42 University of California Publications in Botany [Vol. 10 



15 cm. to 18 cm. high. Stroemfelt does not mention the height of the 

 type in his diagnosis, but his illustration, natural size, is 14 cm. high. 

 The size of our plants, their narrow contorted segments, the character 

 of the receptacles, the large firm holdfast and dendroid habit, agree 

 so well with the description and figures of Stroemfelt as to leave little 

 room for doubt as to the affinity of the two. Comparison with the type 

 material may, however, show that the two sets of plants are distinct, 

 as Stroemfelt does not mention the cryptostomata nor the caeco- 

 jtomata, nor does he mention the color, which is very striking in our 

 plants. As the plants hang on the rocks after the receding tide, the 

 shape, appearance, color, and abundance of receptacles reminds one 

 of settled swarms of Italian bees. 



Fucus evanescens f. marginatus f. nov. 



Plate 42 



Frons distincte caulescens, 20-50 cm. alta, regulariter et multo 

 dichotoma, evidenter rigida coriaceaque, yalde contorta, luteola, siccata 

 olivaceo-brunnea, stipite tereti, superne inter ramos diminuante ; seg- 

 ments 9-15 cm. latis, strictis, supra axillas non angustatis truncatis, 

 vulgo linearibus, costa paululum prominente, parum evanida, alis 

 inaequaliter attritis, cryptostomatibus fere absentibus ; receptaculis 

 1.5-3 cm. longis, definitis, integris aut alte bifurcatis, anguste ellipticis, 

 in margine dispositis ; conceptaeulis valde prominentibus. 



Growing on rocks in the middle littoral belt. Sitka, Alaska. 



Type, Gardner, no. 3997 (Herb. Univ. Calif., no. 201163), July, 

 1917. 



The distinguishing characters of this form are absence of con- 

 ceptacles from a complete margin of the receptacle, particularly of 

 the rounded terminal portion, coupled with extreme sparsity of both 

 cryptostomata and caecostomata. 



It is difficult to decide to which species this form is most closely 

 related. It has evident affinities with Fucus evanescens Ag. and with 

 F. spiralis L. Its color and consistency, coupled with the slightly 

 vanishing midrib in the terminal segments, seem to ally it with F. 

 evanescens, but, if allied with this species, it would be through f. den- 

 droides of Stroemfelt as its nearest relative. I have not seen the type 

 nor any authentic specimens of f. dendroides, but since no one has 

 questioned the validity of the form as belonging to F. evanescens, I 

 am retaining it as such and grouping certain Alaska specimens with it. 

 The narrow, rather long, considerably contorted and profusely branched 



