ECHXNOCYAMUS PUSILLUS. 55 



animal is examined from the ventral surface, the young teeth thus acqui- 

 ring an oblique position. Moreover, the tooth does not form a plane 

 triangle, its upper or rather inwardly directed surface being concave, 

 the opposite convex, PI. IX, fig. Ill a — d. If we subject the convex 

 surface to a careful investigation we see clearly that a keel runs ivoxw 

 the top to the middle of the base and that another low ridge passes 

 along the base and close to it, crossing the keel, fig. c — d. This being 

 so, I am much inclined to believe that also the teeth originate as small 

 tetrahedrons. During the increase the tooth gradually assumes the shape 

 of an elongate isosceles triangle with the long sides curved inwardly. 

 In contrast to the alveoli, plates etc., which are composed of a calcareous 

 net-work with unequal and irregular meshes, the tooth in this condition 

 presents itself as a thin imperforate lamina of great transparency. 



Already in a very early state, when the deposit reaches a size 

 of only 0,008 mm., two new very minute triangular laminae appear placed 

 side by side at the base of the first lamina and on its inner concave 

 surface, fig. 111., b — c. The sides of these triangles are unequal, but 

 in such a way that they form together an equilaterally triangular 

 lamina, their adjacent sides being subsequently overlapped and during 

 arrowth more and more united. The coalescence begins with the ang-les, 

 which are directed towards the top of the future tooth, and extends 

 by degrees backwards, the bases still remaining separate for some time. 

 The lamina, arising in the way just sketched, arranges itself close to 

 the inside of the primary lamina, assumes the same curvation and in- 

 creases forwards as well as backwards, so that the paired base ex- 

 tends further back than that of the primary lamina. Successively new 

 laminse arise inside the former and increase in the same way, the result 

 being that the tooth gradually elongates and gets thicker. At the same 

 time as this is going on, the long sides of the triangular laminas be- 

 come more and more. bent towards each other, so that they finally meet 

 and partly join with their anterior parts. Owing to this, each lamina 

 almost acquires the shape of a hollow cone, and the whole tooth pre- 

 sents itself as a series of cones pushed one into the other, PL IX, fig. 

 112. During this developmental process, the increasing tooth becomes 

 slightly arcuate in the direction of its length. Seeing that new laminae 

 incessantly arise at the base, the tooth is necessarily forced to protrude, 

 so that its top gradually advances to the centre of the buccal membrane, 

 which still remains unpierced. Such a tooth of a sea-urchin fifty days 

 old has a length Of 0,08 mm. and a breadth of 0,024 mm., and its top 



