No abstract

SUMMARYTen accessions of Napier grass (Pennisetum purpureum Schumach.) were evaluated for their dry matter (DM) yield and forage quality in a semi-arid lowland (Katumani) and a relatively wet highland (Muguga) over seven growth cycles from 2011 to 2013 in tropical Kenya. Three biomass yield clusters were identified from the 10 accessions as high-yielding (HYC), medium-yielding (MYC) and low-yielding (LYC) clusters for both sites. Total biomass (shoot and root) yields (t ha−1) over the seven growth cycles were 25.3 for HYC, 22.2 for MYC and 19.6 for LYC at Katumani and 40.0, 41.4 and 29.1 at Muguga. Total biomass yield averaged over the study period was DM 22.4 t ha−1 at Katumani and 36.8 at Muguga. Rainfall productivity was higher at Katumani (28.8 kg ha−1 mm−1) than 20.8 kg ha−1 mm−1 at Muguga. Neutral detergent fibre (NDF) was lower in LYC, which was more leafy than the other clusters and there was little difference in NDF between the two sites.

Napier grass (Pennisetum purpureum Schumach.) is an important fodder and relatively drought-tolerant crop in tropical and subtropical regions, especially in developing countries. For this and other species, tools are needed for identifying drought-tolerant cultivars to aid selection for semi-arid environments. We determined tissue water status, carbon assimilation, biomass yield and forage quality for Napier grass cvv. Bana and Atherton grown in bins and subjected to three soil-water supply levels (100, 50 or 25% of field capacity) in glasshouses set at either low (15−25°C) or high (25−35°C) temperature regimes, over three growing cycles. Our aim was to explore whether differences in leaf water potential (LWP) and carbon assimilation rates could be reliable indicators of the relative yield potential and forage quality of the two cultivars in environments prone to water and heat stresses. At the low soil-water supply of 25% and low temperature, Bana had lower (more negative) LWP and relative water content (RWC) than Atherton, while at 50% and 100% soil-water supply, Bana had a higher tissue water status. Under the high temperature regime, Bana had consistently more positive LWP and RWC than Atherton, but the differences were not significant. The two cultivars had a similar CO2 assimilation rate (A) and there were no significant differences in the total dry matter yields over the three growing cycles. Water-use efficiency for above-ground biomass (kg ha–1 mm–1) was similar for both cultivars and was 28.5–35.1 under the low temperature regime and 16.9–22.9 under the high temperature regime. Neutral detergent fibre (NDF) was often higher for Bana at low water supply and low temperature than for Atherton, but the trend was reversed under the high temperature regime. Digestibility was generally improved under water-stressed conditions, and there was a positive correlation between NDF and both LWP and RWC measured at midday, but only under the low temperature regime. We conclude that LWP, RWC and A, alone or together, are inadequate for selecting cultivars for dry and hot environments, because cultivars may differ in other mechanistic responses to water stress and high temperatures.

The three water stress indices were poor, whereas vigorous early canopy development (determined as LAI) was a more reliable predictor of productivity potential of Napier grasses. In these dry environments, therefore, early rapid canopy development can be an effective indicator of yield potential and a credible selection criterion. © 2016 Society of Chemical Industry.