ANATOMY AND CLASSIFICATION OF THE AEBNICOLIDJl. 443 



might be gills, but Ehlers (1892) was not disposed to take 

 this view. No good description has been hitherto published, 

 and no histological details exist up to the present. 



On the first caudal segment of the large specimen lent to 

 us by the authorities of the Harvard Museum there exists 

 a pair of rudimentary gills. This is clearly shown by the 

 one on the right side being composed of four distinct 

 branches bearing short processes {End. Br., PL 24, fig. 31). 

 These two additional gills agree in position with the normal 

 ones. 



Just dorsal to each of the last four or five neuropodia of 

 this specimen, and of another from Captiva Key (PI. 24, fig. 

 33), there is a small tubercle (sometimes a slight depression), 

 distinct, from its size and colour, from the neighbouring 

 scattered papillae. The imperfect state of preservation of 

 these specimens prevents us from making an exhaustive 

 histological study of these structures, which agree in position 

 with the " Seitenorgane " of Capitellids. It is very possible 

 that examination of fresh or well-preserved material will 

 result in the discovery of a sensory epithelium on these 

 tubercles. There are also segmentally arranged outgrowths 

 on the tail (PL 24, figs. 31, 32). Two of these, as we have 

 already stated, are accessory gills ; and just as beneath the 

 last few pairs of fully developed gills there are peculiar 

 tubercles, so beneath these rudimentary gills on the first 

 caudal segment there is a tubercle, or rather two, one above 

 the other. The three structures, gill and tubercles, are 

 placed on a thickened annulus clearly corresponding (PL 24, 

 figs. 31, 32) to one of the chfetigerous annuli, as is further 

 shown by their relation to the internal septa (fig. 30). The 

 more dorsal of these tubercles on the left side was 3 mm. 

 long, and the one beueath this 2 mm. long. In section these 

 tubercles are found to be hollow vascular outgrowths of the 

 whole thickness of the body-wall, and though we have not 

 ascertained that the epithelium has specially sensory elements 

 or is ciliated, yet it is undoubtedly composed of columnar 

 glandular cells. These tubei'cles are cirriform structures, but 



