Capsicum species are usually self-compatible and self-pollinating crop (Raw, 2000; Onus and Pickersgill, 2004). Some percentage of outcrossing is associated with insect pollinators than with wind (Raw, 2000). Genetic advance (GA) under selection refers to the improvement of characters in genotypic value for the new population compared with the base population under one cycle of selection at given selection intensity (Singh, 2001). Therefore, genetic advance (GA) is important to predicting the expected genetic gain from one cycle of selection. Heterosis is one of the most important genetic tools for the manifestation of different cross combinations in crops and determining their potential for commercial exploitation under different environmental conditions (Ashfaq et al., 2013). This breeding method has been used extensively in agriculture to increase yield and to enlarge adaptability of hybrid varieties in a number of crop species (Meyer et al., 2004). The percentage increase or decrease in traits among the hybrids over the mid-parent and better parents’ performance is used to assess the overall performance of these hybrids (Inamullah et al., 2006). Heterobeltiosis expresses the betterment over the better parent; likewise, standard heterosis expresses the heterosis over the standard varieties (Shrestha et al., 2011). The major goal of the research was to improve the red fruited pepper genotypes in relation to aroma and β-carotene; however, this report is aimed at evaluating the hybrid field performance as compared to the parents (Shombo, Tatase and Nsukka yellow pepper) and the estimation of heterosis

The experiment was conducted at the Botanical Garden, University of Nigeria, Nsukka. The seed sources of the parental accessions were from a three year evaluation programme at the Faculty of Agriculture Farms, University of Nigeria, Nsukka. The F₁ hybrids were generated between the highly aromatic genotype, Nsukka Yellow Pepper (Nsky - rw) and some red and fat fruited genotypes (Shombo and Tatase) with a view to improving the red fruited genotypes   in  relation  to  aroma  from  the  Nsky - rw  parent  and improving the carotenoid content of the aromatic yellow pepper genotype (Nsky – rw) from the red fruited genotypes (Shombo and Tatase). Hybridization was conventionally done by hand pollination (Morakinyo and Falusi, 1992) (Table 1).

Seeds from successful crosses were used to generate the F₁ plants. The F₁s were evaluated in the subsequent year and the seeds from F₁  plants were used to raise the F₂ plants. The parents, F₁ and F₂ plants were evaluated in the field using randomized complete design (RCBD). Cultural practices were maintained manually. The weeds were removed by manual weeding with small hoe.  The data on plant height, mainstem length, mainstem internode length and canopy diameter were metrically measured in centimeters (cm). The parameter on the number  of  branches per plant, number of leaves per plant, number of nodes per branch, number of nodes per plant, number of nodes on main stem were collected by numerical count at one hundred to one hundred and five days after planting (DAP). The leaf area  (cm²) measurement was taken using the autoamated leaf area meter in the Physiology Laboratory of the Department of Crop Science, University of Nigeria, Nsukka.

The fruit attributes and yield were also studied and data were collected on fruit diameter (cm), fruit length (cm), fruit stalk length (cm) and Pericarp thickness (mm) - measured with Venier Caliper.The number of fruits per plant was by numerical count. Fresh fruit weight (g) per plant was also determined. The fruits were processed and the seeds were sun dried for five days in the Botanical garden; the average of the numerical count of seeds was recorded as number of seeds per fruit. Seed weight / fruit (g) was also determined. The yield in tonnes per hectare was estimated from fruit yield per plant using plant population per hectare.

Statistical and genetic analysis of data

Data collected on the quantitative characters were subjected to analysis of variance (using GenStat Discovery Edition 4 software) to test for significant differences among the treatments, while significant means were separated using Fisher’s Least Significant Difference (F-LSD) at p < 0.05.

Estimation of  Heterosis/ Hybrid vigor was achieved using the formula:

Differences  between selfed (F­₂)  and outcrossed (F₁) progeny were estimated according to Hedrick and Kalinowski, (2000) using          

D = 1 - w_s / w_o,

D = Inbreeding effect.

