General Headquarters, East Africa,
27th October, 1916.

Sir,
In my last despatch I described the brief but important operations which ended on 21st March in the occupation of the Kilimanjaro-Aruscha area. The strategy involved in those operations was determined, for me by the military situation I found existing on my arrival in British East Africa in February. The opposing armies had massed on the Taveta and Longido fronts; the rainy season was expected in a few weeks, and there was no time or necessity for radical alteration in the plans on which my predecessor had been working. When the operations came to an end it was necessary without delay to dispose my forces most advatageously with a view to their health and comfort during the approaching rainy season, and it became necessary to study the important question of the strategy to be followed in the future operations.

Reorganisation of Forces.
(2) Preliminary to both matters, however, was the question of reorganisation of the East African forces, which I deemed necessary not only for the vigorous prosecution of the coming campaign, but also to secure the smooth and harmonious working of a most heterogeneous army, drawn from almost all continents, and speaking a babel of languages. I decided to abolish the two Divisions formed by my predecessor and to organise my forces into three Divisions, two of which were to consist of the contingents from the Union of South Africa, and the third was to include the Indian and other British forces. The Union Divisions were again so organised that each should eventually contain a mounted and an infantry brigade, so as to secure the necessary mobility to enable us to cope more expeditiously with the enemy askari army of fleet-footed Africans. In these alterations, as well as in all other important matters which I have had from time to tdme to submit for the sanction of the War Office, I have found the uniform and prompt support of the latter, for which I cannot be sufficiently grateful, and to which, the success achieved in this campaign was in no small measure due. At the end of March, then, the East African Force—apart from lines of communication troops, under Brigadier-General W. F. S. Edwards, D.S.O., as I.G.C.—was organised as follows: —

The First Division, under Major-General A. R. Hoskins, C.M.G., D.S.O., comprised the First East African Brigade, under Brigadier-General S. H. Sheppard, D.S.O., and the Second East African Brigade, under Brigadier-General J. A. Hannyngton, C.M.G., D.S.O.

The Second Division, under Major-General J. L. van Deventer, comprised the First South African Mounted Brigade, under Brigadier-General Manie Botha, and the Third South African Infantry Brigade, under Brigadier-General C. A. L. Berrange, C.M.G.

The Third Division, under Major-General Coen Brits, comprised the Second South African Mounted Brigade, under Brigadier-General B. Enslin, and the Second South African Infantry Brigade, under Brigadier-General P. S. Beves.

The Second South African Mounted Brigade arrived in May and was ready to take the field in the latter half of June.

Having completed the above reorganisation I disposed the infantry units as far as possible at suitable points on high and dry ground at Moschi, Himo and Mbuyuni, with only advance guards along the deadly malarial line of the Ruwu, facing the enemy forces in the Pare Mountains. The First Mounted Brigade was pushed on to the Aruscha area, which was reported to be most suitable for horses, and at the end of March the whole brigade had arrived there.

General Plan of Invasion.
(3) The most important problem for consideration was the strategy to be followed in the coming campaign. As a result of the preceding operations we had just barely entered the enemy territory, which stretched out before us in enormous extent, with no known vital point anywhere, containing no important cities or centres, with practically no roads, the only dominant economical features of the whole Taeing the two railway systems. Faulty strategy at the beginning, a wrong line of invasion once entered upon, might lead to months of futile marching and wasted effort. All our information credited the enemy with the two-fold intention of conducting an obstinate and prolonged campaign in the Pare and Usambara Mountains, and thereafter retiring to fight out the last phases of the campaign in the Tabora area, from which much of his supplies and most of his recruits were drawn. Careful consideration was given to the various alternative lines of invasion that presented themselves.

(4) There was, in the first place, the possibility of advancing inland from the coast along the existing railway lines, which had been adopted with such signal success in the German South-West Africa campaign. An advance from Tanga was, however, ruled out because I considered the place of no importance after the Tanga railway had been reached further north. Much, on the other hand, was to be said for an advance inland from Dar-es-Salaam, the capture of which would have great political and military importance, and would much facilitate the transport and supply arrangements for the campaign into the interior. It was, however, also ruled out, partly because the prevalence of the S.E. monsoon at that period makes a landing of a large force on that coast an operation of great difficulty and even
danger, partly because a prolonged campaign on the coast immediately after the rainy season would mean the disappearance of a very large percentage of my army from malaria and other tropical ailments.

(5) In the second place consideration was given to the question of an advance on Tabora by Victoria Nyanza, which we controlled, and Muanza, which would have to be wrested from the enemy. This plan had the advantage of presenting a comparatively short line of advance, and of promising to strike at the main recruiting ground of the enemy forces, as the German askaris would be loath to remain in the field after their homes and families had fallen into our hands. Its adoption, however, would involve the transfer to a distant theatre of a large part of our forces while the enemy army would remain concentrated and ready to strike at our railway communications with the coast. But my main objection to adopting it was the consideration that to occupy so huge a territory as German East Africa within reasonable time a simultaneous advance from different points along different routes was essential. Now in the Eastern Lake and Uganda area we already had a force of about 2,000 rifles; in addition the Belgians had a very large force in the West in the neighbourhood of Lake Kivu with which they were prepared to invade the Ruanda and Urundi districts if we could, assist them with the necessary transport and supply arrangements via Victoria Nyanza. For the occupation of the western parts of German East Africa it was therefore only necessary to make these arrangements, and thereby to set the Belgian and British forces simultaneously in sympathetic motion in the Ruanda and Bukoba districts, respectively. This was done, and with the best results, as will be described later.

(6) There remained, then, the third and last alternative of either striking at the main enemy forces in the Pare and Usambara mountains along the Tanga railway line, or of launching an attack against the interior and the Central Railway from Aruscha. A movement against the enemy concentration along the Tanga railway had, however, several grave disadvantages. It was the step desired and expected by the enemy, as the massing of almost his entire fighting force in that area showed. It would involve a prolonged and costly campaign over terrain which nature and art had prepared admirably for defensive purposes. And at the end of such a campaign the entire enemy territory would still remain unoccupied, as the operations would have been conducted lengthwise all along the border. On the other hand
an advance from Aruscha into the interior, if it was not to be a mere temporary raid but a secure and permanent occupation of the country, had to be in such force that it could meet any counter-attack by the enemy, who would in such counter-attack have the advantage of his two railway systems and so be practically moving on interior lines. Such an advance in force, therefore, ran the risk of weakening our forces in front of the enemy in the Pare and Usambara mountains and of giving him an opening to attack our vulnerable communications both with the interior and the coast.

(7) In spite of these difficulties powerful arguments weighed with, me in finally deciding in favour of an advance into the interior. I was informed that the violence of the coming rainy season would be mostly confined to the Kilimanjaro-Aruscha area; that further West and South the rainy season was milder and would not markedly interfere with military operations; and therefore an advance into the interior would prevent our operations being brought to a complete standstill during the rainy months of April and May. In addition to this the enemy had made the mistake of retiring South along the Tanga railway with practically his entire fighting force, and the door to the interior stood wide open and unguarded operated between Kilimanjaro and Meru mountains against General Stewart's advance from Longido, and were expected by me to fall back on Aruscha and obstruct our advance in that direction, joined the enemy's main force at Kahe. A small detachment at Aruscha fell back before the advance of our mounted scouts, and when the mounted brigade arrived at Aruscha at the end of March there was for the moment nothing to prevent an immediate movement into the heart of the enemy country. I
decided to push the whole of the 2nd Division into the interior under van Deventer, and for the present to keep the other two divisions with
.me in rain quarters facing the enemy concentration South of the Ruwoa. In this way it would be possible to occupy a valuable portion of the enemy country within the next two months; and if, as I expected, this move would and must have the effect of compelling the enemy to withdraw large forces from the Pares and Usambaras to stem the tide of invasion into the interior I could, if necessary, strengthen van Deventer still further and yet have sufficient troops left to make a comparatively easy conquest of these mountains against the enemy's weakened defence. These anticipations were fully realised, as will be seen from the sequel.

Van Deventer's March to Kondoa Irangi.
(8) By April 1st the Headquarters of the 2nd Division, together with the 1st South African Mounted Brigade and two batteries of artillery had reached Aruscha, while two battalions of the 3rd South African Infantry Brigade were on the way. On the same day General van Deventer reported that his scouts had engaged the enemy six miles north of Lolkissale, an isolated rocky hill in the Masai Steppe some 35 miles southwest of Aruscha. Further reports showed that this force consisted of a detachment of the enemy which had taken up a position covering the water springs on the Mil, and that no other water was to be found in the vicinity. I therefore issued instructions that the movement southward should be initiated by the occupation of Lolkissale. This operation was carried out with great skill by the 1st Mounted Brigade. On the morning of the 3rd April three regiments of South African Horse moved out from Aruscha, and during the night of 3rd/4th April surrounded Lolkissale. The enemy held the mountain with considerable determination, and fighting continued all day on the 4th and 5th, but at daybreak on the 6th the whole force, consisting of the 28tfh Field Company and Kaempfe's Detachment numbering 17 whites and 404 askaris with porters and two machine guns, surrendered. Our horses had been without water since noon of the 3rd. A large quantity of stores, ammunition, pack animals, etc., fell into our hands, while from information obtained from prisoners and captured documents it was ascertained that the enemy contemplated reinforcing Ufiome and Kondoa Irangi, and that the garrisons at these places had received instructions to hold out as long as possible. As it was evident that the bulk of these reinforcements must be sent from the troops on the Usambara Railway, and that several weeks must elapse before they could arrive, I decided to press forward the movement southwards of the 2nd Division as rapidly as possible, and ordered General van Deventer to send his mounted troops to occupy Ufiome, Umbulu and Kondoa Irangi before the enemy could reinforce them. The remainder of the 2nd Division to follow in support of the mounted troops.

