KUNGL. SV. VET. AKADEMIENS HANDLINGAR. BAND 51. N:o ll. 13 
of coenenchym, and the anthocodiae are mainly placed on these thickened, arched parts, 
whilst the side parts and the lower parts of the branches contain few or no anthocodiae. 
In the coenenchym between the coelenterons we find a richly developed system of endo- 
dermal cell-strings and canals, which normally bring about an indirect connection between 
the coelenterons (P1. III fig. 28). Thanks to the numerous polyps, a great crowding 
together is caused in the inner coenenchym, and, in consequence, a more direct connection 
results. The direct canals, which present themselves in the case of the species here 
mentioned are, however, comparatively few. The direct canals have appeared particu- 
larly in places where a powerful indirect connection exists. Lower down on the branch 
or the colony, where part of the younger polyps have disappeared and but few new 
polyps have made their appearence, the space and the coenenchym between the coelen- 
terons are better developed, and the direct canals disappear, whilst the indirect connec- 
tion continues with copious development of canals and strings. 
But in the case of Gersemia fruticosa, the polyps are placed differently. The bran- 
ches are more extended and are longer, and the anthocodiae are placed not only on the 
terminals of the branches, but also along the sides for nearly the whole length of the bran- 
ches, generally a considerable number at the same height. This has the result that the 
number of the coelenterons remains constant, and that new coelenterons of polyps from 
the lower part of the branches appear in the place of the coelenterons, which disappear. 
We find in the case of this species, too, a system of endodermal strings and canals, which 
constitute an indirect connection between the coelenterons of the polyps (Pl. III, fig. 
30). But this form of connection is often replaced by direct canals, which arise in con- 
sequence of the crowding between the coelenterons. This occurs when within the 
branches a great number of coelenterons are added. When the number of the coelen- 
terons is reduced again by disappearence of the younger coelenterons, the direct canals 
also disappear (P1. III, figs. 30, 32). The crowding between the coelenterons also can 
be repeated many times on the same branch. The crowding can either arise amongst 
all coelenterons in the branch or only amongst a part of them. The direct canals partly 
occur between coelenterons of older polyps and coelenterons of younger ones, partly only 
between coelenterons of older polyps. A direct connection between the same coelenterons 
of older polyps may be repeated many times on the same branch. When new polyps 
arise often, the indirect canals and strings are numerous, and the budding of the new 
polyps has arisen from these. The direct canals do not denote the way of budding of 
new polyps. The direct canals between the younger coelenterons and surrounding older 
coelenterons are numerous. 
In the case of certain stunted, small specimens, for instance Gersemia fruticosa 
forma arctica and forma loricata, the direct canals between the coelenterons occur as large, 
wide openings. Within these specimens the trunk and branches of a colony are but little 
developed, the coenenchym of the trunk and the branches also is unimportant. The 
number of polyps is relatively large and the crowding between the coelenterons is great. 
Through the cessation of increase in the colonies the crowding between the coelenterons is 
retained and the direct canals at first developed also retained are within the whole trunk 
or entire branches. The indirect cell-strings within the coenenchym between the coelen- 
