To M. Chabas belongs the honour of being the first to clear up the relations between the uten and kat. The history of this discovery is an interesting proof of the fruitlessness of the purely empirical form of metrology which confines itself to the measuring of buildings, and weighing of ancient weight-pieces and coins, unless its path is made clear by means of the light derived from ancient records. The names uten and kat had been long known, as both of them recur frequently on the walls of the temple of Karnak (_Temp._ Thothmes III. 1700-1600 B.C.), and Egyptian weights were in the museums of Europe, but nevertheless “the exact relation of the one to the other remained unknown until it was fortunately disclosed by a passage in the Harris papyrus, which contains the annals of Rameses III. (circ. 1300 B.C.). From this it appears that the Uten contained ten Kats[294].” The uten therefore is the tenfold of the kat: Nissen[295] thinks that the latter was perhaps originally a gold weight (_vielleicht ursprünglich ein Goldgewicht_). These two units served for the weighing of gold, silver and copper, and there seems to be no difference noted in the documents between the units used for each purpose. In the lists of booty we read of such sums as 3144 utens of gold and 36692 utens of electrum. In lists of prices of commodities kats and utens of silver and copper are frequently mentioned. The weight of the kat has been fixed by Lepsius at 9·096 grammes (142·1 grains) and that of the uten at 90·959 grammes (1421·2 grains). But as it often happens in the case of coins that one well-preserved specimen is a better index of the normal standard than any that can be attained by taking the average of 100 bad specimens, so in the case of weights, one good specimen, made of some hard and imperishable substance, will give us a truer representation of the standard unit than the average of a large number of weights made of some less durable material, and carelessly executed, and meant merely for traffic in goods of little value. If such a weight as we have supposed is inscribed with its name, and we can also get some indication that it has all the authority that belongs to a weight used for official purposes, its value becomes still greater. Such a piece fortunately exists in the Harris Collection. It is a beautifully preserved serpentine weight, and weighs 698 grs. Troy. Allowing for its extremely slight loss we may suppose its original weight to have been about 700 grs. It bears the inscription, _Five Kats of the Treasury of On_. This gives 140 grains Troy as the weight of the kat[296]. This inscription also proves that the kat was the unit. For if as is commonly stated the uten is the unit, of which the kat is simply the one-tenth, we must naturally expect to find this weight described as ½ uten rather than as 5 kats. This is confirmed by a statement of the grammarian Horapollo (or Horus, who although writing about 400 A.D. nevertheless preserves