DEVELOPMENT OF THE SKULL IN THE OSTRICH TEIBE. 135 



It is well ossified in the mature chick of the Rhea, and clamps the quadrate cartilage 

 very strongly ; it also helps the prootic capsule to carry the single condyle of the os 

 quadratum, whilst the hinder part of its lower margin forms a strong eave to the 

 tympanic chamber, but does not carry the fibro-cartilaginous tympanic ring ; this part 

 is untouched when the squamosal is slipped off in the macerated skull (Plate IX. 

 fig. 2). 



The free or Meckelian part of the mandible has its usual splints ; viz., three external, 

 the dentary, surangular, and angular (Plate IX . fig. 8, d.a.sa.) ; and two internal, the 

 splenial and the coronoid ; they are well developed at the time of hatching, but are free. 

 In the old bird the dentary keeps its distinctness, and the splenial does not make haste 

 to coalesce with its neighbours. 



There are no splints to the enfeebled hyoid arch, and it is instructive to see how sud- 

 denly the development of the second and third poststomal arches is stopped, whilst the suc- 

 ceeding ones exist only for a few days, even as simple cellular tissue. The descending 

 plate, which grows from the investing mass to form the piers of both the mandibular 

 and hyoid arches, is in all fishes subdivided so as to form an immense proximal piece for 

 the attachment of the segmented hyoid crus. But directly we come to the air-breathing 

 groups, with open nostrils and a tympanic cavity, then we see this mass giving most of 

 its proliferating tissue to the formation of the mandibular pier, the proximal part of the 

 hyoid arch being converted into the auditory columella or stapes. In mammals, where 

 the hyoid arch is not unfrequently attached to the skull, this is attained, not by a 

 resumption of the proper suspensorium, but by a direct downgrowth from the epiotic 

 region of the auditory cartilage. 



Bromceus, "A." 



My youngest specimen of the Emu is an embryo one week short of the full period of 

 incubation ; it is of the species Bromceus irroratus of Sclater and Bartlett. 



I have been very fortunate in obtaining five different stages of development of the 

 skull in the genus Bromceus ; some belonging to B. irroratus^ and others to B. novce- 

 hollandice. Moreover the study of this genus is very important, as, like the Dinornis, 

 it belongs to the Casuarine group, in which the mammalian characters become most 

 unmistakeable. Already in Bromceus, A., the ossific process is far advanced, and in 

 some respects further than in the young puUi of the Rhea, which had been incubated 

 eight weeks, or one longer than in this instance. The basioccipital (Plate XI. fig. 2. 

 b.o.) is much larger than in the Rhea, and is less evenly oblong, swelling out between 

 the entrance of the internal carotids (Plate XI. fig. 2, i.e.), and then wedging itself in 

 between the ovoidal basitemporals {ht.), and applying itself by a narrow but transverse 

 surface to the end of the basisphenoid, this synchondrosis being well seen below. A 

 narrow band of cartilage separates the last bone from its arches, the exoccipitals {e.o.), 

 and they have already reached their huge key-stone, the superoccipital {s.o.). Behind, 

 the opisthotic forms an upper lobe of the exoccipital (Plate XI. fig. 4, op.) united by 



