Session 3 Genetics, Provenance Trials, and Vegetative Propagation

Session Papers

Provenance Trials

Variability of Faidherbia albida in Progeny Trials in Burkina Faso

A. Billand¹

Abstract

Faidherbia albida progenies were selected from provenances of Kongoussi (Burkina Faso), Matameye (Niger), and Kagnobon (Senegal) through the 1985-87 IRBET/CTFT field provenance trials. The Kongoussi provenance was planted in 1987 and the others in 1989 at Gonsé station near Ouagadougou, Burkina Faso. Single-tree balanced complete block designs were used. Variability of height and collar diameter between progenies is analyzed. Yearly values for individual and provenance heritability for growth (height and collar diameter) and branching (length, number, and height to first branch) are presented. Heritability decreased regularly from 6 to 42 months for the Kongoussi provenance, but this provenance consistently had the highest heritability values. Matameye had the lowest values. Branching parameters had low heritibility values (under 10%).

Introduction

Breeding strategy of Faidherbia albida at IRBET was defined by de Framont (1985). Range-wide provenance trials were established in 1985,1986, and 1987 at the Gonsé and Dindéresso field stations to compare 35 provenances from 8 countries (IRBET/CTFT Annual Reports 1985-88). Early results led to new trials with 25 other provenaces on three climatically different sites in 1990 (Billand 1991). Juvenile growth showed strong heterogeneity within provenances.

Three progeny field trials were set up in 1987 and 1989 to study phenotypic variability, with the objective of proportionally quantifying genetic and site effects. The first progeny trial was set up in 1987 to examine the Kongoussi provenance (used in our studies as a local control) from Burkina Faso (Billand 1988). Two other progeny trials were set up in 1989 to study provenances showing the best growth (Matamaye, Niger and Kagnobon, Senegal).

Materials and Methods

Seeds and Site

Seed for the trials was obtained from various seed centers (Table 1). All three trials were established at the Gonsé station, 30 km east of Ouagadougou. Gonsé has an altitude of 315 m, average annual rainfall of 800 mm, and average temperature of 28 C (maximum 35°C, minimum 22'C). The ferruginous soils have a clayey-sand texture, with a buried laterite layer.

Experimental Design

Trials consisted of 20 progenies with single-tree balanced complete block designs (i.e., each block had one tree from each progeny). The Kongoussi provenance had 20 progenies in 24 blocks, Matameye 19 progenies in 41 blocks, and Kagnobon 21 progenies in 40 blocks. Tree spacing was 4 m x 4 m, and planting was done in Jul 1989.

Table 1. Information of the Faidherbia albida progenies, Gonsé, Burkina Faso, 1985-87.

Measurements and Analysis

Measurements of the 1987 trials were taken every December/January at 6, 18, 30, and 42 months. Heights were measured with a cm ruler; collar diameter with a mm calipers. The 1989 trials were measured at planting and at 6 and 18 months. Branching variables (height to first branch, length of longest branch, number of branches, and canopy diameter) were collected at 6 months.

The analysis of variance was computed on each variable with a General Linear Model. Genetic models for heritability calculations considered each progeny as a half-sib family. Each family had the same mother tree, and pollination was assumed to be panmictic.

Narrow sense heritabilities were computed as follows:

h²_SS = V_A/V_P

where V_A = additive variance
                = 4σ²_A (half-sib hypothesis)
    and V_P = phenotype (total) variance.

Precision within heritabilities was calculated with the formula (Wright 1976) [sic]:

Where b = number of blocks and
             f = number of families.

It should be noted that an unknown proportion of full-sibs and offspring from self-breeding leads to an overestimation of heritability (Kedharnath and Vakshasya 1977).

Results

Mortality

At the age of 42 months, progenies from Kongoussi provenance had mortalities ranging from 6% (progeny 16) to more than 50% (progeny 1). Mortality was low at 6 and 18 months but increased at 30 and 42 months. Progenies from Matameye and Kagnobon had less than 8% mortality.

Growth

Progenies from Kongoussi provenance at 42 months ranged in average height from 1.93 ±10.08 m (progeny 16) to 1.11 ±0.07 m (progeny 4) and collar diameter ranged from 4.45 ±0.59 cm (progeny 16) to 2.47±0.48 cm (progeny 4) (Table 2).

Height increased by 200% between 6 and 18 months, by 50% between 18 and 30 months, and by 30% between 30 and 42 months. The most significant period of diameter growth appeared to be one year later, between 18 and 30 months, representing 50% of the total diameter measured at 42 months.

Progenies from Matameye provenance at 18 months had heights ranging from 0.73 ±0.03 m (progeny 5) to 0.51 ±0.02 m (progeny 12) and collar diameters ranging from 1.40±0.02 cm (progeny 1) to 0.92±0.13 cm (progeny 12). As with Kongoussi provenance, increments between 6 and 18 months were larger for diameter than for height.

Heights at 18 months for Kagnobon provenance ranged from 0.96±0.10 m (progeny 17) to 0.67±0.06 m (progeny 3), diameters ranked from 2.29 ±10.30 cm (progeny 6) to 1.62+0.23 cm (progeny 11). Analysis of variance showed better growth within Kagnobon progenies than within those from Matameye.

Table 2. Heights (±5% confidence interval, in m) of progenies of Kongoussi provenance at 6,18,30, and 42 months. Gonsé, Burkina Faso. 1987-1990.

Branching

Branching parameters had low but significantly different values. Branching parameters, total height, and collar diameter were analyzed with principal components analysis based on 21 progenies of Kagnobon provenance at 6 months. Three groups of variables sufficiently described these 21 progenies. Two growth parameters (height, collar diameter) and two branch length parameters (canopy diameter, length of longest branch) described 53.3% of total variability on axis 1. Height to first branch described 18.9% of variability on axis 2, and number of branches described 11.3% of variability on axis 3.

