Nghiên cứu chọn tạo và biện pháp nâng cao năng suất, phẩm chất dòng dưa hấu (citrullus vulgaris l ) tam bội in vitro tt tiếng anh

MINISTRY OF EDUCATION AND TRAINING
CAN THO UNIVERSITY

SUMMARY OF DOCTORAL THESIS
Specialization: Crop Science
Code: 62 62 01 10

TRAN THANH TRUYEN

INVESTIGATE ON BREEDING AND MEASURING
TO IMPROVE YIELD AND QUALITY OF
TRIPLOID WATERMELONS (Citrullus vulgaris L.)
IN VITRO

Can Tho, 2019


The thesis was completed at College of Agriculture and
Applied biology, Can Tho University

Scientific supervisors: Prof. Dr. Lam Ngoc Phuong

The thesis is defended in front of the University Examination
Council in Can Tho University
Time:…………..……………
Date:…………….……………..

Reviewer 1:..........................................................................
Reviewer 2:..........................................................................

Further information of the thesis could be found at:

1. Learning Resource Center of Can Tho University
2. National library of Vietnam


LIST OF PUBLIC WORKS
1. Tran Thanh Truyen, Lam Ngoc Phuong and Ngo Phuong
Ngoc. 2014. Effect of Benzyl adenine (BA), Indole butyric acid
(IBA) and activated charcoal on the shooting and emergence of
triploid watermelon (Citrullus vulgaris Schrad.) in vitro.
Journal of Science, Can Tho University 2014: 168-172.
2. Tran Thanh Truyen and Lam Ngoc Phuong. 2017. The result
of planting triploid seedless watermelon lines in Mekong Delta.
Journal of Agriculture and Rural Development (Crop), Volume
1: 108-114.

1


CHAPTER 1: INTRODUCTION
Watermelon (Citrullus vulgaris L.) is one of the most popular crops in the world.
Since the reign of King Hung Vuong, Vietnamese people have been growing
watermelons as one of the traditional crops, till then, the production of watermelons in
Vietnam keeps going on and it was ranked at the fifth in exports (FAOStat, 2012).
Watermelon is an ideal tropical fruit for hot summer days, however, the significant
disadvantage of the diploid watermelons is that they contain too many seeds (Donald,
2012). Therefore, seedless triploid watermelons (3X-triploid) have become a favorite
choice due to its seedless convenience and sweet flavour. Although the price for these
seedless fruits is quite higher than the usual diploid watermelons, it is still acceptable
for the customers (Marr and Gast, 1991; Maynard et al., 2002).
Triploid watermelon is formed by the hybridization method crossing the tetraploid

inbreds with diploid watermelon inbreds (Andrus et al., 1971; Kihara, 1975; Guo et al.,
2011). The tetraploid watermelons are formulated by treating diploid seedlings with
colchicine or oryzalin, in nurseries or in vitro (Kihara, 1951; Raza et al., 2003; Noh et
al. 2012, Lam Ngoc Phương and Nguyen Kim Hang, 2010). Nevertheless, tetraploid
watermelon plants produce a small amount of seeds (lower 10 times than the diploid
watermelon plant seeds) and have low seed rate germination. In other words, growing
tetraploid watermelon plants requires high investment costs (Tran Khac Thi et al.,
2008). According to Compton and Gray (1992) and Compton et al., (1993), the method
which adapts tissue culture techniques in tetraploid and triploid lines shows that it can
reduce the time for propagation (1-2 years) and more convenient for the preservation
process than the traditional one.
In the Mekong Delta, seedless watermelon production has been increased recently
but its production expense is high-priced because of using imported varieties mostly,
expensive seeds due to its high-demanding in preservation and losing germination
easily (Marr and Gast, 1991). As a consequence, it is clearly that growing new seedless
watermelon plants in Vietnam has become a necessary mission. Therefore, I decided to
carry out the research: "Investigate on breeding and measuring to improve yield and
quality of triploid watermelons (Citrullus vulgaris L.) in vitro".
1.1 Aims of the thesis:
Main objective: to hybrid triploid seeds of Vietnamese origin.
Specific objectives:
(1) produce the tetraploid watermelon with colchicine and oryzalin in vitro.
(2) find the suitable environment for cloning and evaluate the ability to grow and
develop the tetraploid watermelons in vitro, and to hybrid triploid seeds in the field
conditions.
(3) find the suitable medium for clonal culture and select triploid watermelon lines
for in vitro conditions, evaluate the growth, development, yield and fruit quality of two
triploid (3x) cultures in field conditions.
(4) invest the effects of nitrogen fertilizer rates and plant density on yield and water
quality of the tissue-cultured TriP1 triploid watermelon line.

1.2 Meaning of the thesis:
i) To produce triploid watermelons originated from Vietnam.
ii) To apply tissue and cell culture techniques to shorten selection time, multiply
lines and reduce the cost of tetraploid and triploid seedlings.
2


iii) To identify the suitable environment for propagation of tetraploid and triploid
watermelons.
iv) To investigate and evaluate the growth and fruit quality of two triploid
watermelon lines forming transplanted tissue in the field.
v) To have initial evaluation of nitrogen fertilization and planting density on TriP1
triploid watermelons for field planting.
1.3 New findings of the thesis
i) Creating tetraploid seedlings by tissue culture and multi-nucleotide treatment;
planting trees in the field.
ii) Hybriding triploid hybrids from tetraploid flowers with diploid pollen.
iii) Finding in vitro culture medium suitable for shoot multiplication, the formation
of tetraploid, triploid, and rapid multiplication of seedlings; reduce the initial
investment in seedless watermelon production.
iv) Investigating and evaluating the growth and fruit quality of two watermelon
lines forming transplanted tissue in the field in Can Tho and Hau Giang.
v) Initially evaluating nitrogen fertilization rates and planting density is appropriate
for the growth, yield and quality of the transplanted tissue culture in the field.
CHAPTER 2: LITERATURE REVIEW
2.1 The scientific basis to make polyploid with colchicine and oryzalin
The chemical formula of colchicine is C22H25O6N which is extracted from a plant
called Colchium autumnale L. These plants are usually grown on the Mediterranean
coast. The pure colchicine powder is ivory white, soluble in water, alcohol, chloroform.
Colchicine is highly durable and can be sterilized in the autoclave, but it is easily

decomposed in the light so it should be stored in the dark. Colchicine is a poison that is
paralyzing and should be very careful when working with it (Tran Thuong Tuan, 1992).
Oryzalin has the chemical formula as C12H18N4O6S, which is a Dinitroaniline
herbicide. It is an orange-yellow solid with a melting point of 141-142°, its molecular
weight is 346.35 and it is dissolved in water at 2.5 ppm under 25°C. Oryzalin can easily
soluble in organic solvents such as acetone, ethanol, methanol and acetonitrile, but less
soluble in benzene and xylene. It is insoluble in hexane. Oryzalin is used to stimulate
multiplicity in many crops (Morejohn et al., 1987). Oryzalin has been shown to inhibit
cell division in plant species and was first registered in the United States in 1974.
The effect of oryzalin on cell division is similar to that of colchicine, which affects
the disulphite bond of the protein and the ribose molecule of the ribonucleic acid. In the
process of cell division, the decongestant stops working because it forms a
microtubule-chemical complex, which prevents later division of the cell and the later
partisation of the cell while chromosomes and chromate formations are still normal. In
the middle of the meridians, the chromosomes are not distributed on the equatorial
plane of the spindle, the chromosomes are shortened and thickened. Disagreement of
the opposite-polar chromosomes in the lateral phase did not occur, the cell did not
divide, although the number of chromosomes was doubled. The size of these cells is
larger than that of normal cells. When stopping the action of chemicals, the cell is able
to divide normally and form tetraploid cells (Tran Thuong Tuan, 1992; Dolezel et al.,
2004). However, the activity associated with microtubules of oryzalin is tighter than
colchicine. Plants treated with oryzalin at micromolar concentrations (μM) gave the
same results as those treated with colchicine at millimolar concentrations (Dolezel et
3


al., 2004). On the other hand, oryzalin mainly penetrates through the cut surface and
enters the epidermis and through the cuticle on the surface of the leaf cells (Allum et
al., 2007).
2.2 Determination of polyploidy

