SummaryThis study assessed simple macro-propagation methods, which build on methods reported for enset multiplication, for producing banana seedlings in four different Musa cultivar use groups across four unique agro-ecologies (900-1,815 m a.s.l.). The methods consisted of a substrate of loosened soil or a soil-and decomposed farmyard manure mixture under either a semi-cylindrical tunnel made of wooden/stick frames covered with knitted elephant grass stems or a 5 cm thick mulch cover of spear grass/elephant grass. A standard macro-propagation unit, made of wooden planks and thick plastic polythene sheet covering, with sawdust as substrate, served as a control. The average number of harvested plantlets per corm, irrespective of cultivar and site, varied between 7.5 under semi-cylindrical tunnel without manure and 12.6 under the standard macro-propagation unit. In general, and across sites and cultivars, there were no significant differences (P>0.05) between macro-propagation methods in the mean number of harvested plantlets. Irrespective of method and cultivar, fewer plantlets were harvested at the high altitude sites. Significantly more plantain plantlets (12.1-14.5) were produced at low altitudes (900 and 1,066 m) while dessert (12.8) and cooking (12.7) types performed better at 1,700 m. Significantly fewer plantlets per corm were produced by the ABB types (7.9), while the highest numbers were realized for the plantains (12.2). The net profit from sale of plantlets from the simple macro-propagation units was comparable and sometimes higher than that from the standard unit. The high initial cost and skills needed for establishing the standard macro-propagation unit have often hindered its adoption. The low cost, use of local materials and comparable returns from the simple macro-propagation units suggest that they could be a good alternative for banana seedling production under small-scale farmer conditions.