Heritability

Heritability was calculated from the time of planting to 42 months for Kongoussi provenance (Table 3), and for Kongoussi, Matameye, and Kagnobon at 6 and 18 months (Table 4). Heritability for Kongoussi provenance decreased regularly each year. Heritability for height and collar diameter was highest for Kongoussi and lowest for Matameye.

Table 3. Narrow sense heritability (h2ss±0) for total height and collar diameter for the Kongoussi provenance from 6 to 42 months, Gonsé, Burkina Faso, 1987-90.

Table 4. Narrow sense heritability differences between the Kongoussi, Matameye, and Kagnobon provenances at the time of planting, and at 6 and 18 months, Gonsé, Burkina Faso, 1987-90.

Branching variables from Kongoussi and Matameye provenances had low heritability (values under 0.10). Height to first branch for Matameye was an exception with a narrow sense heritability of 0.28 ±10.03.

Conclusion

Fart of the phenotypical variability observed in field trials can be explained by genetic effects. Values of narrow sense heritability presented here are close to those cited in literature from outside Africa. In Africa, only exotic species such as Eucalyptus spp have been studied. Van Wyk (1977) calculated a heritability of 0.16 for height on 6-month old Eucalyptus grandis in the Republic of South Africa.

A principal component analysis concerning seven growth and branching variables on twenty one 6-month old progenies from Kagnobon provenance showed that three groups of characters described the progenies—growth and branch length; height to first branch; and number of branches.

Yearly increments for collar diameter were strongly reduced for Kongoussi provenance between 30 and 42 months, but height increments at the same ages were not reduced. This was probably caused by competition between the trees. This seemed to vary between progenies and could be a factor for selecting trees for high density field plantations.

The trials studied here were young and had not flowered, so interesting characters such as pod and leaf production or yields of associated understory crops could not be measured. Yearly measurements from youth to maturity will allow the establishment of juvenile-adult correlations that could be used to shorten breeding cycles.

Progeny of mother trees from these provenances are well identified and their seeds are available. They could be used to examine variability at the progeny level with electrophoresis, as has been done at the provenance level (Zeh-Nlo 1989).

References

Billand, A. 1988. Première évaluation d'un essai comparatif de descendances d' Acacia albida à l'âge de 6 mois au Burkina Faso. Technical Note no. 88/04, IRBET/CTFT. Ouagadougou. Burkina Faso: Institut de recherche en biologie et écologie tropicale/Paris, France: Centre technique forestier tropical. 4 pp. (Limited distribution.)

Billand, A. 1991. Le point sur cinq ans d'essais comparatifs de provenances de Faidherbia albida I I'lRBET/CTFT au Burkina-Faso. Cahiers scientifiques no. 11, CTFT. Nogent-sur-Marne, France: Centre technique forestier tropical.

Billand, A., and De Framont, H. 1990. Variabilité génétique de Faidherbia albida (Syn. Acacia albida) en essais comparatifs de provenances au Burkina Faso. Presented at the Symposium on Physiologie des arbres et arbustes en zones arides et semi-arides, 26 Mar to 6 Apr 1990, Nancy, France.

De Framont, H. 1985. Eléments de réflexion pour l'élaboration d'une stratégie nationale de sélection et d'amélioration génétiques des arbres forstiers au Burkina Faso. Technical Note No. 85/4. Ouagadougou, Burkina Faso: Institut de recherche en biologie et écologie tropicale/Paris, France: Centre technique forestier tropical. (Limited distribution.)

IRBET/CTFT. 1985-1988. Rapports annuels d'activité. Ouagadougou, Burkina Faso: Institut de re-cherches en biologie et écologie tropicale/Nogent-sur-Marne, France: Centre technique forestier tropical). (Limited distribution.)

Kedharnath, S., and Vakshasya, R.K. 1977. Estimates of components of variance, heritability and correlations among some growth parameters in Eucalyptus tereticornis. Pages 667-676 in Proceedings of the Third World Consultation on Forest Tree Breeding, 21-26 March 1977, Canberra, Australia. Canberra, Australia: Commonwealth Scientific and Industrial Research Organization.

Van Wyk, G. 1977. Progress with the Eucalyptus grandis breeding programme in the Republic of South Africa. Pages 639-643 In Proceedings of the Third World Consultation on Forest Tree Breeding, 21-26 Mar 1977, Canberra, Australia. Canberra, Australia: Commonwealth Scientific and Industrial Research Organization.

Wright, J.W. 1976. Introduction to forest genetics. New York, USA: Academic Press, Inc. 463 pp.

Zeh-Nlo, M. 1989. Contribution à l'étude de la diversité génétique d'Acacia albida. Approche par électrophorèse enzymatique. DEA Thesis, University of Nancy, Nancy, France.

Footnote__________

1 Institut de recherche en biologie et écologie tropicale (IRBET)/Centre technique forestier tropical (CTFT), 01 B.P. 1759, Ouagadougou, Burkina Faso.

Billand, A. 1992. Variability of Faidherbia albida in progeny trials in Burkina Faso. Pages 71 -75 in Faidherbia albida in the West African semi-arid tropics: proceedings of a workshop, 22-26 Apr 1991, Niamey, Niger (Vandenbeldt, R.J., ed.). Patancheru, A.P. 502 324, India: International Crops Research Institute for the Semi-Arid Tropics; and Nairobi, Kenya: International Centre for Research in Agroforestry.