Scurvy and Lieutenant Evans

When the Terra Nova expedition left civilisation in 1910, vitamin C was not to be discovered for another twenty years. Whether scurvy was a deficiency disease or due to a food toxin was still controversial, despite a preponderance of evidence favouring a dietary deficiency. Whilst it is a controversial issue whether the men at Cape Evans possessed sufficient vitamin C in their diet, vitamin C does exist in seal meat; one commentator, Susan Solomon, believes that the fresh meat consumed by the men at Cape Evans would have given them sufficient reserves of the vitamin to keep them healthy during the journey to the pole and even on the return (Solomon Reference Solomon2001: 122–124). Though Lieutenant Evans’ extreme debilitation might at first argue against this, Evans’ companions Crean and Lashly showed no symptoms of scurvy upon their return to base. For this article I will narrow the parameters to one specific question: if all the men had the same diet, why did Evans succumb to scurvy when his companions did not?

The swift onset of Evans’ scurvy, first noted by Lashly on 22 January 1912 (Cherry-Garrard Reference Cherry-Garrard1994: 407), 91 days into the journey, has generally been attributed to two factors: Evans’ possible genetic predisposition to the condition (‘individual susceptibility’), and the physical labour involved in both his previous sledging trips and on the southern journey itself. With regard to that journey, Evans’ was unquestionably a far harder task than that of most of his companions. Whilst the others walked alongside a sledge hauling pony, Evans’ navigating team had to build snow cairns to guide the ponies (Evans Reference Evans1949: 205) and to man haul a sledge over the Great Ice Barrier for over 300 miles. When the ponies were shot at the foot of the Beardmore Glacier, Evans then had to endure over 200 further miles of man hauling, 100 miles of this to reach an elevation of 9,000 feet, before being sent back to base.

Such efforts certainly contributed to Evans’ later debilitation. In a US experiment from the 1960s, a walk of 10 miles a day added to a vitamin C-deficient diet led to scurvy symptoms becoming apparent far more quickly than in previous studies, ‘skin changes in 8 to 13 weeks, and gum changes in 11 to 19 weeks’ (Carpenter Reference Carpenter1988: 204) compared to a previous average, without exercise, of 30 weeks for visible debilitation (Carpenter Reference Carpenter1988: 202). Was this excessive labour the sole cause of Evans’ early breakdown?

It would be convenient to assume this, and leave things there, and one would if it were not for the case of Evans’ colleague Lashly, whose experiences mirror Evans’ so closely that he appears a kind of ‘test control’ in this scenario. An unvarying member of Evans’ motor party and man hauling teams, Lashly had no navigation duties but otherwise went through the same physical ordeal as Evans. Not only did Lashly suffer no equivalent debilitation, but at the end he still had abundant strength. On 22 February, on the return to Hut Point, Evans, strapped to the sledge, was close to death, and Lashly, by contrast, writes that he could walk and even run:

Both Lashly and Evans were impressively strong men. Lashly, as a stoker rating, had developed his physique through years of the toughest job on the ship, whilst Evans enjoyed entertaining others with feats such as lifting a man by the belt with his teeth (Evans Reference Evans1952: 41). Yet under similarly gruelling conditions, the 31 year old Evans had collapsed and the 44 year old Lashly had remained strong to the finish. Evans later attempted to address this discrepancy in his 1921 expedition narrative South with Scott:

On the ‘spring sledging journey’ to Corner Camp, 9–15 September 1911, Evans spent a mere five nights under canvas (Gran Reference Gran1984: 124–125). Why should Evans blame such a short trip for his scurvy, and could five extra nights in low temperatures really explain the difference between his health and Lashly's?

There has been a subtle but persistent undercurrent in the accounts by Evans’ expedition contemporaries that Evans’ condition may have been due to his own behaviour. In a letter to Atkinson Cherry-Garrard states, regarding the portrayal of Evans in his forthcoming expedition narrative The worst journey in the world:

If Evans was not in any way to blame for his scurvy, if it was solely due to excessive physical labour, then why should Cherry-Garrard talk here of ‘criticism’? The clue lies in the reference to the ‘medical side of the Expedition’: Cherry-Garrard is saying that Evans’ scurvy was not the fault of Wilson, who had worked hard to alert the officers of the Terra Nova to seal meat's reputed anti-scorbutic properties. In a 1905 article on scurvy in the Discovery expedition, Wilson wrote:

Wilson's persistent advocacy of fresh meat on the Terra Nova expedition led to good natured mockery from the others: in the South Polar Times he is given the ironic theme song ‘I cannot eate but lyttyl meate’ (South Polar Times 1914: 132). The geologist Frank Debenham wrote that Wilson habitually ‘kept a particularly sharp look-out for the possible symptoms of scurvy, and this made him the subject of many schemes to circumvent his regulations as to the issue of tinned sausages and other toothsome but possibly scorbutic rations’ (Debenham Reference Debenham1998: 139–140).

Was Evans one of those who ‘schemed’ to go against Wilson's edict, refusing fresh meat in favour of tinned food? It is worth noting that in The worst journey in the world (1922) Cherry-Garrard refers to the onset of scurvy in terms of diet alone. Consider this observation concerning the health of the Northern Party: ‘Fresh seal meat brought in from outside reduced the scurvy symptoms’ (Cherry-Garrard Reference Cherry-Garrard1994: xcvi). Cherry-Garrard also states that the Weddell seal provides, amongst many other things, ‘an antidote to scurvy’ (Cherry-Garrard Reference Cherry-Garrard1994: 165). This clear link between fresh seal meat and scurvy prevention would give any careful reader the hint that Evans’ scurvy was not an inevitable consequence of polar exploration. The photographer Herbert Ponting's film of the expedition, The great white silence (1924), also contained a title card explaining that fresh seal meat was necessary to guard against scurvy. Consider also Cherry-Garrard's comment on Atkinson's finding Evans at Corner Camp:

As Atkinson's note to Wright referred only to bringing ‘apples, oranges and onions’, is the inclusion of ‘seal meat’ here a mistake? I think it anything but an error. Another meaning for the word ‘wanted’ is ‘lacked’. Read in this light, this sentence becomes Cherry-Garrard's hint that Evans had succumbed to scurvy as the result of a lack of seal meat in his diet.

Not everyone enjoyed the taste of seal meat. We know that Lashly liked it: during his first Antarctic expedition he wrote, on 7 April 1902, ‘We get seal meat every other day. It is very good’ (Ellis Reference Ellis1969: 36). However, even the generous natured Wilson admitted that its flavour often left something to be desired:

In an interview of 11 April 1912, Evans does admit to a lack of fresh meat in his diet in relation to his scurvy, but blames his previous sledging trips into the interior for the deficiency:

So was Evans prevented from eating fresh meat by too long a period away from base, as alleged here? Besides the journey to Corner Camp, Evans’ other sledging trip (for survey work near the Turk's Head Glacier) lasted from 24 September to 13 October 1911. During 26–29 September, 5 October and 9–13 October Evans was stationed at either Hut Point or Cape Evans, where standard meals were possible (Gran Reference Gran1984: 131–135). This leaves 11 days on tinned or preserved food. Add to this the previous five nights on the Corner Camp trip, and we may estimate that Evans had been deprived of ‘fresh meat’ for a total of around 16 days more than Lashly. Again, why should this short period without ‘fresh meat’ make such a difference to Evans’ health?

In his editorial commentary on the Reader's Digest edition of Scott's last expedition, Debenham states that Evans did not eat his seal meat, but that this was not his fault:

This theory of an ‘allergy’ (taken literally as a medical term, as opposed to a profound dislike) is rather unlikely. Evans states that seal meat was the ‘principal diet’ at Hut Point during March–April 1911: ‘we were never in any great want of good plain food’ (Evans Reference Evans1949: 104). Had he been allergic to seal meat, the main fare for six whole weeks, this would have been the natural moment to mention it. Since he does not mention an allergy here or anywhere else, we can infer that Evans was capable of eating seal meat. The question is whether he chose to do so.

As it happens, Debenham changes his story in the Sydney Morning Herald in 1959, more than a year after Evans’ death. His interview is candid and engaging, not least on the subject of scurvy, on which he states:

This is the reporter's dream moment of unguarded indiscretion. The official narrative (or ‘the Reader's Digest version’, if you will) gives way to the spectacle of a personality trait becoming a potentially fatal flaw: ‘Teddy really was a very naughty boy.’

Those familiar with Evans’ life and career will acknowledge the truth of Debenham's description. From an early age Evans displayed a stubborn, mischievous streak which saw him expelled from public school and briefly sent to a ‘school for troublesome boys’ (Evans Reference Evans1952: 18). Though he later calmed down and shot to prominence as a cadet on the training ship RMS Worcester, he remained headstrong upon assuming command: a boatswain on the Terra Nova, according to family legend, ‘believed Evans was a self-opinionated officer, who would not be advised or told anything, and would not listen to what [Boatswain] Feather was telling him about the supply chain and arrangements’ (Stein Reference Stein2011).

