Click on image for list of participants
Ennedi Expedition, Chad
13th - 30th January, 2016

We have completed a superb exploratory voyage to the Ennedi, visiting the extraordinary engravings of Niola Doa, also taking in a good measure of the rock art sites and spectacular landscape along the South Western edge of the plateau. Like on past Chad expeditions, ground logistics were provided by S.V.S..

View Slide-show

Our party met at the Novotel in N'Djamena, arriving the night before trip start with one or the other of the fairly limited flight options. The airport was a pleasant surprise, the obstacle course through the construction site experienced a year ago was replaced by a small but neat and functional modern terminal building.

Day 1. – N'Djamena - Camp before Bitkine

After all the late night flight arrivals we had a comfortable lazy morning while Andrea sorted out the necessary paperwork for our journey. After loading the cars in the garden of the hotel, we were ready for departure by mid-morning.

While the most direct route from the capital to the Ennedi is via the Bahr el Ghazal, our objective was to spend as much time as possible in the Ennedi, and like in many places in Africa the shortest is not necessary the fastest. We took the approximately 900 km surfaced road towards Abéche, a detour of about 300 kilometres, but still taking less time than the direct route (not to mention the comfort). With several of us having traveled already along the dusty Bahr el Ghazal track, we did not take much convincing to take the longer but faster way. We crossed the city and set out on the road, making good progress among the entertaining assortment of traffic. We made a short roadside stop to have our lunch of pre-packed sandwiches, then continued East at a steady pace.

Initially we traversed a dull flat and uninspiring shrub country, but by mid-afternoon we reached a much prettier area of typical African landscape with small traditional villages among scattered granite knolls and denser vegetation.

We stopped at the village of Bolongue to buy some firewood. It was a much neater place than anything I've seen before in Chad, with a well kept main road and basic but clean and orderly houses alongside.

After another forty kilometres we reached the impressive and very picturesque range of higher granite hills before the town of Bitkine. We have covered about With the sun rather low behind us, we turned off the road to search for a suitable camping spot, to settle in before sunset.

After some searching among the low hills and bush we found a flat sandy patch that appeared to be away from any village. The landscape and vegetation around our campsite was a bit out of place for a Saharan expedition, it had more than a passing resemblance to the Erongo Hills in Namibia.

Day 2. – Bitkine - Camp after Abéche

As we had our breakfast at dawn our camp was surrounded by a large herd of cattle tended by a group of curious children - obviously a village was rather close by, just out of sight among the low trees and shrubs. We packed camp, and were ready to depart soon after sunrise.

Getting back to the road, we continued towards the East, passing close by the most prominent of the granite hills and stopping for some photos at the most scenic spots.

After passing the nondescript Bitkine we continued among the granite hills and small villages for sixty kilometres, until we reached the larger town of Mongo. The town itself was hardly an exciting place, but in the middle of it completely out of place, there was a brand new flashy Total gas station complete with a functioning ATM and a store that was equipped with all sorts of goodies, including some top end vintage french Champagne.

Soon after Mongo we reached the impressive Abou Telfan mountains, a granite range rising to 1300 metres, much higher than the small hills we passed up till now. Again the landscape was very reminiscent of Namibia, complete with huge rounded granite boulders littering the slopes like on the Brandberg. During a toilet stop, I could not resist the temptation to climb to the closest boulder with a good shelter underneath. It was empty, but nearby on the top of a smaller boulder I found several man-made grinding basins, and the ground was littered with quartz waste flakes - clearly an ancient habitation site. I will not be surprised if a future survey does find some rock art sites in the area.

After the mountains the country flattened again, and we continued among an almost continuous stream of small villages. We passed a large cistern holding water from the past rains with livestock converging on it from afar. We were still in predominantly cattle country, but here we also saw a herd of camels, the first ones since leaving N'Djamena.

We again passed a series of picturesque granite hills with huge rounded boulders, surrounded by the small thatch roofed villages. Cattle crossed the road constantly, and we passed a steady stream of trucks coming the opposite way, loaded to the brim with raw cattle hides, one of the main produces of the region. By midday we reached the town of Mangalmé, an incongruous mix of tradition and progress with brand new solar powered streetlights lining the road, likely the result of some aid project with a touch of vested interest (why this town and no other is anybody's guess). Soon after the town we pulled off the road to take a lunch break in the shade of a cluster of trees.

After our lunch stop we soon reached Oum Hadjer, the last major town before Abéche. We crossed the bridge over the almost completely dry Oued Batha, the Northernmost larger river that does actually flow for parts of the year after the rainy season, and continued Eastwards over progressively drier looking country. There were still cattle about, but we encountered more and more camels, driven in large herds.

By mid-afternoon we reached Abéche, signaled by a prominent new mosque in the western suburbs. It did not appear to be a very inviting place, as we had enough fuel we did not stop to linger but continued straight through town, leaving the tarmac road and turning North towards Fada.

We managed to cover another 25 kilometres before sunset caught up with us. Spotting a cluster of low granite knolls ahead, we soon turned off the road to make camp behind the hillocks.

Day 3. – Abéche - Camp before Ennedi

This day was to be another long drive, we were hoping to reach the edge of the Ennedi by nightfall. With an early start we packed up camp and set out along the track leading north. Being unsurfaced, it was much slower going, but still better than the track than runs through the Bahr el Ghazal. As we progressed, the landscape changed very quickly from the dry savanna we traversed from N'Djamena to Abéche to a flat grassland with just a few trees and shrubs. We passed another major well by the roadside, with several mounted horsemen corralling the herds of cattle and sheep competing for the water.

Until the town of Biltine about 75 kilometres North of Abéche the land was populated continuously, with herds of cattle and the occasional small fields and a steady row of sometimes quite spectacular granite outcrops.

After Biltine we reached the flat grasslands, villages became scarce and cattle almost completely disappeared, to be replaced by camels everywhere. We continued on this flat country for nearly 150 kilometres before stopping for lunch among a grove of larger acacias before the town of Kalait.

