Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
HORTSCIENCE 54(2):311–316. 2019. https://doi.org/10.21273/HORTSCI13487-18
Fruit Cluster Pruning of Tomato in an
Organic High-Tunnel System
Brian A. Mitchell
1
and Mark E. Uchanski
Department of Horticulture and Landscape Architecture, Specialty Crops
Program, Colorado State University, Fort Collins, CO 80523
Adriane Elliott
Department of Soil and Crop Sciences, Colorado State University, Fort
Collins, CO 80523
Additional index words. fruit load management, fruit thinning, truss pruning, hoop house
Abstract. This research assessed fruit load management and production techniques for
cultivating indeterminate tomatoes in a high tunnel under intensive organic manage-
ment. The successful production of high-quality, high-yielding crops is important for
fruit and vegetable producers, especially growers using high tunnels. High tunnels are
well-suited to organic farming and can be used to grow many valuable specialty crops.
Fruit load management is practiced in fruit production (e.g., apples, peaches, and
grapes), but there is lack of consensus concerning the effectiveness of fruit cluster
pruning on tomato (Solanum lycopersicum) and its impact on fruit yield, quality, and
marketability. In addition, there is no published research on tomato cluster pruning in
certified organic systems or intensively managed high tunnels (e.g., densely planted,
trellised, vegetatively pruned plants) for the Front Range of Colorado. In 2016 and 2017,
a randomized complete block design was used to test the effects of cluster pruning within
a high tunnel on certified organic land at Colorado State University’s (CSU’s)
Agricultural Research, Development, and Education Center, South. Two treatments
and three tomato cultivars were selected for the study; the treatment–cultivar combi-
nations were replicated six times within a high tunnel. The treatments involved reducing
fruit loads to three fruit and six fruit per cluster, whereas plants with unpruned clusters,
which naturally developed as many as 10 fruit, served as the control. Tomato cultivars
evaluated were ‘Cherokee Purple’, a widely studied heirloom, and two hybrids: ‘Jet Star’
and ‘Lola’. Parameters measured included total yield, individual fresh fruit weight,
soluble solids content (SSC), marketable yield, and nonmarketable yield. Individual fresh
fruit weight increased for both hybrids in the three-fruit treatment, averaged over two
growing seasons. ‘Cherokee Purple’ did not respond to the cluster pruning treatments.
There was no decrease in total yield, across all cultivars, between treatments and the
unpruned control. However, ‘Jet Star’ yielded more than the other two cultivars. In
addition, SSC and marketability measurements were more influenced by cultivar than
cluster pruning treatments. ‘Lola’ had a significantly greater SSC than the other two
cultivars. ‘Jet Star’ had the greatest marketable yields of all cultivars tested whereas
‘Cherokee Purple’ produced greater nonmarketable (cull) yields. Cluster pruning
produced larger organic tomatoes without reducing yield or quality for two of the three
cultivars used in the study. Cultivar selection remains one of the greatest factors in
determining yield, quality, and marketability of a crop.
A pound of organic heirloom tomatoes
costs $5 at the Larimer County Farmers’
Market in Fort Collins, CO. Nonorganic
heirlooms range from $2–6.50/lb, whereas
other tomatoes at the local market, both
organic and noncertified, range from $3–5/
lb (Colorado Department of Agriculture,
2018). Trends in the national fresh tomato
(Solanum lycopersicum) market indicate the
increase in demand for both organically
grown and heirloom types. In addition, or-
ganic tomatoes are often sold at a 15% to
20% price premium over conventionally
grown crops (O’Connell et al., 2012). The
market for organic produce in general has
shown strong growth during the past two
decades and currently makes up 5.3% of total
food sales of the United States (Organic Trade
Association, 2018). The objective of this study
was to evaluate differences in yield and
quality of tomatoes using two fruit cluster
pruning treatments and three cultivars of in-
determinate tomatoes grown in an intensively
managed, certified organic high tunnel.