(9) The 1st Mounted Brigade continued its advance to Ufiome on the 7th, encountering the enemy's patrols at various points of the route and dispersing them with loss in killed and prisoners. On the 10th the enemy were located holding a kopje in the vicinity of Ufiome, and on the llth the Brigade advanced from the Tarangire Biver. This movement resulted in the occupation of Ufiome on the 13th, the garrison of about 20 whites and 200 askaris retiring into the mountains, leaving 30 prisoners, some wounded and a large quantity of supplies in our hands. The enemy was pursued for 20 miles south to Kieesse and Ssalanga, retiring in disorder.

(10) As the horses of the 1st Mounted Brigade were greatly exhausted by the continuous marching and fighting a halt was made at Ssalanga until the 17th. The 4th South African Horse had in the meantime been sent by me to join the 2nd Division, and on its arrival was directed on Umbugwe with instructions to clear Umbulu of the enemy. The 10th South African Infantry and 28th Mountain Battery were also detailed by General van Deventer to follow in support of the 4th South African Horse. Umbulu was finally occupied on 11th May, about one company of the enemy being driven out with loss. The advance southward continued on the 17th, and contact was made with the enemy four miles north of Kondoa Irangi on the same day. Fighting continued till noon of the 19th, when our troops occupied Kondoa Irangi with no casualties, having inflicted a loss on the enemy of 20 killed, and 4 whites and 30 askaris captured. The enemy succeeded in destroying the wireless station and a portion of his supplies, but left behind about 80 rifles with much ammunition, and 800 head of cattle.

(11) General van Deventer reported after this action that his horses were so exhausted that he would not be able to move until remounts arrived. He had lost hundreds of animals from horse sickness during his advance of some 200 miles from Moschi in the last four weeks, and his troops were worn out with ceaseless marching and fighting. I therefore decided that the 2nd Division should concentrate at Kondoa Irangi with detachments at Ufiome and Umbulu, and send patrols towards the Central Railway, Ssingida, Mkalama and Handeni.
During the remainder of the month and the first few days of May this concentration was gradually effected. The expedition, conducted by van Deventer with his usual dash and resourcefulness, had secured important results at a trifling cost. Within a month of the battle of Kahe we had taken possession of the high, healthy and fertile plateau which connects Aruscha with the Central Railway, and had occupied the dominant strategic points for any further advance, whether that was to be in the direction of the Central Railway, or westward to Tabora, or even eastward towards Handeni and the Eastern Usarnbara.

(12) Meanwhile, by the middle of April, the rainy season had set in with the greatest violence in the whole area from Taveta to Kondoa Irangi. The numerous rivers came down in flood and swept away almost all our laboriously built bridges, the roads became impassable mud tracks, and all transport became a physical impossibility. The rains fell steadily day after day, sometimes as much as four inches in one day, and the low-lying parts of the country assumed the appearance of lakes. Fortunately, the railway had by this time reached Taveta, where sufficient supplies could be dumped for our resting troops. The extension of the line was energetically continued to join the Kahe-Moschi railway, although for long distances the track was practically under water and the attention of thousands of labourers was constantly required to prevent its disappearance in the mud. Van Deventer's Division in the interior was cut off, and managed to live for weeks on such supplies as could be collected locally, or could be carried by porters from Lolkissale for a distance of 120 miles. The strain and privation were, however, bound to be reflected in the general state of health of the troops.

(13) Meanwhile, also, the enemy had realised the tremendous threat which this expedition constituted against his whole scheme of defence, and, thanks to the onset of the rainy season bringing General van Deventer's movement to a standstill, he was able to take measures to avert the danger to his rear by hurriedly transferring a great part of his force from the TJsambara to the Central Railway, moving by rail to Mcmbo, thence by road to Morogoro or Kilossa, and again by rail to Dodoma. This movement placed him in a position to concentrate some 4,000 men against the 2nd Division, which was at the time so weakened by sickness and unavoidable detachments that it could barely dispose of 3,000 rifles in its isolated position at Kondoa Irangi. The enemy, perceiving this, felt encouraged to assume the offensive, and advanced from the Central Railway in the early days of May, arriving on the 7th within six miles of Kondoa Irangi. General van Deventer gradually withdrew his advanced posts in face of this movement, keeping touch with the enemy, and finally disposed his force in defensive positions on a perimeter of about five miles frontage round Kondoa.

(14) On the 9th the enemy drove in our outlying picquets south-east of the village, and at 7.30 p.m. began an attack which lasted for nearly eight hours. This attack was pressed with determination, the enemy making four separate onslaughts, the brunt of which fell on the 11th South African Infantry, supported by the 12th South African Infantry. In some places the enemy repeatedly charged right up to our positions. Firing finally ceased at 3.15a.m. on the 10th, when the enemy withdrew, leaving three whites and fifty-eight askaris dead on the ground, and five wounded as prisoners. There were numerous signs on the ground of further casualties. Our own losses were two officers and four other ranks killed, one officer and seventeen other ranks wounded. From, information obtained as a result of the fighting, it was found that the enemy had about twenty five companies engaged, under the personal command of Colonel von Lettow, the German Commander-in-Chief. Hie force was organised as three battalions and one smaller detachment. One battalion commander, Von Kornatzky, was killed, and another, Von Bock, wounded.

(15) With this defeat, the enemy's last hope of successful resistance to any large portion of our forces was extinguished. He continued in position round Kondoa during the remainder of May and the greater part of June, keeping for the most part to the thick bush, and engaging, in desultory fighting and occasional long range bombardment. General van Deventer was unable to assume the offensive on any large scale on account of his weakness in horseflesh, the heavy sick rate amongst his men, and the great difficulties of supply over a line of communication of two hundred miles of quagmire; and had therefore to content himself with minor operations and enterprises while reorganising his forces and calling in his detachments from elsewhere. The 10th South African Infantry Regiment and 28th Mountain Battery, arrived from TJmbulu on 22nd May. I had already decided to strengthen the 2nd Division with two more battalions, the 7th and 8th South African Infantry Regiments, and additional artillery and machine guns, all from the 3rd Division, and these reinforcements eventually reached the Division on 23rd May and following days.

Occupation of the Pare, Usambara and Handeni Areas.
(16) Such was the position when, towards the end of the second week in May, the rains abated, the ground once more began to harden, and it became evident that a general movement would soon again be possible. The direction of that movement was settled for me by the necessity of clearing the enemy from the Pare and Usambara mountains before the further invasion of German East Africa could safely proceed. The general conception was to move Eastward along these mountains and at a point opposite Handeni to swing South and march towards the Central Railway in a movement parallel to that of van Deventer. The concentration of the enemy forces in front of Kondoa now made the occupation of the Pares and Usambaras comparatively easy, but the advance had to be rapidly executed to forestall any return movement of the enemy from Kondoa to the Handeni or TJsambara area. Moving through the Masai Steppe along the old caravan route, from Kondoa to Handeni, the enemy could reach the latter place in twelve days, and in two or three days more could be on the Tanga Railway at Korogwe. It was therefore advisable for my advance to reach the Western Usambara in a fortnight; further, if it could reach Handeni before the arrival of strong enemy reinforcements I would have a second force almost the same distance from the Central Railway as that at Kondoa, and it would be impossible for the enemy to make effective resistance to the simultaneous advance of both columns situated 170-miles apart. The nature of the country was, however, such as almost to preclude all rapidity of movement. The Pares and Usambaras are huge blocks of mountains with fertile valleys; the Southern slopes are precipitous, and immediately below
xuns the Tanga railway, while further South dense bush extends for 15 to 20 miles to the Pangani, an impassable river flowing almost parallel to the railway and the mountains. The enemy held the mountains and the railway and had outposts along the Pangani river.

Our advance was expected to follow the railway, which had been fortified at all convenient points for a hundred miles; and the enemy had therefore every reason to expect that the force opposing us, consisting of from 1,200 to 5,000 troops with field and naval guns, would render our progress sufficiently slow to enable him to send any necessary reinforcements. I therefore decided on the following dispositions for my advance. The main column with most of the artillery and transport was to proceed down the inner or left bank of the Pangani, somewhat in advance of another smaller column following the railway line, while a third small column was to start from Mbuyuni and enter the North Pares from the North side through the Ngulu Gap, joining the centre column at Same Pass between the middle and South Pares. In this way, with my flanks well forward in the mountains and along the Pangani, any real resistance of the enemy in his well prepared positions in the centre along the railway would become hopeless.

The advance commenced on 18th May by the movement of Lieut.-Colonel T. O. Fitzgerald's battalion of the 3rd King's African Rifles from Mbuyuni to the Ngulu Gap, and on the 22nd May Brig.- General Hannyngton's brigade moved from Pvuwu along the railway, while Generals Sheppard's and Beves' brigades moved down the Pangani river accompanied by Major-General Hoskins and myself.

