14# 1.74
322 LA 528 7.4 6.7* 35.2 24# 2.34
87312 LA 1508 6.9* 6.1* 19.7 15# 1.18
870762 LA 1508 6.7* 6.1* 33.4 19# 1.96*
870761 LA 1508 6.6* 5.9 37.8 23# 2.23*
88031 LA 483 6.8* 6.0* 27.9 21# 1.78
Purdue 812a 7.2 6.4 28.3 40 1.81
___________________________________________________________
* = Significantly (P=.05) superior to Purdue 812 within trial.# = Significantly (P=.05) inferior to Purdue 812 within trial.
a = Mean values across five trials.