Once settled at the base, Evans’ brashness did not endear him to Scott. The leader's harsh verdict was that Evans’ ‘boyish enthusiasm carries all along till one sees clearly the childish limitations of its foundation’ (Scott Reference Scott2008: 463). This is not an entirely fair assessment of Evans’ abilities; he was at his best during the gale of 2–3 December 1910, when both his organising the crew in bailing out and his successful unclogging of the main pump in freezing shoulder high water (Taylor Reference Taylor1916: 433) helped save the Terra Nova from foundering. This may even explain Scott's naming the base ‘Cape Evans’ on 4 January 1911, ‘in honour of our excellent second-in-command’ (Scott Reference Scott2008: 70). However, by October 1911 Scott would write in annoyance that Evans’ was ‘a sort of character which plants itself into a corner and will stop there’ (Scott Reference Scott2008: 303).

Ultimately, Scott's distaste for Evans came down to a personality clash. Scott, a scholar manqué, respected men such as Wilson and Bowers, men characterised by humility, intellectual curiosity and self-discipline. Evans, by contrast, lived for the adrenaline rush and those rare moments of crisis when he could seize the initiative: in an atmosphere of dull routine he must have found it hard to hide his boredom. Throughout his life and naval career Evans would frequently display the traits of the schoolboy: a need for excitement, an aversion to the mundane and an often blatant disregard for those orthodoxies he considered pointless or irrelevant.

It would have been characteristic of the headstrong Evans to have disregarded Wilson's advice, so was Evans’ collapse triggered by a deliberate, long term avoidance of fresh seal meat? Seen in this light, Evans’ puzzling emphasis on the physical toll taken by the five day trip to Corner Camp finally makes sense. He must have known that, sooner or later, questions would be asked about his eating habits. In his 1912 interview he implicitly exaggerated the duration of his sledging journeys: by doing so he could raise the issue of his diet, but make the fresh meat deficiency appear the result of necessity rather than choice.

Though Debenham alleges that Evans deliberately avoided seal meat, does Evans ever admit it himself? While apparently no straightforward confession from Evans exists, there is an oblique suggestion of one in, of all places, Evans’ children's fiction. In middle age Evans turned his hand to adventure serials with a dashingly heroic but often imperialist flavour (understandable at the time but uncomfortable for the modern reader). Of particular relevance is The mystery of ‘the Polar Star’, published in 1927, in which a young midshipman encounters an ice-locked British ship in the Antarctic to find its crew suffering from a familiar illness:

‘Refused to take advice’: if Evans’ own illness was brought on solely by over exertion, why does he state here that scurvy is caused by choosing tinned food in preference to seal meat, and ignoring medical advice? Within the context of Evans’ own experiences, this passage has the ring of a mea culpa. I believe it is as close to a public admission as Evans would ever come that he might bear some of the blame for his own illness.

Were Cherry-Garrard and Ponting correct to insinuate that Evans’ scurvy was due to his avoidance of seal meat? Was Evans entirely responsible for the onset of his own illness? It would be tempting to come to such a clear cut conclusion, but sadly we cannot know for sure. To state definitively that Evans’ refusal of the seal meat must have triggered his illness is not safe, given both the lack of evidence and subsequent scientific findings. For a start, when Evans is described as having refused seal meat, what is meant by ‘seal meat’? Is this an all-inclusive term which covers the organs as well as the flesh, or does it mean solely the flesh (the ‘very poor steak’ described by Debenham)? It would seem to be specifically the flesh that Evans avoided. We cannot be certain that Evans refused the seal liver as well, and the issue of the seal liver is crucial, as it was the chief source of vitamin C in the men's diet.

If a daily consumption of 15–20 mg of vitamin C per day is ‘enough to ward off scurvy’ (Carpenter Reference Carpenter1988: 231), then seal liver is a particularly rich source, providing between 14–30 mg of the vitamin per 100g if lightly cooked (Carpenter Reference Carpenter1988: 232). Seal meat is a less potent source, offering around 0.5–2.5mg of the vitamin per 100g (Carpenter Reference Carpenter1988: 232). In South with Scott Evans hints at seal liver not being to everyone's taste (Evans Reference Evans1949: 55) but in passing records two instances of his having eaten it (Evans Reference Evans1949: 195, 205). Seal liver provided anywhere from 6 to 60 times the amount of vitamin C provided by seal meat, so if Evans ate seal liver as often as his colleagues, and apparently ‘seal liver fry’ was one of the ‘standard’ breakfast dishes (Cherry-Garrard Reference Cherry-Garrard1994: 167, 193; Debenham Reference Debenham1992: 106), then he too would have absorbed the expedition's main source of vitamin C. All we can state with safety is that Evans refused to eat seal meat despite Wilson's medical advice, and that it is possible that this omission of a minor source of vitamin C was the tipping point which triggered his illness.

Certainly Debenham, along with Cherry-Garrard and others in the expedition, linked Evans’ refusal of the seal meat with his scurvy. They would have observed that Lashly ate his seal meat, did the same amount of physical labour and remained in good health, and they were aware that if Evans had returned to Cape Evans equally fit and well, the crisis which put Atkinson under strain (and which, I will argue, pressured him into making a fatal error) would never have arisen.

Indeed, Evans himself appears to have connected his refusal of seal meat with his subsequent illness. A sense of guilt at having endangered Crean and Lashly would explain why Evans urged his companions to abandon him on the Ice Barrier: ‘I endeavoured to get them to leave me when they came in with their suggestions, but it was useless to argue with them’ (Evans Reference Evans1949: 253). Lashly wrote that ‘[Evans] wished us to leave him, but this we could not think of’ (Cherry-Garrard Reference Cherry-Garrard1994: 414). Evans also wrote a note to explain the situation and protect his companions should they return without him (Ellis Reference Ellis1969: 145–146). If Evans’ diet was the element which triggered his scurvy, and again, we cannot know for certain that it was, then it must be said in his moral defence that on 13 February 1912 he was willing to pay the ultimate penalty for his misjudgement. Thankfully, there was no need for him to do so: he had the supreme good fortune to be accompanied by Crean and Lashly, two men fit and courageous enough to risk their own lives to save his.

We must also examine the extent to which Evans’ scurvy affected the chances of the polar party's safe return. Whilst Evans’ illness certainly placed a burden on the men back at base (and, as we shall see, on Atkinson in particular), the conventional narrative states that at this point Evans’ scurvy was responsible for the loss of a ‘vital message’ which resulted in the deaths of the polar party. As we shall see in the next section, this ‘vital message’ never existed in the first place, and is in fact an error dating from 1979.

The myth of Scott's ‘verbal message’

The story goes that, before the second returning party parted from the polar party on 4 January 1912, Scott gave important verbal instructions to Lieutenant Evans to convey back to base. These were, apparently, crucial last minute revisions to previous orders: for the first time, the dog teams were now to be used as insurance for the polar party's swift return to Cape Evans. Huntford's Scott and Amundsen (1979) is the earliest source for this story:

Later comes the devastating finale that Evans’ near fatal scurvy had left him too delirious to relay these verbal instructions, which in turn left the scientists ignorant of Scott's final orders. ‘In that crisis, Scott's vital message that the dogs were to meet him between 82° and 83°, casually mentioned just before parting with Evans, was forgotten’ (Huntford Reference Huntford1999: 520).

As a result of this fatal ignorance of Scott's ‘vital message’, or so Huntford's story goes, a young, inexperienced scientific assistant, Cherry-Garrard, was sent out with the dog teams. He knew to proceed to One Ton Depot at 79° 29′ S, 119 miles from Hut Point, but had been told that there was no urgency to travel beyond the depot. Consistent with his orders he ventured no further, but turned back for Cape Evans, unaware of the polar party's desperate struggles a mere 61 miles away.

What story could more colourfully illustrate the moral ‘For want of a message, the kingdom was lost’? Here are seven repetitions of the tale in subsequent expedition histories.

Limb and Cordingley, Captain Oates, soldier and explorer (1982): ‘They. . . turned northwards towards Cape Evans, carrying a message from Scott that the dogs were to be brought out to meet the returning Polar party between 82° and 83° S – a good deal further south than previously envisaged’ (Limb and Cordingley Reference Limb and Cordingley2009: 195).

Preston, A first-rate tragedy (1997): ‘Evans was also carrying an oral message which would play its part in the disaster ahead. Scott had changed his instructions yet again for the dogs. Meares was to bring the teams out to meet the returning party between 82° and 83° S towards the middle of February, to enable the returning Polar Party to be in time for the Terra Nova. . . Scott's great mistake was to assume that Evans would deliver the message in time’ (Preston Reference Preston1997: 180).

Lagerbom, The fifth man, Henry R. Bowers (1999): ‘A final word from Scott to Evans once more changed the plans for the dogs. Evans carried with him the message from Scott that the dogs were to greet the returning polar party between 82° and 83° South latitude, much farther than had earlier been determined’ (Lagerbom Reference Lagerbom1999: 173).