(17) The enemy's first position was reported to be at Lembeni, at which place the railway takes a sharp bend in towards the mountains and the ground is most suitablefor defensive action. I trusted, however, to turn this position neither directly by Fitzgerald's column forcing
its way through the Ngulu Gap, or indirectly by the continued advance of the Pangani column past the enemy's position. The turning movements proved successful, and the enemy evacuated the Lembeni position on 24th May; on the following, day Hannyngton occupied Same station without opposition, and on the 26th May Fitzgerald's column joined Hannyngton's and thereafter formed part of it. Hannyngton was ordered to proceed on the 28th over Same Pass along the road which passes through the South Pare mountain, and thence through the Gonja Gap between this mountain and the Usambara on ito Mkomazi river. This move would prevent the enemy from making a stand on the railway along the South Pare mountain, and would at the same time clear the enemy out of the Gonja Gap. It was completely successful; on the 29th Hannyngton reached Uonia and two days after the Mkomazi road bridge.

(18) Meanwhile the advance of the main column continued steadily along the Pangani, the advanced guards and mounted troops continuing to keep touch with the enemy's rearguards, and I soon discovered that it was his intention to make his next stand near Mikotscheni, at which place the Pangani river rejoins the railway close to the mountains. On the 29th May the advanced troops came up against this position and drew fire from a naval 4.1 inch gun and two field guns. On the 30th May the 2nd Rhodesian Regiment attacked the position in front while the rest of General Sheppard's brigade made an arduous but successful turning movement by our left. The enemy retired in the night along the railway, leaving part of a new bridge in process of construction behind him. Buiko station was occupied by us the following day.

(19) Leaving a rear guard of two companies in front of Hannyngton at Mkomazi, the enemy's main body retired along the railway to Mombo station, whence a trolley line proceeds to Handeni. They then followed this trolley line and entrenched themselves at Mkalamo where this line crosses the Pangani river. This retirement made it clear that the enemy was not going to make a stand in the Usambara, but intended to retire to Handeni and on to the Central Railway. I decided, therefore, to cross to the right bank of the Pangani with the main column and to leave the further clearing of the TJsambara district to Hannyngton. The rapidity of our advance had exceeded my best expectations. We had reached the Usambara in ten days, covering a distance of about 130 miles over trackless country along the Pangani river and through the mountains.

(20) As at this point a short pause in the operations was necessary to enable the German bridge over the Pangani to be completed, and
to give the railway time to catch up with the advance, I proceeded on June 2nd via Moschi to Kondoa Irangi, to visit the 2nd Division and to arrange personally the plans for future co-operation between my two widely separated forces. On my return on June 7th I found that the German bridge over the Pangani had been completed and another smaller one made close to Buiko railway station, roads had been cut through the bush, and another 30 miles South had been covered by the main column along the right bank of the Pangani.

(21) I had instructed General Hannyngton with his brigade to proceed down the railway line with Mombo as his objective. He advanced to Mazinde station on 8th June and occupied Mombo Non 9th June, meeting with only slight opposition and capturing a machine-gun from the enemy. The enemy retired South along the railway. On the same date the main force of the enemy was encountered by our main column entrenched at Mkalamo, and the 1st East African Brigade had a sharp action, lasting till nightfall. The enemy retired in the night, leaving numerous dead on the ground. At Mkalamo the trolley line from Mombo to Handeni was reached on the 10th, and thereafter the advance to Handeni continued for a considerable distance along its route. The trolley line leaves the Pangani at Luchomo and from that point proceeds in a Southerly direction to Nderema, 2 miles West of Haudeni. Between Luchomo and Nderema is a dry belt of 32 miles, the only water being found by digging in a dry river bed at Mbagui, 22 miles South of Luchomo. To cross this distance General Sheppard was sent forward with two battalions to press the enemy back until Mbagui was reached on the 13th. From there he worked forward to within five miles of Handeni, where the enemy was on 15th June found to hold a strongly entrenched position. It was therefore decided to send Beves' brigade from Mbagui by a more Westerly route through Gitu to Ssangeni on the Mssangassi river, 10 miles West of Handeni, where good water was found on the 17th June, and on the following day the brigade was launched against the enemy's Southward line of retreat from Handeni at Pongwe and another point 4 miles North of Pongwe. At both places the enemy's retreating forces were beaten with heavy loss and driven into the bush, a pom-pom gun being subsequently found abandoned in the bush by the enemy.

On the following day Handeni and Nderema were occupied by Sheppard. On the same day Colonel J. J. Byron's battalion (5th South African Infantry) was sent in pursuit of the enemy to occupy Kangata, 8 miles South of Pongwe. They found the enemy in a concealed entrenched position in dense bush, and in the fight which ensued lost heavily, but held on staunchly until night, when the enemy retreated. At Kangata the main column for the first time since leaving Kahe came into a made road the main road between Handeni and Morogoro having marched for about 200 miles along routes prepared by themselves, mostly by cutting through the bush.

(22) During these operations General Hannyngton had occupied Wilhelmstal unopposed on the 12th June, and advanced along the Tanga railway as far as Korogwe on the 15th, where the wagon bridge had fortunately been saved by his special exertions. From this point he was instructed to move along the Korogwe-Handeni road and to rejoin with all speed the 1st Division, which was now nearing Handeni. He reached Handeni on June 20th, the day after its occupation by Sheppard.

(23) The advance of the main column in pursuit of the enemy continued, and he was next reported as holding a strong position on the Lukigura river. I therefore divided my force in the hope of getting round his position with a flying column and compelling him to stand to fight. General Hoskins with two South African Infantry battalions, a composite battalion of Kashmir Imperial Service Infantry, 25th Royal Fusiliers, and a small body of mounted Scouts, marched on the night of 23rd June to a point on the Lukigura river, north of the bridge held by the enemy. This force crossed the river the next morning, and then got astride the road behind the enemy's position. The remainder of the 1st Division under command of General Sheppard advanced direct on the enemy's position. At mid-day on 24th June both columns engaged the enemy on three sides, and after some resistance defeated him, with a loss of 7 whites killed and wounded, 14 white prisoners, 30 askaris killed and many wounded and captured, together with the capture of two machine guns and parts of a third, one pompom and much ammunition. The Fusiliers and Kashmiris specially distinguished themselves in this action, in which only the dense bush enabled the enemy force to escape from complete capture.

(24) We had now reached the eastern slopes of the Nguru block of mountains, and immediately in our front was the high Kanga mountain. There was every indication that the enemy was massing in great force in both mountains in front of us, as well as on our right flank, and that any further movement, would have to slow down. Our transport had reached the utmost radius of its capacity, and the troops had been on half rations for sometime. They also required rest and reorganisation. Several units were reduced to 30 per cent, of their original effectives, owing to the ravages of malaria, and the difficulties of evacuating the sick were as great as those of forwarding supplies and reinforcements. Since 22nd May the troops had marched considerably over 200 miles in difficult country, often having to cut their way through almost impenetrable bush, and constantly engaging the enemy in his prepared rearguard, positions. The march was rendered more arduous by most serious transport and supply difficulties, and, for the last 80 miles since leaving the Pangani, frequent shortage of water for both men and animals. Besides, I deemed it necessary, in view of the ever growing supply difficulties, to repair and restore the Mombo-Nderema trolley line before moving further.

Further, it was necessary for the execution of my plans that the 2nd Division should be more advanced before the combined movement against the enemy's main, forces on the Central Railway should begin. I therefore formed a large standing camp on the Msiha river, some 8 miles beyond the Lukigura, in which to rest and refit the troopsrior to the next phase of operations.

Occupation of Coastal Area to Bagamoyo.
(25) The pause on the Msiha river enabled me also to deal with another matter which was rapidly becoming urgent. I had deliberately left the East Usambara area alone while pushing the enemy forces in front of me back as fast and as far as possible. The situation on my left flank towards the sea would either clear itself up by the retirement of the small enemy forces in that area, or, if necessary, they could be dealt with at a more convenient time. The railway line beyond Korogwe and the lower reaches of the Pangani river were, therefore, for the present left unoccupied. Steps were, however, taken to seize Tanga.

On 16th June the 5th Indian Infantry, moving south towards the border, occupied Mwakijembe, which the enemy had held strongly for a long time as a base from which to raid and bomb the Mombasa railway. The enemy force of about one company retreated towards the coast north of Tanga. Arrangements were then made by the Inspector-General of Communications for the landing of a force under Colonel C. U. Price, C.M.G., at Kwale Bay, 8 miles north of Tanga, and a simultaneous attack on that port by land and sea. This force, after slight opposition, arrived before Tanga on the 7th July simultaneously with the Navy, and occupied it practically without opposition. The enemy, consisting of two companies, was expected to retire towards Pangani, but did not do so, and continued to hang about in the vicinity, and on several occasions even indulged in some sniping into the town. At the same time the small force of about two companies
which had retired before Hannyngton from Korogwe along the Pangani, returned and showed signs of aggressiveness. Small raiding parties kept interfering with our telegraph line, and convoys between Korogwe and Handeni, and finally, early on the morning of the 13th July, a determined attack was made on the road bridge at Korogwe, which was, however, successfully beaten back.

(26) The time had come to secure my rear and left from this guerilla warfare. Accordingly I ordered the Inspector-General of Communications, General Edwards, to make the following dispositions:—To send part of the 5th Indian Infantry from Tanga, along the railway to Muhesa; to send the 57th Rifles from Korogwe along the railway also to Muhesa, with a small detachment on their left in the direction of Amani; from Muhesa the 57th Rifles to prceed to the coast at Pangani, which was to be seized in co-operation with the Navy. In the meantime another detachment under Lt.-Col. C. W. Wilkinson, consisting of Railway Sappers and Miners, Jhind Imperial Service Infantry, and other details, was to proceed from Korogwe down the Pangani River to deal with the enemy force which had attacked the bridge, and which was reported to be at Segera Hill some distance down the right bank of the Pangani. All these movements were duly and successfully executed. At Amani about 25 enemy whites surrendered without opposition. Col. Wilkinson surprised and defeated the enemy at Segera Hill at dawn on the 15th July, and captured from them a Hotchkiss gun in good order, with
ammunition, and thereafter pursued the enemy south towards Hale and Kwa Mugwe (Hoffman's plantation). The 57th, after reaching Muhesa, proceeded to Pangani, which had been previously occupied by the Navy on the 23rd July. In the meantime, as I thought an effort should be made to capture these enemy parties, I had directed General Hannyngton's brigade to return from Lukigura to Handeni, and from there to march along the old caravan route towards Pangani, so as to intercept the retreating enemy and to clear the country of all raiding parties. He reached Ngambo about midway between Handeni and Pangani on the 21st July, but found the enemy had already
slipped through, part proceeding to the coast, at Mkwadja, and the greater part retiring south along a track which proceeds by Rugusi and Manga (about 40 miles south-east of Handeni), in a southerly direction towards Mandera, on the Wami River. Accordingly I ordered General Hannyngton to send Lt.-Col. W. J. Mitchell, with a detachment of the 40th Pathans, after the enemy on this route, and to return with the rest of his brigade, as well as the 57th Rifles, to Lukigura, which was reached in time for them to take part in the operations through the Nguru Mountains. Col. Mitchell, in the meantime, had overtaken the enemy at Manga, at the same time as a cooperating detachment of the Cape Corps, sent from Kangata; the enemy was beaten and driven south to Mandera. Sadani Bay was occupied by the Navy on the 1st August, and a detachment of the West India Regiment was landed and moved south and then westward towards Mandera to co-operate with Mitchell in clearing the enemy from the lower Wanii River. This was successfully carried out, and thereafter the combined force marched southeast to Bagamoyo, which had been brilliantly occupied by the Navy on the 15th August, with the capture of a 4.1 inch naval gun in good order with ammunition. From Bagamoyo this force was to form part of a larger movement for the investment and capture of Dar-es-Salaam. The military operations on the coast and parallel to it were, subject to the I. G. C.'s orders, under the command of Col. C. U. Price, C.M.G., and were ably carried out.

Operations in Western Lake Area.
(27) To gain a complete picture of the state of the campaign in the northern parts of German East Africa at the end of June it is desirable at this point to consider the operations which were in progress in the west, in the neighbourhood of the Great Lakes. During the months in which my main columns were operating in the Kilimanjaro, Kondoa and Usambara areas and pressing their advance to the Central Railway, the "Lake Detachment," consisting of the 98th Infantry, 4th Bn. King's African Rifles, Baganda Rifles, Nandi Scouts, and other small irregular units had not remained inactive.

Previous to the inception of active operations in East Africa the task originally assigned to the detachment had been the defence of the Uganda and British East Africa frontiers on both sides of Lake Victoria, and this task had been faithfully carried out for many long months. Although no engagement of importance took place, there was constant activity, and minor affairs of posts and patrols on the 300 miles of front were of almost daily occurrence. This necessarily entailed a continued state of vigilance and strain and demanded a high state of efficiency on the part of all ranks. That this was maintained is amply shown by the success achieved whenever opportunity offered. An instance of this had occurred just before my arrival, when the small post of one officer and 35 men at Machumbe had utterly defeated a raid of the enemy, causing him a loss of 3 whites and 22 blacks killed and 1 white and 31 blacks captured.

Apart from the minor operations of the Lake Detachment, my principal concern in the west was to make the necessary arrangements to.
facilitate the advance of Major-General Tombeur's Belgian forces. As an advance from his headquarters at Kibati, north of Lake Kivu over the barren region of active volcanoes and in face of strong German opposition was impracticable, an arrangement had been concluded whereby part of General Tombeur's force was to move north-east to Lutobo, in order to advance from there in a southerly direction against Kigali, the capital of the rich German province of Ruanda. To enable him to do so it was also agreed that the base for this force should shift to Bukakata, on Lake Victoria, 150 miles further east and that we should be responsible for the transport and supply arrangements from this base. Owing to a variety of causes, the organisation and execution of these transport and supply arrangements proved a matter of considerable difficulty; and in consequence I sent Brigadier-General the Hon. Sir Charles Crewe,. K.C.M.G., C.B., of my staff, to the Lake area, to keep in touch with General Tombeur to advise me in regard to all necessary requirements, and to push the arrangements on as fast as possible. All difficulties were eventually overcome by General Tombeur and my representatives, and towards the end of April the advanced Belgian column under Colonel Molitor arrived at Kamwezi, 10 miles south-east of Lutobo. Thereafter rapid progress was made, and Kigali was occupied on the 6th May.

The occupation of Kigali made the position of the German forces further west on tne Belgian border untenable, and enabled General Tombeur to push forward columns both from the north and the south of Lake Kivu. It also became possible for Colonel Molitor's column to resume the advance to the southern end of Lake Victoria, and on the 24th June the Kagera river was reached.

(28) As the Belgian advance towards Lake Victoria progressed during April, May and June, our troops further north on the Kagera line increased their activity against the enemy opposed to them, and began gradually to drive him from his advanced posts. This withdrawal enabled our forces to become more concentrated, and finally it was found possible to release sufficient troops for an operation against Ukerewe Island. This island, the largest in Lake Victoria, lies immediately to the north of the German port of Mwanza, and produces much of the rice which forms the staple diet of a large part of the enemy's native troops. The island is within a few hours of Mwanza, and forms a favourable base for an operation against that town.

The operation for its capture was skilfully carried out on 9th June by Lt.-Col. D. R. Adye, commanding the Lake Detachment, in conjunction with the Naval Flotilla on the lake under Commander Thornley, R.N. The enemy was completely surprised, eight German whites, about 60 blacks, and two small field guns being captured.

(29) As the withdrawal of the enemy from the Northern Kagera river and Karagwe district became accelerated, it also became possible to concentrate our scattered posts in that area into a mobile fighting force which could act more effectively against the retreating enemy. For this purpose, Brigadier-General Sir Charles Crewe was appointed to the Lake command in the middle of June. With his mobile column he first occupied Bukoba and K.aragwe districts, and then proceeded south to arrange a combined forward movement with the Belgian forces. The advanced parties of the Belgian column had in the meantime reached Namirembe, at the southwest corner of Lake Victoria, at the end of June, the main body further west being hotly engaged with the German forces retreating from the north. Sir Charles Crewe came to the sound conclusion that the course which promised the best results was a movement of his force against the important fortified town of Mwanza, the occupation of which would give us an excellent base at the south of the Lake for the forward movement of the combined British and Belgian forces to Tabora. Accordingly, on the 9th, 10th, and 11th July, he embarked his force, consisting of about 1,800 rifles, at Namirembe and TJkerewe .Island, and on the night of the 11th landed a column under Lieutenant-Colonel C. R. Burgess at Kongoro Point, east of Mwanza, and the following day another column, under Lieutenant-Colonel H. B. Towse, further north at Senga Point. By the skilful disposition and movement of both columns—the one from the east, the other from the north-east— to Mwanza, he made it impossible for the enemy to withstand his advance; and the threat to the enemy's retreat from Burgess' column made the enemy evacuate the town on the 14th July. Most of the whites escaped down the Gulf in the s.s Mwanza and Reinrich
Otto and the steam pinnace Schwaben, with some lighters and boats, while about 400 to 500 askaris escaped down the main Tabora road. The enemy destroyed the powerful wireless station, but left a 4.1-inch naval gun in our hands. The pursuit was continued next day, both by a force moving down the Tabora road and by another embarked on the s.s. Winifred, which was disembarked some 22 miles south of Mwanza. Some distance south 5 German whites were captured, and the enemy steamers and lighters were found abandoned; much baggage and stores and ammunition, a Colt gun, and even much specie were found abandoned by the enemy in his headlong flight. The pursuit was continued as far south as Misungi, opposite the southern end of Stuhlmann's Sound. The s.s. Mwanza and the lighters have since been salved, and are now in active use. Our total losses in this operation were quite insignificant, while the enemy had been skilfully ousted from one of his most important strongholds. The rapidity with which the enemy abandoned his valuable Lake Provinces and Mwanza was a clear indication that the eventual retreat would not be towards Tabora, but further east towards Dar-es-Salaam, or south towards Mahenge.

Van Deventer's Advance to Central Railway.
(30) I now turn back to review the main operations further East, and shall begin with van Deventer's advance to the Central Railway. On the 24th June the 1st and 3rd Divisions came to a halt at the foot of the Nguru mountains. On the same day van Deventer, with the 2nd Division, attacked the enemy positions all along the line round Kondoa Irangi and succeeded in occupying them with comparatively small loss. For some time information had beer, received to the effect that a considerable transference of enemy forces from Kondoa to the Nguru front was in progress, and the enemy at Kondoa had been displaying a certain nervous activity and aggressiveness which are often the prelude of preparations for a retirement. After the action of the 24th June van Deventer proceeded to collect sufficient transport and supplies for the forward movement to the Central Railway. My orders to him were to clear his right flank towards Ssingida, to move a small column along the Saranda road towards Kilimatinde, and to move his main force towards Dodoma and further east on the road to Mpapua. My object was not only the occupation of the Central Railway, but more especially the movement of van Deventer's force to the east so as to get into closer co-operation with the force at the Nguru mountains in dealing with the main enemy forces as they fell back to the Central Railway.

