d must have stood entirely above this straight
horizontal line, which line itself reaches above the middle height of
the upright bar forming the I. Immediately above the horizontal line,
for a space about an inch or more in depth, and some ten or twelve
inches in length, there has been a weathering and chipping off of a
splinter of the surface of the stone, as indicated by its commencement
in an abrupt, curved, rugged edge above. This lesion or fracture of the
stone has, I believe, originally given rise to the idea of the semblance
of this terminal letter of the inscription to an R. Probably, also, this
disintegration is comparatively recent; for in the last century Lhwyd,
Sibbald, Maitland, and Pennant, all unhesitatingly lay down the terminal
letter as an I. But even if it were an A or an R, and not an I and
hyphen point, this would not affect or alter the view which I will take
in the sequel, that the last word in the inscription is a Latinised form
of the surname VICTA or WECTA; as, amid the numberless modifications to
which the orthography of ancient names is subjected by our early
chroniclers, the historic name in question is spelled by Ethelwerd with
a terminal R,--in one place as UUITHAR, and in another as WITHER.[142]
Altogether, however, I feel assured that the more accurately we examine
the inscription as still left, and the more we take into consideration
the well-known caution and accuracy of Edward Lhwyd as an archaeologist,
the more do we feel assured that his reading of the Cat-stane legend,
when he visited and copied it upwards of a hundred and sixty years ago
is strictly correct, viz.--

IN OC TV
MVLO JACIT
VETTA F.
VICTI.

_Palaeographic Peculiarities._

The palaeographic characters of the inscription scarcely require any
comment. As in most other Roman and Romano-British inscriptions, the
words run into each other without any intervening space to mark their
separation. The letters all consist of debased Roman capitals. They
generally vary from two and a half to three inches in length; but the O
in the first line is only one and a half inch deep. The O in TVMVLO in
these ancient inscriptions is often, as in the Cat-stane, smaller than
the other letters. M. Edmond Le Blant gives numerous marked instances of
this peculiarity of the small O in the same words, "IN HoC TVMVLo," in
his work on the early Romano-Gaulish inscriptions of France.[143] Most
of the letters in the Cat-stane inscr