412 REPORT— 1903. 



Sileliesier Excavation. — Report of the Gommittee, co7isisiing of Mr. 

 Arthur J. Evans (Chairman), Mr. J. L. Myres (Secretari/), 

 ami Mr. E. W. Brabrook, appointed to co-operate with the 

 Silchester Excavation Fuiul Gommittee in their Excavations. 



The excavations in 1902 were begun on May 15, and continued without 

 break until November 17, under the direction of Mr. Mill Stephenson. 



To suit the convenience of the tenant, operations were for the most 

 pai't confined to a narrow and irregular-shaped field of some four acres in 

 the south-east quarter of the town, adjoining and lying westwards and 

 .southwards of the present churchyard. 



The area in question consists as to its northern half of a gentle slope 

 southwards, but it then rapidly descends, and its southern end is only 

 slightly raised above the level of a small brook which rises close by. 



The eastern margin of the field was explored in 1890, when part of a 

 walled enclosure, containing two square temples, which partly underlie 

 the churchyard, was excavated and planned. A small corridor house to 

 the south of the temple enclosure was discovered in 1896, when the 

 ground there was explored in view of an extension of the churchyard. 



The operations of 1902, for the reasons above cited, did not enable 

 the executive committee to excavate any entire insula, but portions of at 

 least four were included in the area placed at its disposal. 



The first work undertaken was the tracing of the temple boundary 

 wall northwards. This was followed up to and across the modern road 

 crossing the town and into the field beyond. Here it was found to turn 

 at a sharp angle in the direction of the east gate, thus showing that the 

 main street through the town from west to east was defiected from a 

 straight line before passing out eastwards. At the upper corner of the 

 temple area was uncovered a small apsidal building of uncertain use. 



The buildings of which remains were uncovered in the field adjoining 

 the churchyard were eleven in number. They included five more or less 

 complete houses, and as many other structures. Two of the latter are of 

 a character not hitherto found at Silchester : one being a semicircular 

 building entered probably by a wide arch, and forming a kind of alcove ; 

 the other a long narrow gallery with a colonnade or portico along one 

 side. Both buildings may have stood in a pleasure garden belonging to 

 the house found in 1896 south of the temple enclosure. 



Besides the field next the churchyard, the exploration was undertaken 

 of a section of the pasture west of it. Tliis Ijrouglit to light two more 

 houses and two otlier smaller structures, one perhaps a group of three 

 shops. The houses belong to the indu/a (XX VIII) south of XXVII, 

 which was explored in 1901. 



The several houses discovered in 1902 are of interest, firstly, for their 

 comparatively small size, and secondly, on account of the transition from 

 the corridor to the courtyard type which most of them present. Three 

 at least of tliem had winter rooms warmed by hypocausts, but in two 

 in.stances the heating arrangements had been cleared out and the rooms 

 put to other uses. 



The number of buildings in the circumscribed area dealt with left 

 comparatively little space between for garden ground, and consequently 

 the number of pits and Avells and of objects found was unusually small. 



Part of a large inscribed slab, with remains of finely cut letters five 



