No abstract

Vines are the major source of planting material in sweetpotato. Extended dry spells hinder conservation of vines and in turn affect the availability of planting material at the onset of rains in southern Africa. In some cases, improved sweetpotato germplasm has been lost by smallholder farmers in Mozambique due to prolonged dry spells. Small to medium roots provide an opportunity to conserve germplasm and get planting material at the beginning of the rainy season. The objectives of the study were to (i) measure sprouting ability of diverse germplasm of sweetpotato - farmer varieties, improved clones and released varieties and (ii) estimate their ability to provide planting material for the next crop in southern Mozambique. Trials with 29 genotypes were established in a randomized complete block design with two replications at Umbeluzi Research Station and Nwalate Farm in 2015, 2016 and 2017. At harvest, 14 small to medium roots were selected and stored in small plastic dishes filled with dry sand at Nwalate Farm. After four months in storage, 10 similar roots were taken and planted in 1 m row plots arranged in a randomized complete block design with two replications. The trials were irrigated to initiate sprouting and support plant growth during the first four weeks. Data collected were analysed using SAS 1996. All the tested genotypes sprouted after sowing. The number of sprouts per root were significantly affected by the genotype, location and genotype x location x year interactions. Caelan had the most sprouts per root. Sprout length measured at six weeks after sprouting was also significantly affected by genotype, location, year and genotype x location x year interactions. Caelan had vines each long enough to provide 10 cuttings of 10 cm length for rapid multiplication. The number of cuttings depended on the growth habit of the variety. Irene, a popular variety in Mozambique, is erect and bushy and could only provide four cuttings over the same period. Growth habit especially under a changing climate should be considered in breeding programs as an option of facilitating a sustainable and easy seed system.

Drought has negative effects on sweetpotato production. Two experiments with two watering treatments (irrigated and water-stressed) were conducted at Umbeluzi Research Station in 2015. The objectives were to (i) determine response of 48 sweetpotato germplasms to mid-season drought, (ii) determine best traits for improvement of storage root yield under mid-season drought and (iii) assess the selection criteria for identifying drought tolerance in sweetpotato germplasms. The irrigated and water-stressed trials received 640 and 400 mm of water, respectively, throughout the season. Water stress was imposed from 30 to 70 days after planting. Each treatment had two replicates arranged in a randomized complete block design. Data collected on storage root and vine yield and derived drought tolerance indices including harvest index were subjected to analysis of variance in R. Sweetpotato germplasms with high storage root yield under mid-season drought were associated with a high harvest index. Harvest index stability and the geometric mean are key to identifying cultivars with high and stable storage root yield under both treatments. MUSGP0646-126, Irene and Ivone combined both low TOL, SSI, HI and high yield storage root yield across the treatments and over seasons. The use of drought and harvest indices is encouraged for selecting improved cultivars for varied production environments and their regular use in accelerated breeding schemes is suggested.

Optimisation of sweetpotato production potential is not realized in sub-Saharan Africa where dry spells are common after the growing season. Lack of seed cultivars at the beginning of the rainy season is the challenge and identifying those cultivars with the ability to survive a 4 to 7-month dry spell is an important objective in these regions. To this end, the purpose of the present study was to estimate survival of vines of different cultivars under drought stress. Three harvesting times set as 5, 9 and 11 months after planting were established at Umbeluzi, Nwallate and Gurue in 2015. For each harvesting time, 37 clones were evaluated in a randomized complete block design with two replications. Data collected on yield, vine length and number of sprouts were analysed using SAS 1996. Resisto had long stems but no vines after 9 months at any of the sites. The ability of some clones to reach 9 months and 11 months with few vines at Umbeluzi and Nwallate offer sources of planting material for the subsequent season. Sprouting had broad sense heritability above 50%. Sprouting is known to be an important aspect of cultivar survival. While we show that vine length and thickness could be traits responsible for sweet potato drought tolerance