Abstract

Groundnut production is largely constrained by biotic stresses with groundnut rosette virus disease seriously contributing to losses in yield in Nigeria and sub Saharan Africa. This study was conducted to understand performances and correlations between aphid resistance, rosette resistance and other quantitative characters. Two aphid resistance, one rosette resistance, one aphid susceptible and one rosette susceptible lines were used as parents to develop F1s, F2s, BC1P1 and BC1P2. The seventeen generations obtained were evaluated along with three checks in three replications using randomized complete block design. The F₁ generations had higher mean performance compared to the parents suggesting heterosis. The segregating generations (F2s, BC1P1 and BC1P2) had mean values higher or lower than the parents due to transgressive genes. The genotypic correlation coefficients in contrasting magnitudes and directions exceeded those of the corresponding phenotypic correlation coefficients for most of the character pairs indicating that the correlations were more genetic than environmental in the three sets of crosses studied.

1. Introduction

Groundnut is an oilseed crop in Nigeria appreciated for its abundance of dietary protein rich oil, carbohydrate and other essential mineral nutrients. It is mainly used for consumption, raw material in the confectionary industry and protein source in animal feed industry ¹. These multiple uses of groundnut make it an excellent cash crop for domestic and international trade ². As a legume, it has a soil fertilization quality through the activities of nitrogen fixing bacteria resident in the root nodules ³.

Despite the economic importance of groundnut, its production is constrained by several abiotic and biotic stresses among which is the groundnut rosette virus disease. Groundnut rosette virus disease (GRVD) has been recognized in all groundnut growing countries on the African continent, including its offshore islands such as Madagascar, but not anywhere outside Africa ⁴. GRVD is responsible for an annual groundnut loss of worth US$ 150 million ³. Nigeria alone lost about 0.7 million hectares of land to groundnut rosette virus epidemic which amounted to US$250million ⁵. The disease results from the synergistic interaction of three viral components; groundnut rosette virus (GRV), its satellite RNA (Sat-RNA), and groundnut rosette assistor virus (GRAV) ⁶. The disease is spread by groundnut aphid.

The most effective, economic and sustainable method of limiting both the spread of the aphid and rosette viruses is to develop high yielding varieties that are resistant to the vector and the disease. ^{7, 8}. Although efforts have been made at the Institute for Agricultural Research, (I.A.R) Samaru, Nigeria to develop resistant varieties, there are often cases of break in resistance ⁹.

Correlation is a biometric tool that brings out the extent of association between two or more characters with the end to provide information that forms the bases for selection in a breeding program ^{9, 10}, Characters that show similarities in magnitude and direction as indicated by the correlation coefficients are simultaneously considered for selection for further resistance studies ^{11, 12, 13}. Therefore, the present study was carried out to study genotypic correlations, phenotypic correlations between aphid resistance, rosette resistance and other quantitative characters in groundnut based on these objectives;

1. To study the mean performances of aphid resistance, rosette resistance and other quantitative characters

2. To explain associations among aphid resistance, rosette resistance and other quantitative characters.

2. Materials and Methods

The field evaluations were conducted at the Institute of Agricultural Research (I.A.R), Samaru with an altitude of 686m above sea level, lat 1111’N, long 0738’E during the 2014 growing season in the Northern guinea savanna zone of Nigeria, with a mean annual rainfall of 1050 mm distributed within five months. The soil type is loamy.

The plant materials for this research consisted of four entries obtained from West and Central Africa groundnut improvement program, Mali (ICGX-SM00020/5/9, ICGVIS07899, ICGX-SM0017/5/P10/P1 and ICGX-SM0020/5/P4/P1), one local variety (MANIPENTA) and three checks (SAMNUT 22, SAMNUT 23, and SAMNUT 24). Lines ICGX-SM00020/5/9 and ICGX-SM0020/5/P4/P1 are resistant to groundnut aphid, line ICGVIS07899 is resistant to rosette while lines ICGX-SM0017/5/P10/P1and MANIPENTA are susceptible to both groundnut aphids and rosette. The genetic populations were developed through crossing of the resistant P₁ and susceptible P₂ parents to obtain the F₁s. The F₁s were advanced to obtain the F₂s. The F₁s were also crossed to the recurrent parents to obtain BC₁P₁ and BC₁P_2. The resulting generations (P₁, P₂, F₁, F₂, BC₁P₁ and BC₁P₂) were evaluated with the three checks. Table 1 gives information about the genotypes.

