ox MERLIA NOKMAXI. 677 



believe, from the West Indies, and two of my specimens of 

 the real Merlia from Porto Santo, on which to base his 

 conclusions. If I had had such scanty material, I also might 

 ensily have come to the same conclusion, because Merlia often 

 grows on a spreading crust of Polytrema or other Foranii- 

 nifera, and the cancellated structure of the latter appears 

 to belong to the Merlia; but having had hundreds of 

 specimens to deal with, I have found Merlia growing on 

 shells, on gneiss, on spreading laminse of thin incrusting 

 Corallines, etc. Foratninifera with a superficial reticulate 

 pattern with nodal tubercles nre not uncommon. Dr. 

 Harmer showed me one he had found among the " Siboga " 

 Polyzoa. Professor Stnnley Gardiner collected a similar 

 species incrusting Lithothanmiun froin Providence Reef. I 

 myself dredged a Carpenteria-like species off Christmas 

 Island. In all these instances the Foraminiferal nature is 

 clearly shown by the finely perforated floor at the base of 

 each mesh. At one time I myself thought Merlia was a 

 Foraminiferan, not only on account of the calcareous skeleton, 

 but also because the masses of crypt-cells (in gl^'cerine) 

 looked like lobose pseudopodia, the maltha in which they 

 were imbedded being invisible. 



At the gi'owing edges of the sponge and in whole young- 

 specimens there is simply a delicate network of calcareous 

 bars. In the case of young specimens on thin bivalve shells 

 one can often see through the whole structure. The soft 

 sponge tissues cover the calcareous network, which has no 

 basal lamina, and there is no trace of any other organism what- 

 soever. Even this strong circumstantial evidence in favour of 

 the theory that Merlia is a sponge is unimportant in com- 

 parison with the positive evid^-nce afforded by the soft tissues. 



The growing edge of the skeleton shows the delicate poly- 

 gonal network, with half- or open polygons at the edges, and 

 here and there isolated small lumps or even bars (PI. 38, fig. 5). 

 In the older parts the slender bars of the delicate meshwork 

 have increased in height and the meshes have become pits, 

 and still later tubes with one or more tabulas. 