Wheeler, Cherry: a life of Apsley Cherry-Garrard (2001): ‘Evans had more verbal orders from Scott about the dogs: they were to come further south to meet him on his way back, and hurry him back to Cape Evans before the ship left. These orders were forgotten in the ensuing drama’ (Wheeler Reference Wheeler2002: 131–132).

Smith, I am just going outside. . . (2002): ‘Evans also carried another message for either Meares or Atkinson to bring the dogs out to 82° or 83° in mid-February to help the returning party catch the Terra Nova. . . Scott happily assumed that Evans, Crean and Lashly would have an easy 750-mile ride home. It was a dangerous assumption’ (Smith, M. Reference Smith2002: 198).

Fiennes, Captain Scott (2003): ‘Scott gave various messages to Teddy Evans to take back to Cape Evans. One, for Meares, updated his previous three instructions on what he wanted the dog teams to do. This last order cancelled the previous ones: Meares was to come out and meet Scott between 82° and 83° on the Barrier at some time towards the middle of February’ (Fiennes Reference Fiennes2004: 307–308).

Crane, Scott of the Antarctic (2005): ‘On 4 January. . . he had given Teddy Evans the last of a series of instructions taken back by the returning parties, ordering that the dogs should be brought out to meet him somewhere between 82° and 83° S′ (Crane Reference Crane2006: 555).

The story has even found its way into a recent novelisation of the expedition, Robert Ryan's Death on the ice (Ryan Reference Ryan2009: 452-453). In all of these accounts Evans’ scurvy, which led to this ‘vital message’ being lost, has been framed as a misfortune that could have struck any overworked member of the expedition. However, if Evans’ own behaviour was at least partially the cause of his illness, the picture changes. At a stroke, the loss of Scott's ‘vital message’, a message which, had it been delivered in time, could have saved four men's lives, could be attributed to Evans’ decision to ignore Wilson's medical advice. It is a serious charge indeed.

However, is it the truth? Did Scott's ‘vital message’ to Evans even exist? With a charge as grave as this, it is incumbent on us to examine the foundations for Huntford's version in detail.

First of all, a practical query. If Scott gave Evans alone the ‘verbal instructions’ which were subsequently lost in the confusion caused by Evans’ illness, then how would posterity know of them in the first place? There is no mention of these instructions in Evans’ writings; there is no record in Scott's journals either, a curious omission, given Scott's tendency to document his plans.

This tendency is amply illustrated in Evans’ South with Scott, which provides a list of Scott's orders as handed out to members of the team before his departure on the polar journey. Evans states that Scott ‘gave me all his instructions to the various parties to read’ (Evans Reference Evans1949: 167). These instructions are certainly, in Evans’ words, ‘explicit and comprehensive’ (Evans Reference Evans1949: 167); in them Scott second- and third-guesses events to work out probable outcomes, attempting to give every possible guidance to those away from his command. It seems highly uncharacteristic for Scott to turn away from pencil and paper just when they were most called for. It is not as if he were no longer interested in written communication: he did at this point entrust a letter for his wife to Evans’ care (Crane Reference Crane2006: 537). If Scott had changed his mind on the dogs as late as January 1912, and wished Evans to carry a message to this effect back to base, why would he not have spared five minutes to make a permanent record of such a crucial order?

We must also consider the final conversation between Scott and Evans on the polar plateau. In South with Scott Evans gives no specific details of this, but Sara Wheeler proposes a theory of their final exchange in a footnote:

The full quotation from Scott's letter to Kinsey, dated 28 October 1911, is:

‘I am going to take some steps concerning this’; sadly, Scott does not state exactly what he had in mind. Regarding the interpretation that Evans was to be ‘sent back in disgrace’, Wheeler has probably taken for guidance a letter from Atkinson to Cherry-Garrard, dated 5 December 1919:

That Scott regretted his appointment of Evans as second in command is indisputable. However, despite his criticism, Scott does not mention a decision to dismiss Evans in his journals. There is no evidence of any officer being aware at the time of such a decision: Atkinson only considered the possibility a full seven years afterwards. To the best of my knowledge, there is no extant letter from Scott that explicitly states an intention to send Evans home early.

That said, such a decision, to order Evans home, is not entirely implausible. There is a precedent in Scott's sending Shackleton home (albeit as an invalid) from the Discovery expedition in 1903. It is therefore possible that Scott's final conversation with Evans included the information that Evans was expected to rejoin the Terra Nova and sail home in early March. The idea of Scott brusquely dismissing Evans on the polar plateau is unlikely: would Scott really have entrusted a letter for his wife to Evans’ care, had he just sacked the man? However, it cannot be ruled out altogether.

Why raise the issue of whether Evans was sacked on the polar plateau? I do so because one thing should be apparent to even a casual reader: the theory of Scott's dismissal of Evans cannot conceivably stand in conjunction with Huntford's story of the ‘verbal instructions’. Wheeler alleges that Scott gave both sets of orders to Evans before Evans’ departure, but who in their right mind would deliver vital life-and-death orders together with a dismissal in disgrace? There would be the obvious danger of malice: the sacked man might well suppress these life saving orders out of spite. Even without this, one could reasonably imagine that the victim would be distraught by the news. How could Scott expect a freshly dismissed and probably devastated subordinate, ordered home ‘in disgrace’, to keep verbal instructions regarding the dispatch of dog teams in seven weeks’ time uppermost in his mind? No; Scott would never have been such a fool as to give crucial verbal orders to a man he had just sacked.

There is a further clue that suggests inaccuracy here: it is Scott's journal entry for 10 March 1912, when the polar party arrived at Mount Hooper Depot to find that it had not been restocked.

If the ‘verbal instructions’ had been given to Evans on 4 January, then why does Scott not write that it was Evans who must have had ‘a bad trip home’? It would seem logical to infer, if the instructions had been given to Evans, that the dogs’ absence would point to Evans as the weak link in the chain of communication. However, Scott's concern is not with Evans but with Meares. Why should that be?

The answer to this puzzle, and the key to the story of the ‘verbal instructions’, lies in the pages of Evans’ South with Scott, specifically the written memorandum given by Scott to Meares regarding the dog teams. It is dated 20 October 1911, eleven days before Scott's teams set off towards the pole, and must have been given to Meares at Cape Evans.

This, unquestionably, has to be the original source for Huntford's story of the ‘verbal instructions’. The parallels between these orders and Huntford's précis are unmistakable; here are the instructions, not verbal but written, to bring the dogs as far as the mid-point between 82° and 83°, to ‘meet’ the polar party and give them a chance to reach the ship before it sailed. In Huntford's book, these early written instructions became last minute verbal instructions. The story of Scott's last minute change of mind is an error on Huntford's part. It is no minor error, as we shall see.

Huntford's first error lies in the date and place the order was issued: not January 1912 on the polar plateau, but October 1911 at Cape Evans. His subsequent error concerns the date Scott expected the dogs to meet the polar party. This was not ‘mid-February’, as Huntford alleges, but the start of March. This is crucial because by 1 March the returning polar party had already passed the Mid-Barrier Depot at 81° 35′ S, 245 miles south of Hut Point. This means that the polar party's progress was actually ahead of Scott's projected estimate of 82° or 82° 30′ S (around 300 miles south of Hut Point).

When Huntford's ‘verbal orders’ are taken out of the story, we can see the truth of the matter. Scott's orders were simple. His contingency plan, which dated as far back as October 1911, was for the dog teams to meet the polar party at around the start of March 1912 and help bring them safely back to base.

It should here be stressed that, as with all aspects of polar exploration in this era, this plan could not be considered completely fail-safe. It demanded a great deal both of the dog teams and of the polar party in terms of navigation. On the almost featureless white plain of the Great Ice Barrier, both parties would have had to ensure absolute accuracy to avoid missing each other on the route, with the further possible complication of poor visibility in blizzard conditions, or travel at differing times of day. However, it should also be stressed that the polar party's navigation was accurate: their final camp, roughly 11–12 miles due south of One Ton Depot, was located easily by the search party in November 1912, and proves that they were on the right track.

Had an accurate navigator been sent south from Cape Evans, and had visibility been good, the dog teams and the polar party would have had a good chance of encountering each other en route as Scott had intended. Had everything gone according to plan, the dog teams would have arrived when they were needed, and though it would have been too late for Petty Officer Evans, who had died on 17 February, the lives of Scott, Wilson, Bowers and possibly Oates could have been saved.

Scott's orders for the dog teams

Why, then, were Scott's written orders from October 1911 regarding the dog teams not put into action at Cape Evans in February 1912? Was there confusion regarding Scott's expected date of arrival? Cherry-Garrard had written up Scott's 8 May 1911 discussion of the forthcoming southern journey in the South Polar Times, prefacing this report with

Debenham wrote in his journal that ‘this lecture was a suggestive one rather than a final one’ (Debenham Reference Debenham1992: 102). In it Scott speculated that, following Shackleton's average speed, the party would take 144 days to return to ‘Cape Armitage’ on 27 March, assuming a starting date of 3 November 1911 (Debenham Reference Debenham1992: 103). At this stage he is still uncertain of the roles of both the dogs and the motors, and places little confidence in either. However, these speculations cannot have been a source of significant confusion for the men at base. Scott's specific orders of October 1911 would unquestionably have superseded these preliminary statements, and it is Scott's written orders that should have been followed by the men at Cape Evans during February 1912.

