TH. MORTENSEN, 



(Schwed. Siidpolar-Exp. 



there is at the lower edge of each plate a much smaller tubercle, the whole area 

 thus carrying" four vertical series of tubercles. According to AGASSIZ (Revision of 

 Echini p. 396) there are, besides the primary series, »two irregular vertical rows on 

 each side of the median furrow, with a few miliaries scattered irregularly between 

 the secondary tubercles towards the poriferous zone». Even in the largest specimens 

 examined by me (26 mm. in diameter) I find only one secondary tubercle on each 

 plate; the second of these small tubercles is thus evidently found only in still larger 

 specimens (comp. below, the specimens from Altata?). — The median furrow is nar- 

 row, but mostly quite distinct. The pores are described in the »Revision» as sad- 

 joining; ridge separating pairs of pores prominent*. This is quite correct, and it 

 need only be added that there is a distinct depression all round each pair (Fig. 4). 

 (The difference in the shape of the pores between this species and E. mitrix has 

 been emphasized in the Report on the Echinoidea of the German Southpolar-Expe- 



The apical system (PI. IV Figs. 5—6, 9—10; PI. XIV Fig. 6) is comparatively 

 small, ca. 41—44 % of the horizontal diameter of the test. The ocular plates are in 

 larger specimens generally all in broad contact with the anal system. DODERLEINS 

 statement (loc. cit.) that he finds »bei einem ziemlich grossen Exemplar ... das 

 Analfeld vollstandig eingeschlossen von den Genitalplatten», evidently refers to the 

 specimen of E. nutrix and thus has no bearing on any variability in this character 

 in canaliculata. Further, when AGASSIZ says in ^Revision of Echini* p. 396: »Genital 

 plates in contacb, which cannot mean anything but that the oculars are excluded 

 from the anal area, I would suggest that »genitab is a lapsus calami for »ocular». — 

 The ocular plates are distinctly broader than high, lozenge-shaped, a somewhat un- 

 common form; the inner angle is generally cut off to some extent, the more so the 

 older the specimen; the ocular pore is situated in the outer angle, not near the 



dition; Fig. 4 to compare with Taf. X Fig. 7 of the latter 

 work.) — In the larger specimens there are 5 — 6 ambu- 

 lacral plates to each interambulacral plate, at the ambitus. 



Fig. 4. Ambulacral pores of Attstro- 

 cidaris canaliculata. 



The interambulacra are especially remarkable by 

 the broad, deep, sharply cut, median furrow, which I 

 find to be distinct in all the specimens before me (PI. 

 IV Fig. 4). (Regarding the quite young specimens 

 comp. below.) The areoles are rather deep, only those 

 nearest the peristome confluent. The circle of tubercles 

 surrounding the areoles rather prominent; outside this 

 circle a few smaller tubercles occur at the outer edge 

 of the plates and some more adjoining the median 

 furrow. These tubercles are considerably smaller than 

 those of the edge of the areoles. 



