Proceedings
of The World Avocado Congress III, 1995 106 -113
AVOCADO
BREEDING IN ISRAEL
Emanual Lahav, Uri Lavi and
Chemda Degani
Agricultural Research
Organization,
The Volcani
Center,
Bet Dagan 50250, Israel.
Shmuel Gazit
Hebrew University
Faculty of Agriculture
Rehovot, Israel.
Summary
The breeding project is aimed at producing
new cultivars that significantly over perform present commercial cultivars harvested
in the same season. The ideal cultivar is a high-producer under Israeli growing
conditions, with high-quality fruit of medium size, a small seed, and high
export quality.
The project is based on open pollinations as
well as on controlled crosses performed in caged trees with bees as the
pollination vector. Some of the hybrids are differentiated from selfs by the
use of isozyme, genetic markers. So far about 200 crosses have been made with
about 30,000 seedlings.
Analysis of
36 traits of the progeny seedlings resulted in the following conclusions.
1. Girdling is the best method to shorten the juvenile period in the
avocado.
2. There are significant
statistical differences in flowering age and fruiting age between various
progeny populations
3. The time until first
flowering is the limiting factor in evaluation of the seedlings.
4. In all the 36 traits tested,
no significant differences were detected between grafted and ungrafted
seedlings.
5. Parental selection should
not be based solely on cultivar performance since a significant non additive
genetic variance exist in the avocado.
6. Skin color, flowering group
and anise scent are coded by several loci having several alleles in each locus.
The various phenotypes probably result from various heterozygous combinations
in several loci.
Three
selections have been made and released for commercial planting. These are
'Iriet', Adi' and 'Gil'.
The Israeli avocado breeding project started in
the early seventies. The breeding plots are located at the Akko Experiment
Station in the Western Galilee and at the Agric. Res. Org. in Bet Dagan. Seeds
were collected from crosses and selfings by caging trees under a net, using
bees as the pollen vector. The harvested seeds were sown in a nursery and one
year later transplanted into breeding plots at distances of 4m between double
rows of 2 x 1m. The seedlings were evaluated according to either measurements
or visual scoring about 40 traits were evaluated (Lahav et al. 1995).
To distinguish between hybrids and
self-pollinated seedlings, the progeny were characterized by isozyme analysis
of leaf tissue for the following enzyme systems: leucine aminopeptidase (LAP;
EC 3.4.11.1) (Degani et al., 1986), malate dehydrogenase (NIDH; EC 1.1.1.37)
(Degani and Gazit, 1984), phos- phoglucoisomerase (PGI; EC 5.3.1.9) (Goldring
et al., 1985), phospho-glucomutase (PGM; EC 2.7.5.1.) (Torres et al., 1978) and
triosephosphate isomerase (TPI; EC 5.3.1.1) (Goldring et al., 1987).
The most important obstacle in the project is
the long juvenile period. We have studied the juvenile period of several
progenies and recorded their flowering age and fruiting age.
The mean flowering age ranged between 4.3
years for progeny of 'Rosh- Hanikra 11' x 'Ettinger' and 8.6 years for the
self-pollinated progeny of 'Nabal' (Table 1). Flowering age ranged from 3 to 11
years (one record of first flowering after 14 years); fruiting age ranged from
4 to 11 years (one seedling yielded 14 years after planting). The SD in
flowering age ranged from 0.6 for the self-pollinated progeny of 'Anaheim' to
2.1 for the progeny of 'Tova' x 'Ettinger'. In fruiting age, the smallest SD
was 0.7 for the progeny of 'Anaheim' and 2.0 for the progeny of 'Ettinger' x
'Tova' and 'Tova' x Fuerte'.
To assess the impact of the length of the
juvenile period (flowering age) on the time needed for the first fruit
production (fruit age), the difference between fruiting age and flowering age
was calculated for each seedling (Table 2). Referring only to those seedlings
that both flowered and fruited, 55% to 95% of the seedlings, first fruiting
occurred in the same year as flowering (zero difference). Fewer than 5% of the
seedlings showed a difference of 4 to 6 years. About 5% to 35% of the seedlings
fruiting either 1 or 2 years after first flowering.
The juvenile period was shortened by the use
of autumn girdling (Lahav et al., 1986). Girdling improved each of the measured
parameters, and the earlier the girdling, the greater the effect (Table 3). The
earliest (September) girdling increased the proportion of seedlings flowering
from 47% to about 100%; nearly tripled the flowering intensity (from rating 1.0
to 2.7); highly increased the proportion of seedlings setting fruit (10.8% vs.
65.4%); and, perhaps most significantly of all, resulted in a 7-fold increase
in number of fruits per tree (1.3 vs. 9.3). In breeding evaluation, one fruit
is of little or no value, but 9 fruit permit a significant appraisal.
