144 MR, K. J. TILLYARD ON THE RECTAL BREATHING-APPARATUS 



In order to understand the somewhat complicated manner in which the 

 tracheae are distributed, let us look at the diagram in Plate 22. fig. 22. 

 We see from this that the primary efferent tracheae approach the circum- 

 t'erence o£ the gill-basket at the positions 1, 3, 5, 7, 9, 11 on the clock-face, 

 though the main gill-folds, as already stated, lie in the positions 2, 4, 6, 8, 

 10, 12. The secondary efferent tracheas, formed by the bifurcation of any 

 given primary trachea, enter the basket therefore slightly to right and left 

 of one of the positions represented by an odd number on the clock-face. 

 Nor do the two secondary tracheae from any given primary trachea enter at 

 the same transverse level — one is always a little in front of the other. This 

 is shown in the diagram by representing these tracheae by alternate complete 

 and dotted lines, the dotted ones not being at the same level as the 

 complete ones. 



After giving off branches to the cross-folds, the secondary tracheae run on 

 to enter the main folds at their bases. This the}^ do by branching into 

 a series of gill-tracheae, which spread out at different levels, and then bend 

 sharply inwards to enter the main folds. Each series of gill-tracheae forms a 

 small fan or pencil of rays, and enters the main fold at a convexity, or crest, 

 of the undulations of its base-line. 



It follows from this that each main longitudinal fold receives tracheae from 

 two sources which supply it alternately, viz., a series entering from the right 

 and one from the left. In the case of the mid-dorsal longitudinal gill-fold 

 (position 12) the series to the right is a set of gill-tracheae given off by the 

 series of secondar}^ efferent tracheae which form the more dorsal or left-hand 

 branchings of the upper series of primary efferents given off by the right 

 longitudinal dorsal tracheal trunk ; while the series to the left is derived 

 correspondingly from the upper series of primary efferents given off by the 

 left dorsal trunk. Thus the gill in position 12 receives its tracheae alter- 

 nately from the positions 1 and 11, or thereabouts. A similar arrangement 

 holds for each of the other five main gill-folds. 



The main longitudinal gill-folds are older, both phylogenetically and onto- 

 genetically, than the cross-folds (see p. 187), and are the only gills present in 

 the rectum in the newly-hatched larva. 



The cross-folds are never very large, nor do they project far into the 

 rectum. They are always very crumpled and irregular. As they are Avell 

 supplied with tracheae, they certainly aid in respiration. Probably the 

 twelve sets of cross-folds taken together scarcely extract more than 50 per cent, 

 of the amount of oxygen extracted by the six main folds in the same period 

 of time. 



The principal interest of the cross-folds is their relationship to the gills in 

 the Duplex System, to be dealt with later (p. 150). 



We shall now turn to the study of the two types of gill known to exist, 

 which belong to the Simplex System. These are (1) the Undidate Type, 



