54 TRANSACTIONS—PERTHSHIRE SOCIETY OF NATURAL SCIENCE. 
map make this unlikely. In a plan made about 1834 the island is 
given, but, so far as can be judged from the details, it was then in its 
earlier stages and in rather a fragmentary condition. The Bridge was 
finished in 1771, and, though it has doubtless assisted in the for¬ 
mation of the island, the probable cause was the erection of some 
river wall or other on the left bank above the Bridge. In 1889 Mr. 
R. Dow investigated the flora and found that upwards of 13.0 species 
of plants occurred on the island. Some of these were quite recent 
arrivals, as e.g. the Blue Lupin i^Litpinus perennis)^ which, though long 
known to abound on islands farther up the river, reached Perth only 
a very few years ago. At present this island affords material for an 
interesting study. The portion above the Bridge shows two stages 
in the formation of a stanner. Part of it is bare loose shingle, and 
part has a good number of plants, but has not yet quite got to the 
completely clothed condition. The portion immediately below the 
Bridge has advanced a step further, and is rapidly becoming covered 
with plants, although it has not yet attained any great altitude 
above the water. Finally, the main island which lies immediately 
south of the second portion is in course of being destroyed, but will 
probably have its place taken by the advancing second portion, its 
materials in the meantime going to form new shingles and stanners 
further down, and some at least of its plants finding elsewhere a new 
home. 
The origin and nature of the backwaters alluded to above must 
now be considered. A backwater is an old channel through which— 
either from natural or artificial causes—the river has ceased to flow- 
Whatever the cause may be, the eventual result is essentially the same. 
A natural backwater is generally formed in this manner. When a 
stanner is made it is usually nearer one shore than the other, and 
from the cause of its origin, there is a tendency for shingle to accu¬ 
mulate at the upper end of the narrower channel and form a bar, 
which, eventually prevents any but flood water passing down. Con¬ 
sequently a quiet lagoon is created, open to the river at its lower end 
but cut off (unless the river is in spate) at the upper. To this lagoon 
seeds and roots of water and marsh plants are brought by various 
agencies, the river itself amongst others, and find a congenial home. 
To it also the floods bring and deposit fine sediment, the bar pre¬ 
venting the passage of coarser material, which it retains to itself. 
Thus, in course of time, the bed of the lagoon becomes muddy, a 
process which is assisted by the natural annual decay of the plants. 
In the meanwhile the bar has itself become clothed with land plants, 
and has become an essential part of both the mainland and the 
stanner. Then the silting up of the lagoon goes on, the lower and 
formerly open end gets also cut off from the river, the shallower 
