A major problem in using larvicides to control Culicoides is to recognise the sites requiring treatment. The relationship between plant cover and breeding of Culicoides (mainly C. impunctatus Goetgh.) was accordingly investigated in moorland areas of Scotland. Two sites were examined on Soutra Hill, Midlothian, and one each on Bannachra Muir and at Luss, Dunbartonshire. Three vegetational zones were recognised on Soutra (site 1): bogland (I), acid grassland (III) and a zone (II) characterised by vegetation preferring a much wetter habitat than either of the foregoing. C. impunctatus was virtually the only Culicoides species found in zone I; it was relatively less abundant in zone II, where C. albicans (Winn.) reached its peak, and occurred most densely at the bogland edge of zone III, where it was associated with C. cubitalis Edw., C. heliophilus Edw. and C. obsoletus (Mg.). Further into the grassland, C. impunctatus decreased while the other three species became more numerous and other species, of the group of C. pulicaris (L.), appeared. Within each main zone there were no significant differences between the various floristic groups as regards larval densities.

Fluctuations in population density observed in late autumn were regarded as more apparent than real, and attributable to larval movement in response to an oscillating factor, possibly soil-water level. This movement was recognisable in all zones but was most strongly marked where the ground was uneven.

On Soutra (site 2), where only zones I and III were sampled, there was remarkable consistency among the larval densities of the various groups of samples examined. This applied whether the samples were classified on their angiosperm flora, moss cover or spatial arrangement. The main source of variation was within groups and it was found that two adjacent small plots (12 yd. & z.Times; 12 yd.) which were macroscopically identical showed consistent differences over a period of ten months, covering the transition from one generation to another.

On Bannachra Muir, the larvae of C. impunctatus were concentrated in an area covered by Juncus articulatus and Sphagnum where the water flowed down the hillside near the surface. C. impunctatus avoided both an area of drier leaf mould and a stagnant ditch. C. albicans, C. heliophilus and C. obsoletus preferred the ditch to the hillside flow.

Around a hillside trickle at Luss, larvae of C. impunctatus and C. truncorum Edw. were abundant in subequal numbers. In spite of the restricted size of the area (about 10 sq. yd.) they never occurred in equal numbers in any set of samples, but one or the other predominated. C. truncorum showed a preference for the very wet parts.