Abstracts
Budwood from seven different clones of rubber tree [Hevea brasiliensis (Willd. ex Adr. de Juss.) Müell. Arg.] was planted in replicated trials in four different test sites in the plateau region of the state of São Paulo, Brazil. The clones represented a range of imported germplasm, whereas the locations were selected to represent a range of rubber growing areas. Girths were measured for six years consecutively, before the initiation of tapping for latex. Total number of latex vessel rings (lvr) and bark thickness (bt) were measured at six years. The largest average for all characters was observed in Matão. Overall means for girth, total number of latex vessel rings and bark thickness at six years were 35.26 cm, 11.30 units and 4.83 mm, respectively. Both the test sites and clones showed statistically significant differences in girth, lvr and bt values. There were significant clone x site interactions. Girth at one year was not a reliable predictor of future field performance. Broad sense heritability for girth was 0.16, and for lvr and bt it was 0.28 and 0.40, respectively, at six years. Repeatability was quite high, from 0.52 to 0.75 for all characters in all years. When two clones out of seven were selected, expected genetic gain in girth was about 2.0%; for lvr and bt it was 7.0% and 14.6%, respectively.
Enxertos de sete clones diferentes de seringueira [Hevea brasiliensis (Willd. ex Adr. de Juss.) Müell. Arg.] foram plantados em quatro diferentes locais da região do planalto do Estado de São Paulo. Os clones em estudo, na sua maioria de origem asiática, e os locais selecionados representam as diferentes regiões de heveicultura do estado. O perímetro do caule foi obtido por seis anos consecutivos, antes do início da sangria para obtenção do látex. O número total de anéis de vasos laticíferos (lvr) e a espessura de casca (bt) foram analisados somente no sexto ano de idade. Os resultados das análises mostraram que os clones em Matão apresentaram as maiores médias para todos os caracteres estudados, com média geral de 35,26 cm para perímetro do caule, 11,30 unidades para lvr e 4,83 mm de espessura de casca aos seis anos. As interações do clone-local foram significativas. O perímetro do caule relativo ao primeiro ano de avaliação não foi eficiente para predizer o desempenho futuro do vigor dos clones no campo. A herdabilidade no sentido amplo no sexto ano de idade para perímetro, lvr e bt foi 0,16, 0,28 e 0,40 cm, respectivamente. As repetibilidades foram altas, com intervalos de 0,52 e 0,75 para todos os caracteres ao longo dos anos. Quando 28,0% dos clones foram selecionados, o ganho genético esperado para perímetro, lvr e bt foram 2,0, 7,0 e 15,0 por cento, respectivamente.
Growth trends, genotype-environment interaction and genetic gains in six-year-old rubber tree clones (Hevea) in São Paulo State, Brazil
Paulo de Souza Gonçalves1, Ondino Cleante Bataglia 2, Wagner Rodrigues dos Santos 3, Altino Aldo Ortolani 4, Ivo Segnini Jr.5 and Edson Hideaki Shikasho6
1 Programa Seringueira, Instituto Agronômico (IAC), Caixa Postal 28, 13001-970 Campinas, SP, Brasil. Fax: (019) 2423602. E-mail: paulog@cec.iac.br. Send correspondence to P.S.G.
2 Seção de Fertilidade de Solos e Nutrição de Plantas, IAC.
3 Programa Seringueira, IAC.
4 Seção de Climatologia Agrícola, IAC.
5 Engenheiro Agrônomo, Gerente de Divisão Seringueira/Cambuhy Empreendimentos Agropecuários Ltda.
6 Bolsista do Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq).
ABSTRACT
Budwood from seven different clones of rubber tree [Hevea brasiliensis (Willd. ex Adr. de Juss.) Müell. Arg.] was planted in replicated trials in four different test sites in the plateau region of the state of São Paulo, Brazil. The clones represented a range of imported germplasm, whereas the locations were selected to represent a range of rubber growing areas. Girths were measured for six years consecutively, before the initiation of tapping for latex. Total number of latex vessel rings (lvr) and bark thickness (bt) were measured at six years. The largest average for all characters was observed in Matão. Overall means for girth, total number of latex vessel rings and bark thickness at six years were 35.26 cm, 11.30 units and 4.83 mm, respectively. Both the test sites and clones showed statistically significant differences in girth, lvr and bt values. There were significant clone x site interactions. Girth at one year was not a reliable predictor of future field performance. Broad sense heritability for girth was 0.16, and for lvr and bt it was 0.28 and 0.40, respectively, at six years. Repeatability was quite high, from 0.52 to 0.75 for all characters in all years. When two clones out of seven were selected, expected genetic gain in girth was about 2.0%; for lvr and bt it was 7.0% and 14.6%, respectively.
