on the coasts of France. But shells that have been obtained from the British shores are still extremely rare, and when one is perchance discovered, the fortunate collector is sure to notify his success in some scientific publication. Of its habits little or nothing is known, as is likely to be the case with a burrowing mollusc, unless some enterprising naturalist chooses to devote his energies to the study of some particular species.
At fig. 2 is seen a shell which is notable for the numerous sharp ridges which traverse its surface, and which are cut into teeth like those of a saw. This is the Irus-shell (Venerupis irus), a shell which, on account of its remarkable shape and the ridges of its surface, is very easily identified. It is seldom found in deep water, preferring those parts of the shore which are just below low-water mark, and being often found in spots that are left day at very low tides. The empty shells are not at all unfrequent, especially on the southern coasts of England; but a living specimen is seldom to be found, unless especially sought for. Limestone rocks are favourite localities with the Irus-shell, and it is sometimes discovered in sponges and sea-weeds, like the Bock-borer which has already been described. As may be seen b}r a reference to the illustration, the siphons are united for half their length, and fringed at their tips. One of them is remarkable for the fact that it has a kind of supplementary tube which extends beyond the fringes. In the Mediterranean, the Irus is produced in vast quantities in some places, and the empty shells are flung in heaps upon its shores.
Here I may mention a fact the knowledge of which is useful to all collectors of shells. Generally, when a shell is found on our shores, it is presumed to be a genuine British species, and is classified as such; on the same principle that any plant, insect, reptile, bird, or mammal is considered as indigenous if it be discovered in a living state within the four seas. But this ought not to be the case with shells. A collector might
