BIRDS OF ETHIOPIA AND KENYA COLONY 141 



3. G. c. pyrrhorvhoa. — Madagascar, Mauritius, and Eeimion. Char- 

 acterized by having the under tail coverts deep buff, while the other 

 races have these feathers white or only slightly tinged with buff. 



4. G. c. seycheUarum. — Seychelle Islands (at least He Aride). 

 Characterized by having the under tail coverts buffy, almost as much 

 as in pyrrJiorrhoa and the upper wing coverts more brownish than 

 in hrachyptera and wings slightly shorter (150-170 millimeters). 

 This form is recognized by Sclater ^^ who gives as its range the 

 Se3'chelle Islands. However in his review of GalUnula chloropus ^- 

 Hartert writes that birds from some of the islands are like African 

 ones {lyrachyptera) and expresses some doubt as to the possibility of 

 a small island like He Aride having a form distinct from that of the 

 rest of the Seychelles. Later ^^ he includes all the islands in the 

 range of the race, but remarks that the color characters apply 

 particularly to specimens from He Aride. Series from the Sey- 

 chelles some day will show whether this race is valid or not. In 

 the Museum of Comparative Zoologj^ there is a specimen (No. 

 11292) of seycheUarum collected by Layard (and labeled " South 

 Africa,'' but which probably came from the Seychelles). This 

 bird agrees with Hartert's description, but is even browner above 

 than liis account would indicate. 



C. H. B. Grant ^^ has given an extended, detailed account of the 

 plumages and molts of the typical race of this moorhen. He has 

 shown that the birds molt all the remiges at once, as do many ducks, 

 some rails, swans, and anhingas. In the extensive series in the 

 Museum of Comparative Zoology are specimens of the Madagascan 

 race, pyrrhorrhoa^ and of the North American form cachinnans, in 

 which all the remiges are growing in simultaneously, sliowing that 

 this drastic, sudden replacement of all the flight feathers is not con- 

 fined to the typical subspecies, but is probably common to all the 

 forms. It is rather surprising that it has not been recorded for the 

 North American bird, although much has been written of this form.^^ 

 Bent does not mention it in his detailed account of cachinnans. 



Grant's account of plumage variations in chloropus applies equally 

 well to hrachyptera. The number and width of the white stripes on 

 the flanks, the amount of whitish on the middle of the abdomen, and 

 the whiteness of the under tail coverts vary without respect to sex, 

 age, or season, although it is true tliat the white flank stripes become 

 wider on males than on females, but only in extreme cases. The 

 series before me shows considerable variation in the amount of yellow 

 on the bill; the two extremes are, (1) yellow confined to the tip (7 



" Syst. Avium Ethiop., 1924, p. 108. 



^-Nov. Zoo]., vol. 24, 1917, pp. 267-270. 



"Vog. pal. Fauna, p. 1843, 1921. 



" Ibis, 1914, pp. 298-304. 



1^ Bull. 135, U. S. Nat. Mus., 1926, pp. 331-352. 



