Collections Online

The base has two features of interest apart from its great size. The first concerns the upper torus, which lacks the customary fillet above the roll. There is no indication that the base has been affected by wear or erosion, the profile being uniform on all the specimens, few as they are, and the stone is sound. At the Flavian date proposed from the colonnade an upper fillet would certainly have been expected. It appears on the bases of the portico of the Temple of Sulis Minerva at Bath (B.W.Cunliffe, Roman Bath [1969], Pls.1; 74, 7.8 and 7.9 for other bases from the temple) and is observed in the probably Neronian bases of the primary portico of the public baths of Silchester (Arch., LIX [1905], 342, Fig.3a) which are among the earliest dateable Bath stone columns in the country. The upper torus of a base from the forum-basilica of Caerwent, however, is of precisely the same form and presumably of the same period as the Caerleon specimen (Arch., LXI [1909], Pl.92, Fig.E). The second feature is the circumferential groove around the lower torus. This seems to be an early feature, being found on the of the Bath temple bases before cited. Another example, from Silchester, is illustrated (Arch., LIII (1893), Pl.24, Fig.1) and doubtless others are to be found: on the whole, attention has been paid far more to capitals than to bases, especially abroad, and further research on bases is desirable. The mode of turning such masses of stone is obscure, though Pliny (Hist. Nat., XXXVI, 13, 19) mentions it in connexion with the columns of a labyrinth on Lemnos, which seems to have been worthy of remark because the spindles of the lathe were so balanced that the columns were turned merely by the agency of a boy driving them round:

'columnis . . .quarum in officina turbines ita librati pependerunt, ut puero circumagente tornarentur'.

Presumably the method was otherwise general. It is interesting to note the use of the word 'hung' (pependerunt) which may mean that the columns were turned vertically rather than horizontally, as is the general rule. A square chuck-hole (7x7x10cm) esists in a basal fragment of one of the Caerleon columns. The lathe could have had iron-shod bearings, rather like those of a pivoted door, or axle.