90 CEYLON MARINE BIOLOGICAL REPORTS. 



and fasten on to them. C. davifera then often gets somewhat longer vertical axes, and the whole plant 

 assumes an appearance more closely resembling Turner's Fimus davifer, PI. 57, vol. I., and Weber v. 

 Bosse's, fig. 4, PI. XXXIII. 



If it grows directly on the rock and more isolated, it appears as in fig. 13. That for such a pro- 

 nounced littoral alga, which grows squeezed in between Actinians and corals, such a mode of growth offers 

 certain advantages, seems evident to me. Partly because, as mentioned above, the surface exposed to 

 the waves is relatively small compared \vitli the well-developed adhesive system, partly because the 

 exposure of the assimilation system — and this, I think, is worthy of special note — becomes under such 

 circumstances the most favourable, since it is practically spread out in one level and thereby as many 

 branchlets as possible are exposed to the light. That, for instance, the taller vertical axis with pro- 

 nounced uviform shape, as C. uvifera (fig. 15), cannot be a suitable form amongst corals will easily be 

 understood, for if these vertical axes are squeezed in between equally high corals, only the uppermost 

 branchlets, that is the points alone, can be exposed. Therefore short axes with few side branchlets must 

 be considered as especially suitable shapes for such habitats. The reduction in the assimilation system 

 that results from tlie diminished number of branchlets is compensated by their being larger and coarser. 

 It is also in such localities tliat we meet with forms with very large ball-shaped branchlets, Kutzing's 

 Chauvinia macrophysa. 



These observations of mine concerning the occurrence and mode of growth of C davifera in Ceylon, 

 correspond with Madame Weber v. Bosse's note on C. davifera f. macrophysa (''Monographic desCauler- 

 pes," p.362) where we read : "J'ai recueilli la plante sur des recifs de corail : la f ronde de la f . davifera etait 

 enclavee entre des morceaux de corail, les ramules s'etaient par consequent tournes du cote de la lumiere, 

 s'etaient rapproches du cote frontal de la fronde et en meme temps agrandis." This seems to indicate 

 that the assimilation branches were all as in a level at the same time as they were enlarged. 



I have therefore come to the conclusion, as a result both of my own observations and of such 

 notices in botanical Uterature as the above, that C. davifera may be characterized as the coral reef 

 Caulerpa par preference. 



The changes C. davifera undergoes when it grows deeper down can in general be characterized by 

 the vertical axis, as well as the cyhndrical parts of the assimilation branchlets, being lengthened. The 

 /. remota (fig. 14), described by myself, is such a pronounced form from a deep pool. Its vertical axes 

 are very lengthy, are not recumbent, nor root-taking. Between this and the surface form intermediate 

 forms can be found. The /. remota as a form Hving in deep pools is never exposed to the tearing influence 

 of the swell to the same extent as the main form. Its root system is relatively weaker, but the assimi- 

 lation branchlets are longer, slenderer, and more pUable, so that when they wave to and fro they also can 

 derive every benefit from the light. 



That these various forms arise directly through the influence ofUght, so that, for instance, 

 obscuration favours the lengthening of the axes, while bright or intense light causes shortening of the axis 

 system, seems very probable. But this can only be determined by direct experiments. A comparative 

 examination shows, in any case, that the one type is almost exclusively at home in the upper littoral 

 region, the other in somewhat deeper localities. For analytical ecology, however, it is sufficient to state 

 that both kinds of forms in this organization have certain advantages suitable to the different surroundings 

 amid which they live, and can therefore with good cause be considered as adaptations or ecologisms. 



This C. davifera organization shows itself still more closely an ecological adaptation if we compare 

 it with C. uvifera. In the special section I have dealt in more detail with the differences which exist in 

 Ceylon between these two species. In this place I may call special attention to the fact that G. uvifera 

 wiU be looked for in vain on the coral reefs of the south-west of Ceylon. It is this species however that 

 aeems to be the commonest in tlie north. At the present time I do not venture to decide whether this 

 different distribution of the two forms is to be exclusively attributed to the different nature of the 

 shores ; viz. , that on the south-west and south coasts there occur only chffs and rocks, which are favourable 



