No. 2.] SHUFELDT ON NORTH AMERICAN TETRAONID^. 331 



of tlie Tetraonidcc. They termiuate by dilated extremities of nearly simi- 

 lar shapes, Cupidonia being an exception ; the ends of the apophyses 

 of the sup erior pair in this bird being rounded posteriorly (Plate XI, 

 Fig. 82 ; see Plate XIII, Fig. 91, Lagoims, for the common pattern). These 

 processes arise from a common stem, and their shafts are flat internally, 

 with a raised ridge extending the entire length externally. The "body " 

 of the sternum is, as a rule, very narrow, and notably concave anteriorly, 

 becoming nearly flat behind, where it is produ ced beyond the keel for 

 a greater or less distance. 



The manner in which this part terminates varies in the dififerent 

 Grouse. 



In Centrocercus it is nearlj^ square across ,• in Lagojnis rouudlj^ notched 

 in the middle line, as it is in Bonasa ; in Tetrao canadensis it is broadly 

 cordate; while in Tetrao ohscurus, Ctqndonia, and Pedieecetes it is dis- 

 tinctly cuuiform. The body is very narrow in Bonasa, approaching 

 the Odonfophorincc, where it seems really to be nothing more than a 

 good ribbon»like finish to the superior border of the keel. In these 

 birds, too, we are struck with the double carinal margins anteriorly 

 formed by the projecting ridges, and the long spicula-form costal pro- 

 cesses that extend nearly half-way ui^ the shafts of the coracoids. 



So much do the sterni of the Grouse resemble one another in spe(;ies 

 of average size that it would puzzle one not a little to tell them apart 

 if they were separated from the skeleton, and we were not allowed to 

 examine them in connection with other diagnostic features of the osse- 

 ous parts of the species to which they might belong. 



Cupidonia and Pedioecetes are particularly alike, but the former could 

 be recognized by its superior xiphoidal processes, Bonasa by the nar- 

 row body, Centrocercus by its size in the larger specimens, and so on. 



We will still continue to consider such of the vertebral column as is 

 confluent in the old bird, or rather such vertebrte as become confluent 

 and are more or less embraced by the ossa innominata, as the sacrum, 

 and composed of sacral vertebriie, attempting to make no such divisions 

 as Professor Huxley did, in his Anatomy of Vertebrated Animals, of 

 this compound bone. 



There are sixteen of these segments that are to be so reckoned in 

 Centrocercus, but it is only in the " bird of the year" that they can be 

 counted with anything like accuracy, and even then great care must be 

 exercised, and various pelves examined and comj)ared with the younger 

 birds at diflereut stages and ages. 



The first sacral vertebra i)ossesses free i)leurapophyses, whose luema- 

 pophyses do not reach the costal borders of the sternum, but articulate 

 in a manner to be described farther on. Regarding the pelvis from be- 

 low iu Centrocercus, we note that the anterior four sacral vertebrte have 

 their combined par- and diapophysial processes thrown out as braces 

 against the expanded anterior iliac wings. After this the ilia change their 

 form to accommodate themselves to the basin of tbe pelvis, which they 



