PHANEROGAMS— PROFESSOK BAYLEY BALFOUR. 89 



insequilateralia subcrassa nervo medio fusco dimidiatis venis utrinque 3-4-pinnatis ; folia 

 ad ramos spinosos minora foliolaque pauciora. Pedunculi axillares griseo-tomentosi 

 f— | poll, longi ad apices ramorum defoliatorum pseudofasciculati, fasciculi racemum 

 laxum formantes v. subpaniculati. Involucellum annulare trilobatum puberulum sub- 

 caducum. Capitula globosa ; bracteolse spatbulato-lineares apice ciliatae. Mores albi 

 inodori. Calyx -^ poll, longus bracteas sequans, lobis rotundatis brevissimis ciliatis. 

 Corolla \ poll, longa, lobis oblongis obtusis tubo dimidio brevioribus. Stamina triente 

 corollam excedentia. Ovarium stipitatum glabrum ; stylus infra apicem affixus dimidio 

 ovario longior. Legumen ignotum. 



Nom. Veen. Tamhor (Schweinf.). 



Socotra. A handsome tree of the hill slopes of Haghier. B.C.S. nn. 

 212, 345. Schweinf. nn. 459, 519. Hunter n. 17. 



Distrib. Endemic. 



We only obtained specimens of this handsome tree in leaf, and with unde- 

 veloped flower buds, and whilst I could not identify it as a described species, I 

 hesitated to found a new one upon our imperfect material. Schweinfurth 

 fortunately got specimens in full flower, which warrant the constitution of a new 

 species, and I have adopted the name he has suggested. 



The plant belongs to the gummiferous section of the genus. Perhaps its 

 nearest allies are, as Schweinfurth points out, A. arabica, Willd. (Sp. iv. 1085), 

 and A. Wightii, Baker (in Hook. Flor. Brit. Ind. ii. 298). In the inflorescence, 

 size of flower, ciliated calyx, and unarmed flower-branches, it comes especially 

 near A. arabica, though it has more numerous flowers in the heads, and more 

 stamens in each flower. In A. horrida, Willd. (Sp. iv. 1082) there is an 

 inflorescence such as our plant shows, and in A. nilotica, Desf. (Cat. Hort. 

 Par. ed. ii. 208, A. vera, Willd. Sp. iv. 1085), unarmed flowering branches are 

 seen. The exact affinities cannot, however, be determined until we obtain fruit 

 and seeds. 



Like many plants from this region it shows adventitious twigs differing very 

 markedly from the adult form, possessing features which are probably those of 

 the young seedling plant. Only on such twigs (B.C.S. n. 345, Schweinf. n. 459) 

 do we find spines ; the ordinary branches of the adult plant are characteristically 

 bare of them. 



Camels are particularly fond of the twigs of this tree. 



3. Acacia sp. 



We have specimens in leaf of another species of Acacia belonging to the 

 gummiferous section of the genus. It is a small dwarf spiny shrub of the lime- 

 stone plains near the sea. We collected it near Galonsir. Our guides told us 

 it was one of the gum-trees. I have not been able to match it with any 

 described species, and our fragmentary specimens do not permit of a new 

 specific determination. Schweinfurth, to whom I sent a portion, supposes it to 



TRANS. ROY. S0C. EDIN. VOL. XXXI. M 



