Thursday, October 2, 2008

Contact details for Source of the Oxus blog

To contact authors use email

Friday, July 18, 2008



Written by
Bill Colegrave
Dillon Coleman

JULY 2007


Note on Names

I. Introduction

A. Objectives
B. Expedition Members
C. Summary of the Expedition Agenda

II. Background

A. Regional Geographic Perspective
B. Overview of Oxus River Exploration

III. Logistics & Preparation

A. Access to the Wakhan
B. Timing
C. Visas
D. Maps and Local Information
E. Transport & Local Support

IV. The Journey

A. Kirghizstan
B. Tajikistan and Lake Syr Kol
C. Into Afghanistan and the Wakhan
D. Wild Thyme and Precipices
E. The Ice Caves
F. Lake Chaqmak
G. Return to Ishkashim
I. The Oxus Northward

V. Afterword

Appendix 1 - Research & Planning
Appendix 2 - Wakhan Expeditionary Team


The name Oxus will be used for the river throughout this report other than when specific contributory streams need to be identified. This is the name by which generally the great river is known historically. The Dari name Amu Darya (river) is in use in Dari/Farsi speaking areas, but so also is the Tajik name, Panj (five). So Oxus seems a reasonable and well understood compromise. Part of the river seems to still have this name or a derivation of it in Tajikistan. Oxus may well be a derivation of the Kirghiz words Aksu, meaning white water, which would be wholly appropriate.


For centuries, the Oxus River has played an important role in Central Asian history and geography and has inspired many an explorer, traveller and writer. Our expedition, to the Pamir Mountains and the Wakhan Corridor where the Oxus flows, followed similar routes to those travelled by 19th century explorers such as Younghusband, Wood, Curzon, Dunmore and Littledale. It also used almost identical means of transport and supply (horse, donkey and foot) as the access to the area is unchanged in the last 200 years. If anything, the region is now less accessible for political reasons, but in all other respects the Wakhan remains completely as it was in, for example, 1895 and very little has changed in most of the Pamir.

A. Objectives

Beyond the basic, yet powerful, need just to visit the Oxus, this famous and romantic river, the expedition was planned around a number of more specific purposes:

• reach and photograph the ice caves at the far eastern end of the Wakhan Corridor in northeast Afghanistan, discovered by Curzon in 1894, which have since been regarded as the primary source of the Oxus River;

• visit the other sources and tributaries of the Oxus, including the lakes Syr Kol and Chaqmak and the Pamir, Little Pamir and Bartang Rivers.

B. Expedition Members

The three of us --- with different nationalities, professional backgrounds, temperaments, interests and skills --- complemented one another superbly and this allowed us to achieve all the mission’s objectives. The team consisted of:

Bill Colegrave, 59, FRGS, English – Financier and publisher, frequent traveller to Asia and other adventurous destinations. From the beginning, the expedition was, for the most part, Bill’s inspiration.

Antony Kitchin, 56, Irish -- Farmer and property investor. Amateur geologist. Always ready for a new adventure.

Dillon Coleman, 56, FRGS, American -- economic development advisor in emerging markets; recently completed 19 months in Afghanistan where he was inspired to visit the Wakhan and the Oxus.

For three individuals who hardly knew one another before setting off, we proved to be most agreeable travelling companions and formed a real bond through our mutually shared experiences.

C. Summary of the Expedition Agenda

The expedition lasted a total of 21 days beginning with our departure from London and ending with our return back to the UK. This included 14 days in the Wakhan, 11 of those days trekking on foot and horseback to reach various destinations at the eastern end of the Wakhan.

Chapter IV of this report provides a detailed account of the expedition’s activities. However, in broad terms, the itinerary was as follows:

Table 1 - Expedition Agenda
June 26 • Flight from London to Bishkek via Moscow on Aeroflot
Kirghizstan & Tajikistan
June 27-29 • arrival in Bishkek, Kirghizstan
• drive to Osh, then onward to Murghab and Langar in Tajikistan, including a visit to Syr Kol on the Tajik-Afghan border
June 30 – July 1 • cross into Afghanistan at Ishkashim
• enter the Wakhan Corridor and drive to Sarhad
July 2 – 12 • travel by foot and horse to the eastern end of the Wakhan, including visits to the ice caves and Lake Chaqmatin
• return to Sarhad
July 13 • return drive to Ishkashim from Sarhad
July 14 – 15 • drive from Ishkashim to Khorog, Tajikistan and onward to Dushanbe
July 16 • flight from Dushanbe to London via Moscow on Tajik Air and Transaero Airlines

[Central Asia showing the Wakhan Corridor, the tongue of Afghanistan that divides Pakistan from Tajikistan and China. The Southern boundary of the Wakhan corridor is in effect the watershed between the Hindu Kush and the Pamir and at the Eastern end between the Hindu Kush, The Karakorum and the Pamir.]


A. Regional Geographic Perspective

Our journey began in northern Kirghizstan and continued southward through that country, crossing eastern and later western Tajikistan, and including the Wakhan Corridor in Afghanistan. Throughout this entire region the geography is primarily mountainous, but it was the rivers and lakes linked to the Oxus that most attracted our attention.

Map 1 on the following page shows the entire region traversed during our journey.

Mountain Ranges

The Pamirs are the large mountain system at the core of the region we traversed. This system consists of a series of mountain ranges, beginning with the Trans Alai Range in the north on the Kirghiz-Tajik border and ending in the south on the Afghan-Pakistan border, thus including the

[Map 1 - Pamir Mountain Region]

entirety of the Wakhan Corridor in Afghanistan, our primary destination. Most of this mountainous area lies in the Gorno-Badakhshan Autonomous Region (GBAO) of Tajikistan. Generally speaking, the component ranges of the Pamirs run parallel to one another in an east-west direction. The exception is the Sarykol Range which follows a more or less north-south line along the Tajik border with China.

The Pamir region, approximating a square of 150 miles on each side, has an average altitude at 13,000 –15,000 feet, higher even than the Tibetan plateau. Due to this height, this mountainous area has historically been known to as the “roof of the world”. It is also referred to as the “Pamir knot” because it is the hub from which the five highest mountain ranges in Asia, indeed in the world, radiate:

• The Hindu Kush, the real geographical frontier of the Indian sub-continent, running to the southwest of the Pamirs;

• The Karakorums, a mighty range of brown and black mountains that includes five of the world’s highest 15 peaks, to the southeast;

• The western extremity of the Himalayas, also to the southeast;

• The Kun Lun Range which guards the deserts of Xingjian and Chinese Turkestan, to the east; and

• The Tien Shan Range, the “celestial mountains separating China from the rest of Central Asia, to the northeast.

Tucked in between a number of the high Pamir Mountain ranges, another of the region’s unique geographical features can be found --- wide, high altitude mountain valleys, each of which is called a “pamir”.

Usually such valleys are V-shaped created by river erosion deep between two mountain ranges; instead a “pamir” is a flat valley located at high altitude between mountain ranges as a result of retreating glaciers in past millennia leaving behind rock and silt moraine that filled the valley and then levelling it off. Curzon determined that there were eight such “pamirs” in the Pamir Mountain region.

Thus, the Pamir Mountain Range takes its name not from its peaks (as is usually the case), but from these unique valleys, possibly the only range in the world to be so named. Even though the average altitude of the principal “pamirs” (i.e., the valleys rather than the mountains around them) is 13,000-14,000 feet, the easy terrain, wide and flat, provides an opportunity for subsistence farming based on livestock, now mainly yak and goat.

The Pamirs and these other great mountain ranges are at the very centre of Asia and historically, due to their height and scale, they naturally tended to separate the great regional empires. There is not a natural conduit of any sort; instead it is an area in which invading armies would get absorbed and neutralised. There is no easy or obvious route either through the Pamirs or around the conjunction of mountains to the East. As a result, successive attempts at the colonisation of India all involved a circuitous route around the Pamir Knot, such as those made by Timurlane, Babur and the Mughals.

Rivers and Lakes

The Oxus River, the focus of our journey, is today generally known as the Amu Darya and is the longest river in Central Asia. In broad terms, it can be said that the river rises in the Pamir Mountains of Afghanistan, traverses a total length of approximately 1,500 miles (2,400 kms), and empties into the southern end of the Aral Sea. In recent decades, so much water has been diverted from the Amu Darya in Turkmenistan and Uzbekistan for irrigation purposes that the river virtually ends in the sands of the Kyzyl Kum desert and the trickle that reaches the Aral Sea has not been sufficient to replenish it, resulting in a major environmental. However, the Kazakh Government is now taking steps to reverse the contraction of the Aral Sea and is having some success.

There are several significant tributaries which join together to form the Oxus and the 19th century saw considerable debate as to which of these was the true “source” of the river. The regional Map 1 shows these various tributaries of the Oxus. Map 2 on the following page provides a more detailed picture of the Wakhan Corridor of Afghanistan.

In the mid-1890s, Lord Curzon made a lengthy and well-reasoned argument that the true source of the Oxus River were two ice caves at the base of a glacier on the slopes of the Hindu Kush Mountains at the eastern end of Afghanistan’s Wakhan Corridor. This river of glacial water, known in these upper reaches as the Wakhjir River, flows northwest down the Wakhjir Valley. As it turns west at Bozai Gumbaz, it is joined by the Little Pamir River flowing in from the east across the Little Pamir. The waters of the Little Pamir River are snow-melt flowing down from the south-facing slopes of the Wakhan Range, part of the Pamir Mountains, to the Little Pamir valley at a point just to the west of Lake Chaqmak.