Many tactics are used to grow high-
quality fresh produce for direct markets such
as farmers’ markets, community-supported
agriculture (CSA), and local restaurants. The
use of a high tunnel is one of the most
successful tactics for growing high-quality,
high-yielding, and valuable crops. A high
tunnel, or hoop house, is a temporary, mov-
able, or semipermanent structure that may be
single- or multispan (i.e., many connected
structures). High tunnels may be covered in
polyethylene film, insect netting, or shade-
cloth, or left bare, and are typically unheated
and passively ventilated. Crops are usually
planted directly in the soil within high tun-
nels, although containers and soilless media
are used occasionally. A high tunnel provides
multiple benefits for the grower over open-
field cultivation, including season extension,
insect exclusion, reduced disease pressures,
protection from environmental damage (such
as hail avoidance), and greater marketable
yields. Furthermore, high tunnels provide
valuable services in organic agricultural sys-
tems because they can improve the quality
and yield of vegetables over open-field grow-
ing systems (Rogers and Wszelaki, 2012).
Tomatoes are the most popular and often
the most profitable plant for high-tunnel
growers. They are an excellent crop choice
as a result of their high value in the fresh
market and the crop’s ability to produce high
yields (Mefferd, 2017). Indeterminate toma-
toes produce fruit throughout the growing
season and benefit from both the high tunnel
environment and trellising. Indeterminate
cultivars of tomato allow growers to meet
consumer demand consistently. High tunnels
provide a protected microclimate and longer
harvests, and allow for continuous produc-
tion of fresh fruit (Rogers and Wszelaki,
2012).
Fruit cluster pruning of tomatoes has been
shown to influence total yield, marketability,
individual fruit fresh weight, and various
indicators of quality, such as dry matter and
SSC. Fruit cluster pruning is used to limit the
number of fruit per cluster and reduce com-
petition to increase individual fruit weight
(Hanna, 2009). Studies across the world
(Turkey, South Africa, the Netherlands, Italy,
and the United States) have produced con-
flicting results on the efficacy of cluster
pruning of tomatoes in various controlled
environments (Table 1). Specifically, there
is a lack of research on cluster pruning in
intensively managed organic systems, in high
tunnels, and on the Front Range of Colorado.
Cultivar selection is a critical component
for growers who are concerned about the
maximization of yield, the fresh market
appeal of their products, and the general
performance (e.g., pest and disease resis-
tance, physiological traits, and quality) of
their crops (Healy et al., 2017). Growers
evaluate cultivars for their desired fruit qual-
ities, shelf life and storage characteristics,
and potential yield (Hanna, 2009). Although
most hybrid cultivars of tomatoes are bred for
increased yield, visual appeal, and shipping
durability, open-pollinated heirloom culti-
vars of tomatoes are valued for their unique
colors, shapes, flavors, and legacies. How-
ever, heirloom tomatoes generally lack uni-
formity, have thinner skins, lack disease
Received for publication 10 Aug. 2018. Accepted
for publication 17 Dec. 2018.
We thank the Department of Horticulture and
Landscape Architecture, the Agricultural Experi-
ment Station, and the Statistical Laboratory at
Colorado State University, Fort Collins, CO.
1
Corresponding author. E-mail: Brian.Mitchell@
colostate.edu.
This is an open access article distributed under the
CC BY-NC-ND license (https://creativecommons.
org/licenses/by-nc-nd/4.0/).
HORTSCIENCE VOL. 54(2) FEBRUARY 2019 311
resistance, and have lower yields than most
modern hybrid cultivars (O’Connell et al.,
2012).
SSC, represented in Brix, is a common
measurement used by tomato growers, proces-
sors, and fresh market growers. In the fresh
market, soluble solids levels provide an approx-
imation of how sweet the tomato may taste;
however, many components define the overall
flavor of each tomato, including sugars, acids,
volatiles, and other compounds. An SSC mea-
surement in Brix, using a refractometer, is a
quick, reliable, and inexpensive field test for
quality. Although measurements can be an early
indication of sweetness and flavor, SSC can
fluctuate as a result of many factors, including
crop, cultivar, maturity, growing environment,
and storage conditions (Kleinhenz and Bum-
garner, 2015). Healy et al. (2017) found that
tomatoes had a greater SSC when grown in
a high tunnel compared with open-field
production.
