The focus of plant breeding programs is on multi-environment experiments, making their prediction accuracy, with respect to observed values, a crucial determinant for varietal selection and identification of suitable environments. The current findings indicate variation in a set of okra accessions for yield and resistance to aphids in different environments, thus confirming the need to assess genotype by location interactions (Pacheco et al. 2015). However, the variability in the response in contrasting environments, especially with respect to the local varieties, provides a highly relevant set of database on the effects of genotype and location on 10 okra accessions. The accessions evaluated in this study are a combination of aphid-resistant accessions identified following a series of previous screening in one environment (Abang et al. 2014b, 2020), all with significant contributions to food, nutritional security, and sustainable livelihoods in diverse agro-ecological landscape of Central Africa. The Central African agro-ecology is represented in the four agro-ecologies where the germplasms representing the genotypes tested, were evaluated (Robiglio and Sinclair 2011).
The interactions between environment and genotypes did not significantly affect aphid abundance. WAABS for the BLUP of genotypes also showed that the accessions were moving closer to the origin (center of the biplot). The closer things are to the origin, the less distinct they probably are. On the contrary, the different environments were moving away from the origin suggesting that the different environments were more distinct confirming their diverse agro-ecologies. Cameroon has five agro-ecological zones and earns the title ‘Africa in miniature’ because of its highly diverse ecological landscape which exhibits all the major climates and vegetation of the continent (ZEF et al. 2017; IUCN 2018; Nfornkah et al. 2021). Three environments (Buea, Evodoula and Foumbot) have varying rainfall and seasonal patterns all with humid equatorial climates, while Maroua with Soudano-Sahelian climate is dryer and semi-arid (ZEF et al. 2017). This diverse ecological landscape and the genotypes had no effect on aphid abundance as indicated by the best linear unbiased predictors (BLUP) where both the location and genotypic variances were insignificant. Based on the AUIPC, one accession (VI036213) was resistant, six (VI039614, VI041210, VI051114, VI060688, VI060794, VI060818) were moderately resistant, three (Gombo caféier, VI033824, VI057245) moderately susceptible and two (Local and VI033805) susceptible. The non-significant effect of environment or genotype on aphid abundance (P > 0.05), despite different resistance status, suggests that the type of resistance demonstrated by these germplasms is termed tolerance (Painter 1951; Kogan and Ortman 1978; Horber 1980; Smith 1989). Hence, we found that genotypes with higher aphid abundance were more productive. Tolerance is obtained as a result of plants ability to continuously grow and produce despite insect damage. The plants is capable of keeping its fitness by growing and producing irrespective of damage from herbivory (Boege 2005), compensation by replacement of plant biomass lost (Strauss and Agrawal 1999), through accelerated rate of photosynthesis and stored resource mobilization (Ramula et al. 2019).
The current work opens the avenue for promotion of germplasms having appreciated performance in yield and its stability, and aphid resistance or tolerance targeting locations with similar agro-ecologies as those in Cameroon where the genotypes were tested. The tested varieties or accessions were previously selected based on their resistance to aphids, a trait in which variability comes from genetic differences, with very little contribution from environmental factors. We noticed however only one resistant accession (VI036213) with the lowest aphid abundance, although not very productive. The local variety was the most tolerant as it accumulated the highest number of aphids per leaf, but was the most productive followed by genotype VI060794, one of the new accessions which was also moderately resistant during the current study, making it a candidate for dissemination. These two genotypes (Local and VI060794) have very high breeding values and also with significant effect of the genetic variance. The local genotype recorded the highest leaf area and plant height which are important indicators of plant vigor which enhance tolerance. Vigorous plant can withstand high pest pressure and continue to grow and produce significant yield. Insect response to herbivory have been reported to be influenced by external factors such as various nutrient types (Chapin and McNaughton 1989), light (Anten and Ackerly 2001), and also by internal characters (McConnaughay and Coleman 1999). Among the plant attributes measured were leaf area, plant height, pod width and length. While plant height was significantly influenced by the environment and pod length influenced by genotype, pod width was influenced by both variances. Nevertheless, the heritability was higher for pod width suggesting it to be the plant attribute that led to the observed significant effect of genotype on yield. The breeding value for this parameter (pod width), was highest for VI060794 and must have contributed to bring its yield closest to that of the tolerant local genotype. While the local genotype had demonstrated tolerance to aphids, VI060794 on the other hand was moderately resistant due to relatively lower aphid abundance. Lower aphid abundance and higher yield reveal other mechanisms of resistance, which could be antibiosis or antixenosis. In case of antibiosis, the VI060794 accession and the other genotypes that were resistant or moderately resistant may have had a negative effects on the biological parameters of an insect using them as hosts (Smith 1989) leading to negative effects on that insect life traits (Painter 1951). Such effects can be lethal, irregular growth rate and behavior, malformation, decreased fecundity or reduced fertility (Metcalf and Luckmann 1994). On the other hand lower aphid abundance can be due to non-preference or antixenosis, where the genotypes were not preferred hosts for feeding and oviposition (Kogan and Ortman 1978; Smith 1989). However, mechanisms or categories of resistance of these genotypes have been reported to be either by tolerance or antibiosis as none of them have shown antixenotic properties (Abang et al. 2016), and their antibiotic properties have been linked to plant metabolites (Abang et al. 2018).
