6o 



ECHINOIDEA. II. 



have constaiith' foxind both the iuiier piates of I and V developed; but the piates I. b. i and V. a. i 

 are generally very small and easily overlooked. The piates I. a. i and V. b. i may be ver\- unequally 

 developed, one of them siniulating the labrnni, but the presence of the pore at its inner end shows 

 its real natnre. Generally onl\' the four larger of these piates bear distinct pores and tnbe-feet, in the 

 otlier piates only qnite rndimentary pores are present, sonietinies the pore has even quite disappeared. 

 Resides the snpposed coalescence between the two inner piates of ambulacra I and \', Loven, j^oints 

 out (Op. cit. p. 36) as another pecnliar feature in this species, that the inner piates of ambulacra II and 

 IV are not in accordance with the general rule that the piates La, II. a, III. b, IV. a, V. b are the 

 largest. I have constantly found the inner piates of the paired ambulacra to be in accordance with 

 the rule, only as to ambulacrum IV I have sometimes been unable to see it distinctly. As it seems ver\' 

 unlikely that all the specimens examined by Loven should happen to be abnormal in this respect, I 



must venture to suggest that Loven has overlooked some of these small 

 piates, which mav, indeed, be rather difficult to see. (I have found them 

 easiest to discern when e.xamining the denuded test in alcohol; on dried 

 tests, treated with alcohol-glycerine it is almost impossible to trace the limits 

 between the small piates). A very small plate may sometimes be found be- 

 tween the inner piates of the ambulacra I and II on one side and IV and 

 V on the other side (PI. VIII. Figs. 5, 8, 9, 1 1 ). It must doubtless be regarded 

 as the rndimentary inner plate of the interambulacra i and 4. Whether this 

 plate was really absent in Lov én' s specimens or perhaps was overlooked, 

 it follows from its occasional (not ver}- seldom) occurrence that the piates 

 interpreted by Loven as No. i of the interambulacra i and 4 (On Pourta- 

 lesia. PI. II. 9I are really No. 2. In the figure 9 copied from the quoted figure 

 of Loven, I have shown m\- interpretation of these piates. (Comp. Figs. 10, 11 of 

 Poiirf. pJiinle). Upon the whole there is so great variation in the development 

 of the piates of this region that it is scarceh' possible to find two specimens 

 quite alike in this respect. Such extensive variation in structures of consider- 

 able morphological importance is of no small interest, and it is shown hereb\- that the mutual relation 

 of the piates in this region cannot be relied upon for specific differences, in any case for this .species, 

 and for the other species it will also be necessary to be ver\- cautions in the use of such characters. 

 The fignres 4 — 6, S — ri. PI. \'III show some of the variations in the structure of this region found in 

 P- Jeffreysi. (These specimens otherwise are all quite typical P. Jeff'rfysi; all variations may be found 

 in .specimens from the same station). 



The primary tubercles form distinct longitudinal (from a morphological point of \iew : tran.s- 

 verse) series on the sides at the anterior end of the test. These series generally are very prominent 

 on the piates of the anterior series of the two antero-lateral ambulacra (II and IV|, each plate bearing 

 one series in the middle, the tubercles increasing somewhat in size from the anterior towards the 

 posterior edge of the plate. On the piates of the posterior series of these two ambulacra the tubercles 

 are more irregularly arranged, and on the posterior part of the test they are upon the whole quite 

 irregularh- arranged, though sometimes there is a tendenc\- towards a serial arrangement. The piates 



Fjg, 9. .\ctinal plastrou of 



Poui-lalesia Jeffreysi. .\fter 



Loven. 