Shape characteristics, growth and development: leaf morphology is a useful
criterion for identifying tetraploid sources before recombinant selection (Jaskani et al.,
2005). The tetraploid watermelon leaves are larger, thicker and darker than the diploid
plants’ (Kihara, 1951). In addition, the length width ratio of tetraploid leaves
(1.00±0.02 and 1.02±0.02) were lower than those of diploid leaves (1.28±0.02 and
1.30±0.02) on both Giza 1 and Giza 21 watermelons under in vivo conditions (Nasr et
al., 2004), the tetraploid leaf area was larger than the diploid leaves of all observed
varieties. The TPS watermelon has larger flowers, the fruit weight is also large, average
4 kg and has about 70 seeds/fruit, dark red fruit pulp, hard and sugar ratio about 12.5%
(Karchi et al., 1981). In addition, shell thickness and number of seedlings of tetraploid
plants were higher than those of diploid (Ahmad et al., 2013). Flowering time is also
later than diploid (Pradeepkumar, 2011).
Stomatal coefficients/leaf area and stomatal length: the difference in stomatal
number is the criterion to distinguish diploid (2n) and tetraploid (4n) plant cells as well
as to identify polymorphisms (Hamill et al., 1992). The size of stomatal cells on
tetraploid sugarbaby watermelon (as determined by nucleotide DNA analysis) is greater
than that of diploid plants (Thayyil et al., 2016). This method is fast, inexpensive, does
not require sophisticated equipment and has a high accuracy rate (in some cases up to
90%, Cohen and Yao, 1996). However, this is just an indirect method for evaluating
polyploidy. If the plant is in the state of polyploidy, this method is not reliable. It
should be combined with other methods (Chen et al., 2006).
Counting Chromosomes: Chromosomal counting is one of the most direct and
accurate methods to determine polyploidy levels, but early and preliminary selection
will be time-consuming. In addition, chromosome counting will also be timeconsuming, laborious, and difficult to perform on plants with large chromosomes and
small cell sizes. In particular, this method is difficult to implement for watermelons
because of the small chromosome size (Jaskani and Khan, 2000).
Analysis of flow cytometry with Partec Ploidy Analyzer. The method of analysing
flow cytometry is more advantageous than chromosome counting (Leus, 2005;
Loureiro et al., 2005). It is convenient to sample preparation and is quick to carry out
because it requires only a small amount of sample. One of the reasons flow cytometry

is more advantageous is because it only requires a small sample size and is quick to
prepare. The method is a simple analytical process that performs analysis on multiple
samples at the same time. Different types of tissue can be analyzed such as leaves,
roots, stems, petals, seeds and fruits, etc. without cell division stage identification. The
cost of sample analysis by analysing the flow cytometry is acceptable. The initial cost
for the analyzer is noticeable, however, the materials and chemicals involved are
inexpensive. Currently, in the world, flow cytometry analysis is one of the most
important tools in evaluating the multiplicity of produced seed specimens, and many
researchers study into this method (Faten et al., 2012; Jaskani et al., 2005). However,
this method meets many obstacles such as expensive equipment and unobservable
characteristics of the chromosomes.
4


2.3 Triploid watermelons (seedless watermelons): triploid watermelon was first
produced by Kihara and Nishiyama in 1939 by using colchicine to produce 4x form and
hybridize with diploid pollen. For watermelon to grow to normal size without hollow
hearts inside it is necessary to perform artificial pollination (Mark Arena, 2012;
Johnson, 2014). Therefore, it is essential to plant alternately between 25% and 33% of
diploid watermelon plants to ensure fruit pollinating success (Olson et al., 2012;
Fiachino and Walters, 2003). Triploid watermelon hybridization requires four steps: (1)
selection of diploid varieties, (2) production of tetraploid plants, (3) development of
pure tetraploid and diploid varieties and (4) cross-breeding of triploid hybrids
(tetraploid mother plant x diploid pollen) and planting triploid hybrids (pollination will
complement diploid pollen) (Wehner, 2008).
2.4 The Role of Micropropagation in Crops: micro-propagation is a powerful
tool to multiply herbaceous plants (medicinal plants, fruit trees, food crops, ornamental
plants, etc.), reducing the cost of F1 hybrids, especially for triploid hybrids such as
seedless watermelon. It is highly beneficial to compare with the use of other traditional
methods, contributing to the protection of food security and the fight against global

climate change. This method has overcome the limitations of annual seed production
such as: poor grain yield, high price, low seed germination and weak seedlings
(Nguyen Bao Toan, 2005; Lam Ngoc Phuong, 2012).
2.5 Impacts of planting density and nitrogen fertilizer on yield and quality of
watermelons: watermelon planting density varies in each region and with each variety.
The shorter the distance, the higher plant density, which will increase the yield of
seedless watermelons compared to low density (Motsenbocker and Arancibia, 2002;
Walters, 2009; Strang et al., 2005). Fertilizers are significant in increasing yield and
quality of watermelons. Providing adequate protein will result in effective
photosynthesis, strong growth and dark green leaves. Increasing the nitrogen
concentration will raise the yield of fruit, Brix degree and lycopene content in
watermelons. However, when the nitrogen content is too high, watermelon plants will
grow strong leaves, but may affect tree stiffness. As a result, the plants become
succulent, so they are vulnerable to pests and diseases. The plants will also fail to fruit,
and the fruits will take long time to ripen, become tasteless and are difficult to store
after harvested. Protein deficiency results in poor growth, shortness, small leaves and
small fruits (Pham Hong Cuc, 2007).
CHAPTER 3: MATERIALS AND METHODS
Experimental study consists of 4 main contents (Figure 3.1) and each period has its
own experiments:
3.1 Induction of tetraploid watermelon in vitro by using colchicine and oryzalin
on diploid watermelon varieties.
3.1.1 Experiment 1: Induction of tetraploid watermelon in vitro by using
colchicine.
Materials: watermelon seeds of two clean-bred diploid varieties including TPS,
TPT. In which: TPS variety is derived from Sugar Baby variety with round or oval dark
green fruit and red flesh. TPT variety is originated from the breed of Thanh Long
variety with oval green fruit with dark green stripes and red flesh. These varieties were
provided by the Biophysics-Biochemistry Laboratory. The watermelon seeds of the two
clean-bred diploids were sterilized and cultured in MS environment in vitro.

5


Experimental design: the experiment is carried out under the methodology of one
completely randomized factor formula, which includes 3 levels of colchicine
processing time (4, 6, 8 days) and one controlled factor (without treatment) with 5
reduplications, each repeats with 2 pots, each pot with 10 top bud samples in vitro.
Monitoring indicators: indicators on the effects of colchicine on growth (shoot height,
shoot/sample quantity, leaf quantity) and on tetraploid plantlets identification after
treatment (amount of stomata/mm2, the proportion of plantlets with polyploidy
phenotype and determination of tetraploid watermelon by flow cytometry)
3.1.2 Experiment 2: Induction of tetraploid watermelon in vitro by using
oryzalin.
Materials: watermelon seeds of four clean-bred diploid varieties including TPS,
TPB, TPT, TPX. In which: TPS and TPT varieties are used as in experiment 1. TPB
variety is originated from Bao Long variety with oval dark green fruit with stripes and
red flesh. TPX variety is originated from Xuan Lan variety with oval light green fruit
with green stripes, yellow flesh. These varieties were provided by the BiophysicsBiochemistry Laboratory. The watermelon seeds of the four clean-bred diploid varieties
were sterilized and cultured in MS environment in vitro.
Experimental design: the experiment is carried out under the methodology of two
completely randomized factors formula, which includes 4 diploid watermelon varieties
TPS, TPB, TPT, TPX and 2 levels of processing oryzalin (48 and 54 hours) and one
controlled factor (without treatment) with 5 reduplications, each repeats with 2 pots,
each with 10 top bud samples in vitro. Monitoring indicators: similar to experiment 1.
3.2 Experiment 3: Evaluation of shooting and growth of two tetraploid
varieties TPT and TPS in medium supplemented with BA
Materials: the tissue-cultured single shoots of two tetraploid watermelon varieties,
TPT and TPS, were selected by analysing flow cytometry, cultured in MS environment
3 weeks before the materials experiment.
Experimental design: the experiment is carried out under the methodology of two

completely randomized factors formula with 6 treatments including 2 tetraploid
watermelon varieties TPT and TPS, and three BA concentration levels (0; 0.5; 1 mg/L)
and 3 reduplications, each with 2 pots, each pot with 4 samples. Monitoring indicators:
number of shoots increased/sample, shoot height, leaf/sample increased.
3.3 Experiment 4: Evaluation of shooting and growth of two tetraploid
varieties TPB and TPX in medium supplemented with BA
Materials: the tissue-cultured single shoots of two tetraploid watermelon varieties
TPB and TPX (simlar to experiment 3).
Experimental design: the experiment is carried out under the methodology of two
completely randomized factors formula with 4 treatments including 2 tetraploid
watermelon varieties TPB and TPX, two BA concentration levels (0.5; 1 mg/L) and 3
reduplications, each with 2 pots, each pot with 4 samples. Monitoring indicators:
similar to experiment 3.
3.4 Experiment 5: Evaluation of rooting and growth of tetraploid variety TPT
in medium supplemented with IBA and NAA
Materials: single shoots of tetraploid watermelon variety TPT tissue-cultured in
MS environment after 3 weeks with relatively identical height and leaf quantity.