INTRODUCTION
The rubber tree [Hevea brasiliensis (Willd. ex Adr. de Juss.) Müell. Arg.] is one of the most recently domesticated species (Gonçalves et al., 1990a) of great commercial importance. Indigenous of the tropical rain forest in the Amazon Basin, rubber has been traditionally cultivated in the equatorial region, in a zone lying between 10o north and 10o south of the equator. Cultivation was mainly confined to this area as it provided appropriate environmental conditions. Rubber began to be cultivated outside this region in Brazil to avoid the main limiting factor of Hevea cultivation, South American Leaf Blight, a disease caused by the fungus Microcyclus ulei.
Rubber tree cultivation has become intensive in places as far as 20o to 21oS latitude, in the State of São Paulo. Such plantations are being established in environments grossly different from these to which the species is best adapted. It has also been observed that genotypes which perform well in one environment may not perform as well in relation to a given set of genotypes in other environments.
Genotype-environment interaction studies began only in the early seventies in Malaysia (Tan, 1995). In Sri Lanka, genotype-environment studies were started in 1975 by Jayasekera (1983, 1984) and by Jayasekera et al. (1977, 1984, 1994). The studies in Indonesia (Daslin et al., 1986), Nigeria (Onokpise et al., 1986) India (Meenattoor et al, 1991) and Brazil (Gonçalves et al., 1990b, 1992, in press) are more recent. Although the importance of genotype-environment effects has been clearly demonstrated, the amount of variation attributed to each kind of interaction (edaphic, climatic, etc.) and the stability and adaptability of individual clones under varying conditions remain to be investigated. This would help to determine the range of conditions under which each clone may be used.
MATERIAL AND METHODS
Clonal material
A large-scale program of clonal propagation of rubber trees in the State of São Paulo was made (Gonçalves et al., in press). The genetic material originally consisted of seven clones (Table I). Some clones, such as PB 235, GT 1, RRIM 600 and IAN 873, are recommended production clones, whereas others like RRIM 701, PR 255 and PR 261 are used as parents in rubber breeding programs. The rubber clones were budded onto established rootstocks (Tjir 1 x Tjir 16) in the nursery. One and a half-year-old rootstock seedlings raised in nurseries were used to budgraft clonal materials. Budgrafting was made on December, 1989. The successful budgrafts were uprooted and planted in polyethylene bags. After the first flush of leaves developed, the plants were established in the field.
Table I - List of rubber clones, parentage and places of origin.
1
GT - Godang Tapen; RRIM - Rubber Research Institute of Malaysia; IAN - Instituto Agronômico do Norte; PR - Proefstation voor Rubber; PB - Prang Besar; Tjir - Tjirandji; FA - Ford Acre.2Control clone.
The sites
During summer, 1990, four contrasting test sites in the plateau region of the State of São Paulo were selected (Table II). These sites represent the most important continental climate non-traditional rubber production area. Chemical analysis was performed on the arable layer of these soils (Table III). The experimental design at each test site was randomized complete blocks with three replications. Thirty plants were used in each plot in all locations except Matão. At Matão, a larger plot size of forty plants was used. Plots thus consisted of five rows of six or eight plants each.
Table II - Geographic, climatic and edaphic characteristics of the four test sites.
1
In accordance with Brasil (1960).2Clay soils with indistinct horizon differentiation developed from basic igneous rocks. The soils are slightly acid to neutral with high base saturation and high contents of iron oxides.
P resin = Resin extractable phosphorus; O.M. = Organic matter; S = Sum of bases; CEC = Cation-exchange capacity; V% = Percentual of soil base saturation.
