172 Walter Rothschild. 



fossil or in Struthio, while in Casuarius and Dromaeus they are 

 shorter in proportion to their width and differ in other respects. 



Next to the structure of the columnar (mammillary) layer 

 the arrangement of the pore-canals and their openings is of the 

 greatest systematic importance. In the fossil the pore canals 

 seem to be for the most part simple and run straight to the outer 

 surface, where they open either singly or in small groups 

 (sometimes in pairs) in very slight depressions. A few of the 

 openings are very much larger than the rest. In Aepyornis the 

 arrangement is very similar except that there is a tendency for 

 the pores to open in irregular rows of four or five, at the bottom 

 of slight grooves. In Struthio there is great difference in the 

 arrangement of the pore-openings in the different species. Thus, 

 in Struthio camelus the surface of the shell is smooth and the 

 pores open singly or in ill defined groups which are not situated 

 in a distinct depression in the surface. In Struthio australis the 

 pores open in small groups at the bottom of slight depressions and 

 are usually linked with one another by shallow grooves. In 

 Struthio molyhdo'phanes they open in close set groups at the 

 l^ottom of shallow but well defined depressions forming as it were 

 small cribriform plates: each group seems to be the result of the 

 branching of an originally single pore-canal forming a small 

 sieve-like area. Struthio camelus resembles the fossil most 

 closely in the arrangement of the pore canals, indeed this species 

 perhaps differs less from it and from Aepyornis in this respect 

 than it does from the other species of Struthio. In Bhea the 

 pores usualh^ open in pairs, sometimes in threes and fours at the 

 bottom of short straight grooves which tend to run parallel with 

 one another; in Dinornis the arrangement of the pores is 

 astonishingly like that in Rhea. In Casuarius and Bromaeus 

 the roughness of the surface of the egg makes it difficult to 

 observe the arrangement of the pores. In the former the aper- 

 tures seem to be of irregular shape and uniformly scattered over 

 the surface: in the latter, while the openings are similar, there is a 

 tendency to arrangement in irregular lines. 



To sum up it would appear that in these two frag"ments of 

 egg-shell we have evidence of the former presence in Northern 

 Africa of a hitherto unknown bird which laid an egg considerably 

 larger than the largest produced by any modern ostrich, or than 

 that of the fossil Struthiolithus chersonensis and only inferior in 

 size to that of Aepyornis titaU' Its affinities, so far as they can be 

 made out, seem to be rather with Aepyornis and Struthio than 

 with the other Ratites: with Struthio the relationship was 

 probably very close. 



It, to be hoped that further search will lead to the discovery of 

 some actual remains of this bird itself, for without these no satis- 

 factory knowledge of its affinities can be attained. Moreover it 



