122 BOTANICAL MAGAZINE. [Vo i. xix, 



plants of both first and second categories mixed together in the section 

 Arthorocardia Dene, as well as in the germs Arlhrocardia Arescb. But it 

 was clearly perceptible that the former comprised more species of the first 

 category and the latter more of the second. Areschoug classified several 

 species, which belong, in my judgement, to the third category under "species 

 inquiridas." It is beyond doubt that the Arthrocardia meant by both 

 writers differed somewhat. Schmitz united Arthrocardia to Cheilosporum 

 probably because he took a species of the first category as the representa- 

 tive of the former genus. 



This was one of the most vexatious problems I had to solve during 

 the study of the Corallinae of Japan. 



After the anatomical researches on all species at hand I came to the 

 conclusion that the plants belonging to the first and the third category had 

 the propagating cells originated in the medulla while those belonging to the 

 second category had them in the cortex. All of them had nearly equal 

 structure of artieuli and genicula, except in minor points. 



The propagating cells of those species which should be brought under 

 the Euamphiroa Dene, and the Eurytion Dene, are always generated in 

 the cortical part, however the conceptacles may be seated in deep portions. 

 The propagating cells of Corallina, Jania and the Eucheilosporum Dcne # 

 are constantly of medullary origin. 



If the internal structure of the fronds had been counted as the most 

 important matter in the systematic survey, the plants of the three categories 

 should have been classified in the genus Arthrocardia Aresch. If, however, 

 the positions of the propagating cells were taken as of more value than 

 the vegetative characters, a part of the questionable members must be 

 classified in a group related to the Eurytion or the Euamphiroa ; the other 

 in a group near Cheilosporum. 



In my former paper " Enumeration of the Corallinaceous Algre" I 

 referred Amphiroa aberrans, the type of the second category, under Am- 

 phiroa, but separated it from all the rest under the subgenus Pseudarthro- 

 cardia. I believed Amphiroa corymbosa should stand closely to it. Cheilo- 

 sporum yessoense, Cheilosporum latissimum, etc., which belong either to 

 the first or to the third category, brought under the subgenus Arthrocardia 

 under Cheilosporum. Amphiroa californica, Amphiroa frondescens, etc., 

 were transferred to Cheilosporum and mentioned near them. 



Madam Weber van Bosse re-established Arthrocardia Aresch. in her 

 recent work. She distinguishes that genus from Amphiroa by the internal 

 structure, and from Cheilosporum by the position of conceptacles. The 



