A number of aspects of the symbiosis between Centrosema pubescens Benth. and Rhizobium strains were studied.
In tests with 31 Rhizobium isolates from the cowpea cross-inoculation group including five from C. pubescens and C. plumieri (Turp.) Benth., all isolates nodulated cowpea (Vigna sinensis (L.) Endl. Ex Hasak.) and phasey bean (Phasealus lathyroides L.) but C. pubescens and C. plumieri were very specific in their nodulation requirements. A “tropical legume miscellany”, within which completely reciprocal cross-inoculation may not occur, is proposed to embrace the cowpea, soybean, lupin, Caragana, Sesbania, Dalea, and possibly the bean and some other groups. Reciprocal nodulation occurred with two Rhizobium strains from C. pubescens and three from C. plumieri.
Twenty-one Rhizobium isolates from C. pubescens from different parts of Queensland were tested on this host in sand culture in the glasshouse. A range of effectiveness was recorded although all strains produced a marked plant response. A highly significant correlation was obtained between weight of nodules per plant and aboveground plant weight. Strains also differed in mean nodule weight and number of nodules formed. Nodule production by some strains, however, varied markedly with small changed in the environment, this affecting their relative effectiveness. Two isolates from C. pubescens and three from C. plumieri were effective on both hosts. No interference with nodulation by a highly infective Rhizobium strain inoculated onto seed occurred from a high concentration of a non-infective strain added to the sand. The agricultural significance of high and low specificity in nodulation is discussed.
Commercial seed of C. pubescens was shown to contain stable sparsely nodulating and profusely nodulating lines. With most Rhizobium strains plants of sparsely nodulating lines showed poor nodulation and growth in nitrogen deficient conditions but certain Rhizobium strains were found to nodulate these lines satisfactorily. Good plant growth ensuing. Profusely nodulating lines produced a greater number of nodules than average plants in commercial seed lots but total nodule weight per plant and plant weight were unaltered by the profuse nodulation.
Field study of the effect of phosphate on nodulation and growth of C. pubescens indicated that the major effect of phosphate appeared to be exerted through the plant. Beyond extremely low phosphate levels increasing nodulation with increasing phosphate application was almost entirely due to the increases in root length effected by the treatment. There was, however, and indication of increase in nodulation intensity in some instances. An increase in mean nodule weight occurred also with increasing phosphate application.
The structure of typical nodules of C. pubescens conformed with that published by other workers for similar types of nodule. Pseudo-nodules formed following inoculation with some Rhizobium strains from other hosts had a modified lateral root structure.
Examination of rhizobial proliferation around C. pubescens seed in sterile sand showed a relatively slow growth of the bacteria around the seed. Emerging roots supported a high rhizobial population. Diffusates from the testa of germinating C. pubescens seed and of subterranean clover seed contained a water soluble, heat stable toxin toward a wide range of Rhizobium strains, gram positive and other gram negative bacteria but lucerne seed showed little such activity. Depression or stimulation of multiplication of particular bacteria around planted seed was related to their sensitivity to the diffusates as shown by plate assay. It is suggested that bacteriotoxic seed diffusates may play a significant part in the selective growth of organisms around germinating seeds.
Following field observation of poor nodulation of legumes sown on hot days the effect of high soil temperature on rhizobial survival was studied. Eighty-seven Rhizobium strains showed maximum temperatures for growth on agar of 31-38.4°C, 32-32.7°C, 36.5-42.5°C and 30-42.0°C for clover, pea, medic, and tropical legume strains of Rhizobium respectively. Strain of bacterium, initial concentration of inoculum and the period of subjection to high temperatures were shown to be significant factors in the survival of legume bacteria in sterile Wallum-heath sand and on planted C. pubescens seed. Data on temperatures in bare and soil shaded in various ways were obtained.
In studies of field growth and nodulation of C. pubescens in northern and southern Queensland a synchronization of nodulation and plant growth was observed. Nodulation continued beyond flower production and throughout growth. After stolon formation there was a progressive elimination of older nodules from the main root system but there was no indication of a progressive elimination of smaller nodules. Nodulation and nodule inactivation was considered to be mainly related to vegetative plant growth or a criterion such and Net Assimilation Rate rather than plant factors such as flower production. Flower production, however, did appear to temporarily check nodule formation and growth. Only a small proportion of the plant nitrogen was contained in the nodules and only four to twenty per cent. in the below ground parts. Cutting to simulate heavy grazing resulted in a loss of two-thirds of the plant roots by weight within 30 days and inactivation of sloughing of a major proportion of the nodules.