236 TRANSACTIONS LIVERPOOL BIOLOGICAL SOCIETY. 
rina has led me to support the view advanced by Tiede- 
mann, Muller, Agassiz, and much more recently by 
Ludwig, viz., that all the pores which traverse this plate 
open directly into the water-tube (Pl. XXXVI, fig. 1, mp). 
Hoffmann, supported by Greeff and Teuscher held on the 
other hand that some of the marginal pores open into the 
coelom, and others into the axial perihemal canal. A 
condition of things differing from either of the latter is 
presented by one of my series of sections, cut from the 
disc, 6 mm. in diameter, of a specimen of Asterias rubens. 
In this three or four of the madreporic pores opens directly 
into the lacunar system of the body-wall (Pl. XXX VL, fig. 2). 
In young specimens of Cribrella, 2 mm. in diameter, 
Durham (1) found but a single pore opening into the 
cavity of the axial perihemal canal, into which the water- 
tube also opens; and Cuénot (2) shows that in some species 
this state of things may continue throughout life. 
Communicating with the exterior through one of the 
madreporic pores in the specimen just alluded to is an 
elongated glandular structure, lodged in one of the lacunze 
of the body wall (Pl. XXXVI, fig. 2,9b). It is composed 
entirely of small rounded cells with comparatively large 
nuclei, and does not present any cavity or lumen. I am 
not at present able to offer an opinion as to whether this 
organ should be regarded as an example of the structure 
described by Greeff and Ludwig, and alluded to by Carpen- 
ter (8) as a diverticulum of the water-tube, or as an 
independent and perhaps hitherto undescribed structure. 
It is certainly independent of the water-tube, through 
lying in close proximity to it. Still more perplexing is 
its occurrence 1n only one specimen out of a dozen carefully 
examined. In its passage downwards to the oral aspect 
of the disc, the water-tube is supported by the free edge 
of a projecting fold formed by the junction of the rays on. 
