34!6 THE JOURNAL OF BOTANY 



The nearest ally of this plant seems to be S. remotispiciila 

 Laeaita, but it may be separated from that b}^ its whole habit being 

 more divergent, its numerous sterile branches, its more zig-zag scape 

 Avith branches more spreading, its + patent and shoi-ter spikes, its 

 larger spikelets and bract proportions. From the numerous forms of 

 S. virgata distinguished by its revolute-margined leaves, smaller and 

 less curved and lighter-coloured spikelets, bract proportions etc. 



Distribution. Jugo-Slavia. Dalmatia. Kiirste bei CannosaN.W. 

 von Ragusa ! 1906, A. Ginzberger and R. Wettstein. Dr. Ginz- 

 berger writes : " Cannosa is the Italian name of a village whose 

 South-Sclavian name is Treteno : it is situated on the eastern coast 

 of the Adriatic, thirteen kilometres to the north-Avest of Eagusa. 

 The coast-rocks wdiich the Statice inhabits fall steeply to the sea and 

 consist of limestone." 



Explanation of Plate 5Qo. 



1. Stat ice anfracta C. E. Salmon ; 2, outer bract ; 3, middle 

 bract ; 4, inner bract ; 5, bracteole ; 6, calyx — ^all enlarged four 

 times. 



ANTITHAMNIONELLA, A NEW GENUS OF ALGiK. 



Bt Lilian Lyle, F.L.S. 



In October 1921 I gathered in Guernsey an epiph^'^tic alga 

 belonging to the Uliodophycece,, which proved very puzzling. The 

 plant branches alternately and bears in addition whorls of small 

 ramuli at each joint, thus indicating affinit}^ with Antitliamnion. 

 Farlow% indeed, in his Ilarine Flora of Neic England (p. 121), had 

 used the presence of these wdiorls as a distinction between that and 

 Callithamnion; the filaments of Antitliamnion, he says, "are of two 

 kinds, the main filaments being indefinite and the branches definite, 

 so that we have indefinitely elongating stems clothed with short 

 definite branches, or, to use the expression of Nageli, Avith leaves." 



The triangular division of the tetraspores, however, distinctly 

 excludes the Guernse}^ plant from Antitliamnion. With Callitliam- 

 nion and Bpermotli amnion it agrees in the triangular division of the 

 tetrasporanges, but from the former it is separated b}^ the A^erticillate 

 character of the ramelli, the absence of cortication in the older parts, 

 and the presence of discoid rhizoids; from Spermotliamnion, though 

 it agrees in the possession of discoid rhizoids, it differs in the mode 

 of branching and shape of tetrasjjoranges ; the general character 

 of the species of this genus is more rigid and lax than that of the 

 alga in question. 



It is difiicult to account for the presence of this alga in British 

 Avaters. The only plants approaching it in appearance or structure 

 belong to the Southern Hemisphere, S. Africa, and Cape Horn. 

 A. sarniensis belongs probably to sOme region hitherto uuAvorked for 

 algse, and has travelled to the shores of the Channel Islands by one 

 or other of the means of dispersal possible for algse-^/. e. currents, 

 ships, intestines of birds, packing, etc. 



