298 ROBERTS, Birds of the Granite Belt ^^t.^^fx 



slightly decurved point, and also in Brevirostris, but in this case 

 the tip beyond the notch is not decurved. In many of the other 

 species this notch is also represented by a greater or less number 

 of serrations. At times, too, a heavier underbeak is required, 

 and this is got by various modifications; in Clithonicola and 

 yllbogidaris it has the shape of a miniature cricket bat, looked at 

 from the side, and laid face upwards, but in Brevirostris it is 

 obtained by reinforcing the edges. In Pusca the beak is as broad 

 as it is high at any given jjoint. 



In regard to the wing, much is made of the wing formula, and 

 particular attention has been paid to the length of the second 

 primary compared with the secondaries. In order to make clear 

 which are primaries and which are secondaries a plate is given 

 of the wing of a young pigeon which has recently shot its quills. 

 Let it be said at once that of all the external differences this is 

 as debatable as any, for it is really in many cases a hard matter 

 to detect the ditTerence, which is only a fractional part of an 

 inch. And their meaning and application anatomically is com- 

 plicated by the fact that the last two j)rimaries are undergoing 

 atrophy in the majority of birds, and the reason is this. The 

 secondaries are attached to the bird's fore-arm, the primaries to 

 the rudimentary hands and fingers; of these latter there are 

 three — two having a single joint, and one, two joints. The second 

 joints of two have disappeared, but in the other it is still present, 

 though undergoing degeneration : hence the structures which it 

 carries, and these are the last two primaries, are also undergoing 

 degeneration. In some birds the process is complete, and the 

 bird has nine primaries instead of the usual eleven. As an 

 exami)le of a nine-primaried bird, the swallow may be cited, so 

 that although use is a factor, no one can say where the atrophic 

 inherited element ends and the hypertrophic from use begins. For 

 the jnirpose of clearness of e\j)ression the ala sf>iiria is neglected. 

 Certainly we are willing to admit that the second primary of 

 . \lbo<iularis is distinctly longer than the first secondary, and that 

 this is reversed in Fiisea. We also admit that Brevirostris fol- 

 lows the formula of .Ubo(/itlaris, but the position in regard to the 

 AcaHthicae and Gcohasilcus is not so clear; Chrysorrhous cer- 

 tainly has a second primary, which can be said to be longer than 

 the secondaries, but this we have not been able to confirm in 

 regard to Re(/iiloides, Litieata or PnsiUa: in fact, this j)()int in 

 them is not constant. One other point needs mention, and this in 

 regard to the method adopted for comparing the feathers men- 

 tioned: it is done by noting which feather projects the further 

 in the folded wing, but as the wing in this position lies with its 

 iiones folded parallel the point of origin of the first primaries is 

 lower than that of the corresponding secondaries, and hence a 

 fallacy may creep in. 



In regard to the legs, feet and tail, wc are forced from observa- 

 tion to take up the attitude that modifications of these are simply 



