Habitat preferences of the Wren on English farmland

DURING THE MONTHS January-March 1963 England suffered its coldest winter since 1740 and several resident species were reduced in numbers.The British Trust for Ornithology's `Common Birds Census', begun as a pilot survey in 1961, was able to compare the results of repeat censuses on sample plots totalling 9,980 acres of agricultural land for the breeding seasons of 1962-63.Between these two seasons the density of Wrens Troglodytes troglodytes fell from 3.3 males to 0.7 males per 100 acres (or 8.2 and 1.7 per 100 hectares respectively), a reduction of 79 per cent.(Williamson and Homes 1964).More recent winters have been mild and the Wren population on farmland has shown a steady annual increase.The species was still below the 1962 population level in 1966 (Bailey 1967), but surpassed this density after a further steep increase in the following year (Bailey and Batten 1968).The census data permit an examination of the recovery of the Wren with a view to investigating its habitat requirements and preferences in the agricultural environment.The assumption is made that the few Wrens remaining after the hard winter, together with the new intake in 1964 and subsequent breeding seasons, would settle in the most suitable haunts then available; and only when the optimal sites had been filled would competition force an increasing number of birds to accept less suitable, and perhaps even marginal, situations. METHODThis study could not have been made without the careful field-work of many observers who contribute annually to the Common Birds Census in various parts of the country.Nor would it have been possible without the use of the mapping method, and as this was not introduced until 1964, it was not possible to carry the examination back to 1963 or earlier.The mapping method requires that song and other territorial activities should be plotted on a large scale map (25 in.= 1 mile) of the census area on a minimum of eight visits in each breeding season, in successive years.The field data are later transferred to species maps, on which individual territories can be recognised when the records of singing males form discrete groups or `clusters'.Thus the number and distribution of the territories can be ascertained, and one year's results compared directly with another's.The observer is also asked to supply a map showing in detail the main physical features of his census areaincluding the crops, situation of streams, ditches, pools, gardens, spinneys and so on, and the dominant tree and hedgerow cover.Tracings are made of each year's species maps when the analysis