direction as that pursued by the major planets, their orbits are
inclined at a great variety of angles to the general plane of the
planetary system, and some of them are very eccentric -- almost as
much so as the orbits of many of the periodic comets. It has even been
conjectured that the two tiny moons of Mars and the four smaller
satellites of Jupiter may be asteroids gone astray and captured by
those planets. Two of the asteroids are exceedingly remarkable for the
shapes and positions of their orbits; these are Eros, discovered in
1898, and T. G., 1906, found eight years later. The latter has a mean
distance from the sun slightly greater than that of Jupiter, while the
mean distance of Eros is less than that of Mars. The orbit of Eros is
so eccentric that at times it approaches within 15,000,000 miles of
the earth, nearer than any other regular member of the solar system
except the moon, thus affording an unrivaled means of measuring the
solar parallax. But for our present purpose the chief interest of Eros
lies in its extraordinary changes of light.

These changes, although irregular, have been observed and photographed
many times, and there seems to be no doubt of their reality. Their
significance consists in their possible connection with the form of
the little planet, whose diameter is generally estimated at not more
than twenty miles. Von Oppolzer found, in 1901, that Eros lost
three-fourths of its brilliancy once in every two hours and
thirty-eight minutes. Other observers have found slightly different
periods of variability, but none as long as three hours. The most
interesting interpretation that has been offered of this phenomenon is
that it is due to a great irregularity of figure, recalling at once
Olbers' hypothesis. According to some, Eros may be double, the two
bodies composing it revolving around each other at very close
quarters; but a more striking, and it may be said probable, suggestion