After lunch we entered Kalait, a rather new and unattractive boomtown that is one of the main trading centres for Libyan goods. For practical purposes this is the border of Chad proper (like Faya in the West), further North it is a frontier zone under tribal control. The big Libyan trading trucks are free to cut across the desert till here, the first larger place with customs formalities. As a consequence there is plenty of fuel and other goods available, we filled our tanks and topped up our supplies before moving on. A particularly amusing feature of the town is the "public water supply". The main well is about 2 kilometres out of town, and a steady train of donkeys trod the route, fetching the water in large plastic guerbas.

Once we left Kalait we reached the edge of the uninhabited desert, though still there was some meagre grass and other vegetation about. We encountered the first gazelle, then many more, none of whom were particularly concerned about the approaching cars. Apparently they were not accustomed to being hunted, unlike the ones we have seen in the Tibesti.

We drove for about 80 kilometres Northwards, keeping a lookout for any suitable campsites as the sun was getting rather low. Just as we were ready to make camp on the featureless plain, a pair of low granite hills appeared on the horizon. We just reached them before sunset, startling a golden jackal which chose the same spot for the evening. It rapidly disappeared among the granite boulders, while we all scrambled to make use of the last light of the day and the first proper desert camp.

From the top of the low hill above the camp, it was just possible to make out in the fading light the first ramparts of the Ennedi, about 50 kilometres away.

Day 4. – Camp before Ennedi - Fada - Mourdi Depression

As we were having breakfast and packing camp, Andrea made a very interesting discovery on a flat granite outcrop a few hundred metres away while searching for a good toilet spot. The rocks were full of deep, well made grinding basins. Curiously I could find no surface ceramics or tools around the rock, the only trace left by the ancient inhabitants were these basins, very similar to the ones seen at Abou Telfan mountain two days earlier.

We quickly covered the remaining short distance to the edge of the plateau, passing the first eroded sandstone towers and the last outcrops of the ancient granite basement. Suddenly the landscape changed as we passed the first ramparts, all of a sudden the grey rocky terrain was replaced with golden sand everywhere.

We made a stop at one of the first sand engulfed sandstone outcrops, which had a large natural arch at the rear end. It was quite a challenge to gather our party at the end of the short break, with everyone dispersing to frolic in the sand and the first proper desert landscape.

Only 40 kilometres remained to Fada, the capital of the Ennedi region. The town was a surprisingly neat and orderly place compared to the other Northern settlements in Chad. It is centered on the old French fort, with a large tree lined main square which was very quiet when we reached it before midday. We did not linger for long, we fueled the cars and reported at the Prefecture, then we were again on our way towards the North.

The track leading North towards Ounianga and the Mourdi depression was surprisingly good. The pass leading up to the Ennedi plateau was actually surfaced, it was bizarre to drive up a stretch of perfect fresh asphalt with just the dusty piste before and after the ascent. We did not go far, 15 kilometres beyond Fada we stopped for our lunch break at the foot of a sandstone outcrop with a large natural arc at its end, guarded by a rather friendly family of camels.

Not very surprisingly there was some rock art at our lunch spot. At the edge of the arch there were a couple of crude camel engravings, but on the far side of the rock we also found a panel of paintings, very faint but apparently a herd of cattle. I have only found out after returning and looking at the photos with DStretch that we have found the first of the very characteristic Ennedi mounted camel paintings.

After lunch we continued across the top of the plateau. Here the landscape became less dramatic, presenting an eroded but generally flat rocky country, like the tops of all the Saharan sandstone plateaus. The only difference was the noticeably denser grass and tree cover.

The crossing of the plateau took much less time than anticipated, after about an hour's drive we reached the broken sandy country with the familiar sandstone towers marking the Northern edge of the Ennedi.

After crossing a couple of sand filled valleys we reached the Northern edge of the plateau, with a steep sandy pass leading down into the flat Mourdi depression beyond. As we descended in the soft sand I did wonder about how will we get up on the return, but it was quickly forgotten as we encountered a herd of gazelle stirred up by a passing pickup truck.

With the sun low, we only drove for a short distance beyond the bottom of the pass, to be sufficiently away from the traveled track. We found a pleasant sandy patch for our campsite, bordering the foothills of the plateau overlooking the green valleys of the Mourdi depression.

Day 5. – Mourdi depression - Diona

As we packed our camp in the morning, we noticed that one of the trees nearby was not an acacia as it appeared in the dusk, but a huge Calotropis procera. It is one of the commonest shrubs in the Sahel and Central Sahara, but I have never seen one grow to this size.

The Mourdi depression is a 250 kilometre long and about 50-60 kilometre long valley bordering the Northern side of the Ennedi plateau. On the North it is flanked by the Erdi plateaus, and several large watercourses drain into it from the South, supporting large trees along the riverbanks. Between the wadis the ground is flat and rocky with the occasional sandy patch and dune fields, a rather difficult going as we soon found out heading towards the village of Diona about 25 kilometres North of our objective, Niola Doa, and a good 120 kilometres west of our camp. As we bumped along the rocky terrain, we passed several ancient cemeteries with many round tumuli dotting the landscape.

Mid-morning we reached the Oued Horta with the well of Tébi in the middle of the watercourse, where we stopped to re-fill our water canisters. It is a permanent well marked on the IGN map, one of many dotting the wadis along the central part of the Mourdi, permitting a number of semi-permanent nomadic settlements, most of which are too small to be noted on the map.

Beyond the well the terrain became sandy, we passed some proper barchan fields before reaching another large wadi with many trees half submerged in the yellow sand, dotted with hamlets of a couple of houses. It was very difficult soft going, while we managed to avoid becoming stuck, it was almost constant first & second gear. Our route followed a well traveled track along the wadi banks, leading to the village of Diona. However at some junction we managed to miss the right fork and it was a good 20 kilometres past the proper longitude that we started having doubts about the direction.