Below Bozai Gumbaz, this combined river is known as the Wakhan. Further downstream, at Qala-e-Panja, the Wakhan is joined by the Pamir River flowing in from the northeast; the Pamir River rises at the western end of a lake by the name of Syr Kol located in the Great Pamir valley. The Wakhan becomes the Panj at Qala-e-Panja and then generally is called Amu Darya from the great bend Northwards at Ishkashim.

About 55 miles (90 kms) further downstream, at Qala Wamar, the capital of the Roshan district, the Amu Darya is joined by the Bartang River flowing in from the east. The Bartang also originates in the Little Pamir valley near the source of the Sarhad River; it flows out of the eastern end of Lake Chaqmak and follows a long, circuitous route across Tajikistan to reach Qala Wamar. Shortly after the confluence with the Bartang, the Amu Darya begins to flow

[Map 2 - Wakhan Corridor of Afghanistan]

roughly in a westerly direction, until below the Uzbek city of Termez, the river turns northwest and flows between the Kara Kum and Kyzyl Kum Deserts. The Amu Darya becomes a delta just south of the Aral Sea.

Although Lakes Chaqmak and Syr Kol are only some 18 miles (30 kms) apart, their discharges take totally different routes, one east and north through the Tajik Pamirs, the other west and north via Ishkashim and Khorog. They meet about 240 miles (400 kms) later at Roshan. The reason for this phenomenon is that these lakes lie on either side of the central Pamir watershed.

As we departed for Central Asia, we assumed the Curzon was correct about the true source of the Oxus; as far as we knew, his position had not been strongly challenged. Table 2 summarizes our understanding about the Oxus and its tributaries and also provides the alternative names used for the various branches of the river.

In addition to the rugged and high mountain systems described above, the Oxus River provides yet another natural boundary separating the Central Asian countries from Afghanistan and the Indian Subcontinent. As a hurdle which all invaders had to cross, it inevitably became a significant part of Asian history and the river developed a certain mystique giving rise to its reputation as a barrier between cultures and races. As Curzon put it, “[i]ts waters tell of forgotten peoples and whisper secrets of unknown lands.”

Table 2 - The Oxus and its Major Tributaries
Primary name: Also known as: Sourced from: Confluence with other tributaries at:
• Oxus River (the ancient Greek name)
• Amu Darya (name used by Afghans beginning at Qala-e-Panja; used by Tajiks below Ishkashim) • Wakhjir River (name used in the Pamir-i-Wakhjir Valley)
• Wakhan River (name used from Bozai Gumbaz to Qala-e-Panja)
• Ab-i-Wakhan (name some-times used above Qala-e-Panja)
• Panj (Persian name used below Qala-e-Panja) Ices caves at the base of a glacier at the eastern end of one of the eight “pamirs”, the Pamir-i-Wakhan or Wakhjir Valley This is generally accepted as the prime source of the Oxus River; therefore, all other tributaries join it
Little Pamir River • Sarhad River
• Bozai Darya
• Chelap River Snow melt from the Wakhan Range forming a stream known as the Chelap (or Chilab) which flows through the Burgutai ravine onto the Little Pamir Flows into the Wakhjir River at Bozai Gumbaz
Pamir River • Panj Oxus River
• Panj
• Ab-i-Panj Waters flowing from the western end of Syr Kol, a lake on the Great Pamir (also known as Lake Victoria or Wood’s Lake) Flows into the Wakhjir River at Langar (Tajik side) or Qala-e-Panja (Afghan side)
Bartang (name used below Lake Sarez in Tajikistan) • Aksu River (from Lake Chaqmak to Murghab)
• Murghab River (from Murghab to Sarez Lake) Waters flowing from the eastern end of Lake Chaqmak in the Little Pamir Flows into the Oxus (or Amu Darya, depending on usage) at Vomar, Roshan, in Tajikistan

The Oxus is not as a major commercial and transport artery. At no time did we see, on the upper reaches of the river, either a boat of any sort or even a fisherman. In the regions we explored, the Oxus flows swift and strong almost immediately from its origins in the Pamirs and forces its way through very difficult gorges where navigation is impossible. Its waters are more navigable below Ishkashim and there is some river traffic. However, this has been much reduced in recent times as the diversion of water for irrigation has severely reduced river levels.

B. Overview of Oxus River Exploration

Early explorers such as Marco Polo, Benjamin Goez and Hueng-Tsuang gave some insight into the geography of the Pamir and Wakhan region but so many different names were used and there was so little attention to recording distance and direction that the maps were at best guesswork. The area had long been a part of the folklore and religion of Hindus, Zoroastrians, Persians and Buddhists. The Hindu Purana pointed to this part of Central Asian as the Aryan Paradise; the Persians saw it as one of the foundations of their race; Magian Zoroastrians had similar traditions and the Buddhists venerated the area as the land above its most sacred place Mount Meru. All considered the Pamirs as providing the source of the four great rivers of Asia; Ganges, Indus, Oxus and Sita (possibly the Yarkand River, although no rivers of any size ever flowed east out of the Pamir). Hueng-Tsuang claimed that the fount of these rivers was the Dragon Lake.

In reality little was known when Lieutenant Wood, R.N. ventured into the Wakhan in 1839 and, choosing to follow the northern Panj branch of the river, found Lake Syr Kul. This was presumed to be the Dragon Lake and the source certainly of the Oxus. His Journey to the Source of the Oxus (1843) remains a travel classic.

The whole history of Oxus exploration then became bound up in what has become known as the Great Game, a period of shadow boxing between the Russian Empire, whose representatives had gradually colonised most of Central Asia in the mid 19th century and the Government of British India, keen to maintain the integrity of its northern boundaries.

Until the discovery of the ice caves by Curzon in 1895 the leading contenders for the source of the Oxus had been:

  • Lake Syr Kul (or Victoria) promoted by Lieutenant Wood, probably the first modern era European to enter the Pamirs.
  • Lake Chakmaktin, promoted by Lord Dunmore and Colonel Trotter.
  • A lake high above the Wakhan-i-Pamir claimed by Francis Younghusband.
  • The Sarhad or Little Pamir River, promoted by several travellers such as Littledale and Major Montgomerie’s team.

Some of the theories required Syr Kul or Chakmak Lake to have more than one egress creating rivers on opposite sides of the watershed, which is intuitively unlikely.


We include here some practical information with regard to preparing for a journey such as this. Further helpful resources, such as a list of books, websites, and contacts for logistical support, are included in Appendix 1.

A. Access to the Wakhan

At the present time, there are really only two legal ways to reach the Wakhan Corridor: (i) fly to Kabul and then take a three day journey north by 4-WD vehicle to Ishkashim, the gateway town at the western end of the Wakhan; or (ii) fly to either Dushanbe in Tajikistan or Bishkek in Kirghizstan and travel south by 4-WD vehicle to the international frontier at Ishkashim.

It is certainly physically possible to enter Wakhan via one of the passes from Pakistan, such as Baroghil, Irshad, or Kilik. However, given today’s volatile security situation in this region, permission for such entry will most likely not be granted by either the Afghan or Pakistani authorities. The motives of anyone attempting mountain crossing will surely be misinterpreted and, irrespective of the personal risk, the opportunities for subsequent travellers will be prejudiced.

This is one area of the world where access is substantially more difficult now than it was, say, 100 years ago when Curzon or Younghusband could expect to reach the Little Pamir in a few days from Pakistan or even China. The geography and means of transport are unchanged from Curzon’s time, but the political barriers are much greater.

In our case, we chose to fly to Bishkek and to drive south through Kirghizstan and Tajikistan and enter Afghanistan at Ishkashim. This allowed us to visit Lake Syr Kol, one of the sources of the Oxus River on the Tajik-Afghan border, and it also provided an opportunity to see more of the Pamir Mountains which are located primarily in the Gorno-Badakhshan Region of Tajikistan.

B. Timing

Our expedition was expected to take three to four weeks from beginning to end and it needed to be completed sometime in the two summer months of July and August.

An important practical consideration was the level of water in the rivers at the western end of the Wakhan Corridor. There is a road of sorts between Ishkashim and Sarhad, where the Wakhan range intervenes to stop vehicle traffic. However, this road is not always passable because of floods from the snow and glacier melt. Previous visitors to the Wakhan region advised that the later in the summer we travelled, the better, as the level of the rivers is highest in the spring and falls for the rest of the year until they freeze over in the autumn. Furthermore, all advice agreed that by mid-September the weather would be too cold for comfort.

It turned out that the advice we received may not have been entirely correct. Local information during our journey later suggested that the rivers continue to rise in some places until mid or late July due to the glacier melt-off which gains pace through the summer, long after most of the snowmelt is over.

C. Visas

Obtaining the necessary visas is a complicated process, particularly where there are three travellers involved, each with a different nationality. The following notes reflect the conditions as they existed in the summer of 2007.


Visas for Kirghizstan were reasonably easy to obtain in both London and Washington, but the expedited service is expensive (£120 in London). No introductory letters are required. The consular staff are helpful, informative and proud of what they have to offer in their country.


Obtaining travel documents for Tajikistan proved to be much more problematic. It was necessary for us to obtain two documents: (i) a visa to enter the country and (ii) a special permit to enter the Gorno-Badakhshan Autonomous Region (GBAO) where the Pamirs are located.

Double entry visas were not available through the Embassy in Washing ton, but were eventually in Brussels. This meant that passports had to be couriered across the Atlantic, then to Brussels and back. Overall cost for three double entry visas was over $500.


The Afghan visas were the easiest to secure. The consular staffs in both London and Washington were friendly and helpful. They assume that you are going for work, usually security, and so ask for employers’ names and references; these are unnecessary for tourists. They may request a letter of introduction or proposed itinerary. The visa takes just 24 hours and will be available once you show that you have paid £30 into their bank account, for which they provide the necessary forms.

