A study of the gene B6m, previously described as a modifier, revealed that it is a recessive resistance gene of moderate effect when homozygous. Its value in enhancing resistance to bacterial blight when in combination with other genes, in particular B2, is emphasized. The symbol for the gene is simplified to B6.

Results of genetical tests in the Sudan are consistent with those of Arnold (1963), who was unable in Tanzania to transfer the bacterial blight intensifier gene, B6, as a separate entity to local cotton. Confirmation was obtained of the presence of B2 in Mwanza Local. The high resistance of UKBR selections from Mwanza Local was found to result from the interaction of B2 with a gene, or complex of genes, occupying the same locus as B6, or closely linked to it. This B6-type gene was obtained by exerting steady selection pressure over a number of years and gradually increasing resistance; it therefore reveals that during the last 50 years blight resistance in African Uplands has evolved per se. Whether all of Knight's (1957) B genes have arisen in such a way, or by point mutation, is conjectural.

In a replicated trial the high resistance to bacterial blight of the Allen derivatives Reba W296 and Albar strains was confirmed. Only one major gene, Lagière's B9, was detected in Reba W 296. The very high resistance of this variety is due to B9 fortified by genes of small effect.

The key to resistance in Albar derivatives is the major gene B2: the presence of B3 awaits confirmation. It was found that Albar 49, maintained in Uganda, had deteriorated in resistance to leaf infection in the Sudan. Possible reasons for this change are discussed. All Albar derivatives selected in East Africa for stem and boll resistance were found to be variable for leaf resistance under Sudan conditions.

A high level of resistance to the bean common mosaic virus strains NL3 and NL4 when inoculated separately or together was confirmed in the cultivars Turkish Brown, Valja and 1750–73. Following crosses between the susceptible cv. The Prince and all three resistant cultivars, inoculation of F2 and F3 populations revealed that the cvs Turkish Brown and Valja each differed from The Prince in being homozygous for a major recessive gene conferring resistance to both NL3 and NL4. Homology tests indicated that the same gene was present in both Turkish Brown and Valja. Polygenes in Turkish Brown and Valja also appeared to confer some resistance to NL3, or environmental variation led to continuous distribution patterns in the F2's of resistant parents crossed with The Prince. Although a major recessive gene controlled resistance to NL3 in cv. 1750–73, it differed from that in the two other resistant cultivars. In addition, cv. 1750–73 carried a dominant gene determining reaction to the virus; probably the I gene originally found in Corbett Refugee. An attempt is made to relate these results to Drijfhout's (1978) gene-for-gene model. Using Drijfhout's gene nomenclature all the cultivars tested appear to carry the strain nonspecific gene bc-u. Further genetical tests will be necessary to determine with certainty which, if any, of Drijfhout's strain-specific resistant bc-genes are present in the three resistant cultivars studied here.

Exotic accessions of Phaseolus vulgaris selected for seedling cold tolerance by Austin & Maclean (1972) in a test in controlled environments were grown in a field experiment at Wellesbourne in 1974. Yields of up to 4·;7 t/ha at a plant density of 34 plants/m2 were obtained. In 1975, six selected accessions were grown at 21 plants/m2 at six different sites. Mean site yields ranged from 1·0 to 3·1 t/ha. The six accessions all had better emergence and higher yields than two commercial cultivars used as controls. The genotype × site interactions for yield (and for yield components) were small. The variation in yield between accessions (from 2·0 to 2·5 t/ha) was smaller than the variation in yield between sites. Variation in yield components tended to be additive between sites, and compensatory between accessions. There was a negative correlation between crude protein content and yield across accessions; the correlation between protein content and yield across sites was not significant. Accessions differed significantly in the amount of green leaf still attached to plants at harvest. This variation was not related to maturity as judged by moisture content of the seed. There was a negative correlation across accessions between the weight of leaf per plant at harvest and harvest index (the ratio of seed weight to stem weight).

In a hybridization programme to study the genetics of resistance to bacterial blight of cotton within Gossypium herbaceum, a major gene, thought to be homologous with Knight's B9, was found in Wagad 8. The strain Chinese Indigenous was also found to possess a major gene for blight resistance but this gene was non-homologous with B9In addition genes there were strains of cotton. to the aforementioned major genes of small effect in both

Numerous dwarf bean accessions were screened for resistance to three strains of bean common mosaic virus (BCMV). Many lines were resistant to a strain of BCMV isolated at Wellesbourne (NVRS) and the Dutch NL4 strain, but only a few lines showed resistance or tolerance to the Dutch NL3 strain. The NVRS strain was found to belong to BCMV Group IVa.

Hybridization experiments showed that resistance to the NVRS strain of BCMV in the dry bean cvs Seafarer and Aurora was controlled by a dominant gene and that in Seafarer this was probably the dominant I gene originally found in Corbett Refugee.