Tomato vegetative growth is pruned reg-
ularly when the crop is grown in a controlled
environment. Tomato plants develop side
shoots in their leaf axils; these axillary shoots
will continue to grow and produce leaves and
fruit clusters if not removed. Axillary shoots
should be removed regularly (often repeat-
edly) to maintain a single leader, or main
stem. This is an effective way to maximize
production space, and can improve the yield
and quality of tomatoes (Maboko et al.,
2011). It is also common practice in intensive
greenhouse operations to remove any leaves
below the lowest ripening cluster. The lower
canopy receives little sunlight and is unnec-
essary for continued plant growth and fruit
production. In addition, the removal of su-
perfluous leaves allows the plant to allocate
more energy to vertical growth and fruit
development while also reducing disease
pressure by maximizing air flow.
The overall objective of this study was to
evaluate the differences between two tomato
fruit cluster pruning treatments and a control
on three cultivars of indeterminate tomatoes:
Cherokee Purple, Jet Star, and Lola. We
hypothesized that 1) cluster pruning would
decrease total yield, 2) cluster pruning would
increase the quality of the organic tomatoes,
and 3) cultivars would respond differently to
the cluster pruning treatments for the param-
eters analyzed in the study. To test these
hypotheses, differences between cluster
pruning treatments and cultivar performance
were assessed in terms of total yield, indi-
vidual fresh fruit weight, SSC, marketable
yield, and nonmarketable yield.
Materials and Methods
Site description, high-tunnel description,
and cultural practices. This study was conduct-
ed in semipermanent high tunnels on certified
organic land at CSU’s Agricultural Research,
Development, and Education Center, South
(ARDEC S.) (40.610012, –104.993979; alti-
tude: 1523 m) in 2016 and 2017. CSU in Fort
Collins, CO, has 557 m
2
of high-tunnel space
across seven structures dedicated to research in
vegetable crop breeding, vegetable cropping
systems, cover crops, and cyanobacterial fer-
tilizer (Carey et al., 2009). The soil in the high
tunnels at ARDEC S. is classified as a Nunn
clay loam (Soil Survey Staff, Natural Re-
sources Conservation Service, U.S. Depart-
ment of Agriculture, 2018). Soil samples
werecollectedtoadepthof20to30cmeach
year before planting and were tested at the CSU
Soil, Water and Plant Testing Laboratory. Soil
was analyzed for pH, salts, lime, texture,
organic matter, and nutrient content to deter-
mine recommended preplanting compost ap-
plication and fertilizer rates during the growing
season.
The certified organic high tunnels were
prepared each year by tilling the soil 0.15 to
0.30 m deep with a rototiller (Harvester 722;
BCS, Portland, OR). All practices aligned
with U.S. Department of Agriculture guide-
lines for organic production. After tillage, 1.2
to 1.5 m
3
of plant-based compost (A-1 Or-
ganics, Eaton, CO) was incorporated into the
topsoil. Six lines of drip tape 13.7 m long
Table 1. A literature summary of tomato fruit cluster pruning controlled environments.
Site
Growing
environment Tomato cultivar
Fruit no./
cluster
Saglam and
Yazgan (1999)
Turkey Low tunnel Vivia 4, 6, 8
Hanna (2009) Louisiana, U.S. Greenhouse Geronimo, Trust, Quest 3, 4
Maboko et al. (2011) South Africa Hydroponic/shade FA593 4, 6
Fanasca et al. (2007) Netherlands Greenhouse Cederico/Maxifort rtstk. 3, 6
Fig. 1. The randomized complete block design used within the high tunnel in 2016.
312 HORTSCIENCE VOL. 54(2) FEBRUARY 2019
were run lengthwise within the high tunnel,
corresponding to the blocking described
later. The drip tape was attached to 1.9-cm
black plastic irrigation headers at each end of
the high tunnel and emitted 500 L/h/100 m.
The high tunnels were 6.1 m wide, 15.2 m in
length, and had twin-walled polycarbonate
end walls with a 1.8-m square roll-up door on
one side. Insect netting was used on both high
tunnels and was attached to the sides with a
system commonly used in the nursery and
greenhouse industry. The netting was fine
meshed and served to exclude insect pests.
The high tunnels had six metal cables
running from the end bows that functioned as
a trellising system. Extra metal cables were
strung perpendicular to the six main trellis
cables for additional support. Trellising
spools with white twine and hooks on each
end were used to train the single-leader
tomatoes. Trellising clips were used to attach
the plants to the twine and were added every
15 to 30 cm throughout the growing season.