The findings
in this experiment agreed with our previous observations of differences in the
length of the juvenile period among various crosses (Table 4). Selfed 'Rincon'
seedlings rated highest in all 4 parameters measured. 'Tova' x 'Regina' rated
lowest in the flowering comparisons, while 'Ettinger' x 'Fuerte' seedlings
rated lowest in number of fruit. No significant interaction was found between
seedling origin and girdling date for any of the parameters studied.
Most fruit-tree breeding projects utilize the selection of the best
performing nonjuvenile seedlings, concerning agriculturally important traits.
These seedlings are then vegetatively propagated, usually by grafting, to allow
a more thorough assessment. Since the first-stage selection is carried out on nongrafted
seedlings, it is highly important for the breeder to be aware of any potential
effect on performance caused by either the rootstock or the rootstock x scion
interaction. Obviously, in cases where such an effect is expected, much caution
is needed at the seedling stage. Therefore we have studied the potential effect
of grafting on the performance of avocado seedlings.
Thirty-five traits having gradual trends in
evaluation were compared. For 17 traits no significant differences were found
between the performance of the original seedlings and their grafted duplicates.
These traits were tree size; foliage density; leaf anise flowering time; length
of pedicel; fruit stalk thickness; damage by snap picking; fruit skin gloss;
surface, and ease of peeling; seed weight; flesh fibers, bitterness, sweetness,
and darkening; time from harvest to softening and shelf life. For eight traits
significant main effect differences were detected between the performance of
the seedlings and those of their grafts (Table 5). For the remaining ten traits
a significant interaction was detected between the performance of the original
seedlings and that of their grafted trees (Table 6).
A significant interaction suggests that the
difference in performance between the original seedling and its grafted
duplicates is not a general effect of the rootstock but is rather limited to
some seedlings and their grafts. We had expected environmental effects to cause
some differences between the performance of seedlings and that of their grafts
in traits having low heritability (Lavi et al., 1993) and also to find
differences due to assessment errors by the experimenters, but such differences
were not detected. Rootstocks might have affected scion performance positively
or negatively, especially in regard to productivity. In this study no
significant rootstock effect on productivity was found. In all the cases the
differences between the seedlings and their grafted duplicates was relatively
small in relation to the 2 to 7 degrees of evaluation.
Three cultivars have been selected in the
project. 'Iriet' (Lahav et al., 1989) a relatively small tree, pear shaped
fruit with black slightly pebbly peel with gloss. Fruit weight 300-500 g.
Harvest season February to May; 'Adi' (Lahav et al., 1992) strongly resembles
Hass but light green in color. Average weight 230 g, seed size less than 10%.
Long harvest season from November to April. 'Gil' (Lahav et al., 1995) pear
shaped with a short neck, the skin is black slightly pimpled with medium gloss.
Average weight 300 g. Harvest season January to March.
References
Degani, C. and Gazit, S. 1984. Selfed and crossed proportions of
avocado progenies produced by caged pairs of complementary cultivars.
HortScience 19(2): 258-260.
Degani, C., Goldring, A., Gazit, S., and Lavi, U. 1986. Genetic
selection during the abscission of avocado fruitlets. HortScience 21(5): 1187.
Goldring, A., Zamir, D., and Degani, C. 1985. Duplicated phosphoglucose
isomerase genes in avocado. Theor. Applied Genet. 71: 491-494.
Goldring, A., Gazit, S., and Degani, C. 1987. Isozyme analysis of
mature avocado embryos to determine outcrossing rate in a 'Hass' plot. J. Amer.
Soc. Hort. Sci. 112: 389-392.
Lahav, E., Zamet, D., Gazit, S., and Lavi, U. 1986. Girdling as a means
of shortening the juvenile period of avocado seedlings. HortScience 21:
1038-1039.
Lahav, E., Lavi, U., Zamet, D., Degani, C., and Gazit, S. 1989. Iriet-
A new avocado cultiver. HortScience 24: 865-866.
Lahav, E., Lavi, U., Degani, C., Zamet, D., and Gazit, S. 1992. 'Adi' a
new avocado cultivar. HortScience 27: 1237.
Lahav, E., Lavi, U., Mhameed, S., Degani, C., Zamet, D., and Gazit, S.
1995. 'Gil' a new avocado cultivar. HortScience 30: 158.
Lahav, E., Tomer, E., Gazit, S., and Lavi, U. 1995. Performance of
avocado (Persea americana Mill.) and mango (Mangifera indica L.)
seedlings compared with their grafted trees. J. Amer. Soc. Hort. Sci. 120:
265-269.
Lavi, U., Lahav, E., Degani, C. and Gazit, S. 1992. The genetics of the
juvenile phase in avocado and its application for breeding. J. Amer. Soc. Hort.
Sci. 117: 981-984.