6 NATURE 
| May 5, 1870 
white. We are in the vine-hills again. . . . Cactus, 
oleander, orange, and pomegranate—all these appear.” 
He then passes through the Basque country, St. Sebastian, 
and Biarritz, which the author considers “ the very pearl 
of a summer resort.” The work is ended with a fable 
recorded by Ford : “When San Ferdinand captured 
Seville from the Moor and bore the conquest to heaven, 
the Virgin desired her champion to ask from the Supernal 
Power any favour for Spain. The King asked for a fine 
climate and sweet sun : they were conceded. For brave 
men and beautiful women: conceded. For oil, wine, and 
all the fruits and goods of this teeming earth. This 
request was granted. ‘Then will it please the beauteous 
Queen of Heaven to grant unto Spain a good Govern- 
ment ?? ‘Nay, nay, that can never be. The angels would 
then desert heaven for Spain !’” 
The book is plentifully interspersed with good illus- 
trations. : 
LETTERS TO THE EDITOR 
[ The Editor does not hold himself responsible for opinions expressed 
by his Correspondents. No notice is taken of anonymous 
communications. | 
The Sources of the Nile 
From Mr. Keith Johnston’s communication to NATURE of the 
14th April, it appears that he agrees with me in opinion—though 
quite independently of me and by a different process of reasoning 
--that the great river Kassabi, Kassavi, or Kasai, of South- 
Western Africa, instead of flowing to the north and north-west, 
as it has hitherto been shown to do in all maps, has its course 
north-eastward as far as about the meridian of 27°30’ east of 
Greenwich, where it is joined by the river system of the Cham- 
beze. Such being the case, the only material question between 
us is with respect to the lower course of the united stream of the 
two rivers, which Mr. Johnston carries round by a sharp curve 
to the north-west and west, so as to join the Zairé or Congo 
river, whereas I regard it as continuing northwards, and uniting 
with the Albert Nyanza, so as to form the upper course of the 
Nile. 
While thus disputing the claim of the Kassavi and Chambeze 
to be the head-streams of the Nile, Mr, Keith Johnston adyo- 
cates the rival claim of ‘‘the feeders of Lake Liemba.” Ido 
not, however, understand him to mean that these rivers, four in 
number, are to be considered the head-streams of the main body 
of the Nile, or to be anything but tributaries of that river, as is, 
in fact, shown in the map accompanying his paper. It being 
upon this point that the whole difference between us really hinges, 
I beg to be allowed to offer the following observations on the 
subject. 
Dr. Livingstone, the discoverer of Lake Liemba, describes it 
as lying at an elevation of 2,800 feet above the sea, on the 
northern slope of the Balungu upland, ina hollow with preci- 
pitous sides 2,000 feet down, and as going away in a river-like 
prolongation, two miles wide, N.N.W., to Tanganyika, of which 
he rightly considers it to be an arm ; the difficulty with respect 
to the elevation of the latter having been removed by Mr. Findlay, 
who makes it to be 2,800 feet, or the same as that of Lake 
Liemba. 
Captain Burton, the discoverer of Tanganyika, says respecting 
this great lake: ‘‘ The general formation suggests, as in the case 
of the Dead Sea, the idea of a volcano of depression—not like 
the Nyanza or Ukerewe formed by the drainage of mountains. 
Judging from the eye, the walls of this basin rise in an almost 
continuous curtain, rarely waving and infracted, to 2,000 or 3,000 
feet above the water-level ;” from which description it is evident 
that Liemba and Tanganyika are portions of one continuous 
fissure or ‘‘crack” in the table-land, of which table-land the 
elevation is 5,000 feet, or perhaps more, above the ocean.—Dr. 
Livingstone says 4,000 to 6,000 feet, sloping towards the north 
and west, but he had not seen any part of it under 3,000 feet of 
altitude. Further, Lake Tanganyika is described by Dr. Living- 
stone as passing northwards, bya river named Loanda, into Lake 
Chowambe, which lake he identifies with Sir Samuel Baker's 
Albert Nyanza. 
The last-named expanse of water was found by its discoverer 
to have an elevation of 2,720 feet above the ocean, with a pre- 
cipitous cliff of 1,200 to 1,500 feet on the east shore, whilst on 
the opposite side the faint blue mountains rose about 7,000 feet 
above the water-level. 