As it was about lunchtime anyway, we stopped in the shade of the largest tree we could find, in a very pretty area of golden dunes dotted with acacias and balanites trees. During the break we consulted the maps and satellite imagery, and it became clear that we should have turned South at a small village we passed about half an hour earlier.

We have continued East beyond the right track for about 30 kilometres, far enough that the Southern tip of the Erdi Ma was visible from the top of a high dune near our lunch spot. However the detour was not a major setback, we were only about 15 kilometres to the North West of Diona, lying somewhere between us and the Ennedi plateau lining the Southern horizon.

We decided not to re-trace our tracks but to cut right across to Diona over the undulating sandy country. This was not without some risk, as none of us passed this way before, but all turned out well, the terrain was easily passable and after half an hour we could spot the radio mast marking the small military post on the outskirts of the village. We reached Diona by mid-afternoon, immediately going to the Préfecture to see the head of the village and collect a local guide.

It turned out that the village chief was out and only due back in the evening, so we proceeded to the well in the riverbed in the middle of the village to fill all our empty containers, as there was no water to be found in the vicinity of Niola Doa.

Once we had our fill of water, we drove a few kilometres South to make camp just before sunset among a very picturesque grove of acacias among the dunes, while Andrea returned to the Préfecture to sort out the formalities.

Day 6. – Diona - Niola Doa

In the morning we only had to collect our guide, then were ready to go. For a while the track followed the West bank of the wide Oued Baragué, then crossed the very soft riverbed and turned up towards the rocky flanks of the Ennedi plateau. I expected to approach Niola Doa in the riverbed, but our guide explained that the sand was too soft and impassable for cars and beasts alike, the only possible way was across the rocky ridge, then down into the valley again to cross over to the area of Niola Doa. It was a rather slow bumpy going, sometimes at a walking pace or slower, lengthened by the need to stop to pump up our tyres which were let down to cope with the soft sand in the Mourdi.

With slow but steady progress we reached the col overlooking the Oued Baragué again by mid-morning. As we stopped to take some photos, we were passed by a large herd of sheep tended with two shepherds, soon followed by an older man leading two camels heavily laden with grass is sacks, collected on the pasturages of the plateau. He showed no surprise at our presence, we exchanged greetings then both went our separate ways.

We had to descend again and cross the riverbed, which was a bit tricky with the tyres pumped hard for the rocks, but all cars managed without any mishaps. Now we only had a short six kilometres to go till we reached the sandy plain of Niola Doa, the main objective of our trip. We homed in on a cluster of rocks at the Southern end of the plain, and all of a sudden we were standing in front of the famous and unique panel of engravings, not quite believing yet that we were really here.

The panel was already in the shade, we could see that the best light is in the early morning, just after sunrise. This was no major concern, as our plans were to spend as much time in the area as necessary to locate all of the known sites and photograph them in the best possible light. As none of us, including our guide, knew the precise location of all, we allotted three days for the area, allowing for plenty of time for searching large areas if necessary. As many of the Eastern sites were within walking distance of the "French" panel, we decided to make camp behind the adjacent cluster of rocks.

The "French" panel was discovered by lieutenant Courtet in 1954, and was visited and published by a group of French geologists a year later (hence it's name). Another panel of the same type of engraved figures was noted by Courtet a couple of kilometres further North, it was published by the Austrian ethnographer Peter Fuchs in 1956 (the "Fuchs" panel). In 1993 an Italian party re-visited the area and found a major new panel of engravings unreported by any previous explorers, though strangely it was in plain sight of the French panel. Immediately after parking our cars we set out to find this "Italian" panel, which did not prove to be difficult, the cluster of rocks were clearly visible a mere two hundred metres away, with the engravings clearly visible as we approached. This panel was partly in sun, partly in shade, looking at the sun angles a mid-afternoon time was ideal here. With a little search we also found the two unfinished figures not far from the main panel.

We returned to the cars for lunch and a break, which I used to scout the low hills to the south west of the French panel, where Choppy placed a small shelter called "Abri des trois vaches" (shelter of the three cows). There was a conspicuous cluster of rocks on the top of a ridge about 500 metres from the cars, I was fairly certain they must be the site. On reaching the hilltop, I found a shelter that has a single strange, undecipherable painting on the ceiling. Even DStretch does not help much, one can only see the bizarre thing clearer, but that does not make it any more comprehensible.

On the far side of the ridge crest there was another, more spacious double shelter, with the left part a tunnel leading through the rocks. There were a few crude engraving on the bottom, and above the entrance I could spot the faint outlines of three cattle (DStretch shows that they are in fact four). The identity of the site appeared clear, there were further faint paintings of humans and cattle along the shallow shelter to the right. It was only at home, checking positions and publications that I realised something was amiss. None of the scenes matched the copies form "Abri des trois vaches" (nor any other published shelter), and I have traversed a little North of West, rather than South west as intended. Apparently I have stumbled upon an unreported site.

At the end of the break we returned to the Italian panel, which was now in a bit better light. The main panel with the characteristic large figures was still not perfect, but the light was ideal for the incomplete figures on the adjacent rock to the right. We made a mental note to return the next afternoon two hours later for the best light, and then set out with two cars towards the Northern part of the Niola Doa plain.

Our first stop was a large shelter, known to our guide and clearly visible from a distance under a large conspicuous boulder a little over a kilometre from our camp. There are some engravings on the side of the rock and around its base, including a possible "Niola Doa type" figure. However the main attraction is the shelter (in reality a tunnel under the boulder) known as "Abri au Bidon", the ceiling of which is full of paintings in the characteristic Ennedi cattle pastoralist style. With the best effort I could not identify with any degree of confidence the vessel or container after which the site was supposedly named by the Italian party in 1993.

