504 Howe: Phycological studies 



thallus surface, mostly, however, bounded by narrow reticulately 

 disposed series of sterile cells. [Plate 30 ; plate 33, figures 1-5.] 



La Paz, Vives 2oe (type) and lib. 



No. 2oe, the larger of the two specimens seen, is antheridial and 

 11&, the smaller, is sterile. The plants are somewhat Rhody- 

 menioid in habit, but their structure is rather more suggestive 

 of that of plants of J. Agardh's subgenus P odeum of Gracilaria, 

 though the medullary cells are more nearly empty than is cus- 

 tomary in this group. However, the antheridia appear to har- 

 monize much better with those of Gracilaria, so far as described, 

 than with the wholly superficial or exserted and uninterruptedly 

 expansive antheridia of Rhodymenia. So far as we are aware, the 

 antheridia have been described in only three species of Gracilaria, 

 viz., G. confervoides* G. armata,] and G. compressa.% 



The antheridia of Gracilaria confervoides are described and 

 figured as occupying nearly closed pear-shaped cavities, the bases 

 of which penetrate the subcortical layer. Those of G. Vivesii are 

 more like shallow saucers, often irregular in outline, and they are 

 confined to the rather thin cortex and are covered only by the 

 hyaline outer walls of what were originally epidermal cells; they 

 seem to agree essentially with the descriptions of the antheridia in 

 G. armata and G. compressa, and similar antheridia are present in 

 a specimen from Barbados in the herbarium of the New York 

 Botanical Garden, distributed as "Gracilaria multipartita J. Ag. 

 fronde latiori" (Vickers, Alg. Barb. 136). 



The two nearest relatives of Gracilaria Vivesii are probably 

 G. Cunninghamii Farl.§ (J. Ag. Sp. Alg. 3 4 : 93- 1901), based 

 upon material from Santa Barbara, California, and Gracilaria (?) 

 peruana Pice. & Grun., from Paita, Peru. Gracilaria Vivesii 

 differs from G. Cunninghamii in its brighter red (less brownish) 

 color, in being regularly dichotomous instead of often somewhat 

 tri- or polychotomous, in its less elongate, less cuneate, and broader 



* Thuret, in Le Jolis, Liste Alg. Mar. Cherbourg 134. 1863; Thuret & Bornet, 

 Etudes Phyc. 80-82. pi. 40. f. 1, 2. 1878; Buffham, Jour. Quekett Micros. Club 

 II. 5: 294. pi. 13- f- n, 12. 1893. 



t Thuret, Ann. Sci. Nat. IV. 3: 22. 1855; Thuret & Bornet, loc. cit. 



% Thuret & Bornet, loc. cit. 



§ This specific name appears to have originated through the misreading of a 

 different genitive specific name accompanying a specimen sent to Agardh by Mrs. 

 G. A. Hall. 