The Albert Nyanza is, however, in nowise a continuation of 
the system of which Liemba and Tanganyika form parts; for 
whilst the direction of these two lakes is from north to 
south, or nearly so, the general bearing of the Albert Nyanza 
is from about north-east to south-west ; so that, as is shown on 
Mr. Keith Johnston’s map, the former joins the latter at an 
angle of 45°; whilst the main body of water extends probably 
a hundred miles beyond the junction, or to about the 28th 
meridian, And the Albert Nyanza does not terminate here ; for, 
in the latitude of Karagwe, between 1° and 2° S., it was said by 
the natives to turn to the west, in which direction its extent was 
unknown even to Rumanika, the King of Karagwe. 
Quite independently, then, of the question of the junction of 
the joint stream of the Kassavi and Chambeze with the Nile, 
there is this preliminary question, which I would propound for 
Mr. Keith Johnston’s consideration and answer :—Where would 
he place, even if only conjecturally, the head of this unknown 
western extension of the Albert Nyanza?—or, in other words, 
where would he trace the western limits of the Upper Nile 
basin ? : 
As regards his objection to ‘‘ the northward wall-like continua- 
tion of the Mossamba mountains on the 20th meridian to beyond 
the equator,” shown in my sketch-map of the Upper Nile basin, 
in Part xv. (for March 1st last) of the ‘‘Illustrated Travels,” I 
must explain that the same is so marked merely conjecturally, 
and that I do not think of maintaining it against—I will not say 
proof—but any reasonable argument. In my former maps of 
1849, 1859, and 1864, I placed the conjectural western limits of 
the basin of the river somewhere about that meridian on or near 
the equator, thence continuing to about 10° N. lat., where the 
line was made to curve inwards towards the valley of the river. 
When I found the Kassavi, which I look on as the head-stream 
of the Nile, actually rising in the Mossamba mountains, on 
about the meridian thus indicated, I naturally extended my con- 
jectural limits of the basin of the river along the same meridian 
from the equator southwards. But I repeat that, beyond what 
we actually know, all the rest is purely conjectural. If there is 
reason to carry the limits of the basins of the rivers of the West 
Coast of Africa to the east of the 20th meridian, on the equator 
or even as far south as the fifth parallel of south latitude, I have 
nothing to object to it, except that care must be taken to leave 
sufficient space for the western flank of the basin of the River 
Nile. 
And this brings me to what I regard as an insurmountable ob- 
jection to Mr. Keith Johnston’s hypothesis. By causing, as he 
does, the Kassavi and Chambeze, after their union on the me- 
ridian of 27° 30’, to make a curve round to the north-west and 
west, so as to form the main-stream of the Congo River, he 
actually brings the course of this supposed river within 150 miles 
of the south-western extremity of the Albert Nyanza, as laid down 
onhis own map. But to enable him to do this, he must, in de- 
fiance of Sir Samuel Baker’s authority, deny the great westerly 
extension of this immense body of water ; and by closing it up in 
that direction he renders it merely a ‘‘ back-water” to Tanga- 
nyika, as Captain Speke imagined it to be to his Victoria Nyanza, 
instead of its being the main-stream ofthe Nile. Mr. Keith Johnston 
evidently has misgivings on this head, for he says : *‘ If, however, 
the Albert Nyanza prove to have a great south-westerly extension, 
this one difficulty would be removed,”-—namely, the sole diffi- 
culty in the way of the junction of the Kassavi and Chambeze 
with the Nile, for which I contend. 
The argument founded on the comparative levels of Lakes 
Moero and Tanganyika I fail to appreciate. If the upland, 
2,000ft. below which Liemba and Tanganyika lie, has a general 
elevation of 5,000ft., the waters of Lake Moero, instead of pass- 
ing as they do through the crack in the mountains of Rua, could 
not by any possibility unite with those of Tanganyika, except by 
means of a similar crack in the mountains forming the western 
side of the latter; unless, indeed, Moero were supposed to lie on 
the upper level, to which supposition Mr. Johnston’s argument 
is diametrically opposed. 
In support of my own argument that the united stream 
of the Kassavi and the Chambeze continues northwards to” 
join the Nile, instead of turning round to the north- 
west and west to join the Congo, I have really nothing to 
add. My opponent himself carries the joint stream for me to 
within 150 miles of the known south-western extremity of the 
Albert Nyanza: it is for him to show how the two are to be 
