176 ORTHOGENETIC EVOLUTION IN PIGEONS. 



librium; in the organism we have a fluctuating equilibrium kept up between loss 

 and replacement. 



A form that is constant in nature is a form in which the characters have reached 

 a condition of balance — of stable equilibrium. If we select a variation in one direc- 

 tion — deal with a single character — we find the variation can be maintained only 

 by the continued aid of selection. Without this aid regression carries the form 

 towards the type. 



The variation is transmissible, but liable to regress, because the other charac- 

 ters have not moved with it. For stability, the organism must change as a whole. 

 The center of gravity for the whole organism must shift; all of the parts must be 

 readjusted. 



Artificial selection is one-sided, and works with forms that are the expression 

 of weakness, abnormal strength, and with features that are in reality deformities 

 due to internal derangement or abnormal external influences. 



In ontogeny we have invisible changes going on for a longer or a shorter time, 

 and then the outward change appears as the summation of the internal steps. This 

 is illustrated in color-patterns that come on at the end of periods of apparent rest. 



Ontogeny teaches us, then, that there are no disconnected jumps in its processes. 

 The visible surface-changes are often separated by intervals of apparent rest; but 

 these periods of seeming inactivity are filled with subtle internal processes that bind 

 all the external form-changes into one unbroken sequence. The invisible work 

 going on beneath the surface follows steadily in a definite direction, culminating 

 at the appropriate times and places in all of the outer and inner form and structure 

 characters peculiar to the species. Every organ and tissue has its history running 

 continuously back through serial form-changes and differentiations that become 

 more and more simple, until finally even the rudiments sink out of sight and are 

 lost in primordials indistinguishable with the most powerful microscopes. 



When we reflect that the periods of development required to individualize the 

 rudiments of organs must correspond in many cases to ages of phylogenetic history, 

 we begin to realize that the time of specific characters does not date from the moment 

 of becoming visible, but from the very first inception of the primordial. 



If we shrink from the idea of a primordial carried along in the germs of succes- 

 sive generations for long periods without revealing itself in a visible character, it 

 may help us if we recall the fact that primordials that have already come to full 

 development as visible characters may remain latent and invisible for many genera- 

 tions, and then reappear in perfect forms. 



The guinea-mark thus comes as a "mutation" in a young Zenaida, but it has 

 a premarked adumbration in the breast-feathers of many Zenaida young, in the 

 young turtle-dove, and other species. Moreover, it lasts here only in the first 

 feathers, being lost in the second plumage. But these marks in the juvenal feathers, 

 although they disappear in the adult, are yet hereditary, for they appear in the 

 juvenal feathers of the second generation. 60 They are therefore referable to the 

 germ-cells rather than to any special conditions of environment. 



w If these first feathers had been pulled out at short intervals, in an early period, I believe the marks could have 

 been made la disappear gradually, and show its continual gradual elimination, just as continuity can be shown in the 

 diamond-dove and other forms. (See Chapter X). 



Note. — The manuscripts and records used in the preparation of this chapter were found in folders designated as 

 follows: C 21, CC 2, CC 3, C 000, CC 0, CC 000, H, H 12, I, K, L 10, Misc., W 2, W 6, WW 1, WW 6, XS 1, XS 3. 

 XW 2, Z 3, Z 8, Z 9.— Ed.) 



