you are currently viewing: {Location Field}

The Battle of Isandlwana

Isandlwana - The Missing Five Hours

Ron Lock – Peter Quantrill September 2010
download this article as a pdf

Appendix A: Campbell Collections map numbered 2, titled ‘Military Survey of the Country Around Isandlwana’ dated 13th November 1879.
Appendix B: Campbell Collections map numbered 3, titled ‘Military Survey of the Battlefield of Isandlwana’ dated 11th November 1879.
Appendix C: Campbell Collections map 2 superimposed with modern roads and topography.
Appendix D: Photograph assessed to be the ‘ridge/hill’ that Raw and Roberts were ascending when contact in the area of the crest and reverse slope with the Umcityu regiment was made.


Arguably few battles in British military history have resulted in greater historical debate or more divergent held viewpoints than the Battle of Isandlwana fought in Zululand on 22nd January 1879. Controversy that followed the overwhelming British defeat by the victorious Zulus centred round such issues as the roles played by Lord Chelmsford, the Lieutenant General Commanding in South Africa; Brevet Colonel Anthony Durnford, late Royal Engineers (RE) commanding No 2 Column and Brevet Lieutenant Colonel Henry Pulleine, commanding 1/24th Regiment around which the defence of the camp was framed. Opinion further differs on the shortage or otherwise of the ammunition supply to the firing line. Discussion continues on the intent of the Zulu army to do battle on the 22nd together with the pre-battle doctoring of the amabutho. (Regiments.) Perhaps the greatest debate centres round determining the location and immediate intention of the Zulu army when first located by Lieutenant Charles Raw of the Natal Native Horse. (NNH)

The aim in the first part of this refreshed thesis is to re-visit and evaluate key primary source material. This will track early Zulu movement out of the main Zulu bivouac, recognised to be located in the deep Ngwebeni Valley, and into pre-planned bivouac positions under the Nqutu range of mountains. This occurred in the early hours of 22nd January although it is accepted that significant number of ‘stragglers’ might not reach their destination before daylight.

The second part will examine in greater depth source material that, in our opinion, has never been rigorously evaluated, namely professional Forensic Document examination to determine the ownership of the handwriting of the annotated Isandlwana maps held at the Campbell Collections, KwaZulu Natal University, Durban. Near similar maps are also to be found at the Royal Engineers Museum, Chatham. These annotations display with some precision, the locations of named Zulu regiments in bivouac out of the Ngwebeni valley, (more comprehensively named in the Chatham maps), together with an indication of the ground location where the NNH initially probably encountered the Umcityu regiment.
In addition we reflect upon the apparent empathy and goodwill that existed between Brigadier General Sir Henry Evelyn Wood and Lieutenant Colonel Edward Congreve Langley Durnford, Royal Marine Artillery, younger brother of the slain Anthony.

Battle Primary Source Reports - 0530 hrs to 1030 hrs 22nd January

Accepted convention holds that the greater part of the Zulu army camped the night of the 21st January 1879 in the area of the Ngwebeni Valley, apparently with the intent of attacking the British camp on 23rd January. Reasons advanced are that the 22nd was not a propitious date to attack due to superstition surrounding the ‘unfavourable’ state of the moon. In addition, the time factor had prevented the ‘doctoring’ of the amabutho to prepare them for battle on the 22nd.

Various reported Zulu sources indicate that some regiments had indeed been briefed that no attack was to take place on the 22nd. It is reasonable to assume that such briefing of the regiments would have taken place prior to their arrival in the extended Ngwebeni valley, on 21st January.

Others hold the opinion that it was only as a result of Lieutenant Charles Raw, NNH, stumbling upon the Zulus at some point on Mabaso Hill, overlooking the deep Ngwebeni Valley, that a Zulu attack was precipitated. From this vantage point, it is contended, his troop fired down on the assembled host below, thus causing the Zulu attack to be prematurely mounted at or near mid-day on 22nd January. This has been the accepted tenet and popular view of the battle from the work of the late Donald Morris (The Washing of the Spears.) His views then gained traction with similarly minded researchers and writers. We question the credibility of the theory that Raw actually discovered the Zulu army as described above.

Based on our work reflected in Zulu Victory, The Epic of Isandlwana and the Cover-Up, we continued to revisit the principal primary sources. In addition we analysed further available fresh primary source evidence subsequent to the above publication, in particular the analysis of the ‘Missing Five Hours,’ an invented term that describes the critical period immediately after daylight on 22nd January. This thesis, together with any new perspectives gleaned, is done in a genuine spirit of enquiry.

Our conclusions are very different from accepted convention and in several respects have also evolved from those in Zulu Victory. We further emphasised that hearsay evidence was generally weak and unacceptable for analytical purposes. We advanced that hearsay may be defined as relaying gossip information by the ‘he said, she said’ method. The danger of this practice, of course, is the potential for unreliability; the person telling the story was not actually present when the story unfolded. Often a story can pass around a circle of friends and end up as a mangled form of the truth. And the telling of the Battle of Isandlwana has, over the years, relied on more than its fair share of hearsay, rather than analysis and conclusions drawn from primary and secondary source material.

As a starting point in developing our argument, it is generally accepted that the Zulu army was initially ‘doctored’ at kwaNodwengu prior to its initial move to the Ngwebeni Valley area. It would therefore be neither practical nor necessary to undergo further ‘doctoring’ a mere five days or so later. However, some modern historians have argued that when the Zulu army was about to commence battle, one last ritual was essential, namely a ritual led by the iziNyanga. [Witchdoctors - Herbalists] This may well have been the case in Shaka’s time, but was no longer practised by Cetshwayo. To further clarify the issue, we were helpfully assisted by Amafa, (The KwaZulu Natal Heritage Council) in seeking the opinion of a recognised and highly respected cultural authority, namely Prince Mangosuthu Buthelezi. His maternal great-grandfather was none other than King Cetshwayo and whose paternal great-grandfather was the Anglo-Zulu War Commander-in-Chief, Mnyamana Buthelezi. Whilst during that period in history no corroborative written reports were available, cognisance should be taken that there can be no finer source on the history of the Zulu nation than the Prince. This was his reply.

The Zulu army was ritually strengthened at kwaNodwengu. Once this had been done, the army was despatched and ready to do battle. There would have been no further rituals performed for either individual regiments or the army as a whole.

Thus, though a further  ‘doctoring’ may have been desirable in the heightened state of preparation prior to an attack, it was nevertheless not mandatory.

 Post-war interrogation reports did indicate that some regiments were briefed that it was not the intention to mount an attack on the 22nd and we accept this. However, for reasons that we will develop, when the Zulu commanders radically changed their plan of attack, (and evidence to support this will be advanced,) it was simply not possible to disseminate fresh detailed battle orders within the time constraints in the usual way of bringing key regimental commanders together. It must be conjectured however that the overall plan of attack had already been disseminated to individual regiments, hence their pre-planned bivouac positions in the early hours of the 22nd for which evidence will be led.
 Nor was the state of the moon of itself a decisive factor. The battle of Nyezane, fought on the same day as Isandlwana, indicates a Zulu readiness to attack on the day of the ‘dead moon’ if driven to do so by a tactical opportunity. 
Much of the historical theorising on the battle also tends to focus on events following the arrival of Colonel Anthony Durnford at Isandlwana Camp at approximately 1030 hrs. However, there is a plethora of mutually supporting primary source material that indicates deliberate Zulu preparatory manoeuvre being variously reported from about 0530 hrs onwards, preceding by several hours and therefore not in response to Durnford’s own arrival with the greater part of No. 2 Column. Preoccupation with Durnford’s supposed role and influence tend to suppress the significance of these other reports. Three main considerations set the scene and context for continued study.

