Lunar orientation in sandhoppers is affected by shifting both the moon phase and the daily clock

The use of a moon compass in the sandhopper, Talitrus saltator, is well known, but the mechanism that allows compensation for the changes in moon position during single nights and from night to night is still poorly understood. In a moon phase shift experiment, we exposed sandhoppers to a dim nocturnal light simulating the moon. The false moon was kept in phase with the true moon for controls and gradually shifted for experimental animals so that, at the time of the test, the false moon calendar was set up to coincide with the first quarter for one experimental group and with the last quarter for the other. During the test under the full moon, the experimental groups behaved as expected on the basis of the moon phase shift, whereas controls showed an intermediate orientation. In a daily clock shift experiment, the sandhoppers were subjected to an artificial light:dark treatment, which was coincident with the natural 24 h light:dark cycle for controls, but was slow or fast shifted by 3 h for the experimental groups. When tested under the full moon, the slow- and fast-shifted experimental groups oriented in a direction that was appropriate to the position of the moon some hours later and before the test, respectively. We conclude that the sun and moon compasses are not ruled by independent rhythms and propose that three interacting rhythms (a circadian, a circalunidian and a circalunar one) are involved in the functioning of the moon compass.