. THE GREEN AI<GAE OF NORTH AMERICA 391 



form-orbicular with cordate base, 1-7 cm. wide, entire or some- 

 what lobed, compact, surface smooth or slightly wrinkled, gen- 

 erally zonate ; stipe cylindrical or flattened, ^-4 cm. high, from 

 a short base ; filaments slender, 6-24 //. diam., in the interior up 

 to 35 n, cylindrical or slightly torulose ; color olive or ashy 

 brown. Fig. 147. W. I. 



Externally distinguished by the shorter stipe, relatively 

 broader lamina, and smoother surface ; but as there is much 

 variation in all these characters, inspection of the filaments is 

 necessary for anything like certainty. 



4. A. LONGICAULIS (Kiitz. !) Murray and Boodle, 1889, p. 

 70; P. B.-A., No. 1479; A. Mazei, Murray and Boodle, 1889, 

 p. 71 ; Rhipilia longicaulis Kiitzing, 1858, p. 13, PI. XXVIII, 

 fig. 2 ;* Flabellum cuneate or oblong, from a usually long stipe, 

 up to 15 cm., sometimes but not always from a basal rhizome ; 

 filaments cylindrical, except for sharp constrictions above the 

 forkings, 28-70 ^ diam., the outer filaments sometimes down to 

 20 p.. W. I. 



Distinguished from A. nigricans by the cylindrical filaments ; 

 in external characters the two are very similar, but A. longicau- 

 lis usually has a longer stipe, less developed rhizome, and more 

 regularly cuneate flabellum. A. Icvis is generally a smaller 

 plant and has a thinner flabellum, but sometimes it is necessary 

 to determine the size of the filaments, which are much smaller 

 in the latter species. 



3. PENiciUvUS Lamarck, 1813, p. 297. 

 Frond penicillate, with distinct, calcified stipe, from the top 



from the same locality are large and well developed, and agree well in 

 habit with A. asarifolia. The filaments of the latter are torulose, 

 but in this respect A. levis varies considerably. In the same paper, 

 p. 28, fig. i, Bbrgesen notes under the name of A. comosa (Bail, and 

 Harv.) Murray and Boodle, a single specimen from the Danish West 

 Indies. Chlorodesmis comosa Bailey and Harvey has been reported from 

 many stations. in the Pacific, but this seems to be the only definite report 

 of it in the North Atlantic. As Bbrgesen speaks of his plant differing in 

 some particulars from the ordinary form, it may be well to suspend judg- 

 ment for the present. Moreover, the inclusion of Chlorodesmis comosa 

 in Avrainvillea by Murray and Boodle seems rather unnecessary, and it 

 would certainly give the genus a much broader and less definite exten- 

 sion than that commonly received and here used. At p. 36, fig. 6, Bbrg- 

 esen mentions as Avrainvillea sp ? a form which may well be included 

 in A. levis as here understood. 



*See Howe, igosa, p. 586; 1907, p. 510, as to the type specimen of Rhi- 

 pilia longicaulis ; Murray and Boodle having first used the binomial 

 Avrainvillea longicaulis, must be considered its authors, although the 

 name does not belong to the plant to which they applied it. 



