44 Mr. R. Kiikpatiick on 



work became visible. On breaking a crust, the appearance 

 presented was that of little square blocks of reddish-orange 

 jelly in white porcelain-like " cells " or pots superposed one 

 above the other, in from two to six storeys. 



On examining the first thin sections made from a specimen 

 decalcified by dropping alive into Flemniing's solution, I 

 realized that Merlia included siliceous as well as calcareous 

 elements in its composition. A specimen decalcified whole 

 presents a curious appearance, viz., of numerous closely 

 packed but separate moniliform cylinders, about a millimetre 

 or more in length, hanging down from a flat lamina. The 

 lamina and the bulk of the layer of beads in the plane just 

 below it compose the ectosome atid choanosome of a siliceous 

 sponge. All below the first layer is composed of hollow 

 cylindrical cell-masses separated by very deep constrictions, 

 and joined each to each merely by a narrow thread of tissue 

 whicii had passed through the central hole, which is often, 

 though not always, present in each tabula. The cells com- 

 posing these masses are large, elongated, usually pyriform or 

 fusiform cells applied like an epithelium, two or three cells 

 deep, to the surface of the cavities of the calcareous honey- 

 comb. A measured cell was 41 fj, long, 10 fi broad at the inner 

 end, and 3*5 fx broad at the outer end next the calcareous wall ; 

 the clear nucleus was 3*5 /x in diameter, and almost concealed 

 by the crowd of deeply stained spheroidal granules each about 

 1 ft in diameter. Above, it was stated that the bulk of the 

 uppermost layer of beads" was composed of ordinary sponge- 

 tissue ; at the base of each of these upper beads is a layer 

 of the large elongated cells, which rested on the upper surface 

 of the highest tabula. 



A surface view of a decalcified specimen shows, below 

 the ectosome, node-like masses of soft tissue joined to each 

 otiier by 5-7 radiating spokes; the nodes are the sponge- 

 mass(s which dip into the upper spaces of the honeycomb, 

 and the radii consist mainly of flagellated chambers lying 

 between the surface-tubercles of the calcareous framework, 

 the clear spaces between the radii being the gaps left by the 

 dissolved tubercles. To what extent other tissues enter into 

 the formation of the radii I have not yet discovered; but, in 

 places, there can be seen, below the flagellated chambers, 

 fusiform cells apparently in continuity with the cells on the 

 surface of the uppermost tabulae. 



These large granular cells appear to be calicoblasts formed 

 t'n situ, and not to be sponge " archgeocytes " which have 

 grown down into empty cavities — even to the fifth floor — of 

 the calcareous honeycomb. The term *' calicoblast " is here 



