BY O. W. I>E VIS. 



25 



example of the femur of Dromornis than anything else. It is 

 therefore a matter of regret that a cast of th(3 Dromornis femur 

 cannot be placed, for purposes of comparison, l)eside the fossil 

 under examination. The original is in the Sydney Museum, and 

 Mr. Haswell courteously regrets that he has not a facsimile at 

 command. However, in the figure given of it by its describer, we 

 have a sufficient corrective to any bias in favour ot' its identity 

 with the fossil before us This bone is in much the same peculiar 

 state of mineralization as the great majority of the Darling 

 Downs fossils. It consists of somewhat more than the upper 

 third of the left femur, minus the upper part of the head (h) 

 and trochanter (t.ma.) These have been lost by abrasion 

 while projecting above the surface of the creek bed. It measures 

 five inches in breadth anteriorly from the head to the lower end 

 of the great trochanter, four inches posteriorly from the head to 

 the upper angle of the trochanter, and five inches externally 

 across the trochanter. These are precisely the measurements 

 yielded by a specimen of Dinornis crassus, Owen. The shaft, at 

 its place of fracture, is rather more rounded than in D. crassus, 

 measuring two inches two lines in breadth, and half an inch ten 

 lines in fore and aft thickness. Its section is a full, irregular 

 oval, as in D. elephantopus, very dissimilar to the pure oval of 

 Dromornis, but somewhat less unlike that of the emu's femur, 

 in which the inner side is rather more convex than the 

 outer. The base of the head presents a strong annular con- 

 striction (a.c.) which, as in Dinornis, renders the head quite dis- 

 tinct from the neck (n.), and contrasts it with the subsessile 

 heads of Dromoius and Dromornis. The depression for the 

 ligamentum teres is in the emu scarcely perceptible, the ligament 

 in this bird being attached to the centre of the almost con- 

 tinuously convex articular surface of the head ; behind it a 

 smooth concave tract slopes down towards the pneumatic 

 foramen. In Dinornis crassus there is a subcentral pit nearer 

 to the hinder part of the periphery of the head, and excavated 

 to a moderate depth. In the fossil (L. ter.) it is in a similar posi- 

 tion, but deeply sunken, and its hinder edge is raised into a 

 rough ridge. In neither D. crassus nor in the fossil is there a 

 concave slope behind the ligament pit. The neck (n) of our 

 subject is distinctly longer and narrower than in 1). crassus, 

 and consequently more divergent in both respects from that of 

 Dromoeus. The neck at its junction with the epitrochanterian 

 surface (ep. t.) is far more deeply hollowed chan that in the 

 emu, and therefore conspicuously unlike that of Dromornis, in 



