STYLASTERIDAE (HYDROCORALS) FROM SOUTHERN SEAS 41 



are numerous on all sides of the terminal branches, but disappear gradually on the posterior and 

 anterior sides of the branches towards the central parts of the flabellum, although they are present 

 along the sides of the meshes in older parts of the colony. Surface of the coenosteum vermiculated, 

 without nematophores. Gastrostyle pine-cone shaped with bristling spines. Dactylostyles wanting. 

 Dactylozooids monomorphic, gastropolyps with four to six rudimentary tentacles. 



Conopora pauciseptata n.sp. 



(PL IV, figs, i, 2) 

 St. 2493. 42° 03-9' S, oo° 03-5' E, 472 m. 



Only a single fragment of a colony (PI. IV, figs. 1, 2) was found in the catch. The main stem is 

 23 mm. high with a basal greater diameter of 6 mm. Only the basal parts of broken side branches 

 have remained ; their arrangement and construction show that the ramification has been in one main 

 plane, and the stem and branches of the fan exhibit a distinct anterior side with numerous cyclosystems 

 in one main row, whereas the posterior side is devoid of cyclosystems. On the other hand, the stem 

 and branches also have lateral rows of cyclosystems along both sides towards the anterior side. 



The surface of the coenosteum exhibits a peculiar vermiculated or striated structure, which is 

 conspicuous before cleaning with sodium hypochlorite, and which is reminiscent of a finger-print, 

 the same structure observed in Errinopsis reticulum (see PI. Ill, fig. 2), but even more conspicuous in 

 the present species. The finer lines are caused by rows of microscopic pores in the coenosteum through 

 which the surface epithelium communicates with the tissues of the intersecting canals of the stem and 

 branches. 



The coenosteum is not so hard as in most Stylaster species, but is softer and more easily ground, and 

 although it is not so porous as in the species of Conopora previously investigated it compares well with 

 their somewhat more compact construction. The fragment shows no indications of symbiotic worms, 

 but this may be fortuitous. The dimensions indicate that the fragment comes from an old and rather 

 large colony, and one might at all events have expected to find traces of the said symbiosis. This is the 

 second Conopora colony observed without symbiotic worms, and it is of interest to note that the 

 ramification and general flabellate shape of the two colonies agree (Broch, 1942, p. 88, pi. xiii, fig. 37a), 

 although they evidently represent different species. 



The cyclosystems of the median, anterior row of the stem and branches have on the whole many 

 fewer dactylopores than the lateral systems. The accompanying table gives a variation from 3 to 6 

 (average 4-5) in the anterior row, whereas the lateral rows show a variation from 5 to 11 dactylopores 

 (average 8-i). 



Numbers of 



dactylopores ...3 4 5 6 7 9 I0 ll Total 



Anterior row 2442 I2 



Lateral rows — 1 2 6 1 1 6 2 2 30 



Complete cyclosystems occur apically on the branchlets ; otherwise a single, complete system may 

 quite exceptionally be found on the lateral rows. Generally, however, a diastemma is developed at 

 the upper (adcauline) side (Fig. 8). It is not possible to state with certainty whether this diastemma 

 exists from the beginning. But its mode of distribution in the different cyclosystems indicates that 

 a diastemma is caused by and developed during growth of the coenosteum (the colony), and that the 

 growth of the coenosteum gradually obliterates the adcauline dactylopores, whereas the abcauline 

 side of the cyclosystem remains unaltered. In this way the number of dactylopores will also diminish, 

 a fact consistent with conditions in the Discovery colony fragment. 



