Mli. P. H. CARPENTEU ON TUE GENUS ACTINOMETRA. 453 



median line of each plate remain in connexion with one aiaother. 

 Not unfrequently the triangular interradial process, which ia deve- 

 loped from a secondary calcareous deposit on the ventral side of 

 the original basal, becomes more or less completely united with 

 these primary bars connecting the lateral portions of the basal. 

 The latter retain their primitive relation to the first radials ; 

 for they remain united with them aloog the inner margins of their 

 dorsal surfaces ; and as they partially cover in the dorsal aspect 

 of the bifurcating nerve-cords which enter the central canals of 

 the first radials, the author has called them the basal bridge. 

 The interradial processes of the rosette of A. rosacea may also ex- 

 hibit departures from their normal triangular shape : not unfre- 

 quently they become long and spout-like, with inflected parallel 

 margins, which are so applied to the projecting and similarly in- 

 flected outer edges of the adjacent openings of the central canals 

 in two contiguous radials as to convert the interradial furrow 

 lying between them into a complete axial interradial canal, pre- 

 cisely similar in character to the axial radial canals already 

 described. 



This occasional spout-like condition of the interradial processes 

 of the rosette of A. rosacea is of considerable interest, as it is 

 normal in Actinometra, in which genus also the ends of the alter- 

 nating radial and interradial processes are always connected by a 

 basal bridge, the unabsorbed remains of the outer margins of 

 the embryonic basal plate. 



From the outer ends of each of the interradial processes of the 

 rosette of Actinometra, where it unites with the two bars of the 

 basal bridge proceeding from the radial process on either side, 

 there extends, for a longer or shorter distance towards the peri- 

 phery of the calyx, a prismatic or cylindrical rod, which is received 

 in an interradial furrow on the ventral surface of the centrodorsal 

 piece. These five rods, to which, taken together, the author has 

 given the name of the basal star, vary very greatly in the degree 

 of their development, not only in different species, but in diiferent 

 individuals of the same species, and to some extent also in the 

 same individual. 



The reason of this is that these rods are not like the other 

 pieces of the skeleton, calcifications in a nucleated protoplasmic 

 network, but they are simply formed by a more or less complete 

 deposition of calcareous matter in the five interradial planes around 

 the fibres of connective tissue which effect the synostosis of the 



