19. On the Structure of Larvæ of Hispine Beetles.—II.

S ummary (1) In some species of the genus Cephaloleia the thoracic spiracles are ventral, i. e. , they open on the ventral surface. In larvæ of a similarly flattened type such as, for example, Chelobasis and Arescus , the thoracie spiracles open dorsally. (2) The legs show a complete gradation from a well‐developed condition to one of great degeneracy, where the claws have disappeared altogether. (3) In the pupæ there are usually five pairs of spiracles on the first five abdominal segments, but there may be a sixth or seventh pair occurring on the sixth or seventh segments respectively. (4) It is possible that two or more different species of the same genus live together in the same mine or under the same sheltered condition, and this may compel allied but distinct species to interbreed, producing a great number of varieties. Probably this hypothesis accounts for the existence of a large number of colour‐patterns in the species of Arescini. (5) Confined within a small space the larvæ may react, under the stress of a parasite, in directions which may not be in accordance with the behaviour expected in the cases where they are capable of freer movements. (6) In the same species structural variation may occur in the immature stages. (7) The form of the full‐fed larva is the most constant character, and therefore important for the determination of relationship. The form in the earlier stages may be somewhat different from that of the full‐fed. (8) The value of a character depends on the measure of its constancy: the more constant a character is the greater is its value showing relationship. Persistency of an organ or a character is of interest to the systematist. The selection of characters by the systematist has, therefore, an experiential basis, and is not merely arbitrary. If in a certain sense some arbitrariness can be ascribed to the systematist's selection of characters, it is an arbitrariness that is inevitable and common to all human activities.