678 PHILOSOPHICAL TRANSACTIONS. [aNNO 1683. 



had our Arabic learning, especially the astronomical. And it must needs be as 

 ancient as any Latin translations are out of Arabic, of astronomical tables and 

 other astronomical treatises. Which could neither be well understood in 

 Arabic, nor translated into Latin, without the use of such figures; which 

 occur frequently in those authors. But I do not remember, that I have 

 any where seen any monument of them more ancient than the mantle-tree 

 here described. 



The sides of the chimney, by which the mantle-tree is supported, are of 

 stone: but the mantle-tree itself is of wood, oak, or some other hard wood, 

 which by being kept always dry and smoked, is become as durable. And it may 

 yet (to all appearance) continue for some hundred years more. It is all over 

 as black as ink, by age and smoke. The length of it, AB, (fig. 10, pi. 22,) 

 is 5 feet 9 inches: its breadth or depth, at the ends, as AC, BD, is one foot, 

 or rather 11 4- inches; but at the middle, as EF, somewhat less; being some- 

 what hollowed, archwise. It is carved, from end to end. The lower part of 

 it is abated, as in the mouldings of other chimnies. On the front of the upper 

 part, is in the one half, the sculpture of a dragon's head, and wings. In the 

 other half, there is here only shown in the fig. on three squares parted from 

 each other by a deep furrow or channel, the date as it is here expressed; 

 and, on a fourth, a flower; on a fifth, the two letters W. R. with an escutcheon, 

 representing I suppose the name of him to whom it then belonged. And then, 

 in two lesser squares as the space would permit, one over the other, flowers, as 

 before. The letters and figures, on their several squares, are not engraved or 

 cut in, but prominent, by way of bas-relief, the wood being abated round about 

 them. The o, over the A, is a round o; but that over the M, is a square O; 

 and part of this has been lately pared off" with a knife, by somebody it 

 seems who had a mind to see of what colour the wood is underneath; and it 

 appears there not so black as the rest, but fuscous of a dusky smoke colour. 

 And this, as I remember, is all the defacing that appears in the whole 

 mantle-tree. 



Hence it appears, that the use of such figures here in England, not only in 

 astronomical tables, and other like pieces of learning, but even on ordinary 

 occasions, is at least as old as the year 1133 ; which was the 33d year of king 

 Henry the first. And I judge it to have been yet somewhat older, because 

 the shape of the figures, though not come just to the shape which we now 

 use, was even then considerably varied from the shape of the Arabic figures; 

 which argues they had then been for some time in use; such change of shape 

 in figures and letters coming on gradually with time. 



The foot of the figure 3 being turned backward, makes it more resemble the 



