it which is ten times larger than the orbit of the earth.
Suppose now that the sun were suddenly to be extinguished; light takes
about eight minutes to travel from the sun to the earth, and consequently
we should not get news of the extinction for some eight minutes; the sun
would appear to us to still go on shining for eight minutes after he had
really been extinguished. Saturn being about ten times as far away from
the sun, the news would take eighty minutes to reach Saturn; and from the
earth we should see Saturn shining more[3] than eighty minutes after the
sun had been extinguished, although we ourselves should have lost the
sun's light after eight minutes. I think we already begin to see
possibilities of curious anomalies; but they can be made clearer than
this. Instead of imagining an observer on the earth, let us suppose him
removed to a great distance away in the plane of the two orbits; and let
us suppose that the sun is now lighted up again as suddenly as the new
star blazed up in February 1901. Then such an observer would first see
this blaze in the centre; eight minutes afterwards the illumination would
reach the earth, a little speck of light near the sun would be
illuminated, just as we saw a portion of the dark nebula round Nova Persei
illuminated; eighty minutes later another speck, namely, Saturn, would
begin to shine. But now, would Saturn necessarily appear to the distant
observer to be farther away from the sun than the earth was? Looking at
the diagram, we can see that if Saturn were at S{1} then it would present
this natural appearance of being farther away from the sun than the earth;
but it might be at S{2} or S{3}, in which case it would seem to be nearer
the sun, and the illumination would seem to travel inwards towards the
central body instead of outwards. Without considering other cases in
detail, it will be tolerably clear that almost any anomalous appearance
might be explained by choosing a suitable arrangement of the nebulous
matter which we suppose lighted up by the explosion of Nova Persei.
Another objection urged against the theory I have sketched is that the
light reflected from such a nebula would be so feeble that it would not
affect our photographic plates. This depends upon various assumptions
which we have no time to notice here; but I think we may say that there is
certainly room for the acceptance of the theory.

[Illustration: FIG. 6.]

[Sidenote: Did the nebula cause the outb