ORNITHOLOGICAL EXPLORATIONS. 53 



am not so certain in regard to fig. 2, text. That bird was picked out 

 of a lot of similar ones wbicli I examined when visiting the rookery 

 on Toporkof Island on the 4th of June, 188 ?. Here are the remarks 

 concerning these specimens, written down on the spot : " The bill is con- 

 siderably shorter and lower, and the casque is arched and not straight 

 as in the old birds ; the colors are less bright, the red being j)aler and 

 more orange, and the green more olive, the casque alone being clear 

 apple-green ; the rosette is less vivid, the red especially less conspicu- 

 ous and more pinkish ; the naked eye-ring is narrower and more dull 

 colored, while the feet are more orange." These were evidently birds 

 of the foregoing year, and their presence on the rookery so late would 

 seem to indicate that they were going to breed that season, an assump- 

 tion the more probable, as the eggs in the ovary were quite swollen. 

 It may be, therefore, that the early-hatched birds breed the next season, 

 while those reared late in the year — and downy young are found in the 

 nests as late as the end of September — pass the first summer without 

 partaking in the reproduction. 



Towards the end of the breeding season, and a little before the 

 moulting of the contour-feathers commences, the green color of the de- 

 ciduous parts becomes more brownish, the rosette shrinks, becoming 

 rather pinkish of color, the edges of the different pieces are raised and 

 hard, and the grooves between them deepened. The distinction be- 

 tween the apical and basal parts of the lower mandible, which until 

 then has only been indicated by the difference in color, is now clearly 

 marked by a sharj) and deepened line. 



The first part to come off is the horny orlet, the process usually be- 

 ginning in such a way that the orlet bursts on the culmen where it is 

 quite narrow, and much more so than shown in Bureau's figure (fig. 4a, 

 pi. iii, B. S. Z. F., 1879), and from the culmen the two detached halves 

 of the orlet flake off* downward to the tomia, exposing the membranous 

 orlet, which is now covered with rows of about 1'""' long tvhite feathers. 

 Next comes the nasal cuirass, which, unlike the forujer, first detaches 

 itself along the upper margin of the nostrils. This nasal cuirass is 

 separated from the adjacent pieces all around by deep furrows, except 

 at two narrow points behind and in front of the nostrils, and in loosen- 

 ing, these narrow bridges break irregularly. Simultaneously the basal 

 part of the mandible is shed, first loosening along the feather border, 

 then along the tomia until it finally breaks at (he lately formed line 

 across the mandible. In Bureau's figure the horny selvedge is repre- 



