488 EEPOET— T881. 



seaweeds, being too sliingly and sandy. On the dry sandy plains the 

 vegetation typical of the desert regions on the mainland reigns. Small- 

 leaved, stunted, and woody bushes and herbs, often so rigid as to become 

 spiny, or fleshy plants without foliage-leaves prevail. Leaving the plains, 

 and passing to the hill slopes and valleys, plant life is more vigorous, 

 but in no place sufficiently so to call for the designation of forest, nor 

 is there anything in the way of fine timber. But in the valleys, wherever 

 there is any degree of moisture, small trees of some 20 to 25 feet, Avith 

 smaller shrubs packed so densely as to exclude the light from above, 

 linked together by far-reaching lianes, and underlain by a thick under- 

 scrub of fern and herb, make an almost impenetrable thicket, and produce 

 a verdure quite tropical in its luxuriance. Once out of the valleys 

 upon the plateaux and the scene is essentially different. Wide barren 

 stretches of grey limestone extend on every side unrelieved, save by an 

 isolated Draccena, or tree Euphorbia of stiff erect habit, looking like 

 the remnant of the vegetation of some old geological epoch ; or where 

 a lake-like depression, with its brown earth sparingly coated with 

 green herbage, intervenes. And again, reaching the higher altitudes 

 on the granitic range, the vegetation impresses one at once with its sub- 

 temperate character. The arborescent type has almost entirely dis- 

 appeared. Twiggy, narrow-leaved herbs form a dense deep carpet on 

 the soil, interrupted here and there by a j^rotruding lichen-covered 

 boulder, and for all the world like the covering of heather on a Scottish 

 moor ; whilst within the shade of the boulders, or in the moisture of the 

 overhanging clifis in the ravines, bright green herbs nestle in beds of liver- 

 wort and moss, so that it would requii'e no very great effort to believe one 

 was exploring an Alpine crag in a temperate region. 



Aromatic odours are a marked feature of many plants, and also the 

 occurrence of gums and resins, which in some cases form natural exu- 

 dations in the form of tears. The common desert characteristics of a 

 glaucous grey colouration or a hairy pubescence mark also many of the 

 plants on the shoi'e-plains and on the plateaux. 



I shall not give any statistics regarding the flora, nor shall I attempt 

 any detailed account of its affinities at the present. Our collections, 

 amounting to about 700 species, of which 550 at least are Phtenogams, are 

 only in course of being worked out, and I am daily expecting a con- 

 signment of this gathering from Schweinfurtb, which he tells me includes 

 probably 200 species, that we did not obtain, and I consider it better 

 therefore to delay until the materials for the estimation of the whole flora, 

 as we are able to know it, are at hand. 



The flora is, as you will readily believe, a pretty extensive one. There 

 is in it a goodly number of cosmopolitan and tropical weeds, but there 

 is a fair proportion of endemic genei'a and species. The orders most 

 abundantly represented appear to be Leguminosas and Graminea?, closely 

 followed by Compositae, Acanthace£e, Cyperacere, and Euphorbiaceffi. 

 There is a fair number of ferns, a few orchids or palms, whilst of cellular 

 Cryptogams lichens are exceedingly abundant. 



Of individual plants interesting for various reasons, let me merely 

 mention a few,- — and firstly, on morphological grounds, may be noticed 

 the ' Camhane ' tree, a new genus of Cucurbitaceas. This plant differs 

 from the ordinal characters in being a tree with a stem often four or 

 five feet in diameter at the base, rapidly tapering, and forming a very 

 soft juicy wood. Another plant of interest, on morphological grounds, is 



