80 TH. MORTENSEN, (Schwed. Siidpolar-Exp. 



be taken into account any more in serious scientific discussions. — It may be ob- 

 jected that the figure represents A. cordatits, while my observations are made on 

 A. cavernosas, and that the differences might thus be due to specific differences. 

 This objection does not hold good, for two reasons: I) it is extremely improbable 

 that two so nearly related species as A. cordatus and cavernosas should prove so 

 highly different in this early stage, and 2) I have examined a corresponding stage 

 of A. cordatns and found it exactly agreeing with A. cavernos7is. 



Pedicellariae are developed already in this stage, both globiferous, rostrate and 

 tridentate, the latter both 2- and 3-valved; they are all of the same structure as in 

 the grown specimens; the rostrate pedicellariae are only comparatively very large. 

 The disk of the tubefeet of the frontal ambulacrum is small, the rosette plates are 

 not yet developed. The tubefeet around the mouth show the first indication of the 

 terminal prolongations. The spicules are few and small. 



The next stage, the youngest of the free- 

 living specimens found in the material, is repre- 

 sented by specimens of 4 mm. length. There is 

 thus a small gap here. Meantime the develop- 

 mental processes, which have been going on while 

 the specimens have grown from 2.5 mm. to 4 mm. 

 length, are easily enough understood. There is 

 a complete series of specimens of the following 

 sizes. In PL IX are represented a series of figures 

 of young specimens (4, 5, 5.5, 6, 6.5, 7, 8, 8.5, 9, 

 12, 16 and 18 mm.) showing the changes under- 

 gone by the test during growth. The tests have 

 been cleaned and treated with alcohol-glycerine and 

 show the limits of the plates upon the whole very 

 clearly. It has, however, been thought necessary 

 also to give some text-figures illustrating the minor 

 features, which cannot be seen exactly in the phototypic plates, but which are ne- 

 cessary for gaining a clear conception of the growth changes of the species. 



In the specimens of 4 mm. an important change is to be noticed, viz. that the 

 transverse branch of the fasciole has been developed (Fig. 19), the whole of the 

 fascioles of the species having thus been formed, viz. the peripetalous fasciole (anterior 

 part of the primary fasciole + the transverse band) and the latero-anal fasciole (the 

 posterior part of the primary fasciole). The transverse band is seen to pass between 

 the apical system and the periproct, traversing the two pairs of interambulacral 

 plates which have meantime developed between the periproct and the madreporite. 

 As growth goes on this transverse fasciole becomes stronger, but it keeps its posi- 



Fig. 19. Abatus cavernosus. 4 mm. 

 Showing the transverse band of the 

 fasciole. 



