SALMACIS. 315 



genus recognizable and valid. A closely related form has long been distinguished 

 as rarispina, but the examination of our M. C. Z. material confirms Mortensen's 

 statement that there is no constant difference between the two forms. Indeed 

 the difference which his specimens showed in the number of ambulacral and 

 interambulacral plates and which led him to retain rarispina as a variety of 

 bicolor is not shown by our specimens, and it is not practicable to distinguish 

 the variety. So far as can be determined from the available material, and the 

 published notes on the different species, the form of the test is strikingly variable 

 in bicolor, virgulata, Alexandri, and perhaps also in sphceroides, the vertical diame- 

 ter ranging from little more than one half to nearly three fourths of the horizontal. 

 We note also that most specimens fall into one of two groups, the one with the 

 vertical diameter from .55 to .60 h. d. and the other with it about .70 h. cl. 

 Ramsay, who appears to be the only writer to have examined large series of 

 living Salmacis, speaks emphatically of the diversity shown by S. Alexandri 

 in the form of the test (1885, Cat. Ech. Australian Mus., p. 48). The question 

 of the relation of A lexandri to virgulata appears to be a debatable one; the former 

 is at least a subspecies characteristic of the Australian region, and as a con- 

 necting series of specimens is lacking, those at hand being easily recognized, it 

 has seemed well to let Alexandri stand as a valid species. These two species 

 (virgulata and Alexandri) are distinguished from the other members of the genus 

 by the absence of bands on the primary spines, but this is of course, not so 

 important a character as those which distinguish Dussumieri. In this species, 

 the test appears to be always flattened, so much so that the vertical diameter 

 is less than one half the horizontal, and ocular I is insert or very nearly so. 

 Moreover the ambulacra are highly modified, so that there is a primary tubercle 

 close to the poriferous zone, only on every other ambulacral plate, excepting 

 only the oldest (near actinostome) and youngest (near ocular plate). Morten- 

 sen (1904, Dan. Exp. Siam: Ech., p. 73) refers to "a very dark colored form" 

 of Dussumieri in the collection of the British Museum from Tuticorin (southern 

 India) and he adds that Bell's S. sulcata from Zanzibar is similar. He says 

 further that "it can only be regarded as a color variety." A fine specimen of 

 this form from Zanzibar (PL 111, figs. 4-6) is in the collection of the M. C. Z. 

 and its color characters are so strikingly different from Dussumieri that it must 

 rank as a new species, for which we suggest the name erythracis (epvOpos, red + 

 OLKLS, a pointed instrument) in reference to the more or less conspicuously ver- 

 milion-red bases of the secondary and miliary spines. There is little question 

 about the status of the form, which Loven has showed Linne designated as 



