Control of Fruit Size in Greenhouse Tomatoes
By Stuart Field and Mike NicholsMassey University
Palmerston North, New Zealand
Fruit size in tomatoes is normally controlled by choice of variety and by a process called truss thinning, in which the number of fruits allowed to develop on a fruit truss is controlled by the removal of any flowers in excess of the pre-determined fruit number. This normally means that mean fruit size is controlled by reducing fruit numbers per truss in the winter, and increasing them in the summer.
However, mean fruit size is not the factor which influences market return, as it the the actual size of the individual fruit which is critical (ie fruit size distribution). Increasing the fruit number per truss during the summer (when growing conditions are optimum) may keep the mean fruit size constant, but there will be an increased range in fruit weights due to the additional number of fruit per truss.
Our study was to examine the relationship between fruit thinning and fruit size (weight) of the individual fruits down the truss in order to develop a strategy to increase the market yield of greenhouse tomatoes.
The experiment was conducted at the Plant Growth Unit (PGU), Massey University, Palmerston North during March through to December 2001. A greenhouse of 12m X 8 ½m and growing height of 3m was selected. A NFT (nutrient film technique) system was constructed with seven-meter long channels made from pandafilm running parallel across the width of the greenhouse in double rows spaced at 40cm on a slope of 1 in 35. These double rows were spaced at 1m intervals.
Seeds of the tomato cultivars: Alboran (Standard), Ophir (Beefsteak) and Cherita (Cherry) were all sown on March 1, 2001, into a peat based seedling medium. The seed was germinated on a heated (22°C) capillary bench in a greenhouse. On March 13, 2001, the seedlings were pricked out and transplanted singularly into one hydroponic cell. These transplants were placed on a wire mesh bench so the roots would not grow out of the trays. The greenhouse was maintained to heat at 14°C and vent at 20°C. The transplants were liquid fed every two to three days until planting to ensure continued growth and development.
The seedlings were transplanted out into the NFT channels on April 11, 2001. Each cultivar was grown at a density of 27 plants per double row (2.76 plants/m2). On fruit formation Alboran was thinned to three, four or five fruit per truss, Ophir thinned to one, two or three fruit per truss, and Cherita thinned to four, eight or 12 fruit per truss. The truss thinning was achieved by thinning from the distal end of the truss to the required truss numbers. Bumblebee hives were put into the greenhouse every two months to promote good fruit set on all tomatoes. An EC of 3.0-4.0 mScm-1 and a pH between six to seven was maintained daily by the addition of water or a nutrient, acid, or base solution. Greenhouse temperatures were maintained to ventilate at 22°C and heat at 15°C throughout the whole growing season.
Neem oil with an Azadirachtin content of over 1500 ppm was sprayed once or twice a week to control whitefly infestations. Removal of the bottom leaves up to the last fruiting truss was performed to eliminate any onset of disease. Fresh red fruit weight was recorded at each fruit position in the truss.
We are unclear whether our results are consistent for all cultivars but it is clear from our preliminary results that :
- the larger fruited (beefsteak) types show a much greater reduction in fruit weight (size) as one moves down the truss, than the small fruited (cherry) types, while the standard varieties are intermediate.
- In general the size of the first fruit on the truss determines the size of the other fruits down the truss.
Thinning to less fruit per truss resulted in significantly heavier proximal fruit (Table 1 & 2). This difference in proximal fruit weight was then observed down the truss as shown in Figures 1 and 2. This indicates that by reducing the number of fruit on a truss or sink size, the remaining fruit (with more photosynthetic now available to them) increased in proportion to each other within the truss, thus keeping the same slope.
Table 1. Weight of first fruit on truss for Alboran fruit thinning treatments. Figures within columns followed by different letters shows a significant difference at P d” 0.05 using LSD test.
Table 2. Weight of first fruit on truss for Ophir fruit thinning treatments. Figures within columns followed by different letters shows a significant difference at P d” 0.05 using LSD test.
Each truss on the plant was measured for all three Cherita fruit thinning treatments. (Figs. 3 to 5). It was shown that all trusses up the plant followed the same slope in each of the three fruit thinning treatments. There was also a trend of the earlier trusses having a heavier proximal fruit, and thus heavier fruit down the truss, compared with the later trusses. This was probably the case of the later trusses developing under a high plant fruit load with increased competition for available assimilate, while the earlier trusses had less competition for available assimilate.
Fruit thinning to eight fruit per truss in Cherita resulted in significantly smaller fruit than thinning to 12 fruit per truss as was shown with the proximal fruit (Table 3). No reason can be given for this except that in both blocks of plants, the eight fruit per truss treatment plants were by chance randomly positioned on the eastern part of the glasshouse where solar radiation is lowest. Therefore the reduced solar radiation would cause fewer assimilates to be produced causing lighter proximal fruit.
Table 3. Weight of first fruit on truss for Cherita fruit thinning treatments. Figures within columns followed by different letters shows a significant difference at P d” 0.05 using LSD test.
Let us look at an example.
Fruit weights of standard tomato varieties get smaller by about 15 g per fruit as one moves down the truss. If we require a 75 g tomato, and we have a fruit size of 100 g for fruit No. 1, then the fruits down the truss are:
| Key: | A—5 fruit/truss spring |
| B—5 fruit/truss summer | |
| C—5 fruit/truss winter | |
| D—7 fruit/truss summer | |
| E—3 fruit/truss winter |
Assume that the premium grade of tomatoes ranges from 55-90 g, then although we get an increased total yield under good growing conditions by increasing fruit number/truss (D) we do not increase the yield of premium size.
The answer lies not in increasing the number of fruit per truss, but to increase the number of trusses/m2. This can only be achieved by increasing the number of stems per plant (or /m2).
Similar results occur when a similar study was undertaken with beefsteak and with cherry tomatoes.
The difference being that reduction in weight per fruit as one moves down the truss is larger for the large fruited beefsteak type and smaller for the small fruited cherry tomatoes.
It is interesting to note that the yield per truss, and therefore per plant for cherry tomatoes is lower than for standard types, which are in their turn lower than for the beefsteak types.
For this reason it is possible that attempting to standardize fruit size by using “truss tomatoes” types (with their more even fruiting down the truss) may result in lower yields.
Clearly to control fruit size through the year requires a combination of thinning trusses, but more importantly the manipulation of truss numbers per m2.
If there is a premium price paid in the markets for specifically sized tomatoes, then it is clearly economically relevant to develop production systems which will maximize productivity of the premium sizes.
Our work suggests that this is most likely to be achieved by manipulating stem numbers (truss numbers), rather than by manipulating fruit numbers per truss.
Although we did not examine the effect of stem density in our study, it is likely that to maximize productivity, the amount of foliage (leaf area) should be seasonally adjusted, possibly by doubling stem numbers in the summer, compared with winter, with spring and autumn stem numbers being intermediate.
Under these circumstances a constant number of fruit per truss may prove to be the ideal solution?
References
Bangerth, F. 1989. Dominance among fruits/sinks and the search for a correlative signal. Physiologia Plantarum, 76: 608-614.
Bangerth, F. and Ho, L. C. 1984. Fruit position and fruit set sequence in a truss as factors determining final size of tomato fruits. Annals of Botany, 53: 315-319.
Bohner, J. and Bangerth, F. 1988. Effects of fruit set sequence and defoliation on cell number, cell size and hormone levels of tomato fruits (Lycopersicon esculentum Mill) within a truss. Plant Growth Regulation, 7: 141-155.