D. Maps and Local Information

The best maps available of the Wakhan region are those produced in the mid to late 1980s by the Soviet military. These can be purchased from several online dealers (e.g.,, The maps are quite detailed and several sheets are needed to cover the entire Wakhan. They are very expensive (ranging from $60-100 per sheet). The maps are purchased as a digital file which must then be printed. The drawback of these maps is that all labels are in Cyrillic, so it is best to have key areas and landmarks translated beforehand. Furthermore, the Soviets used a different system of measuring longitude which does not coincide with one’s GPS readings.

The best map of the entire Pamir region is one entitled “The Pamirs” which is distributed by Gecko ( This map was produced with support from UNESCO and the Swiss Development Agency and covers the Pamir Mountains from the Kirghiz border in the north to the Pakistani border in the south. It is indispensable for travel in this area.

Both and, the websites of Julien Dufour and Robert Middleton, respectively, are immediate and valuable sources of current information about the area.

E. Transport & Local Support

Travelling in the remote areas of Kirghizstan, Tajikistan and the Wakhan Corridor of Afghanistan requires appropriate vehicles or pack animals, as the circumstances demand, and someone who knows the way. We benefited greatly from a small group of Tajiks, Afghans, and Wakhis who made getting around possible.


We contacted Ergash Fayzullobekov on an introduction from Robert Middleton. Ergash is an employee of the Mountain Societies Development & Support Program which is funded by the Aga Khan Foundation, the most visible agency at work in both Tajikistan and Northern Afghanistan. Ergash and his driver Hossein really knew their way around Tajikistan and kept us on the right way even when it was difficult to tell if there was a road at all. And they were very adept at dealing with the police or military checkpoints that popped up occasionally in the GBAO region.


No journey of this nature in the remote Wakhan could have been accomplished without local support. Three Afghans --- Ghulam Sakhi Danishjo, Mohammed Nadir and Mohammed Shafi --- served as the expedition’s arrangers and fixers and, importantly, the go-between with the local police and tribal leaders. All are employees of the Great Game Travel Company and proved to be a real key to the expedition’s success.

Wakhi and Kirghiz

Our journey into the eastern Wakhan (that is, beyond Sarhad) required horses and handlers, cooks, and camp organizers. There were 16 of these who joined us at Sarhad and made their homes there or in the surrounding villages. They were all Wakhis, meaning that they are ethnically Tajik. Most were in their early to mid 20s, although several were considerably older.

Our formal communication was limited since these men knew only a few English words. But as the days passed, we began to discern their different personalities and a simple rapport developed between us based on a few words, lots of hand signs, and the challenging time we spent together.

These “mountain men” proved to be brilliant horsemen and incredibly well suited to the high altitudes and rugged terrain we encountered. Able to walk for hours in the thin air, their fitness was simply amazing. And we appreciated their sincere and unfailing efforts to ensure our safety in perilous situations that came all too often. In short, they got us through.

As described later, two Kirghiz joined the expedition at Bozai Gumbaz. We were forced to take them with us for several days given that we were, by then, in Kirghiz territory.

Appendix 2 provides a list of expedition team names and photos.


A. Kirghizstan

Wednesday, June 27. We landed at Bishkek’s Manas Airport at about 5:30AM and were quite pleased (and relieved) to see our Tajik guide, Ergash Fayzellobekov and the driver, Hossein, waiting for us with their Toyota Land Cruiser. We did, however, soon discover that neither of them had ever been to Kirghizstan before.

After a short stop for breakfast, we quickly got underway from Bishkek (2,625 ft; 800 m) and headed westward along the M39, the road that eventually reaches Tashkent. After about 35 miles, we turned south on the M41 and it was not long before we were into the foothills of the Kirghiz Alatau Range. We wound through deep, narrow ravines with a racing stream along one side. The road was in good condition and well maintained with many sections having been repaired with international assistance. We were soon climbing and within a few hours crossed the Tuu Ashu Pass (11,765 ft; 3,586 m) that took us down into the Suusamyr Valley, a high steppe plateau 7,220 ft (2,200 m) above sea level. The road ascended again into the Suusamyr Range and we crossed another high pass, the Ala-Bel (10,446 ft; 3,184 m).

As the afternoon progressed, we passed the Toktogul Reservoir which powers a large hydro-electric station. This lake is fed on its eastern end by the Naryn River flowing westward from the Tien Shan Mountains. Our road followed along the river after it exited the south side of the reservoir. By late afternoon, we dropped down to the marvellous Kara Suu valley, an aquamarine bath between the green swathed mountains. It is much more agricultural here, with rice paddies and fields of corn.

We all noted that Kirghizstan, based on the day’s experience, appeared to be a land of green mountains, a big contrast from the brown black peaks of the Karakorum and Hindu Kush.

That night we reached Osh, the second city of Kirghizstan. A lone traveller would have found it hard to find hotel rooms as the city does not seems to cater much for travellers. Our accommodation, the grandly named Taj Mahal Hotel, was located in an unmarked building above a bank. We had a street-side dinner of mutton and chicken.

B. Tajikistan and Lake Syr Kol

Thursday, June 28. The following day, we continued southward on the M41 headed for Murghab in Tajikistan; much of the drive was through the Alai Range. By early afternoon we reached the Tajik frontier where we encountered some snow. The road here, unpaved and very rough, runs through the Kyzyl-Art Pass (14,049 ft; 4,282 m) which lies between the Kirghiz and Tajik border posts; these are very isolated facilities. The Kirghiz guards were happy to see us and readily posed for photographs. A few hundred yards further were the Tajik guards, who were young boys. Ergash had brought them fresh bread, but they remained quite sullen.

Technically, we were now in the Pamir Mountain region. It is generally thought of as beginning along the Kirghiz – Tajik border where the Trans Alai Range lies.

We descended from the pass and were on flat steppe land. The road was straight as an arrow for miles. We passed an old Soviet military installation that had two huge spheres which undoubtedly had been some kind of radar or listening device. At this point, the road ran close to the Chinese border. While the actual border was some miles away (approximately 10), there was a barbed wire fence just off the road. This fence most likely creates a no-mans land up to the actual border and was built in Soviet times as a defence.

Eventually Lake Kara Kol came into view. It is the highest lake in Central Asia (12,845 ft; 3,915 m) and was created by a meteor approximately 10 million years ago. The water is salty and lifeless and the village of Kara Kul, which we passed on the eastern side of the lake, appeared lifeless also. The basin of the lake sits in the first of the broad “pamir” valleys we were to traverse, the Khargush Pamir.

[photo - Lake Kara Kul]

We soon encountered the first people that we had seen in Tajikistan, a group of Kirghiz nomads with herds of goats and yaks. We stopped and the children came running. This gave us the opportunity to distribute some of bangles and other cheap jewellery, sunglasses, and small toys that we had brought along for such occasions.

[photo - Kirghiz children close to Murghab]

By mid-afternoon we crested the Akbajtal Pass (15,272 ft; 4,655 m) which lies approximately midway between Lake Kara Kol and Murghab. Here we passed the only car we were to see in Tajikistan outside of Murghab. It was another vehicle of the Mountain Societies Development & Support Program.

We descended from the pass onto a flat plain with mountains on either side. It was very barren, but the colours were fabulous. Some 30 miles (50 kms) south of the pass we skirted along the western opening into another “pamir” valley, the Rang Kul Pamir.

A heavy rain started and continued all the way to Murghab, where we arrived at 9PM. Murghab (11,975 ft; 3,650 m) was awash and unwelcoming. It took a long, wet time to find accommodation and food. It rained about an inch in Murghab that night, rather a lot for a town that has an average rainfall of only three inches for the whole year.

That night two of us suffered considerably from the headache and nausea of altitude sickness. It had been a mistake to travel in one day from Osh at about 2,500 feet, through the Kyzyl-Art Pass at 14,000 feet and Ak-Baital Pass at over 15,000 feet, to Murghab at about 12,000 feet. Too much altitude too quickly.

Friday, June 29. Very soon after departing southward from Murghab on the next day, we traversed a second “pamir” valley. This was the Sarez Pamir, a great wide peaceful, pale green flat land between the mountains. The scenery was spectacular. The floor of the valley was just packed sand and we roared along in the vehicle for 10-15 miles.

At the end of the plain, we began to ascend and continued over grassy, but very rocky and rutted terrain. Here there were no tracks and we were all searching for the route. Our objective was Syr Kol, a large lake on the Tajik-Afghan border that is the source of the Pamir. The going was very difficult, but Hossein, our driver, was brilliant behind the wheel. This would have been a great advert for Toyota; our 11 year old Land Cruiser took considerable abuse and never whimpered.

During the morning, we saw just two groups of people. The first was a large Tajik family living in a semi-permanent stockade. The men were busy forming cow manure into large bricks and setting them in the sun to dry for fuel. Then, about 11AM, we happened upon a Kirghiz nomad family living in yurts. They quickly invited us to have lunch; it seemed as natural as saying “good morning” that they should offer to share their food with strange travellers. This was our first introduction to the milk products that dominate their diet; it included the double yak cream that would be the envy of three star Michelin chefs, as well as every creamery in Padstow and around. These Kirghiz move four times a year with their herds of yaks and spend the harsh winters in Murghab.

By then Ergash was having trouble finding the route, but the Kirghiz gave us fresh directions and we pushed on. We knew instinctively that, if we travelled west with the Wakhan range to the south, we would eventually reach the watershed and streams would start to flow toward Syr Kol, thereby giving us better orientation. The going was difficult; there were many small rivers and often it took several attempts to find crossing points. We had to be careful as we had only the one vehicle and could not afford to get stuck.