We continued to the Northern edge of the plain, towards where our guide recalled the location of the Fuchs panel. We found nothing at the outcrop, but a little to the West we saw a double shelter, and took the opportunity to check it out. It was definitely a worthwhile exercise, as in the left shelter we found the panel with a female figure wearing a strange headdress reported by Guido Faleschini in 1994, but the precise location became lost with his passing a few years ago. The right shelter was even more interesting, on the wall we noted two very faint plump figures which did not appear in any publication. One was an unusually fat male figure of what Bailloud calls the Tamada style, with the characteristic matchstick head surrounded by a faint orange coiffure (only visible with DStretch), holding bow and arrows. The other figure was of the Hohou style (a little resembling the Niola Doa figures, minus the large round head), the only one I'm aware of in the Northern Ennedi.

With the little deviation our guide found his bearings, and on the second attempt led us straight to the Fuchs panel, just a few hundred metres to the East of where we first looked. This panel appears incomplete, parts of the figures are just completed in outline with no body decoration, but the adjacent rock with a panel of fine cattle is very significant, as there is a clear association of a Niola Doa type figure with the cattle (squashing any speculation that they could be much more ancient than the cattle pastoralist art of the region). The light here was very unfavorable, but now we knew the location to return to the next morning.

We knew that there was a panel of engravings with large cattle no more than 200-300 metres to the North west of the Fuchs panel, however neither our guide nor Andrea knew its location. We spread out to search the low rocky plateau above the Fuchs panel, and it did not take more than 30 minutes for one of us to stumble upon the site on the plateau top.

Near the cattle engravings we found a very strange geological phenomenon - a perfectly round hemispherical basin in the sandstone, apparently the remnants of some concretion the like of which I have never encountered before.

We still had about an hour and a half till sunset, we decided to give the "lost sites" a try. Two panels of Niola Doa type figures (including one of the finest) were seen by sergeant Bousso in 1964 to the West of the principal site (partially published by Paul Huard in 1966), however no one saw them again until they were re-located by Alec Campbell and David Coulson in 1996. We had a good idea about their general location, so made a quick dash to the Western part of the plain. As we approached the presumed location, there was a large towering rock formation, and we found the lesser of the two panels on its side in perfect light.

Unexpectedly the shelters around the base of the towering rock were full of paintings. We only made a brief survey, as our plan was to return here to camp the following day.

A short distance beyond the rock tower we found the fine panel of engraved cattle also reported by Bousso. While there were no human figures, these cattle were clearly the same as those associated with the Niola Doa type figures at the Fuchs panel.

While we were busy with the other sites, our guide easily found the main panel of the locality on the side of a fairly inconspicuous rock surrounded by a low dune. It was already partially in the shade, we did not go close to leave the sand undisturbed for next days' photo session, but even from the distance we could see that indeed it was one of the finest of the Nola Doa engravings.

Among the rocks beyond the main panel, we found two shallow shelters, both with large and well preserved panels of paintings. Clearly there was much left to see and photograph for the next day when we were to return.

Before returning to camp we made a little detour, to try to find the several Niola Doa type figures located somewhere on a flat rock surface to the North East of our camp. Andrea saw them about 15 years earlier, but he only had a very vague idea, nevertheless just before sunset we stumbled upon a panel of engravings with some very eroded Niola Doa type human figures. They were not the ones we were looking for, but those must have been close, and now we had a bearing to walk to from camp at sunrise for the best light. It was a very productive afternoon exceeding all expectations, having successfully re-located practically all of the important sites we were looking for, so we could spend a leisurely next day photographing them in the best light.

Day 7. – Niola Doa

We were up at dawn to capture the French panel in the best light. As the sun rose, it provided a perfect contour lighting, bringing out the finest details of all the figures.

After a lengthy photo session, we started to walk to the flat panel we saw the evening before, less than a kilometre away. Near half way we passed a small panel of engravings, mostly cattle and a few goats. Nearby there were other markings in the rock, made visible by the low morning light. It was something I have read about many times, but never witnessed: grooves created by a glacier during the Ordovican, when most of what is now the Sahara was under the south polar ice cap.

The panel we found the previous evening was now in a perfect light. There was a large Niola Doa type figure, unfortunately very eroded with the head completely missing. It is surrounded by several smaller figures, including one of the rare type of which two may be observed on the French panel, interpreted by some as drummers. It was clear however that these were not the panel we were looking for. Subsequently Roberta Simonis confirmed that indeed we did find a new, previously unrecorded panel.

While the rest of us were photographing the panel, Ursula wandered off towards the North, and finally found the panels reported by the 1993 Italian party. These engravings were much better preserved, we could photograph them in ideal lighting conditions. The four large Niola Doa type figures are surrounded by a multitude of lesser figures.

About two hundred metres to the North east we found another area covered with engravings, also reported by the 1993 Italian party. Here there were no Niola Doa type human figures just cattle, mostly in crude outline with a few simple human figures.

Walking back towards camp we found a few more scattered engravings near our first find, while nothing dramatic or spectacular, probably these too were unrecorded. One panel was clearly modern, as a pair of cars may be recognised among the animals and humans.

By the time we returned the French panel was in very different lighting, with the sun much higher. Some details were in perfect contour lighting, especially the little "drummer" figure at the lower right of the panel.

The Italian panel was still completely in the shade, we took some photos under these conditions, in case some details stand out better without sunlight.

Following the sun, we drove to the Fuchs panel, reaching it just as the sun was shining nearly edge-on over the main panel. We had to conclude that while some details stood out better, due to the many irregularities in the rock the panel was better visible overall while in the shade rather than in direct sunlight.

We climbed up again to the panel of large cattle, which was now mostly on the sun. Here too we found some details to be better visible, but not a significant improvement.

While we were busy with the sites, Somein, the most inquisitive of our drivers wandered off, and returned with the news that he found a small site with paintings in the adjacent valley just a few hundred metres away. We all set out to see it, it turned out to be the shelter named "Petit abri" by Choppy. Along the way we also passed a fine panel of camel period engravings.

