172 MR. R. J. TILLYARD ON THE RECTAL BREATHING-APPARATUS 



appear to be in the nature of rudimentary supports developed in conjunction 

 with the main efferent trachese and their accompanying hypobranchial tissue. 

 But, uninteresting as the structure o£ the basal pad in Dendroceschna may 

 be, it nevertheless allows us to make certain very obvious deductions. The 

 first is that it is almost certain, with this example before us, that all basal 

 pads arose merely as swellings in an originally uniform epithelial syncytium. 

 The second is that they were probably called forth to act as supports for the 

 bases of the projecting gills, and play no part in the physiological process 

 of the extraction of oxygen from the water in the gill-basket. The third 

 deduction is that the basal pads never i^ossessed separate cell-territories, and 

 therefore that it is useless to try, as Sadones has done, to indicate such 

 territories around each nucleus. 



I should like here to digress in order to institute a comparison between ' 

 the basal pads of the gill-basket and the well-known fibrillar columnar 

 epithelium which supports the " fields ^^ bearing the teeth in the gizzard of 

 Odonate larvse. There also the need is for a support, and the need is supphed 

 by the specialization of the epithelium underlying the chitinous cuticle. In 

 the gizzard, however, the epithelium is formed of separate cells. Hence we 

 find the columnar support evolved from it also formed of separate cells. 

 These cells, however, agree with the syncytium of the basal pad in having 

 their nuclei enlarged, and in showing a distinct fibrillar structure and a 

 turgescent appearance. The function of these enlarged areas, whether in 

 the gizzard or in the gill-basket, is clearly that of a cushion or support for a 

 structure which might otherwise collapse. 



To return to the basal pad. Apart from those of the Implicate Type in 

 the Duplex System, the basal pads of all forms examined by me show a high 

 degree of specialization. Let us take the pad found in Austrogomplius 

 ochraceus (Plate 22. fig. 22). At the bases of both main longitudinal folds 

 and cross-folds, lying on the outer side of the gill-wall, one finds these very 

 prominent pads in all sections. They lie not only along the projecting gill- 

 fold itself, but also partly along a small portion of the circumference of the 

 rectum. In many sections the pad may be seen to be continuous between a 

 main fold and an adjacent cross-fold. In the enlarged figure of a basal pad 

 selected from the section diagrammatized in Plate 22. fig. 22 (fig. 25 of the 

 same Plate), representing the pad marked * in the diagram, the pad will be 

 seen to rear itself high up on one side against the wall of the main fold, 

 while another and shorter portion bends upwards against the base of the 

 cross-fold. Between, a portion of the pad lies on the rectal circumference. 

 If this pad be followed through a number of sections, it will be seen that the 

 part supporting the base of the cross-fold increases as the cross-fold comes 

 to be cut more centrally ; afterwards it decreases as the section of the cross- 

 fold decreases. 



The pad is continuous at both ends with the epithelial syncytium, and 



