1882.] ANATOMY OF THE TODIES. 445 



The triceps is Y-shaped at its scapular origin, and receives no 

 tendinous slip from the humerus. 



The construction of the syrinx may be best understood from an 

 inspection of the accompanying figures (p. 444), representing the an- 

 terior (A) and posterior (B) views of that of Todus dominicensis. The 

 trachea terminates below in an ossified bony box, formed of three 

 or four modified rings (probably bronchial) fused together, as is very 

 evident in the posterior view : as may also be seen there, the two 

 preceding tracheal rings are coossified with this box in the middle 

 line posteriorly, though in front they are quite free from it. The 

 box is deeply notched in front, a narrow pessular bar running back- 

 wards from the apex of the notch, forming a three-way piece. The 

 bronchial semirings succeeding the box have the normal character. 

 The lateral muscle of the trachea continues downwards to terminate 

 just on the upper limits of this syringeal box. The syrinx of Todus 

 viridis is constructed on exactly the same plan. 



Comparing these figures with those of Momotus lessoni given by 

 Garrod ', it will be seen that Todus differs from Momotiism its syrin- 

 geal box being deeply notched anteriorly, and much more perfect 

 posteriorly, the two parts being united by a pessular bar unrepre- 

 sented in Momotus. In fact it resembles that of the Alcedinidae or 

 Galbulidse rather than that of the Momotidse. The chief difference 

 from the former is that in the Alcedinidae the intrinsic muscle, 

 often very broad, passes down over the syringeal l)Ox to be inserted 

 on one or more of the movable bronchial semirings, instead of 

 ceasing before doing so, as in Todus. In Galbula there is a bony 

 box nearly similar to that of Todus, but with its sides more strongly 

 concave below, and produced downwards anteriorly into strongly 

 projecting points ; the lateral muscle only passes on to the loiver 

 margin of the box, thus stopping short, as in Todus and Momotus, of 

 the movable bronchial semirings. 



As regards the pterylosis, there is a strange oversight on the part of 

 Nitzsch ^ and Murie ^ as to the condition of the oil-gland, both these 

 observers stating it to be nude. In fact it is, in all the four species 

 of the genus, provided with a very well developed, and even long, 

 tuft of plumes, therein completely differing from that of the Momo- 

 tidse, in which the tuft is either altogether absent or quite rudimen- 

 tary^. In both T. viridis and T. dominicensis I count twenty 

 remiges, ten being secondaries ; Nitzsch and Murie give nine, having 

 apparently failed to observe the most proximal, smallest one. Niizsch's 

 figure of the pterylosis in Todus, having nearly certainly been con- 

 structed from an examination of the skins only, is not quite accurate — 

 it making the outer pectoral branch to the inferior tract too markedly- 

 divergent, and not showing the weaker lines of contour-feathers that 

 run from its apex to the hypopterum. The connexion between the 

 dilated part of the main pectoral tract as it passes on to the breast 

 and the patagial feathering is also made unduly important in his 

 figuie, this connexion in reality consisting only of some slight, scattered, 



' L. c. p. 428. 3 i. c. p. 679. 



* ' Pterjlography,' Ray Soc. ed. p. 88. * 6/. Garrod, I. c. p. 427. 