6


Experimental design: the experiment is carried out under the methodology of two
completely randomized factors formula with 6 treatments, including two IBA
concentration levels (0 and 1 mg/L) and three NAA concentration levels (0; 0.2; 0.5
mg/L) and 3 reduplications for each treatment, each with 2 pots, each pot with 4
samples. Monitoring indicators: root quantity (root), root length (cm).
3.5 Experiment 6: Evaluation of rooting and growth of tetraploid variety TPS
in medium supplemented with IBA and NAA
Materials: single shoots of tetraploid watermelon variety TPS tissue-cultured in
MS environment after 3 weeks with relatively identical height and leaf quantity.

Experimental design: the experiment is carried out under the methodology of two
completely randomized factors formula with 6 treatments, including two IBA
concentration levels (0 and 1 mg/L) and three NAA concentration levels (0; 0.2; 0.5
mg/L) and 3 reduplications for each treatment, each with 2 pots, each pot with 4
samples. Monitoring indicators: similar to experiment 5.
3.6 Experiment 7: Evaluation of rooting and growth of four tetraploid varieties
in medium supplemented with IBA
Materials: single shoots of four tetraploid varieties tissue-cultured in MS
environment after 3 weeks with relatively identical height and leaf quantity.
Experimental design: the experiment is carried out under the methodology of one
completely randomized factor formula, including 4 tetraploid watermelon varieties with
3 reduplications, each reduplication with 2 pots, each pot with 4 samples. Monitoring
indicators: similar to experiment 5.
3.7 Experiment 8: Evaluation and selection of tetraploid varieties tissuecultured and hybridization of triploid seeds on fields
Materials: seedlings of four tetraploid watermelon varieties domesticated on
experimental fields in Can Tho University.
Experimental design: the experiment is carried out under the methodology of one
factor of latin square formula including 4 tetraploid watermelon varieties tissuecultured with 4 reduplications in such a way that each reduplication has all 4 tissuecultured
tetraploid
watermelon
varieties
correlative
to
10
plantlets/varierty/reduplication. Experimental location: Hau Giang province.
Monitoring indicators: growth indicator (vine length, number of leaves/vine quantity,
and fruit indicators (successful hybridized fruit quantity, fruit weight, yield) and fruit
qualities (fruit pale thickness, Brix degree, triploid seeds/fruit/variety).
3.8 Experiment 9: Evaluation of shooting and growth of triploid varieties
tissue-cultured in medium supplemented with BA

Materials: Collected from experiment 8, single shoots of triploid watermelon seeds
are sterilized and transferred into MS environment in vitro. In which, TriP1, TriP2,
TriP3, TriP4 are hybridized seeds from tetraploid mother varieties TPX, TPT, TPB and
TPS.
Experimental design: The experiment is carried out under the methodology of two
completely randomized factors formula with 16 treatments including four BA
concentration levels (0; 0.5; 1 và 2 mg/L) and 4 triploid watermelon varieties. Each
treatment has 3 reduplications, each reduplication with 2 pots, each pot has 4 samples.
Monitoring indicators: number of shoots increased, shoot height increased

7


3.9 Experiment 10: Evaluation of shooting and growth of TriP1 triploid
watermelon variety in medium supplemented with BA and activated charcoal
Materials: tissue-cultured TriP1triploid watermelon variety
Experimental design: The experiment is carried out under the methodology of two
completely randomized factors formula with 8 treatments including four BA
concentration levels (0; 0.5; 1 và 2 mg/L) and two activated charcoal concentration
levels (0 và 2 g/L). Each treatment has 3 reduplications, each reduplication with 2 pots,
each pot has 4 samples. Monitoring indicators: number of shoots increased, shoot
height increased.
3.10 Experiment 11: Evaluation of rooting and growth of the three triploid
watermelon varieties in vitro in medium supplemented with IBA
Materials: 3 tissue-cultured triploid watermelon varieties: TriP2; TriP3; and TriP4
Experimental design: the experiment is carried out under the methodology of one
completely randomized factor formula including 3 triploid watermelon varieties (TriP2;
TriP3; and TriP4) in environment supplemented with IBA 2 mg/L with 3
reduplications, each reduplication has 2 pots, each pot has 4 samples. Monitoring
indicators: root number/sample quantity, root length, shoot height.

3.11 Experiment 12: Evaluation of root development and growth of TriP1
triploid watermelon varieties in medium supplemented with IBA and activated
carbon
Materials: tissue-cultured TriP1 triploid watermelon variety.
Experimental design: the experiment is carried out under the methodology of two
completely randomized factors formula with 10 treatments, including five BA
concentration levels (0; 0.2; 0.5; 1 and 2 mg/L) and two activated charcoal
concentration levels (0 and 2 g/L) with 3 reduplications. Each reduplication has 2 pots,
each pot has 4 samples. Monitoring indicators: root/sample quantity, root lenght, shoot
height.
3.12 Experiment 13: Evaluation of the growth, yield, productivity and quality
of two triploid watermelon (3x) varieties tissue culture in the field.
Materials: seedlings of tissue cultured triploid watermelon varieties: TriP1, TriP2,
TriĐC (control) domesticated on experimental fields in Can Tho Univeristy. In which,
TriĐC is Mat Troi Do seed, which belongs to Syngenta Company (growing time: 60-62
days (dry season) or 65-67 days (rainy season), average weigth: 3-4kg/per fruit, Brix
degree: 12-13%, red flesh, high adaptability, can be planted year round, easy to fruit).
Experimental design: the experiment is carried out under the methodology of one
completely randomized factor formula including 3 treaments and 2 hybridized triploid
watermelon varieties (TriP1, TriP2) respectively and 1 tissue cultured control seed
variety (TriĐC) after 3 weeks of domestication, with 4 reduplications, each
reduplication with 20 plantlets/variety. Experimental location: Can Tho (autumn-winter
crops, 2013) and Hau Giang (winter-spring crops, 2013). Monitoring indicators: similar
to experiment 8.
3.13 Experiment 14: Study on the impact of nitrogen fertilizer (N) content and
planting density on yield and quality of TriP1 triploid watermelon tissue cultured
Materials: seedlings of tissue-cultured TriP1 triploid watermelon variety
domesticated on experimental fields in Can Tho Univeristy.

8

Experimental design: the experiment is carried out under the methodology of two
completely randomized factors formula including 4 treaments with 2 content figures of
nitrogen fertilizer (150 and 200 kg N/ha) and 2 planting density figures (8.750
plantlets/ha and 10.000 plantlets/ha). Each treatment has 3 reduplications, each
reduplication with 7-8 plantlets correlative to planting density.
N fertilizer in experimental design is combined with P (200 kg P/ha) and K (150 kg
K/ha) and is fertilized 3 times: basal fertilizing: ¼ total content, initial top dressing: ½
total content, growing fruit top dressing: ¼ remain content. In particular, use Phu My
nitrogen fertilizer (46% N) for N, Phu My big grain kali fertilizer for K and Long
Thanh fine-grained phosphorus fertilizer for P. Monitoring indicators: similar to
experiment 8.
3.4 Data analysis methods: data were processed on excel and the statistical
analysis of experimental data was conducted using the SPSS 16 software. Duncan test
was used to compare of experimental averages for randomized complete block design
and completely randomized design, then the LSD test was used for the 1, 7, 8 and 13
experiment.
3.5 Experimental layout diagram
(1) Making tetraploid watermelons in vitro

Treatment with colchicine (4, 6, 8 days).