The five parents, three F₁s, three F₂s, six backcrosses along with the three checks making a total of twenty entries were evaluated at the Institute for Agricultural Research farm Samaru (lat 1111’N, long 0738’E). The experiment was laid out in a Randomized Complete Block Design (RCBD) in three replications with one row plot each of 5m in length and an inter row and intra row spacing of 0.75m by 0.25m. The rows containing the genotypes under test were flanked by two infector rows of MANIPENTA a highly susceptible cultivar. This technique, called infector-row technique was described by ¹⁴. An alley of 1m separated one block from the other. MANIPENTA was sown two weeks earlier to allow buildup of infestation. Aphis cracccivora were collected from infested groundnut Arachis hypogaea and cowpea Vigna unguiculata plants in fields within Zaria and environs. These colonies were maintained on susceptible local groundnut genotype MANIPENTA. Three wingless (apterae) aphids were introduced on the tender leaves of 14 day-old seedlings of each of the twenty genotypes under trial. Each genotype was observed for the presence or absence of the aphids. Plants with no aphid were re-infested 7days after the first infestation. It is rare to find plants without aphids in choice test because the aphids were free to roam to find suitable plants within the field. No measure was taken to confine the aphids within the field. Data were collected from 40 plants for non-segregating populations (P₁, P₂, F₁) while 100 plants and 80 plants were considered for F₂ and BC₁ P₁ and BC₂ P₂ segregating populations respectively. Plant height, Days to 50% flowering, Days to maturity, Number of mature pods per plant, Number of seeds per pod were counted while Shelling percentage, pod yield per plant, and Hundred seed weight were measured using Mottler PM16-N weighing balance (ISC070501 Model).

The disease severity was recorded as the percentage of the amount of plant tissue that was diseased with green or chlorotic rosette. Reaction to rosette was scored based on the scale of 1 (No apparent rosette symptoms), 3 (10% - 20% rosette symptoms), 5 (20% -60% rosette symptoms) and 7 (60% -80% rosette symptoms). Disease severity index was then obtained using the formula described by ¹⁵

This is the ratio of number of plants that showed green or chlorotic conditions to the total number of plants in the row expressed in percentage. The percent disease incidence was scored based on the scale recommended by ICRISAT ³

Aphid infestation index for each line was calculated by the formula and scale developed by ^{16, 17}.

Data collected from different genetic populations were subjected to mean performance, genotypic correlations and phenotypic correlations. Mean separation was undertaken using Duncan’s Multiple Range Test. Means with the same letter are not significantly different at 5% probability level. The Least significant difference was noted.

Genotypic and phenotypic correlations were used to investigate the association among characters studied. Correlation coefficients were calculated from component of variance and covariance according to ¹⁸ the genotypic components were computed by equating the genotypic variances and covariance to the expected mean square and set products, and hence genotypic correlations were computed by this formula:

(1)

Where

(2)

(3)

(4)

(5)

Phenotypic correlations were computed using the following formula:

(6)

Where

(7)

(8)

(9)

(10)

3. Result

The mean performance of the parents, F₁s and the segregating generations (F₂, BC₁P₁, BC₁P₂) for days to fifty percent flowering, aphid infestation index, rosette disease incidence, rosette severity index, plant height, days to maturity, number of mature pods per plant, pod yield per plant, number of seeds per plant, shelling percentage and hundred seed weight are presented in Table 2.

There were significant differences between the parents for all the traits measured in the three sets of crosses. Among the sets, ICGVIS07899 was the best performer in terms of days to fifty percent flowering (36.67days), aphid infestation index (2.16%), rosette disease incidence (2.78%) and rosette severity (0.93%) followed by ICGX-SM0020/5/P4/P1. The tallest (43.78cm) and the shortest (31.56cm) plants were ICGX-SM0020/5/P4/P1 and Manipenta. The earliest maturing (98.67days) was Manipenta followed by ICGX-SM0017/5/P10/P1. The latest was ICGX-SM0020/5/9. ICGVIS07899 exhibited superiority in terms of numbers of matured pods per plant (35pods) and pod yield per plant (27.78pods) with ICGX-SM0017/5/P10/P1 ranking second. With regards to number of seeds per plant (18.56seeds) and shelling percentage (74.69%), Manipeta was also the highest. The heaviest in terms of hundred seed weight (59.07g) was recorded in ICGX-SM00020/5/9 and ICGX-SM0017/5/P10.

