556 ON THE PTERYLOSIS OF THE BARBETS AND 1oucaNS. [May 5, 
Of these species the last only (under the name of Bucconides roset- 
collis) is figured by Nitzsch, and, as I believe, inaccurately. 
Featier-tracts of Megalema asiatica. 
The right-hand figure shows the ventral aspect, the left-hand figure the dorsal. 
The accompanying drawings illustrate the pterylosis of Mega- 
lama asiatica. The drawings are copied from the late Prof. Garrod’s 
MS. As will be seen on comparing them with Nitzsch’s figures 
of Megalema armillaris, there are considerable differences, which 
of course may possibly exist between allied species. My own 
observations upon the first four species of my list and those of 
Mr. Forbes (in MS.) upon Megalema virens agree so entirely with 
each other and with Garrod’s sketch that I cannot but think that 
Nitzsch has fallen into error. 
The chief difference between us—it will be observed—concerns 
the spinal tract. In all the species of Megalema and Cyanops to 
which I have referred the posterior part of that tract is, as 
Nitzsch has correctly indicated, not in connection with the anterior 
fork ; but instead of being a straight band ending at the base of 
the oil-gland, it forks some little way in front of that gland and 
surrounds it. Another peculiarity of Megalema (not figured by 
Nitzsch in WM. armillaris) is a lateral band on either side which 
commences at the fork of the anterior part of the dorsal tract and 
runs down to a point about ona level with the middle of the poste- 
rior fork. This is quite distinct. from the more conspicuous 
femoral tract, excepting in WM. asiatica, where the lateral tract 
joins the femoral posteriorly. This lateral tract is figured by 
