During October 1993, a Geophysical survey was carried out by Peter Cott of the Temple site at Gosbecks Farm. The survey used a Geoscan RM4 Twin Probe array, with manual data logging. The resistivity method employs two remote probes placed at least 15m from the area to be surveyed, and a pair of portable probes mounted on a simple frame, which also carried the electronic circuitry and readout. Using a set of marker ropes, readings were taken at 1m intervals. 42 20m squares were surveyed. The plot of the raw data readings appears in Figure X. This is made up of 42 squares of computer print-out, pasted together to form the overall picture. It can be seen that three surrounding walls, which are well known from many aerial photographs, have been detected with varying degrees of success. This could be due to robbing in antiquity, and it is worth mentioning that Jenkins, who excavated the site in 1842 recorded that the farmer subsequently removed 40 cartloads of stone and rubble. The ‘great ditch’, which also shows up on aerial photographs is probably represented by the wider, more diffuse plots which enclose the centre of the site. These positive readings could be caused by the amount of rubble in the ditch overcoming, as far as the instrument is concerned, the amount of moisture retained. No direct indication of the Roman celle was obtained, even though the small surrounding ditch for this, too, is prominent in the photographs. Hull, who drove a trench diagonally across the site from NW to SE in 1936, also failed to find any strong evidence of it. Outside the walls, to the south, there are strong features which are not seen on the photographs, and there are more strong plots outside the east walls. The resistivity method has detected the main known features of the Temple site, and has shown that there are other features as yet unexplored surrounding the site. <1><2>