First:  Having conjecturally decoyed half of No 3 Column (and two thirds of its artillery) from the camp in order to join Major Dartnell in an effort to locate, engage and destroy a strong Zulu presence approximately ten miles east of the camp in the Mangeni Falls area and,
Second: The reconnaissance of the Isandlwana Camp made by a mounted party that included Mehlokazulu kaSihayo, one of the commanders of the Nkobamakosi Regiment in the very early hours of 22nd January and,

Third: In the knowledge that the camp strength was substantially reduced by observation of a strong column leaving in the early hours of the 22nd on the track leading to Mangeni Falls.

That Consequentially

  Ntshingwayo kaMahole, overall commander of the Zulu Regiments, presented with a sudden and unexpected tactical opportunity, changed his battle plan from remaining concealed until 23rd, to one of mounting a deliberate attack as soon as possible on 22nd January. 
To assist with evaluating this reasoning we have divided the battlefield into two areas of engagement.

 First: The ‘Eastern Area’ covering the eastern and deeper reaches of the Ngwebeni Valley, together with the iThusi, Qwabe and Nyezi areas.
Second: The ‘Western Area’ comprising the areas that include the central point of Magaga Knoll (also known as Mkwene), the location of Barry’s NNC picquet, together with the large open, undulating terrain to the north and west.  As attacking operations developed during the day, the two areas effectively merged into one single co-ordinated battle zone. It would now be prudent to initially analyse British primary source reports.

Eastern Area

Trooper Barker, Natal Carbineers.

“ ……[we] arrived on the hill [assessed to be Qwabe] about sunrise [0522 hrs] After being posted about a quarter of an hour we noticed a lot of mounted men in the distance and on their coming nearer we saw that they were trying to surround us….. we discovered they were Zulus. We retired to Lieut. Scott about two miles nearer the camp [assessed to be Conical Hill] and informed him of what we had seen, and he decided to come back with us but before we had gone far we saw Zulus on the hill we had just left and others advancing from the left flank  [an area including iThusi Valley] where two other videttes (sic), Whitelaw and another had been obliged to retire from. Whitelaw reported, a large army advancing ‘thousands’ I remember him distinctly saying ….this would be about eight a.m.”
…….. shortly afterwards numbers of Zulus being seen on all the hills to the left front.”  1

This report calls into question how thorough and vigilant these videttes had been at the earliest stages of daylight. The words ‘large’ and ‘thousands’ within the context of the source and allowing for exaggeration, show that by first light more than a Zulu reconnaissance patrol was sighted, and was viewed as having aggressive intent by the vedettes expressly sited to detect such manoeuvre that could continue until virtually unobserved.

 In Barker quoting ‘A large army’ and ‘thousands,’ the vedettes positioned across a frontage of a mile or so were compelled to abandon their posts and in doing so, large tracts of dead ground were exposed for the commencement of Zulu manoeuvre. Indeed this dead ground remained out of sight of the British until contact with the two troops of NNH at midday or thereabouts, as will be evidenced later. Here it would be prudent to add how extensive the shallow areas and ‘dead ground’ exist along the full length of the Ngwebeni streams on the tablelands. 

J.A.Brickhill, Interpreter.

On the morning of 22nd January between 6 & 7 O’clock in the morning the Zulus showed in considerable force at the southern end of Ingutu Mountain.  

Again, referring to Raw and Roberts, he records:
At about eleven a.m. a party of them were sent back round the way they came, round Isandhlawana, & from there round the Northernmost point of Ingutu. 

Brickhill went on to add:
Shortly afterwards another force came into sight about the middle of the hill and intervening space was speedily filled in. 

The intervening space may be interpreted as the vast space between iThusi Hill and Barry’s picquet situated on Magaga, namely the Nyoni Ridge. This is indicative of the arrival in strength of regiments, or major elements, having exited the area near the Ngwebeni Valley area and advanced to position themselves on the escarpment overlooking the Isandlwana camp within an hour or thereabouts of first light. Brickhill, as interpreter, was located centrally to a fairly commanding position in front of the Columns Office.  2

Captain Edward Essex.
75th (Stirlingshire) Regiment, serving as the Director of Transport for No 3 Column. 

“…… until about eight A.M., when a report arrived from a picquet stationed at a point about 1,500 yards distant, on a hill, to the north of the camp, that a    body of enemy’s troops could be seen approaching from northeast.”  3

 Essex possibly confused a picquet with a vedette. The distance described by Essex makes it probable that the hill described was iThusi on which a vedette was positioned. The significance of the report lies in its timing, and its coincidence with movements referred to by Barker. All l well before the arrival of Durnford.

Lieutenant J.R.M.Chard, RE.
Time approximately 0930 hrs by estimation.

  I also looked with my own, [field glass] and could see the enemy moving on the distant hills, and apparently in great force. Large numbers of them moving to my left, until the lion hill of Isandhlwana, on my left as I looked at them, hid them from my view. The idea struck me that they may be moving in the direction between the camp and Rorke’s Drift. 4

Further prime source corroboration of substantial Zulu deployment prior to Durnford’s arrival. Chard clearly made the point of the possible commencement of the deployment of the Zulu right horn. The ‘far distant hills’ may indicate the iThusi area.

Western Area

Lieutenant W. Higginson, 1/3rd Natal Native Contingent (NNC.)
The first intimation we received about the Zulus was at 6 a.m when. Lt. Honourable Standish Vereker came into camp and said that the Zulus were appearing on the extreme left, and nearly opposite his outlying picket [Assessed as being somewhere north of Magaga Knoll and south of the Nqutu Range of hills.] …… Soon afterwards Colonel Pulleine sent me and Sergt Maj Williams came with me. We found Captain Barry [Comment: Commanding the picquet] and Lt Vereker watching a large body of Zulus on the extreme left of the camp, and they informed me that a large force of about 5,000 had gone round behind the Isandula Hill.  5

This report, made shortly after first light, indicates substantial Zulu deployment sighted within view of Magaga Knoll, together with an approximate’5, 000’ moving westward, therefore well clear of the Ngwebeni Valley with the possible intent to envelop Isandlwana. This occurred before Durnford’s arrival thus indicating Zulu aggressive movement, not only to deploy, but to do battle on the 22nd, confirmed by the actions observed both on the eastern and western areas. The estimated size of the Zulu force estimated by Higginson also indicates a deployment of a major functional part of the Zulu army. It follows therefore, that a deliberate plan by the Zulu High Command was already in place with the right horn located out of the Ngwebeni valley and in position north of Magaga Knoll (Barry’s picquet) at first light 22nd January    

Lieutenant Hillier, Lonsdale’s Natal Native Contingent. (NNC)
 At half past seven a.m. Lt. Veriker [sic] of the NNC who was on picquet duty with Captain Barry rode into camp and reported to Colonel Pulleine that the Zulus were advancing on the camp in large numbers6

This report corroborates that of Lt. Higginson, in that Zulu deployment was taking place in the open and in view of the camp’s outposts.
The words advancing on the camp are unambiguous and show aggressive intent to attack. Note the time: 0730 hrs 22nd January. 

