[ 49 ] 
<l him put live coals into an iron veffel (chafing-difh), 
<c and hold it clofe to the wax, when the wall, bv 
<f being heated, begins to fweat ; then let it be made 
“ fmooth : afterwards let him rub it with a * candle 
" and 'f- clean linen rags, in the fame manner as they 
“ do the naked marble ftatues. This the Greeks call 
“ xavo-is. The coat of Carthaginian wax (thus put 
“ on) is fo ftrong, that it neither buffers the moon 
<c by night, nor the fun- beams by day, to deftroy 
“ the colour.” 
Being fatisfied, from this paffage in Vitruvius, 
that the manner of ufing wax in Exp. 8. was 
right, I was now to find if the wax-varnifh, thus 
burnt into the pidture, would bear wafhing : but 
here I was a little difappointed ; for rubbing one 
corner with a wet linen cloth, fome of the colour 
came off ; but wafhing with a loft hair pencil dipped 
in water, and letting it dry without wiping, the co- 
lours flood very well. 
A board painted, as in Exp. 8, was hung in the 
moft fmoaky part of a chimney for a day, and ex- 
* This account of the method of polifhing [painting] walls 
coloured with vermilion, gave me great fatisfadfion, as it proved 
the method I had taken in experiment 8. (which 1 had tried 
before I faw or knew of this paffage in Vitruvius) wa6 right. The 
ufe of the candle, as I apprehend, was to melt the wax on the 
walls, where by accident the brufli had put on too much, or afford 
wax where the brufh had not put on enough, or had left any part 
bare. 
f The rubbing the wall with a linen cloth, while warm, will 
do very well, where there is only one colour to be preferved ; but 
where there are many, as in a landfchape, it will be apt to take off 
fome, or render the colouring rather faint ; which I found by 
wiping the wax off from a painting while it was hot. 
V O L. LI. II 
pofed 