Tomato fruit cluster supports were used to
redistribute fruit weights from plant stems to
the trellises and were added as fruit ripened.
The reduction of the number of fruit per
cluster can increase fruit size dramatically,
and the weight of large fruit can cause
damage to the cluster and stem (Saglam and
Yazgan, 1999).
Water was supplied to the plants with a
drip irrigation system. The tomatoes were
watered once or twice daily for 15 to 30 min
using an irrigation controller. In 2016, Organic
Materials Review Institute (OMRI)-approved
fish emulsion fertilizer (Alaska Fertilizer 5–1–
1; Pennington, Renton, WA) was used for
supplemental nitrogen fertilization throughout
the growingseason and was applied every 2 to
3 weeks from July tomid September. In 2017,
the same fertilizer was used from mid June to
mid July. A different fish emulsion product
(Drammatic ‘‘One’’ Plant Food 4–4–0.5;
Dramm Corporation, Manitowoc, WI) was
used from August until mid September. The
fertilizer change was an economic decision;
however, all treatments received the same
rates of nitrogen, phosphorus, and potassium
fertilizers. Organic fertilizers were applied
through the irrigation system using a siphon
injector or by hand with a watering can. In mid
July 2017, a small fertilizer injector (Dosatron,
Clearwater, FL) replaced the previous system.
Woven weed fabric was used to control weeds
within the high tunnel. Weeds were pulled by
hand or mowed in and around the high tunnel
as needed. Predominant weeds included bind-
weed, canada thistle, and grasses.
The tomato cultivars evaluated were
Cherokee Purple, a popular heirloom used
widely in research; Jet Star, an F
1
hybrid
cultivar previously trialed at CSU’s CSA
program; and Lola, an F
1
hybrid cultivar bred
for intensive greenhouse cultivation, which
differs from high-tunnel production. Seeds
were sown in a greenhouse on 17 May 2016
and 13 Apr. 2017. The seedlings were hard-
ened off in a shaded, protected outdoor space
for a week before planting. The seedlings
were transplanted into a high tunnel in June
during both years of the study. All tomato
plants were trained to have a single leader
beginning on 8 July 2016 and 19 June 2017.
Fruit clusters would often develop stems or
leaves that would to grow to obscure and
possibly damage the cluster; this excessive
vegetative growth was pruned off all plants
during the study. Leaves were removed
throughout the growing season as they
senesced, showed signs of disease, touched
the soil or weed barrier (to manage dis-
ease transmission preventively), or if they
hampered data collection or harvesting.
Sufficient leaf cover was left to shade de-
veloping and ripening fruit to prevent sun-
scald and uneven ripening. Nitrile gloves
were worn while pruning, and a 70% etha-
nol solution was sprayed on small pruners
between plants for good sanitation. It was
necessary to prune side shoots, unnecessary
leaves, and excess vegetative growth on
fruit clusters once a week. A pollination
wand (Garden Pollinator model VBP-01;
VegiBee, Maryland Heights, MO) was
used to manage pollination. It was used
in the mornings every other day on me-
dium speed near each mature flower on the
cluster.
Fig. 2. A visual representation of the cluster pruning treatments (on ‘Lola’) in the study. Clockwise from
top left: three-fruit treatment, six-fruit treatment, and an unpruned cluster (control), which could set up
to 10 fruit, depending on the cultivar and growing season.
Fig. 3. ‘Jet Star’ had a greater total yield than both ‘Cherokee Purple’ and ‘Lola’ when results were
averaged over treatment, year, and block. Data are shown as mean ± SE. Dissimilar letters represent
differences at a= 0.05.
