Soil Biol. B&hem. Vol. 29, No. 314, PP. 313-316, 1997 0 1997 ElsevierScienceLtd. All rights reserved Printed in Great Britain PII: zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCB SOO38-0717(%)0016!M 0038-0717/97 $17.00 + 0.00 ECOLOGICAL PREFERENCES OF SOME EARTHWORM SPECIES IN SOUTHWEST SPAIN E. G. SANCHEZ, B. MUNOZ, M. H. GARVIN, J. B. JESUS and D. J. DIAZ COSiN Departamento de Biologia Animal I. Facultad de Biologia, Universidad Complutense, 28040 Madrid, Spain zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONM (Accepted 26 June 1996) Smmary--A total of 80 samples were taken from three SW provinces of Spain (Caceres, Badajoz and Huelva), from which 28 earthworm species were collected and 22 soil factors analyzed. The zone stu- died has a continental climate with scarce rainfall, high temperatures during the summer months and Mediterranean vegetation. Attempting to characterize the ecological preferences of those species present in more than 5% of the samples, we used the technique of ecological profiles, to point out species’ preferences for such factors. 0 1997 Elsevier Science Ltd INTRODUCTION The results obtained about the relationship between distribution of earthworms and soil factors by different authors as Nordstrom and Rundgren (1974), Phillipson et al. (1976) and others, were ana- lysed by Mascato et zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA al. (1987). They attributed the contradictions between some studies to the fact that the samples were taken from areas with different biotic and abiotic characteristics, that the same edaphic factors were not analysed and that use was not made of the same statistical techniques. This study continues that initiated by other authors: Trigo et al. (1989); M. J. I. Briones (Unpublished Ph.D. Thesis, University of Santiago, 1991) and B. F. Souto (Unpublished Ph.D Thesis, University of Santiago, 1992), in western peninsular Spain, an area sutliciently variable and extensive to yield results about earthworm behaviour in a range of circumstances. Soil factors considered are basi- cally the same and the statistical technique, also used by the before mentioned authors, is that of ecological profiles (Gauthier et al., 1977; Daget and Godron, 1982). Therefore, the objectives were to determine the species autoecology in an area of Spain with a different climate from those more humid areas, stu- died by Trigo et al. (1989), M. J. I. Jones (lot. cit.), and B. F. Souto (lot. cit.). MATERIAL. AND METHODS A total of 80 samples were taken from 1987 to 1990 in southwestern Spain (Caceres, Badajoz and Huelva) (Fig. 1). Geological and botanical features were taken into account to select sampling areas. After selecting the sampling area, two sampling points were chosen: one next to a river or humid zone and the other in the dominant vegetation. Earthworms were extracted from the top 30 cm of 50 x 100 cm2 plots by the formalin and hand sorting separation method (Bouche, 1972). The soil at each of the 80 sites was sampled to a depth of 30 cm and 1 kg of this soil was taken. Twenty-two soil factors were determined by conven- tional methods: humidity (HUM), porosity (POR), aeration (AER), gravel (fraction > 2 mm) (GRA), sand (fraction between 0.05-2 mm) (SAN), silt (fraction between 0.002-0.05 mm) (SIL), clay (fraction <0.002 mm) (CLA), water retention ca- pacity at pF 2.5 (l/3 bar) (PF2) and pF 4.2 (15 bar)(PF4), pH of soil in water (PHH) and in KC1 (PHK), pH of litter in water (PHL) and in KC1 (PKL), soil temperature (TEM), litter temperature (TLI), carbon (C), organic matter (OM), nitrogen (N), sodium (NA), potassium (K), calcium (CA) and magnesium (MG). (Units: % for HUM, POR, AER, GRA, SAN, SIL, CLA, C, OM and N; “C for TEM and TLI; meq 100 g-’ for NA, K, CA and MG.) Data processing was carried out first by means of ecological profiles technique. The purpose of this analysis was to study the behaviour of each species in relation with each factor with variation intervals divided into classes (Daget and Godron, 1982). A canonical correspondence analysis (Jongman et al., 1987; Ter Braak, 1986) was also used. Only results obtained by the ecological profiles method are shown since both data analyses gave similar results. 313