parts tend to vary in the same way, and homologous parts tend to cohere.  Modifications in hard parts and in external parts sometimes affect softer and internal parts.  When one part is largely developed, perhaps it tends to draw nourishment from the adjoining parts; and every part of the structure which can be saved without detriment to the individual, will be saved.  Changes of structure at an early age will generally affect parts subsequently developed; and there are very many other correlations of growth, the nature of which we are utterly unable to understand.  Multiple parts are variable in number and in structure, perhaps arising from such parts not having been closely specialised to any particular function, so that their modifications have not been closely checked by natural selection.  It is probably from this same cause that organic beings low in the scale of nature are more variable than those which have their whole organisation more specialised, and are higher in the scale.  Rudimentary organs, from being useless, will be disregarded by natural selection, and hence probably are variable.  Specific characters—­that is, the characters which have come to differ since the several species of the same genus branched off from a common parent—­are more variable than generic characters, or those which have long been inherited, and have not differed within this same period.  In these remarks we have referred to special parts or organs being still variable, because they have recently varied and thus come to differ; but we have also seen in the second Chapter that the same principle applies to the whole individual; for in a district where many species of any genus are found—­that is, where there has been much former variation and differentiation, or where the manufactory of new specific forms has been actively at work—­there, on an average, we now find most varieties or incipient species.  Secondary sexual characters are highly variable, and such characters differ much in the species of the same group.  Variability in the same parts of the organisation has generally been taken advantage of in giving secondary sexual differences to the sexes of the same species, and specific differences to the several species of the same genus.  Any part or organ developed to an extraordinary size or in an extraordinary manner, in comparison with the same part or organ in the allied species, must have gone through an extraordinary amount of modification since the genus arose; and thus we can understand why it should often still be variable in a much higher degree than other parts; for variation is a long-continued and slow process, and natural selection will in such cases not as yet have had time to overcome the tendency to further variability and to reversion to a less modified state.  But when a species with any extraordinarily-developed organ has become the parent of many modified descendants—­which on my view must be a very slow process, requiring a long lapse of time—­in this case, natural