_On the variation of the species_; _their arrangement and affinities_; _value of the characters used_; _changes during growth_.--Owing to the great variation in external characters, to which almost all the species are subject, and likewise to the genus being a very natural one, that is, to the species following each other in close and natural order, it is not easy to exaggerate the difficulty of identifying the species, except by a deliberate examination of the internal and external structure of each individual specimen. Every one who has collected sessile cirripedes must have perceived to what an extent their shape depends on their position and grouping. The surface of attachment has a great effect on that of the shell; for as the walls are added to at their bases, every portion has at one time been in close contact with the supporting surface; thus I have seen a strongly-ribbed species (_B. porcatus_) and a nearly smooth species (_B. crenatus_) closely resembling each other, and both having a peculiar appearance, owing to their having been attached to a pecten. Dr. Gray has pointed out to me specimens of _B. patellaris_, curiously pitted like the wood to which they had adhered; and numberless other instances might be added. Quite independently of the effect produced by the surface of attachment, the degree to which the longitudinal folds and ribs are developed on the parietes, is variable in most of the species, as in _B. tintinnabulum_, _vestitus_, and even in _B. porcatus_; the presence or entire absence of these ribs often surprisingly alters the whole aspect of the shell. The persistence of the so-called epidermis is in some degree variable; and in _B. lævis_ we have groups of specimens absolutely naked, and others uniformly clothed with a brown membrane. Again, some species in certain localities are all subject to the disintegration of the entire outer lamina of the walls; and in such cases (as with _B. perforatus_) there is not the smallest resemblance between the corroded and perfect specimens. The size of the orifice, and consequently of the operculum, compared with the shell itself, varies accordingly as the shell is more or less conical or cylindrical; in the latter case, the summits of the radii are generally more oblique and the aperture consequently more deeply toothed than in the more conical varieties. Size is a serviceable character in some cases, but very many specimens are required to ascertain the average or maximum size of each species, for there is no method of distinguishing a half-grown from a full-grown specimen; and I believe, as long as the individual lives, so long does it go on moulting and growing. _Colour_ is of very considerable service; though the precise tint varies greatly in almost every species; and what is a far more serious evil, the majority of the species have their white or nearly white varieties, the latter sometimes as numerous as the coloured ones: in _B. perforatus_,