Evolution and Symmetry in the Order of the Sea-pens. 121 



A remarkable illustration of the variability of the spicules of this genus 

 was obtained by the examination of four specimens obtained by the " Siboga " 

 Expedition. These specimens were all dredged up at the same time in the 

 anchorage at Amboyna, and, although they vary in length from 40 mm. to 

 113 mm., there can be no doubt that they belong to the same species. A 

 preparation of the cortex of the rachis of the two larger specimens shows a 

 dense felt-work of spicules, and, when these are teased out, it is found that 

 the great majority of them are fiat rods with round bifurcated extremities ; 

 but, in addition, there are many twins, triplets, and other varieties. In the 

 two smaller specimens, however, the spicules are much less densely clustered, 

 and the predominating type is a small, flat rod without bifurcated extremi- 

 ties. An even more remarkable illustration, however, was found in the 

 examination of a number of specimens of Sderobelcmnon burgeri from the 

 same dredging in the Molo Strait. This species, although it usually exhibits 

 an incipient bilateral symmetry, is so closely related to Veretillum that some 

 specimens are almost exactly intermediate in character between the two 

 genera. The spicules of the rachis are, as in Veretillum, thin, flat plates 

 with a great variety of outline ; but the majority of them are irregularly 

 round, oval, or dumb-bell-shaped, frequently divided by lines into twins, 

 triplets, quadruplets, and multiplets, but rod- or spindle-shaped spicules are 

 scarce. With so much variety in the spicules of a single specimen there was 

 found great variation from one specimen to another. In one of two 

 specimens from the same station, of approximately the same size, the spicules 

 were few in number, and the larger ones 0'4 mm. in diameter, and in the 

 other the spicules were numerous, and the larger ones only 0"04 mm. in 

 diameter. This difference between the two specimens is so striking that, 

 according to the general practice of systematists, they would undoubtedly be 

 placed in distinct species, but fortunately there are 60 other specimens of 

 various sizes from the same locality, and an examination of these shows that 

 no two specimens are alike as regards the spicular armature, and that there 

 are many intermediate cases between the two that were first mentioned. There 

 can be no doubt therefore that in S. lurgeri we have an example of a species 

 in which the spicules are so variable that they cannot be regarded as being 

 of any value for the separation of species. 



The presence or absence of spicules in the expansible parts (anthocodise) of 

 the autozooids has been used by some authors to distinguish genera from 

 one another. Thus KoUiker distinguished the genus Clavella from Lituaria 

 by the presence of spicules in the anthocodise of the former and their 

 absence in the latter, but the only known species of the old genus Clavella 

 resembles L. phalloides so closely in other respects that it cannot be 



