A wide variety of habitats is found on the Howbery Park site, ranging from formal mowed lawns, to woodland, riverbank, and wild areas. A particularly productive area for butterflies was created artificially, and accidentally, in the 1970s, when an area roughly 100m by 30m was covered in a deep layer of sand for the purpose of performing experiments on meandering of rivers. This has produced a flora of close-cropped fine grass and bird’s-foot trefoil, and is noticeably warmer than other areas. It is particularly favoured by Skippers, Common Blues, and occasionally Brown Argus and Small Copper.

The standard of recording has altered over the 10-year period 1990-99. From 1990 to 1994 only brief notes of species seen (with dates) were made during a few sporadic lunchtime walks around various parts of the site. No records were made in 1992 as I was working overseas. From 1995 onwards the recording became more systematic, walking a more or less constant route around the site, with visits every one or two weeks as far as possible throughout the flight season, and more careful recording of numbers seen. However, I have not attempted to work to the standards of an official transect – that would be too demanding in the limited lunchtimes available. Instead, a fairly standard route is walked, recording any butterfly within sight that is identifiable, on whichever dates throughout the year are possible. Sightings were entered on the standard Butterfly Conservation "Butterfly Site Recording Form", and sent to the Upper Thames recorder, Jim Asher. Most of the results presented here relate only to the five-year period 1995 to 1999 when systematic records were kept.

A caveat about numbers of individuals needs to be given. From 1995 to mid-1996 large numbers (over about 10) of a species seen flying together were estimated rather than counted, and it appears by comparison with later years that they were often greatly over-estimated. This should be taken into account when interpreting those tables that show numbers of individuals.

This five-year view of a single unremarkable site obviously has less statistical value than the longer, more extensive, regional and national surveys, but it nonetheless gives some feel for the trends locally.