Benham. — On the Shin of Notornis. 149 



reddish -pink — the same colour as the greater part of the 

 beak — without any trace of orange, such as is shown in the 

 figure of the second edition, while the colour is a brighter red 

 than in that of the first edition. 



The shape of the bill in the figure of the second edition is 

 entirely wrong so far as the curvature of the lower jaw is con- 

 cerned. It is represented as being concave downwards, giv- 

 ing a curved sharp point to the beak. On the contrary, the 

 curvature is slightly convex downwards, so that the angle at 

 the tip is greater ; and the beak is blunter. The woodcut on 

 page Ixxv. is accurate. 



The shape of the nostril is variously described. We find 

 this statement on page 91 : " Nostrils oval " ; and on page Ixxv. : 

 " Nostrils round." As these statements occur in the diagnosis 

 of the genus it is important to put on record the real facts. I 

 find the nostril very nearly circular ; the longitudinal diameter, 

 however, just exceeds the vertical, the measurements being 

 j^g in. and y^e ^''^- respectively. 



With regard to the shape of the tarsi, these are not cylin- 

 drical, but laterally compressed. The account of the scales 

 given in the text (2nd ed.) does not agree with the figure. 

 As a fact, the text is correct and the plate wrong. Each of 

 the sides of the tarsus, as well as the front, is clothed with a 

 series of transversely elongated scales, the three series being 

 separated by a series of much smaller and more irregular ones, 

 posteriorly and antero-laterally. I counted fourteen of these 

 scales on the front series. 



The number of scales or scutella on the middle toe, includ- 

 ing that which ensheaths the base of the claw, is 26 (Buller 

 gives 23) ; on the inner toe, 18 (instead of 15) ; on the outer 

 toe, 25 (instead of 21) ; and on the hinder toe, 7 (instead of 

 5). No doubt some variation in these numbers is to be ex- 

 pected, but I have thought it worth while to give them for 

 the present bird. As to the shape of the claws, the artist 

 has drawn them too pointed ; the tip, like that of the beak, is 

 blunt. Moreover, the hind toe is placed too high up the foot 

 in the figure (2nd ed.) — as a fact, it rises from the same 

 general level as the other three toes, as shown accurately in 

 the figure in the first edition. 



The shape of the wing is much less definite and less com- 

 pact than would appear from the figure. It is, in reality, 

 more rounded posteriorly as it lies against the body. I may 

 mention here that the wing is provided with a spur. 



Turning now to the general colouration of the plumage, the 

 following points may be noted. A comparison of the plate in 

 the second edition with the bird itself, freshly killed, and with 

 our oil-paintings, done from the Dresden specimen when it 

 arrived in Dunedin, shows very considerable differences. The 



