An experiment in low‐input maize production in KwaZulu∗

Publisher Website
Google Scholar
Abstract
The experiment aims to establish the minimum inputs needed to achieve each of three predetermined target yields for dryland maize grain production in the Nhlangwinl ward of KwaZulu in southern Natal. Three farming systems were identified and assigned target yields based on the objectives of the farmers concerned, and an interaction trial was planted to establish optimal fertilization under a variety of cultural practices (weed control, pest control, plant population). The first year's data show that without weed control, yields were consistently below 0,5 t ha−1 even at moderate fertility levels. Although cutworm control significantly increased seedling survival rate, this was not reflected in the yield. Responses to low levels of fertilization were absent, and it was only at moderate input levels that significant crop yield responses were measured. For the subsistence farming system (where the aim is to feed the family) a yield target of 1 t ha −1 was set, which was nearly achieved using weed control, cutworm and stalk‐borer control, and no fertilizer. A return of 42c to 58c per hour of labour expended was achieved. A yield target of 4 t ha −1 set for the commercial farming system was not achieved by any of the treatments. For the semi‐commercial farming system (where a small quantity of produce is sold, over and above family requirements) a yield target of 2 t ha −1was set. This was achieved using weed control, cutworm, and stalkborer con trol, and a fertilization level of 250 kg ha −1 2:3:2(22) (0,5% Zn) and a topdressing of 100 kg ha −1 of limestone ammonium nitrate (28% N). A return of 79c to 83c per hour of labour expended was achieved. Further trials will show whether these yield levels occur consistently in the long term.

This publication has 0 references indexed in Scilit: