86 PORIFERA. III. 



two-pointed teeth, the exact shape of which may be seen in the figures. Such are the facts generally, 

 but it is however not always possible to decide, which of the branches belong to the axis, the branches 

 often being so uniformly developed that they all seem alike. The chela is somewhat contorted, so 

 that when seen from the end all six branches become visible, and we get a figure almost like a six- 

 rayed aster. (PI. IX, fig. if). The length of the chela from one end of the shaft to the other is 0-028 

 0-035 mm and the thickness of the shaft is 0-004 0-005 mra . That these bodies are transformed chelae is 

 seen also from the developmental stages, of which some were found (PI. IX, fig. ig); these consist of a 

 curved axis with two lateral dilatations near each end, and they resemble to some degree develop- 

 mental stages of other chelae, only that the tooth is here not curved forwards. - These chelae remind 

 one somewhat of the chelae in H. Schmidti Tops., but are different in many respects; they are very 

 interesting and take an intermediate position between the previously described, more or less trans- 

 formed chelae; H. crux has normally shaped, but spined chelae, H. Schmidti has likewise spined chelae, 

 but much more transformed, in H. aenigma they are still more transformed, and finally we find in 

 H. vidua the chelae transformed almost to spined staves. The chelae occur in the dermal membrane 

 forming a dense layer, and they are also found scattered in the tissue lower down in the sponge. 



Locality: Station 89, 64 45' Lat. N., 27 20' Long. W., depth 310 fathoms. The station lies in the 

 Denmark Strait. 



Remarks: While at present only few species are known with spined chelae - - I think only 

 the above mentioned and Pseudohalichondria cla-vilobata Cart (Ann. Mag. Nat. Hist 5, XVIII, 1886, 

 454, PI. X, fig. 8) - there are on the other hand some fossil forms, described by Hinde and 

 Holmes (Journ. of the Linn. Soc. XIV, 1894, 214, PI. XI, figs. 814); the authors figure seven chela:, 

 which they think belong to four species, two of which are named as Pseudohalichondria defonnis and 

 oamaruensis. The forms are all referred to the genus Pseudohalichondria Cart, evidently only on 

 account of the spined chelae. As already mentioned by Topsent in his work from 1904, there may 

 be some reason to think, that the described chelae belong to species of Hymedesmia. The chelae were 

 found in lower Tertiary strata in New Zealand. It is somewhat curious, that such chelae are known 

 as fossil, since the spination and higher or lower transformation of the chelae must certainly be 

 considered as a feature of recent origin. Probably therefore the genus Hymedesmia, containing only 

 incrusting forms of a simple structure, is an old genus. 



34. H. filifera O. Schmidt. 

 PI. Ill, Fig. 12, PI. IX, Fig. 2. 



1875. Desmacidon filiferum O. Schmidt, Jahresber. der Comm. zur wissensch. Unters. der deutsche Meere 



in Kiel fur 1872 73, 1875, 117, Taf. I, Fig. 6. 

 1903. Hymedesmia Jilifera, Thiele, Arch, fur Naturgesch., Jahrg. 1903, I, 391, Taf. XXI, Fig. 25 a c. 



Incrusting, but not thin; surface smooth, bearing a number of thin, cylindrical oscular and pore- 

 papilla. The dermal membrane solid, with horizontal spicules. The skeleton formed of dermal spicules 

 strongly developed, the main skeleton rather weak. Spicula: megasclcra; the skeletal spicules entirely 

 spined acanthostyli with the basal end not or slightly thickened, ovj o'2j'"'", not divided info two 



