254

THE PLANET MARS.

Once more, let us suppose the Moon to describe its orbitrouud the Earth in a single day, and let us also (in order toavoid certain complications) suppose its path to he in theplane of the equator, a position which very nearly correspondsto that of the satellites of Mars . In this case a very curiousresult would follow ; viz., that the Moon would be seen by onehalf of the Earth , but never by the other half; and any placethat would see it would always see it in one fixed positionin the heavens.*

But lastly, let us suppose the Moon to circle once round theEarth in less time than the Earth occupies in one revolutionupon its axis. What would then happen ? Its real motionfrom west to east being greater than the apparent motion fromeast to west imparted to it by the Earth ’s rotation, it wouldappear to go round the sky from nest to east, with a velocityequal to the excess of the one motion over the other, instead•of apparently travelling from east to west, as all the otherheavenly bodies do. Now this is just what happens in thecase of the inner satellite of Mars , and for an exactly similarreason. The rotation of the planet would make it appear to goonce round from east to nest in 24 hours 37| minutes; whileits own motion would take it rouud from west to east in 7 hours39j- minutes. The combined effect is, that in about 11 hours fit seems to go once round the Martial sky from nest to east.

The outer satellite travels round its orbit in about 30 hours

* This may perhaps be best understood by a simple example. Let theMoon , for instance, be supposed to be upon the meridian of any place atany given time ; then it is not difficult to realize that it would (upon theabove supposition) just keep up with the movement of the place producedby the rotation of the Earth , so that it would appear to remain fixedupon the meridian in question. Similarly for other positions in which itmight be continuously seen from other places of observation.

t The former velocity being rather less than |rd of the latter, inas-much as 24 h 37 m is rather more than 3 times 7 h 30 m , it follows that thevelocity corresponding to a rotation in 7 h 39™ will be diminished byrather less than one-third part. We must consequently increase thisperiod by barely one-half (or, in other words, increase it in a ratiorather less than that of 3 to 2) in order to obtain the duration of theapparent circuit of Phobos . This gives, as above stated, about 11hours.