We only had a single major site left on the Eastern part of the plain, the "Abri de Colonne" (shelter of the column) just a short distance to the South from the Fuchs panel. We returned to the cars and drove to the site that was readily visible, the largest rock on the plain with a deep hollow in its side, adjacent to a very prominent column-like rock.

The "Abri de Colonne" is the largest shelter with paintings on the Niola Doa plain, mostly executed in the characteristic Ennedi cattle pastoralist style. There are paintings with varying degrees of weathering at all more sheltered locations of the large rock, but the best are in a little recess in the middle, where a protected ceiling section contains some very well preserved scenes.

A natural arch divides the shelter along the Eastern face of the rock into two sections. At the left of the arch there is just a single panel of paintings depicting some giraffe, possibly somewhat older than the rest of the pastoralist paintings.

Having completed our round of the Eastern part of the Niola Doa plain, we returned to camp for an early lunch. I have again used the time to survey some clusters of rocks a few hundred metres to the South of our camp, and did come across a shelter with some rather weathered paintings.

I did not recognise much on the spot, but later the shelter was identified as what Choppy called "l'abri de l'animal gravé" on account of a rather archaic looking engraving of what appears to be a cattle with forward curving horns, subsequently covered by later pastoralist paintings. Choppy placed the site 1 kilometre to the south of the French panel, but in reality it is just 350 metres to the South west.

After lunch we packed up and moved camp a few kilometres to the west to the edge of the rocks hiding the "lost panel" that we found the previous afternoon. The new campsite was much nicer than the previous one, with good shade, several trees and a wide view over the dramatic landscape of the Western plain littered with solitary and clustered rock towers.

We had the afternoon free to explore the area, with plenty of time left until the main panel was to be illuminated in optimal lighting. We dispersed in all directions, some going for the rock art, some just for the landscape. I started off at the big rock with the second smaller lost panel, where there were many more paintings than what I've found during the cursory look the previous evening.

Moving on to the main "lost panel" of Bousso, it was just coming out into the sun, with some of the main figures in perfect contour lighting, bringing out all the fine details. However it was still a couple of hours until the whole panel would be well illuminated.

Adjacent to the main "lost panel" towards the plain, there is a large rock with a shelter underneath, with a large panel of well preserved pastoralist paintings, which to my knowledge have not appeared in any of the publications.

Roberta Simonis informed me that there are two more unpublished shelters with paintings near the southern edge of the plain. I set course for a likely group, passing a small watercourse with plenty of green vegetation, including a plant with small red flowers and little green leaves that I have not seen before in the Eastern Sahara. It was not difficult to find the first shelter, it contained mostly unimpressive weathered cattle, but there was a very fine scene of a camel rider with a conical head, a type appearing sporadically at several places around the Ennedi.

The other site was found very close, on the far side of the same rock island. It was a more complex site with several panels, starting with a shallow shelter with mostly cattle scenes.

The more interesting panels were on a thin rock wall nearby that appeared as if it were about to fall down if viewed on the edge. Along its side facing the shelter there are numerous faint but very fine scenes of human figures in the Tamada style.

The unexpectedly quick finding of these sites left some time for surveying the broken country further west, at the foot of a medium-sized hill closing off the lain towards the south. We teamed up with Andrea to search the rocks, we found nothing on the lower level but we spotted a largish and promising shelter in the hillside, at the edge of a terrace at an elevation about 25-30 metres above the plain. Climbing up, there was a commanding view from the terrace over the entire western plain.

There were indeed paintings in the shelter on the rear wall, though their state was disappointing. At first only two red animals could be discerned, the first apparently without a head. After some searching a very faint scene with an archer and some animals was also found plus some traces of paint which do not present a coherent picture even with dStretch. However the other figures are quite interesting, especially the archer, which is very different from the ones commonly encountered in the Ennedi, and appears much older than the pastoralist paintings on the plain below.

As I descended from the hill and made my way towards the camp, I passed a couple of low acacia trees which were occupied by a flock of small birds who created a huge rabble, their chirping audible from a long distance. After our return Koen de Smet helped to identify them as fulvous babblers (Turdoides fulva), a close cousin of the jungle babblers which inhabited our Delhi garden. We called them the real angry birds on account of their looks and the huge rattle and commotion they could make.

Mid-afternoon some of us returned to the Italian panel with just one car to photograph it in the best light conditions. Along the way we met a flock of crowned sand grouse (Pterocles coronatus) who unusually did not appear very concerned about our approach, we could get quite close before they took off.

The Italian panel was now in perfect light, with all parts evenly illuminated by the sun. We spent plenty of time photographing all the details, though not forgetting that we were yet to have a similar session at the "lost panel".

While we were busy taking the photos, Ursula found a very strange, dense black piece of rock that appeared to have a molten crust on one side, and an iridescent luster with an angular crystalline structure on the other fractured sides. There was much excitement that she may have found a meteorite fragment, but there was no way to tell in the desert, so it was packed up to be examined later.

By the time we returned the main "lost panel" was also in perfect light, a lengthy photo session ensued to capture the panel in all detail.

The light was now also becoming suitable to photograph the two shelters with paintings just behind the panel of engravings which were exposed all afternoon to the sun. The first is literally the back of the same rock, with some engravings on the shelter floor and a very nice well preserved panel of cattle and human figures across a large area (a few cattle figures were copied by Bousso and reproduced by Choppy). The most interesting element is a row of relatively young human figures with spears and round shields.

The other shelter is just a dozen metres away, and it contains a large and well preserved panel of cattle (with a few human figures), the most interesting element being a herd of calved (?) enclosed in a corral.

We still had time to return to the lesser "lost panel", now in its full glory in the light of the setting sun, like the evening before.

With sunset approaching, everyone slowly started converging on the campsite, making use of the last light of the sun to capture the amazing landscape of rock towers, now casting long shadows over the plain.

