SOME SPECIES OF SALICORNIA 179 



flower only reaches about one-third of the way up the segment, 

 whilst in the upper half of the spike it reaches about half-way up. 

 The seed is greyish, not black nor strictly tuberculate, but covered 

 with short stout hairs. 



Eouy (op. cit. p. 60) gives a variety defiexa ; but perhaps this 

 is not distinct or more than a mere state. 



3. Salicornia pbeennis. 



S. ijerennis Mill. Gard. Diet. ed. 8, no. 2 (1768) ! 

 S. radicans Sm. Eng. Bot. t. 1691 (1807) ! 

 Arthrocnemum fniticosum var. radicans Moq. loc. cit. 

 S. sarmentosa Duval-J. in Bull. Soc. bot. Fr. 174 (1868) ! 

 S.fruticosa auct. mult., pro parte. 



S. fnUiculosa Tineo, Cat. Plant. Hort. Reg. Panorm. 280 (1827), 

 might, judging simply by the description, be referred to this 

 species ; but herbarium specimens seem to be distinct both from 

 >S'. perennis and from S. fruticosa. 



S. perennis is with certainty only known to occur in England, 

 France (including south France), Spain, and Algeria, though it 

 may be ultimately proved to have a wider range than this. 



It is often a social plant, growing in matted clumps. Isolated 

 tufts are sometimes a metre or more in diameter. At maturity, it 

 is a tufted dwarf shrub, spreading centrifugally by means of freely 

 rooting branches. The central flower of the cymes is larger than 

 the lateral ones, which only reach about half-way up the segment. 

 The seeds are nearly globular, and covered with slightly curved 

 hairs, which are rather longer than those of the following species. 



Eouy (op), cit. p. 60) divides the species into three varieties, 

 which are perhaps mere soil forms. 



4. Salicornia lignosa. 



S. lignosa Woods, Bot. Gaz. 31 (1851) ! 



Giirke {Plantce EuropcBCB, II. i. 158, 1897) doubtfully refers this 

 plant to the preceding species ; but the two, would appear to be 

 quite distinct. If the two plants are to be referred to the same 

 species, Woods's plant should be given a varietal name — S. perennis 

 var. lignosa. 



Up to a year ago, this species was only known to occur in the 

 south of England. In the spring of 1910, however, I found the 

 plant on the shores of the small Sebka, near Oran, in north-west 

 Algeria. Specimens from this locality are now in the herbarium 

 at Cambridge ; and the plant was also collected in the same 

 locality by Professor C. Schroter and Dr. E. Riibel, of Zurich. I 

 submitted a specimen to Mr. A. Bennett, who wrote : — " Yes ! It 

 seems to me to be exactly like Woods's plant." Doubtless it will 

 prove to be spread over a wider area when the characters which 

 distinguish it from S. fntticosa and S. perennis become more 

 generally known. 



At maturity, it is a prostrate dwarf shrul), leaving the ground 

 by a single stem, which grows mainly in a unilateral direction. 

 The branches, although they lie flat on the ground, like those of 



2 



