e
fine remark of Malebranche, that labor and attention are a silent prayer
which we address to the Master of truth: the labor is a prayer, and the
discovery is an answer granted to it.

When this wholly spontaneous character of discovery is not recognized,
and when it is thought that the observation of facts naturally produces
their explanation, it must needs be granted that a discovery is
confirmed by the very fact that it is made. But this is by no means the
case. Hypothesis does not carry on its brow, at the moment of its birth,
the certain sign of its truth. A flash of light crosses the mind of the
savant; but he must enter on a course, often a long course, of study, in
order to know whether it is a true light, or a momentary glare. Every
supposition suggested by observation must be confirmed by its agreement
with the data of experience. Let us listen to a great discoverer--
Kepler. He is giving an account of the discovery of one of the laws
which have immortalized his name.

"After I had found the real dimensions of the orbits, thanks to the
observations of Brahe and the sustained effort of a long course of
labor, I at length discovered the proportion of the periodic times to
the extent of these orbits. And if you would like to know the precise
date of the discovery,--it was on the eighth day of March in this year
1618 that,--first of all conceived in my mind, then awkwardly essayed by
calculations, rejected in consequence as false, then reproduced on the
fifteenth of May with fresh energy,--it rose at last above the darkness
of my understanding, so fully confirmed by my labor of seventeen years
upon Brahe's observations, and by my own meditations perfectly agreeing
with them, that I thought at first I was dreaming, and making some
_petitio principii_; but there is no more doubt about it: it is a very
certain and very exact proposition."[164]

All the logic of discoveries is laid down in these lines; and these
lines are a testimony rendered by one of the most competent of
witnesses. You see in them the conditions of a good hypothesis: Kepler
has long studied the phenomena of which he wishes to find the law; he
has studied them by himself, and by means of the discoveries of his
predecessor Brahe. The law has presented itself to his mind at a given
moment, on the eighth of March, 1618. But he does not yet know whether
it is a true light, or a deceptive gleam. He seeks the confirmation of
his hypothesis; he does no