Chemical and ecological studies on plants and soils of ultrabasic and other areas on the island of Rhum, Scotland

Abstract

The climate, geology and vegetation history of the Island of Rhum are briefly described. Rhum has a large area of ultrabasic rock, mainly over 350m which has distinctive soils and vegetation. The vegetation of selected sites on several rock types was described from 32 transects with a total of 285 1x1m quadrats. At each quadrat a soil sample was taken and analysed chemically. The vegetation and soil data were ordinated (by DECORANA) and classified (by TWINSPAN). Twelve vegetation classes emerged which were related to previously described communities. The soil factors which significantly correlated (rs) with the ultrabasic classes were: total nickel, chromium and cobalt, pH and exchangeable nickel and calcium. There is a wide range of soils on the ultrabasic and most are unusual for this type since they have low Mg/Ca quotients and a sandy texture. The above-ground parts of six species were analysed chemically and samples from the ultrabasic soils had fairly high concentrations of Ni and in some cases high Fe. Correlation coefficients for soil-soil, soil-plant and plant-plant elements provided some insight into selectivity and possible mechanisms of adaptation to the ultrabasic soils. Soil solutions were extracted from 21 samples using a centrifugation technique and analysed chemically. Four of these analyses from contrasting soils were used as a basis for culture media for experiments to test the importance for plant growth of certain ions (Ca2+, Mg2+, Ni2+, H2PO4 and Zn2+). RGR’s were measured in culture media with several ions varied in a factorial manner and there were two harvests (after 3 and 6 weeks). Ni2+ (0.2 mg 1-1) was mildly toxic to a non-ultrabasic race at Harvest 1 but there was no effect at Harvest 2. The implications of this are discussed. The non-ultrabasic race had a reduced RGR in the Ca and Mg concentrations (combined) in the solutions simulating those in the ultrabasic soils, but grew best of all the races in the solutions simulating those from its site of collection. The plants grown in the culture media were analysed chemically and compared with the same species which occurred in the field. The experimental plants had lower concentrations of Fe, and higher of K, but in other respects were similar to the field plants. Field experiments and observations were made on the effects of added major nutrients to barren areas and erosion. Unfortunately most of the nutrient addition experiments were lost, but earlier work was confirmed that nutrient-addition causes an increase in plant cover on expo9sed barren ultrabasic areas. Erosion is still rapidly occurring in some places. No single factor can explain the distinctive vegetation of the ultrabasic areas on Rhum, but the following singly and in combination are probably important: high Ni (with its greatest effect in dry spells); fairly high soil pH (for Rhum); low major nutrients; soil physical factors, particularly its sandy texture, frost-heaving and erosion (probably resulting from past-burning and grazing). The effect of a high Mg/Ca quotient seems not to be important on Rhum.