THE PROBLEM OF THE ORIGIN OF SPECIES. 



33 



But is there any direct proof that the transformation is actually making progress 

 to-day? May not these transitional steps go on appearing generation after genera- 

 tion, without ever making any permanent progress? 



We have to concede that we can not follow the processes that reveal themselves 

 in steps. We .can at most only see what is done, not the doing. We are entirely 

 in the dark as to the time required to carry the change through a single row of 

 feathers. But we know that this has been done in three other species of the same 

 family. We see after it is done, not before, that the transitional steps appear in 

 the next and last row. Moreover and this is as close as we can hope to get to 

 actual seeing we find that progress of just the kind we are looking for is certainly 

 made in passing from the Juvenal to the adult plumage. This is an ontogenetic 

 change of a few weeks, which we can easily demonstrate by experiment to be pro- 



TBXT-FIOUBE 10. Juvenal male crested pigeon, Ocyphaps lopHole*. Age 



15 days. Natural size. Hayashi del., 1> 



This shows, in the first row of coverts, the tr.-insitiim l>etween lonn 

 spots (longitudinal streaks of black) and cross-bars (better than 

 female Juvenal in pi. 8. ED.). In tin : .. .Tons-bur* are 



subterminal instead of terminal law in <i< ///,! . Compare 

 with adult OeypAapg (pi. 3) and I.<>i>l>iihn / (text-figure 8). 



gressive and continuous. The corresponding phylogenetic advance has left no other 

 record, and hence we only know that it took time that it was not a momentary 

 salt. In the adult plumage, one or two full steps are taken beyond the Juvenal 

 stage, and taken precisely at the points premarked by transitional steps. The 

 number of transitional steps is increased at the same time. 14 



As the next and last illustration, I take a case in which the bars are verging to 

 complete obliteration. The well-known wild stock-dove (Columba cenas) of Europe 

 may serve as a convenient and instructive example. In this pigeon we find that 

 reduction of the chequers has swept over the whole bar, leaving nothing except a 

 few obsolete spots, which we recognize as vanishing elements of bars formerly 

 more highly developed and homologous with those of the rock-pigeon. 



14 One point here should not escape attention, namely, that the transitional steps in Ocyphaps form a linear series; 

 hut there is nothing artificial or arbitrary about it. It is a small-number series, each element of which stands in an 

 appointed place and marks the height to which the transformation process rose at that point in its course. Such a 

 series can not be open to the objections which de Vries has very justly made against large-number series, tin 1 element* 

 of which are collected at random and then arranged arbitrarily to display transitional continuity. In the OcypHapi 

 series there is some fluctuation, the series varying in length, but always advancing in one predetermined direction, 

 like a tidal flow guided along a prepared channel, and flowing to varying distances, according to the initial momentum. 



