Cost aria , Undaria , and Laminaria . 
705 
soon as the blade has attained to the length of 7*5 mm. or more, a con¬ 
spicuous number of mucilage glands may be clearly seen under a hand lens. 
As an exceptional case I found a frond with the blade about 2 cm. in 
length and the midrib running half of its whole length—without the slightest 
doubt of its being Undaria —in which no gland could be detected. 
The glands are far more numerous in the embryonal blades than they 
are in the youngest ligules of a more developed stage. They are denser in 
the marginal region than in the middle of the blade and the transitional 
region. Along the margin of a blade about 2-5 cm. in length I counted 
120 glands in a square millimetre, taking both surfaces together. 
The formation of the glands takes place after the hyphal structure has 
appeared in the meridional area above the transitional region. But when 
it has commenced its work, the glands are generated in the two-layered as 
well as in the more complex area (Fig. 34, PI. LV). Hence, the formation of 
the glands is related to the age of a plant and not to the structure. In other 
words, the glands are formed after the epidermal layer has attained to 
a certain stage of development. An analogous case has been pointed out 
in the formation of cryptostomata in Costaria. 
In my former paper 1 I have remarked that each glandular cell, as 
a rule, originates from a single cortical cell which is in contact with the 
epidermal layer. In the majority of Laminariaceae, the peripheral layers 
of the blade have distinct features from those of the underlying tissue in the 
cellular arrangement and in the cell-contents. In my use of the term ‘ cortical 
cells * I follow the practice of many algologists and mean the cells which 
constitute the layers interposed between the epidermal layers and medulla. 
It is a well known fact, however, that in the stipes the peripheral cells are 
hard to distinguish from the underlying layer, as the former give rise to the 
latter by the formation of tangential walls. German botanists include the 
whole portion of such tissue, in the stipes as well as in the blades, which has 
nothing to do with the hyphal cells, under the name ‘ Rinde ’. Setchell 2 
applies the term ‘limiting layer ’ to the peripheral portion of the ‘ Rinde’- 
This layer is characterized by having its cells much smaller than those 
of the underlying layer. In a young blade which has a single layer of large 
parenchymatous cells between the well defined epidermis and medulla, he 
calls the interposed layer ‘ cortex ’. And, in a more complex structure, 
he applies the term ‘ outer cortex ’ to the layers of small cells beneath the 
epidermis, and ‘ inner cortex ’ to those of large parenchymatous cells near 
the medulla. The outer cortex in his sense is nothing but the ‘ limiting 
layer ’ minus the epidermis. Barber 3 also noticed in the stipes of Sacco- 
rhiza bulbosa that the peripheral two layers are constructed of much smaller 
1 Yendo: Mucilage Glands of Undaria. Ann. of Bot., vol. xxiii, No. xcii, 1909, p. 621. 
2 Setchell: Concerning the Life-history of Saccorhiza , p. 200. 
3 Barber: 1 . c., p. 53, Fig. 11. 