Lt.-Col. A. J. Taylor was on 26th July sent with one infantry battalion, one mounted squadron and an artillery section to Ssingida, which, after some skirmishing on the way, was occupied on the 2nd August. A post was left there, and the balance of the column marched south to Kilimatinde. A similar sized columns, under Lt.-Col. H. J. Kirkpatrick, was on 14th July sent direct towards Saranda. Little opposition was encountered until they reached Mpondi, about twenty-four miles north-east of Saranda station. Here, in a country covered with very dense bush where scouting was well-nigh impossible, they suddenly found themselves under heavy machine gun fire from a well-prepared enemy position. There was no alternative but to go straight for the enemy in a frontal attack. The attack was successful, Mpondi was occupied the same afternoon, our losses being eight killed and nine wounded. The advance was continued next day, and on 31st July the Central Railway at Saranda was occupied, as well as Kilimatinde, seven miles further south.

(31) Van Deventer's main column, moving south along the Dodoma road, occupied Chamballa (Jambalo) unopposed on the 18th July and Aneti on the 19th July. The country further south was reported to be waterless and the enemy to be entrenched at the water-holes at Tissa Kwa Meda and Tschenene. Van Deventer therefore divided this force into two columns, and ordered General Manie Botha to move the Mounted Brigade by Tissa Kwa Meda and Njangalo towards Kikombo station on the Central Railway, while General Berrange, with two infantry battalions, a motor cycle corps and mounted scouts, was ordered to move by Tschenene and Meia Meia towards Dodoma. On the 25th July Tschenene was occupied with small loss, notwithstanding the strong enemy entrenchments, the success being largely due to the excellent work of the Armoured Motor Battery, which engaged the enemy at close range. On the 27th July Meia Meia was occupied, and part of an enemy mounted detachment was captured without any loss to us. On the 29th July Berrange occupied the Central Railway at Dodoma.

In the meantime the First Mounted Brigade had occupied Tissa Kwa Meda after a sharp engagement on the 22nd July. From here Brigadier-General Manie Botha, who had rendered great service at the head of this brigade, returned to the Union of South Africa on private business, and his place was taken by Brigadier-General A. H. M. Nussey, D.S.O., who had been van Deventer's Chief Staff Officer. After occupying Naju and Membe the Mounted Brigade on the 28th July reached Njangalo, where the enemy was driven from a strong position with the loss of a machine gun and 1,500 head of cattle. Kikombo station was reached on 30th July.

(32) By the end of July a hundred miles of the Central Railway was thus in our possession. Practically every bridge or culvert was found blown up, but our advance had been so rapid that the enemy had had no time for further destruction of the track. General van Deventer spent the following week in concentrating his forces, now scattered along the railway from Saranda to Kikombo, at Njangalo, which is on the main road to Mpapua. In the meantime serious attention was given to the transport and supply situation, which —already grave enough at Kondoa with a transport distance of 200 miles from the Moschi railhead—had now become still graver by the addition of more than a hundred miles, and for the immediate future presented the baffling problem of having to provide for another 120 miles in the advance to Kilossa. How this problem was solved, and van Deventer's force could be supplied for the advance to Kilossa, and even beyond to the Great Ruaha river, will be explained later.

The concentration of his Division at Njangalo was completed on the 9th August, and the advance was resumed on that date. But the sequence of events requires me now to turn to the operations through the Nguru mountains.

Advance through Nguru Mountains.
(33) The general situation in German East Africa in the first week of August may be summarised as follows: — Van Deventer had occupied the Central Railway from Kilimatinde to Dodoma; in the Lake area the British and Belgian forces were well south of Lake Victoria and preparing for a combined move towards Tabora. Further west a Belgian force had crossed Lake Tanganyika and occupied Ujiji and Kigoma, the terminus of the Central Railway. In the southwest General Northey's force had occupied Malangali after a brilliant little action, and was prepared to move towards Iringa, seventy miles further north-east. All coast towns as far south as Sadani had been occupied, and a small column was working its way southward to the Wami River and clearing the country, between the Nguru mountains and the coast. The time had therefore come for the First and Third Divisions to resume the advance to the Central Railway. Hannyngton's brigade had rejoined the First and Enslin's Mounted Brigade had joined the Third Division at Lukigura.

(34) For a distance of about forty-five miles the main road to the Central Railway passes close under the Nguru and Kanga mountains. The enemy had skilfully disposed about twenty companies or 3,000 rifles, with much heavy and light artillery, in the mountains and athwart the main road, which had been entrenched along; the numerous foothills which the road crosses. If we forced our way down the road against these formidable obstacles or moved by our left flank through the bush and tall elephant grass, part of the enemy force in the mountains on our right would get behind us and endanger our communications. It was therefore essential to advance by way of the mountains themselves and to clear them as the advance proceeded southward. This could best be done by wide turning movements through the mountains, which would have the effect of threatening or cutting off the enemy's retreat if he delayed his retirement unduly.

The main block of the Nguru mountains on the west is divided from the Kanga mountain and foothills of Nguru on the east by the rough valley of the Mdjonga river, which flows from Mahassi at the northern entrance to the mountains due south towards Turiani, where the main road round Kanga crosses it. Into this river two streams run from the north-west through gaps in the Nguru mountains, the one entering the valley near Matamondo, the other by Mhonda Mission Station, near Turiani. Along both these streams rough mountain footpaths pass . to the track which follows the course of the Mdjonga river. The enemy held the Mdjonga valley strongly from Mahassi-to
Turiani, and a turning movement would have to be further west so as to close in either at Matamondo or Mhonda Mission. My information was that both the Mdjonga track and the Mhonda footpaths were capable of carrying wheeled traffic. I therefore decided on the following dispositions for the advance. While General Sheppard's brigade was to make a feint from Msiha camp directly against the enemy's position at Ruhungu, on the main road, he was to move the bulk of his brigade by his left flank so as to arrive at Russongo river, six miles behind the Ruhungu entrenchments.

General Hannyngton's brigade was previously to have moved to Mahassi, and from there, accompanied by General Hoskins, was to advance along and clear, the Mdjonga valley. Brits' Division was at the same time to make a detour to the north by the Lukigura valley, and then, turning west through Kimbe, to enter the mountains further west of Mahassi and emerge from the mountains through the Mhonga gap behind the enemy's forces disposed along Kanga and the Mdjonga valley.

(35) On the 5th August General Enslin moved with the 2nd Mounted Brigade from Lukigura via Kimbe, and the following day entered the Nguru mountains some eight miles west of Mahassi. On the 6th Beves' Brigade followed the same route, while General Hannyngton marched along mountain footpaths straight from Lukigura to Mahassi. On the 7th General Sheppard moved out from Msiha camp. General Hannyngton worked his way down the Mdjonga valley and found no strong opposition until he reached Matamondo on the 9th. In the meantime Enslin had been moving rapidly through the mountains, and had arrived in the Mhonda gap and proceeded to occupy Mhonda on the 8th. He sent back word that the route through the mountains was entirely impracticable for wheeled traffic of any description. In consequence all our transport was sent back to Lukigura to follow Sheppard along the main road. Hoskins had also returned to rejoin Sheppard, and in view of the strong opposition Hannyngton was meeting at Matamondo and the impracticability of the mountains, I directed General Brits to take Beves' brigade down the footpath to Matamondo to reinforce Hannyngton.

One of Enslin's mounted regiments had lost its way in the mountains, and had also finally emerged at Matamondo. With the balance of his brigade, Enslin passed through the Mhonda gap and seized a series of positions across the road by which the enemy had to retire. These, however, he found it impossible to hold in view of the smallness of his force and threatened enemy attacks on his flanks. He, however, maintained his position at Mhonda Mission, and thereby forced the enemy everywhere to abandon his defence in the mountains and retire as fast as he could. If the terrain had permitted of the original scheme being carried out, and the whole Third Division had proceeded to Mhonda, the retreat of the enemy from these mountains would probably have been impossible.

(36) After stubborn fighting at Matamondo on the 10th and 11th the enemy was driven south with great loss, and a machine gun was captured from him. Our loss amounted to about sixty killed and wounded. On the 11th General Sheppard had worked his way through the dense bush round the enemy positions on the slopes of Kanga and had arrived at the Russongo river only to find the enemy gone. On the 12th I directed him to proceed due south by Mafleta to the Wami river at Kipera so as to be well on the left flank of the retiring enemy; he reached Mafleta on the same day, and on the following day occupied Kipera, where a small enemy patrol was driven off and a light bridge over the Wami was saved. On the 12th and 13th the other brigades had reached Turiani, the enemy having fallen back some miles further south. It was becoming clear that we were now dealing with only part of his force, and that the balance had retired, further south towards the Central Railway either in the direction of Morogoro or Kilossa. Our progress was, however, very much hampered by the numerous rivers flowing from the Kanga and Nguru mountains, over all of which the bridges had been destroyed and had to be rebuilt by us, including some of very considerable dimensions.

