AMPHIURIDAE 281 



none, likewise almost regularly throughout. He does not mention the occurrence of two 

 tentacle scales here and there, alternating with one or none ; but this is by no means of 

 rare occurrence, and this latter case leads to that of typical Belgicae {Mortenseni) : two 

 scales regularly throughout the arm, excepting on some of the proximal joints. As a 

 matter of fact, we have all transitions from the occurrence of two tentacle scales regu- 

 larly throughout the arm to one or none, and as the specimens agree completely in all 

 other characters, it is impossible to draw a line between them anywhere ; consequently 

 both Mortenseni and alternans become synonyms of Belgicae. 



On the whole A. Belgicae is perplexing in showing such great variation in regard to 

 the tentacle scales, for the number of these scales, two, one, or none, is otherwise of 

 primary classificatory importance in the genus Amphiura. Not that this species is other- 

 wise difficult to recognize. As a matter of fact, apart from the variation in the tentacle 

 scales, its general appearance is quite characteristic: 

 the long, stout, conical spines, almost constantly four 

 in number, the large irregular scales in the centre of 

 the disk, generally without recognizable primary 

 plates, and the rather thick skin obscuring the plates 

 of the mouth frame (this is also pointed out by 

 Lyman as characteristic of A. tomentosa); further, 

 the arrangement of the tentacle scales, when there are 

 two of them, is characteristically different from such 

 species as A. Eugeniae, as seen in Figs. 18 and 21. As 

 for the young specimens identification is often dif- 

 ficult — but this holds of most Ophiuroids in their Fig. 18. Amphiwa Belgicae, Koehler. 



. Part of oral side. 10. 



young stages. 



The specimen from St. 190 is exceptional in having only three arm spines. One 

 specimen from St. WS 27 is unusual in the spines (five) being somewhat flattened and 

 even slightly widened at the point, not conical as is otherwise usual in this species. Both 

 specimens being in all other characters typical, I can only regard them as individual 

 variations of A. Belgicae. 



Hertz regards A. Mortenseni (Belgicae) as only a variety of Eugeniae. I do not think 

 this correct, as is evident from the description and figures of A. Eugeniae given below. 



A. Belgicae is viviparous and hermaphrodite. Whether it is protandric I have been 

 unable to settle definitely ; it seems rather to be a more complicated case. Whereas in 

 adult specimens it is easy enough to ascertain the hermaphrodite character of the 

 gonads, both eggs and sperms being found within the same gonad (though sometimes 

 only eggs are distinct), it seems as if the gonads in the younger specimens are of separate 

 sexes. In one young specimen, 4 mm. in diameter of disk, I found very distinct male 

 gonads, partly with ripe sperms, besides gonads with young eggs only. Young ones may 

 be found in specimens of only co. 6 mm. diameter of disk ; it appears that they lie within 

 the gonads, the development thus being intra-ovarial. The adult specimens also strongly 

 convey this impression. In young specimens there are only one or two embryos in each 