Lieutenant C. Pope’s Diary. 2/24 Regiment, portion of which read:
“ Alarm- 3 Columns Zulus and mounted men on hill E. Turn Out 7,000(!!!) more E.N.E., 4000 of whom went around Lion’s Kop.[Isandlwana Hill] Durnford’s Basutos, arrive and pursue.”  7

 Pope, by direct personal observation, provided confirmatory evidence that a large Zulu force was sighted. Furthermore, the deployment was taking place prior to Durnford’s arrival. This is a valuable, and completely uncorrupted, collateral source report.  

The Zulu War Diary of Lieutenant Richard Wyatt Vause, NNH, recorded:
Durnford ordered me to ride back to meet our wagons as the Zulus were seen in our rear and he expected they would try to cut them off.  8
The time was between 1015hrs and 1045 hrs, with Vause expressing the opinion that the right horn was perceived to be a threat and well deployed at that time.

Thus from both the Eastern and Western area reports, it may be concluded that significant elements of the amabutho were in the process of, or had already, deployed out of the Ngwebeni Valley. Within an hour or two of daylight, they were discernible from the camp and its remaining outposts, moving deliberately and in strength. It is logical to conclude that at the very least, some of the key preparatory moves for a deliberate attack on the camp were underway.
How then could Lieutenant’s Raw and Roberts, commanding two troops of the NNH who were yet to arrive at the camp with Durnford, ‘discover’ the main amabutho sitting quietly in the Ngwebeni ravine? This is a direct contradiction of the aforementioned primary source reports (and importantly not hearsay) that large numbers of Zulus were already deployed out of the valley area and positioned on the Nqutu Plateau.
 We will return to events surrounding Raw later.  

It has also been argued that these reports are merely indicative of earlier Zulu reconnaissance and we accept that some reconnaissance was mounted. However this general view is unsustainable on more thorough evaluation. Most of the amabutho had left the deeper valley and were aggressively not only deploying, but also pushing back the outlying vedettes on the Eastern Area, and in visual view of the picquet on the Western Area. These very actions contradict the military definition of either a reconnaissance or fighting patrol, the numbers involved being well in excess of such definition.
In addition such a move – except if linked to an already taken decision to attack - would have precipitated the Zulus into the tactically vulnerable situation of having disclosed the location of at least 5-7,000 warriors. This would have been a powerful indicator that the main army was indeed close to the Isandlwana Camp, rather than still opposite or near Chelmsford’s force in the Mangeni Falls area.  


Most contemporary Zulu witness reports tend to be selectively compiled. We are therefore not provided with any sequence of Question and Answer that may enable us to form an independent view. The interrogator recording key points often packaged the report in a manner designed to ease the task of the military reader that, in some instances, obscured a clear meaning. Instances occurred whereby the prisoner quoted ‘Lord Chelmsford’ by name, being obviously guided by the interrogator. Very few high-ranking commanders were formally questioned, the exception being Mehlokazulu, son of Sihayo and one of the commanders of the Ngobamakhosi Regiment. There are two recorded interrogations of Mehlokazulu that differ substantially. The first took place at the time Mehlokazulu was expecting to be found guilty of a capital offence. He was under the threat of capital punishment and was not to know that circumstances would turn out very differently for him and it is reasonable to consider that he was experiencing a degree of anxiety that probably limited the extent to which he commented in this first interview. Here he confirmed that there was no intent to fight on the 22nd     
 He further confirmed that his part in the attack began when three mounted troops- black and white – attacked us first9

‘Us’ in this context, is Melokazulu’s wing of the Nkobamakosi regiment that was positioned to the east of the Umcityu regiment and out of the Ngwebeni valley.  

This mention is often taken to be yet another clear reference to Raw and Roberts’ Basutos, NNH, and is routinely used to attribute the direct cause for a general Zulu attack being mounted. Developed thought along these lines further supposes that the NNH attack took place overlooking the Ngwebeni Valley; the Zulu army, silent and in array, being surprised and caught unprepared in its overnight bivouac; and so on. However the Raw ‘theory’ could only work if virtually the whole Zulu army moved back to the Ngwebeni valley and hid after a period spent moving about on the Nqutu Plateau and  Nyoni Ridge. 

 This version is popular, much accepted and much quoted. Less known or analysed are the implications of Mehlokazulu’s second longer interview report,  compiled by the Governor of the jail at Pietermaritzburg and subsequently made available in full by Lieutenant-Colonel E. H. Steward, by then the Commanding Royal Engineer, of The South African Field Force, having replaced Colonel Hassard. This report is dated 28th November 1879. It was provided after Mehlokazulu was released a free man, all charges having been dropped and the possibility of capital punishment no longer an issue. He offers much more.

Mehlokazulu’s Second Interrogation Report:

 He [referring to Cetshwayo] then gave Tsingwayo orders to use his own discretion and attack the English wherever he thought proper [Indicating clearly that Ntsingwayo was at liberty to attack as and when he thought fit] and if he beat them he was to cross the Buffalo River and advance on Pietermaritzburg, devastating the whole country and to return with the spoil.  I caught up with the Zulu Army at the bottom of the Ngutu Mountains, about eight miles from Isandhlwana, where they had encamped. We learnt from our scouts that the English were encamped at Isandhlwana, but did not know that the army had been divided, as we did not send spies into their camp

We need to know that Mehlokazulu was not himself privy to all the tactical information received by the Zulu battle commanders. He is also referring to the position of the amabutho by onset of darkness 21st January and not the very different situation that became clear on the early morning of the 22nd.
We slept that night at the above-mentioned place. In the morning Tsingwayo called me and said. ‘ Go with three other indunas and see what the English are doing.’ 

I called the indunas and started off at a good pace. We were all mounted. When we got to the range of hills looking on to Isandhlwana, we could see the English outposts [mounted men] quite close to us, and could also see the position of their camp. The outposts evidently saw us, for they commenced to move about, and there seemed to be a bustle in the camp, as some were inspanning the wagons, and others were getting in the oxen. We immediately went back, and I reported to our commander Tsingwayo, who said, ‘All right, we will see what they are going to do. I went away and had something to eat, as I had no food that morning. Presently I heard Tsingwayo give orders for the Tulwana and Ngyaza regiments to assemble. When they had done so he gave orders for the others to assemble and advance in the direction of the English camp. We were fired on first by the mounted men, who checked our advance for some little time.”  10

This general narrative is corroborated by other independent British primary sources already quoted herein indicating that the amabutho had commenced deployment with battle intent, fairly soon after first light on 22nd January. The terminology ‘advance’ is also arguably indicative of preparation to ‘attack.’

There is also evidence that Mehlokazulu was, at some stage, interviewed by Brigadier General Sir Evelyn Wood VC who, in 1879, commanded the Northern Column (No 4 Column) and later reconstructed as The Flying Column.

In H.P. Holt’s book, The Natal Mounted Police, (NMP) published in 1913 and dedicated to Major General J.G.Dartnell NMP, Holt refers to Isandlwana and states:
It is a wonder that the whole force was not exterminated, for what Mehlogazulu [sic] a son of Sirayo, afterwards told General Wood, it appeared that the chiefs of the neighbouring impi decided to postpone such an easy task until they had first ‘eaten up’ the main camp.

As far as we are aware, this is the first publicly communicated instance that corroborates a possible meeting between Wood and Mehlokazulu.The resultant information gleaned would place Wood in a relatively unique position to pass measured judgement on tactical aspects of the battle. This will be shown in the military maps annotated by Wood in 1880 and will be discussed further.     

