Biological nitrogen fixation in drought resistant tepary bean (Phaseolus acutifolius A. Grary var. Latifolius) in the semi-arid makueni district of Kenya.
Laboratory trials were carried out to assess infectivity of tepary bean (TB) by the various selected rhizobium strains. Strains 446 and 3254 were obtained from the Microbiological Resources Centre (MIRCEN), University of Nairobi; while 578 and 579 were from Iserlohn, Germany. A local isolate, KTBI was obtained from Kiboko soils. All treatments except the nitrogen treatment GUNF and the control nodulated under greenhouse conditions, though there were variations in the proportions of modulated plants. The highest nodulation proportion (100%) was observed with strain 3254 followed by 446 (75%), then 578 and 579 (both at 25%). The lowest nodulation proportion was observed with strain KTBI (20%). Field trials were conducted at KARI/ICRISAT station; Kiboko to compare the dry matter yields of tepary bean inoculated with various rhizobial strains. Plants were sampled at three stages during growth and separated into several fractions viz. root, shoot and nodules. There were no significant differences in the dry weight and mean pod dry weight from treatment 32543 were significantly higher (p<0.05) than for all the other treatments. Sample diffusive resistance readings were high in TB (about 0.17) compared with indigenous crops such a maize (0.150). This was indicative of good water economy by the plant as exemplified by the corresponding low relative transpiration rates. DNA samples from the various treatments did not give good amplification due to the scarcity of specific primers. However, one primer produced good amplification revealing a close phylogenetic relationship between rhizobial strains 578 and 579 due to the similarities of their bands while strains 3254, 446 and KTB 1 had quite dissimilar bands. N analysis was carried out at the division of environmental and applied biology, University of Dundee; both the % N (2.55) and mean N per plot (19.26g) for treatment 3254 were significantly higher than for all other treatments. All data processing was carried out using the Microsoft Excel and means separated using Turkey’s procedure.