Having walked many miles in these same areas of Dimmit and La Salle counties, I often have found myself trying to imagine how I would describe the walk at the end of a hard day. Thick brush near drainages (Plate 1) and on rocky ridges would have made a big impression because of the work involved in hacking through. Such areas of resistance would be remembered at the day's end. However, much of the region, though covered with frequent shrub clumps, is very open to a man on foot or horseback. The north part of Chaparral Wildlife Area, where presumably no major manipulations (Whittaker et al., 1979) have occurred, remains very open in aspect. In wet years, anyone interested in pasturage would certainly be inclined to emphasize the grass component when traveling the region. During droughts, on the other hand, even low shrubs would dominate ones perception (Plate 2). Lehmann (1969) emphasizes this particular point. All things considered, it must be concluded that today's woody vegetation has a long history in the region and that it has been compatible with high grass production in the past. Of course, woody species have increased in density relative to grassland in many areas subject to overgrazing over the past century.

Contemporary ranchers of the region heard from fathers and grandfathers about the open conditions of the past. Virtually all of those early ranchers arrived near the end of a period of intensive sheep ranching on the Rio Grande Plain (Lehmann, 1969). It may be that overgrazing by sheep actually suppressed many woody shrubs and helped reduce woody plant cover near the beginning of the cattle era. In a semi-desert region of Australia, three dominant shrubs have declined under grazing by sheep and rabbits, due to suppression of juvenile recruitment (Crisp, 1978). Indeed, overstocking of sheep in drought-prone South Australia threatens to drive various woody species to extinction by suppression of seedling recruitment (Lange and Willcocks, 1980). Several decades of seedling suppression in South Texas would have enhanced the pattern of scattered older individuals characteristic of episodic recruitment in trees and shrubs. It is worth noting that once sheep were removed from South Texas, brush increased in density, particularly where overgrazing was severe (Lehmann, 1969).

While it is clear that mesquite and other woody shrubs of the Rio Grande Plain are not invading from afar, the question of "noxiousness" is much more complex. A rancher's goal is to convert as much of the primary production of the range as possible into beef at a minimal cost. The Journal of Range Management and similar range science publications have repeatedly documented what every South Texas rancher knows: if you remove mesquite and associated woody shrubs, grass production (particularly imported African grasses) and cattle carrying capacity increase--at least for a while (if it is not a drought year). The obvious conclusion, that mesquite inhibits beef production (and is therefore noxious) seems reasonable but is, when critically viewed, not the only valid interpretation of the facts.

Most experimental studies of mesquite removal suffer from an insufficient duration (many are two year master's degree projects) and only evaluate the end agronomic result (production of grass or livestock) without investigating details. Other such studies are difficult to interpret because of the complexities inherent in field manipulations (e.g., Tiedemann and Klemmedson, 1977). I am unaware of any experimental work which demonstrates continued high grass productivity under both continuous grazing and continuous mesquite suppression over 5, 10, 15 year periods. In the absence of such data, long range, economic projections from short term studies are suspect. Indeed, in the absence of long duration studies, we cannot exclude the alternative hypothesis: mesquite is beneficial rather than noxious with respect to forage and beef production.