Dec., 1910.] Self-Dividing Laminae of Certain Kelps. 
225 
the superficial and hvpodermal cells lining the gap divide chiefly 
with periclinal walls until the newer tissue comes to have a fan- 
like structure, similar to that seen in Nereocystis (Fig. 19). By 
the same process described in that plant though on a much larger 
scale, the furrow is strongly widened and deepened. This results 
as in Nereocystis not only in attenuating the lamina locally but 
in filling up the middle region with young cortical tissue (Fig. 
20) which has no strength and is easily torn apart by wave action. 
After ripping apart the cortical cells exposed from a to b 
(Fig. 20) take on the aspect of epidermal elements with which is 
associated a tendency of the contiguous tissues to close around the 
injured portion. The two lamina edges resulting from the entire 
process of division are lined with new tissue, the superficial layer 
of which becomes epidermal in character and remains so. These 
edges show for some time a groove running in them which is wholly 
accounted for by the peculiar splitting process obtaining in this 
plant. 
GENERAL CONSIDERATIONS. 
That in all cases the split actually progresses through the blade 
is clearly shown. The exposure of the inner tissues lying just 
behind the apex of the advancing cleft is evidence enough to prove 
that the fissure is cleaving the blade and is not a pseudo-cleft 
brought about by intercalary growth of the portion divided by the 
perforation. This latter, however, is of much importance in the 
apparent elongation of the split. 
As to the rapidity of advance of the cleft, nothing whatever 
was determined. It is doubtful that this can be worked out even 
in the field with any degree of accuracy. The facts of intercalary 
growth and mechanical ripping would tend to vitiate any measure- 
ments that might be made. 
In the study of the various genera special attention was given 
to the advance of the distal end of the split, or the division of 
the blade. In each genus however observations on the proximal 
end of the cleft were made, which not only showed that the stipe 
was dividing, but the splitting process was in each instance 
identical with that described for the blade. Owing to the thick- 
ness and compactness of the stipe the changes in it are much slower 
than in the blade. The elongation of the stipes is chieflv bv inter- 
calary growth. 
An arrangement of the five genera on a basis of specialization 
in the splitting would be as follows: Postelsia, Nereocystis, Les- 
soniopsis, Macrocystis and Dictyoneuron. The relative arrange- 
ment of the first three might be questioned but that the process 
in Macrocystis and particularly Dictyoneuron is a very definite 
and specialized one admits of no doubt. 