Meares and Simpson were aware of Scott's most recent instructions. We know that Scott left orders with Meares regarding the dog teams in October 1911; from South with Scott we also see that he was in the habit of sharing his plans with his officers (he ‘gave me all his instructions to the various parties to read’ (Evans Reference Evans1949: 167)), so his statement to Simpson of 3 October 1911, ‘I think you are fully aware of my plans and wishes’ (Scott Reference Scott1911a), would indicate that he shared his plans with Simpson just as he had with Evans.

In Scott and Amundsen Huntford writes that ‘On January 5^th, Meares finally arrived [at base] with the dogs. He had nothing to do except wait for orders from the south’ (Huntford Reference Huntford1999: 519). In fact, more was expected of Meares than idleness. Scott's orders include the following instructions:

Meares and Gerof arrived back at base on 5 January rather than 19 December as originally planned (Evans Reference Evans1949: 186). Due to the need to allow the dogs sufficient rest, the dog teams could not go out again to restock One Ton. However, Scott had already allowed for this eventuality in his instructions left for Simpson, dated 3 October 1911:

So Simpson knew at this early stage that he was responsible for ensuring that the ‘relief stores’ reached One Ton Depot by any means necessary. Scott's later orders sent back with Day, dated 24 November, informed Simpson that the dogs would definitely not be available:

Scott's detractors argue that Meares’ late return in January 1912 fatally disrupted Scott's plans, but in fact this late return came as no surprise to Simpson. Scott had explicitly prepared him for this possibility. A man-hauling team was accordingly dispatched with supplies, leaving on 26 December and arriving at One Ton on 9 January (Smith, J.M. Reference Smith2010: 38). Unfortunately, when Cherry-Garrard, Gerof and the dogs arrived at One Ton on 4 March the men found no dog food there, which seriously limited their prospects of proceeding further south.

Why was there no dog food at One Ton? There are two possible scenarios. Did Simpson did not carry out Scott's instructions, failing to send out dog food with the man-hauling team, or did the man-hauling team mistakenly unload the dog food at an earlier depot? We do not know. However, Scott himself cannot be blamed for the absence of dog food at One Ton. Had his orders been followed accurately, the relief dog teams would have had sufficient supplies to proceed further south.

The second element in Scott's orders which was apparently disregarded was the projected departure date of the dog teams for the south. Scott's instructions to Meares were as follows:

Meares undoubtedly knew what was required, and Simpson, being ‘fully aware of [Scott's] plans and wishes’ (Scott Reference Scott1911a), would have known this as well. However, Cherry-Garrard states that Atkinson acted on Scott's verbal instructions to ‘come as far as you can’, given before the first returning party's departure on 21 December 1911:

Here, at least, Scott's use of verbal instructions makes sense. After having written out orders to Meares, he probably saw no reason to transcribe the same orders again, and intended these words to Atkinson as a mere reminder. However, Cherry-Garrard's account implies that Atkinson was unaware of any other orders for the dog teams, and based his subsequent decisions solely on Scott's verbal command to ‘come as far as you can’. Did Meares or Simpson neglect to pass on Scott's detailed written orders of October 1911 to Atkinson, leading Atkinson to believe that Scott's vague verbal order was his only guide?

Although Cherry-Garrard apparently believed that Scott's orders had been unspecific and had left Atkinson in confusion, I think he is in error. Atkinson's letter to Wright of 22 February, ‘I want you please to take my team south to meet the last party’ (Atkinson Reference Atkinson1912b), echoes Scott's instructions that the dog teams ‘should aim at meeting the returning party about March 1 in Latitude 82 or 82.30’ (Evans Reference Evans1949: 188). Atkinson also wrote to Simpson that Wright should ‘go south to meet Captain Scott’ (Hooper Reference Hooper2011: 201). The emphasis on the relief teams ‘meeting’ the polar party is evidence that Atkinson knew of Scott's original written orders to Meares.

However, although Atkinson must have been aware of his leader's written instructions, he did not follow Scott's order to leave for the south within ‘the first week of February’ (Evans Reference Evans1949: 187). The dogs, having returned on 5 January, should have recovered their full strength within four or five weeks and been ready for a fresh start in early February; however, Atkinson and Gerof did not leave Cape Evans to head south with the dog teams until 13 February. Moreover, Atkinson did not leave immediately after that date. Cherry-Garrard states that the men left only because the sea-ice between Cape Evans and Hut Point was beginning to break up. The actual journey southwards was scheduled to start from Hut Point ‘in about a week's time’, presumably on or shortly after 20 February (Cherry-Garrard Reference Cherry-Garrard1994: 427).

Some examination of Atkinson's background is important at this point. Though equivalent to a lieutenant in rank, Atkinson's naval title was ‘Surgeon’; he was a medical or scientific officer. This distinction is crucial because a surgeon did not undergo the same training as a naval officer. A medical officer's appointment depended solely on the candidate passing examinations in medicine, science and languages (Navy List 1907: 864–866). The navy expected its medical officers to care for invalids; it did not expect them to hold positions of command, and provided no specific training in this regard.

Atkinson had never served as a full-time cadet in a naval college or on a training ship. As a teenager he had had five years’ membership in the school's officer cadet corps, the Rifle Volunteer Corps at his school, Forest Grammar (Sewell Hawkins Reference Sewell Hawkins2009: 4), but this was an extension of the typical Victorian grammar school or public school experience, and did not come remotely close to the culture of rigid obedience and harsh discipline experienced by a naval cadet. Atkinson later trained as a doctor at St Thomas’ Hospital in London, and after qualifying in 1906 joined the navy two years later. His first position was as a junior doctor in the Royal Naval Hospital at Haslar, Gosport. Afterwards he had a six-month stint as a junior ship's doctor on the HMS Achilles, followed by scientific research at the shore posting of HMS Pembroke (Jones A.G.E. Reference Jones1992: 16). Atkinson's appointment to the Terra Nova expedition in 1910 as shore party surgeon and parasitologist was therefore his first prolonged experience of naval life and of an officer's responsibilities.

Atkinson had no experience preparing him for command, let alone the task of giving life-or-death orders in his leader's absence, yet he would soon be forced to do just that. On 19 February Crean stumbled into Hut Point with the news of Evans’ illness, news which diverted the dog teams from their original intentions to a more immediate rescue.

Atkinson judged Evans's condition so serious that he could not be left without professional care: indeed, Taylor wrote that Evans could not walk and reminded him more of ‘a corpse than a live man’ a week later (Taylor 1916: 421), whilst Gran found Evans barely recognisable (Gran Reference Gran1984: 177). Had Atkinson left for the south, no-one else at base would have had the medical knowledge to deal with Evans’ possibly life threatening illness. As the sole doctor Atkinson would have to remain with Evans, and someone else would have to accompany Gerof on the journey out to meet the polar party. In the next section we shall assess the men available for this task, and the evidence that, as a result of the new circumstances, Atkinson altered Scott's original orders concerning the dog teams, with tragic consequences for the polar party.

Atkinson's ‘verbal instructions’

With Atkinson occupied caring for Evans, only two individuals remained for the task of taking the dog teams south: Meares, the obvious choice to go, had received news of his father's death and so had to return home on Terra Nova (Fiennes Reference Fiennes2004: 340). The first candidate was Wright; the second, Cherry-Garrard. Wright had mild myopia, but was physically fit and had successfully navigated the first returning party over 470 miles back to base (Bull and Wright Reference Bull and Wright1993: 222). By contrast Cherry-Garrard's myopia was more severe than Wright's, and he found navigation extremely difficult. Scott was well aware of the importance of navigation skills, and wrote on 12 June 1911 that ‘every officer who takes part in the Southern Journey’ (Scott Reference Scott2008: 222) should possess some knowledge of navigation, including meridian altitude observations. Cherry-Garrard records in his journal on 15 June that Lieutenant Evans approached him with a meridian altitude and asked whether he found it straightforward. Cherry-Garrard found it anything but straightforward, and struggled on his own to learn navigation before finally approaching Scott, who wrote in a letter of 26 October that:

It is easy enough, in retrospect, to state that Cherry-Garrard should have approached one of the officers and requested a course of instruction. Unfortunately, Cherry-Garrard was extremely self-effacing: he had no wish to impose his needs upon others at the best of times, and on this expedition scientific research occupied the greater part of the men's schedules. So he tackled the problem in private and, for the most part, unaided (he records taking a lesson in the use of the ‘sundial’ from Evans on 20 September) before admitting defeat.

