Research Article (DOI: http://doi.org/10.15580/GJPBCS.2016.1.111015157)

Assessment of Lobé Oil Palm (Elaeis guineensis Jacq.) Population from Cameroon crossed with Deli Testors

Noumouha Epa N. Ghislain*¹, Allou Désiré², Adon Benjamin³, Konan Jean-Noël², Issali Auguste Emmanuel⁴, Sékou Diabaté², Konan K. Eugène P.⁵ and Simon-Pierre Assanvo Nguetta¹

¹Université Felix Houphouët-Boigny, UFR Biosciences, Laboratoire de génétique, 02 BP 801 Abidjan 01 (Côte d’Ivoire).

²Centre National de Recherche Agronomique (CNRA), Station de Recherche de La Mé 13 BP 989 Abidjan 13 (Côte d’Ivoire).

³Dekel-Oil BP 133 Cidex 03  Abidjan Côte d’Ivoire.

⁴Centre National de Recherche Agronomique (CNRA), Station de Recherche sur le Cocotier Marc Delorme, Port Bouët 07 BP 13 Abidjan 07, Côte d’Ivoire.

⁵Centre National de Recherche Agronomique (CNRA), Direction régionale Abidjan, 08 BP 33 Abidjan 08, Côte d’Ivoire.

INTRODUCTION

The genetic improvement of oil palm (Elaeis guineensis Jacq.) in Côte d’Ivoire is based on a reciprocal recurrent selection scheme. This breeding strategy, adopted on the La Mé station since 1957, uses two groups of populations with complementary production components (Meunier and Gascon, 1972). The Group A is characterized by palms with a small number of large bunches (Deli) and Group B, the La Mé and Yangambi populations (West Africa), includes palms with a large number of small bunches. This selection scheme is comprised of several successive breeding cycles. Each cycle includes progeny tests (Group A × Group B crosses) to identify the best parents, which were then recombined within each group (within-group recombination) to make up the improved populations or basic populations for the next cycle.

The implementation of this breeding strategy made it possible to perform a progression of 60 % on the oil yield in 50 years of selection (Durand-Gasselin et al., 2009). However, this progress was accompanied by a reduction of the genetic base of populations used (Cao, 1995). This situation can hamper genetic progress in the long-term. To widen the genetic base and especially to enrich agronomic qualities of populations of breeding scheme, it is necessary to use additional materials. These new materials are introduced into the reciprocal recurrent selection scheme during the within-group recombination phases (Jacquemard, 1995). These materials derived from surveys of wild oil palm groves and exchanges between research stations. Most often, these populations are evaluated before their introduction into the reciprocal recurrent selection scheme.  It is in this

context that Lobé population, beforehand improved by PAMOL plantations of Cameroon was evaluated in crosses with Deli testors derived from the second cycle of reciprocal recurrent selection scheme. A wide range of agronomic traits of major importance for oil palm breeding was measured, in order to evaluate Lobé × Deli crosses: bunch production components, vertical growth rate and vascular wilt tolerance. This last selection criterion is very important in tropical Africa, where the disease may trigger substantial damages (sékou et al., 2009). This paper describes the characteristics of Lobé × Deli crosses and searches out the best parents of Lobé population able to enrich agronomic traits of populations of Group B valued in the selection scheme.

MATERIALS AND METHODS

Materials

Planting material: Lobé population originated from Ekona region of Cameroon. It was beforehand improved by PAMOL plantations of Cameroon. It was introduced at the La Mé station between 1968 and 1970, in the context of a program of exchange of planting materials, in the form of recombinations between the best parents of PAMOL plantation. Thirteen palms were chosen in these recombinations on the basis of heritable characters (bunch number, percentage of pulp on fruit and vertical growth) to represent this population. These palms were crossed with twenty-three Deli parents to produce forty Lobé × Deli progenies. The Deli parents were chosen in the second breeding cycle of selection scheme. Their agronomic traits being acknowledged, these parents were used as testors. The LM 2 T × DA 10 D and LM 2 T × DA 115 D crosses of the first breeding cycle of selection scheme were used as controls. These hybrids were largely diffused in the village and industrial groves in Côte d’Ivoire.

Study site: This study was conducted at the La Mé station of the CNRA (Centre National de Recherche Agronomique). It is located in the Gulf of Guinea (5° 26’ N; 3° 50’ W; 23 m). This station is located in ombrophilous zone of forest, under an equatorial climate of transition. This climate is characterized by four seasons. The two seasons of rains are dominant (March-June and September-November) and are intercalated by two dry seasons (December-February and July-August). The last 15 years of the station were characterized by an average annual pluviometry of 1600 mm, an average hydrous deficit of 307 mm and a monthly average temperature of 27 ° C.

