第174章

In the chapter on "The Instability of the Homogeneous," it wasshown that a uniform force falling on any aggregate, produces unlike modificationsin its different parts -- turns the uniform into the multiform and the multiforminto the more multiform. The transformation thus wrought, consists of eitherinsensible or sensible changes of relative position among the units, or ofboth. Such portion of the permanently effective force as reaches each differentpart, or differently-conditioned part, may be expended in modifying the mutualrelations of its constituents; or it may be expended in moving the part toanother place; or it may be expended partially in the first and partiallyin the second. And if little or none is absorbed in re-arranging the componentsof a compound unit, much or the whole must show itself in motion of suchcompound unit to some other place in the aggregate. and conversely. Whatmust follow from this, in cases where none or only part of the force generateschemical re-distributions, what physical re-distributions must be generated?

Parts that are similar to each other will be similarly acted on by the force,while parts that are dissimilar will be dissimilarly acted on. Hence thepermanently effective incident force, when wholly or partially transformedinto mechanical motion of the units, will produce like motions in units thatare alike, and unlike motions in units that are unlike. If then, in an aggregatecontaining two or more orders of mixed units, those of the same order willbe moved in the same way, and in a way that differs from that in which unitsof other orders are moved, the respective orders must segregate. A groupof like things on which are impressed motions that are alike in amount anddirection, must be transferred as a group to another place, and if they aremingled with some group of other things, on which the motions impressed arelike one another, but unlike those of the first group in amount or directionor both, these other things must be transferred as a group to some otherplace -- the mixed units must undergo a simultaneous selection and separation.

Further to elucidate this process, let me set down a few instances inwhich we may see that the definiteness of the separation is in proportionto the definiteness of the differences among the units. Take a handful ofpounded substance, containing fragments of all sizes, and let it fall graduallywhile a gentle breeze is blowing. The large fragments will be collected onthe ground almost immediately under the hand; somewhat smaller fragmentswill be carried a little to the leeward; still smaller ones further away;and those minute particles we call dust, will be drifted far before theyreach the earth: that is, the segregation is indefinite where the differencesamong the fragments are indefinite, though the divergences are greatest wherethe differences are greatest. If, again, the handful be made up of distinctorders of units -- as pebbles, coarse sand, and dust -- these will, underlike conditions, be segregated with greater definiteness. The pebbles willdrop almost vertically; the sand, falling obliquely, will deposit itselfwithin a tolerably circumscribed space beyond the pebbles; while the dustwill be blown almost horizontally to a great distance. A case in which anotherkind of force comes into play, will still better illustrate this truth. Througha mixed aggregate of soluble and insoluble substances, let water slowly percolate.

There will in the first place be a distinct parting of the substances thatare the most widely unlike: the soluble will be carried away; the insolublewill remain behind. Further, some separation, though a less definite one,will be effected among the soluble substances; since the first part of thecurrent will remove the most soluble in the largest amounts, and after thesehave been dissolved, it will continue to bring out the remaining less soluble.

Even the undissolved matters will have simultaneously undergone some segregation;for the percolating fluid will carry down the minute fragments from amongthe large ones, and will often deposit those of small specific gravity inone place, and those of great specific gravity in another. To complete theelucidation we must glance at the obverse fact; namely that mixed units whichdiffer but slightly, are moved in but slightly different ways by incidentforces, and can therefore be separated only by such adjustments of the incidentforces as allow slight differences to become appreciable factors in the result.

The parting of alcohol from water by distillation is a good example. Herewe have molecules consisting of oxygen and hydrogen, mingled with moleculesconsisting of oxygen, hydrogen, and carbon. The two orders of molecules havea considerable likeness of nature: they similarly maintain a fluid form atordinary temperatures; they similarly become gaseous more and more rapidlyas the temperature is raised; and they boil at points not very far apart.