Coloured Pearls. 263
produce the colour of certain kinds of fluor-spar and quartz. It is a curious fact that the Unto rectus is
usually white in the Ohio river, while in the more northern waters it
is generally of a fine rich purple or salmon colour. Again, the Unio ligamentinus has
probably never been found, either pink or purple in the Ohio, while at
Grand Rapids, in Michigan, the pink and salmon shells are quite common.
In like manner the Margaritana margaritifera—or common
Pearl-mussel—has usually a fine purple nacre in the Columbia river and
its tributaries, while it is almost universally white in the rivers of Pennsylvania, Connecticut,
and Massachusetts, as is also the case in Northern Europe. Instances
like those seem to prove that the colour is caused by certain foreign
substances, which are present in some streams and not in others.
It
seems probable that the gold theory of I)r. Lewis may explain the
origin of the purple and violet colours of certain Pearls, inasmuch as
it is well known that such tints are readily yielded by salts of gold :
thus the magnificent dye termed "purple of Cassius" is a compound of
the chlorides of gold and tin. According to the American conchologist,
Mr. Tryon, jun., the Pearls found