1906.] RESPIRATORY SYSTEMS IN THE OPHIDIA. 527 



gntter, which extends beyond the commencement of the liver. 

 This band seems to me to present a stage which I have not yet 

 met with in any other Snake, lying between the more usual tracheal 

 gutter and the rudimentary seam, which in a few Snakes (e. g.. 

 Python sehce and, which is more to the point at the prasent 

 moment, Tarbophis) seems to be all that is left to represent the 

 continuation backwards of the trachea within the lung. 



Tarhopliis obtusus closely resembles Erythrolamprus msculapii 

 in the structure of its respiratory organs, and both Snakes are 

 placed by Boulenger in the same subfamily of Opisthoglypha. 

 It has a very distinct tracheal lung. This tracheal lung, how- 

 ever, as in Erythrolamprus, is of very modest dimensions. 

 Although it presents the characteristic honeycombed structure 

 only, two or three of the cells intervene between the free ends of 

 the tracheal semirings. The tracheal lung is, in fact, very 

 naiTOw. The trachea is continued for a short distance into the 

 lung proper, and shortly before its termination gives off a 

 bi'onchus to the rudimentary left lung. From the point where 

 the tracheal semirings apparently terminate in the interior of the 

 lung a fibrous seam is continued onwards, which recalls at once a 

 very similar seam or ridge in the lungs of the Lacertilian genus 

 Tehis * as well as of Python sebce, and of ErylhrolaTnprus just 

 described. 



In Leptodira hoiambceia, which belongs, lilce Erythrolamjjrus 

 and Tarbophis, to the Dipsadomorphine family of the Opistho- 

 glypha, the lungs are not widely difierent from those of Ta7-bop>his. 

 The advantage of examining a recently dead specimen is chiefly 

 seen in the ease with which the vascular can be marked off from 

 the anangious region of the lung. The minute portion of the lung 

 which in this Snake (and others) serves as an efficient breathing- 

 organ, contrasts with the large extent of the functional lung in 

 such a snake as Bitis arietans. There are about 10 inches of 

 vascular lung in the Puff-Addei- and about 1 k inches in Leptodira 

 hotambceia. Nor is this enormous difference to be explained by 

 relative bulk. Leptodira possesses a rudimentary left lung which 

 in spite of its small size is red, and thus entirely vascular. I have 

 had to remark in other cases upon the vascularity of the rudi- 

 mentary lung in Snakes. Assuming that the point where the 

 rudimentary lung arises marks the line of division between the 

 thoracic and tracheal lungs, Leptodira may be said to possess a 

 functional tracheal lung. As there is no headward extension of 

 the larger lung, it is not possible to fix the boundary of the 

 thoracic and tracheal lung otherwise. There is, however, an exten- 

 sion of lung-tissue further forward than the very short tract lying 

 in front of the bifurcation of the trachea which is vascular. The 

 tracheal lung is very wide. The membranous interval between 

 the dorsal ends of the tracheal semirings is very much wider than 

 the diameter of the trachea itself. This Snake, therefore, in the 



* Milano, Zool. .Jalirb. Abth. f. Anat. vii. 



Proc. Zool. Soc— 1906, No. XXXYI. 36 



