Sponge-fauna of Norway . 437 



tlilu skin to lie immediately upon the extended rays of bifur- 

 cated ternate spicules, which, regularly overlapping, map out 

 the skin into a number of triangular spaces, most of which 

 are singly perforated by a circular pore 0'004 to O'Ol of an 

 inch in diameter. This arrangement is to be seen on both 

 upper and under halves of the sponge ; but in the equatorial 

 recess it is replaced by another. There the skin is separated 

 to a greater extent from the mass of the sponge by the under- 

 lying vesicles of the canal-system ; it is not supported by the 

 rays of furcate spicules, but fine threads, crossing it trans- 

 versely, strengthen, support it, and divide it into a number of 

 more or less oval areas, each of which is perforated by a 

 great number of closely-set pores, which reduces it to a fine 

 network (see Kent, xii. pi. Ixvi. tigs. 3, 4). Of spicules this 

 cribriform floor of the equatorial recess contains chiefly 

 minute spinispirules, and only occasionally quadriradiate 

 stellates. 



On cutting the sponge across, one sees a greyish mass 

 enveloped in a thin skin, but without a cortex, traversed by 

 fascicles of spicules and a great number of vesicles ; the 

 vesicles lie in rows, longitudinally and radiately disposed. 



The Canal- system. — The pores have been already described 

 as distributed generally over the whole surface of the skin, 

 including its conical extensions over the roots of the anchoring 

 fibres. They occupy the triangular spaces in the skin 

 mapped out by the overlapping rays of the furcate spicules 

 below it ; usually there is one pore to each space, rarely two. 

 In the equatorial recess the skin is divided into oval areas by 

 fibrous strings, and in these areas is so abundantly perforated 

 by pores as to be converted into a sieve-like net, in just the 

 same manner as described by Schulze in so many Cero- 

 spongife, and by myself in Tetilla^ as likewise occurs in many 

 Esperue, and probably also in a vast number of other sponges. 

 The curious way in which this cribriform poriferous mem- 

 brane occurs in a recess, while the rest of the sponge is per- 

 forated by single pores, reminds one forcibly of similar 

 arrangements in some of the Esjjerice. The pores, whether of 

 the recess or the general surface, lead directly into spherical 

 or ellipsoidal chambers or vesicles beneath the skin, the first 

 of a series of vesicular dilatations Avhich constitute the in- 

 current canal-system (PI. XVIL fig. 6). For in this sponge 

 the canals are not canals in the ordinary sense of the word, 

 t. e. not continuously open more or less tubular channels, but 

 a succession of vesicles, which seldom open into each other 

 except by narrow sphinctrated orifices. Thus, in a linear series 

 of vesicles representing a canal in other sponges, every 



