so close together that they cannot be seen | the determination of the orbits of other
separately without the aid of a telescope. binary stars of which the components are
We may take, for instance, the very cele- | further apart. Among these we may men-
brated double star Castor, well known as tion the remarkable binary star § Ursæ
one of the Twins. Viewed by the un- Majoris. The distance between the two
aided eye, the two stars look like a single components of this star varies from one
star, but in a moderately good telescope second of arc to three seconds. The first
it is seen that the object is really two sep-recorded measurement of this object was
arate stars quite close together. The
question now comes as to whether the
propinquity of the two stars is apparent
or real. It might be explained by the
supposition that the two stars were indeed
close together compared with the distance
by which they are separated; or it could
be equally explained by supposing that
the two stars, though really far apart, yet
appeared so nearly in the same line of
vision that when projected on the surface
of the heavens they seemed close to-
gether. It cannot be doubted that in the
case of many of the double stars, espe-
cially those in which the components ap-
pear tolerably distant, the propinquity is
only apparent, and arises from the two
stars being near the same line of vision.
But it is, also, undoubtedly true that in
the case of very many of the double stars,
especially among those belonging to the
class which includes Castor, the two stars
are really at about the same distance from
us, and, therefore, as compared with that
distance, they are really close together.

by Sir W. Herschel in 1781, and since that date it has been repeatedly observed. From a comparison of all the measurements which have been inade it appears that the periodic time of the revolution of one of these components about the other is about sixty years. This star has thus been followed through more than one entire revolution. The importance of these discoveries became manifest when an attempt was made to explain the movements. It was soon shown that the movements of the stars were such as could be explained if the two stars attracted each other in conformity with the law of gravitation. It would, however, be hardly correct to assert that the discovery of the binary stars proved that the two stars attracted each other with a force which varies inversely as the square of their distance. Even under the most favorable circumstances the observations are very difficult; they cannot be made with the same accuracy as is attained in observing the movements of the planets; they have Among the splendid achievements of not even the value which antiquity will Sir W. Herschel, one of the greatest was often confer on an observation which has his discovery of the movements of the not much else in its favor. There are binary stars. It was shown by Herschel probably many different suppositions that in some of the double stars one star which would explain all that has yet been of the pair was moving around the other, observed as to the motions of the binary and that their apparent distances were stars. Gravitation is but one of those changing. The discoveries inaugurated suppositions. Gravitation will no doubt by Herschel have been widely extended carry with it the prestige acquired by its by other astronomers. One of the more success in explaining phenomena in the rapidly moving of the double stars lies in solar system. I do not know that any one the constellation of Coma Berenices. The has ever seriously put forward any other revolution of one component around the explanation except gravitation to account other requires a period of 25.7 years. The for the movements of the binary stars, two components of this star are exceed nor is any one likely to do so while gravi. ingly close together, the greatest distance | tation can continue to render an account being about one second of arc. There is of the observed facts; but all this is very very great difficulty in making accurate different from saying that the discovery measurements of a double star of which of the binary stars has proved that the the components are so close. More re-law of gravitation extends to the stellar liance may consequently be placed upon regions.

Except for what the binary stars tell us, | true that in one very minute part of this we should know nothing as to the exist- infinitely small region the law of gravitaence or the non-existence of the law of tion appears to reign supreme. This gravitation beyond the confines of the so- minute part is of course the solar system. lar system. Does Sirius, for instance, There are also a few binary stars in this attract the pole-star? We really do not infinitely small region whose movements know. Nor can we ever expect to know. would admit of being explained by graviIf Sirius and the pole-star do attract each tation, though as yet they can hardly be other, and if the law of their attraction be held to absolutely prove its existence. the same as the law of attraction in the solar system, it will then be easy to show that the effect of this attraction is so minute that it would be entirely outside the range of our instruments even to detect it. Observation is hopeless on such a matter. If we cannot detect any attraction between a star in one constellation and a star in another, no more can we detect any attraction between our sun and the stars. Such attractions may exist, or they may not exist: we have no means of knowing. Should any one assert that there is absolutely no gravitation between two bodies more than a billion miles apart, we know no facts by which he can be contradicted.

If we know so little about the existence of gravitation in the space accessible to our telescopes, what are we to say of those distant regions of space to which our view can never penetrate? Let a vast sphere be described of such mighty dimensions that it embraces not only all the objects visible to the unaided eye, not only all the objects visible in our most powerful telescopes, but even every object that the most fertile imagination can conceive, what relation must this stupendous sphere bear to the whole of space? The mighty sphere can only be an infinitely small part of space. It must bear to the whole of space a ratio infinitely less than the water in a single dewdrop bears to the water in the Atlantic Ocean. Are we then entitled to assert that every particle in the universe attracts every other particle with a force which is proportional to the product of their masses, and which varies inversely as the square of their distance? We have, indeed, but a slender basis of fact on which to rest a proposition so universal. Let us attempt to enunciate the law of gravitation so as to commit ourselves to no assertion not absolutely proved. The statement would then run somewhat as follows:

Of the whole contents of space we know nothing except within that infinitely small region which contains the bodies visible in our telescopes. Nor can we assert that gravitation pervades the entire of even this infinitely small region. It is

It must then be admitted that when the law of gravitation is spoken of as being universal, we are using language infinitely more general than the facts absolutely warrant. At the present moment we only know that gravitation exists to a very small extent in a certain indefinitely small portion of space. Our knowledge would have to be enormously extended before we can assert that gravitation extended entirely through this very limited region; and even when we have proved this, we should only have made an infinitesimal advance to a proof that gravitation is absolutely universal.