Day 8. – Niola Doa - Camp in Mourdi Depression

As usual we were up with the sun for breakfast, with another magnificent view over the plain and the rock towers. Over the previous day we have managed to cover most of the sites on the Niola Doa plain, there was only one shelter with paintings remaining at the north western edge of the plain, thus we were planning to move on after visiting it.

The site was within easy walking distance from camp, just over a kilometre to the North. We did not need to wait for the cars to be ready, we just packed our gear and set out on foot across the sand plain among the rock towers towards the site.

This is another site that is not published anywhere, information on its existence and location was passed on to Andrea. It is a small low shelter, with some excellent Tamada style scenes on the ceiling and the rear wall, overall the best ensemble of paintings at Niola Doa. On account of the figures with spears, it referred to as the "abri des guerriers" (shelter of the warriors).

While waiting for the cars to catch up with us, we had some time to explore the vicinity, and found a shelter nearby with some very faint paintings of cattle.

Some source suggested that there may be another site about 1 kilometre further East, since we were not pressed for time we walked on with the cars following, but aside the beautiful landscape saw nothing else. Most probably it was the same shelter, just incorrectly positioned.

After a quick lunch we left the Niola Doa plain, taking the track back towards Diona. We had a few entertaining moments in the soft sand of the oued bordering the plain to the East, but with a little push all cars got through.

We stopped a couple of kilometres before the Oued Baragué where a few engravings were reported by earlier travelers. After all the wonders we have seen they were rather unspectacular, after making a brief round we continued our return journey.

We crossed the watercourse and ascended the rocky pass along the bumpy track cutting across the big bend of the Oued Baragué. It was slow progress, but at least it was a known quantity, after an hour or so we could look into the lower valley with the lush acacia groves bordering its sides.

Once reaching the riverbank the going became a bit better, we soon reached Diona to drop off our guide and fetch some fresh water before moving on. It was interesting to note several flashy new cars parked outside the thatch huts in the village just like in western suburbia, the proceeds of the lucrative camel trade with Libya.

Soon after leaving Diona we startled a pair of patas monkeys (Erythrocebus patas) who quickly ran up a big acacia as we stopped. We have already seen some from a distance on our onward journey, I was not aware of them living in the Mourdi, so far to the North of their normal Sahelian range. They hid in among the upper branches as We gathered around the tree, patiently waiting to catch a glimpse and take some photos.

Not much worth mentioning happened in the rest of the afternoon. We wanted to cover as much distance as possible, so we just drove continuously among the endless low dunes swamping the meagre vegetation along the valley banks. Late afternoon the wind picked up, and while it did not quite become a full fledged sandstorm, the air was filled with dust and haze.

Click image to see high resolution WMV video (0:45, 56MB). For codec problems or slow connection use links below:
High resolution MP4 (43 MB) | Low resolution MP4 (8.4 MB)

By sunset We reached the Oued Horta, about half-way along our return journey in the Mourdi. We made camp along the steep sand-covered banks, hoping that the lee of the dunes will provide at least some shelter from the windblown sand.

Day 9. – Mourdi Depression - Fada

Fortunately the expected sandstorm failed to materialise overnight, we awoke to a crisp clear morning. It turned out that our camp was just a kilometre from the well of Tébi which was now occupied by a large flock of goats tended by a few girls. While we waited for our turn to top up our water cans, I witnessed a scene that could have come off straight from a rock art scene (there are in fact several such depictions at Uweinat). A girl used a long stick with a hook on the end to harvest ripe acacia pods for the goats, one of their favorite meals.

As we continued in the Mourdi depression, we saw several dorcas gazelle (Gazella dorcas), and came across a set of curious tracks which took some time to figure out: a striped hyena caught a young goat (or gazelle) and carried it back to its den (possibly for its young) holding the prey in its mouth, with the legs dragging in the sand on both sides.

In the last valley before reaching the edge of the Ennedi again (very close to where we camped at the end of day four) we startled another band of monkeys, this time there were at least a dozen if not more. Unfortunately they were rather far and made their getaway as soon as they heard the cars, so I could only take a shot of some fleeing specks plus the fresh tracks. More interestingly, on the far side of the valley we found the field of ancient tumuli which we already noted on our onward journey. This time we could pause to investigate. They were a mix of larger ones made of stone rubble, and smaller ones made of large slabs. As is the case with most Saharan stone monuments, we could not find anything on the surface surrounding them that could give any clue to their age.

The sandy pass where we descended on our onward journey is impassable on the way up (or at least it would have been a lengthy struggle), we continued a few kilometres further west where a firm low gradient sand slope offered a way in among the broken foothills of the Ennedi. We reached a high point from where there was an amazing view over the sand filled valleys and the jagged rock towers.

In the connecting valley we startled three dorcas gazelle (Gazella dorcas) who made a wide dashing circle around us before breaking into a gentle trot, then stopping altogether looking at us with curiosity mixed with apprehension.

As we progressed, we reached an area of rock towers similar to what we have already encountered at Niola Doa. We stopped for our afternoon break in the shade of one of them, where I managed to catch a small and rather docile juvenile spiny tailed lizard (Uromastyx dispar) under a rock just beside where we sat down for lunch.

After the lunch break we joined the piste linking Fada to Ounianga. We passed several amusing concrete road signs informing us that the track ahead might be bumpy...

By mid-afternoon we reached the outskirts of Faya, making camp behind one of the inselbergs to the East of the town. Andrea departed with one car to pick up fuel and settle all chores and paperwork, while we stayed for a relaxed two hours before sunset to enjoy the scenery and the fact that we don't really have to do anything.

Wandering about, I came upon a terrace on the hill adjacent to camp that was littered with fragments of the familiar Silurian Arthropycus trace fossils, well known from the thin marine layers of all the big Saharan sandstone massifs, marking a relatively brief period of higher sea levels.