In spite of this and other difficulties I decided to give the enemy no time, and ordered Enslin's Mounted Brigade to proceed the same day (13th August) round the left flank along the Liwale river to Ngulu on the Mkindu river, where he was to be joined by the 130th Baluchis from Kipera, and thence to make for Kwedihombo and Mwomero, where the roads for Morogoro and Kilossa respectively, leave the Nguru mountains. At the same time Hannyngton's brigade was to work its way south along the main road. On the 15th both these places were occupied by Enslin and Hannyngton after only slight opposition.

(37) The bulk of the enemy force retired along the Morogoro road towards Dakawa on the Wami river, while a few companies went off along the Kilossa road. General Hannyngton was ordered to Mwomero to follow the latter to the Mkundi river, while the rest of the force was ordered to follow the enemy to Dakawa. General Sheppard had been ordered to cross the Wami at Kipera and to move his brigade along the right or southern bank of the Wami to Dakawa crossing. Sheppard and Enslin arrived on opposite banks at the enemy position on the 16th August, but the enemy was strong enough to hold Sheppard at bay some two miles north and at the same time to prevent Enslin from attempting to cross the river, which is both wide and deep. The mounted men got across the river higher up the following day, and the enemy retired precipitately as soon as he discovered the threat to his line of retreat. The Crossing was occupied by us the following morning (18th August). Our losses in this action amounted to about one hundred and twenty, while the enemy had been very severely handled. A halt ensued here, as the bridging of the river was estimated to take four days.

During this time Hannyngton was ordered to move his brigade to Dakawa, and the Cape Corps to take its place in following the retreating
enemy party towards Kilossa. This pause rovides a suitable opportunity to review van Deventer's operations along the Central Railway.

Van Deventer's Advance to Kilossa and Great Ruaha River.
(38) On August 9th van Deventer's Division had been concentrated at Njangalo, while the enemy was reported holding Tschunjo Pass with his left on Gulwe and his right on Kongoa. The advance was commenced on that day, and contact was established with the enemy at Tschunjo on the afternoon of the 11th. The troops had to march from Njangalo to T'schunjo over a waterless area, and went into action without any rest. Fighting continued nearly all night, and next morning the enemy was found to have retired, and was immediately pursued towards Mpapua, where he was again engaged and defeated before nightfall on the same day (August 12th). Fighting and marching had been continuous for forty-two miles. The enemy force from Tschiinjo to Mpapua consisted of twelve companies supported by artillery. Owing to the difficulties of the country the flanking movements were delayed and the advance had to depend for progress mainly on frontal attacks.

On the 15th August the enemy was again engaged at Kidete station, holding a strong position. He was supported by machine, field and heavy guns. On the 16th August the engagement at Kidete was continued until late in the day. The enemy was driven out by a flanking movement by the mounted troops who attacked in rear. Our casualties were six killed and thirty-nine wounded. From 15th August to 22nd August our troops were in daily contact with enemy driving him gradually from Kidete along the railway line to Kilossa and Kimamba, which were both entered on the 22nd August, the day before our advance was resumed on the Wami river.

(39) In reporting these arduous operations General van Deventer says: — "The railway from Kidete to Kilossa for a distance of twenty-five miles follows a narrow defile cut through the Usugara mountains by the Mkondokwa river; every yard of advance was stubbornly resisted by the enemy. Of the more important engagements those on the 19th at Msagara and on the 21st before Kilossa should be mentioned. In all the actions on this advance the fighting consisted of the enemy receiving our advance guard with one or several ambushes, then falling back on a well-prepared position, and retiring from that on to further well-selected ambush places and positions. All the time our less advanced troops were subjected to vigorous shelling by means of long range naval guns. Since leaving Kondoa Irangi the troops who have reached Kilossa by the shortest route have done at least 220 miles. Those troops who have gone via Kilimatinde and other places have done many more miles. Owing to bad roads, shortage of transport and the rapidity of advance, the adequate rationing of the troops was not possible. The underfeeding and overworking are sadly reflected in their state of health. Regarding the animals of my Division, the advance from Mpapua to Kilossa was through one continual fly belt, where practically all the animals were infected. After the occupation of Kilossa it was ascertained that the enemy held Uleia, twenty miles south, in force, and was being reinforced by troops from the Southern Command, who had opposed General Northey's advance. As my Division was now weakened by the absence of the First Mounted Brigade (less one regiment), which had gone to Mlali on 25th August to co-operate with the Second Mounted Brigade, and as my infantry was in an exhausted condition, the Commander-in-Chief's wire of 26th August, asking for an advance on Kidodi and Kidatu, imposed a task which I had not intended to ask from my troops before they had had some rest. The advance was, however, ordered in accordance with, the request of the Commander-in-Chief, the enemy being driven-out of Uleia on 26th August and out of Kidodi on September 10th.

From Uleia to Kidodi the country consists of high mountain ridges running across the road for several miles. These had all been entrenched by the enemy some time ago, so that in the various actions his troops could fall, back from one entrenched position to the next, a mile or so in rear. The operations thus called for an extraordinary amount of mountain climbing and constant fighting. The slight casualties sustained in the various engagements over an enormous track, of country, bristling with dongas and difficulties at every point, were mainly due to the advance being carried out by avoiding as far as possible frontal attacks. Dispositions were made with a view to carry out flanking movements while holding the enemy to the position occupied by him, but this the enemy carefully avoided, and under cover of darkness the engagement was usually broken off and a retreat effected.

The success with which the whole movement from Kondoa Irangi to the Central Railway, thence to Kilossa, and on to the Ruaha river, was carried out is due to the loyal cooperation and splendid spirit displayed by all units under my command.

It is difficult to express my high appreciation of the conduct and spirit of the troops, who all worked with determination and zeal; their endurance and hardships during long marches through dry and waterless stretched on scanty rations form an achievement worthy of South African troops."

Occupation of Morogoro and of Uluguru- Mountains.
(40.) When the advance through the Nguru mountains began I entertained some hope that, even if we failed in cornering the enemy in those mountains, he might still be brought to bay at Kilossa, on the Central Railway. Our information tended strongly to show that, if the enemy retired from the railway, Mahenge would be his next objective; and as the most convenient point of departure for Mahenge appeared to be Kilossa, there was som ejustification for the hope that our rapid advance from the north and west might cut the enemy off in the direction of Kilossa. It may, however, have been the rapid progress of van Deventer towards Kilossa that caused the enemy to retire with his main force towards Morogoro. Whatever the cause, our information did not leave us in any doubt as to the fact that the bulk of the enemy forces had retired to Morogoru. The next move now was to try and bring the enemy to bay at Morogoro, if possible. To this end Enslin, whose brigade had been ordered to the Central Railway on the 21st August and had occupied Mkata station on the 23rd August, was ordered to proceed immediately to Mlali, about fifteen miles south-west of Morogoro, on the road to Kissaka, round the west of the Uluguru mountains. Mlali was successfully occupied by him on the 24th August. General van Deventer was asked to send the First Mounted Brigade, under General Nussey, to reinforce Enslin so that it would be impossible for the enemy to force his way south by that route.

The next point was so to arrange the advance of our other forces from Dakawa as to block also the road leading from Morogoro by Kiroka, round the eastern slopes of the Uluguru mountains, and thus to bottle the enemy up in Morogoro. I was not then aware that a track went due south from Morogoro through the mountains to Kissaki, and that the capture of the flanks of the mountains would not achieve the end in view.

On the morning of the 23rd August our forces crossed the Wami by the mow completed bridge, but instead of moving forward to Morogoro we moved backward down the right bank of the Wami for about nine miles, and from there struck due east so as to cross the waterless belt of about twenty-five miles to the Ngerengere river, north-east of Morogoro. Owing to the nature of the country and the bush, the heat, and the absence of water, the march for that and the following day proved one of the most trying of the whole campaign; but on the night .of the 24th August we were encamped on the Ngerengere river, in the neighbourhood of Msungulu, some eighteen miles north-east of .Morogoro. A mounted detachment under Colonel A. Brink, General Brits' Chief Staff Officer, had preceded us, and had that morning seized Mkogwa Hill, some three miles further south-east on the other side of the river. The move must have been a surprise to the enemy, who, evidently misled by Enslin's march into the belief that the whole force would move to Morogoro by |the west, had .massed his forces on the road between Dakawa and Morogoro and further west along the railway.

Owing to the exhaustion of man and beast, the next day was spent in reconnoitring the country, and on the 26th August the advance was resumed, General Hannyngton being directed to Mikesse station, twenty miles east of Morogoro, and the brigades of Sheppard and Beves moving up the Ngerengere towards .Morogoro. Both places were occupied on the 26th August, only, however, to find that the enemy had gone, the Commander-in-Chief Von Lettow and Governor Schnee with a force on the track due south of Morogoro through the
mountains, and another force by the eastern or Kiroka route, while Enslin was engaged with a third force at Mlali.

At Morogoro I found many proofs of the precipitate flight and demoralised condition of the enemy forces, and I decided to continue the pursuit in spite of the fact that my forces and animals were worn out with the exertions of the last three weeks .and that my transport had reached its extreme radius of action. General Sheppard occupied Kiroka on the 26th, and General Hannyngton was ordered to continue the advance south after the retreating enemy. By the 30th August the First Division had pressed the enemy over the Ruwu, having been continually engaged with him since the 27th.