I do not for a moment assert that our ordinary statement of the law of gravitation is untrue. I merely say that it has not been proven, and we may also add that it does not seem as if it ever could be proved. Most people who have considered the matter will probably believe that gravitation is universal. Nor is this belief unnatural. If we set aside comets' tails, and perhaps one or two other slightly doubtful matters, we may assert that we always find the law of gravitation to be true whenever we have an opportunity of testing it. These opportunities are very limited, so that we have but very slender supports for the induction that gravitation is universal. But it must be admitted that an hypothesis which has practically borne every test which can be applied has very strong grounds for our acceptance: such, then, are the claims of the law of gravitation to be admitted to a place among the laws of nature.

The wondrous series of spectroscopic researches by which Mr. Huggins has so vastly extended our knowledge should also be here referred to. Mr. Huggins has shown that many of the substances most abundant on the earth are widely spread through the universe. Take, for instance, the metal iron and the gas hy drogen. We can detect the existence of these elements in objects enormously distant. Both iron and hydrogen exist in many stars, and hydrogen has been shown, in all probability, to be an important constituent of the nebulæ. That the rest of the sidereal system should thus be com

posed of materials known to be to a large
extent identical with the materials in the
solar system is a presumption in favor of
the universality of gravitation.

question of no small importance. It presents features analogous to certain very interesting problems in biology which the labors of Mr. Wallace have done so much to elucidate. We are told that the fauna and flora of an oceanic island, cut off from the perpetual immigration of new forms, often assumes a very remarkable type. The evolution of life under such. circumstances proceeds in a very different manner to the corresponding evolution in an equal area of land which is connected with the great continental masses. Is our sidereal system to be regarded as an oceanic island in space, or is it in such connection with the systems in other parts of space as might lead us to infer that the various systems had a common character?

In what has hitherto been said, we
have attempted to give an outline of the
facts so far as they are certainly known
to us. Into mere speculations we have
no desire to enter. We may, however,
sketch out a brief chapter in modern
sidereal astronomy, which seems to throw
a ray of light into the constituents of the
vast abyss of space which lies beyond the
scope of our telescopes. The ray of
light is no doubt but a feeble one, but we
must take whatever information we can
obtain, even though it may fall far short
of that which an intellectual curiosity will
desire. The question now before us may
be simply stated. Are we entitled to sup-
pose that the part of the universe acces-
sible to our telescopes is fairly typical of
the other parts of the universe, or are we
to believe that the system we know is al-
together exceptional; that there are stars
in other parts quite unlike our stars, com-ible to us are fairly typical of the general
posed of different materials, acted upon contents of the universe. The strongest
by different laws, of which we have no evidence that can be presented on this
conception? The presumption is that the subject is met with in the peculiar circum-
materials of which our system is com-stances of one particular star. The star
posed are representative of the materials in question is known as number 1830 of
elsewhere. This presumption is strength- Groombridge's catalogue. It is a small
ened by the very important considerations
now to be adduced.

The evidence seems to show that the stars in our system are probably not permanently associated together, but that in the course of time some stars enter our system and others stars leave it, in such a manner as to suggest that the bodies vis

star, not to be seen without the aid of a
telescope. This star is endowed with a
very large proper motion. It would not
be correct to say that its proper motion
exceeds that of any other known star, but
it certainly has the largest visible proper
motion of any star of which the distance

In the first place, let us distinctly un-
derstand what is meant by our sidereal
system. We have already dwelt on the
isolated position of the sun and the at-
tendant planets. The grandest truth in
the whole of astronomy is that which as-is known. The proper motion of 1830
serts that our sun is only a star separated Groombridge amounts to over seven sec-
by the most gigantic distances from the onds annually. It would take between
other stars around. Our sun, indeed, ap- two and three centuries to move over a
pears to be but one of the vast host of distance in the heavens equal to the ap-
stars which form the milky way. We parent diameter of the moon. The dis-
need not here enter into the often distance of this star is much greater than
cussed question as to whether the nebulæ might have been anticipated from its very
are, generally speaking, at distances of the large proper motion. The estimates of
same order as the stars. There seems to the distance present some irregularities,
be no doubt that some of the nebulæ are but we shall probably be quite correct in
quite as near to us as some of the stars. assuming that the distance is not less than
At all events, for our present purpose, we two hundred billions of miles. This star
may group the milky way, the nebulæ, the is indeed ten times as far from us as a Cen-
stars, and the clusters, all into one whole tauri, which is generally considered to be
which we call our sidereal system. Is the sun's nearest neighbor, in our sidereal
this sidereal system as thus defined an system. The proper motion and the dis-
isolated object in space? are its mem-tance of 1830 Groombridge being both
bers all so bound together by the law of assumed, it is easy to calculate the veloc
universal gravitation that each body, what-ity with which that star must be moving.
ever be its movements, can only describe The velocity is indeed stupendous and
a certain path such that it can never de- worthy of a majestic sun; it is no less
part finally from the system? This is a than two hundred miles a second.

