THE DEVELOPMENT OK DONDERSIA. 17 



that at first it is somewhat elongate ;ui<l reaches (on the dorsal side?) 

 almost to the ciliated ring, but later becomes circular (Fig. 10 A). 

 The further development of the larva, as described by Pruvot (A and 

 B) makes his interpretation of this depression (at least in the later 

 stages) appear somewhat doubtful. 



About this stage the embryo appears to become free, for the stage 

 depicted in Fig. 10 .4 is called by Pruvot a larval stage. The 

 aperture of invagination has already narrowed, and is terminal in 

 position. 



The further transformation of the larva is said to take place in the 

 following way. From within the invagination a button like structure 

 gradually grows out, which corresponds to the posterior end of the 

 animal (Fig. 10, B and C), and carries at its end the remains of the 



Fig. 10. — A-O, stages in the metamorphosis of Dondersia banyulensis ; A and B, larvae 

 36 and 100 hours old respectively ; C, young animal, seven days old, immediately 

 after metamorphosis (after Pruvot). e, aperture of invagination ; wk, ciliated ring ; 

 sp, calcareous spicules; dp, dorsal calcareous plates (dorsal spicules). 



blastopore. This latter must be regarded as having been previously 

 displaced inwards. With the buttondike terminal region a conical 

 part is also gradually evaginated (B), and this is destined to yield 

 the greater part of the definitive body. On this the leafdike spicules 

 are already appearing (B, sp). These are said to arise, as asserted for 

 GTii'fo/i (p. 11), as intracellular secretions, only breaking out of these 

 cells as they increase in size. The conical part now grows consider- 

 ably, and new spicules continue to be formed ; the anterior part of 

 the larva, on the contrary, degenerates rapidly, and finally appears 

 only as a small collar at its anterior end. The larva is at last no 

 longer able to swim about in the water, for the cilia degenerate 

 together with the cells that carry them. The two posterior rows of 

 cells also (those of the third section, Fig. 10, A and B) are thrown off. 



