PROJECT: The Australian Mango Breeding Program
Project Officers: V. Kulkarni1, I. Bally2, S. Blaikie3 and P. Johnson4
Location: CPHRF, Darwin, Southedge Qld and ARS Kununurra WA
NT Department of Business, Industry and Resource Development
Queensland Department of Primary Industries
WA Department of Agriculture
Develop improved mango cultivars for the domestic and export markets through a hand-
pollinated hybridisation program. It was recognised that each region may have some
differences in objectives but many objectives would be common to all regions.
The specific aims of the project were initially to:
• Develop hybrid cultivars with superior fruit quality and production characteristics that are suited to
the various mango-growing regions in Australia.
• Generate a minimum of 50 individual hybrids for each parental combination over three years.
• Generate some quantifiable data on the inheritance of characters and the combining ability of
specific mango cultivars.
The broad objectives of the program have been to develop cultivars with the following characteristics:
Dwarfness – Reduced tree vigour and size is desirable, as Kensington Pride is over-vigorous at the
expense of cropping in the hotter growing districts.
High productivity – Mangoes are generally low producers when compared with other species such as
avocado or stone fruit, and among mangoes, Kensington Pride is a low producer, on average
producing between 5 to 10 tons per hectare.
Fruit size (400 g)– The current domestic market has a preference for fruit in the range of 325 to 400
grams, making up trays of 18 to 20 fruit.
Fruit colour (good blush) – Both Australian and export markets prefer fruit with high blush.
Retention of the KP flavour – The Australian market and many export markets recognise the unique
flavour of Kensington Pride as the cultivar’s greatest asset.
Reduced sap burn and post-harvest problems – One of the greatest post-harvest fruit quality problems
with Kensington Pride is skin browning and sap related injuries. Any reduction in this problem will
significantly increase fruit quality.
Longer shelf-life – The shelf-life of Kensington Pride is relatively short; prolonging storage time will
improve access to export markets.
Reduced physiological disorders in the fruit – Physiological disorders in the current varieties pose a
major fruit quality problem that is hard to control using management techniques. Susceptibility to
specific forms of disorder is cultivar related.
Early maturing – Cultivars that produce fruit earlier than Kensington Pride have a distinct market
advantage, especially in the Northern Territory.
Reduced susceptibility to disease – In certain mango growing districts pre-harvest diseases such as
bacterial black spot seriously limit the production of late cultivars. A reduction in susceptibility to this
and other diseases will improve productivity and fruit quality.
Program of Activities:
The Australian National Mango Breeding Program is a long-term project that has been divided into
four phases of development.
Phase 1, Hybridisation
The hybridisation phase was completed in 1997. A total of 1800 hybrids were generated.
Phase 2, Initial selection and characterisation
The second phase of the program involves the initial screening of hybrids for desirable types and
collecting data on the specific characteristics for inheritance analysis. This screening is being done
using the progeny planted at the two sites, Southedge and Coastal Plains. Southedge was chosen
because of its cooler night temperatures and elevation, which stimulated the hybrids to flower at a
younger age than in warmer areas. Detailed evaluation commenced in the 1999-2000 season and will
continue until 2006. However, most trees will have been evaluated by 2003.
Phase 3, Detailed regional testing
The third phase of the program involves the planting of replicated trials in several agro-climatic regions
to compare and evaluate the most desirable selections from the initial screening (Phase 2). Data
obtained from these trials will be used to evaluate the commercial suitability of selections for the
different growing regions in Australia and to prepare applications for Plant Breeder Rights.
The future of the Program was discussed in Darwin in November 2000, with all participating agencies.
An agreement was reached in April 2001 on a set of core principles in respect of a pathway to
commercialisation of potential new varieties coming out of the Program. These principles are set down
in the following section.
Australian Fresh Mangoes (AFM) were selected as the commercial partner for the project.
Phase 4, Market testing
The fourth phase of the Program consists of market testing and commercialisation of potential hybrids.
This will involve testing in domestic and export markets of commercial quantities of fruit produced in
grower-cooperator orchards. It is expected that contracting with grower cooperators for this purpose
will commence in the second half of 2003. A possible first release of a commercial cultivar could
happen as early as 2006.
Results During 2002-03:
Commercial Agreement with AFM
Negotiations with AFM have been protracted and have resulted in the delay or postponement of some
project activities due to financial constraints. The main effect was a delay of 12 months in establishing
trees on grower cooperator’s farm. A conditional agreement with AFM was signed in late May 2003,
with the expectation of the full agreement to be signed by 30 September 2003. The full agreement is
conditional on the success of an HAL project to manage the further testing and commercialisation of
the hybrids. The proposed project will run for six years with a budget of $100,000 a year. This will free
up resources to commence a program of in-depth data analysis and post harvest evaluations and
market evaluations/commercialisation of the best hybrids.