Arguably Cherry-Garrard was too late in informing Scott of the problem. With less than a week to go until the southern journey, Scott would have had no time to arrange a last-minute course of instruction. As a consequence Scott was essentially forced into a breezy dismissal of the potential danger: certainly it was highly unlikely Cherry-Garrard would ever be called upon to be the solitary navigator in any party of men. However, five months later, Cherry-Garrard would find himself in just this situation, and, now that navigation skills were of paramount importance, he found himself irremediably incapable (Wheeler Reference Wheeler2002:132).

Wright was therefore the obvious candidate to replace Atkinson on the southward journey. As well as Atkinson's note to Wright, Gerof also carried a letter from Atkinson to Simpson requesting that Wright be allowed to ‘go south to meet Captain Scott’ (Hooper Reference Hooper2011: 201).

Simpson, however, did not wish Wright to leave. Simpson was to return to India on Terra Nova in March, and had appointed Wright as his successor for head of science until Wilson's return. Simpson wrote in his diary on 23 February that

It cannot be the loss of meteorological data that concerns Simpson here. Such observations are straightforward and would have been within Cherry-Garrard's capabilities. Was it the specific loss of Wright's other data that concerned Simpson, or was it a general fear that if anything happened to Wright the expedition's scientific programme would be impaired? Whatever Simpson's motive, Meredith Hooper notes that his choice of Cherry-Garrard over Wright ‘was made entirely in relation to science’ (Hooper Reference Hooper2011: 203) and Wright wrote in his memoir, ‘Simpson, who had been left by Scott in charge, demanded my body in order to carry on his work during the second winter, leaving the One Ton journey for Cherry and Demetri’ (Bull and Wright Reference Bull and Wright1993: 253). With science as his priority, Simpson assumed that Cherry-Garrard possessed sufficient navigational skill to meet the polar party in Wright's place. Though Simpson sent both Wright and Cherry-Garrard to Hut Point, ostensibly for Atkinson to choose between them, Atkinson knew that Wright would remain at Cape Evans for Simpson's observations.

Cherry-Garrard, however, could not navigate beyond the basic use of a compass. Fortunately, locating One Ton Depot would not require advanced navigational skills. According to Headland (R.K. Headland, personal communication, 15 June 2011), One Ton could be found easily on the southbound course, at a point where the peak of Mount Erebus (an active volcano, often identifiable by a crown of smoke) was just about visible. However, south of One Ton, the terrain soon devolves into a featureless white plain where a thorough knowledge of navigation is crucial. Beyond One Ton, Cherry-Garrard could not easily proceed.

It must have seemed an intractable dilemma for Atkinson. Cherry-Garrard's limitations were obvious, but if Atkinson, the only qualified doctor at base, departed with the dog teams, then Evans might die for want of professional care. On the other hand, if Wright departed for the south, the expedition's scientific results might be compromised.

What had Atkinson planned to do before the Evans crisis arose? His correspondence is enlightening. Atkinson wrote to his parents on 1 February that ‘I may not get home until the ship has left and I come back with the Southern Party’; later, on 11 February, he adds that ‘as I am off in four days for Capt. Scott I must finish this [letter]’ (Atkinson Reference Atkinson1912a). Here we clearly see that Atkinson aimed to meet Scott's party on the southwards route. When the crisis arose, Atkinson wrote to Wright on 22 February, ‘I want you please to take my team south to meet the last party’ (Atkinson Reference Atkinson1912b); there is also a request in Atkinson's letter to Simpson to allow Wright to ‘go south to meet Captain Scott’ (Hooper Reference Hooper2011: 201). There can be no doubt that Atkinson had originally intended to follow Scott's orders to ‘meet’ the polar party on their return, and that, when he had to stay with Evans, his first choice of a substitute had been Wright.

Cherry-Garrard reproduced Atkinson's verbal instructions to him in The worst journey in the world:

There is a discernable shift in the plans at this point. Prior to Cherry-Garrard's being appointed, the plan had unquestionably been for the dog teams to head south in order ‘to meet’ the polar party; now, for the first time, we hear that Scott had given ‘particular instructions’ that the dogs were ‘not to be risked’, as they could be needed later in the season.

The question inevitably arises: if Scott had stated that ‘the dogs were not to be risked’, why had Atkinson not previously taken this into account? In his letters to Wright, Simpson, and his parents, there is no mention of preserving the dogs for later. The plan had clearly been ‘to meet’ the polar party and bring them back to base, irrespective of the distance travelled and the ‘risk’ posed to the animals. Why, at this late stage, does Atkinson suddenly mention orders from Scott not to risk the dogs?

A difficulty arises when we look for corroboration that Scott ordered that the dogs should not be risked. Scott, a man who habitually committed his plans to paper, seems not to have written this down. There is no record of not risking the dogs in the instructions cited in Evans’ South with Scott; no such statement exists in Scott's journals or in Scott's outlined plans for the expedition in the Scott Polar Research Institute archive. Furthermore, such an order would obviously have contradicted Scott's plans. Scott's written instructions to Meares of October 1911 stated that, when it came to meeting the polar party, the dogs were indeed to be ‘risked’ as far as 82° or 82° 30′ S (which meant a journey of around 300 miles from Hut Point). Why should Scott have been concerned for the welfare of the dogs beyond the current ‘season’? The southern journey was the only event which explicitly required the use of the dog teams, since it involved the issue of time expediency, getting the polar party swiftly back to base to avoid a fall in temperatures and in the hope of catching Terra Nova before her departure. For all other sledging trips dogs were convenient, but for the southern journey they were crucial.

Moreover, Scott knew that 14 fresh dogs would be arriving with Terra Nova in 1912 (Bull and Wright Reference Bull and Wright1993: 249). If the previous dogs had all been lost, these would theoretically have sufficed for the greatly reduced needs of the 1912 sledging season. Why, again, should Scott have stressed to Atkinson that it was necessary to preserve the life of every single dog? Had the polar party returned having failed to reach the pole, a second British attempt in 1912–1913 would have been out of the question given both the reduced number of men remaining for the second year and the serious threat posed by the Norwegians. There was no reason for Scott to place the long term preservation of the dogs ahead of the immediate preservation of the polar party.

There appears to be only one source for Scott's statement that ‘the dogs were not to be risked’, and that is Atkinson himself. Cherry-Garrard's account indicates that Scott told Atkinson this as part of a verbal exchange on the Beardmore Glacier (Cherry-Garrard Reference Cherry-Garrard1994: 424) but, as seen previously with Huntford's mythical ‘verbal instructions to Evans’, this makes no sense. If Scott had decided to preserve the dogs for the following season, which would seemingly destroy his previous plans for the dogs to come south to meet the polar party, then why did Scott not spend five minutes to write these new orders down, to ensure that there was no confusion back at base? Furthermore, can we believe that Scott told Atkinson to ‘come as far as you can’, yet added the proviso that the dogs should ‘not be risked’, during the same period of time on the Beardmore Glacier? Scott would surely have been aware of the potential contradiction in these two vague statements. Judging from the extremely thorough orders cited in South with Scott it seems fair to state that, had Scott changed his mind here, he would have written detailed instructions for Atkinson's guidance. Yet Atkinson brought no written orders back to base.

Most importantly, if Scott had given Atkinson verbal orders, at any point in time, that the dogs ‘were not to be risked’, why did Atkinson only remember this caveat when Cherry-Garrard was appointed to take over from Wright? The timing is crucial: Wright and Simpson were both told that the aim was to ‘meet the polar party’, and only when Cherry-Garrard was assigned to the task did Atkinson suddenly switch the focus from ‘meeting the polar party’ to the importance of preserving the dogs for the following season. Furthermore, Atkinson kept these instructions to Cherry-Garrard purely oral, even though a novice would have gained much needed reassurance from explicit orders in writing. Did Atkinson wish to avoid being held accountable for these orders at a later date?

The evidence and the timing point to one conclusion: that Atkinson independently decided to alter Scott's orders at the exact moment when Cherry-Garrard was appointed as leader of the dog teams. The original mission of heading south to ‘meet the polar party’ as far as 300 miles out was now altered to the unloading of supplies at One Ton Depot, only 119 miles out. The task was silently, and fatally, downgraded to fit the abilities of the man chosen for it.

Why should Atkinson misrepresent Scott's intentions in this way? My hypothesis is that Atkinson wished to protect Cherry-Garrard as far as possible. From Hut Point, Cherry-Garrard would have been able to reach One Ton safely, but his limited navigational abilities would have led to serious difficulties on the Ice Barrier itself. He had to be prevented from a quixotic attempt to head out onto the featureless plain in search of the polar party for, without the restriction that ‘the dogs were not to be risked’, Cherry-Garrard certainly would have been tempted to try. Two members of the polar party were of particular importance to Cherry-Garrard: Wilson had been an admired mentor, whilst the fun-loving and dependable Bowers had been his closest companion. With no orders to restrain him, one can picture the young and idealistic Cherry-Garrard striking out across the Barrier in search of his friends, and risking both his own life and that of the dog-handler Gerof in the process. Atkinson had to prevent this, and stating that ‘Scott’ had expressly forbidden such a move would have been the easiest way of reining Cherry-Garrard in.

