250 Prof. P. M. Duncan oii the Salenid^e. 



hence its appellation. He writes {op. cif. p. 261), " The 

 primary spines are enormous, twice the diameter of the test in 

 length, of a brilliant white colour and of all shapes. Some 

 of them are uniformly tapering, others swelling at about one 

 third the distance from the base, others flattened and curved, 

 but all longitudinally striated and loosely covered with sharp 

 spines irregularly arranged along the body of the spines. 

 The secondary spines, as well as the greater number of the 

 spines of the ambulacra, as tar as the ambitus, are short club- 

 shaped, sometimes curved and flattened, longitudinally striated, 

 with slight serrations. These short spines give to the median 

 interambulacral and ambulacral zones the aspect of the corre- 

 sponding zones of Cidaris, but they are not, as in Cidans, 

 arranged in a circle around the base of the primary spines. 

 These small spines, as well as the whole abactinal area, are 

 covered with prominent dark violet pigment-cells, standing 

 in striking contrast to the white primary spines." In the 

 plate (pi. iii. flgs. 8 and 14, op. cit.) A. Agassiz delineates this 

 remarkable variety of spines. Loven does not notice spines 

 at all ; and hence we may suppose that his specimen was 

 denuded. 



In the specimen of Salem'a profundi the spines are all 

 preserved, and are of great interest, both on account of their 

 variety and ornamentation. 



The few largest primary tubercles situated above the am- 

 bitus and close to it bear very long spines, which are -fi inch 

 long, or very nearly four times the length of the diameter of 

 the test (fig. 11). These largest spines are slender, and taper off 

 gradually towards the end, but are slightly swollen midway and 

 towards the succeeding quarter of their length. Usually they 

 are nearly straight, cylindrical, and pin-shaped. They are 

 finely ridged longitudinally, there being ten or twelve low 

 long ridges with corresponding grooves, and the ridges are 

 very minutely aciculate or serrate. They are not flat or much 

 bent ; and the spinules on them are in regular whorls at stated 

 intervals. The spinules are numerous in each Avhorl, and are 

 bent, the point being outwards and nearly parallel with the 

 spine. The whorls succeed a plainly striated region close to 

 the base of the spine above the socket, and are closer together 

 than further up. Then they become fewer and smaller, and 

 finally disappear near the tip, Avhere the delicate ridges or 

 flutings become finer and are either excessively minutely 

 serrate or the termination has a cellular appearance under a 

 moderate magnify ing-power. Pigment granules are seen on 

 the lowest part only of the largest spines ; but most of the fine 

 ornamentation is covered with a fluftV adventitious matter. 



