ABSTRACT

A complete diallel set of crosses was produced from ten diverse parents of varying degrees of resistance and sucseptibility to white rust and assessed to study the inheritance of the disease caused by Albugo candida. Data were analysed using Griffing and Jinks-Hayman methods of diallel analysis. The results clearly indicated that additive genetic variance predominantly controlled the inheritance of resistance to white rust disease. The parent Poorbijaya was the best general combiner for resistance followed by BJ-38. The cross Glossy mutant x BJ-38 was found to be the best specific combination. All the crosses showing superior specific combining ability effects were resulted from poor x good general combiners for the disease resistance. The average degree of dominance over all loci indicated partial dominance of the disease resistance. Heritability estimates in narrow and broad senses were high. Predominance of additive genetic variance coupled with high heritability suggested that simple selection procedures such as pedigree breeding would be useful for improving the level of resistance.

INTRODUCTION

White rust is an important disease of oilseed Brassicas in India and Canada. The disease caused by Albugo candida affects primarily rapeseed (Brassica campestris) and Indian mustard (Brassica juncea). Though some studies showed that resistance to white rust was controlled by few major genes (Verma and Bhowmik, 1989, Paladhi et al., 1993), the resistance to this disease was also reported to be quantitatively inherited conditioned by minor genes (Edward and Williams,1982). In B. juncea, several sources of resistance have been described (Saharan et al.,1988) but information on the incorporation of resistance to agronomically superior cultivars has not been reported . This could be attributed to the scanty informations on the nature of inheritance for this disease. In order to understand and verify the nature of inheritance of white rust and also to formulate an efficient breeding procedure for improvement in the level of resistance, the present investigation was planned.

MATERIALS AND METHODS

Ten diverse genotypes of Indian mustard namely, Pusa Bold, Varuna, Pusa Bahar, Pusa Barani, RH-30, PR-1108, Poorbijaya, Glossy mutant, BJ-17 and BJ-38 of varying degrees of resistance and susceptibility were crossed in all possible combinations including reciprocals. Forty-five F₁s, 45 reciprocal F₁s and their 10 parents were raised in a randomised complete block design with 3 replications. Each entry was swon in 2 rows of 5m length and was spaced at 45 cm row to row and 15 cm plant to plant. The recommended agronomic practices were followed for raising the crop. Though the crop was raised in hotspot for the disease, all the plants in each row were inoculated to ensure proper development of the disease.

The data for white rust infection on leaves were recorded in percentage on sampled plants of every genotype in each replication after two weeks of flowering. The severity of the disease was recorded by using the scale as per Anonymous(1985). The per cent infection index was calculated by the formula given by Singh (1984) which covered a wide range and was subjected to angular transformation as advocated by Fisher and Yates (1957).

The data were statistically analysed for combining ability using Griffing's method 1, model I (Griffing, 1956). For the estimation of second degree genetic parameters, the Jinks-Hayman approach of diallel cross analysis (Hayman, 1954, Jinks, 1954 ) was followed. Heritabilities in narrow sense and broad sense were estimated from the genetic components as suggested by Mather and Jinks(1971).

RESULTS AND DISCUSSION

The analysis of variance showed significant variation among parents and crosses (Table 2). The mean per cent infection index of the genotypes is presented in Table 1. The disease infection in parents ranging from 2.00 to 60.67 showed that the parent Poorbijaya had the highest degree of resistance followed by BJ-38. On the other hand, the parent PR-1108 exibited highest degree of susceptibility. The hybrids between resistant and susceptible parents recorded a range of intermediate disease reaction suggesting polygenic nature of inheritance of the disease. Dominance of white rust resistance over its susceptibility was indicated by the lower mean values of the infection index in majority of the crosses including reciprocals than their mid-parental values.

Table1. Mean of white rust reaction of ten parents (diagonal), their F₁s (above diagonal)

and reciprocals (below diagonal) in 10x10 diallel cross.

-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Parents Pusa Varuna Pusa Pusa RH - 30 PR-1108 Poorbi- Glossy BJ - 17 BJ - 38

Bold Bahar Barani jaya mutant

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Pusa Bold 55.80 52.33 58.87 47.60 52.40 61.13 8.60 54.73 48.87 13.87

