THE ALGAE OF BERMUDA. 57 



4. C. ISTHMOCLADUM Vickers, 1905, p. 57; 1908, p. 23, PL XXVIII; 

 Collins, 1909, p. 388. Shelly and Gravelly Bays, March, Hervey. 

 Probably not uncommon, resembling a condensed form of C. tomento- 

 sum, but with the divisions more or less contracted at the base, 

 gradually increasing in diameter to the next forking. The utricles 

 are larger than in C. tomentosum, approaching the size in C. decorti- 

 catum. 



AVRAINVILLEA Decaisne. 



1. Filaments distinctly moniliform. 1. A. nigricans. 



1. Filaments nearly or quite cylindrical. 2. A. longicaulis. 



1. A. NIGRICANS Decaisne, 1842, p. 96; Howe, 1907, p. 508, PI. 

 XXVIII, figs. 8-25; Collins, 1909, p. 390. Walsingham, Feb., 

 Harris Bay, April, Hervey; Inlet, Cooper's Island, Aug., Collins. 

 Very variable, from small, delicate plants to coarse, heavy and 

 unsightly ones. 



Var. FULVA Howe in P. B.-A., No. 1480; Collins, 1909, p. 390; 

 P. B.-A., No. 2171. Stouter, coarser, with less difference between 

 stipe and flabellum; color more yellowish than in the type. 



2. A. LONGICAULIS (Kiitz.) Murray & Boodle, 1889, p. 70, as to 

 name only; Collins, 1909, p. 391; P. B.-A., No. 2170; Rkipilia longi- 

 caulis Kiitzing, 1858, p. 13, PI. XXVIII, fig. 2. Walsingham, Feb., 

 Mangrove Bay, Feb., Harrington Sound, Nov., Inlet, Dec., Hervey; 

 Fairyland, Dec., Collins. Growing in company wath A. nigricans, 

 from which it is frequently indistinguishable, except that the fila- 

 ments of the latter show distinctly moniliform on microscopic exami- 

 nation, while those of A. longicaulis are nearly or quite cylindrical. 

 The name is a somewhat unfortunate result of following the rules of 

 botanical nomenclature; the authors of the binomial used it for a 

 different plant, A. nigricans. Rhipilia longicaulis, from which it 

 derives its specific name, is according to the type specimen the 

 present species, while the description and figure given by Kiitzing 

 belong better to a third species, A. sordida Murray & Boodle. The 

 arguments in favor of the name used here are found in Howe, 1907, 

 p. 509; those in favor of preferring A. Mazei Murray & Boodle in 

 Gepp, 1911, p. 27; a later summation will be found in Howe, 1911, p. 

 133. 



PENICILLUS Lamarck. 



1. Surface of stipe smooth. 1. P. capitatus. 



1. Surface of stipe rough. 2. P. pyriformis. 



