106 THE AMERICAN NATURALIST. . [Vor. XXXVI. 
beneath, or another root, — these rhizoidal hairs spread, as a 
closely adhering film of whitish fibers, over the surface touched, 
binding the root to it. The root-hairs assume fantastic shapes 
in order to conform with the irregularities of the rock. If a 
root has, by chance, entered the loose sand, it becomes enclosed 
in a compact cylinder of sand, bound 
together by the myriad of branch- 
ing, interwoven hairs. 
The plant has now taken a firm 
hold on life. Supposing that the 
start has been made aloft in the 
branches of the Amphiroa, the elon- 
gating stem now dips downward till 
it strikes the rock. Thenceforth it 
creeps along, taking, as it goes, a 
firm grip upon every inch. At each 
node it bears a leaf, and each of the 
short internodes produces, on one 
side a supra-axilary bud, and on 
the other what Professor Dudley! 
has aptly termed an *'epaulette " 
of six or eight roots; alternating, 
so that, if one internode has its 
roots on the right side, the next 
will have roots on the left. Thus 
the stem, though in itself weak and 
brittle, keeps a close, broad grasp 
upon the rock, while the wiry 
leaves, buoyed by intercellular air- 
Spaces, stream upward sometimes 
Fic. 8.—Seedling (older) with rou, COT @ length of two meters. Before 
these leaves have been whipped to 
tatters by the waves, their place is supplied with new ones 
from the lateral buds. 
When the rhizome has reached a length of, often, only one 
. or two decimeters, it begins to die away behind. At the same 
time the lateral buds begin to push out for themselves, so that 
n ! Cf. Dudley, R. W., in Zee, vol. iv, p. 381. 