As we returned to camp for sunset and dinner, I witnessed an astonishing performance. A little boy of about six-seven descended from the higher parts of the hill with a flock of goats (there must have been a hidden valley with pasturage up there). One section of the descent was a near-vertical rock face with just a few tiny footholds, with a vertical drop of a good 20 metres underneath. The goats, followed by the boy came down that section as if they were walking down a flight of stairs...

Day 10. – Fada - Elikeo - Ovayké

As usual we were up at sunrise, packing camp and having our breakfast in order to make an early start. With our good luck at Niola Doa finding all the sites much faster than expected, we had four days ahead to see the sites of the South western Ennedi (mostly reported by Bailloud) instead of the planned three, and we intended to make the best use of this time.

We engaged a guide in Fada who supposedly had a good idea of Bailloud's sites in the general vicinity. Our plan was to focus on the most important localities with archaic paintings, visiting other sites along the way opportunistically. Our itinerary was dictated by the terrain, which is not as straight forward as it appears on first look. The eroded southern fringes of the Ennedi consist of a series of terraces covered by sandy plains, but the sometimes considerable level changes in-between are not readily evident on satellite imagery. In many cases the narrow passes leading to a lower level are blocked by big dunes, passable for camels but not for cars. Our first target was Elikeo, one of the most important group of archaic paintings, about 30 kilometres to the South east of Fada. While the locality is just 10 kilometres North of Archeï guelta, the true distance between the two is more like 50 on a circuitous and difficult terrain involving a 200 metre drop, it was much easier to access Elikeo from Fada on the same level. Along the way to Elikeo our guide pointed out a group of rocks called Guili Dweli where he knew some paintings. The first site he took us to opposite a fine natural arch was nothing spectacular, mostly crude late camel and horse paintings over a few poorly preserved older cattle and human figures.

On the far side of the rock outcrop our guide showed us another complex shelter which proved to be more interesting. On the outside we could only recognise some traces of paint (though dStretch reveals quite a lot going on...), but inside a deep cave-like recess at the right end of the shelter, filled with rather foul smelling bat guano, there was a pair of curious figures painted in white, which were reproduced by Bailloud both as a tracing and as original photo, referring to the site as Guili Dweli VIII. Back home I have noted that Bailloud also reproduced a photo of the first site we saw as Guili Dweli XI, implying that there were many more sites in the area aside the two we have seen.

As we approached Elikeo the tone of the country changed, with the sandy plain covered with lush grass grazed by camels everywhere. The site itself is located in a horseshoe shaped valley, with a series of deep shelters along a softer strata about 10 metres above the plain level.

The most interesting site is the deep double shelter called Elikeo III by Bailloud at the extreme left of the valley (I & II are lesser sites at the foot of the hill further left, outside the valley). It contains a number of very ancient looking paintings, most importantly on the dome-like circular ceiling of the main left shelter (published in its entirety by Choppy et. al. in Sahara 9). The most readily visible are two large human figures executed in white paint, difficult to make out and even more difficult to photograph. There are also some other large white figures which by now have almost completely disappeared, and several more smaller figures and some enigmatic black shapes at the upper right part of the ceiling. One may just about make out several white figures whose bodies have red dotted and striped body decoration.

There is a well known scene of another group of human figures with dotted and striped body decoration on the left wall of the main shelter, unfortunately very much damaged by flaking plus repeated wetting by past visitors. With dStretch it is visible that these too have a white body (now almost invisible), covered by the red dots and stripes, mixed with figures that are completely white.

The right smaller shelter also contains a number of archaic-looking scenes, unfortunately mostly in a very bad state of preservation.

The centre of the horseshoe is occupied by a very large shelter with a number of smaller sections. It must have been the principal habitation site, as it offers comfortable shady spots throughout the day. Bailloud refers to it as Elikeo V, the largish shelter half-way between this site and Elikeo III must have been Elikeo IV. Our guide said there are no paintings up there, in retrospect it probably would have been a good idea to double-check anyway.

The paintings are dispersed on the rear wall of the shelter, not continuously but in discrete groups. The best is a large panel of Tamada style figures, the women in skirts, the men wilding bows and arrows, all depicted with a rather strange elaborate coiffure. There is also a delightful little scene of a harp player accompanied by a sitting woman, unfortunately in an area that was partially covered by the soot of fires.

Our guide was only aware of these two shelters containing any paintings, but there was a rather prominent one on the far right of the valley, partially obscured by a dune anchored by vegetation. Knowing that Bailloud copied some scenes which we have not yet seen, we checked out the site and indeed found the group of strange masked or animal headed human figures copied by Bailloud (and referred to as Elikeo VI), plus a number of other interesting unpublished scenes.

By the time we finished with all the sites it was past midday, and it started to become uncomfortably hot in the enclosed valley, with the rock walls reflecting the heat. The meagre shade provided by the small acacias were also gone, so we decided to find a shady spot approachable with cars for our lunch rather than have to carry everything up to one of the shelters which were quite far from where the cars could stop. We crossed the plain in front of Elikeo, and moved to a group of rocks visible towards the west, finding a perfect spot at the foot of a cliff overlooking an amazing set of rock towers.

In the afternoon we continued towards Ovayké about 30 kilometres to the west, another locality where Bailloud copied some archaic paintings. We passed along a grass filled valley lined with towering mesas, with grazing camels everywhere. We also passed a relic of the Libyan wars, a Libyan tank destroyed during the battle for Fada.

We reached the well of Ovayké in the middle of a wide dry riverbed by mid-afternoon, taking the opportunity to re-fill our water canisters before moving on to our campsite at a very scenic spot among a group of towering rock pillars.

The location of Bailloud's sites was very vague, and our guide only knew the well, he had no idea about any paintings in the vicinity. The most likely spot appeared to be a "rock city" above the well in the bend of the valley, with the top eroded into parallel streets and avenues, very reminiscent of some sites in the Tassili n'Ajjer. With still a good two hours left till sunset, we set out to explore the rocks in the hope of finding the site we were looking for.