A report by Henry Longcast, Interpreter, when interrogating Zulu prisoner Umtyolalo, again corroborates early Zulu movement.
During the night of the 21st January they [Zulus] were ordered to move in small detached bodies to a position about one and a half miles to the east of Isandlwana on a stony table land about 1,000 yards from and within view of the spot visited by Lord Chelmsford and Colonel Glyn on the afternoon of 21 January. On arriving at this position they were ordered to remain quiet not showing themselves or lighting fires. Their formation was as follows. Centre – Undi Corps. Right Nokenke and Umcityu. Left Mbonambi and Nkobamakosi. 11

The wording ‘Lord Chelmsford’ and ‘Colonel Glyn’ is hardly from the source, but they do enable some position fixing. This substantiates a further indication of a deliberately ordered and organised move onto the wider Nqutu range, and cannot be said to refer to the Ngwebeni valley itself. Thus the main Zulu army was plainly intended to form up across a much bigger area than some analysts suggest.
 We will examine and evaluate this probability further and at this stage speculate on the reaction to these early Zulu movements by the command and control structure of the camp prior to the arrival of Colonel Anthony Durnford at approximately 1030hrs. Here it may be argued that Colonel   Pulleine was simply obeying his instructions supposedly issued to him by Major Clery relating to the defence of the camp.

This was not the impression of Horse Guards who, in an undated letter bearing the letterhead stamp of the War Office, specifically addressed the issue by stating:
Doubtless finding himself Senior Officer on the spot [Durnford] when action had already commenced he according to the custom of the service took Command, but this was now all too late a period to remedy the fatal error of position selected before his arrival.

This may be construed as either criticism of the position selected for the camp or for Pulleine’s conjectured inability to assess that he was, at an early hour, opposed by the main Zulu force. There is seemingly a case to be argued that the fate of the camp was already settled before Durnford’s arrival.

Military Survey Maps of Battlefield of Isandhlwana

Lieutenant-Colonel E.H. Steward, as Commander RE, had ordered Captain T.H. Anstey RE and Lieutenant C. Penrose RE to survey the country about Isandlwana. On 28th November 1879, Colonel Steward recorded:
The brother of Mehlo-ka-zulu … gave them assistance and accurate information with great readiness and good will……  and accurate information respecting the ground occupied by the Zulus, both before and during the battle, has been collected by them. 12
Anstey’s surveys comprised two base maps, one ‘signed off’ on 11 November 1879 and the other closely followed, being ‘signed off’ on 13 November 1879. The baseline mapping was printed on a glaze semi-transparent fabric, using waterproof inks, especially made to enable accurate map tracing before the advent of photographic copying. The same lithographed base maps also exist at the Chatham Museum (initially analysed by historians Julian Whybra and David Jackson) and, differing somewhat, only in the added manuscript annotations at the Killie Campbell Museum, Durban, now known as The Campbell Collections, University of KwaZulu Natal, Durban.

The first map, ‘signed off’ on 11 November 1879 was titled: ‘Military Survey of the Battle-Field of Isandhlwana’ and was originally lithographed in London by the Intelligence Branch, Quarter Master Generals Department Horse Guards, very probably in January 1880. The Campbell Collections map was annotated in ink handwriting and further marked as ‘No 3.’ The Chatham map was also marked outside the map margin, ‘No 3,’ whereas the Campbell Collections map differed in that  ‘No 3’ was in bold print and inside the map margin, written in a different hand. Arrows drawn showed Zulu lines of attack, together with named Zulu regiments. The annotation on the Chatham map had the words ‘N.N. Horse under Durnford’ whereas the Campbell Collections version differed to read ‘Hlubi’s Basutos.’  In addition there were other discernible differences between the two maps.

The Intelligence Branch then updated the same base map in March 1881. The updated version added the ‘Camp Layout together with the Defensive Outposts Night and Day.’ It was not annotated and serves no further purpose for analysis. Thus two lithographed maps exist under the title ‘Military Survey of the Battle-Field of Isandhlwana,’ one annotated by hand and marked ‘No 3,’ the other not.

 The second map, ‘signed off’ by Anstey on 13 November 1879 was also lithographed in London, January 1880, but was titled differently, namely ‘Military Survey of the Country Around Isandhlwana.’
The map held by the Campbell Collections is annotated in ink ‘No.2.’
The handwriting extends beyond the right hand margin, showing in detail the writers interpretation of the battle. In addition it marks some Zulu regiments deployed on the northern reaches of the Nqutu Plateau and under the Nqutu hills. More specifically, it marks boldly the letter ‘x’ as being Where the Basutos fired on the Umcityu.

The Chatham map significantly also marks ‘x’ in the same location, but with a different notation namely, Where the Umcityu first advanced. Ground inspection indicates that both interpretations are inter-related, namely the probability that the Umcityu had advanced from their bivouac, north of the Ngwebeni stream shown on map 2, crossed the stream and contacted Raw and Roberts in the general area of x.  

The Intelligence Branch then updated the same map in March 1881. This updated version was not annotated but was extended to show Chelmsford’s movements in the Mangeni Falls area. It has no bearing on our analysis. Thus four maps exist using the initial two base maps drawn by Anstey and Penrose.
It is not known how many maps, all similar to originals, were lithographed by Horse Guards. However, known sets are held by The Royal Engineers Museum, Chatham and The Campbell Collections in Durban. We have also established that the National Archives hold both base maps that were lithographed in London in January 1880.These maps were deposited with the National Archives by the Foreign and Commonwealth Office, date unknown to us, and not annotated. The maps also have no bearing on our analysis.

In summary, the only difference between the Campbell Collections and Chatham sets of maps lies in the additional added manuscript annotations in handwriting. Otherwise the arrows and lines on No 2 are the same.  In addition, both sets have the letters ‘a a a a ‘ marked by Anstey/Penrose in the deep Ngwebeni valley and extending eastwards. This is shown as the Valley in which the Zulu army bivouacked on the night of Jany 21st –22nd    

Here Anstey is indicating the Ngwebeni valley area in which elements of the Zulu army that could arrive, initially spent part of the night. It does not follow that they necessarily remained there and it also does not follow that the area ‘a a a a’ area is the only place that the Zulus established their bivouac. There is therefore no correlation between the ‘a a a a ‘ depiction and the ‘Raw theory.’
With the proven existence of maps 2 and 3 it is logical to question the whereabouts of designated map No 1. There is no such map held at Chatham or in Durban, and is conjecturally the version used to illustrate the Narrative.         

 So what are we to make of these annotations and the minor differences that exist between the Chatham and the Campbell Collections map annotations? The first critical issue was to check the source of The Campbell Collections maps.

Here it was revealed that they form part of the ‘Wood Papers’ (Brigadier General Sir Henry Evelyn Wood, later Field Marshal) acquired by the then Killie Campbell Africana Museum, as it was then known, in 1982. The Papers purchased from an American collector for the princely sum of $2,300 contained not only a vast selection of Wood’s papers and letters, but also all four of the aforementioned lithographed maps. If this assertion is correct, It follows therefore that the maps, being in Wood’s possession, belonged to him; but for what purpose? It also follows that the annotations on maps numbered 2 and 3 were in Wood’s hand.