Varuna 62.43 57.23 62.33 51.07 60.97 59.67 6.80 53.10 44.40 32.23

Pusa Bahar 50.63 67.20 57.70 65.27 69.53 61.13 8.37 63.67 55.43 33.33

Pusa Barani 54.73 59.80 58.40 53.83 53.93 53.83 6.53 61.07 55.13 32.33

RH-30 54.50 56.63 51.70 55.97 51.83 53.50 7.03 48.97 45.93 38.87

PR-1108 55.83 49.73 65.10 60.57 54.50 60.67 7.80 58.90 62.67 29.07

Poorbijay 26.17 9.13 6.40 2.57 10.10 12.33 2.00 10.60 6.20 4.83

G. mutant 64.27 61.07 65.30 61.27 56.07 63.10 14.50 56.53 55.67 22.90

BJ-17 55.50 58.63 57.10 65.40 58.73 56.03 11.50 45.70 52.47 19.53

BJ-38 19.73 22.17 34.83 42.83 25.20 31.10 6.07 8.60 14.43 14.90

-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Combining Ability Analysis

Mean squares due to general and specific combining ability were significant for white rust infection index (Table 2). This suggested the importance of both additive and non-additive gene action. However, the ratio of the estimated variances (σ²_gca/σ²_sca) indicated that additive gene action mostly contributed towards the inheritance of white rust resistance in Indian mustard. Similar results were also reported by Singh and Singh (1987) and Yashpal et al. (1991) in this crop.

Table 2. Analysis of variance (MS) for combining ability and estimated variance in 10x10 diallel set for white rust infection index.

------------------------------------------------------------------------------------------------------------

Source d.f. MS/Estimated variance

------------------------------------------------------------------------------------------------------------

Replications 2 121.98

Genotypes 99 1339.99**

Parents 9 1241.57**

Crosses 89 1359.30**

Parents vs. Crosses 1 507.36**

Error 198 48.60

General combining ability(gca) 9 4308.92**

Specific combining ability(sca) 45 85.19**

Reciprocal cross effect(rce) 45 35.69**

Error 198 16.20

σ ² _gca - 214.64

σ ² _sca - 68.99

σ ² _rce - 9.74

σ ²_gca/σ ²_sca - 3.11

----------------------------------------------------------------------------------------------------------

** : significant at 0.01 level.

The parents Poorbijaya and BJ-38 showed significant and negative gca effects for white rust infection index (Table 3), thus, they were good general combiners for imparting the disease resistance. The magnitude and sign of gca effects for these parents were also in agreement with their per se performance (Table 1). This indicated that these genotypes could be considered as desirable parents for hybridization. A perusal of the Table 3 revealed that the crosses Pusa Bold x BJ-38, Varuna x

Table 3. Estimates of general combining ability effects (diagonal) and specific combining ability effects (above diagonal) for resistance to white rust in 10x10 diallel cross.

----------------------------------------------------------------------------------------------------------------------------------------------

Parents Pusa Varuna Pusa Pusa RH - 30 PR-1108 Poorbi- Glossy BJ -17 BJ - 38

Bold Bahar Barani jaya mutant

----------------------------------------------------------------------------------------------------------------------------------------------

Pusa Bold 4.89** 3.28 -2.64 -3.53 0.65 2.72 0.41 5.48* 1.20 -11.18**

Varuna 6.81** 5.45* -1.18 4.08 -2.98 -6.92** 1.14 -1.39 -2.70

Pusa Bahar 10.10** 1.93 2.60 2.15 -10.80** 5.25* 0.07 0.89

Pusa Barani 7.40** -0.36 -1.07 -10.93** 4.64 6.77** 7.09**

RH-30 5.51** -2.38 -5.02 -2.12 0.73 3.43

PR-1108 8.47** -6.48* 3.40 4.79 -1.47

Poorbijaya -34.32* -2.26 -2.92 16.68**

Glossy mutant 6.73** -2.14 -14.07**

BJ-17 3.69** -9.80**

BJ-38 -19.31**

-----------------------------------------------------------------------------------------------------------------------------------------------

C.D 0.05 C.D 0.01

(g_i) 1.67 2.19

(s_ij) 5.06 6.65

------------------------------------------------------------------------------------------------------------------------------------------------

* , ** : Significant at 0.05 and 0.01 level, respectively

Poorbijaya, Pusa Bahar x Poorbijaya, Pusa Barani x Poorbijaya, PR-1108 x Poorbijaya, Glossy mutant x BJ-38 and BJ-17 x BJ-38 exhibited significant and negative sca effects for the white rust infection index. The above crosses showed that the desirable sca effects were due to poor x good gca parents for the disease resistance indicating that desirable transgressive segregates may be released from these crosses in subsequent generations.

Genetic Analysis

The estimates of genetic components of variances and the derived values from them are presented in Table 4. The t² estimate (0.4001) to test the uniformity of the Wr, Vr values was not significant suggesting fulfillment of the assumptions. A further verification of the assumptions by the joint regression analysis showed that b (0.9513 ⁺ 0.0556 ) was significantly different from zero but not from 1.0. This satisfied the assumption regarding the adequacy of additive dominance model and could be inferred that non-allelic interaction was absent. The combining ability analysis (Table 2) clearly exhibited the presence of reciprocal differences. In such a situation, it was suggested to replace all the entries in the diallel table by the mean of the F₁ (Christie and Shattuck, 1992).

Table 4. Estimates of components of variance for resistance to white rust disease in 10x10 diallel cross.