HORTSCIENCE VOL. 54(2) FEBRUARY 2019 313
Experimental design, cluster pruning
treatments, and harvests. A randomized com-
plete block design was used for this experi-
ment (Fig. 1). The high tunnels at ARDEC
S. are oriented lengthwise east to west; in
northern latitudes, east–west orientation is
commonly used to maximize light intercep-
tion in the winter. Six blocks ran east to west;
blocking was used to compensate for known
variability in the growing space within the
high tunnel. There were 0.9-m spaces be-
tween blocks and from blocks to the high-
tunnel sides. Clusters were reduced to three
fruits in one treatment and six fruits in the
second treatment. Plants in the control had
unpruned clusters, producing as many as 10
fruit per cluster (Fig. 2). The two cluster
pruning treatment groups and the control
groups were assigned randomly within
each block. Treatments within blocks were
separated by a 1.2-m walkway. Within each
treatment, four-plant groupings of the three
indeterminate tomato cultivars were assigned
randomly. Plants within treatments were
spaced 0.3 m apart. In 2017, the experi-
mental design was the same, but the exper-
iment was conducted in an adjacent high
tunnel to ensure an adequate crop rotation
per certified organic guidelines. The two
treatments and the control were rerandom-
ized within the six blocks; the three cultivar
subsamples were rerandomized within each
treatment.
Fruit were pruned from clusters when they
were marble size (Maboko et al., 2011). The
fruit most proximal to plant stems were
selected to remain on the cluster, and distal
fruit were removed when a choice was avail-
able. King fruit (the result of two or more
flowers fusing together) were was identified
and removed from clusters. These fused fruit
are often first to develop and are often larger
than average, misshapen, and unmarketable.
King fruit were particularly an issue with
‘Cherokee Purple’ in our study. Cluster
pruning was done once a week throughout
the growing season, from 29 July to 28 Sept.
2016 and from 7 July to 27 Sept. 2017.
Harvests were conducted to emulate a
grower harvesting ripe fruit for the local fresh
market. In 2016, there were 11 harvests, with
the first harvest on 30 Aug. 2016 and the final
one on 7 Oct. 2016. In 2017, there were 14
harvests, with the first harvest on 31 July
2017 and the last harvest taking place on 6
Oct. 2017. The data collected for each fruit
included the block number, treatment, plant
number within treatment and cultivar, cluster
number proximal to distal, tomato number
and position on each cluster, whether the fruit
was marketable, notes on physiological dis-
orders or damage, and fruit fresh weight.
Marketability was determined using a com-
bination of the U.S. Standards for Grades of
Fresh Tomatoes and local market standards
(U.S. Department of Agriculture, 1991).
On 23 Sept. 2016 and 8 Sept. 2017, specific
harvests were conducted to assess the SSC of
fruit using the Brix scale, which ranges from
0 to 56. Fruit werecollected during the peakof
the harvest season—a standard practice when
measuring SSC. Representative fruit were
harvested from plants within each experimen-
tal unit. The samples were sealed in plastic
bags and stored in a freezer at 30 C. The fruit,
when thawed, were mashed and combined to
attain an aggregate measure. The pulp was
strained with cheesecloth and the juice
was assessed using a refractometer. SSC was
measured and recorded for each of the 54
experimental units each year. An acceptable
range for fresh market tomatoes in Colorado is
3.5 to 5.3. The Brix scale is used to assess
harvest readiness and to determine quality in
the field and during processing, and represents
a product’s potential sweetness and flavor
(Kleinhenz and Bumgarner, 2015).
Statistical analyses. A two-way factorial
treatment structure with subsampling was
used in our study. Analyses explored inter-
actions and main effects of treatment and
cultivar. If an interaction was found to be
significant at a= 0.05, additional tests were
conducted. The separate linear models in-
cluded block, year, treatment, and cultivar as
categorical variables, and allowed interaction
between treatment and cultivar effects. Re-
sponse variables were total yield, individual
fresh fruit weight, SSC measured in Brix,
marketable yield, and nonmarketable yield.
Blocks were treated as a random effect
whereas year, treatment, and cultivar were
treated as fixed effects in each model. The
experimental unit consisted of four-plant
groupings by cultivar that represented each
treatment and cultivar combination. Years
were combined for a total of 12 replications
for the study; the same two treatments and
three cultivars were used both years.
All data were analyzed using R statistical
software (R Core Team, 2016) and R Studio
(version 1.0.136). Yields was assessed by taking
the mean yield for the plants in each experi-
mental unit. Fresh fruit weight was calculated
by taking the mean yield of the individual fruit
Fig. 4. The treatment effect of cluster pruning on individual fresh fruit weight of organic high-tunnel
tomatoes. ‘Jet Star’ and ‘Lola’ produced larger fruits with the three-fruit treatment. Data are shown as
mean ± SE. Dissimilar letters represent differences at a= 0.05 and results are averaged over year and
block.