w_{s =}  F₂ value

w_o  =  F₁ value

The results of the field evaluation of the parents and hybrids are presented in Tables 2 and 3. There were significant variations across the traits among the parents and hybrids. The values for canopy diameter ranged from 37.9- 66.92 cm. Wider canopies could confer advantage in exposing the leaves to sunlight and more production of the fruit. Equally, high number of leaves together with high leaf area among the F₁ plants of Sho x Nsky – rw is advantageous in attracting more sunlight energy thereby increasing assimilates from photosynthesis. Shombo had significantly the highest value in main stem length across the parent and other hybrid. Its hybrid (Sho x Nsky – rw) equally significantly had the tallest plants with a mean value of 66.11 cm. This value did not, however, vary from Shombo parent and F₂ plants of Sho x Nsky – rw. Tall plants could be disadvantageous in mechanical harvesting but could be preferred in positioning the leaves and tender branches for wider contact with sunlight. Sho x Nsky – rw F₁ plants had significantly the shortest internodes (0.55 cm). These same F₁ plants (Sho x Nsky – rw) also had significantly the highest number of nodes on the main stem, number of nodes/ branch, number of nodes/ plt and number of branches/ plt. This seems to suggest that the inheritance of shorter internodes could result in a significant increase in yield component characters that may have direct or indirect effect on fruit yield. These yield components  have  been  shown  to  contribute  to yield by having high correlations with fruit yield (Nandadevi and Hosamani, 2003).

The higher numbers of fruits recorded in Shombo parent and its F₁ hybrids (Sho x Nsky – rw) were translated to higher cumulative yield of 11.1 and 11.28 t/ha, respectively. The red fruited genotype Shombo could be used as one of the parents in breeding programs due to its high yield. Shukri et al (2015) reported that genetic distance between parents used in breeding program is required in developing new parents with high performance in line with the breeders’ objectives. It had earlier been reported that Shombo has a frutescens origin while Nsukka yellow pepper (Nsky – rw) is purely Capsicum annuum (Morakinyo and Falusi, 1992). This tends to point out the genetic distance between these two parents, thus favouring the observed high performance.   

Canopy diameter, number of leaves/ plt, number of nodes/ plt, leaf area and number of branches/ plt showed positive heterosis over the mid-parents in both hybrids, and all except number of leaves/ plt    in     Tat     x     Nsky-rw,     showed    positive heterobeltiosis over the better parent (Table 4). These traits are the major traits that contribute to yield in peppers. Nandadevi and Hosamani, (2003) reported a high correlation between these traits and pepper yield. It, therefore, implies that the increase in these yield component traits could significantly improve fruit yield among the F₁ plants. The F₁ hybrids of Tat x Nsky-rw had positive heterosis over the mid and better – parents (+42.69 and +34.84, respectively) in plant height. This peculiarity puts these F₁ plants at advantage in hand harvesting since taller plants are preferred. All the major yield component traits showed inbreeding depression in F₂ among Sho x Nsky – rw plants. This is expected as there are allelic segregations at F₂ and pepper being predominantly self - pollinating would have high heterozygosity at F₁. High heterozygosity favours heterosis in many plant species. The F₂ seeds of Tat x Nsky-rw failed in germination; this phenomenon  had  previously  been  observed   in crosses by some researchers (Silva et al., 2017).

The estimation of heterosis over the mid and better parent and inbreeding depression in F₂ for fruit characters is presented in Table 5. The number of fruits/ plt is a major trait contributing to fruit yield in peppers and it has positive heterosis over both the mid and better – parent in Sho x Nsky – rw (+113.65 and +103.42, respectively). It has been reported that number of fruits/ plt could predict yield with a high prediction accuracy (Abu et al., 2015).

Fruit weight together with fruit yield also had a positive heterosis over both the average and better parental values in Sho x Nsky – rw. Superiority over the better parental values (heterobeltiosis) has been reported to be more effective than heterosis, particularly in the breeding of self-pollinating crops (Shukri et al., 2015). It should also be noted that most of the fruit characters had negative heterosis in Tat x Nsky- rw.  Tatase  usually  sets  few  fat  fruits  with  high single weight value. Inbreeding depression was positive among the F₂ plants of Sho x Nsky – rw while in Tat x  Nsky-rw,  there  were  no  survivals. Gonza´lez et al. (2014) associated inbreeding depression with embryo mortality and a reduction in  plant  vigor. This  could explain the germination failure of the F₁ seeds of Tat x Nsky-rw hybrids and the reduction in vigor among the F₂ hybrids of Sho  x  Nsky  – rw. In  conclusion,  the  very  short internode length observed in Sho x Nsky – rw hybrid was advantageous as it had the highest values in major morphological traits with high weightage to yield and in fruit yield. Equally, while Sho x Nsky – rw had the highest yield, Tat x Sho had the highest fruit wall thickness. The combinations of these traits are important to both the farmers and consumers. The genetic distance between the Shombo and Nsukka yellow pepper was advantageous in raising F₁ hybrids with high performance above the better parent.

The authors have not declared any conflict of interests.

The authors appreciate Tertiary Education Trust Fund, Institutional Based Research (TET-Fund - IBR) University of Nigeria, Nsukka for sponsorship of the research.