Was Atkinson a fool, or a villain, to have altered Scott's orders? Neither, but a good man pressured by conflicting responsibilities. In Scott and Amundsen Huntford states that ‘Atkinson, the man in command, although a doctor, was a Naval officer. In a situation calling for judgement and initiative, Naval discipline failed’ (Huntford Reference Huntford1999: 520). This is an error: Atkinson was not a professional naval officer but a medical officer, a distinction crucial in this context. A naval officer, from a formative experience of harsh naval discipline, would obey his superior's commands without question. A medical officer, used to autonomy (and coming from a far less authoritarian civilian background) would expect greater freedom of interpretation.

We have seen that Scott's written orders had been for the relief dog teams to leave for the south in ‘the first week of February’ (Evans Reference Evans1949: 187), but Atkinson took the decision not to leave for the south until at least 20 February. Why the delay? In his 1913 report Atkinson states that he was ‘kept at Hut Point by bad weather’ (Atkinson Reference Atkinson and Scott1913: 299), but Simpson's meteorological reports show that it was not until 16 February that the wind velocity rose significantly to 30–38mph and continued at this rate or higher for the next six days. Previous to 16 February there were periods of relatively calm weather when Atkinson could have made a start for the south (Simpson Reference Simpson1923: 66–67). Atkinson does not cite the dogs’ health as a cause for delay, but states that on 9 February ‘we started landing stores from the ship, and in this all hands were employed’ (Atkinson Reference Atkinson and Scott1913: 298). The meteorological report dates the landing of stores as 7 February (Simpson Reference Simpson1923: 317); as Atkinson placed priority on assisting at the unloading of Terra Nova over leaving for the south, he must have believed he had discretion to interpret Scott's orders as he saw fit.

Moreover, as a medical officer faced with Evans’ life threatening illness, Atkinson was bound by his Hippocratic Oath. His duty to a sick man unquestionably had to override Scott's orders. Simpson's objections must also have weighed heavily with Atkinson: it would not be easy to bear responsibility for failing to complete important scientific studies. In addition, Atkinson had to ensure that, if Cherry-Garrard were sent south in place of Wright, the young man's life would not be endangered by his mission. Juggling his responsibilities to the four various parties, Scott, Evans, Simpson and Cherry-Garrard, Atkinson made the fateful decision that the polar party was the unit least in need of immediate attention, and that the re-stocking of One Ton Depot by Cherry-Garrard would be sufficient to assist them. By making this small compromise, Atkinson hoped to satisfy all parties. Only in retrospect is it clear that the compromise was a dreadful mistake. We can see that Huntford's verdict, ‘in a situation calling for judgement and initiative, Naval discipline failed’ (Huntford Reference Huntford1999: 520), is a misreading of the situation. The evidence suggests that Atkinson did use his ‘judgement and initiative’, and that it was exactly this that led to a fatally wrong decision being taken.

Would Atkinson have acted differently, had Scott made it clear that the dogs might be urgently needed? Perhaps, but this still does not excuse Atkinson entirely, as Scott should not have needed to explain. Scott gave his order as a naval captain to his subordinate, and expected it to be obeyed without question. With hindsight, Atkinson certainly should have placed priority on Scott's instructions to ‘come as far as you can’ and ‘meet the polar party’. Had Atkinson acted with blinkered naval obedience to his captain's wishes, disregarding all other concerns, the remaining members of the polar party might have been rescued in time.

However, Atkinson could not know of Scott's evolving plight, and he made his decision. At 2am on 26 February 1912, Cherry-Garrard obeyed orders and left with Gerof for the south. He had been told that there was no necessity to go further, but in the next section we shall examine the evidence which suggests that Scott and the other members of the polar party expected to be met by the dog teams.

‘The dogs which would have been our salvation. . .’

On 4 March Cherry-Garrard arrived at One Ton, and the nightmarish conditions he experienced there are detailed in his book. The blizzards, the absence of the dog rations necessary to proceed further, and Gerof's sudden incapacitation (Cherry-Garrard Reference Cherry-Garrard1994: 434) came close to overwhelming him. There was no way that he could have gone further. Utterly isolated and painfully inexperienced, he did what anyone else in his situation would have done; after six days of waiting, he left One Ton to bring both himself and the apparently paralysed Gerof safely home. He departed believing that the polar party was certain to return safely. Less than a month later, he would write bleakly in his diary: ‘We have got to face it now. The Pole Party will not in all probability ever get back’ (Cherry-Garrard Reference Cherry-Garrard1994: 441).

In November 1912, a party with dogs and mules journeyed out onto the Barrier to search for traces of the polar party, and Wright spotted the dark apex of Scott's tent jutting from a pyramid of snow. They found it, and the three bodies it contained, eleven or twelve miles south of One Ton Depot.

Atkinson was the first to open the tent; what he found within was a sight so disturbing that Cherry-Garrard would not reveal it in his official account of the expedition, but only in a verbal account decades afterwards. During the filming of Scott of the Antarctic (1948) he told the playwright Mary Hayley Bell that Scott's body was found ‘sitting up facing the opening of the tent. His coat had been pulled aside at the neck, and his blue eyes were wide open; on his cheeks were two frozen tears’ (Bell Reference Bell1968: 184).

In the Tasmanian newspaper The Mercury on 21 April 1913, an anonymous source close to the expedition related that ‘Captain Scott was found sitting in his sleeping-bag with his eyes open’ (The Mercury, Hobart, Tasmania, 21 April 1913: 2). The conclusive corroboration, however, lies in a journal entry by Hooper, one of the members of the search party: ‘Capt. Scott must have been awake when he passed away; his eyes were wide open & he looked as though he must have been in great pain’ (Smith, J.M. 2010: 53). The detail of Scott's open eyes was omitted from the other expedition diaries and official accounts, probably on grounds of taste, but Hooper's written record and Cherry-Garrard's later verbal admission testify to an uncomfortable truth.

Scott's journals revealed that on 10 March, the day Cherry-Garrard turned back for base, the polar party had reached Mount Hooper Depot, 61 miles due south of One Ton. An estimate of the speed of a dog team over the ice is roughly 20 to 25 miles per day. Had the teams which arrived at One Ton Depot on 4 March 1912 been possessed of a leader both capable of advanced navigation and willing to slaughter dogs to provide food for the others, this leader could have started a southern journey on 10 March after the blizzards had passed. Had he done so, he and Gerof could conceivably have met Captain Scott's party around 13 March. It should be remembered that Captain Oates did not sacrifice himself until 15 or 16 March. Seen in this light, the polar party's salvation seems tantalisingly close, a matter of a few days’ journey by a practised navigator.

Did Scott and his men expect to be met by dog teams on the journey back to base? There is evidence to suggest this. On 30 November 1911 Scott wrote that ‘[t]he dogs are reported as doing very well. They are going to be a great standby, no doubt’ (Scott Reference Scott2008: 334). What could Scott mean by ‘standby’, if not that he expected the dogs to provide back up in case of emergency? On 27 February 1912, during the return journey, Scott wrote that ‘we are naturally always discussing possibility of meeting dogs, where and when, &c.’ (Scott Reference Scott2008: 403). Teddy Evans reported that in their final farewell Oates ‘asked me to send him out tobacco and sweets by the dog-teams’ (The Pittsburgh Press, Pittsburgh Pennsylvania, 19 June 1914: 29).

One possible argument for the polar party's faith in the dogs’ arrival is Oates’ decision to keep marching despite the revelation, on 2 March 1912, that his feet were badly frostbitten. One of Oates’ biographers, Michael Smith, argues that if Oates had truly intended his sacrifice ‘he would not have lingered until the entire party was also at death's door. If there was a ‘correct’ moment for sacrifice, surely it would have been somewhat earlier’ (Smith M. 2002: 231). However, Oates had previously stated in public that he was entirely willing to sacrifice himself rather than risk his companions’ safety (Ponting Reference Ponting1922: 288) just as Bowers had previously discussed self-sacrifice with his friend Cherry-Garrard (Cherry-Garrard Reference Cherry-Garrard1994: 537). Why, then, did Oates wait? If the polar party were expecting the dogs, then Oates’ ‘lingering’ would certainly make sense.

The polar party could not have known at this point that no-one was coming to save them. Scott's written orders to Meares had specified that the polar party be met on 1 March at ‘Latitude 82 or 82.30’ (the latter point being approximately 300 miles south of Hut Point); it was now 2 March, and they had just passed Mid-Barrier Depot at 81° 35’ S (245 miles south of Hut Point). The dogs’ arrival must still have seemed possible, and hence I believe that a logical and humane gamble was taken to keep Oates in the polar party, trading slower progress for the chance of Oates’ eventual rescue by the dog teams.

