Woodlice are unique among the arthropods in moulting in two halves. The intramoult is 1.8 days inOniscus asellus and results from the two halves being out of synchrony throughout the cycle. In the integument, the initiation of a new moult cycle is heralded by epidermal cell vacuolation; a little later, the subepidermal tissue proliferates and, for a brief time, macrophages appear. The cuticle layers are produced in order, starting with epicuticular structures such as tricorns and plaques, followed by the biphasic epicuticle, the lamellate pre-ecdysial cuticle and, after ecdysis, the lamellate postecdysial cuticle. Epicuticle is formed in a continuous sheet along the distal epidermal membrane and postecdysial cuticle from fibres formed within the cell body — both features not reported for other arthropods. Immediately prior to ecdysis the epicuticle and pre-ecdysial cuticle are highly corrugated and the epidermis very constricted as the width of the ecdysing gap is increased by the new cuticle components withdrawing slightly beneath the old cuticle. At this time the space between the two cuticles appears devoid of the fluid that once occupied it, although the ecdysial membrane is still apparent. After ecdysis the appearance of the epicuticle changes due to chemical events which make it relatively impermeable. The pre-ecdysial cuticle also changes in appearance at this time due to the physical stress of expansion. Large cuticle precursor vesicles, similar to those of other Crustacea, are particularly associated with postecdysial cuticle production. In addition there are numerous small vesicles associated with pre-ecdysial cuticle formation. The epidermal vacuoles disappear as postecdysial cuticle is produced so that, when it is complete, the epidermis is again a narrow cell layer. Apolysis occurs straight after this and shortly afterwards the cycle recommences. Unlike other arthropods woodlice do not expand to break free of the old cuticle instead they walk free of either half and then expand. Experimental evidence is given which suggests that woodlice do not take in water to achieve this expansion but rather, they appear to employ internal hydrostatic pressure. Using published data from woodlice endocrinological studies and the present work a possible method of hormonal control of the biphasic moult is suggested.