Treatment with oryzalin (48, 54 hours).
2) Multiply the tetraploid watermelons in vitro
Rooting: medium with IBA and NAA
(different concentrations) (3 experiments)

Shooting: medium with BA (different
concentrations) (2 experiments)

(2) Investigation of 4 tetraploid watermelon lines of tissue culture in the field and
hybridization of triploid seeds (1 experiment)
(3) Multiply the triploid watermelons in vitro
Shooting: medium with BA (different
Rooting: medium with IBA (different
concentrations) +/- activated charcoal
concentrations) +/- activated charcoal
(2 experiments).
(2 experiments)
(3) Investigation of 2 triploid watermelon lines of tissue culture in the
field (Can Tho, Hau Giang) (2 experiments).
(4) Investigation of nitrate fertilizers (N) dose and planting density on the
TriP1 triploid watermelon tissue culture (1 experiment).
Figure 3.1: Experimental layout diagram

CHAPTER 4: DISCUSSION RESULTS
4.1 Induction of tetraploid watermelon in vitro by using colchicine and oryzalin
on diploid watermelon varieties.
4.1.1 Amount of stomata/mm2: average stomatal index of diploid watermelon was
higher than that of polyploid in all of four watermelon varieties. Specifically, the TPS
9


variety had an average stomatal index on diploid plantlets of 233.7 ± 24.4
stomata/mm2, while the TPB was 333.3 ± 24.3 stomata/mm2; TPT was 273.3 ± 17.8
stomata/mm2; TPX variety was 324.2 ± 25.7 stomata/mm2; while the average stomata
index in lower polyploidy plantlets: TPS variety was 139.5 ± 19.1 stomata/mm2, TPB
variety was 187.6 ± 27. stomata/mm2, TPT variety was 184.4 ± 27.6 stomata/mm2, TPX
variety was 208 ± 27.2 stomata/mm2. The results showed that the polyploidy plantlets

had less stomatal index due to the longer stomatal length than that of the diploid
plantlets. This result is consistent with the study of Lam Ngoc Phuong and Nguyen
Kim Hang (2010). Nguyen Van Hien (2000) also reported that the stomatal index of
TPS diploid watermelon is higher than that of polyploidy plantlets, but the stomatal
size is shorter.
4.1.2 Polyploidy TPS and TPT watermelons rate (%) after processed with
colchicine: polyploidy plant rate was determined by polyploidy phenotype (big stems,
big and dark green leaves) combined with counting stomatal index to regconize
polyploidy plantlets. The results showed that high rates of polyploid formation occured
in colchicine treatments, and vice versa. Specifically, for TPS variety, the rate of
polyploid was highest in the 4-day colchicine treatment regimen (32%) and this rate
decreased as the colchicine treatment time increased. In which, 6 days treatment was
18% and 8 days was 12%. For the TPT variety, the highest polyploid rate was observed
in the 6-day colchicine treatment (25%), in the 4-day colchicine treatment, which had a
lower rate (19%) and 8-day treatment for the lowest rate (12%).
4.1.3 Polyploidy TPS, TPB, TPT và TPX watermelons rate (%) after processed
with oryzalin: polyploidy plant rate after processed with oryzalin is determined by
polyploidy phenotype (big stems, big and dark green leaves) combined with lower
stomatal index compared with control shoots. The results showed that high rates of
polyploid formation occured in oryzalin treatments. The rate increased when the
processing time increased compared with unprocessed treatments (0% polyploidy
plantlets), statistically significant difference 1%. Specifically: TPS variety, in 48 hours,
reached 20.7% polyploidy plantlets and in 54 hours reached 29.1%. Similarly, TPB
variety reached 12.9% in 48 hours and 16.7% in 54 hours. TPT variety, in 48 hours
reached 12.5% and 54 hours reached 19.3%. TPX variety, in 48 hours of processing
reached 15.1% and 54 hours reached 21.4%.
4.1.4 Determination of tetraploid watermelon by flow cytometry
Treatment with oryzalin: after flow cytometry testing, leaf samples in oryzalin
treated treatments were analyzed with a 1C peak of 62.78-average cell DNA content in
a total of 5 plantlets analysis. In the 54-hour treatment, the tetraploid count was 4%

with a 1C peak of 121.67 (double DNA content compared with diploid), which was 7%
of the 74 polyploid. In the treatment of oryzalin 48 hours did not give the tetraploid but
only the multiply plants accounted for 12% of the 59 polyploid analysis.
Treatment with colchicine: The results of the analysis showed that in the noncolchicine treatment the samples were diploid. In treatments with colchicine 4 and 6
days, there were no tetraploid and polyploid (corresponding to 114 specimens and 83
specimens analyzed). In the 8 day colchicine treatment, the tetraploid plant was 9%
complete in 44 plantlets for analysis.
In conclusion, the multiplicity of diploid watermelon shoots with colchicine and
oryzalin resulted in tetraploid and polyploidy. The results also showed that tetraploid
10


ratios increased with increased treatment time in both chemicals. In that, with short
treatment time on both chemicals do not create tetraploid, but with 8 days colchicine
makes up 9% tetraploid and with oryzalin in 54 hours make up 4% tetraploid, tested by
flow cytometry. This is similar in the study by Nasr et al. (2004) with a duration of 9
days for high tetraploid yield with colchicine levels of 2000 and 2500 μM/L and
oryzalin concentrations of 100 μM/L, while at the same concentrations but with 3 and 6
days without tetraploid plant. However, this rate was low and similar in Koh's study
(2002). The tetraploid effect was 3.3% -5.5% when treating watermelon shoots with
oryzalin (5-60 μM).
4.2 Evaluation of shooting and growth of two tetraploid varieties TPT and TPS
in the medium supplemented with BA
4.2.1 Number of shoots increased: at 3 weeks after culture (WAC), Table 4.1
showed that the tetraploid watermelons had a statistically significant increase of 5%, of
which TPS was 3.1 shoots and TPT gaining 2.1 shoots. Concentration of BA also
influenced the number of shoots increased, statistically significant difference of 1%. In
which, medium supplemented with BA 0.5 mg/L and 1 mg/L gave higher shoots (3.33.6 shoots) than non BA supplemented shoots for the lowest shoots 1.0 shoots.
Interaction between tetraploid watermelon and BA concentration for shoots increased
from 0.7 to 4.3 shoots, but not statistically significant.

Table 4.1: Number of shoots increased in tetraploid watermelon
medium BA (mg/L) at different concentrations at 3 WAC.
BA concentration (mg/L)
Tetraploid watermelon (A)
0
0,5
TPT
0,7
2,4
TPS
1,4
4,3
Average (BA)
1,0b
3,3a
FA
*
FBA
**
FA xBA
ns
CV (%)
36,9

on supplemented culture

1
3,3
3,9
3,6a

Average (A)
2,1b
3,1a

Note: Numbers with the following letters are the same in the same column or row or in the column and
row are not statistically different from the Duncan test; ns: not statistically different, *: statistically
different at 5% significance level, **: statistically different at 1% significance level.

4.2.2 Shoot height: at 3 WAC, shoot height increased between two distinct
tetraploid watermelons at a significance level of 5%, of which the TPT line was 1.25
cm and the TPS line was 0.27 cm. The medium with or without supplement BA
increased shoot height not significantly different, ranging from 0.34 cm to 1.17 cm.
Similarly, there was no interaction between BA levels and tetraploid watermelons on
shoot height, ranging from 0.23 cm to 2.12 cm. In conclusion, medium supplemented
with BA levels of 0.5 mg/L and 1 mg/L gave higher shoot multiplication, but height
and leaf numbers were not significantly different from those without supplementation.
Similar results were found in Nguyen Thi Phuong Thao et al. (2010) and Veysi
Okumus et al. (2011).
4.3 Evaluation of shooting and growth of two tetraploid varieties TPB and
TPX in medium supplemented with BA
4.3.1 Number of shoots increased: at 3 WAC, BA concentrations affected the
number of shoots increased, in which the medium supplemented with BA concentration
11


1.0 mg/L achieved 2.2 shoots higher than that of the medium supplemented with 0.5
mg/L BA gain 1.4 shoots, statistically significant difference at 1% level. However,
between two different tetraploid watermelons, the number of shoots increased without
statistically significant difference (1.8 shoots). Similarly, there was no interaction

between tetraploid watermelon and BA concentration on the number of shoots
increased.
4.3.2 Shoot height increased: at 3 WAC, medium supplemented with BA 0.5 and
1 mg/L gave rise height of 0.16 cm and 0.27 cm respectively, but not statistically
different. Similarly, between two different tetraploid watermelon varieties, the shoot
height was not statistically different. Interactions between tetraploid watermelon and
BA concentrations did not affect the height of shoots.
Thus, the medium supplemented with BA 0.5 and 1.0 mg/L gave effective shooting
on both tetraploid TPB and TPX lines. In addition, the interaction between BA
concentration and tetraploid watermelon for shoot multiplication is not statistically
significant. The above results showed that MS medium supplemented BA 1 mg/L
suitable for shoot multiplication on tetraploid lines under in vitro conditions.
4.4 Evaluation of rooting and growth of tetraploid variety TPT in medium
supplemented with IBA and NAA
4.4.1 Root number: Table 4.2 showed that at 3 WAC, supplemented with 1 mg/L
IBA supplementation for high root numbers (6.3 roots) was statistically significantly
different at 1% additional IBA (1.9 roots). However, medium without NAA gave the
highest number of roots (6.5 roots) statistically significant difference at 1% level
compared with medium supplemented with NAA (0.2-0.5 mg/L) for the lowest root
numbers (2.4-3.6 roots).
The interaction between IBA and NAA auxin influences root formation, IBA 1
mg/L does not incorporate NAA for the highest number of roots (12.8 roots),
statistically significant difference 1% compared with the remaining treatments (control
treatment for the lowest roots-0.2 roots).
Table 4.2: Root number of TPT tetraploid watermelon shoot on IBA and NAA
supplemented media at different concentrations in 3 WAC.
NAA (mg/L) concentration
IBA concentration
(mg/L)
0

0,2
0
0,2c
2,0bc
1
12,8a
2,8b
a
Average (NAA)
6,5
2,4b
FIBA
**
FNAA
**
FIBA x NAA
**
CV (%)
33,2

0,5
3,7b
3,5b
3,6b

Average (IBA)
1,9b
6,3a

Note: Numbers with the following letters are the same in the same column or row or in the column and

row are not statistically different from the Duncan test; **: statistical difference at 1% significance
level.

