ECHINOIDEA. I. 155 



23 or still more plates in succession. This, however, does not make the primary series look very 

 irregular. The secondary tubercles are very few; on the actinal side the largest ones form a rather 

 regular longitudinal series on either side inside of the primary series; here they almost equal the 

 primary tubercles in size. The pores reach quite to the edge of the area. 



The primary tubercles of the interambulacral areas are considerably larger than the ambula- 

 cral ones; they form very conspicuous longitudinal series. The tubercles are largest at the ambitus, 

 but often they decrease only very little in size towards the apical area, but in the common way 

 towards the peristome. Most frequently a primary tubercle is wanting on a few or more plates near 

 the apical area; in the latter case, the plates, as in the ambulacral areas, are placed alternately with 

 plates provided with a primary tubercle. The secondary tubercles are rather few on the abactinal side, 

 and averagely much smaller than the primary ones; on the actinal side they are more numerous, and 

 the largest are of about the same size as the primary tubercles; they form a rather regular longitudinal 

 series inside of the primary one, and on large specimens one more series may be found inside, along 

 the very median line of the area, not, however, very regular. The tubercles outside of the primary 

 series form no longitudinal series. The miliary tubercles are rather numerous, but very small, so that 

 the test looks rather smooth. 



The spines are not very numerous, nor very close-set, but upon the whole long and strong - 

 considerable variation is found, however, with regard to the size. The longest ones are found a little 

 above the ambitus; in some individuals they decrease only very little in size towards the apical area, 

 so that the uppermost spines are of about the same length as those at the ambitus, which gives to 

 the animal a very peculiar appearance. On the actinal side the ends of the spines, as in var. nirdi- 

 terranea, form an even ambulator}' surface; they are likewise flat, almost all of them, truncate, and 

 a little widened at the point. 



The colour of the test is white with a more or less broad, reddish brown band down the 

 middle of each series of plates (PI. I. Fig. 7). The lower side most frequently quite white. The spines 

 are red, reddish brown, or greenish; brown for a smaller part at the base, the rest white; the actinal 

 spines are quite white. In younger specimens the red colour may reach almost to the point of 

 the spines. 



Var. norvegicus (PI. I. Figs. 4, 8. PI. II. Figs. 2, 6. PI. XV. Figs. 2, 16. PI. XVI. Figs. 5, 22. PI. XVIII. 

 Figs, i, 7, 24). The test generally much flattened, in larger specimens slightly conical. The peristome 

 highly varying in size, sometimes very small; the edge of the peristome generally much bent inward. 

 The tubercles rather large and strong. 



The ambulacral areas are very characteristic (PI. XV. Fig. 16. PI. XVI. Fig. 22). The primary 

 tubercles form no continuous series; between every two plates with a primary tubercle one or more (up 

 to 4, most frequently 2) plates are found without such a tubercle. On these latter plates (those above 

 the ambitus) generally only one secondary tubercle is found, placed a little outside of the primary 

 series, and this secondary tubercle is most frequently rather large, almost as large as the nearest 

 primary tubercles. As a consequence of this feature the primary and secondary tubercles form 

 together one longitudinal series, which is very irregular, partly because the tubercles do not decrease 

 evenly in size upward, partly because they are not placed in a straight line. On the actinal side the 



20* 



