They include at least half the sites listed in Table 3. Müller’s tables confirm my impression that Colonial sherds are exceedingly rare in northern Tlaxcala. In brief, many Postclassic villages apparently did not persist long enough to accumulate

any post-Conquest material culture detectable by surface survey. In Table 3 the more damaged sites outnumber those at the opposite end of the gradation. This may mean that erosion started a long time ago, i.e. early in the historical era, or that abandoned terraces are extremely vulnerable to erosion, and preserved only under exceptional circumstances. The gradual transitions between one category and the next suggest that even sites like Margaritas were once terraced. A counter-intuitive observation is that the best preserved Nutlin-3 ic50 sites are often those that experienced renewed cultivation and terracing in the Colonial or Independent periods.

Area A of La Laguna, where metepantles are superimposed on bench terraces, was cultivated as recently as the 1960s. It contrasts with area J, exploited in living memory only for its isolated patches of rough pasture. At Amoltepec the owner (in his eighties in 2003) reclaimed the land by cutting ditches into the eroded hillside, then, in Protease Inhibitor Library the late 1980s, re-shaped it with a bulldozer. The stone walls that survive are those incorporated into the berms scraped up by the bulldozer. In a contiguous sector of the hill recently re-forested with pine trees, no traces of terracing survive. At Ocotelulco and Tepeticpac, the good preservation of terraces may be due to their continued post-Conquest use. Recent cuts reveal Postclassic sherds in A horizons buried by younger terrace fill, which may be Postclassic or later. These two sites form part of the capital city of the

pre-Conquest province (Fargher et al., 2011a and Fargher et al., 2011b, 315–7) and are in many ways exceptional. Some of the risers probably had a defensive role, and Tepeticpac sits on a localized outcrop of less erodible sedimentary rocks. It is one of only two sites in Tlaxcala where I have observed terraces apparently stabilized by the re-growth of natural vegetation. The other one, Zarandelas, isothipendyl is at very high altitude (2900 m a.s.l), again on a geologically peculiar substrate, and the terraces show no clear association with any settlement remains. Both examples underscore how rare an occurrence the natural stabilization of abandoned terraces has been. All documented terraces of Postclassic age in Tlaxcala are of the stone-faced bench type. The more level treads may have been particularly suitable where, apart from crops, they had to support the weight of dwellings. In contrast, terraces without stone walls and with more sloping treads are the dominant field type today, the metepantles being the most common subtype. The partially buried metepantles documented at La Laguna are Colonial or later.