68 



32 interambulacral plates in bicolor, 32 ambulacral and 28 interambulacral plates 

 in rarispina; in a specimen of 56 mm.. 48 ambulacral and 35 interambulacral plates 

 in bicolor, 36 ambulacral and 29 interambulacral plates in rarispina, a very con- 

 spicuous dilTerence. Now the material of bicolor examined is very small , and it 

 may safely be supposed that a rather large amount of variation will occur also 

 here. But the examinations made tend to show that the Ceylon-form of bicolor has 

 on an average more coronal plates than those from Mauritius and the Siamese 

 specimens of rarispina. Other distinctive characters do not seem to occur. Pedi- 

 cellariæ, spicules and structure of spines are quite alike; as regards the colour of 

 the spines it may be said that the light rings are generally broader in rarispina 

 than in bicolor, but it is no constant feature. Neither can any character be taken 

 from the angular pores; they vary very much, may be very distinct or almost (juite 

 obliterated in specimens of the same size. Thus I think it an inevitable conclusion 

 that Salmacis bicolor and rarispina are indeed only one species, of which may be 

 distinguished a forma typica, with numerous coronal plates, as yet known only 

 from Ceylon, and a var. rarispina, with less numerous coronal plates, known from 

 Siani, Mauritius and Zanzibar. 



It may be especially remarked that I have examined the type specimens of 

 both bicolor and rarispina in the Paris-Museum , also with regard to pedicellariæ. 

 Furthermore Dr. Ghavier has been so very kind as to send me the type specimen 

 of rarispina for direct comparison with my Siamese specimens. I can thus say with 

 absolute certainty that they are really S. rarispina. The question of these two 

 „species" I hope thus to have definitely settled. 



The Salmacis rarispina of de Loiuol from Amboina is, indeed, Salm. 

 sphœroides, as supposed by Döderlein. Through the kindness of Prof. Bedot I have 

 had the specimen sent for examination. The specimens mentioned by Koehler 

 (Cat. rais. Éch. de la Sonde, p. 413) will then probably also be sphœroides, as they 

 are said to be identical with de Loriol's specimen. The S. globatrix of Loven can 

 scarcely be anything but S. bicolor. The expression „spinæ basi rubræ" does not 

 agree with any other Sa/macis-species. — The Salm. bicolor named by Bell in the 

 „Alert"-Echinodermala (p. 118) is S. belli Döderl. — Of the specimens named in the 

 „Challenger"-Ech. (p. 113) the S. rarispina from St. 188 is S. belli, those from St. 186 

 (2 small specimens) are, the one ^S. dussumieri, the other perhaps S. belli (no globi- 

 ferous pedicellariæ found), al any case not rarispina. The specimens from St. 203 

 I have not seen. The S. bicolor from Zamboangan is really that species. 



9. Salmacis virgulata Agass. forma typica. 



PI. VI. Figs. 7, 18, 46, 47. PI. Vll. Kig. 40. 



Salmaci.1 virgiiJatus. L. Agassiz & Desor. 1846. Catalogue raisonne des Éeliinides p. 359 (55). 



— conica. v. Makiens. 1866. Ostasiatische Ecliinodermen. Archiv f. Naturgesch. 1866. p. 159. 



