George A. Dyer; Alejandro López-Feldman

“ Seed replacement varies to a much greater extent across regions than across elevations, e.g., rates in west-central Mexico are five times those found in the southeast (Table 2A) . Furthermore, interregional variation is not reduced markedly after controlling seed characteristics associated with replacement, e.g., seed source (Table 2C) . Accounting for both location and seed characteristics is crucial nevertheless because frequent interactions among factors (e.g., altitude-by-region and altitude-by-source interactions) can obscure the patterns of variation. ”

“ A definitive explanation nevertheless requires recognizing that farming decisions are seldom taken in isolation; that a given practice usually depends on many others taken by either the same farmers, their neighbors or more distant members of society; or that crop management is not only caused by environmental conditions but the cause of those conditions. ”

“ While acknowledging these factors’ “unambiguous” influence, Brush & Perales [11] seek to unravel farmers’ present-day contribution to maize evolution by analyzing the way in which the structure of human societies affects that of maize populations. Based on the most thorough survey of maize-farming communities to date, they conclude that in Chiapas, Mexico, the effect of social factors is evident only at the landscape level, while the environment contributes to maize diversity at greater spatial scales. ”

“ In Chiapas, the age, source and replacement rates of seed all vary monotonically with altitude. In lowland localities, for instance, acquaintances outside the family are the most important source of seed, while family sources become increasingly important at higher elevations. Only 3% of seed lots maintained by indigenous highland farmers are replaced annually (i.e., 1 - p = 0.03) ; the average age of a seed lot is 28.5 years. ”

“ Our analysis also highlights the lack of independence among the factors that determine introduction rates, due in this case to altitude-by-region interactions. Namely, the pattern of variation across altitudinal zones that is highly significant at the national level—i.e., introduction rates are most common at mid elevations—occurs only in southeastern and west-central Mexico, while other regions exhibit contrasting patterns: in the north, seed is introduced most in the lowlands, in central Mexico it is introduced most in the highlands. ”

“ Indeed, ethnicity seems to explain considerable variation in replacement rates at every elevation in Chiapas: indigenous farmers consistently replace seed less often than mestizos. At mid elevations, for instance, seed replacement rates are almost twice as high among mestizos as among indigenous farmers (i.e., 0.07 vs. 0.04) . Ethnicity and altitude could have similar roles in seed replacement throughout the southeast, but additional sources of variation clearly are at play within the region. ”

“ The type, source, geographic origin and ownership of seed help explain observed rates. Yet, controlling for the characteristics of germplasm barely reduces interregional differences vastly exceeding variation across elevations. With few exceptions, monotonic altitudinal trends are absent. Causal relationships between management practices and the physical environment could determine farmers’ wellbeing and crop conservation in the face of climate change. Scarce and inconsistent data on management nevertheless could prevent an understanding of these relationships. ”

“ More importantly, discrepancies also suggest that critical causal relationships underlying crop management (and diversity) remain unidentified. Individual factors discussed here offer only conditional explanations; their influence is rarely independent of other factors. For instance, replacement is not consistently lowest in the highlands or unfailingly higher for improved varieties than for landraces; rate differences between two types of seed can be reversed depending on other characteristics. ”

“ While not all sources of variation within this environment have been identified, the region’s mid elevations clearly experience distinct rates of seed replacement. According to ENHRUM data, the average replacement rate at mid elevations (0.16) differs significantly from rates in the highlands and lowlands (Table 2A) . Although the rate decreases when data from other studies in this altitude-by-region environment are considered (0.08) , differences with other altitudinal zones persists. ”

“ Planting hybrids, keeping landraces: Agricultural modernization and tradition among small-scale maize farmers in Chiapas, Mexico. World Dev 39: 1434-1443 .10.1016/j.worlddev.2010.12.010↑ Bellon MR, Adato M, Becerril J, Mindek D (2005) Poor farmers’ perceived benefits from different types of maize germplasm: the case of creolization in lowland tropical Mexico. World Dev 34: 113-129 1 2 3 4 Smale M, Aguirre A, Bellon M, Mendoza J, Manuel Rosas I (1999) Farmer Management of Maize Diversity in the Central Valleys of Oaxaca, Mexico. ”

“ Econ Bot 51: 20-38 .10.1007/BF02910401↑ Snively-Martinez AE (2009) Perceptions of change in horticultural subsistence strategies in a rural Mexican community: San Francisco Pichataro, Michoacan. MS Thesis, Washington State University. ”