130 BR. J. ENT. NAT. HIST., 7: 1994 



In this paper we present data obtained by running paired moth traps, within sight 

 of each other, either side of a sharp habitat boundary between dense woodland and 

 open grassland. The results are discussed in relation to the factors which may affect 

 habitat specialization, including larval foodplants, roosting sites and microclimatic 

 differences between trap sites which may affect flight. 



Methods 



Trapping was carried out using paired light traps between 16 and 30 June, 1989 

 and between 28 June and 6 July 1990, in Juniper Bottom, Box Hill, Surrey. Juniper 

 Bottom is an east-west running valley. The vegetation in the bottom of the valley 

 is chalk grassland, close-cropped by rabbits, with a few standard broad-leaf trees 

 and patches of mixed broad-leaf scrub. It is very rich in terms of number of plant 

 species. Conversely, large areas of both sides of the valley are covered by mature 

 yew woodland. The canopy of this woodland is extremely dense, and the under canopy 

 is dark with virtually no ground vegetation except beneath occasional deciduous trees, 

 such as whitebeam, and below breaks in the canopy caused by the 1987 October gales. 

 The boundary between these yew-covered slopes and the bottom of the valley is sharply 

 defined, consisting of a rather impenetrable natural mixed broad-leaf hedge. The site 

 was chosen for this study because of the contrast between the habitats either side 

 of the hedge. The extreme paucity of the ground vegetation under the yew, large 

 parts of the yew woodland effectively being a natural monoculture, provide a sharp 

 contrast to the species richness of the chalk grassland. A list of the identified species 

 of plants growing within 5 m of the traps in the two habitats is given in Table 1 . 



Up to three pairs of traps were run on a night. Three types of trap were used; 125- W 

 Robinson traps; 100-W 'dustbin' traps and 12-W Heath traps, both traps of a pair 

 being the same type. Robinson and dustbin traps were powered by E650 Honda 

 generators. Heath traps were powered by 12-V car batteries. One trap of each pair 

 was set up approximately 10 m inside the yew woodland, the other being placed 

 approximately 10 m outside the yew wood in the chalk grassland. In 1989, the positions 

 of pairs of traps along Juniper Bottom varied between nights. In 1990, three pairs of 

 trapping sites were occupied each night, the types of trap at each site varying between 

 nights. Traps were run for 3-4 h between 9.30 p.m. and 1.30 a.m. 



When trapping was terminated for a night, the traps were stoppered and transported 

 to Juniper Hall Field Studies Centre, where the catches were scored the following 

 morning. All macrolepidoptera were identified except the pugs (excluded due to time 

 constraints and the inexperience of the scorers with this group). 



In 1990, measurements of windspeed, temperature and humidity were taken, at 

 each trap site, at intervals throughout the trapping period. Windspeed was measured 

 over 10-min periods using a cup anemometer. Temperature and humidity were 

 measured using a whirling hydrometer. Readings of light intensity at different sites 

 in each habitat were taken using a photographic light meter. 



Table 1. Plant species within five metres of (a) the traps inside the yew woodland and 

 (b) the traps in the chalk grassland. 



(a) yew, mosses, dogs mercury, whitebeam, lichens. 



(b) silverweed, nettles, thistles, hawthorn, dogs mercury, hairy saint John's wort, bramble, 

 hogweed, speedwell, bedstraws, dark mullein, marjoram, thyme, ribbed melilot, convolvulus, 

 dead-nettle, docks, plantains, buttercups, beech, fine-leaved sandwort, mouse ear, clovers, 

 dogwood, wood sage, hazel, beaked hawk's-beard, bugle, crosswort. 



