Mr. Kendal s account of the Dighton rock. 185 
that, which in the original is undefined ; if I have rendered incapable 
of being seen in more than one form that, which on the rock may he 
seen in twenty, my excuse must be tound in the extreme difficulty of 
the attempt to make a representation of any object, without giving it 
a shape. 
The figures you vill find it coloured. In this also I have 
followed my original. The figures, though sunk, are almost every 
where of a light but corresponding tint, of the colour of that part of 
the face, upon which they are found ; and, as the surface of the face 
is variously coloured, so are the figures. The. figures 1 near the sum- 
mit, where the rock is REPS are of a ferruginous yellow. ‘This 
lighter tint of the figures. pro juence, to which I have sub- 
mitted. The. figures a are core age creus, and should there- 
fore be shaded on that side, which is next. the light ; but being ren- 
dered by their colour lighter, instead of darker than the surface, below 
which they are, they are apt, when. shaded, to appear embossed, not 
hollowed. 1 observed this delusion, on ibe rock itself, where the touch 
removes its but on the canvass I have omitted the shadows. ee 
‘My sketch contains many dnacturacics.s 3 some in. the ‘colouring ; ; 
some in the outlines and fissures of the rock ;. and some in the fig- 
ures themselves; but notwithstanding, I believe it free from impor- 
tant errors, such as might frustrate my design of conveying a faithful 
idea of the contents, stvle, execution, and condition of the inscription. 
_If Mr. Winthrop’s were the only other copy, I might feel some 
hesitation at the variations of mine ; but the truth is, that all the cop- 
ies differ, ; in extraordinary particulars, from each other and from the 
original. Dr. Cotton Mather, who published, as I am informed, a 
wood-cut of what he called two dines of the inscription, and which two 
ines are to be found in the fifth volume of Jones’ Abridgement of the 
Philosophical ‘Transactions 4: Dr Cotton, Mather substitutes for fig: 
23 
