1 877-] 
On Scientific Method. 
493 
hypothetical elements. Nor have his predictions been alto- 
gether unfulfilled. The most recent addition to the chemical 
elements is the metal gallium : in very many of its proper- 
ties — in fa eft in its general chemical behaviour, so far as this 
has been experimentally examined — gallium corresponds 
very closely with one of Mendelejeff’s hypothetical elements. 
We have here an example of an hypothesis founded on ana- 
logical reasoning. 
But analogy may mislead ; it has often misled men in 
framing hypotheses. As telescopes were made of greater and 
greater power, astronomers found that the nebulae were re- 
solved into clusters of stars. One by one these apparently 
gaseous masses were proved to be really aggregations of solid 
matter. Analogy suggested that all nebulae would be resolved 
when sufficiently powerful instruments could be brought to 
bear upon them. But meanwhile a new method of research 
was discovered; and by the use of speCtrum analysis Huggins 
has proved that certain nebulas really consist of gaseous 
matter, and has therefore shown that the analogy in the 
structure of these bodies was not so complete as was 
supposed. 
Analogy must evidently be used with caution. And here 
again we perceive the need of genius in Science. The ordi- 
nary man may amass faCts, may even trace out a few 
analogies between groups of faCts, but it is only the man of 
genius who will discover the analogy which will guide to 
great generalisations. Very probably even the genius will 
follow many false scents ; but if he be a true student of 
Nature, besides being possessed of the divine gift of ima- 
gination, he will test his hypotheses framed on analogical 
reasonings by appeal to faCts, and he will discover the true 
analogy and frame the correct hypothesis at last. 
Of the vast masses of faCts which are presented to the 
enquirer in each branch of Science there will be some of 
more value — considered as guides in deducing general 
laws — than others. Not unfrequently it happens that it is 
the faCt which somehow refuses to fit in with the generally 
accepted hypothesis which becomes the means of guiding 
the investigator to a new and wider hypothesis. “ When, 
in an experiment, all known causes being allowed for, there 
remain certain unexplained effects (excessively slight it 
may be), these must be carefully investigated, and every 
conceivable variation of arrangement of apparatus, &c., 
tried, until, if possible, we manage so to exaggerate the 
residual phenomenon as to be able to deteCt its cause. . It 
is here, perhaps, that in the present state of Science we 
