he word _island_, they say, is easily deflected.

At the risk of being thought presumptuous, I do not hesitate to say, that
both these alternatives are manifestly erroneous; and, for the following
reason, I propose a third source, which seems to carry conviction with it:
first, from analogy; and secondly, from the usage of the language from
which our English word is undoubtedly derived, the Anglo-Saxon.

First, from analogy. Let us only consider how frequently names are given to
parts of our hills, shores, rivers, &c., from their supposed resemblance to
parts of the human body. Thus, for instance, we have a _head_ land, a
_neck_ of land, a _tongue_ of land, a _nose_ of land (as in Ness, in
Orfordness, Dungeness, and, on the opposite coast, Grinez); also a _mouth_
of a river or harbour, a _brow_ of a hill, _back_ or _chine_ of a hill,
_foot_ of a hill; an _arm_ of the sea, _sinus_ or bosom of the sea. With
these examples, and many more like them, before us, why should we ignore an
_eye_ of land as unlikely to be the original of our word _island_? The
correspondence between the two is exact. How frequently is the term _eye_
applied to any small spot standing by itself, and peering out as it were,
in fact an _insulated_ spot: thus we have the _eye_ of an apple, the _eye_
or centre of a target, the _eye_ of a stream (_i.e._ where the stream
collects into a point--a point well known to salmon fishers), and very many
other instances. What more natural term, then, to apply to a spot of land
standing alone in the midst of an expanse of water than an _eye_ of land?
{50}

In confirmation of this view, let us look to the original language; there
we find the compounds of _eag_, _ea_, _aegh_, the eye, of very frequent
occurrence: all of them showing that this compound _ea-land_ is not only
legitimate, but extremely probable. Thus we find, _eag-aeple_, the pupil of
the eye; _eag-dura_, a window-light, eye-door; _eag ece_, pain in the eye;
_eah-hringas_, the orbits of the eyes. In the last instance, the _g_ is
dropped; and it is certain that _eag_ was pronounced nearly as eye now is.
From all this, is it too much to conclude that _ea-land_ is the same as
_eye-land_? But farther, _Ig_ (A.-S.) sometimes stands by itself for an
island, as also do _Igland_ and _Igoth_, and _Ii_ was the old name of Iona.
Now I cannot find that there ever was the slightest connexion between the
A.-S. _Ig_ and _water_; nor do I believe that such an idea woul