From documents held at Chatham (requiring further examination) it is further conjectured that Colonel Edward Durnford, Royal Marine Artillery, brother of the slain Anthony, in an effort to protect the latter’s reputation, communicated at some stage with Wood. This possibly took place in January 1880 when Wood received an official communication from Horse Guards transferring him from Command Belfast District to Chatham.  13

Chatham also happened to be the barracks of the Royal Marine Artillery and Durnford was already stationed there at a time. This coincided with the availability of lithographed maps from the Intelligence Branch, QuarterMaster General’s Department, Horse Guards, London. These maps would have been available to both Wood and Durnford and it is reasonable to assume that Durnford, rather than Wood, annotated the Chatham maps by hand. It also appears that Wood had provided the Campbell Collections acquired maps on loan to Durnford. The exchange between Edward Durnford and Wood was then confirmed, together with assertion that the handwriting was that of Wood, clearly demonstrated by a letter held in the Campbell Collections from Durnford to Wood and dated 4 October 1880 that read:
I return the two maps [2&3] you so kindly lent me. I see you note the mounted natives under Colonel Durnford as ‘Hlubi’s Basutos.’ The two troops under his immediate command were 1 troop Basuto’s (Hlubi’s) & 1 troop of Edendale men. The 2 troops under G. Shepstone were Sikali men. 14

With the words ‘Hlubi’s Basutos,’ Durnford is referring to map No 2 held by the Campbell Collections and not the Chatham version. With the words ‘I see you note,’ Durnford implies that the author of these words was Wood.

(Comment:Hlubi’s claim, as noted by Wood on map 2 that he, on hearing the firing of the Umcityu persuaded Durnford to retire is discredited by the following evidence:

1.Statement of Lieutenant H.D.Davies commanding Edendale Troop, NNH.
...We looked up to the ridge on our front. And could see the enemy in great numbers, 1500 yards steadily advancing and firing at us. Colonel Durnford gave the order for us to extend our men and wait for the enemy to come within 400 yards of us, then Henderson’s and my Troop to retire …
(TNA(PRO) WO 33/34 Inclosure 2 in No.96.) 

2. Statement of Lieutenant W.F.D Cochrane, Transport Officer, No. 2 Column.
Having proceeded between 5 and 6 miles, a mounted man came down from the hills on the left, and reported that there was an immense “impi” behind the hills to our left. They were in skirmishing order, about 10 or 12 deep, with supports behind. They opened fire at us at about 800 yards and advanced very rapidly. We retired …….. (PRO WO 33/34, Inclosure 1 in No.80.)
At the time of Wood’s conjectured interview, Hlubi ruled over the Isandlwana area as one of the thirteen chiefs appointed by Sir Garnet Wolseley to reign over Zululand.)

In order that there can be no further doubt that the Campbell Collections notations on both maps numbered 2 and 3 were in Wood’s hand, we consulted Mike Irving, Forensic Document Examiner (USA) DIP.IR. (E.E) to obtain a professional opinion. He examined The Campbell Collections maps together with letters bearing the original handwriting of Edward Durnford, Brigadier General Wood, and Lieutenant Alfred Henderson, NNH. Henderson’s handwriting was subjected to scrutiny as some historical sources have suggested that he was the originator of the notations. Irving’s eleven page comprehensive and professional Document Examination report that included three charts, prove conclusively that the handwriting on maps numbered 2 and 3 held by the Campbell Collections were authored by Wood. The report reads:


Based on the documents made available to me and the examinations conducted, I am of the opinion that:

  1. The notations found on the original Maps No 2 and 3 examined at The Killie-Campbell library in Durban South Africa, and the original hand-written letters authored by Brig.Gen. Sir Evelyn Wood are of ‘Common Authorship.’  In other words, the notations found on Maps No. 2 and 3 in Durban have been authored by Brig.Gen.Sir Evelyn Wood.
  2. One can only imagine the circumstances under which the notations on the maps were authored. But the ‘Pictorial’ appearance of the two sets of writings would initially indicate that there are two authors. However, when examining the intricacies of the period style of writing and the presence of comparative inconsistencies in letter forms appearing in the known writings and map notations, issues of comparative pen control and habitual pen movements become apparent. The combinations of all of these issues supports a finding of ‘Common Authorship.’

Also confirmed was that the Campbell Collections maps were original 1880 lithographed versions.

Confirmation that Wood’s hand annotated Maps 2 and 3 now throws considerable additional weight behind the wording and description of the battle. That Edward Durnford and Wood had further communication is evidenced in that the Campbell Collections ‘Wood’s Papers’ also contain two letters that were addressed from Inspector George Mansel, (Second-in- Command to Major John Dartnell NMP) to Colonel Edward Durnford in which, in Mansel’s opinion, Durnford was blameless. 15.

Thus it is apparent that Wood conjecturally showed sympathy towards the Anthony Durnford ‘cause’ otherwise Edward Durnford would not have forwarded the letters to a hostile recipient. This view is further enhanced by Edward Durnford’s closing sentence to Wood that read:
With many thanks for your friendship.
We have already established that Wood, having served at Horse Guards, had the opportunity to acquire Anstey’s lithographed base maps on publication in January 1880. When then did Wood have the opportunity to visit the battlefield having possession of the maps?  
It is recorded in The Campbell Collections, that Wood was granted permission from the Adjutant General Horse Guards to be absent from Chatham to escort Empress Eugenie, mother of Louis Napoleon to the site where her son the Prince Imperial was killed. 16

 It was during this trip that Wood took the opportunity to visit Isandlwana.
Trooper Clarke, NMP, later to rise to the rank of Colonel who was one of 17 detailed from the NMP to act as escort, duly recorded this visit. Clarke kept a diary that read:
We arrived at Isandhlwana on 5th June.  …….. General Wood interviewed several Zulus who had taken part in the battle  …… We left for Rorke’s Drift on 8th June.  17

Thus two clear days were available for a detailed reconnaissance of the battlefield.  Was this the occasion that Wood met Mehlokazulu as earlier indicated in Holt’s work?  And was this the opportunity for him to annotate maps No2 and No 3? It is conjectured that this was the case.
Wood clearly showed more than a passing interest in the aftermath of Isandlwana. This is shown when three days after the battle, Wood wrote a letter to The Deputy Adjutant General, dated 25th January in which, extraordinary in view of the time constraints, he set out a portion of the Zulu army Order of Battle that attacked Isandlwana. It is assessed that these details were presumably obtained from deserters or captured Zulus subjected to interrogation.

Two months later Wood signed an interrogation report of four Zulu’s who served in the Nodwengu Corps at Isandlwana. In addition, it is conjectured that Wood may well have learnt intimate details of the battle from Major C.F. Clery, formerly of 32nd (Cornwall) Light Infantry, Principal Staff Officer to Colonel Glyn, 1/24 Regiment. The former served under Wood when transferred to the Flying Column prior to the Second Invasion.

Thus armed with a detailed and extensive knowledge of the battle,    
Wood’s hand written notes annotated on Map 2 (Appendix A) showed that the amabutho formed up with intent to attack on the 22nd January.
 Map 3 (Appendix B) clearly shows the lines of advance followed by identified Zulu Regiments. The detailed lines, arrows and positions are similar on both the Chatham and Campbell Collections maps.
Map 2 not only shows the positions of the Zulu Regiments in what we would term as a ‘Forming Up Place,’ (FUP) prior to attack, but also of some significance, (as already outlined) the marking of spot x by Wood and annotated thus:
I believe [originators underlining] about where the Basutos fired on the Umcityu. The Zulus attacked exactly as they had bivouacked – all in line of Regiments except the Undi Corps which was ½ mile in left rear of Ngobamakosi in bivouac and in action.
Accepted convention holds that ‘x’ should therefore coincide with the northern reaches of Mabaso overlooking the deep Ngwebeni Valley. It does not.
 ‘x’ in fact is located approximately three miles west south-west of the deep Ngwebeni Valley area (marked ‘a a a a a’ on map 2) as being the Zulu army bivouac on the night of the 21st, and some two miles west north-west of the iThusi feature. It is west of the existing road that leads off the tarred Babanango road to Isandlwana Lodge. With professional help, the location of x was then superimposed on to a current survey map, with allowance being given for a small margin of error. (Appendix C.) The co-ordinates in a GPS reading reflected the following:
S 28 degrees, 19 minutes and 17.3 seconds.
E 30 degrees, 41 minutes and 47.8 seconds.

