With Plate zz and 2 Text-figures) THIS paper gives the results of a year's field work on populations of the wild rabbit (Olyctolagus c. cuniculus (L.)). The experimental work was intended to be carried on over three years, but it has been thought advisable to publish the data already collected, since they cover probably the most important period of the rabbit's life cycle, namely, the breeding season. Few facts have been recorded about the rabbit, though there is much tradition and hearsay. To this extent these results may be regarded as an advance and as forming the groundwork of a fuller and more complete understanding of the ecological and economic status of the rabbit.
GENERAL CONSIDERATIONS I N THE ECOLOGY OF THE RABBITThere is practically no European litefature dealing with the subject. Niethammer (1937) is the only author so far traced who has experimented with marking wild rabbits : his methods were random ones (ferreting) and few observations were made. More work has been done in America, notably on the snowshoe rabbit (Lepur americanus) and on the cottontail rabbit (Sylvilogur poridanus), and in the case of the latter live trapping and marking techniques have been elaborated and census taking has reached an efficient level. Unfortunately, these techniques will not apply to the European wild rabbit on account of the difference in its habits.With the rabbit the main unit of the population is the colony, so that distribution is more analogous to a contour map, where each peak represents a focal point with a high resident population and the slopes from each peak represent the area of influence of each warren. The fact that the distribution of the population is not random means that traps cannot be set out in grids, a valuable procedure, which equalizes the trap efficiency at each of a series of trappings. The only practicable plan WBI to study a single warren, so that the figures derived from it could be applied to larger or smaller w-, or to total areas with known amounts of warren upon them.
TECHNIQUE (i) TrappingThe warren used in the experiments was chosen because it was fairly isolated, the nearest burrow being about IOO yd. away, and because it was in the middle of a gnw field and well defined in itcl boundaries. The actual warren area was about 4 acre in extent (PI. zz, fig. I), contained about 104 burrows, and was overgrown with a thick cover of nettles. D u r i n g the winter of 1938-9 a fence was erected around the warren, made of 6 ft. larch stakes 15 ft. apart, and wire n e b g 19 in. gauge, 1 4 in. mesh and 60 in. wide. This was s u n k into the ground about 9 in. deep and turned inwards a further 6 in. to prevent the rabbits burrowing underneath it.Rabbit' "smeuses" or outlets were set in this fence (Pl. zz, fig. z), and consisted of a short metal tunnel, rectangular in section, set in a hole in the netting fence to provide a way in and out for the rabbits. On such a small area as 4 acre the inhabitants could not get enough food without leaving the enclosure; in fact, the endosure was set as n...