4.4.2 Root length: at 3 WAC, medium supplemented with IBA 1 mg/L for high
root length (3.55 cm), statistically significant difference at 1% level compared to nonIBA medium (1.35 cm). However, with no NAA medium supplemented for high root
lengths (3.05 cm), statistically significant difference was 1% compared with medium
supplemented with NAA (0.2-0.5 mg/L) for root length reaching 2.02-2.28 cm.
12


The interaction between the two types of auxin IBA and NAA influenced the root
length, statistically significant difference of 1%. In which, IBA 1 mg/L did not
incorporate NAA for the highest root length (5.78 cm), statistically different from the
other treatments, in which control treatments for low root length most (0.33 cm). In
conclusion, both IBA and NAA factors affected the growth and development of roots
on the tetraploid watermelon shoots. In particular, medium supplemented with IBA
1mg/L gave high rooting and root length.
4.5 Evaluation of rooting and growth of tetraploid variety TPS in medium
supplemented with IBA and NAA
4.5.1 Root number: at 3 WAC, medium supplemented with IBA 1 mg/L, the result
showed that there were large number of roots (8.1 roots) which was statistically
significant difference of 1% compared with the control medium (4.5 roots). NAA
concentrations affected the number of roots on the shoots of TPS tetraploid
watermelons, in which medium supplemented with 0.5 mg/L NAA gave the highest
number of roots (7.2 roots) statistically significant difference 1% compared with the
control medium (4.8 roots), but not different from the medium supplemented with
NAA 0.2 mg/L (6.8 roots). Interactions between IBA and NAA concentrations had a
significant effect on the number of roots, statistically significant at 1%. In details, IBA
1 mg/L did not supplement NAA with the highest number of root (9.0 roots),
statistically different from the control treatments for the lowest number of root (0.5

roots) and treatments supplemented with NAA 0.2 mg/L without IBA (6.3 root), but
not different from the other treatments.
4.5.2 Root length: at 3 WAC, medium supplemented with IBA 1mg/L for high root
length (3.03 cm), statistically significant difference at 1% compared with no
supplemented IBA medium (1.73 cm). Medium was not supplemented and medium
was supplemented with NAA (0.2-0.5 mg/L) gave non-statistically significant in root
length, ranging from 2.27 cm to 2.51 cm. There was an interaction between two types
of IBA and NAA auxin on root length, 1mg/L IBA did not incorporate NAA for the
highest root length (4.31 cm), statistically significant difference of 1% with all
remaining treatments, in which the control treatment gave the lowest root length (0.43
cm).
In general, the TPS tetraploid watermelons in medium supplemented with IBA
1mg/L gave a high rooting efficiency (number of roots, root length), and the shoot
growth is also good (shoot height, number of leaves). In conclusion, MS medium
supplemented with 1 mg/L IBA gave a high rooting efficiency compared with no IBA
supplemented medium and also higher than NAA supplemented medium (0.2-0.5
mg/L). This result is similar in the study of Khalekuzzaman et al. (2012) which found
that the IBA medium of 1 mg/L gave the highest rooting efficiency (100%), the highest
number of roots (12 roots) and the highest root length (7.0 cm) on Elite F1 watermelon.
Similarly, according to Okumus et al. (2011), the medium supplemented with IBA 1
mg/L also gave high rooting efficiency on the various watermelon lines.
4.6 Evaluation of rooting and growth of four tetraploid varieties in medium
supplemented with IBA
4.6.1 Root number: at 3 WAC, 4 different watermelon lines gave different number
of roots, in which the TPT line gave the highest number (5.0 roots), statistically
significant difference at 1% compared with the other 3 lines (2.0-2.3 roots).
13


4.6.2 Root length: four different watermelon lines also gave different root length at

3 WAC, of which the TPT gave the highest root length (1.38 cm), statistically
significant difference at 5% compared with the other 3 lines.
In conclusion, MS medium supplemented with IBA 1 mg/L gave the effect of
rooting on all 4 tetraploid watermelon lines. However, the difference in rooting
efficiency between the tetraploid lines can be effectively attributed to the formation of
tetraploid lines as follows: TPT>TPX> TPS>TPB.
4.7 Evaluation and selection of tetraploid varieties tissue-cultured and
hybridization of triploid seeds on fields
4.7.1 Vine length, number of leaves: at 21 days after the top had been cut, the
TPB variety gave the lowest plant length (129.32 cm), statistically significant different
at 5% compared with the TPT (208.02 cm) and TPS (205.5 cm), but not different from
the TPX variety (179.91 cm). There was a strong increase in number of leaves in 4
varieties of tetraploid watermelon varieties tissue culture. Specifically, the TPX variety
was 33.1 leaves; the TPS variety was 34.2 leaves; the TPT variety was 33.1 leaves, and
the TPB variety was 30.0 leaves, but between them there is no statistical difference.
Based on statistical results at the time of growth, we can rank order of the growth of 4
tetraploid watermelon varieties as follows: TPS> TPT> TPX> TPB.
4.7.2 Number of hybrids succeeded: using the hybridization method by hand,
after picking to harvest and harvesting all 4 tetraploid watermelon lines produced
triploid seeds. However, the number of succeeded hybrids was low and varied in each
line: the TPX line was 35 trees; the TPS line was 23 plantlets; the TPT line was 21
plantlets; and the TPB line was 15 plantlets out of 40 hybrids per line.
4.7.3 Fruit weight and yield: the fruit weight of 4 tetraploid watermelon lines
ranged from 1.27 kg to 1.44 kg but no significant difference was found between them.
The yields of 4 tetraploid watermelon lines ranged from 22.29 tons/ha to 26.53 tons/ha,
but they were not statistically different.
4.7.4 Fruit qualities: Table 4.3 showed that there was a statistically significant
difference at 1% between the 4 tetraploid lines tissue culture of Brix degree. The TPX
line gave the highest Brix degree (9.3%), statistically different from the other 3
tetraploid lines, with TPS and TPB giving the lowest Brix degree (8.2%).

There was a statistically significant difference at 5% between the 4 tissue culture
tetraploid lines of pale thickness. In particular, the TPB gave the highest pale thickness
(0,86 cm) which was different from the TPT with the lowest fruit pale thickness (0.73
cm), but not different from the TPS (0.85 cm) and the TPX line (0.81 cm).
4.7.5 Triploid seeds/fruit: triploid seeds are seeds with a slightly inflated top on
the shell compared with diploid and tetraploid seeds, often possessing a convex part of
the shell (Figure 4.1). This was the distinguishing feature between triploid and diploid
seeds. Table 4.3 showed that the solid triploid seeds/fruits of the other 4 tetraploid lines
at a 1% level of significance, in which the TPT line gave the highest number of solid
seeds (57.5 seeds), which was statistically different from the remaining lines (34.3-40.0
seeds).
In brief, the four lines of tetraploid tissue culture grown in Hau Giang showed little
difference in their ability to grow (vine length, leaf width, fresh vine weight). Fruit
quality (Brix degree, fruit pale thickness) is statistically different. The combination of
growth, yield and fruit quality showed that the order of the tetraploid lines from high to
14


low is as follows: TPX>TPT>TPB>TPS and all four lines were successfully crossed to
produce triploid seed
Table 4.3: Brix degree (%), fruit pale thickness (cm), and number of seeds/fruit of the four
tetraploid watermelon lines tissue cultures grown in Hau Giang.
Tetraploid lines Brix degree (%) Fruit pale thickness (cm)
Number of seeds/fruit
TPX
9,3a
0,81a
37,3b
TPS
8,2c

0,85a
34,3b
b
b
TPT
8,9
0,73
57,5a
TPB
8,2c
0,86a
40,0b
F
**
*
**
CV (%)
1,32
3,89
14,72
Note: Numbers followed by similar symbols in the same column are not statistically different by
LSD; *: statistical difference at 5% significance level, **: statistical difference at 1% significance
level