Methods

Experimental design: All palms of various Lobé × Deli progenies and controls were planted in two trials (LMGP 131 and LMGP 146). These trials were planted in 1994 for LMGP 131 and 1996 for LMGP 146 according to a Randomized Complete Block Design with 6 replications. Each experimental unit, according to trials is comprised of 10 or 12 palms (either 60 or 72 palms per progeny). These palms were planted at the standard density of 143 palms per hectare after the felling of an old palm grove.

Measured parameters

Bunch production components: The bunch production was observed palm by palm for 7 years from the start of bearing (3 years after planting). Bunches were collected on each tree at a frequency from 2 to 3 times per month. The bunch number (BN) and total bunch weight or yield bunch (TBW) were recorded with each harvest. These observations cumulated on each crop year made it possible to determine the average bunch weight (ABW). The bunch production components of the first 3 years (3 to 5 years) correspond to the young age and the following 4 years (6 to 9 years), the adult stage.

Vertical growth rate: The height of stem (H) was measured at 16 years or 18 years (according to the trials) on each straight and healthy palm, starting from the axil of frond 33 to the ground level. The vertical growth rate was calculated from the formula of Jacquemard (1980):

Vc = H / (N - P0)

Vc: vertical growth rate (cm/year)

N: tree age expressed in years and tenths of years

P0: the age of imaginary take off from the soil of leaf 33 of an oil palm, after which the speed of growth is constant (estimated at 3.5 years)

Vascular wilt susceptibility: Each progeny was represented by 160 seedlings arranged in main plots of 20 seedlings in 8 blocks. The preparation of the inoculum, its application in the pre-nursery at 2-leaf stage and the identification of wilt-infected plants were carried out according to the method of Renard et al. (1972). However, the stock monospore 179 was used for the preparation of the inoculum. The assessment of vascular wilt susceptibility of each progeny was carried out starting from its index (I) which was derived from the following formula:

The wilt index of each Lobé parent (Ip) was obtained starting from the mean index of progenies of this latter. A progeny or a parent with wilt index less than 100 is known as tolerant to vascular wilt. It is susceptible in the contrary case.

Statistical analyses: The progenies data were attributed to various crosses from which they result. The descriptive statistics parameters (average and coefficient of variation) and analyses of variance followed by the test of mean comparisons of Newman and Keuls to the risk of 5 % (Dagnelie, 1998) were used on parameters measured to compare Lobé × Deli crosses to the two controls. The mating systems used to provide progenies tested in the two trials consisted of incomplete factorials with several Lobé parents crossed only once. Moreover, Deli testors which were used in the various crosses are not all identical from parent Lobé to another parent Lobé. Aware of these situations, it is not possible to compare statistically general combining abilities (gca) of Lobé parents. The control cross LM 2 T × DA 10 D being planted in the two trials and even in all genetic trials at the La Mé station, it was used to provide a link between the different trials. Thus, for a given character, a classification of performances of Lobé × Deli crosses was carried out in order to identify parents having good gca. A parent had a good gca when crosses carried out from this one were better classified and the performance of each one of them was higher or equal to that of LM 2 T × DA 10 D control. The classification was applied after the test of Newman and Keuls at 5% likelihood. The analyses were carried out with the Genstat software 10.1 (Lawes Agricultural Trust, 2007).

RESULTS

Characteristics of Lobé × Deli crosses

Bunch production components at the young age: In the two trials, Lobé × Deli crosses were characterized by a small bunch number (BN) compared to controls (Fig. 1 and 2). Their average bunch weight (ABW) was at the same level as those of controls. The bunch production or bunch yield (TBW) of these crosses was lower than those of controls. In the LMGP 131 trial, this small bunch production was estimated at 97% of LM 2 T × DA 10 D and at 68% of LM 2 T × DA 115 D. To the level of the LMGP 146 trial, it corresponded to 94% of LM 2 T × DA 10 D and to 85% of LM 2 T × DA 115 D. The variability was substantial between crosses of the LMGP 131 trial for the bunch production. The control LM 2 T × DA 10 D, LM 10975 T × LM 8933 D,  LM 10964 T × LM 10824 D and LM 10975 T × LM 10818 D Lobé × Deli crosses, presented largest yield bunch (Table 1). These crosses differed significantly from LM 2 T × DA 10 D control. LM 10963 T × LM 7116 D got the smallest value of bunch yield.

Figure 1: Distribution of bunch production components at the young age of controls and Lobé × Deli crosses of the LMGP 131 trial (BN: Bunch number, ABW: Average bunch number, TBW: Total bunch weight).

