118 CEYLON PEARL OYSTER REPORT. 



portion of the choanosome which is free from sand. The cortex (Plate VI., figs. 2, 3, 

 cort.) is about 0'13 millim. thick between the surface tubercles, but much thicker in 

 the tubercles themselves, which are composed exclusively of cortical tissue and 

 spicules. The greater part of the substance of the cortex is made up of asters of 

 various forms and their accompanying scleroblasts, but it also contains bands of fibrous 

 tissue running in various directions. Stout bands of dense fibrous tissue (figs. 2, S,f.b.) 

 also run vertically inwards from the cortex through the outer part of the choanosome 

 to the sandy layer, where they appear to assist in binding the sand-grains together. 

 The vertical spicule-bundles which run into the surface tubercles are also accompanied 

 by similar bands of fibrous tissue (fig. 2), and the same kind of tissue is also developed 

 in connection with the vents and the pore-bearing grooves. 



A noteworthy feature of the cortex is the presence of the very numerous, 

 approximately spherical, vesicular scleroblasts (fig. 3, scl.), about O'OIG millim. in 

 diameter ; each resembles a cystenchyme cell and encloses one of the smaller asters, 

 the ends of whose actines abut against the thin limiting membrane of the cell, or 

 perhaps sometimes project beyond it. Similar scleroblasts may be observed in 

 the choanosome. The outermost part of the cortex is convposed of small-celled 

 chondrenchymatous tissue ; the inner part is more or less fibrous, and between the 

 two Ave find collenchyma with stellate connective-tissue cells. 



The flagellate chambers (fig. 3, Jl.c.) are oval or nearly spherical, closely crowded 

 together in the choanosome and about 0'028 millim, in diameter where least contracted 

 by shrinkage. The state of preservation is not sufficiently good to enable me to make 

 out minute details very satisfactorily, but the chambers are apparently eurypylous. 



The inhalant pores are, as I have already observed, minute openings in the floor of 

 the pore-bearing grooves (fig. 2, m.p.g.). They are very numerous, and 10 or more 

 may be indicated in a single transverse section of the groove. From each pore a very 

 narrow inhalant canal runs vertically inwards and opens, with its fellows, into a 

 system of irregular crypts which lie beneath the thick floor of the groove and from 

 which wider inhalant canals take their origin and run inwards to the choanosome, in 

 which they sub-divide into smaller branches. Stout bands of dense fibrous tissue run 

 across in the floor of the groove from side to side, between the transverse rows of 

 inhalant pores and pore-canals. Probably, by the contraction of these bands ot 

 fibrous tissue, the prominent lips of the groove can be brought together and the 

 groove thus closed.* 



The main exhalant canals are only moderately wide and converge towards the vents. 

 Owing to the state of contraction I am unable to say whether there is a single large 

 vent or a group of small ones on each of the vent-bearing prominences. Sections 

 indicate that there may also be small vents between the conuli, and it appears as if 

 one exhalant canal sometimes opens through several small apertures. 



* A very similar arrangement is found in a very different sponge, Esperella mwrrayi, as described in the 

 Report on the ' Challenger ! Monaxonida, 



