80 DWIGHT 



illustrations of this among them, the Meadow Lark {SturncUa 

 magna) (plate II, fig. 17) the Grasshopper Sparrow {Ammo- 

 dranius savaniianiui passcrimis) (plate II, fig. 3) the Rose- 

 breasted Grosbeak {Habia ludoviciaua), and many others ^\•here 

 bars and spots of light color become singularh- eroded during 

 the breeding season. 



Buff or pale-tinted edgings of dark colored feathers, produc- 

 ing in the plumage a veiled effect, are acquired by man\' species 

 at the time of moult both in the spring and in the autumn and 

 seem to owe their deciduous character as much to their color as 

 to their structure. I ha\-e examined hundreds of such feathers 

 under the microscope and can find little evidence that they wear 

 down to the black or other darker color, because at this point 

 an "interlocking" or strengthening of the barbules takes place 

 as has been stated and even figured. Xo such conditions regu- 

 larly prevail, for the black color often extends distally far beyond 

 the point where the barbules cross and often the breaking off of 

 the barbs either does not reach the black, or on the other hand, 

 the black ma}' be involved to a considerable extent, as for in- 

 stance in the Meadow Lark (plate II, fig. 14). It is significant, 

 however, that each overlying feather tip should reach only to 

 the limits of the black area of the feather beneath, leaving its 

 pale margin wholly exposed to wear. Veiled species are the 

 rule in autumn and the loss of feather edgings produces remark- 

 able color changes in the plumage, although there is no 

 actual pigmentary change in the individual feathers, an im- 

 portant difference to be noted. By wear alone the brown Snow- 

 flake {PlcctropJicnax nivalis), for instance, becomes entirely black 

 and white in the breeding season and the brown young of the 

 Red-winged Blackbird {Agc/ains phiviiiccns) assumes a black 

 breeding dress. In these and many other species the actual 

 shape of the individual feathers is changed but always by de- 

 struction of their substance. It is equally true that whatever 

 pigmentaiy color change takes place in a feather there is always 

 destruction of color, never a recoloration. I find no slightest 

 exception, the apparent exceptions being optical delusions. In 

 proof of one such delusion I need cite but two species : the 