As we approached the lake, we encountered a small military checkpoint. Ergash got us through and soon we could see Syr Kol just to the south. This was an exciting moment, as we had now reached one of the main sources of water flowing into the Oxus River.

Syr Kol lies on the Great Pamir, the largest of the high altitude valleys that give this region its name. Just to the south of Syr Kol, on the other side of the Great Pamir, the Wakhan Range (also referred to as the Nicolas Range in the 19th century) rose up from the flat valley, capped at the top with snowy peaks.

The lake was a somewhat different shape than it appears on the two maps that we used,. For a while, it was hard to see the westward discharge that creates the Pamir River. Soon we saw it: the flow is wide, strong, and fast almost immediately and is full of rapids.

[photo - Syr Kol, or Wood’s Lake Victoria]

The Tajik-Afghan border runs through the middle of Syr Kol. From this point westward the boundary between Afghanistan and its neighbours Tajikistan and Uzbekistan follows right along this river: first the Pamir River till it joins the Wakhan River at Langar and then the Wakhan River / Amu Darya / Oxus (all one and the same) for hundreds of miles until the Amu Darya turns northwest into the Kyzyl Kum Desert.

It was a further six hours to Langar where we were to stay for the night, the last two of which were on reasonably well established roadways which followed the river, sometimes very high above and sometimes close alongside. There seemed to be more livestock such as camel, yak and goat on the Afghan side than the Tajik side but no established roads or tracks were visible. As we descended to Langar we found lush green grass and trees.

(Add photo)

C. Into Afghanistan and the Wakhan

Saturday, June 30. From Langar we continued to follow road alongside the Pamir River southwest toward the Afghan border post at Ishkashim. Across the river, we could see the broad Wakhan valley that the next day we would follow into the corridor. The river bank on the opposite side consisted almost entirely of a broad and deep belt of glacial moraine; this fanned out from the side valleys from which it had been deposited by glaciers in past millennia.

We arrived at Ishkashim by late morning. Ergash was anxious for us to visit a bazaar located behind a high stone wall on the island in the middle of the river that serves as the border crossing. Unfortunately the Tajik border guards said that foreigners were not allowed into the bazaar that day, so we collected our baggage and headed for the nearby border posts. As we did so, we were met by our Afghan guide, Ghulam Sakhi Danishjo, and two of his colleagues.

We cleared the Tajik border checkpoint with no problem, paid off Ergash and Hossein, and made arrangements for them to collect us at Ishkashim in two weeks time. We then experienced some initial resistance from the Afghan guards, saying they did not accept foreigners at the border on Saturday. They relented and we were allowed through, but not without considerable interviews and negotiations with officers from the border control and the Ministry of Interior, during which we were questioned extensively about our personal biographies and the purpose of our trip. To a certain extent, this was understandable because a Russian had recently disappeared in Wakhan and they needed to know as much as they could about the three of us.

Having finally crossed into Afghanistan, we had a quick lunch at Wafai’s Guesthouse which is supported by the Aga Khan Foundation. Over lunch we were somewhat disappointed to learn that our guide Sakhi had no greater knowledge of the eastern Wakhan area than we did, as he had never been there before.

We next reported to the office of the regional Chief of Police in Ishkashim, Mir Abdul Wahed, to have ourselves registered and to obtain letters of introduction to present at police checkpoints further in the Wakhan. As we examined the very large scale Wakhan maps on the wall, Sakhi and the Chief had a conversation in Dari which sometimes bordered on argumentative and went on so long we were soon concerned that we had a big problem. It emerged that the chief was “concerned” for our safety in the Wakhan, particularly from criminal activity, and wondered whether we wanted an armed police escort to accompany us. Sakhi, who is suspicious of the motives of all Afghan police, was against this idea. We obtained the letter of introduction, but declined the escort.

By early afternoon, we left Ishkashim in a Nissan 4-wheel drive vehicle accompanied by a Toyota double-cab pickup truck. For the next five hours we drove across the glacial moraine we had seen that morning from the Tajik side of the border. It was interesting so see how the character of the moraine changed as we drove eastward: some areas consisted of fine, barren sand almost like a beach, others of boulders and rocks, and others of sand with clumps of vegetation. One mile to the next, things changed.

For all its remoteness, this is a gentle and mellow land, lime green in the early summer in the wide flat valley; mild browns as the mountains sides take over from the river and Pamir basin, usually about half a mile wide. The mountains rise about 2,000 feet above the valley, although the much higher peaks of the Hindu Kush are soon visible between the lower peaks around the valley. Sometimes the river spreads to cover the whole valley. Red clothed, red headressed girls tend goats and horses, all a bit furtive and fearful of contact.

There were several places where rushing streams flowed down from the Hindu Kush to the river. Some were several feet deep and required considerable manoeuvring to cross. Usually this required one of the Afghans to wade into the water to find a safe crossing point.

As darkness closed in, we arrived at the little hamlet of Ishmergh; here we were roughly halfway between Ishkashim and Sarhad where our trek in the eastern Wakhan would begin. In Ishmergh, a village of about 25 families, there was a newly built guesthouse and even a new long drop loo with a view over the valley. The dinner ritual that evening was to become a very familiar one: first the aluminium bowls and water pot for hand washing; then the tarpaulins and vinyl “table” cloths, from which fell mill wheels of unraised bread, then the bowls of tasty yogurts, then oily mutton and rice. We were a curiosity for these villages and at dinner many of the men and boys crowded into our room to watch us eat.

Sunday, July 1. Before departing the following morning, we distributed bangles, sweets and medals to the children in the village. It was really a treat for us to see what pleasure these rather simple things brought to these kids.

By 8:30AM we had arrived at a police checkpoint at Kala-e-Panja. It was just across the river from Langar in Tajikistan where we had slept two nights before. Here is the confluence of the Pamir River, flowing down from Syr Kol, and the Wakhan River flowing along the Wakhan Valley. The road now turns south-eastward into this valley.

[photo - Wakhan valley]

Again we were frequently fording streams rushing down from the Hindu Kush. At times the vehicles were up to their undercarriages in water. At one point, the Toyota became stuck in a muddy area and had to be towed out.

By late morning we stopped in the village of Shelk to pick up one of the cooks, Mehraban, who would accompany us on our trek in the eastern Wakhan. The village seemed to have three or four families and a guest house where we were welcomed and graciously offered lunch --- bread, delicious yoghurt and tinned tuna. The village itself was neat patchwork of small fields with its own little irrigation network supplied from a higher stream. There were numerous slate or sod dams that could be opened to divert water into the individual fields. Goats and donkeys are tethered or roam freely.

A few kilometres later, the valley narrowed and the river deepened and roared through a gorge. We came across a house and compound with a satellite dish and two windmill generators. This was a medical clinic supported by the charity ORA International and managed by an Englishman, Dr. Alex Duncan, who lives there with his wife Eleanor and four young children. They have been in this remote area for several years now, with Alex treating Wakhi and Kirghiz patients, conducting medical surveys, and carrying out various medical research projects. When we arrived, he was planning a journey into the Little Pamir to test Kirghiz nomads for disease and opium addiction. As he is working with a people who have never been exposed to antibiotics, he will also be testing them to determine the levels of immunity that they develop.

Alex advised us that the Wakhan River would, most likely, not reach its maximum depth for another two weeks and that we could expect problems in crossing some of the feeder streams which were swollen by snow melt flowing down from the mountains. Several miles further along we did in fact encounter two places where the road was in bad shape and high water made crossing the mountain streams difficult, but our drivers were able to negotiate these trouble spots without incident.

A few hours later, the valley widened out to about three miles and the prospect in front of us changed. No longer was there the flat, but boulder strewn riverside; instead, we faced a mountain wall in the distance.

We arrived at Sarhad (10,675 ft; 3,245 m) at about 5:30PM. Here the Wakhan riverbed is at its widest, perhaps 10 miles across, and the flow is very shallow. On the opposite side to the south we could see a new valley opening, which is the route to the Baroghil Pass leading into northern Pakistan and eventually to Chitral and Mastuj. Sarhad is the farthest point west that Lord Curzon reached in 1894; he turned south here, crossed the Baroghil, and returned to India. Sarhad also marks the end of the road, such as it is, in the Wakhan. From here on, we would have to travel on horseback or by foot.

The guesthouse at Sarhad has been subsidised by the Aga Khan Foundation and the Norwegian Ministry of Foreign Affairs, another reminder of the quiet support given by these organizations in this totally Ismaili area.

There was a long discussion over dinner about our proposed route. It seemed a fairly pointless discussion, as there were only two possible routes. Of these, the lengthier and possibly easier one to the north apparently remained blocked by snow. There was an even longer discussion about our objectives in finding the source of the Oxus. None of our Afghan colleagues has been up the Wakhjir valley (Wakhan-i-Pamir), but they know where it is. Some talk of Lake Chaqmak as a source and even suggest that it empties westwards, which cannot be the case unless it is like the Sonle Lap in Cambodia and changes direction at different times of the year.

[map or photo -- the inner Wakhan; the green line shows the route from Sarhad to the Ice Cave]

D. Wild Thyme and Precipices

Monday, July 2. By very early next morning the team was gathering. Our expedition party was to be 22 members strong: the three of us, our three Afghan guides from the Great Game Travel Co., and the 16 Wakhi recruited from Sarhad and the surrounding villages. The Wakhi also supplied the necessary animals: six horses for riding, nine pack horses, and one donkey.

An hour or so was spent loading the animals and by 8AM the caravan set off due east for the Wakhan Range barrier.

