A similarly indented celt to that found at the Castle Hill is engraved in the "Archæologia," Vol. xvi., Pl. 54, No. 2, and is described in page 362. It is there stated that a ring of the same metal was attached to it, on which was a bead of jet, and that it was found near Tadcaster, in Yorkshire. The glass bead above mentioned may have been similarly attached to the celt in question. In the work entitled "Old England*" it is stated that the weapons of the ancient Britons show their acquaintance with the casting of metals. Their axe-heads, called celts, are composed of ten parts of copper and one of tin; their spear-heads, of six parts of copper and one of tin. Moulds for spear-heads have been frequently found in Britain and Irelandt."

66

Of late years, much has been written on the uses to which these singular implements were applied. The preferable opinion appears to be that they served as hammers, axes, knives, chisels, gouges, and tomahawks, or missiles, according to their respective shapes and materials.

The late Sir Samuel Rush Meyrick considered celts" to have been of foreign manufacture, brought to this island by stranger merchants, perhaps the Phoenicians, and purposely fashioned by them in imitation of the ruder stone implements used by our British ancestors, in order to secure a market by meeting their wants and tastes."

Of the double-pointed pin, found at the Castle Hill, I gave a woodcut in the first edition of this work (p. 84); since that time I have made numerous inquiries as to its probable use, but without success. Instruments something like it (but with an eye or hole through the head, instead of the knob) are engraved in Illustrations of the Remains of Roman Art in Cirencester, the site of ancient Corinium§." They are described in that work as

Part I., pp. 22, 23.

+ See the "Journal of the Archæological Institute," Vol. iv., p. 1, &c.; and p. 327, &c., as to various forms and moulds of celts.

See "Archæologia," Vol. xxx., p. 493.

By Professor Buckman and C. H. Newmarch, Esq., 1850, p. 105.

nail instruments,-"the divided lower extremities serving to extract dirt from beneath the nails, whilst the hole in the top would allow it to be suspended or tied up, perhaps with other articles of a similar nature*."

The bronze tweezers from the Castle Hill are most probably either Roman or Anglo-Saxon; for one of the Cirencester relics, figured in the above-mentioned work, p. 105, is a pair of tweezers, with what is therein considered a nail instrument attached. The following is the description given :-"A pair of well-formed tweezers, like our modern instrument of the name, has an iron axis through its rounded top, upon which the nail instrument freely moves; this is an exceedingly simple instrument, and, like all of its kind from Corinium, is simply ornamented with engraved lines or circles."

Now it is worthy of remark, that the Castle Hill tweezers have a bronze axis through the rounded top, upon which probably was attached an instrument similar to the one mentioned above, which may have been either a nail instrument or a comb-cleaner, or both.

In the "Journal of the Archæological Institute," Vol. viii., p. 188, a very similar pin to the one in question (with a head of green stone) is figured, and described by Mr. Buckman as having lately been found at Cirencester.

There are some Roman tweezers in the British Museum. It is also observable that in a rather scarce work, published by Thomas Browne, M.D., in 1658, and entitled, "Hydriotaphia, Urn-burial, or a Discourse of the Sepulchrell Urnes lately found in Norfolk," in a field of Old Walsingham, the author, in pp. 14, 23, refers to brazen nippers to pull away hair, as found in the

urns.

In the "Journal of the Archæological Institute," Vol. v., pp. 235, 236, there is a lithographic engraving, and also an account of Roman tweezers of bronze with an ear-pick appended, found at Chesterford.-Amongst numerous Anglo-Saxon sepulchral relics found at Little Wilbraham, Co. Cambridge, exhibited by the Hon.

One with an eye was found at Droitwich with Roman relics, and then thought to have been the acus of a fibula. See under "Droitwich."

R. C. Neville at the Society of Antiquaries, 14th January, 1852, was a pair of bronze tweezers, with an ear-pick attached to it.

Before leaving the Castle Hill, I must notice that the workmen found several genera of recent species of sea shells in the native gravel bed under the hill, which are in Mr. Eaton's possession, namely, Turritella, Murex Erinaceus, Buccinum Macula, and Purpura Lapillus. These correspond with some of the several genera of recent species of marine shells which I procured, through the workmen, from the bottom of the gravel beds at Kempsey, Powick, and Bromwich Hill, near this city, all which latter shells, Sir R. I. Murchison, has noticed in his work on the Silurian System*, in proof that an ancient arm of the sea formerly divided England from Wales. Since that work was published I have obtained Turritella and Cardium from Bromwich Hill; Purpura Lapillus and Turbo Littoreus from Kempsey; and Turritella from the gravel bed at Northwick; which last place lies up the Severn, within about a mile of this city.

