and the Absolute Size of the Fixed Stars. 307 
not only of the proportions of the different parts of the orbit, but of 
their actual size, and the weight of the two bodies. Thus the least 
distance of these two suns is only half that of the earth from the 
sun, or a little less than that of Venus, while the greatest distance 
is a little more than that of Uranus ; and the mass of the two stars 
comes out three-quarters of that of our sun, their distance from us 
being 226,100 times our distance from the sun. 
Well, therefore, may Sir J. Herschel have said, “ that no subject 
more worthy of diligent and continued inquiry can possibly be urged 
on the attention of southern astronomers.” 
But the most interesting part of the orbit is still to come, viz., the 
periaster in 1851, and that this be well observed is indeed to be 
earnestly hoped, for the period will be an eminently crucial one; it 
proved so in the case of yy Virginis, imperatively requiring an ex- 
cessive alteration in all the elements except one, as previously cal- 
culated ; and in the case of @ Centauri, the characteristic features 
of the orbit are of a much more violently marked nature, besides be- 
ing represented altogether on a larger scale. 
The extreme importance of obtaining an abundance of observations 
at that epoch may be further indicated by the mere statement, that 
it cannot yet be considered as fully proven, that the law of gravity 
extends absolutely unaltered to the most distant parts of the sky, and 
the only mode of proof open to us, is by observing the double stars. 
It is true that most of the orbits yet computed on the theory of 
gravity have turned out very near the truth, but still not quite so 
near, it must also be confessed, as could have been desired; and in 
the luciferous case of y Virginis, every orbit that has been computed 
for it yet, has persisted in giving a minimum distance of not less than 
0:5”, while observation at the time of the periastral passage made it 
certainly much smaller. 
I do not, of course, by any manner of means, wish to express any 
doubt on these grounds as to the sufficiency of gravity to explain all 
the observed phenomena; a great part of the onus, or the whole of 
it, may rest on the excessive difficulty of the species of observation, 
and their inappropriateness for calculation in all ordinary manners, 
caused by the extreme roughness of even the very best procurable 
data; resembling, indeed, those of the comet of 1556, whose return, 
calculated on such wretched notices of its former perihelion passage, 
we have been looking out for in vain so long. , 
But whatever weight we may attach to the insufficiency of our ob- 
servations and methods of calculation, it is always proper to draw a 
distinct line of demarcation between those things which are proved 
and those which are merely inferred, and not seek to enjoy a triumph 
before the victory has been decidedly achieved. 
The above observations and calculations seemed to many 
abundantly convincing, but the illustrious author of the 
“ Outlines of Astronomy,’ published last year, says, after 