On 9 March the party arrived at Mount Hooper Depot, to find that it had not been re-stocked. Scott wrote, ‘The dogs which would have been our salvation have evidently failed’ (Scott Reference Scott2008: 408). It is at this point that Oates directly asked Wilson whether he, Oates, ‘ha[d] a chance’ (Scott Reference Scott2008: 408). Scott privately doubted it, and wrote ‘With great care we might have a dog's chance, but no more’ (Scott Reference Scott2008: 408). In context, the choice of that particular phrase is rather telling. It would appear that Scott did not entirely give up hope of the dogs’ arrival: ‘Nothing could be said but to urge [Oates] to march as long as he could’ (Scott Reference Scott2008: 409). I believe that Oates’ departure into the blizzard a week afterwards was, at least in part, his acknowledgement that the long-anticipated dog teams would not be coming.

It can justly be asked why, if Scott fully expected the dogs to arrive, he did not make his frustration with their failure apparent in his journals. After all, he had not refrained from noting others’ faults back at base; why did he not make his displeasure clear? Indeed, Scott's criticism at this point is sparse and unspecific (‘generosity and thoughtfulness have not been abundant’ and ‘It is a miserable jumble’ (Scott Reference Scott2008: 471)).

Firstly, it should be remembered that Scott’s journal entries did not always reflect his emotional state. We see this with his response to the news that Amundsen was in the Bay of Whales. Cherry-Garrard's eyewitness account portrays Scott as openly angry, even suggesting that the British ‘fight’ the Norwegians (Wheeler Reference Wheeler2002: 95) yet Scott's journal entry for 22 February 1911 is restrained, even philosophical, on the Norwegian threat (Scott Reference Scott2008: 135, 460). If Scott does not express certain strong emotions in his journal, this does not necessarily indicate the absence of such emotions.

Furthermore, whilst it is impossible to take any discussion of this point beyond conjecture, Scott's reticence in attacking the dog teams for their failure to arrive could be explained in part by his extreme situation. Criticism of others comes easily in a position of comfort and security; by now, Scott was far from feeling comfortable or secure. He was facing the real possibility of death, and his one hope of survival would lie in the selfless efforts of others on his behalf. Could anyone reasonably expect Scott to criticise, in writing, the very dog teams that he hoped were coming to save him?

Scott's entries refer to his expectations only obliquely, as if hesitant to tempt fate. In writing he lowers his expectations of the dog teams to their providing minimal assistance only (‘Providence to our aid! We can expect little from man now except the possibility of extra food at the next depot’ (Scott Reference Scott2008: 406)). Scott seems careful not to expect too much of ‘man’, but the phrase ‘the dogs which would have been our salvation have evidently failed’ (Scott Reference Scott2008: 408) reflects the extent of his hope in them. There is also one last telling phrase in Scott's penultimate entry of 29 March 1912, ‘I do not think we can hope for any better things now’ (Scott Reference Scott2008: 412). By this point Scott had spent perhaps as long as a week with little food and no fuel. He could not travel due to frostbite, and his companions Wilson and Bowers were too greatly weakened to make the 22 mile round trip to One Ton for further supplies. In such a context, what else could Scott possibly have meant by the euphemistic phrase ‘any better things’, if not the arrival of the dog teams?

There is a further possible reason why Scott did not make explicit reference to the dog teams, and that is the issue of his legacy. He did not know what was transpiring back at base: at worst, the dog teams sent to rescue him could have perished during the relief journey. Scott would hardly have wished to leave behind a journal which laid stress upon the dogs’ failure to arrive, giving an impression of ingratitude and criticising men who, for all he knew, had died in the attempt to save him.

Finally, perhaps Scott suspected that he had made a vital error at this point. Who would not be prey to self-doubt in such circumstances? In his farewell letter to Sir Edgar Speyer he writes,

The phrase ‘overshot our mark’ could in isolation be taken as a metaphor for failure, but the supplementary phrase ‘it's a pity to have missed it’ argues that Scott is referring to a specific physical location that he has ‘missed’ or ‘overshot’. This ‘mark’ cannot be One Ton Depot: on this letter Scott gives his location as ‘Lat 79.5’ (or 79° 50′), so he knew he was still some distance short of One Ton (at 79° 29′). I would suggest that this ‘mark’ refers to the point at which Scott could have expected to meet the dog teams. Scott feared that he had deviated from the true route north, and that the expected ‘support’ had consequently passed them by. If this is what Scott meant, his fear was groundless: he had navigated accurately, and was well within sight of where he was supposed to be. The location of his final tent bears this out.

There has been speculation that Scott actively sought death at the very last (Huntford Reference Huntford1999: 524–526; Spufford Reference Spufford2003: 336). Huntford has even recently said of Scott's death, ‘it was a kind of suicide, lying down in the tent’ (New Scientist, London UK, 4 October 2011). Hooper's and Cherry-Garrard's eyewitness testimony suggests the opposite, that, at the moment of death, Scott may have fought to stay awake. Though of course we will never know for sure, it is possible that Scott died hoping to the last that his rescuers would arrive. The appearance of Scott's frozen body, his open eyes, his sitting position facing the entrance to the tent, may well have given that impression to Atkinson, Cherry-Garrard and the others who found him in November 1912.

The aftermath

Cherry-Garrard returned to Britain a broken man. Despite the numerous obstacles in his path, any of which would have legitimately deterred someone of his limited experience from proceeding further, the polar party's loss would gnaw at him for the rest of his life. Certain newspapers made him a target, and in April 1913 he sent a poignant private letter to Atkinson asking him to clarify the ‘dog journey’ to the media:

With media interest so keen, the survivors needed to frame the sequence of events in the best possible light. The first major opportunity came with the publication of Scott's journals, under the title of Scott's last expedition, in 1913. For the second volume Atkinson wrote a report, ‘The last year at Cape Evans’, which described Cherry-Garrard's and Gerof's journey to One Ton. In it, the lack of dog food at the depot is mentioned as a fact rather than an oversight: ‘As there was no dog food in any of the depots except Corner Camp or along any of the route, it meant that 24 days was the limit of [the dogs’] usefulness’ (Atkinson Reference Atkinson and Scott1913: 300).

This would give anyone the impression that such limited use of the dogs had been Scott's plan all along, and that there had never been any intention to proceed beyond One Ton. There is no mention of the fact that Scott had ordered Meares to bring ‘as much dog-food as you can conveniently carry (for third journey)’ to One Ton (Evans Reference Evans1949: 187). This omission, together with Atkinson's statement that ‘strict injunctions had been given by Captain Scott that the dogs should not be risked in any way’ (Atkinson Reference Atkinson and Scott1913: 304), makes Scott appear to have lacked the foresight to prepare the depot for a possible advance to the south. Regrettable and erroneous as this impression was, Atkinson's primary concern was to protect the living, specifically, the young man whom he had sent on a painfully futile quest. Atkinson concluded:

Cherry-Garrard had indeed followed his instructions to the letter, but they were Atkinson's, not Scott's.

Meanwhile Evans, by now promoted to Commander, seized control of public relations with a force that struck many as overbearing. In 1912 Evans had worked to frame the awkward issue of his scurvy favourably to the newspapers; now he increased his media profile, stepping into the spotlight on every available occasion. With his extensive lecture tour he became the public figurehead of the expedition in the absence of Scott.

Privately Cherry-Garrard, Atkinson and Ponting would unite in condemnation of Evans’ actions, interpreting them as blatant self-promotion. Whilst there was certainly an element of this, I would argue that Evans’ efforts to push himself forward were most probably his form of self-defence. He was still suffering from the sudden death of his wife from peritonitis (Pound Reference Pound1963: 125). Now, in the awareness both of his previous possible misjudgement (regarding the seal meat) and that his relations with Scott had been less than ideal, Evans needed to make himself impervious to attack by getting the media firmly on his side.

As well as protecting his own reputation, Evans made efforts to present the men at Cape Evans favourably. In an interview of 15 February 1913, Evans declared that ‘Captain Scott left instructions that no search parties should leave the base to look for him’ and that ‘it was humanly impossible for the base party to save Scott and his comrades’ (Fiennes Reference Fiennes2004: 389). Of course, the first of these assertions was palpably untrue: Scott left no orders forbidding search parties. This story was designed to deflect criticism from survivors, and, sadly, misrepresented Scott as taking no heed of contingency plans.

As for Cherry-Garrard, he would bear a grudge against Evans for the rest of his life. The two men had never been on friendly terms; Cherry-Garrard noted that ‘from the first I had never liked Evans’ (Wheeler Reference Wheeler2002: 79), and in a letter he called Evans ‘the one big failure in [the expedition]’ (Jones M. Reference Jones2003: 128).