Measurements
The girth of each tree was measured. In the first year, diameter was recorded because plants were too small to measure the girth. Measurements were converted into girth, assuming that the stem was cylindrical. The diameter of the plants was measured 0.50 m above ground level by a slide caliper. The other five annual measurements were taken at 1.20 m above the highest point of the bud union. Virgin bark samples were removed from the stem, as plugs, at 1.00 m height, at six years. Longitudinal and transverse sections of the bark were taken and stained with Sudam III and phenic acid to determine the total number of latex vessel rings. Bark thickness was measured with a micrometer and the results were expressed in mm.
Biostatistical analyses
a) Analysis of variance (ANOVA)
The ANOVA model in Table IV was applied for overall comparisons. The model was:
Yijk = m + Gi + Pj + (GP)ij + Rk(j) + Eijk
where, Yijk = observed value of the ith clone of the jth site in the kth replication; m = general mean; Gi = fixed effect of the ith clone (i = 1, 2 ... c, c = 7); Pj = effects of the jth site (j = 1,2 ... s, s = 4); (GP)ij = interaction between ith clone with the jth site; Rk(j) = effect of kth randomized block in the jth location; Eijk = experimental error associated with ijkth observation.
Table IV - Degree of freedom (d.f.), mean square (M.S.) and expected mean square (E.M.S.) for two-factor analysis of variance.
s = Number of sites, p = 4.
g = Number of genotypes = 7.
VS = Variance due to site differences.
VG = Variance due to clonal differences.
VGS = Variance due to clone x site interactions.
VE = Environmental variance among plots.
Variance components were calculated through "Expected Mean Square" components given on the right side of the ANOVA table. Tukeys test was applied to find out which entries were different from each other.
b) Correlation coefficients
To determine the degree of association between given character pairs, Pearsons correlation coefficients rxy were calculated by the following equation:
rxy = (Covxy) /
where, rxy = phenotypic correlation coefficient; Covxy = covariance of variable x and variable y; Vx = variance of variable x; Vy = variance of variable y.
c) Heritability and gain estimates
Broad sense heritabilities (
) and repeatabilities of clonal means (R) were calculated for girth, from age one to six, and for bark thickness and total number of latex vessel rings, both at age six years, using the following equation:
where, VG = variance due to clonal differences; VP = VG + VGS + VE = phenotypic variance; VGS = variance due to clone x site interactions; VE = environmental variance among plots.
Similarly, repeatabilities were estimated from algebraic manipulation of the mean square of the table, according to Turner and Young (1969) suggestion, using the following equation:
where, R = repeatability; MSG = mean square of genotype; MSGP = mean square of clone x site interaction; s = number of sites.
Genetic gain (GGs) was calculated by
GGs = i. R.
where,i = selection pressure.
In formulating the above equations, genetic principles, as discussed in Falconer (1981) and Becker (1984), were followed. Several applications of these principles to forest trees were also consulted (Shelbourne and Thulin, 1974; Burdon, 1977; St. Clair and Kleinschmit, 1986).
RESULTS AND DISCUSSION
Test site performances
There were significant differences among test sites for girth at all ages and for bark thickness and total number of latex vessel rings at six years (Tables V and VI). At 1 year, the relative difference between the best (Matão) and the worst stem growth (Ribeirão Preto) site was 67%. This difference steadily decreased and reached the lowest level at age six (26% between Matão and Ribeirão Preto). Starting at four years, each site maintained almost the same rank order over the years. It is difficult to explain the low growth in Ribeirão Preto. The soil is high in nutrients (Table III) and apparently the environment is not much different from the other sites. The low K/(Ca + Mg) ratio in this soil, associated with high pH and zinc deficiencies in the young stage of growth, may be part of the explanation.
Table V - Girth (cm), total number of vessel rings (units) and bark thickness (mm) means through different ages ( in years) at four sites.
Test sites with the same letter for a given age are not significantly different from each other at the 5% level (Tukeys test).
Numbers within parentheses are coefficients of variation.
Table VI - Results of ANOVA (mean squares) for annual girth, total number of vessel rings and bark thickness for all four sites.
Annual girth
Total number of vessel rings Age 6
Bark thickness Age 6
1
Girth, total number of vessel rings and bark thickness are expressed in cm, unit and mm, respectively.*P < 0.05; **P < 0.01; ns = Not significant.
