THE ALLEGED DESICCATION OF EAST AFRICA 9 
bones, unfortunately no human ones, but he discovered what 
appears to be a rude copper bracelet ; as there is no copper 
found naturally in that country, this betokens trade and inter- 
course with the coast, and with people who had copper to sell. 
These mounds are so numerous, and in addition the large 
number of well-excavated wells, often over forty feet deep, and 
the traces of artificial dams, all go to prove that this area, which 
is now practically a desert, once carried a large and organised 
population. It is now only inhabited by a limited population 
of nomad Somalis, who move about from waterhole to water- 
hole and graze their stock as long as the water lasts. Similar 
mounds are found throughout the Nyika belt as far south as 
Taru, on the Uganda railway, but south of Wajheir they become 
much smaller, and rarely average more than a few feet in height. 
The Somalis say that the wells and the mounds were made 
by the Maanthinle people. This name in the Somali language 
means the tall people ; possibly they consider that people 
who carried out such works must have been of superhuman 
stature. 
In the evidences of the population of the hinterland we 
have, it appears to me, an explanation of the settlements on 
the coast ; and as to their disappearance, I put this down to 
the desiccation of the country. There is no record of these 
settlements having been systematically destroyed by invasion, 
and even if the coast dwellers were annihilated, it is difficult 
to see how the tribes of the interior could have been exposed 
to similar attack. 
We know that in about the year 1589 a terrible horde of 
savages, called the Mazimba, swept up the east coast and 
destroyed Kilwa and nearly destroyed Mombasa ; this force 
swept on up the coast, but were, according to the records, 
finally defeated at Malindi by Mattheus Mendes de Vasconcellos 
with a number of Portuguese soldiers, the Arabs, and some 
3000 Bantu tribesmen recruited in the vicinity and said to be 
Wa Segeju. This invasion may account for the comparative 
scarcity of rums south of Mombasa, but it cannot account for 
the abandonment of the settlements between Lamu and 
Kismayu. The trade of these coast settlements was, I believe, 
the supply of domestic animals for meat during the south-west 
