202 CANARIAN AND MADEIRAN CRASSULACES. 
which these plants have paren it may be worth = to are the 
mixture of species which have been associated with it. There are 
pir packets each labelled E. Dorame. The anit (sof sie 
my notes record) only one sheet from Gran Can It i 
sled ws ith the habit of S. barbatum, and was at ret so  Jabelled. 
Can it represent the Honiwn Bentejui of Webb’s syn. ine 
which nothing definite seems to be known? The second packet 
contains three sheets :—No. dy. a plant with ss flowers and 
i if I rightly remember, the same. No. 3 is marked “© 882 8. 
in ies colour of its flowers, and in the ae ee almost scurfy 
clothing of the flowering becahon. Mr. Gelert, who was kind enough 
Tenerife (Batranco d de Mart cen. Hierro (El Golfo), and Lanzerote 
(El . There are specimens from Tenerife and Hierro at Kew, 
but I sepsall think that they are rightly referred to S. arboreum. 
Indeed, so far as the Barranco Martianez is concerned, I feel 
perfectly sure that 8. arboreum does not occur there, but that S. 
oe was ig for it. I think that the Hierro pen 
: av 
plant. But in 1894 (April ») I gathered a plant at El Dragonal 
Gran Canaria which I was able with considerable assurance to 
specimen was in flower. Bolle says of his Aonium Manriqueorum, 
‘Hab. in Canaria Magna frequens: ie Vega de S. Brigida: Barranco 
de Tenteniguada: El Dragonal: Monte Doramas.’ It is a great 
reer to have solved (as I hope) the problem of the original 
home of S. arboreum. I believe that all the European and 
African localities quoted for it are open to doubt, and I aioe 
i ge garden escapes, naturalized,” would be their proper de- 
scriptio 