Fig. 5. The cultivar effect on soluble solids content, where ‘Lola’ measured greater on the Brix scale than
both ‘Cherokee Purple’ and ‘Jet Star’. Data are shown as mean ± SE . Dissimilar letters represent
differences at a= 0.05 and results are averaged over treatment, year, and block.
314 HORTSCIENCE VOL. 54(2) FEBRUARY 2019
from each plant and taking the average mean
fruit weight for the entire subsample. SSC was
calculated as described previously. An analysis
of variance (ANOVA) and pairwise compari-
sons were used to compare the differences in the
least-squared means of the response variables.
The Type III ANOVA used the Kenward-Roger
approximation for df whereas ttests used
Satterthwaite approximations to df. Pairwise
comparisons were used to separate least-
squared means by treatment and cultivar, and
results were averaged over year. Default P
value adjustments for pairwise comparisons
were made using the Tukey method for com-
paring a family of three estimates. The signif-
icance level was set at 0.05 for all analyses.
Results and Discussion
Total yield. There was a significant effect
of cultivar (P< 0.01) for total yield, as the
three cultivars performed differently, aver-
aged over year and treatment (Fig. 3). There
was no interaction between treatment and
cultivar, and no treatment main effect was
observed for total yield in our experiment.
The results for total yield were comparable to
the findings of Maboko et al. (2011), whose
cluster pruning did not have a significant
effect on tomato total yield.
Cultivar selection had more impact on total
yield than a reduction in the fruit load per
cluster. Despite its compact yet indeterminate
growth habit, ‘Jet Star’ consistently produced
highly acceptable fruit regardless of treatment,
and outperformed both cultivars when consid-
ering total yield. ‘Jet Star’ proved to be a good
selection for organic high-tunnel production in
our study, as demonstrated in the context of
total yield and marketability.
Individual fresh fruit weight. An interac-
tion between treatment and cultivar was
found in the responses for individual fresh
fruit weight (P< 0.01). Significant differ-
ences were found for both the main effect of
treatment (P< 0.01) and cultivar (P< 0.01).
Both hybrid tomato cultivars, Jet Star and
Lola, produced larger fruits with three-fruit
clusters than with six-fruit or unpruned clus-
ters (Fig. 4). ‘Jet Star’ fresh fruit weight in the
three-fruit treatment was 32% greater than
the control and 31% greater than the six-fruit
treatment. The fresh fruit weight for ‘Lola’ in
the three-fruit treatment was 27% greater
than the control and 29% greater than the
six-fruit treatment. There were no treatment
differences in the fresh fruit weight of ‘Cher-
okee Purple’ tomatoes.
Pairwise comparisons for the effect of
cultivar showed that ‘Cherokee Purple’
plants with unpruned clusters produced
larger fruit (P< 0.01) than both ‘Jet Star’
(23% smaller) and ‘Lola’ (31% smaller).
Results were similar for the six-fruit treat-
ment: ‘Cherokee Purple’ produced larger
individual fruits than both hybrids (P<
0.01). With six-fruit clusters, ‘Cherokee
Purple’ fresh fruit weight was 34% greater
than ‘Jet Star’ and 54% greater than ‘Lola’.
Under the three-fruit cluster pruning treat-
ment, ‘Jet Star’ produced a fresh weight
measurement 16% greater than ‘Lola’ (P<
0.01). ‘Cherokee Purple’ clusters with three
fruit almost produced larger fruit than ‘Lola’
in the same treatment (P< 0.07), but results
were not significant at P= 0.05.
Pruning tomato clusters to three fruit in-
creased fruit weight significantly for two of the
three cultivars in our study. This is similar to
previous findings, in which a reduced fruit load
per fruit cluster resulted in greater individual
fresh fruit weights (Hanna, 2009; Saglam and
Yazgan, 1999). The hybrid cultivars Jet Star
and Lola, demonstrated a strong response to the
three-fruit treatment, producing fruit that were
one-third larger than either the six-fruit treat-
ment and the control. Cluster pruning likely
reduces competition for plant resources and
thus increases fruit weight (Hanna, 2009).