It

1

would seem that the velocity may even be | than twenty-five miles a second, then, af much larger than this. The proper motion of the star which we see is merely the true proper motion of the star foreshortened by projection on the surface of the heavens. In adopting two hundred miles a second as the velocity of 1830 Groombridge, we therefore make a most moderate assumption, which may and probably does fall considerably short of the truth. But even with this very moderate assumption, it will be easy to show that 1830 Groombridge seems in all prob. ability to be merely travelling through our system, and not permanently attached thereto.

at least one hundred times as massive as
the system we have supposed would be
required if this star was to be recalled.
The result of this inquiry is really only
to be stated as an alternative: either our
sidereal system is not an entirely isolated
object, or its bodies must be vastly more
numerous or more massive than even our
most liberal interpretation of observations
would seem to warrant. It seems more
reasonable to adopt the first branch of the
alternative. If this be so, then we see
that 1830 Groombridge, having travelled
from an indefinitely great distance on one
side of the heavens, is now passing
through our system for the first and the
only time. After leaving our system this
star will retreat again into the depths of
space, to a distance which, for anything
we can tell, may be practically regarded as
infinite. Although we have only used this
one star as an illustration, yet it is not to
be supposed that the peculiarities which it
presents are absolutely unique. It seems
more likely that there may be many other
stars which are at present passing through
our system. In fact, considering that
most or all of the stars are actually in
motion, it can be shown that in the course
of ages, the whole face of the heavens is
gradually changing. We are thus led to
the conclusion that our system is not an
absolutely isolated group of bodies in the
abyss of space, but that we are visited by
other bodies coming from the remotest re-
gions of space.

the

It may

of objects, beginning at one end with the most diffused nebulosity, and ending at the other with an ordinary fixed star or group of stars. Each object in the series differs but slightly from the object just before it and just after it. It seemed to Herschel that he was thus able to view the actual changes by which masses of phosphorescent or glowing vapor became. actually condensed down into stars. The condensation of a nebula could be followed in the same manner as we can study the growth of the trees in a forest by comparing the trees of various ages which the forest contains at the same time. In attempting to pronounce upon the positive evidence available in the discussion of Herschel's theory, we encounter a well-known difficulty. To establish this theory, it would be necessary to watch the actual condensation of one single nebula from the primitive gaseous condition down to the stellar points. It may easily be conceived that such a process would require a vast lapse of time, perhaps enormously greater than period between the invention of the telescope and the present moment. at all events be confidently asserted that the condensation of a nebula into a star is a process which has never been witnessed. Whether any stages in that process can be said to have been witnessed is a different matter, on which it is not easy to speak with precision. Drawings of the same nebula made at The whole range of astronomy presents different dates often exhibit great disno speculations which have attracted more crepancies. In comparing these drawattention than the celebrated nebular ings, it must be remembered that a nebula hypotheses of Herschel and of Laplace. is an object usually devoid of distinct outWe shall first enunciate these specula-line, and varying greatly in appearance tions, and then we shall attempt to indicate how far they seem to be warranted by the actual state of scientific knowledge. In one of his most memorable papers, Sir W. Herschel presents us with a summary of his observations on the nebulæ arranged in such a manner as to suggest his theory of the gradual transmutation of nebulæ into stars. He first shows us that there are regions in the heavens where a faint diffused nebulosity is all that can be detected by the telescope. There are other nebulæ in which a nucleus can be just discerned; others again in which the nucleus is easily seen; and still others where the nucleus is a brilliant star-like point. The transition from an object of this kind to a nebulous star is very natural, while the nebulous stars pass into the ordinary stars by a few graduated stages. There are, however, good grounds for It is thus possible to enumerate a series | believing that nebulæ really do undergo

with different telescopic apertures. Take, for instance, the very splendid nebula in Orion, which is one of the most glorious objects that can be seen in a telescope. There can be no doubt that the drawings made at different times do exhibit most marked differences. Indeed the differences are sometimes so great that it is hard to believe that the same object is depicted. It is well to look also at draw. ings made of the same object at the same time, but by different observers and with different telescopes. Where we find contemporary drawings at variance-and they are often widely at variance - it seems hard to draw any conclusion from drawings as to the presence or the absence of change in the shape of the nebula.