MR. E. DEGEX ON ECDTSIS. 395 



for the remiges and the major coverts of this portion, it is wholly retained by the 

 median, and in part by the minor series of coverts still, as evidenced through their 

 moult. 



VIII. — The Contoue Feather-Tracts, or Pteryl^e. 

 Preliminary Bemarks. 



No really so precise a description of the order which the moult pursues in these 

 tracts is possible as that which enabled us to follow it for the " alar tracts." 



One reason for this is that synchronous renewal is not circumscribed so definitely on 

 the body as it is on the wing. But on this head I cannot do better than quote 

 Dwight (17), whose statement, p. 84, I. c, under the heading of "Advance of Moult in 

 the Feather-Tracts," is as follows : — 



" The only way to get any idea of how a moult proceeds is to appreciate the fact 

 that it begins almost simultaneously at a number of points in the different tracts, and 

 advances independently from each of them. This is why a bird seems to be moulting 

 at irregular spots all over. There is, as might be expected, a good deal of individual 

 irregularity in the growth of new feathers, but when each tract is studied separately, 

 each will be found to have a definite plan of development ^, which in its turn fits into 

 the general scheme of the process we call the moult." And, again, in connection 

 with this subject on Gulls and Terns fi8), as regards "Sequence of Moults and 

 Plumage of the Laridse," he says (p. 50) : — 



" The moults occur at definite periods, and the feather-growth spreads from definite 

 points in the feather-tracts, so that nothing is a matter of chance unless it be the 

 aiTested development that befalls all organisms, and occasions in birds the retention of 

 old feathers among those that are replaced by new ones at time of partial moult." 



The above statements so exactly describe the conditions that exist in the species 

 under consideration that I can add very little else to them as the result of my own 

 investigation. In consequence, this account can be restricted to the narrowest limits 

 by merely pointing out a few specific deviations from the broad principle. 



A second reason for not being able to make a comparative survey of the moult in 

 the various body-tracts consists in the technical difficulty presenting itself of dealing 

 with them simultaneously, the same as they are in moult. This difficulty is further 

 enhanced by the fact that this not only applies to the tracts themselves, but, as 

 pointed out by Dwight (see anted,), to the different parts of a tract by the new growth 

 which spreads in both a diagonal, transverse, and longitudinal mannner. 



Having, moreover, early recognized the desirability to adequately express in some 

 manner of form the complex order of moult, besides the systematic description of the 

 different tracts, I have prepared a chart (Table VL, p. 398) as the only alternative 

 for intelligibly conveying the relative order of their moult. 



' The italics are mine. 



VOL. XVI. — PART VIII, No. 7. — May, 1903. 3 K 