‘Cherokee Purple’ showed no response to
the cluster pruning treatments for the param-
eter of fresh fruit weight. The fruit produced
by this heirloom tomato cultivar were often
very large, regardless of the number of fruit
per cluster. ‘Cherokee Purple’ with unpruned
clusters produced larger fruits than both hy-
brids, which is likely explained by their pro-
clivity for producing large fruit. The response
was similar with the six-fruit treatment; often,
clusters did not produce more than three or
four fruit, so there was little difference be-
tween the six-fruit treatment and the control.
Soluble solids content. For SSC, no in-
teraction was found between treatment and
cultivar. Neither was there a significant main
effect of treatment; SSC was unaffected by
cluster pruning. However, there was a signif-
icant main effect of cultivar (P< 0.01); ‘Lola’
had a significantly greater SSC than the other
two cultivars (Fig. 5). A larger tomato—from
a three-fruit cluster, for example—might be
expected to have greater water content and
reduced sugars compared with a smaller fruit.
When fruit load and competition for plant
resources are reduced and the sink-to-source
Fig. 6. ‘Jet Star’ produced more marketable yield than ‘Lola’, which produced more marketable fruits than
‘Cherokee Purple’. Data are shown as mean ± SE. Dissimilar letters represent differences at a= 0.05
and results are averaged over treatment, year, and block.
Fig. 7. The cluster pruning treatment effect on nonmarketable yield. The three-fruit treatment increased
nonmarketable yield for ‘Lola’ in our study. Data are shown as mean ± SE. Dissimilar letters represent
differences at a= 0.05 and results are averaged over treatment, year, and block.
HORTSCIENCE VOL. 54(2) FEBRUARY 2019 315
ratio decreases, more photosynthates, acids,
and other compounds are available for the
remaining fruit (Fanasca et al., 2007).
Marketable yield. No interaction was
found between treatment and cultivar for
the response of marketable yield. There was
no significant main effect of treatment, and
marketable yield did not differ between the
two growing seasons. However, regarding
marketability, there was a significant main
effect of cultivar (P< 0.01) and notable
differences between the three cultivars. ‘Jet
Star’ produced significantly greater market-
able yields than ‘Lola’, which in turn pro-
duced significantly greater marketable
yields than ‘Cherokee Purple’ (Fig. 6). In
our study, cultivar selection was the most
important factor when looking at marketable
yield. These results contradict studies in
which cluster pruning increased marketable
yield (Hanna, 2009; Saglam and Yazgan,
1999).
Nonmarketable yield. An interaction was
found between cluster pruning treatments
and cultivar for nonmarketable yield (P=
0.04). The main effect of cultivar was also
significant (P< 0.01). Exploring the main
effect of cultivar with pairwise comparisons,
a significant difference (P= 0.04) was found
between the three-fruit and six-fruit treat-
ments for ‘Lola’. The six-fruit treatment
reducednonmarketableyieldby34%com-
pared with the three-fruit treatment; there-
fore, the three-fruit cluster treatment increased
the nonmarketability of fruit compared
with the six-fruit cluster pruning treatment
(Fig. 7).
Potential nonmarketable yield, or cull
yield, is important to consider when de-
veloping production methods and choosing
cultivars to grow for direct markets. ‘Cher-
okee Purple’ produced fruit prone to various
types of cracking and problems like catfac-
ing and irregular shaping. ‘Lola’ fruit in-
creased in size with the three-fruit treatment,
yet frequently developed large radial cracks;
these cracks could host fungal infections and
render the fruit unmarketable while still on
the vine. ‘Jet Star’ produced the smallest
nonmarketable yields (had more marketable
fruit) and only showed physiological prob-
lems after periods of rain, as in 2017 when
ripening fruit split right before harvest.
Conclusions and Observations
The purpose of our study was to deter-
mine whether cluster pruning would improve
the yield and quality of three different culti-
vars of tomato grown in a certified organic
high tunnel. Cluster pruning treatments did
not increase or decrease total yield compared
with the control; organic high tunnel growers
could reduce fruit loads per cluster and
maintain yield, depending on their cultivar
selection. ‘Jet Star’ produced greater total
yields, averaged over treatment, than both
‘Cherokee Purple’ and ‘Lola’. Individual
fresh fruit weight increased for the two
hybrid cultivars in the three-fruit treatment.