At the outset, all three of us eschewed horses as we still expected most of the expedition to be by foot. However, the morning was spent climbing through the Daliz Pass (13,999 ft; 4,267 m) and within two hours the height and steepness had taken their toll and we were all mounted. The Daliz Pass was to be as high as any on the journey, but it was one of the easiest as the highest parts consisted mainly of a climb through a rocky water course, by now dry, rather than an exposed climb up the side of the mountain. Nevertheless it was still quite tough going for both horses and men.

The top of the pass opens onto rolling grassland, which for the most part is actually wild thyme rather than grass. Here we stopped for lunch, where it was bright and sunny, but windy.

In the afternoon we descended from the Daliz and passed through a region of spectacular scenery. But there was a price to pay for the beauty: it was one of the worst parts of the journey for those who do not like precipices. The paths, generally no more than a foot wide with a sharply angled drop on the outside, wound in and out along the contours of the mountainsides. It was harrowing and one was reminded of Curzon’s words describing his travel in this very area:

“…we again left the main gorge and diverged inland, mounting and descending successive spurs of great steepness and difficulty, down one of which one of our Kirghiz ponies, missing his footing, slipped and fell, and was killed instantly.”

Thankfully our ponies were nimble and surefooted and eventually we found ourselves some 3,000 feet above the Wakhjir River, which we saw again for the first time since leaving Sarhad that morning. From here we worked our way down the mountainside and camped alongside the roaring river in a clump of trees at a place called Borak (or Baharaq).

[photo -- rejoining the Oxus after the Dalez Pass]

Tuesday, July 3. Soon after our departure the following morning, we crossed a very rough bridge over the Borak River that rushes down from the Wakhan Range through a deep gorge. The bridge covered a gap of about 30 feet and consisted of scrub timbers lashed together. It hardly inspired confidence and as a precaution we crossed one at a time and avoided looking at the long drop to the rapids below. This was followed by a long path up the mountainside consisting of one switchback after another; it climbed and climbed until we were several thousand feet above the river. Not long had passed before the path became so steep and narrow that riding was too hazardous and we had to walk up.

Lunch was at a site called Warve Gage. The remainder of the day was one of physical toil for the horses and horsemen and emotional toil for us as we followed the narrow path along the winding contours of the mountainsides, one steep uphill and downhill after another. The Oxus was rarely out of our sight. Somehow it was worse with the river raging loudly through the gorge many hundreds of feet below as the waters’ roar echoed up the hillsides giving apparent substance to the void to our right. It was best not to look down and simply trust the horses.

We continued to be amazed at the physical abilities of our Wakhi colleagues. They walked hour after hour up and down these steep hills and hardly took a deep breath. To add to our amazement, this feat of endurance was accomplished while wearing some of the shoddiest footwear one could imagine: sneakers or locally made slip-ons or tie-shoes, most of which appeared to be several sizes too large.

We camped at Langar (12,323 ft; 3,756 m), an alpine meadow facing the snowy peaks of the Hindu Kush to the south. A beautiful site, but it exposed us to considerable cold and wind.

Wednesday, July 4. Our morning stage was interrupted as we crossed a rudimentary bridge over a rushing stream. The bridge, low in the water and consisting of piled-up rock topped with flat stone, was slippery and one of the pack horses lost his balance and fell off on the upstream side. He was unable to move due to the heavy, wet load and was being pummelled against the rocks. Immediately two of the Wakhi horsemen jumped into the shallow, but ice cold water to calm the animal, remove its load, and lead it to the bank.

Since leaving Sarhad two days earlier, we had looked forward to being out of the mountains and now at last we began to descend. Our downward route was broken by several expanses of flat land and meadows and we took the opportunity to dismount and walk for long stretches. We lunched at Mirzamurad and by mid-afternoon we could see the open vista of the Little Pamir, that wide glacial valley, in the distance. We rode on and later encountered our first group of Kirghiz nomads in the Wakhan. When we finally came in sight of a number of domed burial tombs fashioned out of mud brick, a photograph of which had been included in Curzon’s RGS report, we knew we had arrived at Bozai Gumbaz. To our knowledge, these are the only recorded structures in the inner Wakhan. Nearby we found the ruins of a Soviet military post surrounded by barbed wire.

We camped just beyond these tombs with a magnificent view down the Wakhjir Valley to the southeast. The sheer enormity of the vista before us was the source of its overwhelming grandeur. The Wakhjir River flows out of this valley and we were only a short distance from its confluence with the Little Pamir River (or ‘Scottish’ river from its likeness to the Highlands) lowing west across the Little Pamir. Most entertaining, though, was the thought that our campsite must have been located at, or very near, the place where the Russian Colonel Yanov entertained Captain Younghusband in their celebrated 1891 meeting. This simply had to be the place; given the view, it was hard to imagine that anyone arriving here would have chosen to camp elsewhere.

[photo - ‘Scottish river’ Little Pamir]

Only a short distance to our north was a Kirghiz nomad camp and it was not long after arriving that the local chief and a few others brought us the traditional welcoming bowl of yoghurt. However, this wizened thin man with a woolly hat and the looks of a 55 year old (but who in reality was probably in his 40’s) had other motives, and he was soon engaged in negotiations with Sakhi and his two Afghan colleagues. At issue was an informal understanding the Kirghiz have with the Wakhi that, from Bozai Gumbaz eastward, they (the Kirghiz) are entitled to supply all pack animals to travellers; they were now demanding their share of the business. This was not an attractive prospect to us: we were now accustomed to, and pleased with, the Wakhi and their horses and we had no interest in adapting ourselves all over again.

There was a long discussion in Dari with the Kirghiz. It soon became clear that, as the Kirghiz were about to migrate to other pastures, they were in no position to spare the 15 or more horses and men we required. Their demands were mere posturing because they simply could not supply the goods. Finally, a compromise was agreed whereby we took on two Kirghiz horsemen and left two of our Wakhi at the Bozai Gumbaz camp to rejoin us upon our return. Even these two Kirghiz were to cause us problems.

That evening we bought a sheep and the Wakhi slaughtered it. The mutton was to be an important part of our food for days to come.

E. The Ice Caves

Thursday, July 5. Early the next morning we walked to the nearby Kirghiz encampment where we had been invited for breakfast. Again we took the opportunity to distribute bangles, balloons, and sunglasses to the children and their excitement was heart-warming. Breakfast was taken in the chief’s yurt with some ceremony; he graciously said to us that, “we can hardly ask you to share our grazing without also sharing our home”. There was bread, delicious double cream, and tea. We discussed their upcoming migration, as well as their lack of medical care; sadly, too many Kirghiz women die in childbirth.

We walked ahead of the caravan down to the nearby Little Pamir River (Curzon called this the Sarhad branch, which is hardly appropriate). Right here we could have been in the Highlands of Scotland. The river sliced a neat green swathe through rough, tussocked country, sparkling in the morning sun. It varied in width between about 100-200 feet (30 – 60 metres) and was cleaner than anything we had seen since the uplands of the Tien Shan in Kirghizstan.

We followed the river towards its confluence with the Wakhjir River flowing towards us down the Wakhjir Valley. From our visual observation, it was hard to dispute that, at this time of year, the Wakhjir River appeared to be larger than the Little Pamir branch. We questioned the Kirghiz and the Wakhi about this and there was little dispute. In their opinion, the bright, tumbling brook from farther down the Little Pamir was generally smaller than its Wakhjir neighbour even though the latter disguised its depth by spreading across a wide basin. As one Kirghiz put it: “this one (i.e., the Little Pamir River) can always be crossed by horse, whereas the Wakhjir sometimes can only be crossed by a strong horse and sometimes not at all.”

At the confluence, we also noted the different colours of the water in these rivers as they poured into one another. The Little Pamir River was clear and undoubtedly came from snow melt high up in the Wakhan Range. The Wakhjir River was grey and chalky, typical of glacial melt waters which carry fine silt produced as the glacier advances.

We crossed the Little Pamir River and began our trek down the Wakhjir Valley, or Pamir-i-Wakhan. This was the fourth “pamir”, or glacial valley, of our journey, having already seen the Khargush, Sarez and Great Pamir valleys in Tajikistan. We soon encountered a Kirghiz shepherd and his flock of sheep and goats. Later we saw several groups of Kirghiz, either loading up in preparation for their migration or already underway. The valley was extraordinarily broad, so much so that for a long time that morning we never seemed to move; only by taking a sighting against distant Bozai Gumbaz could one discern progress.

Around midday we climbed up to the north onto a meadow and stopped to eat beside a small Kirghiz camp, a location known as Kiskintash. As we occasionally saw elsewhere, these Kirghiz utilized not only their traditional yurts (jailoo), but also more permanent flat-roofed huts made of mud bricks. There were large stacks of yak dung drying in the sun for later use as fuel.

The view from this location back down the Wakhjir Valley toward Bozai Gumbaz was as beautiful as anything we had seen. This was another Hunza, but without Rakaposhi. The chalky river gave way to moraine and in turn to green banks, to mild green thyme prairie, then to hazy rocky uplands and eventually to snow covered rolling peaks.

After lunch we visited one of the yurts being used for cooking. There were only women inside, three generations of them in fact, and initially they were reluctant to communicate with us. But with time, they began to talk and soon we had a lesson in baking bread in what was, more or less, a Dutch oven, as well as in making yogurt and the fabled yak double cream.

During the afternoon stage, a nascent horsemen’s revolt about our rapid pace down the Wakhjir Valley had to be averted. This mild confrontation was surely fomented by the two Kirghiz who had joined us a Bozai Gumbaz. We had indeed moved down the valley faster than expected; the Kirghiz clearly wanted to extend the trip as much as possible as this would increase their earnings. In the end, the revolt was never much more than a discussion and soon died out.

