Reliable data on the elemental composition of various materials are needed to solve many problems in ecology. Despite the continual development and improvement of analytical methods and measurement apparatus, in many cases, especially when the experimentally determined content of the elements is close to the sensitivity threshold of the method, systematic and random errors can have a large effect on the measurement results [1 ]. It is obvious that in such cases measurements must be performed by several independent methods in order to increase the reliability of the results. In the present work four nuclear-physical methods - neutron-activation (n), y-activation (y), x-ray spectral (p), and x-ray-fluorescence (x) analyses - were used to determine the content of 18 elements in samples of high elecampane, central-Asian mint, field horsetail, mixed grasses, and fruits of the Turkestan dogrose, which were collected on two stationary plots A and B in the Bashkizil'sai section of the Chatkal" biospheric preserve [2]. X-Ray fluorescence analysis was used to determine the elemental composition of soil on the site B, which was not previously studied. The rms errors of the results obtained by different methods and the differences in the accumulation of elements depending on the type of plants, the parts of the plants, and the location of the plants were estimated on the basis of the experimental data. Samples of elecampane, mint, horsetail, and mixed grasses were taken during the budding and flowering periods, and samples of the dogrose fruit were taken during the fall from two sites with light-brown soils in high-mountain meadow-steppes [3t. The sites A and B are located in a thin juniper stand in the foothills (-1200 m above sea level) and on the southern slope (-1270 m above sea level), respectively, of one of the high Chatkal' ridges. The site A has greater moisture and is more shaded than the site B. Samples of soils from site B were obtained from a depth of 10--20 cm by the envelope method. The samples were air dried, pulverized in a mill, and ground into a powdered state in an agate mortar. The amount of material required to prepare samples was determined by weighing. The sample-preparation method is described in detail in [4-71. The investigations were performed on setups assembled on the basis of the VVI~R-SM nuclear reactor at the Institute of Nuclear Physics of the Uzbekistan Academy of Sciences [4], the MT-22S microtron of the Samarkand State University [5], an I~G-4 electrostatic generator from the Joint Institute of Nuclear Research [6], and an RFA spectrometer with a 241Am radionuelide source from the Scientific-Research Institute of Nuclear Physics at Tashkent State University [7], using the SBMT, SVR-7,