On the Bahr El Ghazal, on the contrary, the sudd is chiefly composed of the swimming plants above-mentioned. Their breeding places are Lake Ambadi and the other shallow lakes to the south.

The Ghazal sudd is much lighter in texture than that of the Jebel, and is consequently much easier to remove. At the same time, even in the former river, the sudd is at times dangerous, especially if it forms down-stream of a vessel, and if the latter has to work upon it from its up-stream end. The accident to Gessi Pasha’s expedition in 1880 proves that even the Bahr El Ghazal sudd can be an impassable obstacle under such circumstances.

Before alluding to the work done in removing the sudd, it may be as well to say a few words as to the principal causes of this remarkable barrier.

The Bahr El Jebel traverses the marshes between Bor and Lake No for some 380 miles of its course. South of Shambe the river has never been known to be blocked. On either side of the channel, in these immense swamps, extend large shallow lagoons, some of them covering a square mile or more of area. These lagoons are surrounded on every side by a luxuriant growth of aquatic plants, consisting chiefly of the papyrus and the grass known to the Arabs as the “um sûf” (the mother of wool). Both these plants grow in water, but not in any great depth. The papyrus attains a height of from 15 to 20 feet, with fibrous roots which strike deep into the ground. The “um sûf” rarely exceeds 5 feet in height, and its roots do not extend so deeply as do those of the papyrus. They are, however, very tough and difficult to break or cut through. These roots are bedded in the soil below the water, but the strong gales which blow in these regions loosen their hold to a large extent. If such a storm be accompanied by any rise of the water surface, large masses of these plants are set free from their original position, and begin to float on the surface of the lagoons. Their roots form such a tangled mass that large quantities of the earth in which they are embedded remain clinging to them. These act as ballast, and when the island of papyrus or reeds is detached and, under the influence of the wind, is set drifting about the lagoon, the weight of the earth retains the plants in their vertical position. Their roots, the moment they reach a shallow, act as anchors, and speedily strike down again into the muddy bottom of the lake. Large masses constantly change their position in this way. If the storm cease, they remain where they are. Unfortunately, at the commencement and end of the rainy season, stormy weather is the rule rather than the exception. At such seasons large areas of the marsh vegetation are in motion, driven hither and thither by the wind.

The Bahr El Jebel, where it passes through the swamps, has no banks whatever, and is, as a rule, separated from the lagoons merely by a narrow belt of papyrus. In many places it is in connection with them, and the water of the lakes flows in and out of the river, according to the level of the latter. As the stormy season in these latitudes heralds the approach of the annual rains, the rise of the river follows very speedily. The channel of the Bahr El Jebel being only of sufficient section to carry the low water supply, with the first rise in the levels the river spreads over the marshes, flooding them in all directions, and increasing the depth of water in the lagoons. It thus causes the areas of reed, already detached by the wind, to float still more easily. The continuous gales which prevail set hundreds of acres of these floating masses moving in one direction. Eventually they reach a point on the river where they are forced into the channel. Once there the current speedily carries them down-stream. Ere long their course is arrested by a projection on the edge of the channel or by a sharp bend. It may happen that an area of reed, several acres in extent, bursts into the river in a large sheet, and in such a case it must be necessarily arrested at the first point where the section is contracted. The result is that the channel is quickly blocked, though, perhaps, not at first to any great depth. Masses of weed, however, follow one another in succession, brought down by the stream. The section of the channel being reduced by the first obstruction, the velocity of the water rapidly increases, and these masses, following the easiest course, pass under the obstacle thus created. Each fresh mass arriving is sucked underneath those originally arrested, until at last the whole becomes wedged into one solid block, composed partly of earth and partly of stalks and roots of papyrus and reed, broken up by the extreme compression into an inextricable tangle. So great is the pressure applied by the water, that the surface of the block is often forced several yards above the water-level and is seamed by alternate ridges and furrows. The thickness varies greatly, according to the conditions and sections of the channel. In some cases it is not more than 4 to 6 feet, but it not infrequently obtains a thickness of 16 feet, below water, and occasionally as much as 22 feet have been observed.[229] Underneath this bar the river manages to force an outlet, but with a velocity increased proportionately to the smallness of the aperture. At the same time, the up-stream level rises, flooding the marshes in every direction, the water making use of any side channel that it can find. In time, doubtless, if left to itself, it would desert its original course, and the stream would take an entirely new direction, the original channel becoming permanently blocked. It generally, however, happens from natural causes, such as strong winds or increased heading-up of the water, that these blocks burst, and the obstacle is carried away. On such occasions a great wave passes down the channel, carrying everything before it, and sweeping away any similar blocks which may have been formed down-stream. Only in this way can the clearance of the sudd in certain years which has undoubtedly occurred be explained. Many of these blocks extend for a considerable length, some being as much as several miles long. It is easy to understand that such closures of the river channel cause not only a complete bar to navigation, but also a very serious obstacle to the free passage of the water. More than this, each block thus formed assists in the formation of others, by raising the water level up-stream, and thus assisting the flotation of further areas of papyrus and reed, much of which eventually finds its way into the river.

The movement of these great masses of weed, and the way in which they burst into the river, bears a striking resemblance to the descriptions given of an icefield when in motion. Their steady and resistless movement, the manner in which the fields break up, and the way in which they pile upon one another when an obstruction is encountered to their course, recall irresistibly what travellers relate of the action of the ice-floes when the pack is breaking up.

The sudd in this river, and in a less degree that in the Bahr El Ghazal, constitutes a very real danger to navigation. Should a steamer happen to be surrounded by it when in motion, the compression would certainly strain her framework to the risk of crushing it. Even should she escape this she will most probably be imprisoned for an indefinite time, owing to the river blocking on either side of her. During the stormy season, it should if possible always be arranged that two steamers should work in conjunction, one remaining down-stream, so as to be able to go to the assistance of the other if necessary.

As no fuel is to be met with throughout the whole length of the sudd region, a solitary steamer, if detained for some time by a block, may find herself eventually prevented from proceeding, or from making any attempt to free herself, owing to the absence of any means of generating steam in her boilers.

The Bahr El Ghazal sudd is, as has been said, of a different kind to that of the Bahr El Jebel. It is, as a rule, much lighter in consistency and easier to remove. This is doubtless due to the fact that the velocity of this river is very low, and, consequently, the pressure exerted by the water upon a block is small compared with that caused by similar conditions in the Bahr El Jebel. Again, on the Bahr El Ghazal, the papyrus and reeds do not extend on either side to any distance, nor are continuous and large lagoons close at hand. Between miles 42 and 52, up-stream of the junction with the Khor Deleib, there must, it is true, be always a risk of a block. In this reach the river passes through papyrus swamps, which are miniature editions of the Jebel marshes. Even here, however, the obstruction can hardly attain to the same solidity as in the case of the Jebel sudd, as the sluggish current of the Ghazal would fail to wedge the mass between the banks as tightly as would the rapid stream of the other river. Lake Ambadi appears to be a great nursery for the smaller varieties of the sudd plants. In the beginning of the rainy season these are carried down-stream and, as the channel wanders and twists, they form small obstructions. None of these, however, appear to be lasting, and it is rare that the Ghazal river is blocked for any length of time. It may be closed for a few months and then re-open itself. Thus in March and April, 1900, it was clear throughout its entire length, while in September of the same year it was blocked in more than one place. On this river, as on the Jebel, the sudd frequently sinks to the bottom, and decomposes, gradually raising the bed of the river. In this state it is very hard to remove.