Habitat preferences of three Chorthippus species (Orthoptera: Acrididae) in the Chelmsford area grasshoppers. However, many of the remaining habitats which are suitable for grasshoppers are being destroyed due to increasing amounts of urban development (Marshall & Haes 1988). Despite these changes in the British countryside, species such as Meadow Grasshopper Chorthippus parallelus and Field Grasshopper Chorthippus brunneus arc widespread in distribution and occur in a variety of habitats. C. parallelus, for example, can be found in sandy pastures, woodland rides, roadside verges, wasteland, mires, salt-marsh, and grassy moorland, and its adaptability has enabled it to exploit the 'grasshopper niche' in many habitats (Marshall & Haes 1988). C. parallelus and C. brunneus are very common species in Essex, although Wake (1997) states that the latter species is the more common, occurring in 57 10-kilometre squares. Lesser Marsh Grasshopper Chorthippus albomarginatus, however, is a locally common species in the south-east of Britain (Haes & Harding 1997) with a distinctly coastal distribution in many counties. In Essex, this species is found mainly in coastal habitats (salt marshes and seawalls) or along river courses. C albomarginatus has shown a tendency in recent years to increase its range to more inland habitats in the county (Wake 1997). It is the aim of this paper to ascertain whether habitat preferences exist for the three Chorthippus species in the Chelmsford area. The survey aimed to determine whether densities of grasshoppers were significantly lower on farmland sites than in potentially more favourable swards, such as those found on heathland. Method Sampling of grasshopper populations The method used for surveying grasshoppers in this study was adapted from Ausden (1996), who ascertains that more mobile invertebrates such as grasshoppers may be recorded using randomly placed quadrats. The size of (the quadrats used in the survey was 4m2 (2 x 2m) as Richards & Waloff (1954) state that a large size of quadrat is advantageous w hen surveying adult grasshoppers, particularly if they are present in low densities. Only adull grasshoppers were recorded during Illis study because in this life stage identification can be confirmed without the capture of individuals (Richards & Waloff 1954). Ten quadrats were placed using random co-ordinates in a 100m2 plot (10 x 10m) at each study site on every survey occasion. The comers of each quadrat w ere marked with poles without disturbing the grasshoppers within by casting shadows. Each plot at the study sites was surveyed once in July and once in August in both 2000 and 2001. A total of 40 quadrats were therefore completed for each site. The number of adult individuals of C. albomarginatus, C. brunneus and C. parallelus w ere recorded in each quadrat by brushing the vegetation with a pole to cause any grasshoppers present to jump (Richards & Waloff 1954). The 'flushing' of grasshoppers was conducted in a standardised method ensuring coverage of the whole quadrat by moving from one edge to the other, sweeping the vegetation in a 180" arc. If a grasshopper jumped out of the quadrat it was still recorded, however, if a grasshopper leapt into the quadrat from outside it was not counted. 84 Essex Naturalist (New Series) 20 (2003)