Cherry-Garrard's hostility towards Evans is understandable, given that Evans’ illness, which must have appeared self-induced to some degree, prevented Atkinson from proceeding south as planned. However, in fairness, Evans’ illness was only one of the elements that led to the crisis: in the end it was Simpson's reluctance to send Wright, and Atkinson's endorsement of this, which forced Cherry-Garrard into going south. Even Wright knew of the injustice, and said of Cherry-Garrard's situation in later years, ‘It always worries me. I should have had that job’ (Raeside Reference Raeside2009: 246). However, Cherry-Garrard could not bring himself to criticise Simpson or Atkinson. He counted both men as close friends: he met and corresponded with Atkinson until the latter's death, and in 1939 he appointed Simpson as an executor of his will (Wheeler Reference Wheeler2002: 260). Ultimately it was easier for Cherry-Garrard to heap all the blame on the distant and long-detested Evans than to face the possibility that two men whom he regarded as friends had, however unintentionally, betrayed him.

Unfortunately for Cherry-Garrard, Evans soon managed to place himself beyond all criticism. In 1917, whilst in command of the destroyer HMS Broke, Evans and his colleague Commander Peck (of HMS Swift) routed six German destroyers in their attempt to shell the port of Dover. During the action, Evans ordered the Broke to ram one of the German ships side-on. The skirmish resulted in two German destroyers out of action, 140 Germans taken prisoner and no further enemy attacks upon Dover.

The media were overjoyed by this feel good story of British victory: Evans was popularly dubbed ‘Evans of the Broke’, an unofficial title he bore for the rest of his life. With public opinion firmly on his side, Evans made it clear that he could and would challenge any hint of criticism in Cherry-Garrard's proposed ‘official’ account of the expedition. In April 1919 Atkinson wrote privately to Cherry-Garrard that ‘TE [Teddy Evans] is just itching at present for trouble. If you have anything in the least disparaging about him he will be out for trouble of the worst sort’ (Atkinson Reference Atkinson1919a).

Much as Cherry-Garrard might have wished to make his feelings known, he could never have launched an attack on a national hero. So he backed down, replying to Atkinson ‘Teddy Evans is probably suffering from too many medals. A friend tells me that the greatest mistake in the expedition was that God killed the wrong Evans. He comes out of my book far better than I desire or he deserves’ (Cherry-Garrard Reference Cherry-Garrard1919).

In the event Cherry-Garrard limited himself to alluding indirectly to the possible cause of Evans’ illness and describing Evans at one point as ‘rotting of scurvy’ (Cherry-Garrard Reference Cherry-Garrard1994: 596). It is unlikely that his target was overly concerned. By the time The worst journey in the world was published in 1922 Evans was riding a wave of acclaim. The previous year, as captain of HMS Carlisle, he had helped rescue Chinese shipwreck victims from SS Hong Moh. When survivors refused to leave the wreck due to the rocks and strong currents, Evans dived into the rough sea and swam to the wreck with a rope around his waist. There he sent back two semi-conscious men with a lifebuoy on the rope, then swam back to demonstrate a safe route to the waiting motor-launch. Encouraged, survivors jumped, then swam or used life-lines to get to safety. The Carlisle's rescue work saved 221 people in total (Gawler Reference Gawler1995: 41–47).

Evans was ideal for this crisis: not only was he a strong swimmer and undeterred by personal risk, but, heedless of petty rules as usual, he ignored the unwritten decree that a Royal Naval captain should not leave his ship whilst at sea under any circumstances. For his actions on the Hong Moh Evans was awarded a specially struck Gold Medal for life-saving from Lloyds of London. After World War II he was made a hereditary Labour peer, Lord Mountevans of Chelsea, and died in 1957 at the age of 76.

One example of outstanding bravery was none other than Atkinson, then Surgeon Lieutenant-Commander Atkinson, on the ill-fated HMS Glatton in World War I. In September 1918, whilst docked in Dover harbour, the ship suffered a fire below decks which resulted in exploding munitions: Atkinson responded by heading directly for the danger and carrying three unconscious men up the ladders to safety. When an explosion blinded him he still continued, finding two more unconscious men in the shelter deck by touch alone and dragging them clear. He later recovered the sight in one eye (the other was removed) and was awarded the Albert Medal for life saving (Smith, G., Reference Smith2011), but his ordeal probably contributed to his premature death in 1929 at the age of 47.

What spurred Atkinson on to save further lives even after having been blinded? Perhaps it might seem somewhat romantic, or excessively cynical, to see the post-expedition bravery of both Atkinson and Evans as that of men aware of a debt that needed repaying. At the very least these incidents should provide ample testament to their characters, and it should be understood that this article was not written to ‘pin the blame’ for the expedition's tragedy on either man. If their errors in judgement had tragic consequences far in excess of their deserving, at least Evans’ and Atkinson's moral sense and compassion for others cannot, and should not, be questioned.

It is, however, surprising that the conclusions in this article were not reached earlier. I would argue that this delay in comprehension is in some measure due to Huntford's error, back in 1979, of mistaking Scott's early written instructions for a last minute verbal order. This turned Scott into a man who saw the dogs’ potential for rescue only at the very last minute, and who trusted a matter of life and death to verbal orders alone.

This is Huntford's précis of Scott's policies regarding the dogs:

This statement is demonstrably untrue, and Huntford's error in this regard has done a great deal of harm to the historical legacy of the expedition. By presenting a theory of Scott's ‘last-minute change of mind’ as a matter of cold hard fact, Huntford mistakenly smoothed over a discrepancy in the evidence which, when viewed properly, really should have rung alarm bells in researchers’ minds.

Atkinson stated in 1913 that Scott had ordered that the dogs should not be ‘risked in any way’, and that Cherry-Garrard had obeyed Scott in going no further than One Ton, yet in Scott's journal there are indications that he and the other men fully expected the dog teams to travel beyond One Ton. ‘The dogs which would have been our salvation have evidently failed’ (Scott Reference Scott2008: 408) is a clear statement that the dogs should have come at least as far as Mount Hooper Depot by 9 March 1912. So how can these two accounts, Atkinson's and Scott's, be reconciled? Huntford's scenario of Scott's last minute verbal ‘change of orders’ on the polar plateau in January 1912 tied up these inconvenient loose ends very neatly indeed, but unfortunately Huntford's scenario was pure invention based on an error. It has led a number of polar historians down a regrettable false trail.

Huntford's error has been a slur on Scott's reputation since the publication of Scott and Amundsen over 30 years ago, and for the sake of truth it cannot be allowed to persist in future polar scholarship. Once and for all, Scott gave one set of written orders for the dogs’ journey south to meet the polar party; these were handed to Meares on 20 October 1911, before the start of the southern journey. His verbal statement to Atkinson in December 1911 to ‘come as far as you can’ was not a change of plan, but a simple re-confirmation of what had been earlier written to Meares.

This error, the misattribution of a last minute vital ‘verbal message’ to Scott, has allowed Huntford to state with confidence that Scott's confused orders concerning the dogs ultimately led to the polar party's failure to be rescued. However, as we have seen, the primary evidence shows that Scott left written orders both for re-stocking One Ton Depot with dog food and for the polar party's eventual relief by the dog teams, orders that were not carried out. Furthermore, Atkinson's claim that Scott gave subsequent verbal orders ‘not to risk the dogs’ is contradicted by Atkinson's own written intentions to ‘meet the polar party’ and ‘come back with Captain Scott’, statements committed to paper before the appointment of Cherry-Garrard necessitated a hasty switch of direction and change of story. It is simply not credible that Scott wanted anything other than for the dog teams to venture south and meet the polar party.

So let us put aside Scott's mistakes in planning. Let us put aside also the freak weather conditions, and the unexpected injuries which slowed his team's progress. All of this is immaterial to the fact that on 10 March 1912 Scott and his three remaining companions were still alive and only 61 miles away from One Ton Depot. Scott had left written orders that he should be relieved by this point, and with a practised navigator in command of the dog teams they could have been rescued in time, just as Evans and Lashly had earlier been rescued. The greatest tragedy of the expedition is the fact that Scott's written orders for relief were not carried out by the men back at base.

With hindsight, it is arguably a pity that Scott placed such trust in the scientific staff back at Cape Evans. Leaving the base in the hands of a scientist (Simpson) and a medical officer (Atkinson) meant that, in an expedition with dual objectives, the scientific element of the expedition would take priority. Had a naval trained officer such as Evans or Bowers been in charge he would have ensured, through explicit instructions and supervision of the loading of the sledge, that the dog food for the relief journey specified in Scott's orders would have reached One Ton Depot safely, and, faced with a hypothetical choice between losing a run of scientific data and leaving five men out on the Barrier without relief, Evans or Bowers would doubtless have made the journey out to meet the polar party the highest priority.

This article has outlined the extent to which Scott's orders were fatally compromised by the men at Cape Evans, and how disregard for their leader's instructions led to the eventual tragedy. Perhaps this was a consequence of the divided nature of the expedition: there is always a danger in any project having disparate goals, and one person cannot comfortably ‘serve two masters’. In the crucial moment when Atkinson had to choose between sending Wright or Cherry-Garrard south, there was a struggle for ascendancy in the expedition between the ‘pole-seekers’ and the ‘scientists’. The claim for science ultimately won through. Though the scientists of the expedition undoubtedly did fine and even ground breaking research during their time in Antarctica, I would argue that their victory here came at the risk, and eventual loss, of four lives.