Figure 2: Distribution of bunch production components at the young age of controls and Lobé × Deli crosses of the LMGP 146 trial (BN: Bunch number, ABW: Average bunch number, TBW: Total bunch weight).

Table 1: Classification in homogeneous groups of bunch productions at the young stage of controls and various Lobé × Deli crosses of the LMGP 131 trial.

Legend: TBW: Total bunch weight. Values followed by the same letter are not statistically different according to Newman-Keuls test at 5% threshold.

Bunch production components at the adult stage: Lobé × Deli crosses were also characterized by a small bunch number compared to controls at the adult stage (Fig. 3 and 4). Their bunch was as large as those of controls. The bunch production of these crosses was lower than those of controls. At the level of LMGP 131 trial, it was estimated at 88% of LM 2 T × DA 10 D and at 83% of LM 2 T × DA 115 D. Concerning LMGP 146 trial, it corresponded to 85% of LM 2 T × DA 10 D and to 83% of LM 2 T × DA 115 D. In the LMGP 131 trial, the two controls (LM 2 T × DA 10 D and LM 2 T × DA 115 D) and Lobé × Deli crosses such as LM 10963 T × LM 10815 D, LM 10964 T × LM 10824 D, LM 10973 T × LM 7114 D, LM 10973 T × LM 7034 D, LM 10969 T × LM 10824 D et LM 10975 T × LM 8933 D not differed significantly for bunch yield (Table 2). Regarding LMGP 146 trail, the control LM 2 T × DA 115 D characterized by the largest bunch yield, differed significantly from all Lobé × Deli crosses (Table 3). The control LM 2 T × DA 10 D not differed significantly from LM 10965 T × LM 8933 D.

Figure 3: Distribution of bunch production components at the adult stage of controls and Lobé × Deli crosses of the LMGP 131 trial (BN: Bunch number, ABW: Average bunch number, TBW: Total bunch weight).

Figure 4: Distribution of bunch production components at the adult stage of controls and Lobé × Deli crosses of the LMGP 146 trial (BN: Bunch number, ABW: Average bunch number, TBW: Total bunch weight).

Table 2: Classification in homogeneous groups of bunch productions at the adult stage of controls and various Lobé × Deli crosses of the LMGP 131 trial.

 Legend : TBW: Total bunch weight. Values followed by the same letter are not statistically different according to Newman-Keuls test at 5% threshold.

Table 3: Classification in homogeneous groups of bunch productions at the adult stage of controls and various Lobé × Deli crosses of the LMGP 146 trial.

 Legend : TBW: Total bunch weight. Values followed by the same letter are not statistically different according to Newman-Keuls test at 5% threshold.

Vertical growth rate: The vertical growth rate of Lobé × Deli crosses was higher than those of controls (Fig. 5 and 6). For the LMGP 131 trial, this high growth was estimated at 110% of LM 2 T × DA 10 D and at 116% of LM 2 T × DA 115 D. On the level of LMGP 146 trial, it corresponded to 108% of LM 2 T × DA 10 D and to 123% of LM 2 T × DA 115 D. In this trial, the Lobé × Deli crosses such as LM 10965 T × LM 7047 D, LM 10964 T × LM 7083 D, LM 10966 T × LM 7040 D and LM 5606 T × LM 7047 D not differed significantly from the two controls (Table 4). LM 10966 T × LM 8933 D, LM 10965 T × LM 8933 D and LM 10972 T × LM 9791 D crosses got largest value of vertical growth rate. Concerning LMGP 131 trial, 8 crosses whose vertical growths ranged from 45.69 cm/year to 48.70 cm/year were not significantly different from those of LM 2 T × DA 10 D (46.30 cm/year) and LM 2 T × DA 115 D (43.83 cm/year) (Table 5).

Figure 5: Distribution of vertical growth rates of controls and Lobé × Deli crosses of the LMGP 131 trial.

Figure 6: Distribution of vertical growth rates of controls and Lobé × Deli crosses of the LMGP 146 trial.

Table 4: Classification in homogeneous groups of vertical growth rates of controls and various Lobé × Deli crosses of the LMGP 146 trial.

Legend : Vc: vertical growth rate. Values followed by the same letter are not statistically different according to Newman-Keuls test at 5% threshold.

Table 5: Classification in homogeneous groups of vertical growth rates of controls and various Lobé × Deli crosses of the LMGP 131 trial.

 Legend : Vc: vertical growth rate. Values followed by the same letter are not statistically different according to Newman-Keuls test at 5% threshold.

Susceptibility to vascular wilt: Lobé × Deli crosses were in average susceptible to vascular wilt (Table 6). However, approximately 41 % of these crosses were tolerant to this cryptogamic disease.