(41) It is unnecessary to describe in detail the events of our advance along the eastern slopes of the Uluguru mountains. The enemy fought rearguard actions every day, and held up our advance at every convenient place. Unfortunately the country is very well suited to his tactics. The road passes through very difficult broken foothills, covered either with bush or grass growing from six to twelve feet high, through which any progress was slow, painful and dangerous. The bridging of the Ruwu took several days, and for some distance beyond the road passes along the face of precipitous rocks, round which the enemy had constructed a gallery on piles to afford a track for his transport. As the gallery would not carry our mechanical transport, it 'took us some days to blast away the mountain side and construct a proper road. The gallery would not carry the 4.1 inch naval gun of the enemy, which was found destroyed near the Eluwu. South of the Ruwu, towards the Mwuna river, our advance proceeded not only along the main road to Tuio, but also on a track to the west of it to Kassanga, and to the east of it by the Tununguo Mission Station. The nature of the country and the continual fighting made our daily progress slow, while road making and bridging beaind engaged the attention, not only of the pioneers but of a large portion of the troops as well.

Between the Ruwu and Mwuha rivers the road passes first through swampy country and then over one of the spurs of the Uluguru mountains, which ends with a precipitous face, to the south. Through this spur and down this face a mountain pass was cut in the rock, which took the technical corps, as well as most of General Sheppard's brigade, several weeks, and will remain a notable and enduring engineering feat. Almost every day prisoners were taken, and in one of these daily actions a machine gun was captured. On the 10th September Tulo was occupied, and Hannyngton's brigade, which was leading the advance, moved on towards Dutumi, where the enemy
made a resolute stand for several days, being only finally driven south to the Mgeta river on the 13th September.

(42) I now, turn back to review the operations inside and along the western slopes of the Uluguru mountains. As already stated General .Enslin's Mounted Brigade reached Mlali on the 24th August from Mkata station. Early on the morning of that day the advance scouts of the brigade rushed Jisagale Hill, a small isolated hill athwart the road to the south, and captured an ammunition depot of the enemy, in which about one thousand shells for the naval and other guns of the enemy were found. At the same time one of the regiments galloped up the valley to the north of this hill, just as an enemy force was coming down the Morogoro road, and took up positions in the foothills in the immediate neighbourhood. In the afternoon this regiment, after severe fighting, found their positions in the valley untenable, as the enemy was gradually working round them in the hills and bringing converging fire to bear on them. They retired a short distance to the south, but remained in possession of the road. Fighting continued during the following day, and as the enemy found it impossible to dislodge our men from the road, they destroyed two naval guns, one 3.4 inch and the other 4.1 inch, and retired into the mountains towards Mgeta Mission station, which is situated about ten miles further into the mountains. Leaving their horses behind, the men worked their way after the enemy into the mountains, and on the 27th General Nussey, whose brigade had in the meantime joined that of Enslin, occupied Mgeta Mission, while- Enslin's men, who were moving into the mountains in a more southerly direction with the intention of cutting off the retreat of the enemy, had driven them off Hombossa mountain south-west of Mgeta.

At this stage I arrived with General Brits at Mlali and ordered Nussey to follow the enemy through the mountains along the course of the Mgeta river, while Enslin was ordered back to the track which proceeds round the west of the mountains by Mssongossi river and Mahalaka to Kissaki at the southern extremity of the mountains. In this march Enslin's brigade was joined by Beves' two infantry regiments and was accompanied by General Brits.

(43) It was clear to me from the vast quantities of heavy gun ammunition captured at this and various other points in the TJluguru that the enemy had intended a long and elaborate defence of these mountains, and that it was the unexpected arrival of General Enslin at Mlali and the audacious and successful pursuit into the mountains, combined with the operations of General Hoskins' Division on the other side of the mountains, that had forced the enemy to abandon his plans and retreat towards Kissaki. Nussey, followed only by porter transport, slowly worked his way southward through the mountains, finding much ammunition abandoned everywhere. General Brits, on arriving at Mssongossi river, found that it was impossible to take his guns or wagons any further, and from there they had to return'to Morogoro and rejoin him later at Kissaki by the eastern route. From Mahalaka to Kissaki he followed the elephant track which had been the route of Burton and Speke's journey into the interior in 1857. On the 5th September the neighbourhood of Kissaki was reached without any serious opposition. Nussey had not yet arrived and, owing to the roughness of the mountains and some damage to his wireless, no communication could be established with him. In spite of this, however, General Brits decided to attack Kissaki on the 7th September. Beves was ordered to follow the footpath southward along the Mgeta into Kissaki, while Enslin, with the mounted men, marched round by the right, so as to attack from the west and southwest.

Kissaki was found to be strongly held, the bulk of the enemy being on the right bank of the Mgeta in front of Enslin, while dense bush prevented Beves on the other side of the river from offering any effective assistance to the former. The enemy's superior force therefore found it possible first to- threaten Enslin's left flank by moving between him and Beves, and when Enslin weakened his right flank to reinforce his left, the pressure of the enemy again became too strong on his right. He therefore decided to retireat night, having lost nine men killed, twelve wounded and seven captured. Beves was also ordered to withdraw, and the whole force entrenched below Little Whigii hill, six miles north of Kissaki, and awaited the arrival of Nussey. Nussey, who was in ignorance of these events or the position of General Brits, arrived before Kissaki on the morning of the following day, and an action developed, in which he gallantly held his ground against much superior forces till the evening, when General Brits' messengers reached him with an order to withdraw to Little Whigu. His loss had been twenty-three killed and about the same number wounded.

Although this action could be heard from Brits' camp, it was found impossible, owing to the ruggedness of the terrain and the thickness of the bush, to go to his assistance. If communication between Brits and Nussey could have been maintained there is no doubt a joint attack would have led to the capture of Kissaki, whereas the two isolated efforts led to a double retirement and a regrettable recovery of enemy morale. It was only on the 15th September, when General Hannyngton had already captured Dutumi, eighteen miles further east, that General Enslin, by a flank movement round the north-east of Kissaki to Dakawa and the threat to cut off the enemy's retreat to the Rufiji, compelled him to evacuate Kissaki. The enemy had left behind his hospital full of sick and about seventy-two white Germans, but all supplies had been removed or destroyed. The enemy had now been driven everywhere from the Uluguru mountains, and taken up a defensive line along the Mgeta river south of Dutumi, and further to the west astride the road from Kissaki to the Rufiji. The attack, against him along this line was not pressed, as our men were exhausted and worn out with ceaseless fighting and marching for several weeks through most difficult country on half rations or less, and a thorough rest was imperatively necessary, not only on military but also on medical grounds.

Occupation of Dar-es-Salaam and of South Coast.
(44) Turning now to the coastal operations, which were conducted simultaneously with these movements in the interior, I have already stated that the Navy occupied Bagamoyo on 15th August. At this point General Edwards assembled a force of about 1,800 rifles under Colonel Price for the operations against Dar-es-Salaam. This force was divided into two columns, the smaller one marching south to the Central Railway at the Ruwu bridge with the object, if possible, of seizing that bridge before its destruction by the enemy, and thereafter swinging round towards Dar-es-Salaam; the other and larger column moving down along the coast towards that port. Neither column met any serious opposition on the march, as the enemy, aware of the overwhelming force moving against Dar-es-Salaam, and determined to avoid capture and also anxious to avoid siege operations, against a town containing a large German noncombatant population, had decided not to defend the place, and was everywhere falling back before our advance. Ruwu railway bridge was found completely destroyed.

Southwest of Ruwu a small German force was found, which was driven south with considerable loss, and the column then marched east towards Dar-es-Salaam. In the meantime the coastal column, after occupying Kondutschi and Mssassani Bay, had flung its right wing forward and occupied the Mssimbusi river, which flows round Dar-es-Salaam on the west and north. The Navy, at the same time appeared before Dar-es-Salaam, and on 3rd September the place surrendered, and was occupied by our forces on 4th September. The enemy forces had left a few days before. One 6-inch gun had been blown up, while the rest of their artillery was taken south. The railway station and harbour works had been effectively destroyed; the s.s. Tabora, Kdnig, and Mowe were found sunk in the harbour beyond any hope of being salved; but the Feldmarschall has since been recovered, and at comparatively small expense would again be seaworthy. The floating dock is also being salved.

(45) I considered that the time had now come to occupy effectively the whole of the coast, and accordingly made arrangements, with the Admiral for convoying forces south and co-operating in the seizure of all important points on the coast south of Dar-es-Salaam. In this way Mikindani (13th September), Ssudi Bay (15th September), Lindi (16th September), Kilwa Kissiwani (7th September), Kilwa Kivinge (commonly called Kilwa) (7th September), and Kiswere were all occupied before the end of September. At Kilwa, strong column was landed for operations, which I proposed to conduct against the enemy from that quarter. This occupation of the southern coast not only helped to pen the enemy up in the interior, but was intended to prevent any assistance from reaching the enemy from overseas.

