204 SHOULDER-GIRDLE AND BREAST-BONE. 



(fig. 20). The vertebral ribs (figs. 18 20, v. r.) form the most exquisite jointed armour ; each rib 

 sending backwards a thick, strong flap, which overlaps the succeeding bar : this continuous plate 

 is homologous to the segmented plate which grows from the Bird's rib ; but it has a much greater 

 vertical development. 1 



The Ant-bear and Tamandua are remnants of the third Family of this division of the Edentata ; 

 the scapula (Plate XXIII, fig. 21, Myrmecophagajubata, one-third natural size) would be, if its 

 base were cut away, a very even semi-oval in form ; for there is no angle between the supra-scnpular 

 and pras-scapular margins, and the latter is continuous with front of the large coracoid. All along 

 this supero-anterior edge there is a headland of soft cartilage (s. sc., p. sc., cr.) ; and a tract of 

 cartilage separates-the coracoid from the pra>scapula ; and these two regions arc also more clearly 

 mapped out by the large, oval, coraco-scapular fenestra" (c. s. f.), as in the Sloths. The adze- 

 shaped coracoid (cr.) has its own ectosteal sheath ; this bone helps to form the glcnoid cup (gl.). 

 The area of the " supra-spinous fossa " is about equal to that of the " infra-spinous ;" the meso- 

 scapular " spine " is moderate in height and thickness, and sends an overlapping process back- 

 wards, just below its middle ; it is elegantly curved backwards in its fixed part, and then its free 

 part, the acromion (ac.), has a semilunar form, the curve being in the opposite direction, or forwards. 

 At its root this large acromion sends backwards a rudimentary " metacromion " (m. ac.). The 

 post-scapular region (sc.) is nearly equally divided by a second spine, which has the same back- 

 ward curve as the meso-scapular spine, and converges towards the latter, downwards. The 

 posterior (= inferior of Anthropotomy) margin is elegantly arcuate, but short; behind, it has a 

 small projection, which gives it a sinuous outline : the long sweep of the soft margin of the scapula 

 makes this side to be as much inferior in the quadruped as it is in the erect human creature. 



There are no clavicles nor clavicular correlates in the Myrmecophagina3 proper ; but the 

 Sternum is full of interest to the morphologist. The pra>sternum (Plate XXIII, fig. 22, p. st., 

 Tamandua bivittata, three-quarters adult, natural size) is a thin transversely elliptical plate, which 

 somewhat overlaps the second perfect costal arch below (s. r. 2). Its endosteal deposit has reached 

 the surface and acquired an ectosteal layer, but the whole periphery of the plate is still soft ; the 

 ends of this narrow oval are somewhat squared, to articulate with the sternal moieties of the first 

 thoracic arch (d. r. 1, s. r. 1). There is a Reptilian looseness as to number here, as in the Sloths 

 (these types agree also as to the very reptilian coracoid and its " fenestra") ; for there is an eighth 

 cervical vertebra 2 (c. v. 8) ; its lower part is pointed, and free, and just touches the mamibrium ; 

 it is not ossified at this stage. The meso-sternal segments take on a very remarkable character ; 

 they are short-rounded above, saddle-backed, like the "centrum" of a vertebra; and, together, 

 imitate in an inverted manner the vertebral column of a Mammal very wonderfully, for each 

 segment sends downwards a deep, long, clubbed process, like a vertebral spine. There is to 

 each segment of the meso-sternum one principal " pleurosteon," evidently developed endosteally 

 like that of the Sloth and Armadillo, but becoming very perfect afterwards ; to this main piece 

 there is the superaddition of two pairs of upper epiphyses one pair at each end ; and two pairs 

 of much smaller centres in the unossified inferior process (Plate XXIII, figs. 24, 25, m. st., u. ep., 



The appendages of the ribs are nearly parallel with the ribs themselves in Serptntarlus : in the 

 adult bird they become ancbylosed, and then answer, very exactly, to these flaps of the Tamandua. 

 I am not sure whether this has been noticed before. 