This location matches closely the area described above, and far removed from the deep Ngwebeni Valley.
This is confirmed by Raw’s statement in terms of both distance and contact.
We left camp, proceeding over the hills, Captain George Shepstone going with us. The enemy in small groups retiring before us for some time, drawing us four or five miles from the camp when they turned and fell upon us, the whole army showing itself from behind a hill in front where they had evidently been waiting.  18  (Appendix D)
Brigadier General Sir Henry Evelyn Wood also made a further visit to the battlefield in November 1881 in the company of Colonel Redvers Buller VC, his second- in- command during the campaign.
Quote from Wood’s ‘Midshipman to Field Marshal’ in which Wood made clear that he knew the intimate tactical details of the battle.

Next morning I conducted Sir Redvers over the battlefield of Isandwhlwana [sic] which he had never seen, [By implication therefore that he, Wood, was familiar with the battle prior to this date] and we had the story told by combatants who took part in the fights; Englishmen of the Natal Police, by Basutos, by friendly Zulus fighting on our side, and by two or three mounted officers of Cetewayo’s army, which overwhelmed our forces. Their respective accounts tallied exactly; indeed it seems as if uneducated men who cannot write are more accurate in their description of events than are the Western nations. Corroboration of the location and bivouac of the Zulu army prior to launching its attack may also be found in a pamphlet published by Edward Durnford on 15th November 1880 titled:
Lord Chelmsford’s Statements Compared with the Evidence.

Here it must be said that the contents of Durnford’s pamphlet was published with the intent to restore the contemporary image of his fallen brother Anthony and in response to Chelmsford’s perceived attack on the conduct of Anthony Durnford. However, the following quote from the pamphlet meets the criterion of Zulu dispositions in both the Chatham and Wood’s maps.

Evidence taken on the spot fixes the exact position occupied by the Zulu army on the night of the 21st behind the hills on the left front of Lord Chelmsford’s
Camp – the extreme right 3 miles N.N.E., and almost in direct prolongation of the line of the camps, the left about 6 miles to the left front, and E. of the camp. My informant is a distinguished officer who served throughout the Campaign and who, as an authority, is probably second to none.
When the Zulu army was discovered by Captain G. Shepstone, its advance was direct from the bivouacs, and hence the swiftness of the attack on the company and picquet and camp.

This description is corroborated by all evidence thus far submitted.
 3 miles, magnetic bearing N.N.E. indicates the exact area of the Nodwengu bivouac. The latter could not have camped in the Ngwebeni deep valley as it is already recorded that it was sighted by Barry’s Magaga picquet shortly after sunrise on the 22nd. six miles to the left front magnetic east takes us to the area of the Qwabe valley, where the Ngobamakhosi and Undi Corps, reflected on map 3, bivouacked prior to driving Durnford back at approximately 1130 hrs. The distinguished officer is not named. Conjecturally either Wood or Clery would meet the criteria, but on the balance of probability, and in view of the established relationship, the former.

We then asked Rob Gerrard, AZW historian and lecturer for many years at the well-known Isandlwana Lodge that overlooks the battlefield, if he could identify the exact spot marked ‘x,’
 Rob kindly devoted a great deal of time to ground reconnaissance and, of critical importance, further identified and confirmed considerable dead ground south of the existing Babanango road that could with ease, hide a considerable force of Zulus.

Quote from Gerrard:
Looking at the map you gave me with the spot marked ‘x’ and Wood’s hand written annotation, I would accept the position of the ‘x’ as the approximate location where Raw’s contact with the Umcityu took place. If Raw saw the ‘chest’ and left horn, (the majority of the Zulu army) it is very likely that the ‘head and chest’ advanced through the narrow fold in the ground between the Kwa Nyezi, now filled in because of the road, and the Bizanani hills. The ground is far less rocky and considerably less steep than the route some historians pronounce they took. Having stood on the area of ‘x’ I can confirm that the ground gives credence to Raw’s following statement,
‘ The enemy in small clumps retiring before us for some time, drawing us four or five miles from the camp when they turned and fell upon us, the whole army showing itself from behind a hill, where they had evidently been hiding.’

Following the above assessment, we then took the opportunity to visit the battlefield for a field examination of The Campbell Collections annotated maps, accompanied by Rob whose help greatly facilitated the exercise.  We also, with the aid of a current Survey map and through professional help, superimposed and tracked the modern Babanango road together with the road that leads to Isandlwana Lodge and the battlefield, on to the map. (Appendix C.) In addition we have highlighted and added prominent ground features.

Two days of detailed reconnaissance criss-crossing numerous dongas together with waist high grass confirmed that Wood’s positioning of the area marked as x coincides with the ridge from which the Basutos first saw the Umcityu. This is confirmed by the primary source reports.

Raw described it as ‘a hill.’
J.N.Hamer (a civilian transport officer attached to the Army Commissariat and Transport Department who accompanied Shepstone and Raw) described it as a ‘ridge.’ …….and after going some little way, we tried to capture some Cattle. They disappeared over a ridge, & on coming up we saw the Zulus, like ANTS, in front of us, in perfect order as quiet as mice & stretched across in an even line.  19

Comment: Hamer does not indicate who initially opened fire. But he does confirm that the initial sighting showed that the Zulus were ‘in front of us,’ and that he was not looking ‘down into a valley.’ That description would match x and the ‘ridge.’  
Nyanda (The senior NCO of Robert’s Troop) described it as ‘the crest of a ridge.’ 20
‘The Narrative of the Field Operations connected with the Zulu War of 1879,’ prepared by the Intelligence Department Intelligence Branch of the War Office and published in 1881 described it as ‘ over a small rising ground.’

The location of x, photographed and attached as Appendix D would meet all the above criteria.
The ridge reached by Raw and Roberts is approximately two and a half miles short of Mabaso and the deep Ngwebeni Valley, the supposed Zulu bivouac. On the reverse side of the ridge there is sufficient ‘dead’ ground to conceal, out of view from the Magaga Knoll and iThusi, thousands of Zulus.
 In addition, we located two small Ngwebeni streams, one running west to east just north of  ‘x’ which merged with a south-north stream. This would seemingly tally with Wood’s positioning of the Umcityu and regiments to the east of the Umcityu.
Corroborating evidence is to be found in the statement of a Nokhenke deserter.
On the 21st, keeping away to eastward, we occupied a valley running north and south under the spurs of Nqutu hill, which concealed the Sandhlwana hill, distance from us about four miles and nearly due west of our encampment. 21
The key words are under the spurs of Nqutu hill that match Wood’s annotation on Map 2. Mabaso and the Ngwebweni Valley cannot be described as under the spurs of Nqutu hill. In addition, further corroboration of the distance from the camp is indicated as about four miles, thus matching the location of x.   

The deserter also confirmed that the initial contact of the NNH was with the Umcityu regiment, which tallies with the annotated maps held by Chatham and Campbell Collections.
The statement of Mhoti, umCijo regiment, provides additional Zulu corroboration.
Our whole force advanced towards the camp from the Nqutu range. 22
Note, not the Ngwebeni valley area.

