152 MR. E. T. NEWTON ON THE REMAINS OF A 



project inwards to the same extent, and consequently one sees but little curvature of 

 the inner side of the bone. The form of the trochlear surface towards the back is 

 different (I. c. pi. 6. fig. 5), the outer condyle being broad and having a supplementary 

 groove (compare this with our fig. 11, PI. XXIX.). 



The femur of Gastornis edtvardsii (Lemoine, pi. 1) has the articular head higher than 

 the trochanter, and the superior articular surface is in consequence placed obliquely, 

 thus presenting one of the most peculiar characters of the Croydon femur. On the 

 other hand, the trochanter projects more beyond the outer contour of the shaft, and the 

 anterior crest seems to be less prominent than in the Croydon femur. 



It is evident that the Croydon bird was more nearly related to the Gastornis than to 

 any other known form ; but it will, I think, be sufficiently obvious from the above 

 comparisons that it cannot be referred to any one of the species yet described, the 

 differences being too great to be lightly passed over. Among recent birds some nearly 

 allied genera, such for example as the Duck and Goose, do not present greater differ- 

 ences between their femora and tibio-tarsi than those which exist between Gastornis 

 and the Croydon bird, and it is possible, therefore, that the differences which have been 

 pointed out may eventually prove to be of generic importance. However, it is not 

 deemed expedient to introduce a new genus, but rather, until additional evidence is 

 obtained, to refer the English specimens to the genus Gastornis ; and in order that the 

 name of the gentleman, who has taken so much pains to secure these fossils, may be 

 associated with them, it is proposed to call this new Eocene bird Gastornis klaasseni. 



IV. — Comparison of Gastornis with Dinornis and other Extinct Birds. 



To what group of birds, living or extinct, is the Gastornis most nearly related"? The 

 interest surrounding this question has been fully appreciated by all who have written 

 about these Eocene birds, and it has been treated with especial care by Sir R. Owen, 

 and MM. Milne-Edwards and Lemoine, the more extended material which the last- 

 named gentleman has had at his disposal rendering his remarks of peculiar interest. 

 Seeing, however, that the Croydon remains have the lower end of the tibio-tarsus in a 

 much more perfect condition than any of the continental examples, and that this part 

 of the skeleton is very important for determination, it seems desirable to supplement 

 what has already been done by a further comparison. 



The gigantic size of these bones reminds one so forcibly of the New-Zealand Dinornis 

 remains, that, although their geological horizons are so different, and their geographical 

 positions so widely separated, one is led in the first place to take them for comparison. 

 Dinornis has the supratendinal bridge and canal placed nearer the inner side of the 

 tibia than they are in either of the forms of Gastornis, and these characters are very 

 constant in all the species of Dinornis, although the form of the bone varies. On 

 account of the curvature of the inner side of the tibio-tarsus and the projection of the 

 inner condyle, the Dinornis crassus is the species most like the Gastornis klaasseni; 



