constitution of the great planet and of his satellites which prevent us from affirming that they are at all times equally capable of reflecting light. Some portions are whiter than others, and therefore return a larger fraction of the sunlight which falls upon them. Still, however, we may for the purpose of the present argument remember that as both Jupiter and the moons are illuminated by the same sun, they are both sufficiently nearly of the same brightness. It therefore follows that there would be no gain of lustre to the satellite in being near Jupiter. Note, then, the difference between what would have happened if Jupiter were sun-like, and what does actually happen when Jupiter is merely a planet. In the former case there would be a distinct accession of brightness to the satellite the closer it made its approach. In the latter case there would be no variation of brightness at all.
To follow the argument a step further we must think of what takes place in the eye of the astronomer who is observing the planet. On the retina an image is formed oi the great globe. The image is extremely minute, but owing chiefly to the disturbances of our atmosphere, the genuine image is surrounded with a region in which the nerves of the retina are more or less affected by the light, which ought, if all the sources of disturbance could be excluded, to be entirely concentrated within the image itself. Of course, vision of other objects on this affected part of the retina will be correspondingly impaired. If the image of a satellite fall upon it, then whether it will be perceived or not depends upon whether the brilliance of the little object is sufficient to excite those nerves whose sensibility is somewhat lessened by the stray light referred to As the satellite acquires no increased bril-
