12 University of California ruhlications in Botany [Voi.. 10 



represents the original figure of Agardh and shows very little midrib 

 in the terminal segments nor in other parts of tlie frond. ^\\ own 

 experience with this character is that it cannot be relied upon as having 

 verj^ pronounced diagnostic value, except in a few cases. The type 

 locality is said to be Kamtschatka and, according to Setchell, the type 

 specimen is in the herbarium of J. Gr. Agardh, under no. 00299, with 

 a query. Kjellman has described ten forms of evanescens and reduced 

 J. Agardh 's F. hursigerns to a form of the same species. Stroemfelt 

 (1886, p. 35) added f. dendroides. De-Toni (1895, pp. 201-203) 

 accepts all these forms as valid. Setchell and Gardner (1903, p. 283) 

 added two more. ]\Iost of Kjellman 's forms have been recognized 

 with some degree of certainty in the northern portion of the range 

 considered here. The plants so listed, however, seem almost invariably 

 to be more robust than those of Kjellman 's finding. On the whole, the 

 species as understood in Europe, seems to me to be the best suited to 

 include the most of the forms of Fucus growing in the North Pacific 

 region, having more or less coriaceous fronds for the most part robust. 

 with a brownish or yellowish color, wath relatively few cryptostomata. 



Diagnostic Characters 



In attempting to diagnose our forms, I have made use only of the 

 gross morphological features. I think, however, that it would be quite 

 within the bounds of possibility that histological characters could 

 . be made use of effectually in determining limits. The shape and 

 size of the oogonia and antheridia vary considerably, but I have not 

 carried the study far enough to determine ^v'hether a fixed difference 

 may be present and detectable, or whether the difference may be due 

 simply to fluctuating variation. The only satisfactory way to prosecute 

 such study is by the use of fresh or fixed material. 



The following characters have been used in rendering the diagnosis 

 in this work : 



The lioldfast, in all cases, is a solid, somewhat fibrous disk, larger 

 or smaller according to the species and to the nature of the substratum 

 and the environment, which serves to anchor the plant. The plants 

 growing on tlie open coast, exposed to heavy surf, usually have the 

 most robust lioldfasts. Those plants growing on wood in (]ui('1 waters 

 usually have smaller, poorly developed ones. Plants growing under 

 the latter conditions usually do not persist long. 



The stipe is a term ratlnr loosely applied in this group. Strictly 

 speaking, it should apply to that portion of Ihc frond between the 