Table 6: Vascular wilt susceptibility of various Lobé × Deli crosses

Legend: I: vascular wilt index.

General Combining Ability of Lobé Parents

Bunch production: The LM 10964 T and LM 10975 T parents were characterized by the best general combining abilities (gca) for the precocity of bunch production (Table 7). These 2 parents provided in crosses with Deli testors, best rankings for bunch productions. The additional bunch productions at the young age of these crosses compared to LM 2 T × DA 10 D ranged from 10 to 34 %. However, the performance of cross between LM 10964 T parent and the Deli testor LM 7034 D was low compared to LM 2 T × DA 10 D. At the adult stage, the LM 10964 T and LM 10973 T parents were characterized by the best gca (Table 8). The crosses of these 2 parents with Deli testors had bunch productions at the same level as that of LM 2 T × DA 10 D. However, the performance of cross between LM 10964 T parent and the Deli testor LM 7034 D was lower than that of LM 2 T × DA 10 D.

Table 7: General combining ability for bunch production at the young age of Lobé parents

Legend : TBW: Total bunch weight.

Table 8: General combining ability for bunch production at the adult stage of Lobé parents

Legend : TBW: Total bunch weight.

Vertical growth rate: The LM 10964 T and LM 10969 T parents provided the best gca for this character (Table 9). The crosses carried out with these 2 parents had best rankings. Their growth was at the same level as that of LM 2 T × DA 10 D (between 99 and 104 % of this latter). However, crosses between these 2 parents and the Deli testor LM 8929 D had growths higher than that of LM 2 T × DA 10 D (between 106 and 113 % of this latter). The LM 10970 T parent was characterized by the worst gca. In crosses with Deli testors, this parent provided additional growths ranged from 16 to 17 % compared to that of LM 2 T × DA 10 D.

Table 9: General combining ability for vertical growth rate of Lobé parents

Legend: Vc: vertical growth rate.

Wilt susceptibility: The LM 10964 T, LM 10965 T, LM 10966 T and LM 5606 T parents were tolerant to vascular wilt (Table 10). In crosses with susceptible (LM 8933 D, LM 9791 D, LM 9803 D) and tolerant Deli testors (LM 7034 D, LM 7040 D, LM 7047 D, LM 7083 D, LM 9807 D and LM 9890 D), these 4 parents provided factors of tolerances to their progenies. The LM 10963 T, LM 10967 T, LM 10972 T and LM 10974 T parents were susceptible to this cryptogamic disease. These parents provided in crosses with susceptible (LM 8960 D, LM 8988 D, LM 9349 D, LM 9791 D, LM 9803 D, LM 9818 D, LM 10715 D and LM 11020 D) and tolerant Deli testors (LM 7032 D, LM 7047 D, LM 7089 D and LM 7123 D) factors of susceptibility to their progenies.

DISCUSSION

The study of Lobé population in crosses with Deli material showed that this population provides a low bunch yield compared to controls. This low bunch production is due mainly to their small bunch number because they have bunch as large as those of controls. The high bunch production of which this population made proof during the characterization carried out by Adon (1995) was not observed in crosses carried out with Deli material. However, substantial variability for this character made it possible to make an effective selection at the young age and at the adult period. Indeed, 3 crosses significantly earlier than LM 2 T × DA 10 D controls were identified at the young age. On the level of the adult stage, 8 crosses at the same level that one or the two controls which are references for this character of first crosses cycle (Adon et al., 1998) and even to 2nd cycle were identified. The vertical growth rate of Lobé × Deli crosses was estimated according to trials at 16 years or 18 years. The vertical growth rates of various crosses have been underestimated compared to the usual measurements carried out at 9 years or 10 years. Indeed, according to Jacquemard and Baudouin (1987), the rate of vertical growth gradually decreases from 10 years.

Table 10: Wilt susceptibility of Lobé parents.

Legend: *: The first value between brackets indicates the number of progenies of Deli testor with index less than 100. The second value indicates the contrary case

