^{page 541 note 1} The series of levels and the excavations referred to were carried out with the permission of Colonel J. I. D. Nicholl and Mr. Lewis D. Nicholl, to whom the writer is much indebted. The writer is also under obligation to Mr. Edward Loveluck, A.R.I.B.A., and Mr. Roderick G. Williams, of Bridgend, for much advice and assistance and also to the former for the loan of valuable instruments.Google Scholar

^{page 543 note 1} No adequate map of the area existed, i.e. one contoured to show the sandhills and having the few landmarks which are present among them marked on it, and owing to the difficulty of carrying heavy instruments over long distances of difficult ground, the fixing of sand-hills and sites within them upon the map was carried out with a prismatic compass and a box-sextant, the angles of the latter being used on the spot to check the bearings of the former.Google Scholar

^{page 543 note 2} In order that the angle of dip of the different beds, and their relationship to the surface should be accurately shown the top of one side of the trench was levelled through with a dumpy level, and the beds inserted on a section after the outline of the structure had been plotted.Google Scholar

^{page 543 note 3} All the levelling in connection with this work was carried out by means of a dumpy level.Google Scholar

^{page 545 note 1} This was dated by the late Clement Reid as about 1500 B.C., see Submerged Forests, p. 22.Google Scholar

^{page 545 note 2} A rock lying just over 1½ miles S.W. of the mouth of the River Ogmore.Google Scholar

^{page 547 note 1} i.e. 24th March, 1932.Google Scholar

^{page 547 note 2} Owing to the absence of any bench mark within the area it was impossible to carry through to tie in the final reading. In order to ensure, while crossing the sand-hills, that no loss of height occurred at the frequent changes, on steep slopes with loose surface, large stones were carried from change to change.Google Scholar

^{page 549 note 1} The form of the bed shows that it had not been subjected to erosion since its formation.

^{page 549 note 2} This conclusion has since been corroborated in a very conclusive way. A high tide occurred during 23rd-26th March, 1932, and on the 24th broke through the littoral dune to the seaward of remnant No. 2, and almost completely submerged it.

^{page 549 note 3} This beach could not have been formed after 1578 for a manuscript map of the area bearing this date (The Storie of the Lower Borowes of Merthyrmawr, edited Randall and Rees) is sufficiently accurate near the river to show that the coast-line in its neighbourhood still occupies much the same position.