No abstract

The validity of the concept of "ingle—tree influence circles" was tested in a forest dominated by Tsuga canadensis and Liriodendron tulipifera on steep slopes in the Appalachian Mountains of the eastern USA. Samples of forest floors and the 0—5 cm depth of mineral soil were collected at 135 locations within plots 1.3 ha in total area. Examination of particle—size distribution and pH at 33 of these locations confirmed that parent material to a depth of 100 cm was uniform throughout the study area. Soils were sandy and generally low in nutrients. Levels of pH, Ca, Mg, K, and mineralizable N tended to be higher, and forest floor mass tended to be lower, under the crowns of L. tulipifera compared to T. canadensis. Differnces between tree species were better expressed on areas lacking Rhododendron maximum understories. These results indicate that single—tree influence on soil properties is detectable even in mixed stands on steep slopes, and that the soil landscape may be considered a mosaic of profiles reflecting the occurrence and chemical characteristics of the ground cover vegetation and of individuals of the various tree species present.

No abstract

Shifting cultivation of com (Zea mays) was widely practiced on steep, mesic slopes in the Appalachian Mountains during the period 1800-1930. Contemporary reports often emphasized both the severity of erosion during the period of cultivation, and the rapidity of reforestation following abandonment of agriculture. In eastern Kentucky, old fields that originated during this period are presently occupied by nearly pure yellow-poplar (Liriodendron tulipifera) stands that contrast with the mixed mesophytic forests found on areas that have never been cultivated. Soil properties on four old fields :::::60 yr after abandonment were compared with those on uncultivated areas that were similar in physiography, geology, and, apparently, in original soils and vegetation. This paired-plot comparison examined the effect of both past land use and present vegetation on soil properties.Intensive sampling of the 0-60 em soil depth revealed only minor differences in the properties of the formerly cultivated vs. uncultivated soils. Significantly higher concentrations of coarse fragments (particles > 2 rom) in the 0-5 em depth of cultivated soils were attributed to the erosion of :::::2 em (100 Mg/ha) of fine earth (particles <2 rom) and the subsequent burial of the coarse lag material. Significantly lower concentrations of available phosphorus in the 0-40 em depth of cultivated soils, averaging I mglkg less than in uncultivated soils, may have resulted from depletion during cultivation or from differential retention in the aboveground biomass following reforestation. Exchangeable calcium concentrations and pH of cultivated soils were significantly higher in the 0-10 em and 0-60 em depths, respectively, compared with uncultivated soils. Average plot values of exchangeable calcium and L. tulipifera basal area were linearly related (r 2 = 0.87), as were the average values of these two variables over a downslope composition gradient on uncultivated areas (r 2 = 0. 78). These relationships, together with the greater content of calcium in the 0-60 em depth of cultivated vs. uncultivated soils, suggested that relatively large amounts of calcium were accumulated in the surface soil as a result of the nutrient-cycling characteristics of the L. tulipifera-dominated old-field community.Fields cleared during the episode of shifting cultivation were typically located on landscape positions characterized by favorable water regimes, and by deep, porous soils with uniform physical properties throughout the rooting depth. Surface soils on such sites are resistant to permanent erosional degradation, and are rapidly rejuvenated by processes associated with reforestation. Thus the effects of both past land use and present vegetation on the properties of these soils are for the most part superficial and ephemeral.

Longleaf pine—wiregrass—turkey oak (Pinus palustris—Aristida stricta—Quercus laevis) occurs as isolated islands, 60—4000 ha in area, in a matrix of sand pine—scrub (Pinus clausa—Quercus spp.) on deep sands of the Ocala National Forest, Florida. Striking contrasts in physiognomy and species composition, and sharp, stable boundaries suggested that soil differences determined vegetation boundaries. Examination of soils to a 500 cm depth at 130 locations revealed no consistent differences in profile morphology, particle size distribution, or extractable nutrients. Thus the hypothesis that soil differences were responsible for the distribution of the two communities was not sustained. Organic matter and nitrogen contents of surface (0—60 cm) soils likewise did not differ between plant communities. Differences in surface horizon color were ascribed to differences in soil—forming processes: frequent additions of fine charcoal and active faunal mixing under longleaf pine vs eluviation under sand pine. Biogenic opal extracted from surface soils at 75 locations failed to provide evidence of long—term community stability. Isolines of total opal mass were concentrically patterned around longleaf islands, but opal mass and morphology did not consistently differ across vegetation boundaries.