Total number of latex vessel rings and bark thickness followed more or less the same rank pattern as girth at six years (Table V). The best girth growth test site (Matão) did not always give the greatest bark thickness. Site Indiana had the thickest bark although it ranked the second in girth. The total number of latex vessel rings of the trees was 32% and 18% greater than the overall mean at sites Matão and Indiana, respectively.
The relative difference between the largest girth (clone RRIM 701) and the smallest (clone RRIM 600) girth was 14% at one year (Table VII). At six years, the relative difference was 8%. There were significant differences among the clones for girths at all ages, and for total number of latex vessel rings and bark thickness at six years (Table VI).
Table VII - Overall clonal means of girth (cm) at different ages, and total number of latex vessel rings (unit) and bark thickness (mm) at six years.
1
Rank order at six years.
Starting at age four, clones PB 235 and GT 1 ranked 1st and 2nd. They may be called "consistently top performers" (Tabela VII). Clones IAN 873 and RRIM 600 steadily improved their relative ranks, and thus they had "consistently upgraded performance". Clones RRIM 701 and PR 261 did not maintain their relatively better earlier positions, and fell below the overall average at six years. They were" consistently degrading performers". Clone PR 255 was a consistently poor performer (Table VII).
Total number of latex vessel rings and bark thickness characters did not follow the same ranking pattern as girth at six years. Overall, clones PR 255, PB 235 and GT 1 were among the top for total number of latex vessel rings and clones PR 255, PR 261 and PB 235 were the top for bark thickness (Table V). Clones PB 235, GT 1, IAN 873 and RRIM 600 were relatively higher, but were not equally the best in total number of latex vessel rings and bark thickness.
Clone x site interaction
Significant clone x site interactions were observed for girth (at all observed ages) and total number of latex vessel rings and bark thickness at six years (Table VI). Plantation means of clonal girths, total number of latex vessel rings and bark thickness and their rank orders at six years were determined (Table VIII). Close examination of rank orders of the top three clones shows that clones PB 235, GT 1 and PR 261 were among the "most stable" (least interacting) and clones PR 255, RRIM 600 and IAN 873 were among the least stable (most interacting) clones in their performance across the test sites. Most clones maintained their relative rank orders more or less at the same level across the sites.
Table VIII - Clonal means of girth (cm), total number of latex vessel rings (lvr) and bark thickness (bt) and their rank orders ( R) for each of the four test sites at six years.
Total number of latex vessel rings and bark thickness showed similar clone x site interaction trends as girth at age six. As expected, these three characters were highly correlated (Table IX).
Table IX - Correlations matrix of Hevea characters at different ages.
ns = Nonsignificant; * = P < 0.05; ** = P < 0.01.
lvr = Number of latex vessel rings.
Age to age correlations
Girth of the trees at one year was significantly correlated with girth only up to age four (Table IX). Therefore, the first year girth of clones was not necessarily a good indicator of girth in later years. Only after age two (i.e., second growing season in the field) did tree girth in early years start to show statistically significant correlations with the girths in later years. There were also very high and stable correlations among plant girths at ages 3, 4 and 5. On the other hand, girths were not necessarily a good indicator of total number of latex vessel rings and bark thickness characters (Table IX). Only bark thickness showed statistically significant correlations with the total number of latex vessel rings at six years.
Age dependence of variance components
The percentage contributions of variance components were determined (Table X). Girth was the lowest value component. On the other hand, the error variance (
) increased. These trends in age dependence of VG and VE were similar to those reported by Jayasekera et al. (1994) on a different set of clones of the same species. The relative contribution of clone-environment variance to total variance for girth steadily decreased up to age four, after which it stabilized. This is not surprising, since the clones included in this study represent the best clones from earlier selections, as they are genetically a relatively homogeneous group.
1
Variance abbreviations are the same as in Table IV.2Total number of latex vessel rings.
3Bark thickness.
Studies on the clone-environment variation (VGS) contribution over total variation in Hevea for some characters were conducted by researchers in other countries. In Sri Lanka Jayasekera et al. (1994), who studied 10 clones over eight locations, reported that 19.85%, 17.49% and 10.86% of the total variance components were accounted for by the clone-environment variance component for first, third and sixth year girth, respectively. In Indonesia, Daslin et al. (1986), in a study of 23 clones in two locations, reported that 38.90%, 11.61% and 10.10% of the total variance components were accounted for by the clone-environment variance component of girth, total number of latex vessel rings and bark thickness, respectively.
