40 DR. C. G. SELIGMANN AND MR. S. G. SHATTOCK ON THE 
neck is well developed. The chestnut area of the breast contains 
so many eclipse feathers that the general appearance of this part 
is eclipse, but there are also a few winter-plumage feathers inter- 
mixed. The lower part of the breast, though generally vermicu- 
lated, shows a small number of eclipse feathers. The area on the 
abdomen plucked in July for castration is covered with feathers, 
darker than elsewhere and only partially vermiculated ; it might 
perhaps be described as ‘‘semi-eclipse” in character. A few 
eclipse and partially eclipse feathers persist in the flank. The 
tail contains two good curl feathers, and two others are ridging ; 
all are glossy, as well as the tail-coverts. ‘The bird was killed on 
this day. On the left side there was no trace of testicle, but on 
the right there was a nodule the size of a large haricot at the 
site of the gland, and below this a second, no larger than a millet 
seed. 
Summary of the Effects of Castration upon the Plumage. 
If we take the condition of No. 18 for the seven months 
succeeding the castration carried out early in December 1906, 
whilst the bird was in full male plumage, it appears that at the 
end of July 1907 (when the normal mallard has been in eclipse 
for some weeks) this bird still remained in almost full winter 
plumage. A careful water-colour drawing, made on July 23rd, 
1907, by My. Norman H. Hardy, shows a bird in complete 
winter plumage, except for some small areas of brown on the 
cheeks and round the eyes, and a slightly diffuse tinting with 
brown of some of the vermiculated feathers on the abdomen. 
This bird, however, had passed through its eclipse, and had 
reassumed the full male plumage by September 21st. In July 
1908, 2. e. the following year, it passed into eclipse like a normal 
bird. The record of No. 13, castrated during the latter half of 
September, when the bird was in full male plumage, shows a 
similar result; but in this case the delay in the appearance of 
eclipse feathers was even more marked than in No. 18; while in 
both the birds not only was the eclipse delayed, but it was 
incomplete. 
The behaviour of No. 48, castrated in March 1907, resembled 
that of No. 18, though the delay in the onset of the eclipse was 
not so marked ; in No. 7, the testes of which were removed in 
April 1907, there was comparatively little delay. 
Whilst there can be no doubt that the removal of the testes 
during the period of the eclipse does not retard the assumption 
of the proper male or winter plumage; castration carried out, on 
the contrary, whilst the bird is in full male plumage delays the 
appearance of the eclipse. This delay is not so much a positive 
delay as an abnormal persistence of the winter plumage. For if 
the winter feathers be plucked (7. e. artificially shed), the new 
feathers that replace them are of the typical eclipse kind. 
Text-fig. 6 shows a bird concerning which it is no exaggeration 
