104 FEDERAL REPORTER. �saw such an article in his stock, and that if sueh a faucet had been made it would have been in demand. No sales of such a faucet can be recognized upon Bartholomew's book. �The Jenkins patent bas been in existence sinco 1865. The faucet has gone into extensive use, especially in places •where an article was required which could endure constant hard wear. Clear proof is required that an old patent upon an article used in every-day life, and which has long been in demand by the public, was anticipated by an article made in the city of New York before 1856 ; the knowledge of such anticipation having been ascertained about 1879. If the patent had been anticipated by the Bartholomew faucet, it seems palpable that the manufacturera of such articles would have taken advantage of the fact. �The evidence shows that Bartholomew had the idea of a faucet, the valve being drawn from its seat by a key which rode up the inclines of a V shaped cap; that he made such faucets, and sold a few; that they were not a success, and he did not continue the manufacture. He made, as testified by Waldron, three kinds. A duplicate of one of these kinds is in the case. The inclined surfaces of this kind are almost flat on top, and it is not necessarily a self-closing faucet. The point in this part of the case is whether he made the faucet, with a cap having the sharp and deep V, which is shown by Gen. Mor- rison, and which is also testified to by Waldron and others. It is somewhat significant that the old cap from which the Morrison cast- ing was made is not in evidence. That cap was an important^ part of the Bartholomew faucet, and Gen. Morrison admits that "the V in the new cap is deeper than in the old cap." The testimony of all these witnesses is merely from recollection of the shape of a few arti- cles made from 23 to 25 years before they testified, and is not suffi- cient to destroy the presumptions of a patent upon an article which has been long and extensively used. It may be that the present recollection of these witnesses in regard tp a comparatively insignifi- cant part of their work between 1853 and 1856 is accurate; but, in the absence of specimens of the work made at the time, such testi- mony is an unsafe foundation upon which to rest a finding that the patent had been anticpated. �The Bartholomew self-closing hydrant is also relied upon by the defendant. Bartholomew was largely engaged, between 1846 and 1856, in the sale of hydrants which were made under his patent of August 12, 1846, This patent represents the valve as opened by a lever. "Or, instead of this," the specification says, "the valvestem ��� �