A common approach to ancient map analysis is by georeferencing, a process of semitransparent matching of the old map onto a modern one. As yet, this technique has not been possible with Roy's map and although an overlay can be employed, inaccuracies on the old map often cause individual landmarks to appear some kilometres away from their actual location.
Intelligent comparison and manual highlighting on to the modern map was the only way forward. It is a labour-intensive process which requires interpretation and even some guesswork. Fortunately it became apparent that the eighteenth-century roads almost always have a modern equivalent and usually it is not at all difficult to determine which one it is. However, even if a Roy-road appears to follow a modern course closely in general direction and relationships one cannot be certain of how close it really was; eighteenth century roads were largely un-metalled and no trace can be expected. Ancient bridges are a useful guide and the record of first-hand evidence from the surveyors helpfully suggests that rivers and lochs as well as roads, were the key features in the original survey work. Therefore, in the process of highlighting roads, we could lean on the relationship to watercourses and coastlines.
It is important to clarify the objectives and qualify the process adopted for this site-:
1. This is not archaeology: it is an attempt to identify which roads/paths/tracks are the modern equivalents of the roads on Roy’s map; these are highlighted as they now exist on the modern OS. The precise on the ground course of the old road cannot be presumed from the highlight, and with that in mind a rather wide highlighting pen has been used (about 50m wide at 1/50,000).
2. Minor estate roads within a very local setting have been omitted if they were not clearly part of the larger communication network.
3. Accuracy of the highlight line can be expected when an identifiable bridge, farm, settlement or contour is transected or adjacent; however, some care was required here: farms and settlements could have been renamed or could have moved.
4. If the highlight appears to veer more than 100m from a modern equivalent road this usually means a specific interpretation of a genuine difference in position.
5. If it is evident that the old road ran where no modern equivalent exists then the best guess was made, coursing between relative points such as towns, bridges, coast contours and farms. In these cases rivers and lochs were used a great deal as points of reference and care was taken to avoid unnecessary elevation changes and obstacles. If there were no comparable features to provide any guidance then other old maps were consulted: Thomson, Stobie, Ainslie and first-edition OS; aerial photographs were also examined; occasionally some fieldwork was required. If none of these offered clarification then simple straight lines were adopted for the highlighted course.
In conclusion this exercise is a best-guess: an attempt to provide a general outline of the road system in 1750 and to identify which modern pathways are the remnants of the old road network.