XXI 



so-caljed tertiaries and the Imrrierals ( = the parapteron of 

 Nitzsch) (see fig. 1, p. xi). lu the pajjer mentioned^ Mr. 

 Pycraft very justly doubts the correctness of the termino- 

 logy applied to the set of feathers variously called ^'ter- 

 tiaries"^ or ''tertials/'' and he proposes to call them ""inner 

 secondaries.'"'' According to liis deductions from other writers 

 on the same subject^ the feathers forming the parapteron 

 have apparently no other purpose than to connect the gap 

 between the cubitals and the scapulars. To me it seems 

 that the term parapteron, as applied by Nitzsch, should be 

 retained, inasmuch as this term conveys no confusion as 

 to its meaning; and has the further advantage of leaving 

 open the question of the serial value. To call them 

 '^tertiaries/'' as has been done by certain writers, suggests 

 at once the idea that they are of the same value as the 

 primaries or secondaries, which are flight-feathers of the 

 first magnitude. Mr. Pycraft himself, in his subsequent 

 descriptions of the parapteron, proves that its feathers belong 

 to the series of covei^ts, being indeed a continuation of the 

 major series of coverts of the cubitus, a conclusion which 

 my own studies have am]3ly verified. 



Not only has this set, the feathers of vrhich are serial with 

 the metacarpal coverts, been derived from the upper part of 

 the bird^s body, being identical with the rest of the remigial 

 covert-series, both dorsal and. ventral, as Mr. Pycraft has 

 proved to be the case, but there are two underlying rows 

 as well. The upper series of the last-named can be traced 

 Avith equal distinctness to be a continuation of the ventral 

 major coverts. Lastly, in preparing a wing of a Demoiselle 

 Crane, I discovered that this series of ventral major coverts 

 on the parapteron is accompanied by a series of smaller 

 coverts. 



These turn their dorsal surface upwards, in the manner 

 described by "Mr. Pycraft, and they are nothing more than 

 the ventral median coverts. 



Thus we find a row of dorsal major coverts accompanied 

 by their median coverts, as well as a row of ventral major 

 coverts, also accompanied by their median coverts, and yet 



