But the examination of residual phenomena may also help to free investigators from that tyranny which is exerted by a number of concordant results, all seemingly pointing to but one conclusion.
If experiment after experiment points to one conclusion, and if all, with the exception of perhaps a single residual fact, is in favor of this conclusion, it is hard to resist the temptation to ignore that fact, and adopt what, but for it, is apparently the true conclusion. But this method is not the scientific method. The fact must be examined. It may be that the outstanding fact is finally reduced within the sphere of the previously adopted hypothesis, or it may be that a new hypothesis is suggested which explains this and all the other phenomena.
The great Swedish chemist Berzelius carefully examined the properties of the compounds of a newly discovered element; he determined the chemical and physical characteristics of this element, to which he gave the name of Vanadium. The facts ascertained by the experiments of Berzelius formed a concordant series; so far as these experiments extended, everything appeared to be in keeping with the conclusions arrived at by him. But it was afterward noticed that the crystalline form of certain compounds of the metal vanadium was different from that required by the commonly accepted and, as it appeared, well-established theories concerning the connection between crystalline form and chemical structure. The examination, by Roscoe, of the residual phenomena presented by the crystalline forms of the vanadium compounds led to the astonishing discovery that the so-called metallic vanadium of Berzelius was really not an elementary body, but a compound of the true metal vanadium with oxygen. This peculiar oxide presents most of the physical properties of a metal; indeed, so metal-like is this oxide that the presence in it of oxygen was entirely overlooked, even by so careful a worker as Berzelius.
The researches of Roscoe threw a new light upon the chemical history of vanadium, and at the same time confirmed in a marked manner the law connecting chemical structure with crystalline form.
But, lastly, the study of residual phenomena may aid in freeing our minds from that fascinating, but surely erroneous, idea which a mere superficial acquaintance with natural science tends so much to strengthen, viz., that Nature is, and indeed must be, extremely simple.
The simplicity of Nature is a favorite theme with a certain class of would-be philosophers: it is a doctrine easily accepted, but a doctrine which has led to pernicious results.
Extreme instances of the overruling power of this idea may be found in the fascination exerted over minds, even of the highest order, by numerical analogies, that are really baseless. The seven colors of the spectrum were supposed, even by the great master himself, to have some mysterious connection with the seven tones of music. The number of the satellites of Jupiter added to the single satellite of the earth leaves but one satellite for Saturn, if the perfect number six is to be
