ABSTRACT

Groundnut (Arachis hypogeaa L.) productivity in Nigeria has remained low over the years due to the inherent low fertility status of the soils of the savanna, a region where it is mostly grown by small holder farmers with limited imputs. Groundnut can fix atmospheric nitrogen through symbiotic association with native rhizobia but unfortunately, the amount of N2 fixed is usually not enough due to the presence of ineffective or low numbers of native rhizobia. This study was designed to investigate the comparative response of groundnut to rhizobium inoculation and nitrogen fertilizer with or without phosphorus fertilizer on an Alfisol in the Northern Guinea Savannah of Nigeria. Soil samples were collected from a P-deficient plot on the Institute for Agricultural research/ Faculty of Agriculture (IAR/FOA) experimental farms located in Samaru Zaria and used to conduct two screen house trials using the groundnut genotype “SAMNUT 24” as the test crop. The first experiment consisted of six (6) inoculants (four indigenous rhizobial strains namely: SNN 343, KBU 26, SBG 234, SAMFIX 703 one commercial inoculant (HISTICK) and a reference strain (NC 92) alone, the inoculants each combined with 20 kg N ha-1 mineral nitrogen as starter dose, and mineral nitrogen alone at 0, 20, or 40 kg N ha-1, adding up to fifteen treatments all termed as nitrogen (N) sources. The second factor was phosphorus at three rates (0, 30 or 60 kg 2O5 ha-1) applied as single super phosphate. The N sources (SNN 343, KBU 26, SBG 234, SAMFIX 703, NC 92, HISTICK, SNN 343 +20N, KBU 26 +20N, SBG 234 +20N, SAMFIX 703 +20N, NC 92 +20N, HISTICK

+20 N, 0 kg N ha-1, 20 kg N ha-1 and 40 kg N ha-1) and P rates (0 kg P2O5 ha-1, 30 kg P2O5 ha-1 and 60 kg P2O5 ha-1) were factorially combined (fifteen x three) to give a total of forty five (45) treatments and were laid down in a Randomized Complete Block Design replicated three

times. The second trial was set up to assess the residual benefit of rhizobium inoculation on following crops. Soils previously inoculated with SNN 343, KBU 26, SBG 234, SAMFIX 703, NC 92 and HISTICK and a control, were combined factorially with three levels of P (0, 30 or 60 kg P2O5 ha-1) to give a total of twenty one (21) treatments. The RCBD was also used with three replications. A non- nodulating groundnut variety (ICGL 5) was included as a reference crop to estimate the amount of biological nitrogen fixed. In the first trial, the use of mineral ni