Site Report

WoSASPIN 11408
NX19NE 24 191 991 to 188 990.
For Girvan Mains, West camp (NX 1870 9800), see NX19NE 47.

NX 191 991. The cropmarks of a probable legionary-size Roman temporary camp discovered from aerial photographs in 1976 and 1977 were proved by trenching in October 1977.
Information from J K St Joseph letter, 12 October 1977.

The site occupies mostly level ground divided amongst several fields on Girvan Mains farm, on the N side of the Girvan Water, a length of some 61m of the N arm, the NE angle and a disjointed 350m length of the E arm are known, indicating a total area of not less than 10 hectares (25 acres) and probably as much as 14 to 16 hectares (35 to 40 acres). The two trenches cut in 1977 both showed a V-shaped ditch about 3.05m wide and 1.25 m deep which had gradually silted. (Descriptions indicate the trenches were in areas NX 1923 9932 and NX 1929 9898).
J K St Joseph 1978

RCAHMS aerial photographs (1978) show only an 80m length of the E arm of the temporary camp (NX 193 990). Also shown is an indistinct and formless series of intersecting linear cropmarks in area NX 189 990. These largely occur within the projected area of the camp and could possibly include part(s) of its W arm; at NX 188 990, outside the projected camp area, is the clear linear cropmark of part of another probable Roman work with a rounded E angle and about 60m lengths of the N and E arms showing.
Further excavation by Dr St Joseph in 1980 has apparently proved the NW angle of the temporary camp and its area is now estimated at about 16 hectares (40 acres). No action was taken regarding the linear marks in areas NX 189 990 and NX 188 990 and their significance remains uncertain. The RCAHMS does not as yet hold copies of the CUCAP APs.
Information from G Maxwell (RCAHMS) 4 December 1980

There are no surface indications of the camp or other linear features across arable fields. Dr St Joseph's recent excavations, comprising three slit trenches in area NX 1889 9927, are still detectable in a field of stubble. This being the NW angle of the camp, the N arm now has a complete length of some 360m.
Mr Young (Girvan Mains Farm) the landowner, has a St Joseph AP (U76/ 1976) which shows the NE angle of the camp.
The probable Roman work appearing on APs in area NX 188 990 is significantly placed on a plateau that could accommodate an annexe or separate temporary camp.
Visited by OS (JRL) 8 December 1980

The cropmarks of two Roman temporary camps, as described.
RCAHMS 1983, visited 1982

Two Roman camps lie half a mile (0.8 km) north of the estuary of the Water of Girvan on gravel subsoil in level fields of Girvan Mains Farm. Of the large camp (centre at NX 192 992) much of the N side, about a third of the N and S sides and the rounded NE angle have been recorded on photographs and confirmed by excavation. The NW angle has also been identified by digging. The V-shaped ditch was 3.05m wide and 1.2m deep in maximum dimensions. The central points in the short N and S sides lie respectively beneath farm buildings and a minor road and adjoining cottage garden. Search has failed to find an E entrance, which may indeed have been destroyed by a railway-cutting or by recent roadworks. The N to S axial dimension is about 396m, and the length of the N side is 370m, so the camp will have included about 14.4 ha (36 acres). Of the second camp (at NX 188990), immediately W of the first and on a different alignment, only 137m of the NE side, 213m of the SE side, together with the rounded S angle, has been traced so far. The ditch was V-shaped, not more than 1.37m wide and 1.06m deep, with a squared slot or channel at the bottom in some sections. The topography makes it unlikely that the area will have exceeded 15 acres (6 ha).
The discovery of a fragment of a late first-century glass vessel in the primary ditch-fill of the second camp, combined with the almost square plan of the first, makes it entirely reasonable to assume that these were bases used by the forces of Agricola during the campaigns mentioned by Tacitus (de Vita Agricolae, cap xxiv) as relating to a possible descent upon Ireland.
G S Maxwell and D R Wilson 1987

In April 1993 Scheduled Monument Consent was granted by Historic Scotland to carry out a geophysical survey, by magnetometer, of 1 hectare within the scheduled Roman camps at Girvan, Ayrshire, in advance of the proposed construction of a sewerage pipeline. The work was carried out by Geophysical Surveys of Bradford, under the direction of the author and on behalf of Bullen and Partners, consulting engineers to Strathclyde Regional Council Sewerage Department.
Aerial survey by RCAHMS in 1989 recorded additional linear cropmarkings which substantially altered the interpretation of the cropmark evidence. When plotted by RCAHMS, they appeared to indicate that the W side of this camp lay considerably further to the W than previously thought, the position earlier assumed for the NW angle being probably that of a gate. Moreover, the position of the gate with a traverse that was visible in the W side of the camp suggested that the NW angle might have lain relatively close to the shore, the area enclosed approaching 21 ha.
The survey was carried out using a Fluxgate Gradiometer and recorded two weak linear responses. Both appear to provide evidence for continuations of previously identified elements of the camps into an area where the variable subsoil of the raised beach has not allowed the definition of archaeological features as cropmarks. This new information has profound implications for the interpretation of the sites' layout and helps to confirm that the two would not have functioned contemporaneously. The provision of a beach head at either site would have allowed the camps to fulfill the function of a base for Agricolan combined army and naval operations around the Scottish coast.
The W to E anomaly provides convincing evidence that this camp is formed by a sub-rectangular enclosure, extending almost to the present beach, the W boundary of which was dictated by the presence of the sea cliff, and covering an area of at least 21 ha.
P W Cox 1993.

NX 1870 9930 In October and November 1995, GUARD undertook an archaeological field assessment at Girvan Mains Farm prior to the laying of a new sewage pipe, at the NW corner of a scheduled area protecting one or more Roman temporary camps (NMRS NX 19 NE 24). A trench measuring 75 x 2m was investigated, parallel to the sea wall. No evidence for any significant archaeological remains within the area examined was forthcoming.
Sponsor: Strathclyde Water Services.
K Speller, 1996d

NX 1875 9938 The proposed construction of a treatment works in the vicinity of archaeologically sensitive areas prompted a watching brief on topsoil stripping and trenching for the insertion of a pipeline in the vicinity of a Roman temporary camp, a scheduled ancient monument (NMRS NX 19 NE 24)
A series of ephemeral remains of limited archaeological interest were recorded. This consisted mostly of burnt patches in the subsoil and nothing that could definitely be associated with the Roman temporary camp was encountered. The only finds discovered were two stone tools of unknown date and origin but which typologically could belong to the Late Mesolithic/Early Neolithic period.
Sponsor: Strathclyde Water Services.
D Abernethy, 1996f