HAL Project Proposal
Since the signing of the conditional agreement with AFM in May the technical committee members
have been in regular contact with each other and with AFM to discuss the future progress of the
project and the development the HAL project proposal.
Future milestones for the project will be incorporated in the HAL project proposal. They will include
activities such as establishment of hybrids on grower-cooperator farms, establishing a data recording
system and market testing of the selected hybrids.
Replicated trials of the best hybrids will be established on agency Research Stations where
management of the hybrids can be fully controlled. Data from these trials will be used for detailed
genetic analysis of the new hybrids, for registration with Plant Breeders’ Rights and to develop
information packages on the performance and management of any hybrid for commercial release.
Hybrid Evaluation and Selection
The most promising hybrids have been selected on the basis of a combination of characteristics, from
observations made at the three sites where the material has been planted. The selection has an initial
bias towards fruit attributes for which the most information has been gathered, and the elite hybrids
have been placed in two groups (A and B+) according to their inferred potential.
“A” group hybrids have potential for commercial release and are included in replicated trials currently
planted at Southedge and Coastal Plains, with additional trials to be planted at Kununurra and Coastal
Plains. There are currently six trees in the “A” group and they will be offered to growers-cooperators
for regional evaluation.
“B+” group hybrids have been selected as having a highly desirable combination of fruit
characteristics, however they require further evaluation to confirm their commercial potential. Trees in
this group will be retained for further evaluation as single tree selections.
Heritability Evaluation of Mango Characteristics
Continue the evaluation of fruit and tree characteristics of hybrids planted on Southedge and Coastal
Plains Research Stations. The data will be added to the heritability analysis of these characteristics.
Of the 1800 hybrids generated in the project, 400 remain to be assessed at Southedge Research
Station. The remaining 218 of the hybrids assessed in 2002-03, were re-assessments of the “A” and
“B” groups and a subset of hybrids in a range of families assessed to examine the consistency of
performance between years.
At Coastal Plains Horticulture Research Farm, 634 hybrids were assessed in the 2002 season taking
the total number to 1,063. Analyses of these data will give valuable information about the genetic
effects of each family and the reliability of performance between seasons.
Initial estimates of heritability were calculated for some of the fruit characters in a subset of the
families and were presented at the international conference in Brazil in September (see attached
abstract). Conclusions from the analyses were that fruit weight, width, shape, ground skin colour and
pulp depth have high heritabilities – all above 0.60. These characters are likely to be reflected in
progeny. Also, there were indications that if red or burgundy blush was a feature of either parent then
this character was also likely to be shared by the progeny. Similarly, the Kensington Pride flavour of
the male parent was regularly detected in the progeny. Further, more comprehensive analysis of the
existing data, including all families, will be required to confirm these initial findings and to expand the
application of the study.
“A” group replicated trials - The two new “A” group hybrids selected in the 2001-02 season will be
added to the replicated trial blocks at Southedge and Coastal Plains Research Stations. An additional
replicated trial of the “A” group hybrids will be planted at the Frank Wise Research Farm in Kununurra.
During the 2002-03 season’s evaluations, one “A” group hybrid was demoted to “B+” because of its
large fruit size, pitted lenticel appearance and sunken stem end. One “B+” group hybrid was promoted
to group “A” because of its flavour and high yield.
The replicated trial at Southedge Research Station now contains all the current “A” group trees, a few
“B+” group and parents of the “A” group. There is no room for the addition of more trees in this trial
Four of the “A” group trees fruited in the Southedge replicated experiment for the first time in the 2003.
Replicated “A” group trials have been planned for Darwin and Kununurra. Propagation of hybrids for
inclusion in these trials is currently under way.
Budwood supply scheme - A clean budwood scheme is being developed to initially establish a
germplasm repository, supply the replicated trials and provide grower-cooperators with “A” group
hybrids for regional evaluation. This scheme will then be expanded to ensure sufficient high quality
certified disease free budwood is available of all varieties released to ensure the rapid uptake of the
new variety by industry.
Propagation and multiplication of the “A” group hybrids has primarily been undertaken in Kununurra
where a clean budwood orchard has been established. Some additional multiplication of budwood has
also been carried out in Mareeba. Propagation of selected hybrids has also been undertaken at
Katherine Research Station, which is a clean site for the NT. The protracted negotiations with AFM
which prevented the distribution of hybrid budwood to grower-cooperators, have allowed the
accelerated establishment of the clean budwood block at Kununurra, which now has 50 trees planted
in the field of each “A” group hybrid. We are now able to supply adequate quantities of clean budwood
to all grower-cooperators from this point forward. Extension of propagation in the NT will further
accelerate generation of sufficient number of selected hybrids.