Halo blight resistance was transferred in a backcrossing programme from PI 150414 to a Michigan-type bean. The resistant selections obtained gave yields that were up to 10% higher than that of cv. Seafarer and had seed of good quality that was suitable for baking in tomato sauce. These selections were also resistant to the common strains of bean common mosaic virus (BCMV) and to the lambda race of anthracnose.

In addition, selections with white seed of similar size to Seafarer were obtained from crosses between Seafarer and cold-tolerant lines with large, coloured seeds. In trials over 4 years these selections gave 20% higher yields than Seafarer and were less sensitive to environmental changes. Like Seafarer, they were homozygous for the I gene for resistance to BCMV but were susceptible to halo blight and anthracnose. Although not as suitable for canning in tomato sauce as other material in the National Vegetable Research Station programme, they offer useful parental material for further cycles of breeding.

While the leaf resistance of many of Knight's B genes in a Sakel background is now much lower than that found in his original studies, that of B9K has remained unimpaired and this gene still segregates as a major gene. It is proposed to ascribe to the weak gene transferred by Knight from Wagad 8 (G. herbaceum) to Sakel the symbol B11, as gene homology tests between this gene and other B genes indicate that is non-homologous with them. The strong effect of environment on the expression of B11 was confirmed and it was noted that in Upland stocks B11 was much more effective than in Sakel, whereas the opposite was true of B9K.

While the leaf resistance of Bm Sakel is high its stem resistance is of a low level and it suggested that leaf, stem and boll resistance may be under the same control in some genetic systems but not in others.

Accessions of green and dry-podded Phaseolus vulgaris beans were tested for resistance to seven isolates of bean yellow mosaic virus (BYMV) originating from a wide geographical area. Based on their level of resistance, the beans could be divided into four groups, (i) accessions with a high level of resistance to most BYMV isolates; (ii) accessions containing previously known sources of resistance in germplasm derived from the cv. Great Northern, which were mainly susceptible to the BOS 25 isolate, but had high levels of resistance to most other isolates; (iii) and (iv) groups showing increasing levels of susceptibility with the accession Red Mexican (common) the most susceptible bean tested. The seven BYMV isolates induced distinct reactions when inoculated to arange of plant species, indicating that they possessed different pathogenicity genes.

Resistance to bean common mosaic virus in the cultivars Seafarer and Aurora from the U.S.A. is controlled by the dominant gene I originally found in the cv. Corbett Refugee.

Populations derived from a diallel set of crosses involving five inbred Upland varieties were tested for bacterial blight resistance at Wad Medani in the Sudan and at Namulonge in Uganda. Four of the varieties carried a known B gene for resistance; no B gene was present in the fifth, which was used as a susceptible parent. The importance of genotype/environment interaction was emphasized and it was found that the B genes did not maintain the same dominance relationships at the two sites.

The Reba gene, B9L, segregated as a gene or ‘effective factor’ of large effect but Mendelian segregation was not discernible in hybrid populations not involving Reba. Most of the genetic variance was accounted for by additivity and dominance. Although statistical analysis failed to reveal epistasis there was a strong interaction in the B2 × B6 cross.

At Namulonge, where it was possible to test for leaf and boll resistance, the proportion of dominant and recessive alleles was the same for both phases of the disease.

In five experiments with binary mixtures of a wide range of Upland cotton varieties grown at three sites in Uganda in 1969–70 and 1970–1 there was no evidence to indicate that components interacted to give better yields than would be expected from their yield in pure stand. Moreover, in district variety trials over three seasons there was no evidence that mixtures of several lines of closely related material were more stable than the individual component lines. Problems related to a breeding programme based on the development of multiline varieties are discussed. Although the variable ecological conditions in Uganda should provide an excellent opportunity of developing multilines with a high degree of population buffering, it is concluded that pure lines tested in a wide range of environments are as good as, and are less costly to produce, than mixed populations that have been specially bred and tested. Multilines with components that differ little in yielding ability or stability do, however, provide an opportunity of more rapid seed multiplication, and there is also the possibility that multiline seed issues may show progressive increases in yield with each year of growing. Present practice in Uganda of releasing as seed issues multilines with closely related components that have been adequately tested as individual lines should therefore be continued.

Lagière's B9 from Allen 51–296 (Gossypium hirsutum) was found to be non-homologous with Knight's B genes B1–B7, B9 and Bherb. To differentiate between this gene and Knight's B9 from G. herbaceum the symbols have been changed to B9L and B9K respectively. It was impossible to test Knight's B10 from Kufra Oasis G. punctatum and Lagière's B10 from Allen 51–296, but to avoid confusion it is proposed that these genes should be known as B10K and B10L.