August 6, 1885] 
receive a solution, and this through the aid of astronomical 
observations. 
Romer, a Danish astronomer, determined in 1675 the 
velocity of light by means of the eclipses of Jupiter’s 
satellites. It so happens that the planes in which the 
earth and Jupiter move around the sun, as well as the plane 
in which Jupiter’s satellites move around that planet, coin- 
cide very nearly with each other. As a consequence the 
first or nearest of Jupiter’s satellites passes within the 
shadow of the planet at intervals of 42hr. 28m. 36s., 
and thus becomes obscured. 
Now, if light were to travel instantaneously from Jupiter 
to the earth we should always see this obscuration at the 
moment when it took place. But even if light required 
time to travel, yet if the earth were always at a constant 
distance from Jupiter we should see the obscuration at a 
constant interval of time after its occurrence. Now 
Romer found that when the earth was furthest away from 
Jupiter there was a retardation in the time of the occur- 
rence equal to 16m. 36s., as compared with that when the 
earth and Jupiter were nearest together. 
It will be tseen from the diagram (Fig. 1) that the 
Fic. tr. 
earth and Jupiter are nearest together when the earth is 
between Jupiter and the sun, and that the two are furthest 
apart when the sun is between the earth and Jupiter. 
Hence it follows that the difference in the distances from 
each other of the two planets in these two positions is 
equal to the diameter of the earth’s orbit, or 183,000,000 
of miles. If, therefore, light takes 996 seconds to cross 
this distance it ought to travel at the rate of 184,000 miles 
per second. : 
The velocity of light has likewise been determined by 
experiment. The arrangement for this purpose adopted by 
Fizeau is the one most easily understood. It consists of a 
toothed wheel, which may be made to revolve with great 
rapidity. Now a ray of light is made to pass through one 
of the intervals between the teeth, and to fall upon a reflect- 
ing mirror placed at a considerable distance off in such a 
manner that when the wheel is at rest the ray will be re- 
flected back through the same interval. If, however, the 
wheel is in rapid motion it is possible that during the time 
which the ray takes to travel to the reflecting surface and } 
back again the wheel may have moved so much that the 
ray is caught by the next tooth, and not allowed to pass 
through ; while, if the motion be still more rapid, the ray 
may get through the next interval, and soon. Without 
entering more minutely into the conduct of the experi- 
ment, it will at once be seen that we have here the means 
of measuring the velocity of light. 
By these and similar methods this velocity is now very 
accurately known, and is found to be about 187,000 miles, 
or 300,000 kilometres per second. 
The evidence is very strong that all varieties of light, 
whether red, orange, yellow, green, blue, indigo, or violet 
move through vacant space with the same velocity. 
Having thus briefly replied to the second of these 
questions, let me now return to the first, and inquire as to 
the nature of radiant light. We are able to conceive of 
two, and only two, varieties of progress in space. The 
one of these is the progress of actual matter, the other 
the progress of a form. An arrow discharged from a 
bow, or a bullet from a gun, represents the former of these, 
while the ever-widening circles which follow the plunge 
NATURE 
y 
523 
of a stone into a pool of water represent the latter. The 
progress which is visible when the wind blows along a 
field of corn or grass is another good illustration of a 
moving form. Here the corn or the grass is certainly not 
carried along, and if the wind is so carried, yet we cannot 
see the wind. What we see is an advancing form due to 
the oscillating motion of the various heads of corn or 
blades of grass. In like manner when acannon or a gun 
is discharged at some distance from us the noise reaches 
our ear after a greater or less interval, depending upon 
the distance. Here it would be absurd to suppose that 
certain particles of air had been shot all the way from 
the cannon into our ear with the constant velocity of 1,100 
feet per second—this velocity in the case of a gun or 
pistol being likewise the same as when the most powerful 
cannon is discharged. It is well known that in this in- 
stance a blow is given to the air, thus causing an arrange- 
ment of condensed and rarified particles which progresses 
with a certain definite velocity. The speed of progress 
of this form may either be determined by direct experi- 
ment, or by calculation founded on the well known 
properties of air—the two methods agreeing perfectly well 
together. 
Now in many respects there is a strong analogy between 
sound and light, and these very questions which have 
been asked for sound are equally appropriate in the case 
of light. Can it be thought likely that hot bodies emit 
myriads of very small particles, which pass through space 
with the enormous velocity of 187,000 miles per second ? 
or again, is it likely that this velocity should be precisely 
the same for all bodies and for all temperatures ? 
It is a singular circumstance that the illustrious 
Newton, to whom science owes so much, and one of 
whose achievements was a correct, or nearly correct, 
analysis of the conditions of undulatory motion in air, 
should nevertheless have become a powerful advocate of 
the corpuscular theory of light, thus lending his great 
authority to retard the progress of the rival theory, which 
represents light as an undulatory motion, similar in many 
respects to that which constitutes sound. 
It is to Huyghens in the first place, and to Young and 
Fresnel in more recent times, that we owe the establish- 
ment of the undulatory theory of light upon so firm a 
basis that the older hypothesis is now entirely forgotten, 
or regarded only as a scientific curiosity. 
There are two ways in which a theory may break down. 
Its various assumptions may display a great lack of living 
energy, or, in other words, may exhibit inability to expand 
themselves so as to incorporate a Jarge volume of fact. Each 
new fact would thus imply the construction of a fresh 
assumption, so that there would be as many hypotheses 
as facts. A cumbrous structure of this kind, it is need- 
less to say, would be utterly useless as a scientific 
instrument, and would finally fall to pieces from its own 
weight. 
Another mode in which such a theory may break down 
is by the promulgation of some statement which is ulti- 
mately found to be contrary to fact. The corpuscular 
theory of light has broken down in both of these direc- 
tions. For, in the first place, it had to be propped up by 
many fresh assumptions devised solely for the purpose of 
explaining fresh facts, and wholly useless in any other 
respect. In the next place one of its fundamental state- 
ments was ultimately contradicted by an appeal to experi- 
ment, carried out by M. Foucault, an eminent French 
observer. According to the corpuscular theory, or that 
of emission, the velocity of light ought to be greater 
in water than in air. On the other hand, according to the 
undulatory theory, the velocity in water is less than in air. 
' If, therefore, it can be shown that light moves faster in air 
than in water then the undulatory theory is right ; if the 
contrary, then the theory of emissions is right. Foucault 
succeeded in showing by an experimental method that 
light travels faster in air than in water, and this result has 
