380 Annals of the South African Museum. 



the bare interambulacral area, including the distal half of the 

 genital plate is brown, with 9 or 10 transverse, irregular bars of 

 violet (with little indication of blue), of which the lowest are bright- 

 est and those near the genital plate are faintest; along each margin 

 of the area is a rather broad vertical white stripe, not at all sharply 

 defined but quite evident; the ambulacra are red, as in floridanus. 

 The abactinal, and even some of the actinal, secondaries of inter- 

 ruptus are bright scarlet, but in floridanus they are commonly dirty 

 white, though a few may be more or less red. 



The primary spines of floridanus, when full grown and uninjured 

 are pale greenish at base, particularly the collar; on the actinal 

 side beyond the collar they are shining, pure white; abactinally the 

 greenish passes more or less rapidly but not abruptly into brilliant 

 scarlet-red ; if the spines are very long, the red becomes discon- 

 tinuous distally so that the extreme terminal part of the spine 

 abactinally is pale greenish with well-separated scarlet cross-bands 

 or spots. In some specimens, there is little red and it is nearly 

 all confined to the abactinal surface of the middle third of the 

 spine. More commonly however the red extends even to the collar 

 and sometimes the collar itself is more or less red. But in any 

 case the red is a more or less vivid scarlet. In interruptus on the 

 other hand, the collar of the full-grown spines is usually greenish 

 proximally and underneath but distally, at least on the abactinal 

 ridge, it becomes dull purplish-red and this colour occupies the 

 upper surface of most of the spine; distally it becomes redder and 

 less purple and at the tip of certain spines, especially those that 

 are regenerating, we find red spots on a greenish-ground, very 

 similar to those found in floridanus. The under surface of the pri- 

 maries is always more or less shining white. The amount of red 

 on the spines shows considerably diversity but in any case, it is 

 (except for occasional distal spots as noted) a very purplish red 

 quite unlike the fine scarlet of floridanus. As a result of the colour 

 dillerences interruptus, viewed as a whole, looks quite unlike any 

 specimen of floridanus I have ever seen, and is even more different 

 from the other Recent species of the genus. 



Aside from the colour differences, interruptus differs from florida- 

 nus in the greater stoutness of the primary spines and in the ophi- 

 cephalous pedicellariae. While the thickness of the basal part of the 

 largest primaries in the West Indian species is about 2 mm. or, 

 say, about 2 percent of the whole length, in the African form it is 

 3'5 mm. or about 4-5 per cent of the length. The ophicephalous 

 pedicellariae in both species have stalks about 4 mm. long and 