TPX

TPS

TPB

TPT

Figure 4.1: Triploid seeds produced by the four tetraploid varieties
watermelon
4.8 Evaluation of shooting and growth
of triploid varieties tissue-cultured in
medium supplemented with BA
4.8.1 Number of shoots increased: at 3 WAT, incremental shoots of 4 triploid
lines differed statistically at 1% significance level. The TriP1 line showed the highest
number of shoots (3.3 shoots), which is statistically different from TriP2 and TriP4, but
not significantly different from TriP3. Concentration of BA increased the number of
shoots increased, statistically significant difference was 1%, in which the medium
supplemented with BA 0.5 mg/L gave the highest increase (3.9 shoots) difference was
found between BA 1 mg/L (3.7 shoots) and non-BA supplemented with the lowest
shoot growth (0.8 shoots), but not different from that of the medium supplemented with
BA 2 mg/L (3.8 shoots). There was a significant interaction between triploid
watermelon with BA concentration on shoot increment, statistically significant
difference at 1% level. The TriP3 lines, whose medium was supplemented with BA 1
mg/L and TriP4 tissue cultured which was supplemented with BA 0.5 mg/L gave the
highest increase (4.7 shoots). Statistically significant difference was found between
medium with no supplementation of BA in all four triploid lines tissue culture,
demonstrating the lowest increase (0.7-1.0 shoots) observed.

15


4.8.2 Shoot height increase: at 3 WAC, increase in shoot height displayed by the 4
lines of triploid watermelons had statistically significant difference at 1% level. The
TriP2 line gave the highest increase in height (1.26 cm), which was statistically
different from the other three lines. Concentration of BA influenced height gain

whereas medium without BA supplementation demonstrated the highest increase in
height (1.36 cm), statistically significant difference at 1% level compared to medium
supplemented with BA (0.5; 1, 2 mg/L). The interaction between BA concentration and
triploid watermelon yields maintain statistically significance at 1% increase in height.
Treatment of TriP2 line tissue culture on medium not supplemented with BA gave
highest shoot height of 2.71 cm and reduced shoot height in treatments with
supplemented BA (0.5-2 mg/L) on all 4 triploid watermelon lines.
In hindsight, medium supplemented with BA growth regulator (0.5-2 mg/L) gave 34 times higher shoot multiplication than non-supplemented medium in all 4 lines of
triploid watermelon, especially MS medium supplemented with BA 1 mg/L for high
shoot multiplication. In addition, growth was also different in different triploid lines, in
which the TriP1 and TriP2 lines gave relatively higher shoots but leaf numbers and
shoot height compared with the other two lines.
4.9 Evaluation of shooting and growth of TriP1 triploid watermelon variety in
medium supplemented with BA and activated charcoal
4.9.1 Number of shoots quantity: in 3 WAC, BA concentrations in cultured
medium affected the number of shoots increased by triploid watermelons. Specifically,
at 2 mg/L BA gain 2.9 shoots, statistically significant difference of 1% compared with
control and 0.5 mg/L BA (1.4-2.6 shoots) but not significantly different from medium
supplemented with BA 1.0 mg/L (2.8 shoots). Activated charcoal influences the
multiplication factor. The average of non-supplemented medium was 2.6 shoots,
statistically significant difference of 1% compared to medium supplemented with 2.3
shoots. There was a significant interaction between BA and activated charcoal
concentrations at the 1% significance level. Specifically, the BA treatment of 2 mg/Lno activated charcoal the highest number of shoots (3.3 shoots) compared to the other
treatments but not significantly different from the BA treatment of 1.0 mg/L-no
activated charcoal (3.0 shoots). In addition, the control treatment was not supplemented
with carbon and did not supplement BA with the lowest shoot (1.2 shoots).
Thus, in general, the medium was supplemented with carbon to reduce the
multiplication factor. This may be due to the fact that activated charcoal adsorbs the
number of shoots, but also absorbs organic compounds, growth regulators in the culture
medium (George, 1993). In micro-propagation tetraploid and diploid watermelon, MS

medium supplemented with 1 mg/L BA was highly effective for shooting (Kapiel et al.,
2004). However, the number of shoots obtained was low compared to many reports of
watermelon shoots (Compton et al., 1992) (5-11 shoots); Lam Ngoc Phuong and
Nguyen Bao Ve (2006) (6-8 shoots). Although, treatment with 2 mg BA/L gave the
most shoots, but the shoots was buds. The use of high concentrations of BA for shoot
propagation has been recommended by scientists for morphological disturbances, such
as the formation of multiple shoots, shoot shoots, the formation of different forms of
leaves and stems. Due to the high dose of chemicals added to the environment. The first
experiments on triploid watermelon BA concentration of 5 mg/L produce abnormal
shoots (Lam Ngoc Phuong and Nguyen Bao Ve, 2006).
16


4.9.2 Shoot height quantity: at 3 WAC, it was found that activated charcoal did
not affect the shoot height of TriP1 triploid watermelon. However, the BA
concentration influenced the shoot height difference at 1% level. Specifically, at the
control level for the highest increase (1.4 cm), the statistical difference was 1% level
compared with medium supplemented with BA concentration (0.5-2 mg/L). The field
has a BA concentration of 2 mg/L for the lowest increase (0.94 cm). This means that
when the BA increases, the height decreases. This result was also reported earlier when
the growth regulator of height of watermelon shoot growth was limited (George, 1993).
In general, 1 mg/L BA was suitable for multiplication of TriP1 triploid watermelon. At
this point, exogenous BA has effect on the shooting, stimulating side shoots so the
culture pattern gradually grow small shoots then develop leaves. However, BA shoot
stimulation was not much as the plantlets were in the rejuvenation stage, so the non-BA
medium gave higher leaf height and number.
In conclusion, for both experiments, the medium supplemented with BA gave high
shoot multiplication efficiencies, 2-4 times higher than that without BA
supplementation. According to Vu Van Trong et al. (2007) and Le Duy Thanh (2000)
noted that in addition to the nutrient supply to the culture medium, the addition of one

or more growth regulators is necessary to stimulate growth and development.
Organizing, providing good vitality for the tissues, effecting on protein synthesis and
enhancing the activity of some enzymes. However, this will lead to the phenomenon of
abnormal growth, small leaves, crooked, loss of chlorophyll (Debergh and Maene,
1981) and to reduce this phenomenon must reduce the cytokinin content in the cultured
medium or in combination with the auxin group (Compton and Gray, 1993).
Therefore, the MS medium supplemented with 1.0 mg/L BA was suitable for the
shoot growth of triploid watermelon, the number of shoots healthy, without the
hyperhydricity, although the shoot height was relatively low due to cytokinine. The
focus was on lateral tissue fragmentation, rather than on cell stretching and different
shoot multiplication effects on different triploid melon varieties. At the same time,
concentration of BA 1.0 mg/L and activated charcoal 2.0 g/L gave outstanding number
of leaves, good tree height after 3 weeks of culture.
4.10 Evaluation of rooting and growth of the three triploid watermelon
varieties in vitro in medium supplemented with IBA
4.10.1 Root number: at 3 WAC, there was root formation in all three triploid
watermelon lines cultured on MS medium supplemented with IBA 2mg/L, statistically
significant difference at 5% level. TriP2 gave the highest number of roots of 2.9
distinct roots compared to the other two.
4.10.2 Root length: at 3 WAC, the root lengths of 3 different watermelon triploid
tissue cultures were statistically significant at 1% level. In particular, TriP2 gave the
highest root length of 2.68 cm different from the other two varieties.
4.10.3 Shoot length quantity: there was a statistically significant difference of 1%
after 3 weeks of culturing on the height increment of 3 triploid watermelon varieties. In
particular, the TriP2 variety gave the highest shoot height (1.59 cm) different from the
other two triploid varieties. In conclusion, culture of 3 triploid watermelon varieties in
vitro in MS medium supplemented with 2 mg/L IBA resulted in rooting effect.
However, the root effect (number, length) was low and the effect was different between

17

different triploid varieties, in which TriP2 gave the highest number of roots, root
lengths with shoot height and leaf number.
4.11 Evaluation of root development and growth of TriP1 triploid watermelon
varietie in medium supplemented with IBA and activated carbon
4.11.1 Root number: Table 4.4 showed that in 3 WAC, there were statistically
significant differences in the number of roots between different IBA concentrations and
the presence or absence of activated charcoal supplementation. In particular, the
concentration of IBA from 0.5 to 2 mg/L gave the highest number of distinct roots of
5% compared to the other two concentrations. The roots had higher (2 g/L) roots (6.4
roots) than the non-activated charcoal medium (3.2 roots), statistically different at 5%.
There was a statistically significant 5% interaction between IBA and activated carbon.
In which the IBA 0.5 and 1 mg/L supplemented with 2 g/L gave the highest number of
roots (7.2 and 7.5 roots respectively) statistically different for all treatments no carbon
and IBA 0.2 mg/L with carbon, but not with IBA and supplemented with IBA 2 mg/L
supplemented with activated charcoal 2 g/L.
Table 4.4: Root number of TriP1 triploid watermelon varietie in different medium
supplemented IBA and activated charcoal after 3 WAC
Activated charcoal (g/L) (A)
IBA
concentration
Average (B)
(mg/L) (B)
0
2
e
abc
0
1,6

6,0
3,8b
0,2
1,9e
5,2bcd
3,5b
cd
a
0,5
4,5
7,2
5,8a
cd
a
1,0
4,3
7,5
5,9a
2,0
3,8d
6,3ab
5,1a
b
a
Average (A)
3,2
6,4
F(IBA)
*
F(than)

*
F(IBA × coal)
*
CV (%)
27,57
Note: Figures followed by the following symbols are the same in the same column or row or in the
column and row, which are not statistically different from the Duncan test; *: statistically significant
difference at 5% significance level.

