ANATOMY OF THE SHOE-BILL, 693 



quadrate articulation is enormous and forms a powerful process 

 which runs downwards and forwards to meet and fuse with the 

 tip of the postfrontal. In a smaller skull, like that of the Dissura 

 (PI. LXXXII. fig. 1), the arrangement is practically identical, 

 but the postfrontal, although it meets the squamosal spur, is 

 much more slender, and the frontal factor does not appear to 

 i"each more than half-way down. 



In Tantalus (PI. LXXXII. fig. 2) the brain is relatively 

 slightly longer and larger. The squamosal spur is exactly as in 

 Xenorhynclius and Dissura, but two things have happened to 

 the postfrontal. In the first place it is much shorter, and does 

 not reach the squamosal spur. In the second place the triangular 

 base is very much wider and is dee]3ly notched in front, with the 

 result that it has an anterior and smaller portion corresponding 

 exactly to the anterior margin in all the other birds I have been 

 describing here, but certainly with no squamosal factor, and a 

 longer portion running down towards the point of the squamosal 

 spur, corresponding with the posterior part of the postfrontal 

 in other birds and certainly consisting chiefly of alisphenoid and 

 squamosal factors. 



The condition in Cancroma (PI. LXXXII. fig. 3) can now be 

 followed easily. The brain is still larger I'elatively ; the squa- 

 mosal spur is reduced, and the separation between the two parts 

 of the postfrontal, only just apparent in Tantalus, is well marked. 

 The purely frontal, anterior portion is the stouter of the two. 

 In Ardea (PI. LXXXIII. fig. 1), where again the brain is still 

 larger, the squamosal spur is relatively rather small, but the 

 separation between the two parts of the postfrontal is very 

 wide indeed. 



As this matter appears to be of some interest, and as I 

 have not found it discussed, I shall continue the description 

 outside the immediate relatives of Balceoiiceps. The Pelican 

 (PI. LXXXIII. fig. 2) shows a further extension of the series. 

 The squamosal spur is as in Ardea, but the two portions of the 

 post-frontal are even further separated, and the posterior of the 

 two is reduced to a mere tubercle, intermediate in position be- 

 tween the squamosal spur and what would normally be taken to 

 be the postfrontal. In Plotus (PI. LXXXIII. fig. 3), which has 

 a very large brain indeed, the squamosal spur is small, there is a 

 mere stump to represent the posterior portion of the postorbital 

 process and this is actually nearer the squamosal spur than 

 it is to the anterior representative of the postfrontal. 



I do not suggest that the series, as I have arranged it, is 

 phylogenetic, but it is a striking example of the diSerences that 

 identical morphological material may exhibit in allied birds, and 

 a warning against the hasty drawing of conclusions as to sys- 

 tematic position from the comparison of one or two presumably 

 allied birds. So far as this point goes, Balcenieeps and Scopus 

 stand together as birds with small brains, with the squamosal 

 spur sliglit and the postfrontal process simple. Storks form a 



46^ 



