The first of these: three layers, the central strand, is the core on which the rest of the 

 plant is built up by an elaboration of the branches which it gives off. The central filaments 

 branch trichotomously as a rule, the middle branch of the three continuing its course as a portion 

 of the central strand; while the two side branches push out laterally and themselves branch 

 repeatedly to form the subcortical layer. Each filament of this layer ends in a crown of small 

 branches which are ilie peripheral cells. 



The growth of Halimeda is discontinuous, as is perhaps suggested by its jointed appearance. 

 Whenever the filaments of the central strand have produced a certain length of internode, 

 thej cease growing and form the resting apex of the joint. At this point they are found in 

 close contact, while in the rest of the joint below they always remain c]uite free from one 

 another. At the apex, commnnication is established between the filaments in one of three ways. 



i. In some species, as was shewn by Professor Askenasy, (1. c.) openings or pits are formed 

 in the adjacent walls of all the filaments and thus free communication is established throughont 

 the central strand. The pits are very large and numerous so that little remains of the actual 

 walls, except on the boundary of the strand, where of course the pits are entirely absent. Thus 

 all the filaments of the strand become welded into one continous whole. (figs. 34, 35, 36% 3Ó b , 

 46 and 48). On renewal of growth however, each filament begins again its separate individual 

 course, continues this course unbranched throughout the node, and then, branching trichotomously, 

 begins the formation of the subcortical and peripheral layers of a new internode (joint). 



2. In other species, there is no general formation of pits, but communication is effected 

 by fusion of the filaments in groups of two or three. The separate identity of the fusing 

 filaments is completely lost and at the end of the resting period which follows the formation 

 of a joint, the fused portion continues on as a single filament for some little distance to form 

 the node ; ultimately the filaments branch di-or trichotomously and begin the formation of the 

 new joint (figs. 4 a , 4 b , n b , 16, 30). 



3. The third method of communication resembles the second in so far as it is limited 

 to adjacent filaments. But in this case the number of fusing filaments is always two and their 

 identity is not completely lost, for immediately after the fusion they appear again as separate 

 filaments (fig. 25). 



As to the duration of the resting period, which follows the completion of each joint, nothing 

 is known. It cannot however be at all prolonged, for a plant with small joints has been known 

 to increase in lensrth bv several inches in 6 weeks, when grown under observation at Funatuti. 

 But so far as I know no other experiment has been made on the rate of growth in this genus. 



The branching of the plant as a whole is preceded by the branching of the central 

 strand, which always takes place near the base of the joint. The several portions, 3 — 5, of 

 the central strand run separately to the periphery of the joint in question, where they form 

 separate apical points. At the next period of growth a new joint may arise from each apical 

 point. In many cases however the central strand has branched without the resultant production 

 of new branches of the plant, as the external marking (ribbing) clearly shews. 



Cei.i.-i 1 'N 1 in i's. < )bservations on the contents of the filaments in Halimeda have necessarily 

 been most unsatisfactory, for want of properly prepared material. 1 am only able to confirm 



