Relative vertical movement of land and sea has played a great part in geo? logical evolution. The process is however so slow that its detection by measurement rather than by inference is difficult. Since there are places where such movements have demonstrably gone on during the present and recent centuries and since those places are not very distant from our own shores, it is natural to inquire whether Great Britain shares in the movements. The known elevation of northern Scandinavia at a mean rate of about i metre per century and the depression of the Netherlands coast at rather less than a quarter of that mean rate might have been accompanied by elevation in the extreme north of Scotland and depression in the neighbourhood of the English Channel. The available archaeological evidence is very interesting and significant in this connection. It is quite clear that the land surface in the Thames valley has sunk about 15 feet with reference to tidal levels since the Roman occupation.1 There is to-day a densely populated area of some square miles in south London which is from 5 to 6 feet below ordinary high-water mark. Further, on both sides of the western English Channel there is archaeological evidence of subsidence which may belong to the historical period. The evidence of such things as buried forests is also conclusive, but scarcely concerns the present discussion because it is related to too remote an epoch. An aggregate movement of subsidence amounting to 40 feet in the last ten thousand years would be quite consistent with a contrary movement at the present day. The only method of detecting movement in progress is to keep a careful record of sea-level by means of one or more autographic and continuously acting tide gauges. This was one of the purposes for which the Ordnance Survey erected three tide gauges at Dunbar, Newlyn, and Felixstowe in the years 1913, T9T5> and 1917 respectively. Another purpose was the provision of a datum level surface for the Second Geodetic Levelling of England and Wales, 1912-21.2 The level of the sea at each tidal station is determined in relation to a so-called contact point on the framework carrying the gauge, and the height of the contact point is controlled by frequent levellings to the observatory bench mark, perhaps half a mile away. The latter itself is con? trolled by an annual relevelling to the nearest fundamental bench mark, which may be some miles away. These very necessary precautions ensure the