From David Devereux – 2 July 2020
I’ve just found Gilbert’s piece on Desnes and will look forward to reading that further. My impression of the medieval geography of Kirkcudbright and St Mary’s Isle (Trail) is broadly the same as yours. Based on recent research in Wigtown Bay near Creetown, our local sea level was at least 1m higher 2000 years ago, so that at a very high tide St Mary’s Isle may well have been an island, and Sandside and the area just north and west of Park House inundated. In the town itself there was a tidal creek which ran probably from behind the museum through the parish church grounds / Soaperie Gardens and then through the present Harbour Square. Overall I have the impression of a long promontory from the site of the present town to the tip of St Mary’s Isle, and the attached English Spy’s map of the 1560s (original in the British Museum) seems to support that, although particular details may be inaccurate and drawn schematically – particularly in the layout of the town and the exaggerated northward extent of Manxman’s Lake.