75 



The genus Melebosis Girard is maintained by Pomei, (Op. cit. p. 87) as 

 separate from Salmacis. According to Agassiz (Rev. of Ech. p. 156) Melebosis mira- 

 bilis Gir. is sj'nonymous with S. sphœroides (sulcata); but there is not the least reason 

 to maintain this species as a distinct genus. Neither can the name Diploporus 

 Troschel (for the var. pyramidata of sphœroides) be made use of. 



The Sa/macis-species have for a long time had a very bad reputation as being 

 most difficult to distinguish. Mainly through the researches of Dödf.rlein combined 

 with the investigations here recorded it has now been shown that only five easily 

 distinguished species with some varieties are known; the difficulties are not due to 

 the animals but to the insufficient or bad descriptions of previous authors. It may 

 be desirable to give here an analytical table of the Sa/macis-species '). 



1. Globiferous pedicellariæ of two kinds, large 

 ones without lateral teeth and small ones 

 with an unpaired lateral tooth; tridentate 

 pedicellariæ with short and broad valves. 

 Spines ringed. The pores mostly not reach- 

 ing the edge of the area; tubercles occur 

 outside the pores 2. 



Globiferous pedicellariæ of one kind, 

 without or with 1—1 lateral teeth; triden- 

 tate pedicellariæ with narrow blade. Spines 

 ringed or not 3. 



2. Spines with red base S. bicolor Ag. (with the var. rarispina). 



— — green base S. Belli Döderl. 



3. Globiferous pedicellariæ without lateral 

 teeth ; spines not ringed. The pores reach 

 to the edge of the ambulacral area; no 



tubercles outside the pores S. virgulata Ag. ( with the var. Alex- 



Globiferous pedicellariæ with 1 — 1 lateral andri.) 



teeth. Pores on the buccal side and at the 

 ambitus not reaching the edge of the area; 

 with tubercles outside the pores 4. 



4. A primary tubercle is found on all the 

 ambulacral plates: test rather high, peri- 

 stome not much sunken S. sphœroides Ag. (^with the var. pyra- 



A primary tubercle is found only on midata.) 



every second ambulacral plate; test very 

 low, peristome much sunken S. dussumieri Ag. 



'] UÖDKHLKiN and i)E Mki.ikiik have already j^iven analytical tables of the species of Salmacis: as, 

 however, I have made here several corrections of their views of these species, it seems not superfluous 

 to give a new analytical tabic thereof. 



10* 