Among the F₁ hybrids, ICGX-SM0020/5/9 x ICGX-SM0017/P10/P1 ranked best in the characters studied except in plant height (37.33cm) where it ranked last. ICGX-SM0017/5/P10/P1xICGX-SM0020/5/P4/P1 ranked second in most of the characters performing best in plant height (43.33cm) and last in days to maturity (104.33days) and shelling percentage (63.33%). ICGVIS07899 x Manipenta was the least in performance but ranked second in plant height (41.44cm), days to maturity (104.33days) and shelling percentage (65.25%).

The mean performances of the segregating generations for the characters studied showed that among the F₂s, ICGVIS07899 x Manpenta was outstanding in performance for aphid infestation index (1.92%), rosette disease incidence (30.16%), rosette disease severity (45.50%), plant height (38.11cm), days to maturity (103days) and number of seeds per plant (16.67seeds) with ICGX-SM0017/5/P10/P1xICGX-SM0020/5/P4/P1 ranking second. However, ICGX-SM0017/5/P10/P1xICGX-SM0020/5/P4/P1 ranked first in performance for days to fifty percent flowering (42.33days), number of matured pods per plant (18,33pods), pod yield per plant (21.93g) and hundred seed weight (70.57g). ICGX-SM0020/5/9 x ICGX-SM0017/P10/P1 performed least among the sets of crosses although ranking second for number of matured pods per plant (17.89pods), shelling percentage (64.50%) and hundred seed weight (70.57g).

Among the backcross generations, BC₁P₂ of ICGVIS07899 x Manpenta performed best for days to fifty percent flowering (34.33days) and days to maturity (94days). BC₁P₁ of ICGVIS07899 x Manpenta performed best for number of matured pods per plant (33.78 pods), pod yield per plant (26.14g) and number of seeds per plant (17.33seeds) while BC₁P₂ of ICGVIS07899 x Manpenta and BC₁P₁ of ICGX-SM00020/5/9xICGX-SM0017/5/P10/P1 ranked second for number of matured pods and pod yield per plant(g) respectively. BC₁P₂ of ICGX-SM0017/5/P10/P1xICGX-SM0020/5/P4/P1 was outstanding for aphid infestation index. BC₁P₂ of ICGX-SM00020/5/9xICGX-SM0017/5/P10/P1 ranked first for rosette disease incidence (26.48%) and BC₁P₁ of ICGVIS07899 x Manpenta for shelling percentage (73.77%). BC₁P₁ of ICGX-SM0017/5/P10/P1xICGX-SM0020/5/P4/P1 took the lead for plant height (38.78cm) and hundred seed weight (47.53g). For rosette severity index, BC₁P₂ of ICGX-SM00020/5/9xICGX-SM0017/5/P10/P1 is the best performer with a severity of 39.69%.

The genotypic (upper matrix) and phenotypic (lower matrix) correlation coefficients at 5% and 1% levels of significance among the characters in the three sets of groundnut crosses are given in Table 3.

The coefficients indicated that genotypic correlations were larger than the phenotypic correlations for most of the character pairs. There were also some similarities in magnitude and directions (signs) in genotypic and phenotypic correlation coefficients for the three sets of groundnut crosses for the character pairs studied.

Based on these similarities in magnitude and direction of the estimates of genotypic correlations, aphid infestation index (%) was highly positively correlated with days to maturity (0.88**, 1.00**, 1.00**) but highly negatively correlated to number of matured pods per plant (-1.00**, --1.00**, -1.00**) and pod yield per plant (-1.00**, -1.00**. -0.68**) for the three sets of groundnut crosses studied. Similarly, Aphid infestation index had low positive correlations (1.00**, 0.02, 0.24) with rosette disease incidence.

Rosette disease incidence (%) had consistent positive correlations with number of matured pods per plant (0.33, 0.20, 1.00**), pod yield per plant (1.00**, 0.56*, 1.00**) and hundred seed weight (0.04, 0.65**, 0.06). However, there was negative correlations between rosette disease incidence with number of seeds per plant (-1.00**, -1.00** and -0.32) for the three sets of groundnut crosses tested.

Furthermore, there was positive correlations between rosette severity index with number of matured pods per plant (0.08, 0.11, 1.00**) and pod yield per plant (0.26, 0.45, 0.01) and negative correlations with number of seed per plant (-0.10, -1.00**, -0.03).

Again, days to fifty percent flowering had highly negative correlations with number of matured pods per plant (-0.90**, -0.85**, -0.89**), pod yield per plant (-1.00**, -0.54*, -0.90**), number of seeds per plant (-1.00**, -1.00**, -0.28) and shelling percentage (-0.82**, -1.00**, -0.84**). While positive correlations were recorded between days to fifty percent flowering and days to maturity (0.82**, 0.11, 1.00**) for the three sets of groundnut crosses.

