38 
Tin never, or at all events very rarely, occurs in a native state ; 
in the year 1765 a piece was supposed to be found, but the truth 
of this has been denied; pieces taken from roasting-piaces, where 
the tin has exuded, have been sold as specimens of native tin. It 
is a singular circumstance, that personal ornaments of tin have not 
been discovered. That this metal has been used in the manufac¬ 
ture of objects without alloy is certain, for four ancient paterae of 
tin have been found in Cornwall. The ancients knew that tin im¬ 
parted a hardness to copper, but they also knew of its value in the 
arts. An analysis of some enamels from Nineveh shows that the 
yellow colour, which is an antimoniate of lead, contains tin, whilst 
the white is a glaze of oxide of tin, an invention attributed to the 
Arabs of Northern Africa, in the 8th or 9th century. The ancient 
name of tin, KcurcriTepos , has been supposed by some to be of Celtic 
origin. 
1 to 5. Bronze celts of Class No. 1. No. 2 found in Cheshire, the 
others probably Irish specimens, except No. 5, which is be¬ 
lieved to have been discovered in this neighbourhood; it is 
more highly finished than the other specimens, and deviates 
slightly from the normal type. 
6. Bronze celt of Class No. 3. 
7. Bronze celt of Class No. 4, believed to have been found in this 
neighbourhood—this is of a very elegant form, and exhibits 
great finish after casting. 
8 to 18. Bronze celt of Class No. 5 (palstave), No. 8 obtained in 
Paris; Nos. 12 to 17 discovered in making the railway near 
Bognor, Sussex ; No. 16 has never been finished, it is precisely 
in the state in which it came from the mould, and shows all 
the casting marks; No. 14, on the other hand, exhibits signs 
of having been hammered. 
19 to 22. Bronze celts with loop, Class No. 5. No. 21 shows 
hammering marks, and is a fine specimen; No. 22 was found 
at Charlton, near Downton, in 1863 ; the loop was filed off by 
the finder to ascertain whether the implement was made of 
gold; it was subsequently sold in Salisbury for old metal. 
23 to 30. Socketed bronze celts of Class No. 8. No. 23 found 
near Bath ; No. 24 found near Snowdon, June, 1842. This 
specimen is broken at the socketed end, possibly from the 
effect of a blow when in use, for the fractured edges are as 
much patinated as the other portions of the implement. No. 
26 was obtained in Italy. 
31. Cast of a highly-finished granite celt-mould, in the collection 
of the Bev. E. Duke, Lake House, found near Nine Mile 
Water, in the parish of Bulford, almost opposite the tenth 
milestone from Salisbury to Marlborough, but on the opposite 