4.11.2 Root length: there was a statistically significant differences in root length
after 3 weeks of culture between the mediums with 6.61 cm carbon and 2.08 cm with
no carbon; as well as high levels of IBA (0.5, 1, 2 mg/L) and low IBA. The results are
consistent with many results in vitro. There was a significant interaction between IBA
and activated charcoal concentrations on root length, statistically significant difference
at 5% significance level. The 2 g/L supplementation carbon gave the highest number of
roots compared to the non-carbon supplemented medium with IBA concentrations, the
control medium gave the lowest root length.
Thus, there was a significant difference in rooting rates between treatments with
carbon and no carbon, as well as between treatments with high levels of IBA and low
IBA, and this difference was statistically significant. Therefore, in carbonaceous
medium at any IBA concentration, root numbers and root lengths remain distinct from
no carbon supplement. Similarly, in high IBA medium (0.5, 1, 2 mg/L) also gave the
high root number and root length, significantly different from low IBA concentrations,
whether or not there was carbon.
18


4.11.3 Shoot height: at 3 WAC, the shoot height was higher (4.53 cm) than that
without activated charcoal (3.35 cm), no statistically significant difference of 5%. IBA
concentrations from 0 to 2 mg/L gave no statistically significant increase in shoot

height. There was a significant interaction between IBA and activated charcoal
concentrations on shoot height, statistically significant difference at 5% significance
level. In these treatments, supplementation of carbon at all concentrations of IBA and
non-charcoal treatment with IBA 0.2-1.0 mg/L gave the highest shoot height and
control treatments were the lowest (2.6 cm).
In conclusion, in the above two experiments, mediums supplemented with IBA and
2.0 g/L activated charcoal increased the root effect on triploid watermelon compared
with non-IBA medium. Roots as well as root length increased with increasing IBA
levels from 0.2 to 1.0 mg/L but tended to decrease with increasing to 2.0 mg/L, thus
using IBA 2 mg/L to compare the root effect on three triploid watermelon lines for low
root numbers. Therefore, medium containing 0.5 mg/L IBA supplemented with
activated charcoal 2.0 g/L will be the suitable root-stimulating medium for watermelon
triploid in vitro conditions.
A combination of shooting and rooting experiments found that different triploid
varieties gave different shoot and rooting abilities. In particular, the triploid TriP1 and
TriP2 triploid watermelon varieties gave higher propagation and root formation than
the TriP3 and TriP4 varieties with higher shoot height and leaf numbers. Therefore, two
triploid TriP1 and TriP2 varieties were selected for further investigation of the
possibility of growing in the field compared to the “Mat Troi Đo” triploid watermelon
variety of tissue culture.
4.12 Evaluation of the growth, yield, productivity and quality of two triploid
watermelon (3x) varieties tissue culture in the field.
4.12.1 Vine length: Table of results 4.5 showed that in Can Tho, the length of three
varieties/ triploid watermelon tissue culture was statistically significant at 1% at 21
days after the top had been cut. In particular, the TriP1 variety gave the highest vine
length of 242.51 cm, the TriDC variety was the lowest of 166.93 cm. However, in Hau
Giang, the length of the three varieties/triploid watermelon tissue cultures was not
statistically different, ranging from 186.6 cm to 240.4 cm.
Table 4.5: Vine length (cm) and number of leaves/vine of the three varieties of triploid
watermelon tissue culture grown in Binh Thuy, Can Tho and Chau Thanh, Hau Giang.

Cần Thơ
Hậu Giang
Varieties
Number of
Number of
Vine length (cm))
Vine length (cm))
leaves/vine
leaves/vine
TriĐC
166,93c
46,6b
204,2
33,9
TriP1
242,51a
73,6a
186,6
27,7
TriP2
230,19b
68,4a
240,4
34,3
F
**
**
ns
ns
CV (%)

1,14
8,17
11,83
10,31
Note: Numbers with similar letters in the same column are not statistically different by LSD; ns: not
statistically different, **: statistically different at 1% significance level.

4.12.2 Number of leaves/vine: in Can Tho, the number of leaves/vine of the
triploid watermelon varieties differed significantly at the 1% level in 21 days after the
top had been cut. In this, the TriP1 variety gave the highest number of leaves/vine 73.6
leaves, different from the TriDC variety with the lowest number of leaves/vine 46.58
19


leaves, but not different from TriP2 variety. However, in Hau Giang, the number of
leaves/vine of the three varieties/triploid watermelon tissue culture was not statistically
different, ranging from 27.7 leaves to 34.3 leaves (Table 4.5).
Regarding the growth of the two varieties of triploid watermelon compared to the
control variety (TriDC), can be arranged as follows: TriP1> TriP2> TriĐC in Can Tho.
The results of Enujeke (2013) also showed a difference in the length and number of
leaves/vine of the six watermelon varieties studied in 2011-2012. However, the growth
of these three varieties is not different when grown in Hau Giang.
4.12.3 Fruit weight: Table 4.6 showed that in Can Tho, between 3 watermelon
varieties, fruit weight difference is not significant, ranging from 2.14 kg to 2.2 kg.
However, in Hau Giang, TriP1 yielded the highest fruit weight of 2.24 kg, significantly
different at 5% compared to the reference variety (1.73 kg) and TriP2 (1.55 kg).
4.12.4 Yield: Table 4.6 showed that in Can Tho, yields between the three triploid
watermelon varieties were not statistically significant, ranging from 31.76 to 36.54
tons/ha. Similar results were reported by Joe et al. (2012) when comparing 8 varieties
of seedless watermelons. This finding is consistent with Strang et al. (2004) comparing

eight mini seedless watermelon in Kentucky found no statistically significant difference
in fruit weight between them. The study by Lam Ngoc Phuong and Nguyen Thanh
Thinh (2009) also found that both seedless V1 and V2 seedlings in vitro weighed 3.42
kg and 2.85 kg, respectively.
However, in Hau Giang, yields of three different varieties of triploid watermelon
tissue culture were statistically significant at the 5% level. In particular, TriP1 yielded
the highest yield of 36.91 (tons/ha), which was significantly different from the
reference variety (28.49 tons/ha) and TriP2 (25.56 tons/ha).
Table 4.6: Fruit weight (kg) and yield (ton/ha) of three triploid watermelon tissue
culture varieties grown in Binh Thuy, Can Tho and Chau Thanh, Hau Giang.
Can Tho
Hau Giang
Varieties
Fruit weight (kg) Yield (ton/ha)
Fruit weight (kg)
Yield (ton/ha)
TriDC
2,20
36,54
1,73b
28,49b
a
TriP1
2,17
35,48
2,24
36,91a
b
TriP2
2,14

31,76
1,55
25,56b
F
ns
ns
*
*
CV (%)
2,52
9,95
6,65
6,47
Note: Numbers accompanied by similar symbols in the same column are not statistically different by
LSD; ns: not statistically different, *: statistically significant difference at 5% significance level .

