and use alternative feed sources – especially as the region may wellbe too dry to introduce successfully alternative fodder sources suchas cover crops (see below).In the temperate highland State of Mexico, land fragmentationis a serious issue; average farm size is significantly less than in theother two regions. Drip irrigation systems are common amongstwealthier farmers but poorer farmers remain largely dependent onrain-fed agriculture. Farmers typically use stall-feeding systems tomaintain dairy and beef cattle and ex situ feed is the norm. In situstubble grazing takes place in ejido common lands and in farmers’own fields. Only an estimated 5% of maize stover remains in the fieldand is incorporated into the soil. Cattle herds are under pressurein view of reduced communal grazing areas and livestock feedingdifficulties. There is a vibrant trade in maize stover, and interest-ingly, it is the poorer farmers without cattle who benefit from thistrade in the short term. The promotion of CA would exacerbate thetrade-offs between the retention and feed use of stover.“Maize-livestock systems in the tropical lowlands of Chiapas arecharacterized by high plant biomass production and large num-bers of cattle”. Farmers predominantly feed their livestock in situ– i.e. cattle graze the fields following the harvest and in paddocks.Poorer farmers again tend to have fewer cattle, and often sell stub-ble grazing rights. The relative abundance of biomass means thatthe trade-offs between cover and feed are less pronounced than inthe other two agro-ecological zones, and the prospects of stoverretention are aided by the curtailing of burning crop residues atland preparation.This paper provides a preliminary assessment of maize stoveruse and potential feed–mulch trade-offs. We did not intend toprovide a comprehensive assessment but instead to raise issuesand awareness – complementing studies elsewhere (Erenstein andThorpe, 2010; Erenstein et al., 2011; Valbuena et al., 2012). Indeed,the robustness of the assessment would have been enhanced bycasting the village surveys across a larger population and area – beit in terms of the sheer number of villages or the diversity of agro-ecologies covered. The study format is also less apt in bringing outwithin village diversity. One may hypothesize that poorer mixedfarmers depend more on maize stover as fodder source than non-poor farmers. We did observe that farmers without livestock weretypically