Odo and other Trihitaries of Sohat. 139 " These villages are situated some distance from the river, as the expanse of swamp bordering the Gelo prevents villages being built nearer to its banks. At the height of the rainy season it appears probable that the whole country north of the Gelo and between that river and the Baro is one vast swamp, quite impracticable for transport animals.

" The width of the Gelo west of Lake Tata varies as a rule from (iO to 100 yards. At its junction with the Pibor its width is not more than 30 to 40 yards, whilst its current is very swift.

" As the Gelo is followed along its banks, no other stream appears to flow into it, although several swampy khors issuing from the river are crossed, flowing in a northerly direction. These are reported to join the Bela river and to flow into the Mokwai.

" Along the Gelo, beyond a small patch of cultivation on the river bank near the village of Patok, no other signs of civilization are evident, except in a few diminutive cleared spaces in the woods ; no food is obtainable from these natives. That they must subsist on grain to a large extent appears to stand to reason, so it is possible their fields, like their villages, are concealed in the midst of woods, with which the country abounds. Like the Nuers and Anuaks of the Baro these natives possess small dug-out canoes for crossing the Gelo, and employ them also in their fishing operations.

" To the west of Perbong two other villages, Ametha and Otwol, are reported to exist in Anuak territory hereabouts, and these two probably depend on wells for their water supply, as they must be quite 3 to 4 miles distant from the river, and no signs of tracks leading from or to the Gelo are visible. Goats and sheep are only seen in very small numbers.

" In normal years it is doubtful if this tract of country can be traversed much later than the middle of April with transport animals,* as once the rains set in the whole country is rapidly converted into bog, through which laden animals, especially donkeys, are quite unable to travel. During the dry seasons of the year large herds of elephants roam over these grassy plains and find both food and shade in the forest growth on the north bank of the Gelo. It is to these regions the Abyssinians descend in large numbers yearly from the western edge of the plateau, on ivory hunting expeditions, and traces of recent Abyssinian encampments were on several occasions met with. Before the rains break, however, these parties return to their homes, and we were warned at Gore by Fitorari Hili that we would find the country impracticable for animals once the rains set in, and this we certainly found to be the case." The River Mokwai or Bela appears to flow westwards to the Pibor from the Gurafarda range of hills (Abyssinia) traversing en route the marshy region between the Baro and Gelo ; from the latter river it probably receives a considerable overflow. Its mouth is supposed to be about 8 miles south of the Pibor-Sobat junction, but owing to swamp it has not been accurately determined. It is said to be an important river in flood time.

The Khor Filus enters the Sobat on the left bank about 10 miles from its junction with the White Nile, the following description of this khor, the adjoining country and its inhabitants, is taken from a report by Captain H. H. Wilson, Inspector, Upper Nile Province : — " Starting on April 15, 1902, the expedition marched from the village of Gokjak (some 20 miles from the mouth of the Sobat), and met the Khor Filus at Shol Ajik, some 8 miles inland, thus saving an unnecessary long march along the khor from its mouth, which is 10 miles from the Sobat mouth. The country here is flat and uninteresting, nothing but a vast grass plain, with hardly a tree to be seen. At Shol Ajik trees were met with, the banks of the khor being thinly wooded with a small growth of red 'talh,' with a sprinkling of the unwelcome 'kittr' thorn bush. The khor at this point is some 50 to 80 yards in width, mostly, however, filled up with weeds, the real water channel being clearly marked (though practically dry) by a deep narrow bed, some 15 feet lower than the adjacent banks. On to the village of M'Yolga the same country extends — grass, with a few trees ; in one place there were traces of elephants which had been there in the rainy season. Other game seen was the bastard tiang hartebeeste, which was really all the game seen in any quantity in the country. M'Yolga is a long scattered village on the right bank of the khor, and marks the limit of the Dinka tribe inland. From here onwards for many miles is uninhabited country, from Bia to Nerol being thickly wooded, in places densely, but only occasionally with any undergrowth that makes travelling difficult. Tiang hartebeeste were seen in considerable numbers along the whole route, and also quantities of waterfowl, pelicans, duck, geese, teal, and many and various kinds of crane and heron. In the vicinity of Fanyanglwel, 46 miles south of Sobat, the first signs of Nuer habitation were met with in the shape of their cattle " feriks,' which are the summer residences {i.e., dry season, from January to May), built close to water, and constructed lightly of grass only. No regular huts or tukls were met with until reaching Riul, 7 miles further south, where the banks became low and flat, and trees only at some distance on either side. The khor is here in the rains evidently broad and shallow, and, owing to mud, probably an impassable obstacle under such conditions.

"■ A noticeable feature in the Nuer tukls is the superiority of their construction compared with the tukls of the )inkas, Anuaks, and other tribes of the Sobat, their cattle tukls being in many cases marvels of constructive art, giver • Tliis country was traveled from Marcli to June, 1904, by Mr. McMillan and his expedition. Out of 150 i.niles and the .same her of donkeys only 16 animals survived. Anuak carriers were, however, readily obtainable, from 150 (o 380 being constantly ' oyed.

Width and current.

Cultivation.

Swampy nature of country.

River Mokwai.

Khor Filus.