98 



inellida is distinguished very clearly from all other species of Aphrocallistes by 

 its peculiar shape. In the structure of its dictyonal net and in the shape of its 

 isolated spicules on the other hand, no distinctive character is noticeable. The 

 three specimens collected by the " Investigator " to the West of the Andamans 

 agree in size and shape perfectly with those found in the Philippines and near 

 Japan, which I have described previously. They are tubular, dichotomously 

 branched, circular in transverse section and widen out to a diameter of 8 mm. 

 above. By the repeated bifurcation and the breaking off of every second branch, 

 a zig-zag-shaped tube, 8-15 mm. wide is produced, from the angular bends of 

 which the stumps of the branches broken off", extend obliquely upwards. In the 

 upper parts the branches are generally not broken off and are preserved with their 

 dichotomous ramifications (pi. XV, f. 14). The largest specimen is 10 cm. high. 



The dictyonal net is similar to that of Aphrocallistes heatrix, described on 

 p. 88. Also the slender, hexactine dermalia and most of the robust, rough gastral 

 diactine beams, agree perfectly with the corresponding spicules of that species. 

 The same can be said of the parenchymal oxyhexactines. The scopules also here 

 extend right up to the dermal membrane and are, on the whole, similar to those 

 described above of Aphrocallistes beatrix. Their terminally thickened branches 

 are however never strongly bent outwards as is so frequently the case in that 

 species. 



The parenchymal hexasters, the number of which differs considerably in 

 different specimens and in different parts of one and the same specimen, are 

 somewhat remarkable. They agree on the whole in shape with the parenchy- 

 mal hexasters described above of Aphrocallistes heatrix, but are rarely elongated 

 to such an extent as in that species. In some specimens of Aphrocallistes 

 ramosm the branch-rays of the hexasters all terminate with simple points as in 

 Aphrocallistes heatrix. In another specimen, from a different locality, the branch- 

 rays of some of these hexasters, but by no means of all of them, bear on their 

 ends four, more rarely three, fine, terminal claws, which are either vertical to 

 the ray from which they arise or extend obliquely upwards. These claws are 

 basally straight and terminally shghtly recurved, they terminate in fine points. 

 As these terminal appendages of the branch-rays have the shape of claws (o'"'0. 

 I have named the hexasters supplied with them, onychasters, a term also used 

 in the description of Begadrella on p. 70. A revision of my older slides of the 

 s^ecimejis oi Aphrocallistes ramosus hrought home by the "Challenger" has 

 revealed the fact that there also not unfrequently onychasters, similar in shape 

 and in size, occur together with the parenchymal oxyhexasters, although they 

 are in these specimens not so abundant as in the " Investigator" material. I 

 must suppose that the spicule represented in fig. 10 plate LXXXVI of my 

 " Challenger " Keport on the Hexactinellida as a discohexaster, should in reality 

 have been drawn as an onychaster, that is to say that there were present on the 



