ox MERLIA NORMANT. 695 



sponge?. Yet in this instance I may sav, in view of my own 

 error, that truth is stranger than fictiou. 



It seems to me unlikely that Merlia could form anything 

 else tli;in a thin crust, i.e. tliat it, or forms allied to 

 it, could have formed a laminate or branched growth, for 

 instance, but sometimes the living sponge grows over a dead 

 crust of its own kind. 



Wiiat has happened in one instance may have taken place 

 on other occasions. Amoebocytes of Tetrasonid sponges 

 may have become calcocytes under favourable conditions at 

 any epoch, and these calcocytes may have formed calcareous 

 skeletons, just as Merlia has done. 



(8) Summary. 



Merlia is a vermilion coloured incrustiug Monaxouellid 

 sponge belonging to a new sub-family — Merlinas — of the 

 Haploscleridie. Large granular amoebocytes (calcocytes) have 

 constructed a basal calcareous skeleton formed of vertical tubes 

 divided up by horizontal tabula3. The tubes are built up of 

 columns, each with three vertical wings which unite with 

 wings of neighbouring columns to form tubes. This mode 

 of construction was probably primarily determined by the 

 disposition of the branches of the choanosome which led to 

 the deposition of amoebocytes at the points of bifurcation of 

 the lines of flagellated chambers. 



Apparently the calcocytes become wholly transformed into 

 lumps, conules or flakes. 



The calcareous skeleton shows certain resemblances, espe- 

 cially at the surface, to certain Palasozoic fossils, classed 

 among "Tabulate corals" or Polyzoa. 



There is no dermal epithelium, and the canal system is 

 hymenopylous. 



The sponge has been found in 60-90 fathoms off Porto 

 Santo Island and Madeira. 



A few more words remain to be said. It has been denied 

 that Merlia is a sponge. I can only say that my opinion 



