97 



Plants of this species are collected in several localities, namely 

 Osliima (Idzu), Cape Omaizaki (Totomi), Tosa, Iyo, Cape Nomo 

 (Kjellman) and Kagoshima. The two largest specimens of typical forms 

 in my possession are those from Evagoshima which measure 50 by 35 

 and 45 by 40 cm. 



In the diagnosis of EcTdonia radicosa I described the sori of this plant 

 which is not quoted in de Toni's Syllog. Alg. III. p. 357. I shall, 

 therefore, once more repeat here by the way the line which runs as 

 follows : " Sori on both surfaces of the frond, at first distinct and 



roundish, at length more or less confluent into irregular patches 



The sori are formed not only in the median portion of the frond but 

 also at the base of the lobes. The Zoosporangia are linear and densely 

 packed among cuneiform paranemata ; they are 50-70 y. long and 10// 

 broad. 



Laminaria angusta Kjellm. 



Nom. jap. Naga-hombu. 



Some remarks about the structure of the root and lamina should 

 be added here to the description^ of this species given by Kjellman. 

 The median fascia, which is remarkable in this species in its linear 

 aspect looking like a median costa, is well marked out in wider and 

 especially fructified fronds, in which it is not generally covered by 

 sori. But in sterile fronds and narrower forms it is not so marked. 

 Moreover, lacunas muciferee are not entirely wanting in lamina, as 

 Kjellman asserted in his description of this species. They are some- 

 times apparently wanting in the lamina of some specimens, or if they 

 really exist there, are very difficult to detect ; in others they are clearly 

 seen, in which case they are situated beyond the middle point in the 

 breadth of the middle layer. They, when present, are rather small and 

 are either provided or not provided with small peculiar cells at periphery 

 (Fig. 5). They are always wanting in the stem. But, peculiar to say, 

 lacunar muciferase are again present in root-fibres even in abundance. 

 In a cross-section of a root-fibre we find beneath the cortical layer a 

 very close ring of roundish middle-sized lacunas. 



Narrower forms of Laminaria japonica f. angusta often mislead us 

 in identification, as they assume an appearance very closely resembling 



i) Kjellm. et Petersen Om Jap. Lam. p. 266. 