------------------------------------------------------------------------------------------------------

Components Estimated value

------------------------------------------------------------------------------------------------------

E 15.95**⁺ 2.94

D 311.98**⁺ 9.73

F -293.50**⁺22.46

H₁ 142.72**⁺20.72

H₂ 93.91**⁺17.61

h² 45.92**⁺11.79

Derived values

(H₁/D)^1/2 0.68

H₂/4H₁ 0.16

[(4DH₁)^1/2+F]/[(4DH₁)^1/2-F] 0.18

Heritability(ns) 89.24

Heritability(bs) 95.65

----------------------------------------------------------------------------------------------------

** : Significant at 0.01 level.

All the genetic parameters estimated from Jinks-Hayman model were highly significant for white rust infection index (Table 4). However, the magnitude of additive variace (D) was higher than the dominance component (H₁) indicating predominance of additive genetic variance as also evidenced from the combining ability analysis. The average degree of dominance over all loci estmated by (H₁/D) ^1/2 indicated partial dominance of the disease resistance. The H₂ component was smaller than H₁ indicating unequal proportion of positive and negative alleles in the loci controlling white rust infection. The asymmetrical distribution of genes at loci showing dominance in the parents was evident from the value of H₂/4H₁ (0.16) which was less than 0.25. The significant and negative estimate of F suggested an increase in recessive alleles among the parents. Preponderance of recessive genes in parents was also indicated by the estimate of the ratio [(4DH₁)^1/2 + F ]/ [(4DH₁)^1/2 -F] which was less than unity. The heritability estimates in narrow and broad senses were 89% and 96%, respectively. Predominance of additive gene action and high heritability suggested that simple selection procedures would be worthwhile for improving the resistance level against white rust.

Most of the plant materials of interest to the breeder is selected for traits of economic importance. In the present investigation, ten parents were specifically selected for white rust reaction which could not be regarded as a random sample as opined by Eberhart and Gardner (1966). Hence the inferences drawn about the genetic architecture from this investigation apply only to these parents and their crosses. More detailed study is needed using a large random sample from a population to represent Indian mustard as a whole.

ACKNOWLEDGEMENTS

The first author gratefully acknowledges the financial help provided by the Assam Agricultural University during the course of the present investigation. He also thanks Dr. B.K.Baruah, Senior Scientist (Statistics) of Regional Agricultural Research Station, Shillongani, Nagaon for rendering necessary help in computing the data.

REFERENCES

1. Anonymous, 1985. The Proceedings of Annual Oilseeds Workshop of Rapeseed and Mustard, Indian Council of Agricultural Reseaech, New Delhi, India.

2. Cristie, B. R. and Shattuck, V. I., 1992. The diallel cross : Design, Analysis and use for plant breeders. Plant Breeding Review, 9 : 9-36.

3. Eberhart, S. A. and Gardner, C. O. 1966. A general model for genetic effects. Biometrics, 22 : 864-881.

4. Edward, M. D. and Williams, P. H., 1982. Selection for quantitatively inherited resistance to Albugo candida in Brassica camestris, CGS-1. Cruciferae Newslett., No. 7 : 66-67.

5. Fisher, R. A. and Yates, F., 1957. Statistical tables for biological, agricultural and medical research (5th ed.). Hanfer, New York.

6. Griffing, B., 1956. Concept of general and specific combining ability in relation to diallel crossing systems. Aust. J. Biol. Sci., 9 : 463-493.

7. Hayman, B. I., 1954. The theory and analysis of diallel crosses. Genetics, 39 : 789-809.

8. Jinks, J. L., 1954. The analysis of continuous variation in a diallel cross of Niotiana rustica varieties. Genetics, 39 : 767-788.

9. Mather, K. and Jinks, J. L., 1971. Biometrical genetics : The study of continuous variation (2nd ed.), Chapman and Hall, London.

10. Paladhi, M. M., Prasad, R. C. and Dass, B., 1993. Inheritance of field reaction to white rust in Indian mustard. Indian J. Genet., 53 : 327-328.

11. Saharan, G. S., Kaushik, C. D. and Kaushik, J. C., 1988. Sources of resistance and epidemiology to white rust of mustard. Indian Phytopathol., 41 : 96-99.

12. Singh, D. and Singh, H., 1987. Genetic analysis of resistance to white rust in Indian mustard. In :Proc. 7th Int. Rapeseed Cong., Poland, pp. 126.

13. Singh, R. S., 1984. Introduction to Plant Pathology (3rd ed.). Oxford and IBH Publishing Co., India.Verma, V. and Bhowmik, T. P., 1989. Inheritance of resistance to a B. juncea pathotype of Albugo candida in B. napus. Can. J. Plant Pathol., 11 : 443-444.

14. Yashpal, H. S. and Singh, D., 1991. Genetic components of variation for white rust resistance in Indian x exotic crosses of Indian mustard. Crop Res., 280-283.