The only yield parameter on which significant interactions between environment and genotypes were found was pod size (pod width and pod length). However, the heritability of pod width was 17.7 time higher than for pod length suggesting that any effect of environment on the genotype will affect pod width and consequently pod size and yield. Plant parameters such as pod size are important in the selection of okra species for breeding programs to improve its yield (Falusi 2012). Three genotypes with the higher biggest pod width (Gombo caféier, local and VI060794) were found in the same quadrant with one environment (Evodoula) considered to have good discrimination ability. This discrimination ability seems to be greater towards local genotypes as demonstrated by the WAABS. The BLUP showed that VI060794 and the Local genotype had highest yield but VI060794 was more stable, and no location showed high discrimination ability for this genotype in terms of yield, moreover, the local genotype was unstable, but obtained good discrimination ability in Evodoula only. Out of the four locations, three have a humid equatorial climate, while one location (Maroua) has Soudano-Sahelian climate. The peculiarity of Evodoula with discrimination ability could be attributed to the fact that it is the only environment with a bimodal rainfall with two cropping seasons per year while the other locations have monomodal rainfall with one cropping season. The accessions evaluated in the current study were A. esculentus okra species from Asian; most of the accessions originated from the Philippines, except VI060794 and VI060818 from West Africa. VI060794 and VI060818 are A. caillei and A. esculentus respectively. It is only in West and Central Africa where 10% of world production (Iwu 2014) is obtained, that common A, esculentus and the West African A. caillei are both grown. Both species can now be found at equal percentage in the market (Grubben and Denton 2004) as A. caillei is now replacing A, esculentus in the tropical-humid regions because of its adapts better and tolerates more biotic stresses (Siemonsma 1982). VI060794 is A. caillei and could produce the highest yield in all locations and seasons than other accessions, but not significantly different from the farmers’ varieties. Abang et al. (2020) recently revealed that local okra germplasms in Cameroon are dominated by A. caillei and were more productive than the A. esculentus. This confirms the fact that A. caillei is more adapted and more tolerant (Siemonsma 1982). Furthermore, the values of all the plant parameters were above the grand mean in all A. caillei suggesting that Gombo caféier, the local genotype (tolerant to aphid), VI060794 (high yield and resistant to aphids) which are predominantly A. caillei are more vigorous. These varieties were among the best in terms of plant height, leaf area, pod width and yield. Okra plants vegetative parameters positively correlated yield traits (Akinyele and Osekita 2006). Furthermore, plant height pod, days to flowering, weight and number per plant have been identified as major determining factors of pod yield (Akinyele and Osekita 2006; Saifullah and Rabbani 2009; Falusi 2012; Reddy et al. 2012).
The high discriminatory ability of Gombo caféier may also have been influenced by longer crop cycle as duration to commercial maturity was longest, while that of other genotypes of the same species were among the shortest. The duration of crop cycle therefore shows significant variation based on variety (Bosch 2004). WAASB for duration to commercial maturity identified only one location, Foumbot, in the western highlands (zone III) suitable for this commercial variety. This genotype (Gombo Caféier) could be better adapted to the western highlands, where it originated from the seed company “Grenier du monde rural” (GMR). Furthermore, we noticed that among the traits presented in the current study, the four environments were placed in separate quadrants only for duration to commercial maturity (DCM), and Gombo caféier was almost the only genotype found in the same quadrant with Foumbot confirming its adaptation to that environment. The highly tolerant local genotype and the resistant highly reproductive VI060794 were found in the same quadrant with the semi-arid environment of Maroua, as indicated by the WAASB for DCM. Hence, crop cycle shows overwhelming differences following cultivar, locality and season (Grubben and Denton 2004). Okra is grown annually and with a cycle from one to five months, and varies according to species (Falusi, 2012). In WCA, the varieties cultivated in home gardens and resource-poor farmers’ farms often include the two species of A. esculentus and A. caillei, with the former dominating in dry climates, and the latter in humid climates. On the other hand, market gardeners do not practice intercropping and so grow okra in monocropping, and prefer the uniform and improved A. esculentus with earliness (Iwu 2014). In rural production, farmers cultivate landraces in home gardens in association with other subsistence crops. A. esculentus is generally a short-day species, but its widely geographic range of cultivation indicates that it differ in day light sensitivity. Flower initiation is affected hardly by the length of the day in the tropics. In A. callei, most types show response shorter length of th day. When planted early in the rainy season starting in March, flowering is not possible by November when the rains end, but they remain vegetative with robust leaves and plant architecture, bulky stems, and survive the dry season without irrigation, then reproduce in a period of erratic rainfall. If cultivars with the same precocity are compared, West African (WA) ones usually exhibit extensive period of production than the common species, thus making WA okra suitable for home gardening. Anthesis in common okra may occur early but with a shorter productive period suitable for market gardening (Iwu 2014).