^{page 295 note 1} It is now bounded by a road, the old main road, but there are no signs of ancient fields farther to the east.

^{page 295 note 2} Arch. Camb. 1922, 335–45; 1923, 87–113 and 291–302.

^{page 296 note 1} The writer is indebted for help and advice to Mr. C. A. Ralegh Radford, F.S.A., and Dr. Willoughby Gardner, F.S.A., also to Mr. J. A. Wright of H.M. Office of Works, who has been responsible for all the plans and sections.

^{page 296 note 2} Fields north of ‘Presumed line of old stream’ on the plan.

^{page 297 note 1} Mr. Hugh Hughes states that he has been responsible for thirty or more gates on Caerau farm. Formerly many of the fields had no entrance—another feature of the Aran fields at the present date.

^{page 298 ntoe 1} Report forthcoming in Arch. Camb.

^{page 298 note 2} Sussex Arch. Coll. lxiv, 43 ff.; Antiquity, 1927, 272 ff.

^{page 298 ntoe 3} There is also an isolated round structure, as marked, at the south-western corner of the field system. Time did not permit of its excavation, but its small size suggests that it cannot have been inhabited. It may have been used for storage of fodder or fuel, as in the case of the cleits of Hirta, St. Kilda (see A Last Voyage to St. Kilda, Macgregor, A. A. (1933), pp. 214–15)Google Scholar.

^{page 300 note 1} e.g. Rhostryfan, loc. cit.; see also Wheeler, R. E. M., Segontium and the Roman occupation of Wales, 106 ff.Google Scholar, where the evidence is reviewed.

^{page 300 note 2} Arch. Camb. 1934, 1 ffGoogle Scholar.

^{page 300 note 3} There is no evidence here of pre-Roman agriculture, and any which existed must have been of the most primitive type, which utilized an irregular plot, dug with a mattock. Of this type there are examples on Dartmoor, apparently of the Bronze Age, but so far no such remains have been recognized in North Wales; they leave meagre traces, which are easily destroyed.

^{page 300 note 4} Archaeologia, lxxxiii, 237 ff. There are many interesting parallels in detail between Caerau and Chysauster, but most of them must be due to common circumstances in similar surroundings.

^{page 300 note 5} Ibid. 266 and 277.

^{page 300 note 6} Congress of Arch. Soc. 42nd Rep. (for 1934), 24.

^{page 300 note 7} Loc. cit. 281 ff.

^{page 302 note 1} A similarly flimsy arrangement has been found at an analogous site on Breiddin Hill, Montgomeryshire (report forthcoming in Arch. Camb.).

^{page 302 note 2} In plan, but not in date, this house resembles the hut-group at Pant y Saer, Anglesey, (Arch. Camb. 1934, I ff.)Google Scholar.

^{page 303 note 1} Dr. Thomas stated that the dividing line between the slag and the clay is very clear.

^{page 304 note 1} Journ. Brit. Arch. Ass. 1933, 200 ffGoogle Scholar.

^{page 304 note 2} Trans. Lancs. and Cheshire Hist. Soc. 1901, 14 ffGoogle Scholar.

^{page 304 note 3} A Romano-British Industrial Settlement near Tiddington, Stratford-upon-Avon (Birmingham, 1931), 8 ffGoogle Scholar.

^{page 304 note 4} For the identification of these and other charcoal samples the writer is indebted to Dr. Kathleen B. Blackburn of Armstrong College, Newcastle-upon-Tyne.

^{page 306 note 1} Arch. Camb. 1934, 10Google Scholar.

^{page 307 note 1} On the bottom of the trench there were found an iron nail, the piece of window glass, one perforated slate disc (pl. LIV, 2, no. 4), one sherd, and a few fragments of Samian ware. In the filling there were two Samian sherds with some fragments.

^{page 307 note 2} This is now usually 18–24 in. high, but remains to 36 in. in height between Rooms E and A.

^{page 308 note 1} Average thickness 6 in.

^{page 309 note 1} Antiquity, 1934, 395 ffGoogle Scholar.

^{page 309 note 2} The maximum height is actually 32 in. at the south-west corner.

^{page 311 note 1} 6 in. high and at an angle of 45°.

^{page 313 note 1} Archaeologia, lxxxiii, 242, etc., and fig. 1, Room C.

^{page 313 note 2} In this section the lettering of the rooms has been simplified. For instance, I A = House I, Room A. The exact find-spots are recorded in the description of the rooms above. All the finds have been presented by Mr. Hugh Hughes to the National Museum of Wales. For the drawings of the bronze the writer is indebted to Messrs. L. Munro and G. C. Dunning, F.S.A., and of the querns to Mr. W. F. Grimes, F.S.A.

^{page 313 note 3} Arch. Camb. 1923, 93Google Scholar.

^{page 316 note 1} Arch. Camb. 1923, 90–1Google Scholar.

^{page 318 note 1} They have now been treated for preservation through the kindness of Mr. Alexander Keiller, F.S.A., and Mr. Stuart Piggott.

^{page 320 note 1} Arch. Camb. 1923, 107Google Scholar.

^{page 320 note 2} Ibid. 106.