Later we passed more burial tombs sitting on a bluff overlooking the river. We stopped shortly thereafter at a site known as Karatash where we set up camp.

Friday, July 6. We had been told that the forceful Diwanasu River might impede our progress this morning and we needed to cross it early before the day heated up and increased the snow melt feeding this stream. With this in mind, we reached it about 9.30AM and managed to get all loads across without too much trouble. The crossing was a lively event though, with much shouting at the animals and between the horsemen. A few minutes later we saw the curved horns of a Marco Polo sheep. Sadly this was the only evidence we found of these magnificent animals, so celebrated in the Wakhan.

Later we had lunch was at Bazsil, a meadow just above the Wakhan River. Some three hours further along, about 2PM, we reached the last, long bend in the valley before we expected to see the much anticipated glacier and its ice caves. We were now only a few hours journey away from one of our main objectives.

A nearby site, known as Chalapsu, seemed an appropriate place for the night and we now left most of the expedition team to set up camp while just a few made the final push to the glacier. This small party consisted of the three of us, our three Wakhi horsemen, along with Sakhi, Nadir, Shafi and the one Wakhi who claimed to have some prior experience at this end of the valley.
Based on earlier reports, we knew that the caves should be at an altitude of about 15,000 feet; that would be some 800-900 feet above the Chalapsu camp if our GPS device was correct. Given that the valley ahead appeared to be relatively flat, this meant that the ice caves should be located well above the riverbed.

As we rode beyond the campsite, we found the terrain to be mixed. At first it was not easy: there were streams to cross, as well as rocky fields and other grassy meadows covered in pits that were one to two feet deep. But later we were, for the first time on the journey, riding mainly along the riverbed itself. Here the going was as easy as it had been at any stage in the previous six days. It seemed that all we had to do was to continue up the moraine bed and we would reach our objective.

[photo - approaching the glacier]

Further down the valley we saw tracks up a steep ridge to the left, which should lead to the Wakhjir Pass and the Chinese border, where we understood the Chinese had placed a few concrete posts to mark their territory. This is one of the Wakhan passes that was in fairly regular use in the late 19th century and where Curzon crossed into the Wakhan in 1894.

Eventually the valley narrowed and the top of the glacier began to emerge into view. Since the valley curved to the right (i.e., to the south), its base was still hidden by the sloping hills of the Hindu Kush on the south side of the valley. We continued along the bend and then, having ridden up on the northern flanks of the valley a bit, we suddenly had a full view of the glacier in the distance. It was not high up, but seemingly accessible and low. The valley ended abruptly at the glacial wall; the glacier then led back up into the mountains, sloping much more steeply as it rose and turned right to the south into Pakistan and the Karakorum. For a moment, it was disappointing that the target, the prize, should be so easily obtainable and our step quickened. Unknown to us, though, we still had considerable work to do before we were there.

Even at this vantage point, we could not see anything that resembled the ice caves. We continued to expect that we would have to climb up on the glacier to reach them. At the back of our minds, the thought even occurred to us that perhaps they were no longer there.

Our situation was now a bit precarious. It was already about 4PM and daylight was quickly running out. While the glacier was before us, we simply did not know where the caves were. We needed to get closer, and quickly, but how to do so was not clear. There was no route along the south side of the river, only precipitous rock leading to the upper part of the glacier. Directly in front of us there was a vast, sloping field of boulders leading down towards the glacier; this would be treacherous. The one possibly simple route was to continue up the northern bank till we reached what appeared to be a plateau 400 feet or so above the valley; we could then follow this plateau as it curved around to the glacier and come down to the face of it from above.

At this point, Nadir attempted to force the issue by riding further up the hillside toward the plateau. We followed, but very quickly it became clear that the hill was much steeper than expected and might be too difficult for the horses. The much slower approach across the boulder field would be necessary. We left the horses there on the northern slope of the valley with a couple of our Wakhi colleagues and headed downhill on foot.

This expanse of glacial moraine was a nightmare. There were boulders of all sizes piled up everywhere and no obvious path through the jumble to the glacier. Climbing over and around these smooth (and therefore slippery) boulders, some huge, some small, was awkward and difficult; to make matters worse, we were tired and irritable and in a hurry --- a recipe for injury. Then we ran into two streams rushing between the boulders. They were strong and cold, even for the Wakhis, but not impassable in width and depth; the problem was finding a place to cross. We contemplated trying to skirt the rivers and cross higher up, but decided against this as we didn’t know how far we would have to go. It was guesswork and time was passing. Slowly, but surely, each of us got across without incident.

It took the better part of an hour to cross the sloping boulder field and down to the level of the riverbed. At this point, we could no longer see the glacier; the view was blocked by a huge mound of shale. Our Afghan colleague Shafi scampered up and from that high vantage point could finally see the ice caves --- they were at the base of the glacier where it met the floor of the valley and not higher up as we had expected. Shafi called to us and we followed him up the hill; it was like climbing up a stony sand dune and punished our already exhausted feet and legs.

From the top we looked down on two smaller hills and just beyond there was chalky-grey water rushing out from two caves in the black face of the glacier. These streams joined and flowed down the Wakhjir Valley. Here we were at the source of the Oxus. We stood and photographed and then clambered back down to the riverbed; within minutes we were at the water’s edge just in front of the glacier. The cave on the right (i.e., to the south) was small and still fully intact. The cave on the left (i.e., to the north) was considerably larger; the front part of its roof had collapsed and now lay blocking the entrance. Despite this, it was clear that the larger volume of water flowed out from this cave.

There could be no doubt at all that these were the caves found by Curzon in 1894. In every way, the caves and surroundings we were looking at met the description he had included in his report to the RGS:

“I rode up to the source. There the river issues from two ice-caverns in the rushing stream. The cavern on the right has a low overhanging roof, from which the water gushes tumultuously out. The cavern on the left was sufficiently high to admit of my looking into the interior, and within for some distance I could follow the river… Above the ice-caves is the precipitous front wall or broken snout of the glacier, from 60 to 80 feet in height, composed of moraine ice, covered with stones and black dust.”

But for the fact that the front of the roof of the left cave had fallen, we would have been able to see deep into its interior. In fact, Anthony and Sakhi climbed up on part of the collapsed ice and were able to see a considerable distance inside.

We checked the GPS device as we stood on the side of the river across from the caves. The altitude was 14,900 feet, more or less the height at which we expected to find the caves. Contrary to our expectations, we had not had to climb up on the glacier to reach them. Without realizing it, we had gradually been gaining height ever since leaving the campsite at Chalapsu. The vastness of the landscape had confounded our perception of distance and inclination.

[photo - Bill Colegrave, Antony Kitchin, Dillon Coleman at large Ice Cave]

We lingered only a few minutes to take more photos and to savour having achieved our biggest objective. We now faced the daunting task of getting back to camp before darkness set in. Given the rocky streams, uneven ground, and deeply rutted meadows that lay ahead of us, there were plenty of opportunities for man or beast to be injured as our visibility declined. We were exhausted from the afternoon’s effort, but there was no time to waste and we mounted up. Just riding out of the boulder field near the ice caves was treacherous as the rocks played havoc with the horses’ footing. To add to our challenge, it became rather chilly as the sun dropped behind the mountains. Our horse handlers walked as fast as they could and, through their efforts we arrived at the Chalapsu camp without incident at about 7:30PM, just as the sun’s last rays disappeared.

G. Lake Chaqmak

Saturday, July 7. We were up early today to re-cross the rushing Diwanasu before it became too strong. This we did by about 9:30AM. We marched on, retracing our steps back down the Wakhan Valley, but stopped early, around 3:30PM and set up camp on a grassy bluff overlooking the river. This was Guretuk.

It was a very hot day, perhaps our hottest. Certainly 850 F, maybe more. But late in the afternoon, with the sun disappearing, there was a strong wind and it was cold again.

Shortly before dinner, a young Kirghiz boy arrived in camp looking for a moment’s respite and a cup of tea. He was perhaps 15 years old, with a face so young he could have been mistaken for a girl. But of course no female would be out travelling alone and stopping into a camp of strangers. He was dressed in what appeared to be military fatigues with a red head scarf. After drinking the tea, he mounted his donkey and departed. He was small in statue, but the donkey was so tiny that the boy’s feet were almost dragging the ground as he rode off into the semi-darkness.

In the early evening, a heavy rain began to pour, accompanied by thunder and lightening. This made the horses nervous and there was considerable whinnying.

Sunday, July 8. We awoke to find 2-3 inches of snow on the ground. The mountains around us, Hindu Kush to the south and Wakhan to the north, are lovely at any time, but with the covering of new snow that morning, they had a pristine beauty about them that was breathtaking.

Our Wakhi colleagues were busy spreading out their bedrolls and horse blankets to dry. As always, they had slept outdoors and of course had been drenched, first by the pouring rain and then by the snow. But they reassured us that they had been caught in much worse weather, sometimes in snowstorms up to a meter in depth.

We set off on the final stage back to Bozai Gumbaz and arrived there some four hours later, just before noon. Along the way, we noted that the Kirghiz in two of the villages we had passed on the way out were now loading up their yaks and camels in preparation for their migration. Two other villages had already been abandoned.

We had completed the trip from Bozai Gumbaz to the ice caves and back in 3 ½ days. The Kirghiz said it should take six days, but of course they wanted the pay a more extended trip would bring. As we lounged once again in the warm sunshine at Bozai Gumbaz, we reckoned that five days would have been about right.