Cultivar had an impact on SSC, with ‘Lola’
having greater SSC than the other two culti-
vars. Marketability and nonmarketability
were determined largely by cultivar attri-
butes, weather conditions, and insect and
disease pressure, and less by treatment, as
observed in other studies. ‘Lola’, a hybrid
cultivar used in commercial greenhouse pro-
duction, did not perform well in the context
of our study: Fruit did not ripen in a timely
manner and nonmarketable yield increased
with the three-fruit treatment. In summary,
larger tomatoes were produced in our study
by cluster pruning without a reduction in
yield or quality for two of the three cultivars
tested. Also, the yield, quality, and market-
ability of the organic tomato crops were more
influenced by the effects of cultivar than
cluster pruning treatments.
Literature Cited
Carey,E.E.,L.Jett,W.J.Lamont,Jr.,T.T.
Nennich, M.D. Orzolek, and K.A. Williams.
2009. Horticultural production in high tun-
nels in the United States: A snapshot. Hort-
Technology 19:37–43.
Colorado Department of Agriculture. 2018.
Farmers’ market price reports: 2015 Season
averages—Old Town Farmers’ Market—Fort
Collins. 8 Aug. 2018. <https://www.colorado.
gov/pacific/sites/default/files/Old%20Town%
20season%20ave%202015.pdf>.
Fanasca, S., A. Martino, E. Heuvelink, and C.
Stanghellini. 2007. Effect of electrical conduc-
tivity, fruit pruning, and truss position on
quality in greenhouse tomato fruit. J. Hort.
Sci. Biotechnol. 82(3):488–494.
Hanna, H.Y. 2009. Influence of cultivar, growing
media, and cluster pruning on greenhouse
tomato yield and fruit quality. HortTechnology
19:395–399.
Healy, G.K., B.J. Emerson, and J.C. Dawson. 2017.
Tomato variety trials for productivity and
quality in organic hoop house versus open
field management. Renew. Agr. Food Syst.
32(6):562–572.
Kleinhenz, M.D. and N.R. Bumgarner. 2015. Using
Brix as an indicator of vegetable quality: In-
structions for measuring Brix in cucumber,
leafy greens, sweet corn, tomato and water-
melon. Ohio State Univ. Ext. 18 Sept. 2015.
<https://ohioline.osu.edu/factsheet/HYG-1651>.
Maboko, M.M., C.P. Du Plooy, and S. Chiloane.
2011. Effect of plant population, fruit and stem
pruning on yield and quality of hydroponically
grown tomato. Afr. J. Agr. Res. 6(22):5144–
5148.
Mefferd, A. 2017. The greenhouse and hoophouse
grower’s handbook: Organic vegetable pro-
duction using protected culture. Chelsea Green
Publishing, White River Junction, VT.
O’Connell, S., C. Rivard, M.M. Peet, C. Harlow,
and F. Louws. 2012. High tunnel and field
production of organic heirloom tomatoes:
Yield, fruit quality, disease, and microclimate.
HortScience 47:1283–1290.
Organic Trade Association. 2018. Organic Trade
Association’s 2017 organic industry survey. 14
Mar. 2018. <https://www.ota.com/news/press-
releases/19681>.
R Core Team. 2016. R: A language and environ-
ment for statistical computing. R Foundation
for Statistical Computing, Vienna, Austria. 16
Jan. 2018. <https://www.R-project.org/>.
Rogers, M.A. and A.L. Wszelaki. 2012. Influ-
ence of high tunnel production and planting
date on yield, growth, and early blight de-
velopment on organically grown heirloom
and hybrid tomato. HortTechnology 22:452–
462.
Saglam, N. and A. Yazgan. 1999. Effect of fruit
number per truss on yield and quality in tomato.
Acta Hort. 491:261–264.
Soil Survey Staff, Natural Resources Conserva-
tion Service, U.S. Department of Agriculture.
2018. Web soil survey. 24 Feb. 2018. <https://
websoilsurvey.sc. egov.usda.gov/>.
U.S. Department of Agriculture. 1991. U.S.
standards for grades of fresh tomatoes.
12 Mar. 2018. <https://www.ams.usda.gov/
sites/default/files/media/Tomato_Standard%
5B1%5D.pdf>.
316 HORTSCIENCE VOL. 54(2) FEBRUARY 2019