Bertram Mitford’s Through Zulu Country, page 90, published in 1883, records a meeting with a ‘warrior of the Umbonambi regiment’ that describes the bivouac position on the morning of the 22nd as: We were lying in the hills up there. This description is far removed from any connection with the Ngwebeni Valley, the supposed bivouac location.   King Cetshwayo himself, on page 57 of ‘A Zulu King Speaks,’ edited by B Webb and J.B.Wright, when describing the Zulu encampment (logically sourced from Isandlwana’s returning izinduna) stated:
During the same night that followed upon the day on which my troops took up their encampment at the Ingudu [sic Nqutu] Hill.

Anstey’s base map of the Country Around Isandlwana show a range of hills rising from the Nqutu Plateau. From Tarantala Hill, some four miles to the north/north east of Isandlwana they extend eastwards to the final highpoint of the range that finish approximately north of the spot x. Anstey marked this range as the Ngutu Range. The Ngwebeni Valley cannot be defined as part of the Nqutu range as it is a further two miles east of the last hill in the range. 

The time/distance factor that preclude Raw and Roberts from reaching the Ngwebeni Valley is evidenced in our article published in the Journal of the Society for Army Historical Research, Volume Eighty-Three, Number 334. Here it was established that the distance from the camp, using the route that Raw and Roberts would have taken via Magaga Knoll (where Barry’s NNC picquet was commandeered to accompany the NNH) to Mabaso, was approximately 71/2 miles. Raw, however, describes the distance travelled as four or five miles. Hamer describes the distance as after going some little way. Both descriptions fit Wood’s location of x.

An undated document titled  ‘The Isandlana (sic) Disaster, a Memorandum from the Intelligence Department’ states:
He [Chelmsford] was lead away by the Zulus who decoyed him from the camp. In the mean while the Zulus collected in thousands under the hills near the camp.’ 23

This statement clearly endorses Anstey’s annotated maps. The location indicated is under the Nqutu Range and some distance from the Ngwebeni Valley. It also confirms the official view that Chelmsford was decoyed.

Further substantial evidence is advanced from the initial report on the ‘Isandlana Disaster’ released on 21st March 1879 by the Intelligence Department, Quarter-Master General’s Department, Horse Guards, signed by Lieutenant Walter James, RE and endorsed by Major General Sir Archibald Alison, Deputy Quartermaster General, Intelligence Branch. This document contains three sketches of Isandlana (sic) camp. All three indicate the area of the Zulu bivouac and describes it thus:
Behind the hills B.C. the Zulu army is supposed to have lain on the night of the21st Jan.
BC is shown as a line extending from north of Magaga Knoll to an area assessed to be where x is indicated in the Chatham and Campbell Collections annotated maps; clearly some distance from the Ngwebeni valley.

The report also describes Raw’s initial contact:
Shepstone’s men, while advancing, came upon a small herd of cattle. In trying to cut these off they crossed ridge B, [indicated north/east of Magaga Knoll Appendix D] and came suddenly on the Zulu army, at a distance of about three or four miles from the camp. 24 
Both distance and ridge meet the criteria of the annotated maps.  

 Finally, the contemporary British viewpoint, prepared by the Intelligence Branch of The War Office and published in 1881, titled:
 Narrative Of The Field Operations Connected With The Zulu War Of 1879, confirms that Raw’s contact took place in the area indicated by both the Chatham and Campbell Collections annotated maps.
The Narrative, Quote:
 It appears that Lt. Raw’s troop of Basutos which had been sent out to reconnoitre on the high ground north of the camp, had, after going some 3-4 miles came across a herd of cattle which they had followed over a small rising ground. From the top of this they had seen the Zulu army about a mile off, advancing in line and extending towards its left.

This official report is unambiguous in the description of the contact area and distance, both far removed from the Ngwebeni Valley. It should also be noted that the Zulu army was shown advancing in line, indicative of an attack underway on 22nd January. 

Contrary to popular belief, it is recorded that Shepstone and Raw joined forces with Barton and Roberts just prior to, or on contacting the Umcityu regiment. It is also recorded that the Zulus were the first to open fire  (indicating aggressive intent) and not, contrary to further popular opinion, the NNH.
Raw said:
The enemy had already opened fire upon us; we then opened fire upon them, and retired skirmishing on to the camp. Before this, my troop had been joined by Roberts’.

Nyanda’s (accompanying Roberts) noted:
The distance apart from the two bodies [Roberts and Raw] was about half a mile. We saw a handful (not many) of Zulus, who kept running from us. All of a sudden, just as Mr.Shepstone joined me on the crest of a ridge, the army of Zulus sprung up 15,000 men.

Apart from Raw’s confirmation quoted above, Nyanda further recorded:

The enemy then began firing at about half a mile distance and, we keeping at that distance, dismounted and fired. The question must be posed that if the NNH were looking down into the Ngwebeni Valley, then would the time and distance factor to exit the relatively steep valley allow the amabutho to emerge and skirmish as described by in the Narrative, and by Raw, Hamer and Nyanda?  
It is reasonable to suggest, in view of the importance of the spot marked x, how Wood reached such conclusion and who provided the evidence to convince him? A certain degree of dogmatism is indicated when x is marked with Wood’s underlined words:

I believe about where the Basutos fired on the Umcityu.

Apart from other Zulu unnamed sources, it has already been recorded that Wood met Mehlokazulu who served as induna with the Ngobamakosi regiment.
 In addition, Lieutenant-Colonel Steward’s report on Mehlokazulu added:
The brother of Mehlokazulu, referred to above, on one occasion gave them [Anstey and Penrose] assistance and information …….. and that accurate information respecting the ground occupied by the Zulus, both before and during the battle, has been collected by them.
Thus Wood also had access to the reports collated by Anstey and Penrose.

There were only three known surviving reports/letters on the action from the two troops of NNH that made contact with the Umcityu; namely Raw, Hamer and Nyanda. Roberts and Shepstone were both killed and Barton, to our knowledge, never left a report or letter on the action. Raw in due course served under the direct command of Colonel Buller and overall command of Wood. He served with the Edendale Troop of NNH, commanded by Lieutenant William Cochrane and saw action at Hlobane, Kambula and Ulundi. In view of Wood’s direct interest in Isandlwana, it is reasonable to speculate the possibility that discussions with Raw on the battle may have taken place. This conjecture has been strengthened with fresh source information in the way of family history and photographs in possession of descendant Paul Raw, kindly copied to the authors.
When Wood accompanied the former Empress Eugenie of France to visit the site where the Prince Imperial was killed, Wood, as already recorded, was appointed as her escort. Coincidentally, it was during June 1880 that Raw met and was presented with an ornament depicting an ostrich with crossed palms by the Empress. This is the first indication to our knowledge that Raw met the Empress, during which meeting it is presumed that the presentation was made. June also happened to be the month that the Empress and Wood spent two days camped at Isandlwana. On the balance of probability it is reasonable to assess that Raw was also at Isandlwana as the month coincides with his presentation. It was also recorded that on a reconnaissance prior to the Prince Imperial’s death, Raw was part of the Princes escort. The incident relating to the presentation may be explained as recorded in the Illustrated London News published 28 June 1879.
 After galloping about from point to point, the Prince espied a Zulu on a distant kopjie, and made after him. Off went Lieutenant Raw and the six Basutos after the impatient Prince, and on came Bakers Horse in the wake of the Basutos. The kopje was reached in time for them to see a few scared Zulus making off across the country.’ 
  It is conjectured that the June visit to Isandlwana coincided with Raw’s presentation. It is speculated that this would undoubtedly have given Wood the opportunity for Raw to show Wood the exact contact area that in turn was annotated by Wood as x which, together with all the other correlating factors, add substance to the accuracy of Wood’s maps 2 and 3. In turn, this is confirmed by the Chatham annotations.
It would seem reasonable therefore to surmise that considerable evidence has been advanced to substantiate our interpretation of the battle.