4.12.5 Fruit quality
Brix degree: Table 4.7 showed that in Can Tho, all three varieties of watermelon
triploid tissue cultured for Brix degree ranged from 8.73% to 9.69%, the difference was
not statistically significant. Similarly, at Hau Giang Brix degree of 3 varieties of
triploid watermelon tissue culture ranged from 8.52% to 9.67% and also displayed no
difference that is statistically significant. According to Maynard (2001) sweetness is
one of the major quality elements in watermelon and they are related to the TSS. The
combined results from several studies indicate that the Brix degree in watermelon
triploids ranged from 8.31 to 13.4% (Kee and Ernest, 2005).
Fruit pale thickness: Table 4.7 showed that the fruit pale thickness of all three
varieties of the watermelon triploid tissue culture varied from 0.81 mm to 0.95 mm, but
this difference was not statistically significant when grown in Can Tho. Similar results
20

were obtained in Hau Giang. The varieties of three watermelon triploid tissue culture
variations ranged from 0.82 mm to 0.94 mm, but not statistically significant. Some
studies have shown that the watermelon's hull thickness makes the shells less dense and
useful during rough transportation, but the watermelon pale thickness varies with the
variety (Thomas et al., 2012).
Table 4.7: Fruit quality of three triploid watermelon tissue culture varieties grown in
Binh Thuy, Can Tho and Chau Thanh, Hau Giang.
Can Tho
Hau Giang
Varieties
Brix degree Fruit pale thickness Brix degree
Fruit pale thickness
(%)
(mm)
(%)
(mm)
TriDC
8,98
0,81
8,52
0,94
TriP1
8,73
0,96
9,67
0,82
TriP2
9,69
0,85

9,24
0,83
F
Ns
Ns
Ns
Ns
CV (%)
14,77
16,22
4,91
27,72
Note: Numbers followed by similar symbols in the same column are not statistically different by LSD;
ns: not statistically different, *: statistically significant difference at 5% significance level.

From the above results, the growth (vine length, number of leaves/vine) in Can Tho
of TriP1 variety is better than TriP2 and control varieties but in Hau Giang all 3
varieties are equivalent. In fruit weight and yield, in Can Tho all three varieties are
equivalent, but in Hau Giang TriP1 is better than TriP2 and control. The fruit quality
(Brix degree and pale thickness) of all three varieties showed similar results in both
survey sites (Figure 4.2)

Figure 4.2: Cross section of two triploid strains TriP1, TriP2 and TriDC tissue
cultured
4.13 Study on the impact of nitrogen
fertilizer (N) content and planting density

on yield and quality of TriP1 triploid watermelon tissue cultured
4.13.1 Fruit weight: Table 4.8 showed that the difference in nitrogen yield was
statistically significant at 1%, in which nitrogen fertilization of 200 kg/ha resulted in

higher fruit weight (2.20 kg) than 150 kg N/ha (1.33 kg). Planting density also affected
fruit weight, statistically significant difference at 1% level, with 10,000 plantlets/ha for
fruit weight (2.11 kg) higher than low density (8,750 plantlets/ha) (1.42 kg). Similarly,
the interaction between nitrogen content and plant density affects fruit weight,
statistically significant difference of 1%. The M2N2 treatment (10,000 plantlets/ha+200
kg N/ha) gave the highest fruit weight (2.65 kg) statistically different from the other
treatments, including the M1N1 treatment (8,750 plantlets/ha +150 kg N/ha) for the
lowest fruit weight (1.09 kg).
4.13.2 Yield: Table 4.8 showed that the amount of nitrogen that influences the
statistically significant difference in productivity is 1%. In particular, application of 200
21


kg N/ha gave higher yield than 150 kg N/ha. In inorganic fertilizers, protein is the most
important fertilizer, providing sufficient protein to increase photosynthesis activity,
strong growth leads to increased yields, when increasing protein yield to 1%, the yield
will increase 0.25% (Pham Hong Cuc, 2007, Nguyen Le Hiep, 2010). Planting density
affects the statistically significant difference at 1% significance level. In particular,
high density yielded 19.42 tons/ha higher than low plant density of 13.06 tons/ha. This
result was similar in the study of Hoang Thi Thai Hoa et al. (2012) also found that with
planting density of 9,000 plants/ha, the yield and economic efficiency were higher with
planting density of 6,000 to 8,000 plants/ha. Similarly, there was a significant
interaction between nitrogen content and plant density on the yield, statistically
significant difference at 1% significance level. Specifically, M2N2 gave the highest
yield (23.19 tons/ha), statistically different from the other three treatments, with the
lowest yield (10.85 tons/ha) .
Table 4.8: Fruit weight, yield and fruit qualities of TriP1triploid watermelon tissue
cultured with different content figures of nitrogen fertilizer and planting density .
Fruit
Fruit pale

Yield Brix degree
Factor
weight (kg) (ton/ha)
(%)
thickness (cm)
Nitrogen fertilizer (N) N1
1,33b 13,25b
9,18b
0,80
a
a
(kg/ha)
N2
2,20
19,23
10,45a
0,97
M1
1,42b 13,06b
9,51b
0,92
Planting density
M2
2,11a 19,42a
10,12a
0,85
M1N1
1,09d 10,85c
8,96
0,94

M1N2
1,75b 15,28b
10,07
0,90
N x density
M2N1
1,57c 15,66b
9,41
0,65
M2N2
2,65a 23,19a
10,82
1,04
FN
**
**
**
ns
Fdensity
**
**
*
ns
F(N x density)
**
**
ns
ns
CV (%)
3,11

3,03
4,13
22,9
Note: Figures followed by similar symbols are the same in the same column or row or in the column
and row are not statistically different from the Duncan test; ns: not statistically different, *: statistically
different at 5% significance level, **: statistically different at 1% significance level. In particular, N1:
150 kg N/ha, N2: 200 kg N/ha, M1: 8,750 plantlets/ha and M2: 10,000 plantlets/ha.

4.13.3 Fruit quality
Brix degree: Table 4.8 showed that the amount of nitrogen has an effect on the Brix
degree, an accumulated statistically significant difference of 1%. In particular,
fertilization of 200 kg N/ha for Brix degree (10.45%) was higher than that of 150 kg
N/ha (9.18%). A study on atermelon planted in Cantho showed that the amount of
nitrogen fertilizer increased from 100 to 200 kg N/ha, the Brix degree also increased
from 10.84% to 11.16%, but not different. (Tran Thi Ba et al., 2004). Similarly,
planting density also has an effect on Brix degree, which is statistically different at 5%.
Specifically, the density of N1 (10,000 plantlets/ha) for Brix degree was 10.12% higher
than density N2 (8,750 plantlets/ha) reaching 9.51%. However, the interaction between
nitrogen content and plant density for Brix degree varied from 8.96% to 10.82%, but
not statistically significant.
Fruit pale thickness: Table 4.8 showed that application of 150 kg N/ha and 200 kg
N/ha for pale thickness ranged from 0.80 cm to 0.97 cm, but not statistically
22


significant. Planting density for pale thickness ranged from 0.85 cm to 0.92 cm, but not
statistically significant. Similarly, the interaction between nitrogen content and planting
density for the pale thickness varied from 0.65 cm to 1.04 cm; but not statistically
different. In general, the amount of fertilizer of 150-200 kg N/ha with planting density
of 8,750 plantlets/ha and 10,000 plantlets/ha affect fruit yield and yield. Therefore, the

combination of 200 kg N/ha and density of 10,000 plantlets/ha for high productivity
and fruit quality. This finding was consistent with the study by Maluki et al. (2015)
also found that the increase in nitrogen content (40, 80, 120 kg N/ha) for fruit yield,
yield and total dissolved solids was significantly higher in both trials year 2012 -2013
on TPS variety. As protein improves the photosynthetic activity of the plant, it helps
the plant to grow well, lacking nitrogen for short, small leaves, small fruit (Nguyen
Manh Chinh and Nguyen Dang Nghia, 2006; Jalali and Jafari, 2012). In addition,
increasing plant density means narrowing the distance between plants in the same row
will increase the yield of seedlings in all four varieties tested in 2006 and 2007,
although the increase in density increases the cost of production, but they will yield
higher returns after harvest (Walters, 2009).
In retrospect, two varieties of polyploidization colchicine and oryzalin chemicals in
combination with tissue culture technology yielded a tetraploid ratio of 4%-9% as the
primary source of crosslinking triploid watermelon. Micro-propagation in vitro had
overcome the drawbacks of tetraploid and triploid seeds (including difficulty of
storage, poor germination, etc.) and can produce genetically identical seedlings as well
as shorten the time coupled with disease immunity. This approach consistently yields
greater economic benefits than traditional cloning methods (Figure 4.3).
Diploid seeds germination in vitro
Treatment buds with colchicine 0.01% in 8 days or
oryzalin 0.004% in 54 giờ.
Determination of tetraploid watermelon by
flow cytometry
Multiplication of tetraploid watermelon varieties in vitro
- MS+BA (1.0 mg/L)
- MS+IBA (1.0 mg/L)
Planting tetraploid varieties tissue cultured and hybridization of triploid seed
Triploid watermelon seeds
Multiplication of triploid variety in vitro
- MS+BA (1.0 mg/L)

- MS+ IBA (0.5 mg/L) + 2g/L of activated charcoal
Triploid seedlings tissue cultured
23