22 THE SPONGES OF THE WEST-CENTRAL PACIFIC 



sistency in life and while wet was very spongy, but this sponge is extremely 

 hard to cut even with a very sharp knife, exactly like all typical specimens 

 of the genus Ircinia. Similarly, when dry, it is as hard as wood. 



The surface was much overgrown by a specimen of Phyllospongia; but 

 where not so covered, there are conules 2 mm high and 3 to 4 mm apart. The 

 pores are small, contractile, and quickly closed. The oscules are 3 to 8 mm 

 in diameter and crowded throughout the upper surface of the sponge, about 

 1 to 2 cm apart. 



The ectosome is a thick, fleshy dermis about 100 ju. thick and is loaded 

 with debris, which consists mostly of foreign spicules. The endosome is 

 typical of the genus Ircinia, somewhat cavernous, but otherwise fairly densely 

 crowded with small, spherical, flagellate chambers and permeated by a fibro- 

 reticulation. It is filled even more densely with characteristic filaments. 



The skeleton of this species is characterized by fascicular ascending 

 fibers more than 1 mm in total diameter and by connective fibers of very 

 irregular size, often about 100 /x in diameter. Most of the fibers contain 

 foreign material. The Ircinia filaments are especially interesting in this 

 species. Many of them are quite typical, about 2 /x in diameter throughout 

 most of their length with very large, round, terminal knobs 8 [x in diameter. 

 The total length is (as usual) very difficult to estimate but is probably more 

 than 1 mm. Among them are numerous other strands, or filaments, 3 //, to 

 5 ix in diameter. The latter are full of irregularly distributed, round, green 

 objects, about 1 ^ to 2 ^ in diameter, which certainly appear to be chloro- 

 plasts. These strands, in hematoxylin-stained material, do not take the 

 lavender color as the clear-cut Ircinia filaments strikingly do. One is in- 

 clined to conclude that the somewhat larger strands represent symbiont algae. 

 It is stated in the literature that no incipient or partially formed Ircinia fila- 

 ments ever have been found. It is noteworthy that there are no records of 

 such juveniles. It might be true that first of all the sponge entertains sym- 

 biont algae, digests the protoplasmic portion of them, brings about a chemical 

 change in the cellulose walls that remain, and thus produces the Ircinia 

 filaments. On the basis of this hypothesis, the sponge does not build up the 

 filament from a thinner strand (such as is never found) but reduces it from 

 a thicker strand of very different initial appearance. Confirmation is found 

 in the well-known fact that the chemical composition of typical Ircinia strands 

 is not identical with that of the keratose fibers of the reticulation. The fibers 

 may be dissolved in a strong caustic, which leaves the filaments much less af- 

 fected, and the reaction to staining likewise is different. 



This sponge was first described as Cacospongia irregularis by Polejaeff 

 in 1884, page 63, from Australia. A discussion of its subsequent varied taxo- 

 nomic fate may be found on page 73 of de Laubenfels, 1948, Monograph of 

 the order Keratosa. 



