HOWE: DIMORPHISM IN GALAXAURA 195 



dimorphism in these peripheral assimilatory filaments, one set being 

 long and another short, and the short ones, more or less even-topped, 

 may sometimes be said to form a loose cortex. Whenever repro- 

 ductive organs are found on plants of the "Rhodura" section, they 

 are always tetrasporangia, never antheridia or cystocarps. In the 

 section "Microthoe," one finds a firm, compact, pseudoparenchym- 

 atous cortex — usually firm and coherent, even after decalcification 

 (text-figure 4). In some of the species or forms belonging in this 

 section, the smooth firm epidermis bears, in certain parts of the thallus, 

 numerous long assimilatory filaments, and, when these are particularly 

 abundant, plants of the section "Microthoe" may look much like 

 those of the section "Rhodura," but, generally speaking, it may be 

 said that the firm smooth cortex of plants of the "Microthoe" section 

 and the rough shaggy exterior of plants of the "Rhodura" section 

 give them a very different appearance and it is no wonder that they 

 have been considered not only as different species but also as members 

 of different sections of the genus. But members of the "Microthoe" 

 section, except when apparently sterile, are always either antheridial 

 or cystocarpic — never tetrasporic — just as members of the " Rhodura" 

 section are always tetrasporic and never sexual. And plants of the 

 "Microthoe" section and those of the "Rhodura" section grow often 

 so closely associated — often intertangled in the same tuft — that it 

 seems to be a fair inference that they represent phases in the develop- 

 ment of one and the same species. Two plants from a collection (no. 

 1859) of about 100 specimens made by the writer near Santurce, 

 Porto Rico, in 1903, are shown on plate IV. Not all of the material - 

 in this collection has been examined microscopically, but, roughly 

 speaking, about 80 of the 100 have the "Rhodura" structure, some 

 of them being obviously tetrasporic and others apparently sterile; 

 and about 20 of the 100 are of the "Microthoe" structure, some of 

 them being obviously antheridial or cystocarpic and others apparently 

 sterile. These plants of the "Rhodura" section appear to represent a 

 condition of what Kjellman described as a new species under the 

 name Galaxaura flagellijormis, though usually less "flagelliform" than 

 Kjellman's original; the plants of the "Microthoe" section represent 

 what Kjellman described as a new species under the name Galaxaura 

 squalida. The two forms, as shown in figures 2 and 3 (plate IV), 

 differ much in habit, yet, if we consider only the lower part of the 

 Galaxaura squalida, where the cortex is more or less covered with 

 free assimilative filaments, it looks a good deal like the shaggy tetra- 

 sporic plant, G. flagelUJormis. These Porto Rican specimens lying 

 under the no. 1859 were collected by the writer in his less experienced 

 and less critical days and were put together under one field number as 