66

In the vale between Worcester and Elbury Hill is a farm called Port Fields Farm," in the parish of Claines; the road from it to Worcester (which runs partly in the parish of St. Martin and partly in Claines) is called the "Port Fields Road." This name, as before stated, proves it to have been a Roman port, or military way. The owner of the farm, Mr. John Trevis, informed me that it is called by the above-mentioned name in the earliest of his title deeds, which run back to the time of Henry VIII. About fifty or sixty years ago it got the nick-name of "Skelton's Barn," from the corpse of a person who committed suicide having been found in the barn. The road runs on eastward from this farm, by the Virgin's Tavern," between Leppard and Elbury Hills, and towards Warndon and the Trench Woods.

66

A supposed Roman hypocaust was discovered at the hill, by Sidbury; the particulars of which are as follow:

In January 1843, as the workmen of Mr. Holland, builder, were making an excavation for a building yard, in the marl bank, Vide Vol. i., pp. 532, 533, 534, 554.

+ See p. 9.

just above Sidbury Place, on the south side of the London Road, opposite the Fort Royal, an ancient square underground apartment was discovered. Its walls consisted of bricks and tiles, in alternate courses, set in marly clay; the south-east corner being about seven feet high. There was a double course of tiles between each course of bricks in the walls to the height of about three feet four inches from the floor, and then they ran in single courses of tiles and bricks to the top*. The foot of the high and solid marl bank was excavated in a very square and even manner, to encase the walls of the apartment. These walls, which were eleven inches thick, had been subjected to considerable, although not excessive heat (as they were not vitrified); their interior was quite black, and the marl against them much pulverized by heat. The north and south sides of the apartment were each ten feet wide, and the east and west sides eleven feet. The entrance was at the north side, next the road, and appeared to have been of the whole width of the apartment, except that the wall was nine inches thicker at each side of the entrance, forming square sections of pillars which were two feet four inches broad. The floor was paved with a double course of bricks which were very black. The under course did not reach to the walls by about nine inches on the east, west, and south sides, the intervening part being marl; but the upper course of bricks, covered the whole of the floor, and the flooring came out beyond the entrance about two feet six inches, and terminated at a slight trench. The marl under the floor was also much pulverized by the heat to which it had been subjected. The apartment was only three yards distant from the road, and the floor was upon a level with the road, or nearly so. The covering of the apartment had fallen in, perhaps ages back, and the whole was filled up with bricks, tiles, and earth. In the soil near the western side of the apartment, but unconnected with it, a fragment of a sandstone Gothic moulded shaft was found, and also a piece of blue limestone Gothic tracery; both in the early English style. A Dutch copper coin, with the name Hollandea upon it, was also discovered

* These alternate courses very much resembled those in the Roman pharos at Dover Castle. See "Old England," Vol. i., p. 27.

between the marl and the western wall, to which place it had probably slipped from the upper part of the bank, as the workmen were demolishing that wall, it being evidently of a much later date than the apartment; and foreign copper coins, principally Dutch and German, are frequently dug up in and about this city.

The bricks of the walls and of the floor of the apartment were nine inches long, four inches and a half wide, and two inches thick; and the tiles in the walls were twelve inches long, six inches and a half wide, and about three quarters of an inch thick. There were also some bricks in the walls which were only one inch and a half thick. Upon first seeing the apartment, I was inclined to think that it was an ancient military oven, erected outside the walls of the City, at a short distance (about 220 yards) from Sidbury gate, to supply such troops with bread as might be stationed on the adjoining heights, now called the Fort Royal, &c.; but the late Harvey Eginton, Esq., architect, having examined the apartment with me, suggested that it might have been a Roman hypocaust; and its height, situation, and ancient appearance, strongly favoured that opinion.

I was informed that there was a flue-hole through the top of the wall, at the south-east corner of the apartment, at which a flue might have passed to warm the rooms above; but the top of that corner was broken down by the workmen before I saw it. Whatever other flue-holes there might have been were destroyed when the covering fell in, in days of yore, which brought down all the upper part of the walls, except the corner in question*.

46

The two fragments of stone, in the early English style of architecture, most probably were part of a chapel, which must have formerly stood hereabouts; for Leland, who made his Itinerary" in the reign of Henry VIII., soon after the dissolution of the religious houses, states,"There is a fayre suburb without Sudbury Gate. There is in this suburb a Chappel of St. Godwald. What this St. Godwald was I could not certainly learne. Some sayd he was a bishop."

At the top of the height, above the spot in question, called

*The above particulars I communicated to the Worcester Journals, in January 1843.