Monday, July 9. We had by now accomplished quite a lot in our quest after the various components of the Oxus River system. We had seen two of the Oxus sources, Lake Syr Kol and the ice caves, as well as Wakhjir River, the ‘Little Pamir’ River, and the Pamir River. We had also seen the confluence of the Wakhjir and the Little Pamir Rivers at Bozai Gumbaz and the confluence of the Pamir River from Syr Kol with the Wakhan River at Langar. The one key source that we had not seen was Lake Chaqmak. If we could visit this lake and later see the confluence between the Bartang River and the Oxus at Roshan, we would surely be the only party that had at least seen all the important elements of the Oxus source argument in a single journey.

So, on this morning, instead of returning over the mountain barrier toward Sarhad, we rode east up the Little Pamir valley (the fifth “pamir” we had seen) in the usual sparkling, blue-brown high altitude Wakhan morning. Our objective was to see Lake Chaqmak (or Chakmaktin), but we also had another important mission.

In addition to Chaqmak, we were anxious to see a river which the Wakhis called the Chelap. From our review of maps of the Little Pamir, it appeared that this river presented a most unusual geographical phenomenon. It flows down the south side of the Wakhan Range, through the Burgutai Gorge, and onto the Little Pamir valley just to the west of Lake Chaqmak. However, our maps also indicated that, as the Chelap enters the Little Pamir with its relatively gentler slope, the river splits into a number of channels which then continue straight onto the Little Pamir. But it is not long before most of these streams begin to flow eastward into Lake Chaqmak (and eventually to the Bartang River), while one stream flows westward across the Little Pamir towards Bozai Gumbaz. This is the Little Pamir River.

For an established river such as the Chelap to split into several streams could only indicate that it was flowing across very flat land where there was no obvious direction for it to take. For the streams thereafter to flow in opposite directions almost certainly meant that they had arrived at a watershed area where the earth’s uneven surface forced the various channels to flow in one direction or the other. It is certainly rare and unusual for such events to occur together as the maps showed was the case with the Chelap River.

If indeed the Chelap River fed both the Little Pamir River and the Bartang River (via Lake Chaqmak), this would have remarkable implications for the arguments about the source of the mighty Oxus. This would unify, in the most unexpected way, two of the four claimed sources of the river. And it would potentially call into question many of the most important conclusions by Curzon, Ney Elias and others.

We started north from our campsite, leaving the now abandoned Kirghiz encampment to our left. We crossed several small streams and rode up the wide valley to a plateau, whence we expected to see Lake Chaqmak. We should by now have understood that such expectations are never realised and it was many, many more crests in the rolling “pamir” before we could see the blue sparkling bracelet of the lake.

We kept close to the north side of the Pamir for several hours hoping to run right into the Chelap River as it flowed south out of the Wakhan Range. We marked off each successive small side valley on the map and began to close on the Lake. Ahead and due east, the Little Pamir stretches to the end of Wakhan and terminates at the snow topped mountains that form part of the Tajik, Afghan, Chinese border. These seem to have a much lower snow line than the immediate Pamir peaks and form a distant back drop that can be seen reflecting in the cornflower waters of Lake Chaqmak.

Eventually we did ford one clear ‘Scottish’ river. We had been expecting to cross several more if our expectations about the Chelap splitting into multiple streams were correct. We followed this clear stream across the Little Pamir and from some vantage points could see that it ran uninterrupted towards Lake Chaqmak. We had clearly passed the watershed and all the rivulets, most of which by now were dry, were running east.

As we had still failed to find any river flowing westward, we were increasingly confused by what we had seen on the maps . For the first time we consulted the horsemen who had been to the Little Pamir before. They gave a strong negative and it seemed that perhaps they were right. It was indeed unlikely, whatever the maps seemed to indicate, that any river once established, would divide and then flow in opposite directions on either side of a watershed.

We took our lunch beside the Lake Chaqmak branch. Despite our disappointment at not having confirmed the phenomenon shown on our maps, it was a memorable event, not for the lunch, but for the fishing expedition that followed. The Wakhi managed to catch fish in one of the rivulets by using the tied up ends of a shalwar kameez as a net and even by simply snatching them out of the water in their bare hands from under the reeds.

[photo - fishing by hand in the Little Pamir]

We recrossed the Chaqmak stream and headed west back towards our camp at Bozai Gumbaz. From a distance this section of the Little Pamir looks flat, but it is in fact a series of hidden undulations. At each new crest, we hoped to see a river flowing away from us to the west, but all we found was a few dried up channels that were not conclusive of what the maps showed.

And then we crossed over a small bank and there was the river we were looking for. From the moment we saw it, there was no doubt whatsoever but that it was the final confirmation of what we had seen on the maps. This was not just a meandering stream like the Chaqmak branch, but a river that had cut a 40-60 foot deep valley across the Little Pamir. Our eyes followed the river upstream and it clearly flowed directly from the Chelap source. There was no other possibility. The only way we could have missed this river on our trek eastwards earlier in the day was simply that we had crossed the Chelab above (or north of) the point where it splits into several streams, rather than below.

To be sure, we then followed the new river downstream. A few contributories came from the southern mountains. Quite clearly it flowed west and was indeed the Little Pamir River itself.

[photo - Chelab, the true source of the Oxus? ]

We had now confirmed that two tributaries of the Oxus (i.e., the Little Pamir River and the Bartang River) have the same source. This is an inherently unlikely proposition as it depends on two coincidences, both running counter to reasonable expectations. The first is that a single river, sourced high in the Wakhan Range between the Great and the Little Pamirs, would divide into two or more streams right on the cusp of the Pamir mountain watershed which then forces them to flow in opposite directions. The second coincidence is that the two resulting rivers, flowing in opposite directions for hundreds of miles, would actually rejoin at Roshan to form the Oxus.

In effect two adjacent molecules of water could descend the Chelab to the Little Pamir valley and separate, one east to Chaqmak and one west towards Sarhad. They might then meet again at Roshan some 240 miles downstream in one direction and 250 in the other.

With this in mind, we had reason to believe that we had found a new candidate, the Chelap River, to be the source of the Oxus, perhaps even the principal source. We wondered why past explorers in this region had not made this very argument.

Later that evening, Haji Ossman, the Kirghiz chief for the entire Wakhan region arrived at our camp. Although the next day he was to take a very different line, on this evening he was quite pleasant and only wanted to be seen and establish his authority. We invited him to share the fish that we had caught for dinner, but he declined saying he had to be back that night at his camp, which was more than two hours to the east.

H. Return to Ishkashim

Tuesday, July 10. We began the day with a last visit to the burial tombs near our camp at Bozai Gumbaz and by 8AM had started our journey back to Sarhad. Today we followed a slightly different route than the one that brought us into the Little Pamir; this allowed us to stop mid-morning at a site called Kasheh Goz where Haji Ossman, the Kirghiz chief for the entire region, had his settlement. Here we found 5-6 traditional yurts, a stone enclosure for their animals, a flat roofed hut of mud bricks, and a pitched roofed house for visitors. The latter was the only pitched roofed structure we saw in the Wakhan. Interestingly, one of the yurts had several small solar panels laying on its roof.

We were left for a considerable time in the visitor’s house and provided with bread, yoghurt, and tea. Again, the Kirghiz had an agenda other than social affability. They wanted to supply the horses and handlers necessary for our return to Sarhad. As with our previous discussions on this issue, they were not ready for immediate departure and decided to let the matter drop without confrontation. However, they did emphasize that, for future treks, they expected to get some business and that their domain started at Kasheh Goz and not further east at Bozai Gumbaz.

[photo - Khirgiz chief with his grandson during negotiations]

We were now back on those high mountain trails, narrow and precarious, that had been so challenging on the way out to the Little Pamir. Experience had not made them any easier to deal with and we had some harrowing moments. We camped again at Langar, but this time at a lower level some distance away from our previous site to avoid the chilling winds.

Wednesday, July 11. Today was our last full day on the trail and the experience was a familiar one: tremendous views, the Oxus far below, breathtaking and nerve-wracking trails. By late afternoon, we stopped at Shaur, only a half day’s ride from Sarhad, and camped by a rushing stream.

Thursday, July 12. The last morning’s ride was one of our most stressful. One of our horses stumbled slightly coming over a rocky patch very high up. Four of the Wakhis reacted instantly, grabbing his lead rope, tail, and flanks to stabilize the animal. The danger passed in an instant, but for the rider peering over the edge into the abyss, that second or two seemed like a lifetime.

We climbed up and over the Daliz Pass and, by lunchtime, we were on a long, sloping meadow leading down to Sarhad, several easy miles in the offing. The vista was superb, perhaps because we were now out of the mountains, but also because here the course of the Oxus River opens out from the narrow ravine it had been following into a broad valley several miles across. We stopped to have our last trail lunch, to enjoy the view, and to relax now that the most arduous part of our journey was behind us. After eating, we made group photos and took individual shots with our Wakhi colleagues from whom we were about to depart.

By 2PM we were back at the guesthouse in Sarhad. Sakhi paid the Wakhi, including the $15 per head bonus we had decided on. There were final photos and suddenly they were gone and it was all over. We bathed in the thermal baths again, repacked, and later enjoyed a bottle of cheap Russian vodka procured from the guesthouse manager.

Friday, July 13. We were off early for the drive back to Ishkashim, determined to complete this trip in a single day instead of the day and a half it had taken us on the outward bound journey. There were some poignant moments in the first few miles after leaving Sarhad. We were stopped three or four times by a few of our Wakhi horse handlers who had come out to the road from their nearby villages to say a final good-bye. We took a few moments with each of them, got out of our vehicle, gave them a big hug in the Afghan tradition, and thanked them again. It was very touching. One of the horsemen even invited us for tea is his home and we were afraid of causing offence if we didn’t accept. However, Sakhi explained that we had to press on and he understood.