  1. This reasonable and rational appraisal of the ‘missing five hours’ prior to Durnford’s arrival, displayed Zulu intent to deploy and attack on 22nd January. Arguably this occurred when sufficient force had been assembled by first light on that day to attack the camp in strength and as soon as they became aware that their engagement with Dartnell had successfully drawn Chelmsford from the camp. This tactical opportunity simply could not be missed by the Zulu battle commanders and the regiments rapidly organised to mount a deliberate attack having moved into a ‘Forming Up Place’ on the northern reaches of the Nqutu Plateau and beneath the Nqutu Range straddling two small Ngwebeni streams.


2. That the discovery of portion of the Zulu army did not take place in overlooking the deep Ngwebeni Valley, but rather in the areas marked by Brigadier General Sir Henry Evelyn Wood in his annotated Campbell Collections maps that agree, in essence, with the Chatham annotated maps. The Wood detail is also consistent with conclusions that can be drawn separately by rational evaluation and analysis of most relevant primary and secondary sources.


Our thanks go to The Campbell Collections, University of KwaZulu Natal, Durban, for their unfettered access to the Wood Papers and for allowing us the use of the Wood lithographed maps contained in the Wood Papers for research purposes.

To Forensic Document Examiner Mike Irving for his expert and well- documented opinion together for his interest shown in the project.

We would also like to thank Rob Gerrard, FRGS, a Member of The Guild of Battlefield Guides, for his time consuming task in reconnoitring the battlefield and for accompanying us on our ground reconnaissance in locating the general area of x.  Rob, by virtue of his military infantry service, also brings a ‘soldiers eye’ to the assessment. Both the Wood annotated maps also coincide with conclusions drawn by Rob, based on years of experience and near daily familiarity with the area of ground under discussion.  

Our thanks also go to the advice and help given by Professional Land Surveyor, Barry Bechard, (B.Sc. SUR.)

To Draughtsman Clive Phelps for his professional expertise in superimposing prominent features on the original Campbell Collections map No 2 that include the current Babanango/ Isandlwana Lodge road.

We would further like to thank Nicki von der Heyde, BA HED FRGS, Campaign Trails, who, together with Jennifer Russell, Campaign Trails and Rob Gerrard took the GPS co-ordinates assessed to be the location of x.

To Major [Retd] Paul Naish, African Byways, a member of The Guild of Battlefield Guides, who has served in the Intelligence Field on three continents and who has made a detailed study of the Anglo-Zulu War, for sharing his knowledge and concurring with our conclusions.  

To Paul Raw, for his considerable help in researching and allowing us to share his knowledge of the Raw family history pertaining to his Great Great Uncle, Charles Raw, that include a photograph of the ostrich ornament presented to Charles Raw by Empress Eugenie in June 1880.
To the late John Radburn and to John Young, UK member of the KwaZulu Natal registered ‘Anglo-Zulu War Royal Research Trust’ for allowing us access to documents relating to the post Isandlwana career of Charles Raw, and to John for his helpful comments.

To Julian Whybra, acknowledged Anglo-Zulu War historian and author for his observations and comments. Although there is disagreement with our interpretation as to Zulu intent to do battle on 22nd January, Julian is in agreement with the area of Zulu bivouac on the morning of 22nd January and its consequent discovery by the NNH at the location marked x on the annotated maps.

We would like to thank Keith Smith, acknowledged Anglo-War historian and author for his valued comments. Keith has already suggested in his thesis titled Isandlwana: The Discovery of the Zulu Army, that the initial contact with the NNH was outside the Ngwebeni Valley.

Appendix A

Campbell Collections Map No. 2, titled Military Survey of the Country Around Isandlwana dated 13 November 1879, surveyed by Captain T.H. Anstey and Lieutenant C.Penrose, RE, annotated by Brigadier General Sir Henry Evelyn Wood.

Isandlwana map 1

Appendix B

Campbell Collections Map No. 3, titled Military Survey of the Battlefield of Isandlwana dated 11 November 1879, surveyed by Captain T.H. Anstey RE, Lieutenant Penrose RE, annotated by Brigadier General Sir Henry Evelyn Wood.

Isandlwana map 2

Appendix C

Campbell Collections Map No. 2 superimposed with modern roads and topography

Isandlwana map 3

Appendix D

Photograph assessed to be the ‘ridge’/’hill’ that Raw and Roberts were ascending when contact in the area of the crest and reverse slope (marked x) with the Umcityu regiment was made. (Taken during ground reconnaissance, May 2010, Gerrard, Lock, Quantrill) NOTE: The distance travelled was some four miles from the Camp, west to east, across the Nqutu Plateau. The deep Ngwebeni Valley is approximately two and a half miles further east.

Isandlwana map 4


  1. Statement of Trooper W.W.Barker. Rev. J. Stalker, Natal Carbineers.
  2. The Natal Magazine Volume ii 1879.
  3. WO 33/34, No 69, Chelmsford to Secretary of War, Court of Enquiry.
  4. Chard: Document held in Royal Archives, Windsor Castle. By kind permission of Her Majesty, Queen Elizabeth II
  5. WO 33/34 56333. Enclosure 2 in No 96.
  6. Hillier’s Letter to his father published 28 February 1879 in the Telegraph and Eastern Standard.
  7. Pope’s Diary: Narrative of the Field Operations Connected with the Zulu War of 1879, page 91.
  8. Photocopy of the Vause original diary held by Lock/Quantrill, by kind permission of direct descendant Don Stayt.
  9. Royal Engineers Journal published 2 February 1880.
  10. Royal Engineers Journal published 2 February 1880.
  11. WC 32/7711 Enclosure 3 No 21.
  12. Royal Engineers Journal published 2 February 1880.
  13. KCM 51091  - 89/9/33/2
  14. KCM 51091 – 89/9/32/1(a)
  15. KCM 51091 – 89/9/32/10
  16. KCM 89/9/33/7
  17. Kilie Campbell Africana Library for Research Purposes.
  18. WO/33/34 Enclosure No 91 February 1879.
  19. NAM,6807-386-8-14: Letter from Hamer to his father
  20. WO 33/34 Enclosure 2 in No 91.
  21. WO 33/34 Enclosure 2 in 80.
  22. KCAL: Symons Papers MS 1072.
  23. RA VIC/0 33/92: By kind permission of Her Majesty Queen Elizabeth II
  24. RA VIC/0 34/23. By kind permission of Her Majesty Queen Elizabeth II

Footnote:  Isandlwana Lodge, Campaign Trails and African Byways, when conducting tours to Isandlwana, are currently showing that Lieutenant Raw’s NNH contact with the Umcityu regiment was not in the deep Ngwebeni Valley, but rather as described in this thesis and marked x on the annotated maps.

Copyright: Ron Lock – Peter Quantrill, 2010

No part of this article, including Appendix A,B,C and D may be reproduced, stored in a retrieval system or transmitted in any form or by any means whatsoever without the prior permission in writing from the authors.