M. Hansky finds that the visibility of the "light" is greater during periods of maximum solar activity--that is, at the maxima of sun spots. This he explains by the theory of Arrhenius, in which electrified "ions emitted by the sun cause the phenomena of terrestrial magnetic storms and auroras." "In the same way the dense atmosphere of Venus is rendered more phosphorescent, and therefore more easily visible by the increased solar activity."[47] This seems a very plausible hypothesis. On the whole the occasional illumination of the night side of Venus by a very brilliant aurora (explanation (4) above) seems to the present writer to be the most probable explanation. Gruithuisen's hypothesis (7) seems utterly improbable. There is a curious apparent anomaly about the motion of Venus in the sky. Although the planet's period of revolution round the sun is 224·7 days, it remains on the same side of the sun, as seen from the earth, for 290 days. The reason of this is that the earth is going at the same time round the sun in the same direction, though at a slower pace; and Venus must continue to appear on the same side of the sun until the excess of her daily motion above that of the earth amounts to 179°, and this at the daily rate of 37' will be about 290 days. Several observations have been recorded of a supposed satellite of Venus. But the existence of such a body has never been verified. In the year 1887, M. Stroobant investigated the various accounts, and came to the conclusion that in several at least of the recorded observations the object seen was certainly a star. Thus, in the observation made by Rœdickœr and Boserup on August 4, 1761, a satellite and star are recorded as having been seen near the planet. M. Stroobant finds that the supposed "satellite" was the star χ_{4} Orionis, and the "star" χ_{3} Orionis. A supposed observation of a satellite made by Horrebow on January 3, 1768, was undoubtedly θ Libræ. M. Stroobant found that the supposed motion of the "satellite" as seen by Horrebow is accurately represented by the motion of Venus itself during the time of observation. In most of the other supposed observations of a satellite a satisfactory identification has also been found. M. Stroobant finds that with a telescope of 6 inches aperture, a star of the 8th or even the 9th magnitude can be well seen when close to Venus.[49]