Joty—On the Origin of the Canals of Mars. 259 
breaking strength of basalt, and 3, that of glass. A fluid satellite would not, of 
of course, be stable. In fact, a gramme mass placed upon the innermost point of 
Phobos’ surface must possess a cohesive force binding it to the surface of Phobos of 
about 6 dynes. It is improbable that local stresses due to tidal action on the 
satellite can imperil its stability, seeing that the total stress falls so far short of the 
probable resistance to rupture of the material of the satellite. Further from the 
surface of Mars, the stresses are, of course, much less. 
Effects of Mars’ Atmosphere on the Satellites. 
It appears certain that there is some atmosphere to Mars, and Mr. Lowell 
adopts the current view that this atmosphere is about one-seventh the density of 
our own. The effects of this as a resisting medium to the motion of a satellite 
need some remark. The accompanying figure (3, p. 258) shows the distribution 
of pressure above the surface of Mars. The pressures are given in millimetres of 
mercury supposed to be under terrestrial gravity. The temperature of Mars’ 
atmosphere is assumed at 0°C. The circles show the disk of Phobos at 
two altitudes above the surface. At the height of 65 miles, it is seen that the 
pressure of the resisting medium is 0°8 m.m. 
The effects of a resisting medium upon the fate of a satellite are well known. 
If the angular velocity of the primary is greater than that of the satellite the 
effect will be repulsive, diminishing at the same time the speed of the satellite. 
If the satellite rotates the faster of the two, the effect is the other way. Tidal 
action goes with and accelerates these effects. It is evident from the curve that the 
satellites’ energy will be rapidly absorbed or increased at close distances. A long 
stay in either case near the surface of the primary is impossible. At a distance 
of 65 miles, and assuming the full pressure of the medium to retard the advance of 
Phobos, that is, assuming a perfect vacuum continually maintained in its rear, his 
spiral path will shrink from 75 to 55 miles after some 88 thousand revolutions, 
the satellite taking fifteen years to effect this approach. This estimate is obtained 
by considering the diminished value of the sum of the potential and kinetic 
energy of Phobos when he has fallen to within 55 miles of the surface, and 
assuming that this loss of energy is due to the retarding effect of the atmosphere 
doing work at a rate estimated from the satellites’ velocity when at the mean 
distance of 65 miles. 
During this period and subsequently the nodes of the satellite will probably 
be slowly shifting. The equatorial bulge of Mars, although small, will enter into 
this question, as well, probably, as even any considerable raised surface features 
of Mars, 
2Q2 
