FCHRW/1

TS/1862, 134: `In returning from Brithdir, I paid a visit to the Maen Hir on Gelligaer mountain. This was visited, in 1706, by Edward Lhuyd, who found upon it an inscription'.

Rhys/1875, 370: `On my return through Merthyr Tydvil I went to see the Gelli Gaer stone which was lately figured in the Archaeologia Cambrensis'.

Westwood/1879, 2: `On a mountain near Kevn Gelhi Gaer, not far from Caerphilly on the way to Marchnad y Wayn, stands a maen-hir with an inscription given by E. Lhwyd in Gibson's Camden, p. 616, and in Gough's Camden (ii. p. 498, and ed. alt. iii. p. 127), represented as it appeared in 1693, and as copied in my Plate...At the meeting of the same Association in 1860 it was resolved that the Rev. John Griffiths be requested to take steps for removing the inscribed stone from Gelhi Gaer to a secure place at the cost of the Association. I am not aware that any steps have been taken to carry out this resolution'.

RCAHMW/1976, 36: `first noted in 1693 by Lhuyd'.

Westwood/1879, 3: `It is about 9 feet high, and rather inclines to the north; it is about 14 inches square at the bottom, narrowing upward to 9 inches square at the top'.

Macalister/1949, 152: ` The stone stands in an oblique position'.

Nash-Williams/1950, 132: `Rough pillar-stone'.

RCAHMW/1976, 36: `A roughly squared pillar-stone...The stone, of the local Pennant series and possibly reduced in length, is 2.4m long, 25cm wide on the inscribed face and 40cm thick at the base, leaning to the N. at about 45[degrees] from vertical'.

Macalister/1949, 152: `It has fractured along two cleavage-planes, so that when viewed from east or west its profile suggests a staircase of three steps'.

Macalister/1949, 153: `The inscription...[is] low down on the northern face'.

Nash-Williams/1950, 132: `Inscription...in one line (Fig. 143) reading vertically downwards'.

Nash-Williams/1950, 132, treats the inscription as a name.

The stone has been damaged since the time this reading was to be seen on it; in any case, the drawing is perhaps not very exact of the inscription at any time, for the first letter, according to Lhwyd, was a character which he read t. At present the stone shows ihi; the letter before is gone, excepting the lower part, which may be that of O; the letter before the O still shows traces of being r; this last is preceded by a character which looks a perfect [bookhand-s], and not a part of F. The horizontal bottom of the E still remains; of the first letter there is a part of a curve left which agrees better with Lhwyd's facsimile than with the drawing in the Archaeologia Cambrensis for last April'.

Westwood/1879, 3--5: `The inscription is written longitudinally on the northern face, the tops of the letters being near the north-west angle of the stone, where it is much chipped and several of the letters destroyed, this, as I was informed on the spot, having been done by a party of drunken miners from Dowlais out of sheer mischief. The present appearance of the inscription is given in figure 1 of the accompanying Plate, and is described by Mr. J. Rhys in Arch. Camb. 1875, p. 370. It was read for me, (as given by Gough,) by the late Taleisin Williams at Merthyr Tydvil in September, 1846, `Deffro ini (may we awake)' -- his father having made several rubbings of it.

In the letters of E. Lhwyd, preserved in the Bodleian Library, Oxford, is one dated October 10, 1693, in which this inscription is copied, and it is further stated that...the inscription is Teffro i ti or Deffroi ti (mayst thou awake).' Arch. Camb. 1848, p. 310.

In E. Lhwyd's Archaeologia Britannica, p. 227, a different opinion regarding the inscription was thus expressed: `On a mountain called Mynydh Gelhi Gaer, in Glamorganshire, we find the British name DYVROD inscribed on a stone TEFRAU TI. In the notes on Glamorganshire, in Camden, I have read this inscription, supposing it might have been Welsh, Deffro it ti (mayst thou awake); but having found afterwards that the names anciently inscribed on monuments in our country are very often in the genitive case, as Conbelini, Severini, Aimilini, etc., and most, if not all, Latin, I now conclude it to be a proper name, and the very same that is otherwise called Dubricius.'

In a paper read by the late Mr. Stephens at the meeting of the Cambrian Archaeological Association in 1853, it was stated that the writer preferred the reading of the inscription `VIA FRONTI,' or `Via Frontini,' i. e. the way of Frontinus, supposing it to refer to Julius Frontinus, the Roman general. This reading had been first proposed in 1824, by Dr. J. Jones, in his History of Wales (pp. 17, 327), in which the maen-hir itself was regarded as a miliarium...In Mr. Lukis's Memoir on the Cromlechs, &c.,...this maen-hir is mentioned...It is remarkable in having an inscription in Welsh engraved on its eastern face, which our guide told us reads `DEFROIHI', and means `Awake unto thee'.

The figures which Mr. Lukis has given of this maen-hir, seen from two different points of view, are reproduced at the head of this description (the inscription itself being given as it formerly appeared, although not so stated), whilst the relative position of the maenhir and several adjacent mounds and kists described by Mr. Lukis is shewn in the following woodcut.

According to the late Mr. Thomas Stephens there were no letters visible on this stone when he visited it: `The upper part of the stone has been slit and the inscribed part either destroyed or taken away. This act of vandalism is attributed by the farmers of the locality to a stonemason, named Shon Morgan, who went in a fit of drunkenness to `try the quality of the stone', Arch. Camb., 1862, p. 134'.

Lastly, in Huebner's Inscript. Brit. Christ., p. 22, the ancient representation of the inscription is repeated with the note -- `Quae supersunt vestigia Johanni Rhys visa sunt talia efficere: `tesroihi''; whilst in the text the inscription is read `Sfroihi?''.

Macalister/1949, 152--153: `It suffered severely from the attentions of a drunken miner, who wantonly chipped the letters, some time before 1853. The beginning has thus been lost, and we are dependent for this part of the inscription on the rough woodcut published in 1695: this, however, is close enough to the surviving portion to give us confidence in accepting it. What is left is ....FROIHI; only the last three of these letters are complete, but of the first three there can be no doubt, notwithstanding their mutilation. Nor need we doubt that the letter preceding the F was an E: Rhys (1875) traced `the horizontal bottom' of this letter, but I could not find it. The real difficulty lies in the first letter: Lhuyd first interpreted it as D (1695), afterwards as T. But if it was at all like the character in the woodcut, it must have been an N, of a form similar to the N's in the Eliseg Pillar and the Cenlisini stone at Llanddewi Brefi'.

Nash-Williams/1950, 132: `Round half-uncials...The reading is based on a rubbing made in 1898'.

RCAHMW/1976, 36: `When first noted in 1693 by Lhuyd the inscription near the base of the N. face consisted of seven letters down the face but was severely mutilated before 1862. Only the final three letters survive undamaged, reading IHI with half-uncial H. The complete word (presumably a personal name) as sketched by Lhuyd would seem to have been REFSOIHI, with half-uncial R and S; the lower curve of the O survies, possibly to be read as D. Alternative readings have been TEF(? or S) ROIHI (E.C.M.W. `based on a rubbing made in 1898') and NEFROIHI (C.I.I.C.)'.

Nash-Williams/1950, 132: `partly defaced...The inscription is now mostly obliterated'.