The rate of vertical growth of these crosses must be higher than those which were estimated starting from these two ages. However, all crosses being planted in the same conditions (in the same trial), comparisons between them are still valid even if observations took place beyond 10 years. The Lobé × Deli crosses were in average susceptible to vascular wilt. In contrast, approximately 41 % of crosses were tolerant to this cryptogamic disease. The transmission of tolerance factors being primarily additive Meunier et al., (1979), these tolerant crosses are due to some Deli testors recognized for their tolerance to the disease (LM 7047 D, LM 7083 D), but also some Lobé parents which provided tolerance factors in crosses with susceptible Deli (LM 8933 D, LM 8960 D). The LM 10964 T parent had a good gca for the bunch production at the young age and at the adult stage. However, its performance in cross with LM 7038 D Deli testor was low compared to LM 2 T × DA 10 D. This low production which was observed is ascribable to LM 7038 D testor who is known to transmit a low bunch production to his progenies. On the basis of the vertical growth rate, LM 10964 T and LM 10969 T parents had good gca. However, their performance in crosses with LM 8929 D Deli testor was high compared to LM 2 T × DA 10 D. This high growth rate which was observed is ascribable to LM 8929 D testor who is known to transmit a high growth rate to his progenies. An analysis of selected parents for their good gca for all parameters measured showed that LM 10964 T parent cumulates the maximum of interesting characters. This parent has a good precocity for the bunch production, a good bunch production at the adult stage, an average vertical growth and is tolerant to vascular wilt. Its exploitation in the breeding scheme could guarantee in term (after its introduction into the reciprocal recurrent selection scheme), an improvement of the bunch production to the young age of planting material to provide to the village and industrial groves.  It could also diversify the seeds destined for zones affected by the vascular wilt. Its introduction into the breeding scheme could also make it possible to improve the bunch production and diversify sources of vascular wilt tolerances of recurring populations of group B of recurrent reciprocal selection scheme.

CONCLUSION

On the basis of all parameters measured, Lobé × Deli crosses were characterized by performances lower than those of the controls. However, substantial variability for each parameter made it possible to identify crosses as powerful as or more powerful than controls. The study of general combining abilities of Lobé parents made it possible to select one parent which cumulates the maximum of interesting characteristics. The introduction of this parent into the breeding scheme will be able to make it possible to guarantee the improvement of the precocity of bunch production, bunch production at the adult period and the diversification of sources of vascular wilt tolerance of recurring populations of Group B of scheme selection.

REFERENCES

1.   Adon B., Baudouin L., Durand-Gasselin T. and Kouame B. (1998). Use of non-selected material for oil palm breeding: The Angola origin. Plantation, Recherche, Développement 5(3): 201-207.

2.   Adon B. (1995). Assessment of oil palm (Elaeis guineensis Jacq.) introductions: Use in the reciprocal recurrent selection scheme. Thèse de Doctorat 3ième cycle. Génétique et amélioration des productions végétales. Faculté des Sciences et Techniques. Abidjan, Côte d'Ivoire, 115 p.

3.   CaoT.V. (1995). Organization of genetic variability of oil palm (Elaeis guineensis Jacq.): Consequences for the improvement of populations and varietal creation.Thèse de Doctorat, Institut National Agronomique, Paris-Grignon, 200 p.

4.   Dagnelie P. (1998). Theorical and applied statistics (Tome 2): Statistical inference with one and two dimensions. De Boeck et Larcier s.a., Département de Boeck Université, Paris, Bruxelles, 643 p.

5.   Durand-Gasselin T., Cochard B., Amblard P. and Nouy B. (2009). Exploitation of the heterosis in the genetic improvement of oil palm (Elaeis guineensis Jacq.). Le sélectionneur Francais, 60: 91-100.

6.   Jacquemard J.C. (1995). The oil palm. Le technicien de l'agriculture. Paris, Maisonneuve et Larose, 207 p.

7.   Jacquemard J.C. and Baudouin L. (1987). Contribution to the study of oil palm vertical growth: A descriptive model. Oléagineux, 42(10): 343-349.

8.   Jacquemard J.C. (1980). Method of observation of the height of oil palms. Oléagineux, 35(10): 439-442.

9.   Lawes Agricultural Trust (2007). The genstat system for Windows. Release 10.1. Genstat. 5 The waterhouse, Waterhouse street, Hemel Hempstead HP1 1ES, UK.

10. Meunier J., Renard J.L. and Quillec G. (1979). Heredity of resistance to fusarium wilt in the oil palm Elaeis guineensis Jacq. Oléagineux, 34(12): 555-561.

11. Meunier J. and Gascon J.P. (1972). General schema for oil palm improvement at the I.R.H.O. Oléagineux, 27(1): 1-12.

12. Renard J.L., Gascon J.P. and Bachy A. (1972). Research on vascular wilt disease of the oil palm. Oléagineux, 27(12): 581-591.

13. Sékou Diabaté, Kouakou E. Konan, Désiré Allou, Ouolo A. Coulibaly and Hubert de Franqueville (2009). Performance of two techniques of extraction of rooting phenols for the evaluation of marking of the tolerance to the vascular wilt of clones of oil palm. Sciences & Nature, 6(2): 117-123.