THE CODIUMS OF THE JUAN FERNANDEZ. ISLANDS 59 I 



mature primar\', or pseudoprimary, utricles in the ditterent clusters from any 

 portion of the lobes. N. L. Gardner (Univ. Calif. Pub. Hot., 6: 490—492, 

 pi. 42, figs. 7, 8, 1919) has discussed the relationship supposed to exist between 

 C. diinorphuiit and the plant of the California coast formerly passing under the 

 general name of C. adiuiercns which he describes and figures as C. Setchcllii. 

 The two figures of utricles of C. d'uiiorpJnmi, published by G.ARDNEk, give a 

 clear idea of the shape of the primary utricles, especially of those (his figure 

 8) having the extremely thickened apical membrane (up to 80 a thick) and its 

 stratification as well as the tendencx- towards sculpturing of the external layer. 



Codiiiin adiiacrcns Bailey and Harvey, U. S. Exploring Exp., 17 (Algae): 



165, 1874 (non C. Ag.). 

 Plate 34, figure 4. and plate 40, figures 6—8. 



There are, in the U. S. National Herbarium, 3 specimens, or possibly 

 fragments, of a collection made by the members of the U. S. Exploring Ex- 

 pedition of 1S3S— 42, on the coasts near Valparaiso (almost opposite the Juan 

 Fernandez Islands), which were placed under the old general form name of 

 Codiuin adJiaercns. For reasons not readily understood, they received the her- 

 barium manuscript specific name of •^aiuorplunn,^, a designation which it seems 

 best to allow to remain unpublished. The habit is not entirel}' clear from the 

 specimens. Each may be an erect lobe of a cluster, and this conception seems 

 most likely to be the true one. On the other hand, each specimen may be 

 an individual plant, but in any case it is difficult to determine any exact point 

 (or area.-) of attachment, or any point of disruption from a hypothetical basal 

 thallus possible bearing a number of erect lobes. The existence of these 

 seemingly strongly complanate and probably erect lobes is the distinguishing 

 character of this plant. In these, it seems to difter somewhat from C. diinor- 

 pliuiii Sved., but most definitely from C. cerebrifonne Setchell. It has, also, 

 an entirely difierent utricle cluster from that of C. ccrebriforvie, as comparison 

 of figures on plates 39 and 40 will show. The utricles themselves are more 

 mesophyse» (moderately long and thick 1 instead of ^macrostenophyse;- (long 

 and narrow) as in C. cerebrifonne. They are also psilophyse (hairless). While 

 the utricle clusters of the U. S. E.xploring Expedition plant might be consi- 

 dered close to, if not identical in type, with those of C. diinorphuiii Sved., yet 

 no strictly characteristic groups of the galeate utricle type of C. diniorphuvt 

 have been found, as yet, in the Valparaiso specimens, but they are all sterile 

 and probably young plants. They may, however, best be referred to C. di- 

 morphinn Sved., for the present. There will be, then, at least two species of 

 Tylecoduun, C. diinorphiini Sved. and C. cerebriforvie Setchell, credited to, and 

 presumably endemic to, the Pacific coasts of the South American region. 