Post-harvest disease evaluation - Some preliminary disease screening of the “A” and “B” group
hybrids and their parents will be carried out in the 2002-03 fruit season. Fruit rot will be the initial focus
of this screening to determine if any significant disease resistance is present in the hybrids.
This activity was not carried out over the 2001-02 season due to budget restrictions.
Shelf-life evaluations - Preliminary shelf-life testing of the “A” group hybrids will be carried out in the
coming season to identify hybrids with longer shelf life than the currently grown commercial varieties.
Shelf-life evaluations were not undertaken at Mareeba during the 2002 season due to restrictions on
Market testing - If funding is available some sensory evaluation of the “A” group hybrids will be
carried out in the major domestic and export markets. This information will highlight the particular
requirements of various markets and give the breeding team some guidance in selecting hybrids for
No formal market testing of the hybrids was undertaken over the 2002-03 season. However, Mr P.
Johnson showed a few fruit of a couple of the “A” group hybrids from Coastal Plains to an English
importing company during a visit to London in December 2002. The fruit was well received and the
company wished to be kept informed of future evaluation of the varieties in the European market.
Security - Security recommendations and protocols will be drawn up for the breeding project to
protect intellectual property and preserve commercial confidentiality. These will include
recommendations for security on research stations and information release from the project
Progress has been delayed due to the protracted negotiations with AFM. Research station security is
to be based on the Western Australian model.
Dr Vinod Kulkarni presented a paper titled Inheritance of fruit characters in hybrid mangoes produced
through controlled pollination at the VII International Mango Symposium Recife, Brazil, in September
2002. An abstract is presented below.
Richard I.S. Brettell - CSIRO Peter R. Johnson - AgWA
Vinod J. Kulkarni - NTDBIRD Warren Müller - CSIRO
Ian S. E. Bally - QDPI
Mango fruit quality attributes have been evaluated in hybrids produced by the Australian National
Mango Breeding Program. Since its inception in 1994 the program has generated more than 1800
hybrids from 33 parental combinations using controlled pollination methods. Characterisation of the
fruit was carried out by assessing 24 internal and external attributes, and an estimate was made of the
heritabilities of characters for which the data were either quantitative or were scored on an evenly
ordered hedonic scale. Analysis of the data indicated that many important fruit quality aspects such as
fruit weight, fruit shape, ground skin colour, fruit width and pulp depth have high heritabilities, and can
therefore be readily selected in a breeding program. For non-ordered traits scored in discrete
categories (blush colour, bloom, lenticel colour, embryo type and flavour), an estimate was made of
data consistency from multiple scores for individual hybrids at different times and locations. A relatively
high consistency value was recorded for fruit flavour and in combinations involving Kensington Pride,
between 24% and 47% of hybrids were scored as having Kensington Pride flavour. The embryo type
of hybrids was also recorded and the data are discussed in the context of polyembrony being
controlled by a single dominant gene.
The full paper is available from Dr Vinod Kulkarni.
PROJECT: Improvement of Mango Productivity
Project Officer: V. Kulkarni
Improve productivity in mango by understanding the key factors that affect it and by
addressing limitations imposed by these factors.
Program of Activities:
1. Advise on improvement of flowering, fruit-set and other factors that impact upon production in the
2. Conduct collaborative research with other organisations in several areas including flowering,
which impact upon productivity.
3. Present research results at the VII International Symposium.
4. Present a keynote paper at the VII International Mango Symposium, Recife Brazil 2002.
Mango Flowering – Theories and Practices
While identity of the flowering factor(s) continues to be a mystery, erratic flowering also continues to
be a serious limitation in mango production. Concepts have been developed to attribute flowering to
environmental, genetic, hormonal and nutritional factors. The three factor hormonal theory of flower
promoter, flower inhibitor and bud activity in the floral cycle has however drawn significant support.
What does this theory mean? What information do we have on the flower promoter and inhibitor? Can
we explain the normal observations with the help of this hypothesis? Taking the hormonal theory as a
central theme, is it possible to link the theories together? A wide range of natural and imposed stress
conditions and horticultural practices are reported to induce flowering. Some treatments are reliable in
some situations but not in others. Can we explain these practices, the anomalies associated with them
and other orchard situations with a holistic approach? How can this knowledge be used in developing
strategies for various climatic regions? The review addresses some of the practical issues in the light
of the theoretical background of the 'Tri-factor hypothesis'.
The full paper is available from Dr Kulkarni.