Mr. R. B. Sharpens Catalogue of Accipitres. 95 
the wings are closed, the rufous tint is, in consequence, much 
more permanent. Such a rufous-tinted specimen, obtained 
at Agra, is well described by Mr. Carleylle, as quoted at 
p. 228 of Mr, Hume^s ^ Rough Notes f and a beautiful speci¬ 
men in this phase of plumage was, subsequently to the pub¬ 
lication of Mr. Sharpens volume, presented to the British 
Museum by Captain Pinwill, who procured it in Bengal. 
This is the largest of the Bengal specimens noted in the fore¬ 
going list, and it is also the most rufous example of this spe¬ 
cies which has come under my notice. 
The white abdominal ocellations in S. undulatus are fre¬ 
quently somewhat obscured by the partial overlapping of 
the feathers; and when the plumage is much worn, they are 
rendered still less conspicuous by the fading of the darlc 
coloration of the surrounding portions of the feather. This 
remark equally applies to the adult specimens of the smaller 
race which Mr, Swinhoe obtained in Hainan, and to most 
others of similar size which I have seen from Burmah and 
Central India. 
Of all the specimens of the above races which have come 
under my observation, that in which the ocellations are the 
most conspicuous is a Formosan bird in the Norwich Museum^'. 
In another Formosan adult, and also in an adult male from 
Amoy, which Mr. Swinhoe retained in his own collection, the 
white spots are well marked, but not so remarkably as in the 
Formosan specimen at Norwich. The latter, however, is almost 
equalled in this respect by the example from Siam in the 
Norwich Museum, and also by Lord Tweeddale^s specimen 
from Khalee Gungee, in Assam. The two other examples 
from Assam, which I have examined, do not possess this 
peculiarity; neither have I seen any Indian specimens fully 
equalling in this respect the three specially marked examples 
above alluded to. The two most clearly spotted Indian spe- 
* This specimen is also remarkable for having the upper pale bar on 
the tail more distinct, and the lower one narrower (being only one inch 
in breadth) than is usual in Indian specimens of S. undulatus. I re¬ 
gret that I have preserved no notes as to the markings on the tail in other 
Formosan examples. 
