STYLASTERIDAE 





whatever here or anywhere else to call the tentacles capitate; they are formed quite like the thread- 

 like tentacles in the athecate hydroids. - Further, the gasterozooid shows in its finer structure no 

 difference from what has been described in Stylaster (Eustylaster] gemmas- 

 cens. In the present species it is also broken up at the base into a circle 

 of columns, the number of which seems mostly to be about 6. The 

 structure of the dactylozooid also agrees with that in the other northern 

 species of Stylaster ; but the dactylozooids are somewhat larger in Stylaster 

 non'i'gicits than in the previous species. The large cnidocysts are found 

 in extremely small numbers in the stolons. 



Whilst the preceding species was only represented by specimens Text-fig. D. Diagrammatic me- 

 not very well-preserved, the opportunity was taken to obtain fresh material dian section through the cyclo- 



system of Stylaster (Allopora) nor- 



of Stylaster norvegicus from the Trondhjem Fjord, in the outer part of vegicns. d = dactylozooid, gt = 

 which the species is fairly common in suitable localities. Both male and tentacle of ^ e g asterozooid . & 



= gasterostyle, gw = free part 

 female colonies could be examined and even if most of the questions of the body-wall of the gastero- 



concerning the development of the gonophores must still remain un- 

 answered, yet the investigation contributes a good deal to the understanding of the nature of the 

 gonophore in Stylaster. 



The sexual cells are already present in the youngest developmental stage of the male gono- 

 phore which was found (PI. IV fig. 35) so that their origin cannot be settled. The gono- 

 phore shows clearly, that Hickson (1891 p. 384) was wrong in maintaining, that "the spermarium is 

 covered by a double sheath of very thin etcoderm and endoderm". Neither at this stage or later can 

 anything be seen in the numerous gonophores examined (PI. IV, fig. 37), which could be taken for an 

 endodermal layer between the ectoderm and the generative cells. -- Hickson (I.e. p. 390) maintains 

 as a typical difference between the gonophores in Allopora and Distichopora that the latter genus has 

 no spadix, whilst Allopora has a strongly developed spadix. In Stylaster (Allopora] norvegicus the 

 spadix is strongly developed in the young gonophores but atrophies during the transformation of the 

 sperm cells (PI. IV fig. 38) and lastly disappears entirely in the mature gonophore as is the case also 

 in Stylaster roseus. 



The present species forms in part the basis for Hick son's studies on the seminal duct 

 Unfortunately I did not succeed in finding all the developmental stages of it and it is remarkable 

 that it is not always to be found in almost or quite ripe gonophores in spite of the fact, that the preserva- 

 tion of the material is excellent It seems doubtful, if under all conditions it comes to development 

 even in all gonophores within the same colony; its importance therefore must be reduced in the 

 general considerations on the group. - - The youngest developmental stage found (PI. V fig. 44) appears 

 as a collection of somewhat higher and lighter-coloured cells at the apex of the gonophore. The rudi- 

 ment is distinctly double, for under the thickened ectoderm cells there is a collection of inner cells (I), which 

 have a characteristic, almost fibrillar protoplasmic structure; these fairly large and light-coloured cells 

 are separated from the outer ectodermal layer by a very fine lamella. At a more advanced stage, 

 when the seminal duct is almost fully formed (PI. V fig. 45) the inner cells push the thin lamella in 

 front of them into a conical point, which is surrounded by the more deformed ectodermal cells. The 



The tngolf-Kxpedition. V. ;- 3 



