Field Visit
Date 6 October 2015
Event ID 1005971
Category Recording
Type Field Visit
Permalink http://canmore.org.uk/event/1005971
‘The Roundel’ is a rocky igneous knoll that projects from a point close to the foot of the SE flank of Kirkton Hill 1.3km south of Perth and 2km NW of Bridge of Earn. Survey of the remains of a fort there was undertaken in November 2015 by HES to reconcile the cropmarks of the defences (first recorded in 1975 and partly excavated in 2013) with the fragmentary upstanding remains of banks identified in woodland on the SW flank of the knoll by members of RCAHMS staff at the time of the excavation.
‘Fir Knowe’, as ‘The Roundel’ was then known, is depicted as a roughly oval area of mixed deciduous and coniferous woodland on the first edition of the OS 25-inch map (Perthshire 1866, Sheet XCVIII.13). This map shows an ‘Old Quarry’ on the east side of the knoll and a wall, most likely a plantation boundary, which ran from that quarry, around the northern edge of the woodland to a point on the NW flank where it terminated. These features are still visible though the wall is now ruinous and reduced in places to little more than a stony bank. In the early 1960s the summit of the knoll was chosen as the site of a dwelling and the present owner (Mr Douglas Johnson) informed JRS that at that time there was only a thin cover of soil present and that the foundations of the house had to be excavated into solid rock. Moreover, in order for a garden to be created a huge amount of soil was transported from a development site in Glenearn Road, Perth, over the course of a whole winter. This information helps explain the uppermost scarp which forms the edge of the garden and which gives every indication of being a landscaped feature created by earthmoving machinery. On the NE, the relationship of the slopes suggests that the formation of the platform on which the garage was built may have been a secondary consideration -- though not one that was necessarily separated from the principal earthmoving event by very much time. There are no contemporary reports of any artefacts or structures being found during the course of the building of the house and its garden.
The cropmark evidence indicates that there were originally at least five lines of defence in what is now arable ground immediately to the NE and NW (where there is an entrance) of the knoll. On the wooded NE flank of the knoll itself, where it might have been expected that there would be surviving traces of banks, no visible remains have been identified. However, much of this ground has been disturbed or obscured by the 1960s landscaping, the dumping of field cleared stones or the construction and subsequent collapse of the plantation wall. Original earthworks have almost certainly also been destroyed by quarrying and other episodes of digging.
Two scarps on the SW flank of the hill , the lowermost about 47m in length, appear to represent the continuation of the banks that would have originally accompanied the two outermost cropmark ditches on the NW. The purpose of two other, shorter, scarps further up the slope is less clear, but they, too, may be the remains of original banks.
At the SE end of the knoll there are two scarps, the inner situated below the landscaped edge of the garden and measuring at least 2m in height. Largely obscured by vegetation, including dense brambles, on the date of visit, the north end of this scarp appears to run out from under the landscaped feature and extend south and then west for a distance of at least 35m before disappearing again under it. On the south this scarp is accompanied by another, the east end of which seems to have been destroyed by the construction of the platform on which a brick-and-timber hen-house has been built.
The degree to which the fort on The Roundel has been disturbed since the early 19th century makes it very difficult to assess its original internal size. The dumping of large quantities of soil and the creation of the garden has obscured the natural topography to such an extent that it is not now possible to judge the obvious line which the inner rampart or wall might have followed. And survey of other relatively small multi-vallate forts nearby (for instance Jackschairs Wood; NO01NE 20) has shown that it is not safe to assume that lines of defence need necessarily follow the contours of the knoll on which they are situated. Further, the lack of any visible evidence of defence on a steep slope, for instance the southern flank of The Roundel, cannot be interpreted as evidence that there were no defences here. Given the available evidence, it is likely that the fort was roughly oval on plan and its internal dimensions were about 70m from NW to SE by up to 50m transversely, thus corresponding fairly closely with the size of the fort in Jackchairs Wood.
Visited by HES Survey and Recording (JRS, AM), 6 November 2015.