Restoration of Central Railway.
(46) The restoration of Dar-es-Salaam harbour and the preparation of Kilwa as bases for our operations in the interior are both matters of some difficulty, and requiring some time to complete. Both are being pushed forward with the utmost energy. This is also the place to refer to the restoration of the Central Railway for our supply purposes. While the railway track was largely left undamaged by the enemy, the bridges had been carefully demolished. Between Kilossa .and Dar-es-Salaam alone about sixty bridges, some of very considerable dimensions, had been wrecked. To restore these so as to carry heavy locomotives would take many months, during which period all further operations would have to remain at a standstill and an unbearable strain would be put on our enormously stretched out transport lines from Moschi railhead and Korogwe on the Tanga railway. The difficulty had been solved for General van Deventer by a simple but ingenious device of the South African Pioneers under him. This was to restore the bridges with local material so as to carry a weight of about 6 tons, and to narrow the gauge of our heavy motor lorries so that they could run on railway trolley wheels over the line thus-restored. A motor tractor with trailer carries 10 to 15 tons of supplies. In this way General van Deventer had supplied his division over the railway track for the 120 miles advance from Dodoma to Kilossa, and but for this solution of his transport trouble his advance to the Great Buaha river at this stage would have been a physical impossibility. As soon as Morogoro was occupied, the same treatment was applied to that section of the line, with the result that since the 6th October the railway track has been open for motor traffic from Dar-es-Salaam to Dodoma, a distance of almost 300 miles, and our forces have been supplied from Dar-es-Salaam as sea base. By the end of October the railway will thus be open for motor traffic to Tabora, and the restoration and strengthening of the line for heavy locomotive traffic, for which heavy material has to come up from the coast, can proceed as circumstances permit.

Northey's Advance.
(47) A word more about the Western operations will complete the picture of the military situation in German East Africa by the middle of October. Brigadier-General E. Northey, A.D.C., whose operations have been conducted with remarkable ability and vigour, occupied Lupembe on 19th August and Iringa on 29th August; the latter place would have been occupied much earlier but for my advice to him to slow down while the line of retreat of the enemy's forces from the Central Railway was still uncertain. His Lupembe column is now on the Ruhudje river South-west of Mahenge, while his Iringa column is near the TJlanga river North-west of TMahenge. Ssongea in the South has also been occupied. The importance of his role is becoming more accentuated as the campaign progresses and the enemy forces may intend to retire South.

Advance to Tabora.
In the North-west, as already stated, Sir Charles Crewe's advance troops were at Misungi South of Mwanza on 16th July, while one Belgian column was further West near Biaramulo and Namirembe, and a second Belgian column was at Ujiji on Lake Tanganyika on 5th August. It was arranged between General Tombeur and General Crewe that their columns from Lake Victoria should advance simultaneously to St. Michael and Iwingo respectively on the Western and Eastern roads Southward to Tabora. Difficulties of transport supplies and organisation delayed their advance so that General Crewe only reached Iwingo on 7th August and Colonel Molitor's Belgian column could not be concentrated at St.Michael before the 22nd August. The British column reached Schinjanga on the 30th August. In the meantime the Belgian Ujiji column under Colonel Olsen had steadily moved forward towards Tabora and on 1st and 2nd September fought actions with the enemy to the West and South-west of Tabora. General Tombeur therefore decided to push Colonel Molitor's column Southward with, all possible speed so as to be able to co-operate with Colonel Olsen. Their combined operations caused the enemy to retreat, and the Belgian forces occupied Tabora on the 19th September, while a week later General Crewe's advanced troops occupied the railway at Igalulu East of Tabora. The enemy retired in two columns—one under General Wahle Eastward along the railway and then southward to the Itumba Mountains; the other under Wintgens Southward via Sikonge. At the time of writing this report both columns are approaching the Great, Ruaha river North and West respectively of Iringa, and Northey's and van Deventer's patrols are in touch with them. Their object is evidently to form a junction with the main enemy forces further East.

Portuguese Advance.
In the extreme South General Gil with a Portuguese force has crossed the Rovuma river and occupied certain strategic points to the North of it. The net result of all these operations at the moment of writing is that the Germans have been driven South over the Central Railway and are now disposed as follows: —In the North-east, on the Rufiji river and about 30 miles to the North of it; in the West, along or South and East of the Great Ruaha river and Ulanga rivers. With the exception of the Mahenge plateau, they have lost every healthy or valuable part of their colony. In the East they are cut off from the coast and in the South the Portuguese army has appeared North of
the Rovuma river.

Behaviour of Troops.
(48) It would seem fit and proper to add a few words in recognition of the work done by the officers and men whom I have the honour to command. But in view of the foregoing statement of the main facts eulogy seems unnecessary and misplaced. The plain tale of their achievements bears the most convincing testimony to the spirit, determination, and prodigious efforts of all ranks. Their work has been done under tropical conditions which not only produce bodily weariness and unfitness, but which create mental languor and degression and finally appal the stoutest hearts. To march day by day, and week by week, through the African jungle or high grass, in which vision is limited to a few yards, in which danger always lurks near but seldom becomes visible, even when experienced, supplies a test to human nature often in the long run beyond the limits of human endurance. And what is true of the fighting troops applies in one degree or another to all the subsidiary and administrative services. The efforts of all have been beyond praise, the strain on all has been overwhelming. May the end soon crown their labours.

Special Services.
(49) I am particularly indebted to the following for their services during the operations: —

Major-General A. R. Hoskins, C.M.G., D.S.O., who has commanded the 1st Division and has rendered me the greatest services by the ability and loyal manner in which he has carried out my orders.

Major-General J. L. van Deventer, at the head of the 2nd Division, was throughout these operations in command of a widely detached movement, which he conducted in a manner worthy of the highest praise.

Major-General C. J. Brits, in command of the 3rd Division, has invariably co-operated loyally and ably in carrying out my wishes as intended.

Brigadier-General S. H. Sheppard, D.S.O., has, in addition to his services at the head of his brigade, used his great engineering capabilities to the best advantage on many occasions, thereby enabling our advance to proceed unchecked.

Brigadier-General J. A. Hannyngton has proved his worth as a commander in the Field, having been very largely employed in carrying out independent operations.

Brigadier-General P. S. Beves has sustained his high soldierly record, and the 2nd South African Infantry Brigade under him has borne more than its due share of the labours and hardships of the campaign.

Brigadier-General C. A. L. Berrange, C.M.G., at the head of the 3rd South African Infantry Brigade, has rendered excellent service with the 2nd Division and taken a leading share in all the hard work performed by that Division.

Brigadier-General B. G. L. Enslin, by carrying out two arduous turning movements with his mounted brigade, largely contributed to the rapid clearing of the Nguru and the Uluguru Mountains.

Brigadier-General A. H. M. Nussey, D.S.O., has rendered distinguished service, first as General van Deventer's Chief Staff Officer, and subsequently in command of the 1st Mounted Brigade, in succession to Brigadier-General Manie Botha.

Brigadier-General the Honourable Sir C. R. Crewe, C.B., K.C.M.G., rendered very useful service, first in organising the transport and supply arrangements for General Tombeur's force from Lake Victoria, and subsequently in commanding our advance to Mwanza and Tabora.

My heartiest thanks are due to Rear-Admiral E. F. B. Charlton, C.B., and all ranks of the Royal Navy for the very able and thorough manner in which they have furthered my plans, not only by occupying points on the coast, sometimes even without military assistance, but by enabling a change of base to be carried out first to Tanga and then to Dar-es-Salaam.

The work of the Air Services has been most creditable. In addition to their reconnaissance work, there is evidence to the effect that both material and moral damage has been done to the enemy by their constant bombing raids.

I have already alluded to the amount of engineering work that has had to be carried out. Both in bridge building and road making the engineers and pioneers with the force have worked very hard, and rendered very valuable service.

The Royal Artillery has invariably made the most of any opportunities that have offered for assisting the advance of the infantry.

The Supply and Transport services have spared no effort to cope with the enormous distances and the difficulties entailed in campaigning in such a vast and undeveloped country. The manner and rapidity with which the repairs to the Tanga and Central Railways
have been effected reflect great credit on all ranks of the Railway Services, and in this connection I should like especially to bring to notice the service rendered by Lt.-Col. C. W. Wilkinson, of the Railway Sappers and Miners, and Major J. H. Dobson, of the South African Pioneers, in carrying out the temporary repairs to the Central Railway which have enabled the troops in the interior to be supplied
from Dar-es-Salaam practically within a month of its occupation.

The work of the Medical Units has been very heavy, and all ranks have done their utmost in their care of sick and wounded and in arranging for their speedy evacuation.

The Ordnance Service is to be congratulated on having so successfully met the very varied calls made on it, which success bears testimony to the excellent organisation of that Service.

Great credit is due to the Signal Service for the really excellent way in which communication has been maintained. The operations have been carried on by three widely separated forces, which have each been again sub-divided into two or more columns, and this has strained the resources of ths Service to its furthest limits. It has only been by unremitting efforts that success has been achieved.

My thanks are due to the various Political Officers who have accompanied the columns, and by their work materially assisted the operations by helping to gain the confidence of the natives, which is so important a feature in a campaign of this nature.

The Officers of my Staff have given me every assistance. I would again especially mention the very great debt which I owe to Brigadier- General J. J. Collyer, C.M.G., my Chief of the General Staff, and to Brigadier-General R. H. Ewart, C.B., C.I.E., D.S.O., A.D.C., Administrative Staff, for the tireless energy and unfailing tact with which they have carried out their respective duties, thereby relieving me of all detail work and leaving me free to devote myself solely to the prosecution of the campaign.

Brigadier-General W. F. S. Edwards, D.S.O., has continued to render valuable services as Inspector-General of Communications, and has from time to time had control of minor operations on lines of communication, which he has always, handled to my entire satisfaction.

(50) A despatch giving the names of the officers and men whose services I also desire to bring to your notice is in course of preparation, and will follow at a later date.

I have the honour to be,
Sir,
Your obedient servant,
J. C. SMUTS,
Lieutenant-General,
Commander-in-Chief, East African Force.