i8 7 7-] 
On Scientific Method. 
487 
instruments employed, and errors in our readings of these 
instruments, &c. The result can never be more than ap- 
proximately correct, and the results obtained by the different 
methods will not be exactly the same. We do not therefore 
know the true atomic weight of the element ; but the theory 
of probability enables us to assign to each result its compar- 
ative trustworthiness, and so to deduce the numerical proba- 
bility of the average result being absolutely corredt. 
The application of the theory becomes often very difficult. 
Our knowledge is at the best so limited that it is difficult to 
assign to two propositions their relative probabilities. When 
we deal with simple numbers, as those obtained in the illus- 
tration given, we can apply the theory with comparative 
ease ; but when we come to more complicated questions in 
physical science we find it almost impossible to obtain suffi- 
cient, and sufficiently reliable, data to enable us to estimate 
probabilities. But, as Prof. Jevons has pointed out, “ Nothing 
is more requisite than to distinguish carefully between the 
truth of a theory and the truthful application of the theory 
to adtual circumstances. As a general rule, events in Nature 
or Art will present a complexity of relations exceeding our 
powers of treatment. The infinitely intricate adtion of the 
mind often intervenes, and renders complete analysis hope- 
less. If, for instance, the probability that a marksman shall 
hit the target in a single shot be 1 in 10, we might seem to 
have no difficulty in calculating the probability of any suc- 
cession of hits : thus the probability of three successive hits 
would be one in a thousand. But, in reality, the confidence 
and experience derived from the first successful shot would 
render a second success more probable. The events are not 
really independent, and there would generally be a far greater 
preponderance of runs of apparent luck than a simple calcu- 
lation of probabilities could account for. In many persons, 
however, a remarkable series of successes will produce a 
degree of excitement rendering continued success almost 
impossible.” 
We must be content with partial knowledge. 
In ascending from fadts to generalisations, which general- 
isations are more or less probably true, we must make use of 
hypotheses ; we must accumulate fadts, make an hypothesis 
to explain them, and test the hypothesis by appeal to fadts. 
The investigator of Science must begin with fadts ; he must 
end with fadts; but between the two he must interpolate 
hypothesis. He looks at a number of fadts; gradually he 
sees, or thinks he sees, a light dawning on him — a central 
idea, round which all the fadts group themselves in a 
