We see from this, that, as is usual in like cases, the words have followed lines of their own, and afford but little evidence of the forms of the objects to which they have been applied, excepting in so far that the common word "Klu" or "Clo" for lock and pin, and its connection with the base "Klu," to move, implies that the earliest form consisted of a movable bolt. But, in any case, whether we take the Latin word "Sero", to put, or the Sanskrit "Klu," to move, as independent origins of words for locks, we are carried back to a time when it consisted of a simple bar or bolt put up or slipped through staples to close a door. The passage in the 'Odyssey,' so often quoted in relation to the construction of Greek door locks, does not in reality throw much light upon the subject so long as it is unassisted by archæological discoveries. It has been variously translated,[1] and we are left very much to conjecture for the forms of the most primitive kinds of locks which preceded those of which the relics are to be found in our collections of antiquities. It is noteworthy, however, that the earliest vestiges of apparatus connected with door fastenings in metal, that are discovered, consist of keys, which leads to the inference that the locks themselves may have been made of wood, and have therefore perished. But we have survivals of primitive wooden locks in use at the present time in different countries, which show us, with great probability, the uses to which the keys were put, and it is to these that we must turn in any attempt to trace back the history of the mechanism from the commencement. The process is one, the merits and demerits of which have been too often discussed to need comment here. In the absence of direct archæological evidence we have no alternative but to avail ourselves of survivals as far as possible. The materials, however, in the case of locks are so abundant that it will not be necessary to tax our imagination unduly in order to fill in the links that are found wanting. [1] 'Odyssey,' xxi., 46-50. See translations by POPE, and by BUTCHER and LANG. I put aside all mention of knots and strings which as Mr. SYER CUMING has observed ('Journal of the British Archæological Association,' vol. xii., p. 117) must have formed the fastenings employed by dwellers in tents, and of which the Gordian knot was a complicated example. In early times seals must often have served as substitutes for locks, as we know was frequently the case in ancient Egypt and Assyria. The wooden door must have given rise to a totally different contrivance. It is possible, however, that something analogous to the Japanese book fastening, represented in fig. 1, Plate I., may have been employed under both systems.