194 
HINTS TO COIX-COLLECTOES 
from the centre of wliicli the upper portion is again divided 
by a cross. In each of the upper divisions is a figure like a 
crescent moon to the right, the lower portion of the shield 
being occupied by the letter E, and the whole surrounded by 
an ornamental border. The earKest coin of this series that I 
have met with was found in the extreme south (Kelikarai) 
and bears date 1702, and as I have in my own collection one 
struck as late as 1801 and several intermediate years, the 
series must have ranged over a century. About 1730 a 
PI IV N 61 form of triply-divided shield appears, 
one of the letters E.I.C. occupying each 
division, above which is a waving Kne, surmounted as usual 
by the 4. On the reverse we still find usually the date, but 
occasionally this gives place to the cross lines which appear 
on the early Mysore and other Hinda coins. This form of 
monogram appears to have remained in fashion till the latter 
part of the last century, when we find a retm-n to the earliest 
form, in which the V. E.I.C. appears as sho-mi in Nos. 62, 
63, and 64. On some of these, as for instance No. 62, the 
date still retains its place on the reverse, 
esf 64,Y5. ^liil® 0^ others the " scales of justice " 
take its place with the Persian word Jj^^ 
(justice) beneath. About 1790 a change also took place in 
the style of the Company's coins. The former rough piece, 
unmilled and shapeless, gave way to the cu-oular European 
form of coin, that of 1791 (No. 63) being about the first 
to boast of an even and milled edge. About the same period 
the Company's coat of arms appeared for the first time, and 
with it came innumerable series of coins, the revei-ses of which 
(and sometimes obverses as well) were occupied by the value 
of the piece in the language of the people of the country in 
which it was intended to pass current, some consequently 
With the limited number of specimens availahle for comparison the 
dates of the commoncement and discontinuance of particular series can, of 
course, only ho approxiiuale, and in some inslauces may nol even be that. 
