Grus canadensis. 77 



With regard to the trachea of this species Mr. T. S. Roberts, in the 

 American Naturalist for 1880, p. 109, writes as follows : 



Contrary to the Btatement of Dr. Elliot Goaes, the trachea is convoluted within the 

 keel of the Bternum in G. canadensis as well as in 0. amerieana. This I have deter- 

 mined by the examination of four sterna of Q. oamadensis, three of which were prepared 

 by myself from birds positively identified as G. canadensis by the generally recognised 



external characters Although there is not such a radical difEerence as supposed 



by Dr. Coues, yet the two species are distinct in respect to their trachea and sternal 

 development. A glance at the drawings on page 57 and below will show this at once. 

 They are alike in so far the trachea enters the sternal keel in each. But in G. canadensis 

 the whole sternum is smaller and less stoutly developed ; the coils of the windpipe are 

 confined to the anterior half of the keel, and it is this position alone which is enlarged. 

 There are only about eight inches of windpipe in the keel, and twenty-seven inches in 

 G. amerieana ; the walls of the sternal cavity are much more imperfectly ossified than 

 in G. amerieana, where they are everywhere on the outside dense hard bone. On the 

 whole the entire conformation of the trachea and sternum in O. canadensis is much, 

 simpler than in G. amerieana. 



Although only the anterior portion of the keel is enlarged for the reception of the 

 trachea in G. eamadensis, yet the remainder of the keel is not solid bone ; but, instead, 

 is composed of two frail plates separated by a thin layer of bony meshwork. This 

 light structure of the posterior part of the keel is more pronounced in some specimens 

 than in others, reaching the greatest development yet seen in a sternum which shows 

 also the greatest convolution of the trachea. 



Convolutions or the Tkaohea in the Sternum or Gnus canadensis. 



The specimen figured by Mr. Roberts, an outline of which is here given 

 was selected by him as representing very nearly the average of the four. 

 The entire length of the trachea in Q. canadensis averages twenty-seven 

 inches. 



It is obvious that the extent of the convolutions of the trachea in 

 the various species of cranes depend? greatly on age. Mr. Yarrell, in his 

 well-known work on British Birds, gives drawings of the sternum of G. 

 communis at two difierent ages, and Mr. T. Roberts states, p. 112 : 



In an embryo Crane stated to be the Sandhill, which was just about to break the 

 shell when collected, the trachea does not enter the sternum at all and is perfectly 

 simple. But the anterior part of the keel, which is, of course entirely cartilaginous and 

 very diminutive is, comparatively speaking, much thickened, and a cross section of it 

 shows it to consist of two thin walls separated by a marrow-like substance. In this 

 feature of the sternum, we see the only indication in the embryo of the singular 



