54 Hjalmar Théel, 



of the star, which proceeding appears to take place in a somewhat different 

 manner. In the meantime, the rods themselves unite with their conver- 

 ging points and gradually swell towards the top, so that finally the 

 spherid as seen from above presents itself as three globes closely 

 united and with a minute hole in the middle. This stage, which mea- 

 sures about 0,018 mm. in length and 0,o24/mm. in breadth, is the oldest 

 I have had at my disposal. According to Loven *) the mature spherids in 

 Echinocyamus have a height of 0,09 mm. and a breadth of 0,ii mm. 

 Considering this fact, it is clear that the calcareous covering, in which 

 the spherid is concealed, ought to change from time to time in size, 

 but this evidently cannot take place without a resorption followed by a 

 new formation. 



Almost at the same time as the spherids begin to originate, the 

 first indication of the dentary apparatus appears. A careful examination 

 of the ventral region shows that inside the buccal membrane in each 

 interradium three minute calcareous deposits become visible long before 

 the mouth opens externally, PL VIII, fig. 105 — 106. Of these deposits 

 the middle one in each interradium early assumes a singular shape and 

 structure, being designed for a tooth, fig. 106 t. Shortly after the 

 appearance of the deposits above mentioned, three other calcareous spi- 

 cules of minute size gradually become visible in each radium inside the 

 buccal membrane, but somewhat deeper than the primary deposits, fig. 

 105. Altogether then, the Aristotle's lantern is built up from thirty 

 centres of calcification, fifteen radial and as many interradial, five of 

 which develop into teeth. The twenty-five spicules arise in the same way 

 as the peristomal plates and grow into as many irregular pieces of a 

 meshy tissue, which coalesce or join so as to constitute the alveoli, 

 which form a kind of socket in which the teeth are set. The manner 

 in which the changes of the different spicules take place in order to 

 raise the lantern will be understood from the figures, PL /A', fig. 

 108 — 109. The adjacent surfaces of the alveoli become united by 

 strong transverse muscular fibres. While the alveoli are made up of 

 a calcareous network, the meshes of which are unequal and in certain 

 places almost disappear on account of the vigorous increase of the 

 bars inclosing them, the teeth themselves present a structure quite 

 different. When first distinguishable, they have the form of a ;ni- 

 nute almost equilateral triangle with one angle towards the centre of 

 the buccal membrane and the opposite base placed lower when the 



1) Etudes sur les Echinoïdées. 1874. p. 8. 



