18S7.] OF THE WINGS OF BIRDS. 345 



The remaining; feathers of tlie wing are the coverts; they are hest 

 understood if described from the posterior margin of the vvins^. 



On tlie dorsal side the row of feathers (Plate XXX. fii. «, o) 

 lyinji; next the remiges are the tectrices majores, being qnite definitely 

 related to the remigeal quills, and lying close pressed upon tlieir 

 bases. Each remex is serial with the covert proximal to it, the 

 cubital coverts crossing over the bases of the remiges, the meta- 

 carpal coverts lying; parallel and [)ressed upon two contiguous re- 

 miges (fj/l Plate XXXI. figs, b and c). There is a well-developed 

 covert to every metacarpal exceot the first, wliich possesses only a 

 very small and vestii^ial one, 1', which is completely hidden hy a 

 median covert, 1" (Plate XXXII. fig. 8), which in many birds func- 

 tionally replaces it, the t. mcijor disa])|,ieiring. 



On the ventral surface of the witig is a row of feathers (Plate XXX. 

 fig. b, a), bearing the same relations to the remiges as those just 

 described ; these are the feet rices majores of the lower surface (c/'. 

 Plate XXX. fig. b, and Plate XXXI. fig. b). If tig. b, Plate XXX 11., 

 be examined, which shows the relations of the above feathers in 

 section, starting at the tip of the wing the remicle, or predigilal 2 

 (R'), is seen to have proximal to it a dorsal and ventral covert, form- 

 ing a group of three. The next remex is similar, and so to the 

 1st metacarpal, whose dorsal covert is very small and rudimentary. 

 These relations show that the remicle is a small metacarpo-digital 

 which has probably not been differentiated into a flight-feather. The 

 cubitals sliow the same arrangement except the fifth group, where 

 there are a pair of coverts, but no remex ; this condition is termed 

 aguincubital, and is later described more fully. 



On the dorsal surface the next row of feathers to the t. majores 

 are the median coverts (Plate XXX. fig. a, /3), or tectrices medicp, 

 arranged serially with the other groups. On the cubitus they lie 

 with a reversed overlap to the remiges and t. majores ; those, how- 

 ever, which lie most proximal are luireversed (S. p. 41.5, footnote, 

 and Goodchild, P. Z. S. 188G, p. 191). Those on the mauus lie 

 unreversed, and generally the median covert of the 2ud metacarpal 

 is wanting (cf. Plate XXXII. fii. 8). On the ventral surface of the 

 wing, the next row of feathers (Plate XXX. fig. b, /3) bears similar 

 relations ; they are the tectrices media of the lower surface, and 

 always lie with reversed overlap to tiie remiges and t. majores (S. 

 p. 491). The distal four or five are generally deficient on the manus 

 in the Duck (Plate XXXI. fig. b). In many birds they are nearly 

 all suppressed on the manus. 



The tectrices majores and mediae on the ventral surface have at 

 first sight an anomalous position. Being ou the ventral side of the 

 adult wing, one would expect the backs of the feathers to look 

 ventralwards, whereas they look dorsalwards just as do the remiges. 

 This is pointed out by Sundevall (S. p. 419), who, however, gives au 

 erroneous explanation, saying they are aftershafts developed at the 

 ex jense of the true feather-shaft ; a more probable explanation is 

 discussed later. 



The feathers so far described are seated in the wing-membrane 



