ADDRESSES 61 
But about the directions of rotations of the planets! “Aye, 
there’s the rub.” 
At the time this hypothesis was first entertained, it had 
been, for a long time, a standard doctrine that planets formed 
from coherent rings would rotate in the direction of their 
revolution, that is forward, while planets formed from bodies 
moving in independent orbits would have retrograde rotations. 
If this law holds, our planetesimal scheme of growth, well as 
it seems to work in so many particulars, fails seriously here, 
for six of the eight planets have forward rotations. The two 
others probably rotate obliquely backward, at least their sat- 
ellites revolve in this way. These last make trouble for the 
ring hypothesis, to be sure, but still it has greatly the advantage 
over the planetesimal hypothesis if the reasoning back of the 
alleged law is sound and applicable. It runs in this wise: 
In bodies that rotate as a unit, like a ring, the outer part 
moves faster than the inner part, and besides, every por- 
tion of the outer part goes once around the inner parts in 
every rotation of the ring. If therefore the ring is made to 
collect into a spheroid in any normal way, the spheroid should 
inherit a forward rotation. On the other hand, if a ring- 
like belt is made up of small bodies revolving in independent 
orbits, as do the particles that make up the rings of Saturn, 
the inner bodies must move faster than the outer ones, and 
if these bodies are aggregated in a normal way, it was held 
the resulting rotation must be retrograde. Here then, there 
seems to be a lion in the way of all orbital hypotheses, and the 
planetesimal hypothesis is a most declared type of this class. 
But is the reasoning applicable? In the case on which the 
reasoning is based, the orbits are circular. The whole line of 
reasoning, as well as the ring hypothesis itself, seems clearly 
to have had its ‘initial suggestion in the rings of Saturn, and 
very naturally so, as they seem to be remnants of the process 
of evolution providentially left for our instruction. Roche 
and Maxwell, however, showed on theoretical grounds that 
they teach something very different, and Keeler showed by 
the spectroscope that Saturn’s rings are formed of separate 
solid bodies and not of gases as was assumed in the Laplacian 
hypothesis. The orbits of the particles that make up these 
rings are nearly circular and if massed by some systematic pro- 