Number of matured pods per plant showed positive correlations with pod yield per plant (0.77**, 0.95**, 0.96**) and number of seeds per plant (0.65**, 0.52**, 0.51*). The same trends were observed in the correlations (1.00**, 0.19, 0.69**) between pod yield per plant and number of seeds per plant, and also in the correlations (0.11, 0.004, 0.30) between number of seeds with shelling percentage respectively.

Based on similarities in magnitude and direction of phenotypic coefficient of correlations, there were low positive correlations (0.14, 0.45, 0.16) between Aphid infestation index with Rosette severity index but negative correlations (-0.36, -0.32, -0.64**) with hundred seed weight across the table for the three set of groundnut crosses studied. In the same vein, rosette disease incidence was positively correlated with rosette severity index (0.37, 0.93**, 0.81**) and shelling percentage (0.09, 0.21, 0,05) but negatively correlated with hundred seed weight (-0.36, -0.32, -0.64**) across the table for the three sets of groundnut crosses. Rosette severity index had consistent positive correlations with number of seeds per plant (0.33, 0.52**, 0.06) and shelling percentage (0.18, 0.31, 0.14).

Days to fifty percent flowering showed negative correlations (-0.12, -0.12, -0.51*) with number of seeds per plant. The correlation coefficient estimates (0.22, 0.06, 0.05) observed between plant height and pod yield per plant were positive but low. A consistent negative correlations were recorded for days to maturity with number of matured pods per plant (-0.17, -0.32, -0.48*), pod yield per plant (-0.11, -0.11, -0.40) and number of seeds (-0.20, -0.04, -0.58*) across the table for the three set of groundnut crosses studied. The correlations between number of matured pods per plant and pod yield per plant was high (0.72**, 0.79**, 0.89**) and in the positive direction. There were positive correlations (0.79**, 0.22, 0.68**) between number of seeds per plant and shelling percentage.

4. Discussion

Significant differences in performance among the parents, F₁s, F₂s and backcrosses studied suggest that there was sufficient variability across the generations for the characters studied. The F₁ generations had mean values higher than those of their parental lines suggesting the presence of heterosis. The mean values of the segregating generations were higher or lower than those of the parental lines suggesting the presence of transgressive segregation. Heterosis may result from over dominance of gene pair, depression of the dominant increasing alleles in the parental lines or both while transgressive segregation may result from recombination of additive alleles and epistasis.

Succinctly, the genotypic correlation coefficients exceeded those of the corresponding phenotypic correlation coefficients for most of the character combinations in the three sets of crosses studied. These phenomena can be seen in correlations between aphid infestation index with number of matured pods per plant and pod yield per plant. Correlations between rosette disease incidence and plant height, pod yield per plant and number of seeds per plant, and between rosette disease severity and plant height indicating strong genetic associations between characters studied whose phenotypic expressions where influenced by the environment. Similar results were obtained by ^{19, 20, 21, 22, 23}.

In the three sets of groundnut crosses, aphid infestation index showed positive genotypic and phenotypic correlations with rosette disease incidence but positive phenotypic correlations with days to maturity. Rosette disease incidence had consistent positive genotypic correlations with matured pods per plant, pod yield per plant and hundred seed weight but positive phenotypic correlations with rosette severity index and shelling percentage. Any pair of two mutually and directly correlated characters that are heritable can be considered together during selection.

Similarly, days to 50 percent flowering had highly negative genotypic correlations with number of matured pods per plant, pod yield per plant, number of seeds per plant and shelling percentage. It also had negative phenotypic correlations with number of seeds per plant illustrating that increase in the number of days to fifty percent flowering has a converse effect on the expression of these characters. These findings were in agreement with those of ^{9, 24, 25}. Phenological characters such as days to first flowering and days to fifty percent flowering have converse effect on the expression of yield and yield attributing characters.

Days to maturity showed consistent negative phenotypic associations with number of matured pods per plant, pod yield per plant and number of seeds per plant in the three sets of groundnut crosses. These result were in agreement with those reported by ²⁴. Positive estimates of phenotypic correlations where observed between pod yield per plant and number of matured pods per plant. These findings were at par with those reported by ¹⁰.

Acknowledgements

Authors acknowledge the support of Mr. Paul Bidemi Victor and the technical staff of groundnut laboratory of the department of Plant Science, Ahmadu Bello University, Zaria,

References