Over the years maize breeding by small farmers has been aided to increase yield in rainfed small areas with specific hydrometeorologic, geologic, and orographic conditions (referred to as ecological niches). The selection performed by small farmers is based on such characters as type, health, and length of ear, number of kernel rows, type, filling, color of grain, low proportion of cob, among others. This quasi-empirical process represents a reliable option toward adaptation to environmental changes. In this report we point out the specificity of native maize in large and narrow areas, called horizontal and vertical adaptation, respectively. Maize breeding has traditionally been performed by conventional breeders for horizontal adaptation while small farmers have selected for vertical adaptation. To better understand the potential reach of this process we set up experiments in order to identify the specificity of native maize landraces measured across a range of environments within a defined domain.  The first experiment involved 80 varieties of maize and was established in four ecological niches in the Mexican state of Michoacán: Paracho (PAR), Santa Clara del Cobre (STC), Zitacuaro (ZIT) and Ocampo (OCA), planting varieties in a first cycle into their own micro-region of provenance; then, in a second cycle the three best varieties from PAR, two from STC, two from ZIT and three from SCR were selected and all tested across locations. The varieties of each ecological niche showed the highest average yield in relation to the varieties from other ones. In the second experiment 29 varieties were collected from three levels of altitude L1, 300 masl (8 varieties), L2, 800 masl (9 varieties), and L3, 1200 masl (12 varieties); all varieties were tested in each location, including two soil conditions (shallow and deep soils) in L1. Particular varieties native to the specific areas where they were evalauted outperformed in yield the rest of tested varieties.The results show general agreement with vertical adaptation, except in L2 where L1-deep-soil varieties outperformed the original varieties of L2. Considering the scenarios of a changing climate the potential of this millennial approach represents a reliable option to the sustainability of agricultural activities and the preservation of the agricultural biodiversity.