Estimated heritabilities, repeatabilities and gains
Broad sense heritability (), based on overall evaluation for first year girth, was 0.09 (Table X). The same estimates for girth up to age 6 years varied between 0.10 and 0.16. For total number of vessel rings it was 0.29. These clonal heritability estimates are considered relatively low, mainly due to the relative contribution of to the phenotypic variance. On the other hand, repeatability of clonal means (R), based on overall evaluation, was quite high for girth at all years (Table X, last column).
Genetic gains (GGs) in girth and total number of latex vessel rings (lvr) and bark thickness (bt), at different levels of selection intensity, were also estimated (Table XI). When only the highest girth clone (i.e., only one clone out of seven clones) was selected, with a selection intensity of 0.352, the gain in girth at age six was 0.7571 cm, which is only 2.15% greater than the overall clonal girth means. When only the biggest clone in total lvr, (PR 255) was selected, the gain in total lvr at age 6 was 1.014 units, which is 8.97% higher than overall clonal lvr means. As intensity of selection was relaxed, genetic gains decreased accordingly. For example, when three out of seven clones were selected, the GGs in girth, total lvr and bt were 1.30%; 5.46% and 9.51%, respectively. These gains seem to be small at first sight. However, it should be kept in mind that these values represent gains only from additional clonal selection from previously selected clones, and they do not include gains already achieved from early truncation selections.
Table XI - Genetic gains ( GGs ) at different levels of selection of the top best clones (out of n = 7) at six years of age.
1
n = Number of clones selected.i = Selection intensity (after Becker, 1984).
2lvr = Total number of latex vessel rings.
3bt = Bark thickness.
ACKNOWLEDGMENTS
We thank Cambuh Empreendimento Agropecuários Ltda. and Indiana Rubber Plantation for providing logistical support for the experiments in Matão and Indiana, respectively; Miss Andréa Cardoso Guerreiro for the laboratory work; Mrs. Lígia Regina L. Gouvêa for compiling the data, and Dr. Maria Regina G. Ungaro for providing constructive criticisms of the manuscript. Research supported by Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP), Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) and Secretaria de Agricultura e Abastecimento do Estado de São Paulo (SAA).
RESUMO
Enxertos de sete clones diferentes de seringueira [Hevea brasiliensis (Willd. ex Adr. de Juss.) Müell. Arg.] foram plantados em quatro diferentes locais da região do planalto do Estado de São Paulo. Os clones em estudo, na sua maioria de origem asiática, e os locais selecionados representam as diferentes regiões de heveicultura do estado. O perímetro do caule foi obtido por seis anos consecutivos, antes do início da sangria para obtenção do látex. O número total de anéis de vasos laticíferos (lvr) e a espessura de casca (bt) foram analisados somente no sexto ano de idade. Os resultados das análises mostraram que os clones em Matão apresentaram as maiores médias para todos os caracteres estudados, com média geral de 35,26 cm para perímetro do caule, 11,30 unidades para lvr e 4,83 mm de espessura de casca aos seis anos. As interações do clone-local foram significativas. O perímetro do caule relativo ao primeiro ano de avaliação não foi eficiente para predizer o desempenho futuro do vigor dos clones no campo. A herdabilidade no sentido amplo no sexto ano de idade para perímetro, lvr e bt foi 0,16, 0,28 e 0,40 cm, respectivamente. As repetibilidades foram altas, com intervalos de 0,52 e 0,75 para todos os caracteres ao longo dos anos. Quando 28,0% dos clones foram selecionados, o ganho genético esperado para perímetro, lvr e bt foram 2,0, 7,0 e 15,0 por cento, respectivamente.
(Received March 6, 1997)
- Becker, W.A. (1984). Manual of Quantitative Genetics Washington State University, Pullman, pp. 195.
- Brasil. Ministério da Agricultura. Centro Nacional de Ensino e Pesquisas Agronômicas. Comissão de Solos (1960). Levantamento de reconhecimento dos solos do Estado de São Paulo Rio de Janeiro, pp. 634 (Boletim 12).
