the condition accompanying metamorphosis be that of asphyxia, it is probable that the secretion of the new coat of chitin may figure as an act of excretion of considerable importance. If there be any truth in this suggestion it may prove the means of enabling us to comprehend some points in the development of Insects that have hitherto proved very perplexing. Peyrou has shown[53] that the atmosphere extracted from the bodies of Insects (_Melolontha_) is much less rich in oxygen than the surrounding atmosphere is, and at ordinary temperatures always contains a much larger proportion of carbonic acid: he finds, too, that as in the leaves with which he makes a comparison, the proportion of oxygen augments as the protoplasmic activity diminishes. Were such an observation carried out so as to distinguish between the air in the tracheal system and the gas in other parts of the body the result would be still more interesting. We know very little as to the animal heat produced by insects, but it is clear from various observations[54] that the amount evolved in repose is very small. In different conditions of activity the temperature of the insect may rise to be several degrees above that of the surrounding medium, but there seems to be at present no information as to the physiological mode of its production, and as to the channel by which the products—whether carbonic acid or other matters—may be disposed of. In the order Aptera (Thysanura and Collembola) the tracheal system is highly peculiar. In some Collembola it apparently does not exist, and in this case we may presume with greater certainty that transpiration of gases occurs through the integument: in other members of this Order tracheae are present in a more or less imperfect state of development, but the tracheae of different segments do not communicate with one another, {132}thus forming a remarkable contrast to the amalgamated tracheal system of the other Orders of Insects, where, even when the tracheal system is much reduced in extent (as in Coccidae), it is nevertheless completely unified. _Gryllotalpa_ is, however, said by Dohrn[55] to be exceptional in this respect; the tracheae connected with each spiracle remaining unconnected.