Haast.— Oh Harpagornis. ^3d 



killed or drowned during that operation, because we know from experience 

 that all birds, if possible, retire to most inaccessible or hidden spots to die. 



This mode may also account for the extreme rarity of the remains of 

 Harpagornis, which I am sure flourished in considerable numbers during the 

 Moa Age. 



Mandible of Harpagoenis assimilis. 



Total length measured from point of symphysis along the ramus to 

 posterior end of articular process 4-75 inches, greatest distance between 

 articular processes, measured from the outward sides, 3 inches. 



From a comparison of the general form of this mandible with that of 

 the New Zealand harrier f Circus assimilis) it will be seen that it is some- 

 what narrower in proportion than that of the latter. In this respect it 

 resembles the mandible of the condor ( Sarcorhamphus grxjphus) and grifiin 

 vulture (Gyps fulvus). However, as several others of the vultures have a 

 broad mandible, this character is not of any generic value. But when 

 comparing the shape and size of the articular portion of Harpagornis with 

 that of Circus, the striking resemblance between both becomes at once 

 manifest. The articular part of the mandible in both is well excavated for 

 the mandibular end of the tympanic bone. 



The articular process has the same form in both. A pneumatic canal 

 perforates the surface of this articular process at its base. The articular 

 depressions for the insertion of the pterygoid muscles and for the two 

 strong ligaments uniting the tympanic and squamosal with the articular 

 part of the mandible are well excavated. The three portions of which the 

 ramus consists when extending forward from the articular end are well 

 cemented together, having a long and well defined articular surface for the 

 insertion of the temporal muscle in the central portion. 



The dentary portion is so well united on the right ramus that its 

 junction with the posterior portion is not well visible, whilst on the left side 

 the separation can be easily traced. From here the dentary curves 

 gradually down to the symphysis, which is well channelled, measuring 

 •68 inch across the upper portion. Total length of symphysis, "98 mch. A 

 number of nervo-vascular tracts are seen on the anterior part of the 

 symphysis, together with a series of canals on the lower side. 



RigJit tibia. 



This bone evidently belonged to a full-grown bird, although not aged. 

 The intermuscular ridges, as well as the fibular ridge, are well developed, 

 and all the characters desciibed in my former paper, as observed on the 

 tibia of H. assimilis of Glenmark, are well exhibited. 



Although somewhat shorter than the latter, both the proximal and 



distal extremities are a liitle largei'. 



SO 



