806 W. J. SOLLAS ON THE STRUCTURE AJTD 



canals on the way. These radiating canals arc smaller than the 

 exenrrent ones, and also differ from them in not being continuous for 

 any considerable distance, very few extending from the circumference 

 to the centre. Following one of them as it leaves the exterior sur- 

 face it is found to proceed for a shorter or longer distance inwards 

 and then to terminate in one of the longitudinal or excurrent canals ; 

 but other radiating tubes start afresh from the vicinity of the place 

 where it disappears, and, after proceeding further inwards and 

 crossing several excurrent canals on the way, terminate like the one 

 they have replaced ; and so by easy stages the central cloaca is at 

 length attained. Thus a succession of radiating canals maintains in 

 connexion the exterior of the sponge and the various longitudinal 

 canals. These radiating canals constitute the "incurrent" system; 

 and their external openings are the functional mouths or specialized 

 pore-areas of the sponge, and not " oscules," as is stated in works 

 on palaeontology. 



The interior of the stem is occupied by longitudinal canals in 

 direct continuation with those of the sponge-bod}*. Eadiating canals 

 are not obviously present ; but a number of small openings occur on 

 the exterior, from which, in some cases, superficial branching canals 

 racliately diverge and, after wandering for some distance over the 

 surface, become gradually lost. 



The interstices between the canals are occupied by an irregularly 

 reticulate, originally siliceous skeleton, the examination of which 

 must next engage our attention. 



Minute Structure. — To investigate this the phosphatized specimens 

 from the Folkestone Gault (Plate XXV. fig. 7) were in the first place 

 employed. These specimens had become somewhat worn by the 

 action of water before they were deposited in their latest " gise- 

 ment," and, according to the extent of the attrition they have un- 

 dergone, vary in colour from grey to black, just as may be observed 

 in the case of the coprolites of the Cambridge Greensand. In out- 

 ward form they are globular, ellipsoidal, spindle-shaped, and pyri- 

 form, varying in size from £ to l-o- inch in length and breadth, 

 with no constant ratio between the longitudinal and transverse dia- 

 meters. At one end a round scar or broken stump remains to indi- 

 cate the place of attachment of the pedicel now broken off; at the 

 other extremity is a plain surface in the centre, from which, in some 

 specimens, radiate for a greater or less distance down the sides 

 a number of low smooth rounded ridges, about ^V, inch broad ; 

 these undulate somewhat in their course, and anastomose with each 

 other laterally; between them the surface of the sponge is depressed 

 and minutely pitted. The central plain area indicates the place where 

 the interior excurrent canals originally opened on the surface, or, 

 perhaps, in a few instances, according to Mr. F. G. H. Price, F.G.S., 

 the summit of a cloaca now filled up with foreign matter. The radiating 

 ridges are the phosphatic casts of the exterior excurrent canals, and 

 the intervening depressions the skeletal interspaces, the pitting of 

 which ha3 been produced D3 T the removal, in solution or otherwise, 

 of the skeletal network exposed on the surface. 



