Cost aria, Undaria , and Laminaria . 
703 
considerable speed, the breadth not following the growth in the same ratio. 
Late in spring we find many narrowly lanceolate blades, 30-50 cm. in length 
and 3-4 cm. in breadth, with the base gradually attenuating towards the 
stipe. At the beginning of summer the growth in length seems practically 
to have ceased in the blades and the increase in breadth becomes remarkable. 
The wearing away of the terminal portion of the blade proceeds downwards. 
As in other costated species the membranous part decays off one step 
further than the ribs. Finally the general outline of the blades become 
ovate with a broad cordate base (Fig. 29, PI. LV). 
At the stage when the growth in length of the blade becomes somewhat 
slow, a number of longitudinal ridges and furrows make their appearance on 
the stipe. They run longitudinally from the transitional region and dis¬ 
appear before reaching the holdfast. In dried specimens they are liable to 
be mistaken for the wrinkles caused by shrinking of the stipe on drying. 
The ridges have undoubtedly a certain connexion with the ribs in their 
position, but there seems to be no apparent rule in their relation. The 
intercostal area, however, becomes narrow at the transitional region and is 
always continued into one of the furrows. The ribs themselves have 
a furrow on each at the base of the blade. Some of the furrows are con¬ 
tinuous for the whole way, but others are discontinuous (Fig. 29). 
The cross-sections of the furrowed stipe show the arrangement of the 
subcortical cells in a peculiar manner. In general, the parenchymatous cells 
are arranged radially, as in other members of the Laminariaceae. Towards 
the depressions, however, the cell-rows gradually converge and in the 
elevations they diverge (Fig. 31). In the latter case the cell-rows do not 
radially split to increase their number fastigiately, but each cell below the 
anticlinal point is much stretched tangentially, as shown in Fig. 31. What 
makes it necessary to have such a peculiar construction in the stipe is 
at present beyond our comprehension. For the purpose of mechanical 
strength the structure of the ridges seems to be hardly expedient. Barber 1 
has called attention to the fact that the ridge peculiar to the stipe of 
Saccorhiza bidbosa does not result from a rapid multiplication of the cortical 
cells, but from their increase in size. I mention this here as a curious 
coincidence in such widely different cases. 
Harvey 2 described a perforated specimen of Costaria Turneri as a 
special form, calling it forma pertusa. It is, however, a rather aberrant 
case to find a frond of Costaria Turneri at an adult stage without any per¬ 
foration. Generally speaking, a few holes are completed when a frond has 
attained some 20 cm. or so, and gradually increase in number as the blade 
extends. In the area above the transitional region, the holes are especially 
numerous, but in the upper portions of the blade new holes may also make 
1 Barber : Development of Laminaria bulbosa. Ann. of Bot., vol. iii, No. 9, p. 52. 
8 Harvey : Characters of New Algae, &c. Proc. of Amer. Academy, vol. iv, 1859, p. 329. 
3 A 2 
