288 



ANGOLA 



geographical miles southwards, at the foot of the first terrace 

 from the river Zaire to Cunene. The hilly tract itself, which iu 

 its greater portion is rocky or stony, is often intercepted by 

 extensive sandy plains, rising slightly, and extending inland to 

 the foot of the first chain of mountains, upon which the above_ 

 named trees mostly appear as bushes or low dwarfy shrubs ; and 



them are also observed the M<srua Angolensis and other 

 Capparideae, several LoganiacesB and Acanthaceae, bushes of a 

 species of Balanites^ dwarf specimens of Ilyphcene and a Bau- 

 hinia. Almost the whole of this region, which we may call the 

 inner littoral region, bears a dull desolate aspect ; and it is only 

 in slopes and in deep ravines, where a stratum of red clay may 

 be seen, that a varied and luxuriant vegetation occurs, which 

 gradually increases towards the small lakes and rivulets, and 

 which, on the banks of the large rivers, manifests all the rich- 

 ness and variety peculiar to equatorial countries. The higher 

 slopes of the hills next to the mountains are covered more or less 

 with extensive forests, which, according to diff*erent latitudes, con- 

 sist of Diospyrese and Combretacese, or sometimes of Sterculi^ and 

 Cynomatise, and in the southern districts of Mossamedes, of the 

 Copaifera Mopane^ lately, described by Mr. Bentham in the last 

 published part of the Linnean Transactions. In the hills and 

 undulations of gravelly soil, between the rivers Dande and Bengo, 

 in the north of Loando, and there only, I met with an area of more 

 than ten miles nearly exclusively covered with Hyphsene Palms, 

 which are probably identical with the Hyplicjene cucifera^ Pers. 

 After several years of observation in various places of this regi<:>n, 

 I conclude that these palm-forests covered in past ages a great 

 portion of the coast districts of Angola, where, however, at 

 present they only appear as dwarf bushes without stems, and 

 never blossom. 



This gradual falling off in the vegetation is observable in a 

 similar manner in several other dwarf shrubs belonging to the 

 Eubiaceae, Jasminese, Combretaceae, and especially in the M<£rua 

 Anffolensis, which is the one most frequently found, but which, 

 during my numerous wanderings, I only met with three or four 

 times developed as full-grown trees. The only kind of tree which 

 seems to brave this general dwindling away of vegetation is the 

 well-known Ada nsonia, which is seen in full splendour throughout 

 the district, either singly or in groups. The rains which fall m 

 this region are almost without exception passing thunder-storms, 

 discharging within a few hours immense quantities of water, 



i 



I 



1 



