100 CEYLON MARINE BIOLOGICAL REPORTS. 



3. Dwarf Forms. 



Stunted dwarf forms are often looked upon as a special kind of variation. These, too, are not 

 rare among Caulerpas. Dwarf forms in general are produced by poor nourishment, and it is to be assumed 

 that this has also in some way occasioned the Caulerpa dwarfs. And it seems to be clearly the case in 

 C. taxijolia f. intcrrupta (fig. 6), in which stunting has perhaps been produced by lack of light, owing to 

 the great depth, for Caulerpa, at which it grows (more than 10 m.). In the other cases there are, on the 

 other hand, no directly apparent external circumstances which could justify our caUing the cause poor 

 nourishment. On this point, however, it is impossible to pronounce with any degree of certainty, since 

 our knowledge of the requirements in nourishment of the algse is still so fragmentary. If, now, it is 

 some deficiency or other in the nourishment that has caused them, this type of variation must, according 

 to present day opinion, be classed under fluctuating variability among the different kinds of branchlets, 

 and deserves then to be distinguished at the most as local forms. If, on the contrary, these dwarf form.s 

 occur not simply isolated among the normally developed ones, but show a certain constancy within a 

 given district, it is more probable that we have to do with special races. For it is evident that, since we do 

 not know what influence the different external conditions and especially those of nourishment exert, and 

 since we have no opportunity of conducting cultural experiments, it is absolutely impossible to 

 determine objectively whether we have to do with individual variations or with races with constant 

 character. In such cases it must often be a matter of taste for the taxomonist if the plant in question is 

 to be classed as a race or only as a local form. 



The dwarf forms that I observed in Ceylon were : C. taxifolia, f. interrupla (fig. 6), C. plumaris, 

 f. umhellata (fig. 9), C. Icetevirens f. depauperata (fig. 20) and C. parvula (figs. 43, 44). Of these it seems 

 to me that some doubt arises only in the case of C. parvula (fig. 44), i.e., whether it is not rather to be 

 regarded as a distinct species, since, at Beruwala, where I collected it, it occurred in several places and 

 seemed to be a more or less constant species. I have, therefore, classed it as a separate species. As for the 

 others, they showed all the signs of being mere accidental varieties, which diverge so much, however, 

 that I tliink we are justified in distinguishing them as special forms. 



4. Summary. 



In the preceding we have thus got to Know different kinds of shoot variations which can occur 

 in one and the same Caulerpa. And we liave seen that these are of the following kinds : — 



L Variations whicli depend on the locality, and which are to be considered as adaptations 

 or ecologisms. 



2. Variations which cannot be considered as ecologisms, but which are the result of fluctuat- 



ing variability amongst the different branchlets (= pinnules). 



3. Variations which can be considered as phylogenetic stages of evolution (for instance, that 



the basal branclilets or pinnules are of more primitive form than the upper branchlets). 



4. Bud variations of atavistic origin (C crassicuulis f. mixta?). 



5. Dwarf forms. 



6. The variations which do not fall under any of the above categories may, lastly, be bud 



variations without atavistic origin (= mutations). 



As Reinke has pointed out, it is to be noted especially about Caulerpa tliat no sharp line can 

 be drawn between individual variations and bud variations owing to tlieir characteristic mode 

 of propagation, since all the Caulerpa individuals originate as buds on a common horizontal axis, and 

 thus all variations are bud variations in the widest sense of the term. (Cf. the relation between onto- 

 geny and phylogeny in Caulerpa, p. 16.) 



All tliese factors again work together and cause the multifarious diversity of Cattierprt. And here 

 it is to be noted — as several writers have already pointed out — that all this diversity and all these variations 



