1897 Topsent pp. 425, 440. 



1898 Topsent p. 124. 



1900 (x) Kirkpatrick pp. 128, 134 — 135. 

 1900 Minchin p. 43. 

 1900 Topsent p. 128. 

 1909 Hentschel p. 385. 



Ridley's type is fortunately preserved in the British Museum ; my number B. M. 27 

 refers to this type, registered '82. 2. 23. 278. A second specimen in the British Museum, to 

 which I gave the number B.M. 38, is the type of a "variety" described by Ridley & Dendy 

 in the Challenger Reports ; it is registered '87.5.2.97. Finally there is a third specimen, 

 B.M. 3g, registered in the British Museum '98. 12. 20. 21. It is the specimen described by 

 Kirkpatrick as var. robusfa, specimen r, viz. the one "forming a thin yellow crust on a shell". 

 Ridley stated (1. c. p. 470) about 5. decumbens: "This species appears to be more nearly allied 

 in its spiculation 5. (Alcyonium) purpurea Lamarck, than to any other Indo-Pacific species, 

 but it differs from it in wanting the magnificent crimson color of that form, in its encrusting 

 habit (purpurea being massive), in the inferior diameter of the shaft of the spinulate and the 



superior length of the spinispirular spicule " However, are these differences of specific 



value? I believe not, since by the two other specimens, found afterwards, viz. B.M. 38 and 

 B.M. 39 the original character is slightly modified. The "variety" described by Ridley & Dendy 

 (B.M. 38) is not merely encrusting, but is said to consist "of a number of stout, irregular, 

 anastomosing trabeculae, forming together a sessile, cavernous mass. In the second place Ridley 

 and Dendy state (l.c. p. 229): "there are also slight differences in the proportions of the spicules 

 between our specimen and the type". [Compare figs. 1 and 2 on PI. VIII). The authors arrive 

 at the conclusion that "there are 110 differences sufficiënt to justify us in separating the two 

 specifically". The absence of the "crimson" color can hardly furnish us with an argument, since 

 we have in " Stiberites wilsoni" Crtr. an example that this color may be absent as in S. ïcilso?ii 

 var. alèidus. Now there are in the Siboga-collection some specimens which perfectly correspond 

 to Spirastrella decumbens, in external appearance and in spiculation. The encrusting character 

 is conspicuous in 1031, 1455 a-c and 1945; the spicules of the latter two are figured on PI. YIII 

 (figs. 5 and 3). Specimen 1421, the spicules of which are to be seen in fig. 6, externally resembles 

 more the Challenger-specimen of 5. decumbens (B.M. 38), whereas in spiculation it comes 

 nearer B. M. 27- Comparing them all together we find that decumbens and purpurea are hardly 

 different "species". The greatest difference lies in the size of the largest spinispirae. We will 

 see, however, that even on this account 110 distinction can be made. Por S. decumbens is nut 

 only allied to 5". purpurea. Specimen 1480 indeed is a transition between encrusting forms like 

 1455, 1945 an d pyramiclal forms like 426 a-c The latter are closely allied to - - in fact identical 

 with — Stiberites zvilsoni. Comparing the size of the large spinispirae of this sponge (PI. X. 

 fig. 5) with those of S. purpurea and "S. decumbens ", we feel the supposed specific difference 

 vanish more and more, in iact clisappear. Indeed, either one has to consider U S. decumbens" as 

 identical with S. purpurea or make different species from S. decumbeus-type (B.M. 27), S. decumbens 

 var. (B.M. 38), S. decumbens var. (B.M. 39), S. purpurea (B.M. 12) and the specimens 1031. 

 1455, 1945, 1421, 1402, 1480, 426 a-c etc. The first seems to me at present the better clevice. 



