THE PROBLEM OF THE ORIGIN OF SPECIES. 47 



strongly expressed than that before described in the crescentic marks of the pos- 

 terior band. On the distal side of the lower mark, or half-element, there is a very 

 thin sprinkling of dark pigment, the meaning of which becomes clear in the next 

 feather below. There is also a very weak iridescence overlying this mark and 

 extending over the surface to near the apex of the feather. 



In the third feather the band-character falls still farther below the type, so far, 

 in fact, that had we only this one feather to look at its identification would be some- 

 what doubtful. Close comparison with the first and second feathers, however, is all 

 that is required to make the conditions here intelligible and to let us into their full 

 transitional significance. It is in this feather that the two characters — the bar 

 and the band — are presented in nearly even balance, the former being near the 

 mid-stage of decline, the latter at about the same stage of its rise. What appeals 

 most to our interest here is the union of the two characters within the field of the 

 older one, the individuality of the latter being remolded directly into that of the 

 new character. We can not, of course, see the remolding, but only the several 

 steps in their natural order. Let us look at the picture in this feather, then, as a 

 delineation of a single step in the series — as an image of a character caught by the 

 feather at a moment when the transmutation was in actual progress. 



We noticed that the inclination and width of the band-character and the 

 distance from the tip increased in passing from the first to the second feather, 

 while its sharpness of outline diminished, especially on the distal side, where a 

 sprinkling of pigment appeared, spreading away to near the edge of the feather. 

 These peculiarities are increased or diminished in the same respective senses in pass- 

 ing to the third feather. The sprinkling of pigment is strengthened to a degree that 

 enables us to recognize in the picture as a whole the form of the longitudinal spot seen 

 in the corresponding juvenal feather. In short, we see here our two characters 

 blended into one; or, more accurately speaking, the new band-character arises from 

 the proximal border of the original spot, or bar-character, and hence takes its inclina- 

 tion, outline, and distance from the tip of the feather directly from the spot itself. 



The shape of the spot as a whole is given in plate 17 (first row). It is approxi- 

 mately fusiform and is set obliquely just within the white crescentic edge of the 

 feather. Its long axis is 12 mm. and its short axis 6 mm. Its distal side is strongly 

 rounded; the proximal side is but slightly curved and inclines 36°. In the upper 

 web a small triangular area of pigment is seen, its longer side (4.5 mm.) resting on 

 the shaft. This is a remnant (not a rudiment) of a spot; the proximal border of 

 which shows an obscure band-like differentiation (compare adult condition). 



The proximal border of the spot in the lower web represents the band-character 

 rising into plain view in its lower third, but attaining only a shadowy outline in 

 the upper part, where the pigment is sprinkled in only a little more closely than 

 elsewhere in the spot. This border has a slight curvature, which becomes a little 

 stronger towards the lateral edge of the feather, finally makes an abrupt bend, 

 and runs out into a small and pointed extension, thus giving a rather sharp boundary 

 to this angle of the spot. The extension reaches only 1.5 mm. beyond the border 

 proper, and is therefore quite inconspicuous. It is constant, however, and becomes 

 a convenient landmark for locating the band-character in the first and more obscure 

 stages of its appearance. It sinks almost out of sight in the second feather, measur- 

 ing here only 0.5 mm., and no trace of it is left in the first feather. 