Despite the challenging road, we made good progress throughout the day. Trouble struck at 4PM however; the Nissan 4 wheel-drive vehicle lost power when one of the fan blades sheared off and possibly punctured the radiator. After a failed attempt to tow it to the next village (about three miles), we realized there was no choice but to leave the Nissan there overnight and have it sorted out the next day. The driver, Hakim, and one of our guides, Shafi, stayed behind and the rest of us piled into the Toyota pickup to press on to Ishkashim, about 3 ½ hours away.

We made a short stop at the nearby village which was ¾ of a mile off the road. At first glance, it appeared almost abandoned, but the horn soon attracted a few teenage boys and a village elder. Sakhi explained the situation with the vehicle breakdown and asked the villagers to take food and water to the two colleagues we had left behind --- they readily agreed to do so. Such kindness to travellers, even if they are complete strangers, is commonplace in Afghanistan; certainly nothing like this was likely to happen in the UK, Ireland, or America.

We drove on as darkness approached. Every 100-200 yards we had to slow down for a ditch crossing the road. These were part of a network of small irrigation canals which uses the momentum of the streams rushing down the mountainsides to circulate water in the fields. All afternoon we had seen men and boys with shovels on the roadsides or in the fields who were tending these canals --- rechanneling the water to different fields or repairing the ditches where they had collapsed or filled up.

The road was rough and in the back seat of the pickup we were cramped and continually thrown against one another and the sides of the vehicle. For all the discomforts, there was one constant, the Oxus. The road, more or less, continued to follow the river and as we finally approached Ishkashim, far below we could still see the Oxus shimmering in the moon and starlight, flowing relentlessly westward.

Just before 8PM we arrived at Wafai’s Guesthouse where we had eaten lunch the first day after crossing the Afghan border. He greeted us with great enthusiasm despite the late hour and our unannounced arrival and was soon rushing about preparing rooms and food.

I. The Oxus Northward

After crossing, with some problems, back into Tajikistan we followed the river north to Khorog, the administrative capital of Gorno-Badakshan. It is a grand and magnificent journey following the river as it cuts a swathe through the mountains of Central Asia.

We reached Roshan / Vomar north of Khorog and were able to make a visual assessment of the relative volume of the Oxus that we had followed and its sister, the Bartang River, that had come all the way on the northern route from Lake Chakmaktin via Murghab.

The argument about the true source of the river hung in the 19th century to a great extent on which was the greater flow at this point, the Bartang or the Panj Oxus. Ney Elias considered it to be the latter, whereas Trotter and Dunmore’s informants favoured the former. We were not able to form a view other than observing that there did not seem to be much to choose between them.
We were not able to take more than a cursory view of the conflicting claims.

A full report on this very beautiful part of the trip can be obtained from


As noted in the earlier discussion of our trip to Lake Chaqmak and the Chelap River on the Little Pamir, it seemed to us that the Chelap should be given consideration as the primary source of the Oxus River. After all, this stream splits just above the western end of the lake and then feeds two tributaries of the Oxus, the Little Pamir River and the Aksu / Bartang River. At the time of our visit, we wondered why no previous explorer had thrown this argument into the debate.

Back at home, we continued to explore the literature concerning the primary source of the Oxus. This eventually led us to the Report on the Proceedings of the Pamir Boundary Commission of 1896. In addition to delineating the border between Russian territory (today’s Tajikistan) and Afghanistan east of Lake Syr Kol, the members of this commission also sought to “clear away the mists that have hung so long around the sources of the Oxus”. They argued that since the glaciers and snowfields sitting on top of the Nicolas, or Wakhan, Range between the Great and Little Pamir feed (i) streams flowing down the north side of this range into Syr Kol and then on into the Pamir River and also feed (ii) the Chelap River flowing down the southern side through the Burgutai Pass or Gorge and then on into the Little Pamir River and the Aksu / Bartang River (via Lake Chaqmak), these glaciers should be regarded as the true source of the Oxus River. In their view, nothing else could compete with a source that single-handedly feeds three of the tributaries of the Oxus.

Thus, the position of the boundary commission encompassed and expanded on our own thoughts that perhaps the Chelap River should be considered the primary source of the Oxus. The commission took the argument back up the mountain, so to speak, to the glaciers and snowfields that feed not only the Chelap, but also the streams on the north side of the Wakhan Range flowing into Syr Kol and the Pamir River.

It is an intriguing thought and certainly worthy of consideration. Of course, Curzon attacked the commission immediately because they had directly challenged his own celebrated opinion that the true source are the ice caves at the end of the Wakhjir Valley. Thus, the issue remains open to this day and has never been conclusively answered. As far as we know, Curzon’s report and that of the Boundary Commission are the last writings on the matter. This is not surprising: with the conclusion of the 1895 boundary agreement between the Russians and British, Afghanistan’s northern border was settled once and for all and the location of the main stream of the Oxus was no longer politically important. Furthermore, the Great Game between these two nations was in its waning stages and would die out completely in the early years of the 20th century as momentous events, primarily the First World War and its aftermath, diverted the world’s attention and made issues such as the source of the Oxus seem trivial.

* * * * *

Today the Oxus still roars out of the Pamirs, but now dies out in the Kyzyl Kum desert well short of its former terminus, the Aral Sea. Its glory days were in ancient times, when it separated Central Asian warriors, and in the 19th century when questions over its position and sources energized the Russians and British.

Nonetheless, there are still exciting adventures to be had in this region. Three which occur to us include:

• Travel by horseback up the bed of the Wakhjir River in an attempt to reach the ice caves as Curzon seems to suggest is possible;

• Follow the Chelab River up the Burgutai Pass to its glacier and snowfield source atop the Wakhan Range and investigate further whether this area really could be the true source of the Oxus; and

• Follow the footpath on the Afghan side of the Oxus from Ishkashim all the way to the northernmost point of the river. This journey is the opposite of the one we did on the Tajik side, but is not described in detail in this paper. On the Afghan side there is what looks like a tough but beautiful and very rarely travelled footpath. It is truly exciting journey for an adventurer and their donkey. It is most unlikely that any European has ever done it.

Appendix 1 - Research & Planning

The following sources were valuable in researching and planning our journey:

Reading Material

The Pamirs and the Source of the Oxus -- 112 years on, Lord Curzon’s report to the Royal Geographical Society, written in 1894, remains the best guide book to the area in terms of detailed information regarding passes, rivers and access. It is available in its original, but has also been republished (however, without the valuable map) by Elibron Classics (

Journey to the Source of Oxus -- Lieutenant John Wood’s book covers his groundbreaking 1843 expedition for European travellers from Kabul to Wakhan. This book has also been republished by Elibron.

The Kirghiz and Wakhi of Afghanistan – Adaptation to Closed Frontiers and War, by M. Nazif Shahrani, University of Washington Press (1979, reissued 2002)


• Matthew Leeming ( -- Matthew has co-written the Companion Guide to Afghanistan with Bijan Omrani ( and has travelled in the western end of Wakhan.

• Julien Dufour ( -- Julien had the most comprehensive online review of routes and terrain in the Wakhan.

• Robert Middleton ( -- Robert’s website has the most comprehensive information about Gorno Badakhshan, the Pamir region of Tajikistan. He is now collaborating with Bijan Omrani in writing the Companion to Tajikistan, which will be published by Odyssey.

Logistical Assistance

• The Great Game Travel Company -- Michael Davis and his colleagues in Kabul and Badakhshan arranged the necessary guides and supplies for our journey into the Wakhan. They had previously sent parties into this region, but not into the Pamir-i-Wakhan over the Wakhan mountain passes, so there was extended discussion about time and viability of destination objectives. (

• Mountain Societies Development & Support Program -- Through the introduction of Robert Middleton, we engaged Ergash Fayzullobekov of Khorog in Gorno-Badakhshan to be our guide in Tajikistan. Ergash works for the Mountain Societies Development & Support Programme, an activity supported by the Aga Khan Foundation. (

Other Encouragement and Advice

• The Aga Khan Foundation – AKF’s associates and employees, who were to play a significant part in our journey, were helpful and strongly supportive. It was reassuring to know that they were available for help if necessary via their operations in Ishkashim, Afghanistan and Khorog, Tajikistan.

• Georgiana Cecil, who made one of only two legal journeys that we knew of from the Wakhan into Pakistan, also provided encouragement and advice particularly about the severity of the terrain. Being something of an acrophobe, Bill was worried about this and she provided reassurance, although as it happened the difficult terrain we encountered was beyond where she had turned south to cross the Baroghil Pass.

Appendix 2 – Wakhan Expeditionary Team

Category Name Age Home
Clients Bill Colegrave 59 London
Anthony Kitchin 56 Sligo
Dillon Coleman 56 Raleigh
Guides – Kirghizstan & Tajikistan Ergash Fayzullobekov Khorog
Hossein Khorog
Guides - Afghanistan Ghulam Sakhi Danishjo 37 Shaikh Ali District of Parwan Province
Mohammed Nadir 49 Khosh District of Badakhshan
Mohammed Shafi 53 Darwaz District of Badakhshan
Cooks Khaleq Dad 24 Sarhad
Mehraban 25 Shelk
Horsemen Mirza Mohammad 29 Ptukh
Burhanuddin 22 Panjshir
Niaz Bai 50 Chelkan
Musafer 25 Ptukh
Mohammad Hassan 23 Chelkan
Jama 23 Sarhad
Saifal 50 Sarhad
Shahpoor 25 Chelkan
Satar 29 Neshtkhawr
Menber 22 Ptukh
Murad 45 Ptukh
Hamidullah 30 Chelkan
Shanbe 22 Ptukh
Sayed Qassem 30 Ptukh