16 ORTHOGENETIC EVOLUTION IN PIGEONS. 



ent stages of new characters. Mutation would be equally helpless, and each step 

 would leave a gulf of discontinuity — something that nature seems to abhor. 



Turning from theoretical impasses to the facts, let us compare the two patterns. 

 In the chequered pattern all the feathers are marked alike — no regional differ- 

 entiation. In the other type we have a conspicuous local differentiation (pi. 2 and 

 text-figs. 1 and 2) , suggesting at once a higher stage of evolution. 5 Chequered wings 

 are to be found which vary all the way between a uniform marking and the barred 

 type. If we arrange a number of unequally chequered wings in a series, running 

 from the most to the least chequered, we shall see (pi. 3) that the pattern ap- 

 proaches more and more nearly to that of two bars, as the chequers diminish in 

 size and number. We shall notice that the pigment is reduced more rapidly in the 

 anterior than in the posterior part of the wing. 



As chequers are reduced, they gradually lose their sharp ends and approximate 

 the square or rounded form seen in the elements of the typical bars. The series 

 shows a flowing gradation, that may be read forward or backward with equal 

 facility. Darwin's view takes the bars as the starting-point and reads forward. 

 Taking the chequered condition as the point of departure, the variation runs just 

 as smoothly in the opposite direction. We here meet an ambiguity that is every- 

 where present in color-pattern problems — an ambiguity that is frequently over- 

 looked with disastrous consequences. The only way to eliminate the difficulty is 

 to take our evidence from several different sources, and when agreement is found 

 for one direction and disagreement for the other, the way is clear. 



As an experiment, we may take one or more pairs of pure-bred, typically barred 

 pigeons and keep them isolated from chequered birds for several years, in order to 

 see if the young ever advance toward the chequered type. 



Another experiment should be tried for the purpose of seeing what can be done 

 by working in just the opposite direction. In this case we take chequered birds, 

 selecting in each generation birds with fewer and smaller chequers, and rejecting 

 the others, in order to see if the process of reduction can be carried to the condition 

 of three, two, and one bar, and finally to complete obliteration of both chequers 

 and bars, leaving the wing a tabula rasa of uniform gray color. 



If these experiments are continued sufficiently far, it will be found from the 

 second experiment that a gradual reduction of pigment to the extreme conditions 

 named can be comparatively easily effected, and that the direction of reduction 

 will always be the same, from before backward; while, from the first experiment, 

 it will be seen that it is hopeless to try to advance in the opposite direction, from 

 the bars forward to the chequered condition. No variations will appear in that 

 direction, but such as do appear will take the opposite direction, tending to diminish 

 the width of the bars and to weaken their color. It is in this way that we must 

 account for the existence of some fancy breeds in which the bars have been wholly 

 obliterated. The direction of evolution can never be reversed. I have tried both 

 experiments for eight years, and as both tell the same story as to the direction of 

 variation, I am satisfied that further experiments will not essentially modify the 

 results. Plates 4 and 5 indicate some of the results 6 in the reduction of the chequers. 



'This situation is dealt with more fully in the explanations of plates 2 and 3. — Ed. 



6 Most of the available illustrations for these results exist only in a series of small lantern-slides. The breeding 

 data for the birds illustrated in plates I anil 5 are given in table 70, Vol. 11 ; some further breeding data are also given 

 in Chapter IX, and another illustration in plate 16 of Vol. II. — Ed. 