- Burdon, R.D. (1977). Genetic correlation as a concept for studying genotype-environment interaction in forest tree breeding. Silvae Genet.26: 168-175.
- Daslin, A., Baihaki, A., Danakusuma, T.M. and Haeruman, M.S. (1986). Genotypes x environment interaction in Rubber and their implications in clonal selection. Bulletin Perkaretan4: 23-28.
- Falconer, D.S. (1981). Introduction to Quantitative Genetics. Edn. Longman, London, pp. 340.
- Gonçalves, P. de S., Cardoso, M. and Ortolani, A.A. (1990a). Origem, variabilidade e domesticaçăo da Hevea: Uma revisăo. Pesq. Agropec. Bras. 25: 135-156.
- Gonçalves, P. de S., Cardoso, M., Colombo, C.A., Ortolani, A.A., Martins, A.L.M. and Santos, I.C.I. (1990b). Variabilidade genética da produçăo anual da seringueira: estimativas de parâmetros genéticos e estudo de interaçăo genótipo x ambiente. Bragantia, 49: 305-320.
- Gonçalves, P. de S., Cardoso, M., Santos, I.C.I., Martins, A.L.M., Ortolani, A.A. and Colombo, C.A. (1992). Selection of Hevea mother tree adapted to impredictable annual climatic variability. Braz. J. Genet.15: 137-147.
- Gonçalves, P. de S., Gallo, P.B., Signini Jr., I., Ortolani, A.A., Brioshi, A.P., Landell, M.G. and Souza, S.R. de Components of variance and genotype x environment interaction for annual girth increment in rubber tree. Pesq. Agropec. Bras. (in press).
- Jayasekera, N.E.M. (1983). A basis for selecting Hevea clones stable to unpredictable agroclimatic variabitily. Silvae Genet. 32: 181-185.
- Jayasekera, N.E.M. (1984). A review of studies on genotype environment interactions in Hevea. In: Proceedings of the International Rubber Conference, 75 Years of Rubber in Sri Lanka (Colombo, 1984) Colombo, RRISL, pp. 111-117.
- Jayasekera, N.E.M. and Karunasekera, K.B. (1984). Effect of environment on clonal performance with respect to early vigour and yield of Hevea brasiliensis. In: Proceedings of the IRRDB Meeting on Hevea Physiology Exploitation and Breeding (Montpellier, 1984). IRRDB, Montpellier, pp. 250-255.
- Jayasekera, N.E.M., Samaranayake, P. and Karunasekera, K.B. (1977). Initial studies on the nature of genotype-environment interaction in some Hevea cultivars. J. Rubber Res. Inst. Sri Lanka 54: 33-39.
- Jayasekera, N.E.M., Karunasekera, K.B. and Kearsey, M.J. (1994). Genetics of production traits in Hevea brasiliensis (rubber). I. Changes in genetic control with age. Heredity 73: 650-656.
- Meenattoor, R.J., Vinod, K.K., Krushnakumar, A.K., Seturaj, M.R., Potty, S.N. and Sinha, R.R. (1991). Clone x environment interaction during early growth phase of Hevea brasiliensis I. Clonal stability on girth. Indian J. Nat. Rubber Res 4: 51-58.
- Onokpise, O.U., Olapade, O. and Mekako, H.U. (1986) Genotype x environment interaction in Hevea brasiliensis (Müell. Arg.) Indian J. Genet. 46: 506-511.
- Shelbourne, C.J.A. and Thulin, I.J. (1974). Early results from a clonal selection and testing programme with radiata pine. N. Z. J. For. Sci.4: 387-398.
- St. Clair, J.B. and Kleinschmit, J. (1986). Genotype- environment interaction and stability in ten-year height growth of Norway spruce clones (Picea abies Karst.). Silvae Genet 35: 177-186.
- Tan, H. (1995). Genotype x environment interaction studies in rubber (Hevea) clones. J. Nat. Rubber Res. Kuala Lumpur 10: 63-76.
- Turner, H.N. and Young, S.Y. (1969). Quantitative Genetics in Sheep Breeding. Cornell University Press, Ithaca, pp. 331.
Publication Dates
-
Publication in this collection
06 Jan 1999 -
Date of issue
Mar 1998
History
-
Received
06 Mar 1997
