428 RECTAL GILLS IN LARViB OP ANISOPTERID DRAGONFLIES, 



An examination of Figs. 3, 5, will at once suggest the strong 

 probability that the gas is obtained hy a simple process of diffu- 

 sion through the slender wall of the gill into the underlying cajnl- 

 laries The points in favour of this are : - 



i. The thinness and delicacy of the cuticle. 



ii. The alteration of the undifferentiated rectal epithelium 

 (Fig.66) into a thin syncytium, with elimination of cell-bound- 

 aries. 



iii. The extreme fineness of the walls of the tracheal capillaries, 

 and the complete absence of the spiral thread. 



iv. The fact that the capillaries, instead of running in a 

 definite blood-channel, are embedded in the protoplasm of the 

 epithelial syncytium, thus becoming arranged as close to the 

 external surface of the gill as is possible under the circumstances. 



V. The fact that all the capillaries form complete loops. For, 

 if oxygen were extracted at any given point, or by any given 

 localised organ, we should expect the capillaries to end in that 

 point or organ. The fact that they form complete loops shows, 

 however, that the extraction must take place at all points along 

 the circumference, and pass inwards down both arcs of the capil- 

 lary, to its junctions with the larger efFerents. in other words, 

 there is no circnlatioii of gas in the capillaries in a definite 

 direction round the circumference, but simply a passage of gas 

 from all points inwards towards the efferent trachea3. 



Some such solution as this had evidently been in the minds of 

 most of the earlier students of the subject, until certain appa- 

 rently insuperable difficulties were pointed out. On that account, 

 the Diffusion Theory, as I propose to call it, became unpopular, 

 in spite of Lowne's advocacyO), until Ris again recently reviewed 

 it — without, however, clearing away the initial difficulty of the 

 problem As I intend, in this paper, to support the view ad- 

 vanced by Lowne and Ris, it will be as well if I make their 

 position clear by quoting a translation of the latter 's own 

 wordsiS):— The question becomes more difficult in the case of the 

 Closed Tracheal System, of which our larvae furnish a classical 

 example. What is here wanting is the vis a tergo of the inspira- 

 tion-organ, and we are driven to this, to assume purely chemical 