The rocky area looked very promising, there were shelters all along the sides of the lanes, and many contained stone structures, a sign of past habitation. However all of the shelters turned out to be empty - it was nearly dusk when I came upon a tilted block which formed a shelter underneath, which finally did contain some pastoralist paintings, but not the archaic ones we were looking for.

As the full moon rose, we all converged on the rock spires surrounding our campsite for a beautiful warm evening. Overall it was a very fine productive day, and we could afford another couple of hours in the morning to hopefully find the missing site, which had to be somewhere very close to where we already searched.

Day 11. – Ovayké - Fada - Gaora Hallagana - Sivré

The morning was fairly crisp, nevertheless we were up at sunrise to make most of the day. After a quick breakfast we packed camp and moved with the cars to the "rock castle" to search for the missing site.

We entered the maze and had a good second look at the shelters in the better light conditions, but still did not find the archaic paintings. We did however see many more stone structures, and the ground was littered with lower milling stones, all signs of ancient habitation. One site was a veritable settlement, with a number of circular stone huts and other structures. It was interesting to observe the work of erosion throughout the area, at some point heavy rains have washed out the fine silty sediment that filled the bottom of the "streets" except where a larger stone (or piece of pottery) was lying on the surface, which now all stand on small mud pedestals. It was only back home, looking through the Bailloud volume, that I realised that we have accidentally found the likely location - the group of stone huts we saw and photographed were also recorded by Bailloud, and his photo caption places them adjacent to the missing paintings of Ovayké IV. Probably the scene was so faint (and possibly covered with a dusty crust) that we walked past them several times without ever noticing them.

This cluster of rocks has a magnificent commanding view over the surrounding countryside, and with the well below it must have been a major occupation site throughout the periods of human settlement in the Ennedi. At one end of the rock outcrop we found clear traces of ancient iron works, with the remains of several furnaces embedded in the surrounding sediment.

Bailloud recorded another series of paintings at Soro Kezenanga roughly 2 kilometres to the East of Ovayké including some very fine scenes of the Fada style which feature on the cover of his book. Unfortunately his published map is very sketchy and inaccurate, but a cluster of rocky hills on the far side of the oued with the well looked like a promising spot, at about the right distance. We made a fairly thorough search but aside a herd of goats and a few natural arches we found nothing of interest.

While we searched to the East, Andrea wandered off to an isolated rock in the other direction, and he was luckier. The rock did contain paintings in a series of low shelters all around the rock, though all of them very weathered.

Our guide did not know of any other paintings in the area, so we returned to Fada to drop him off and top-up our fuel, vegetable and water supplies. While Andrea and the drivers went about their chores we lingered at the main square of the town, watching life pass by.

We left Fada around midday along the main track, our plan being to visit the principal sites of Gaora Hallagana and Sivré about 25 kilometres to the west. Both are isolated sandstone outcrops on a generally flat sandy plain bordering the south western edge of the Ennedi. As we neared Gaora Hallagana which is just a few hundred metres off the track, we spotted a single patas monkey (Erythrocebus patas) running up a small acacia. For a while it tried to hide among the dense upper branches, but finally it decided to bolt, jumping down and dashing away with an impressive speed. Fortunately I was just in the right position to snap a couple of lucky shots as it was making its getaway.

Gaora Hallagana is a cluster of rocky hills surrounding a small sheltered plain. Erosion has created some fantastic shapes, with several large shelters and some natural arches all along the base of hills, most of them containing paintings of varying quality. We stopped at the shady side of the hill closest to the road, and after a quick lunch we set about to make an initial survey of the area.

Bailloud recorded at least 14 sites in the area, but there are reproductions of paintings in his book only from three, so identifying the others is rather circumstantial. Gaora Hallagana 1 (roman numerals are used only for those positively identified) is likely to be the site adjacent to where we stopped for lunch, a series of three inter-joined shelters with some rather faint and weathered paintings representing several styles.

The principal site of the cluster is a very large shelter (or more like a true cave, with a deep recess into the rock) at the southern edge of the central "courtyard", partially sheltered by an artifact-stabilized dune in front of it. Bailloud refers to it as Gaora Hallagana IV. There are some fine panels of paintings on the left wall from the late cattle pastoralist period in what Bailloud refers to as the Koko style. Men are holding spears which are clearly metal-tipped, and in a rather unusual manner several figures are presented face-on with their eyes depicted, an unusual element in Saharan rock art. These figures are also associated with some strange maze-like shapes, which are always drawn with one un-broken, never crossing line.

At the rear of the shelter, adjacent to a huge pile of rocks where the ceiling collapsed, there is a very well preserved panel with the same type of figures, huts and cattle, and another harp-player like the one seen at Elikeo V. To my knowledge this one has never been published.

There is a further large panel of paintings on the rear wall of the shallow part of the shelter at the left entrance.

At the rear of the same rocky hill looking over the plain to the South there is another large shelter the roof of which has mostly collapsed (Gaora Hallagana 6), leaving a patch of open sky above and a huge pile of rubble on the floor. There are several panels of paintings on both the left and right walls, the best consisting of several rows of human figures.

A few dozen metres to the right, still on the South side of the hill there is a smaller shelter (Gaora Hallagana 5) with another panel of paintings, mainly cattle in various colours, including some unusual black ones (probably Hotoum style) as the lowest layer.

The western edge of the adjacent hill facing the principal shelter is honeycombed with a number of shelters, caves and inter-joinng tunnels, much like a block of Emmenthal cheese. We found some rather weathered paintings at several locations (Gaora Hallagana 3), and a fine panel of cattle from the late pastoral period (Gaora Hallagana 2).

We still had two of Bailloud's sites to find, including one with some rather interesting archaic paintings, but we also wanted to make a survey of the Sivré sites which are the most important early period paintings in the Ennedi. As we did not know their exact location, we had to calculate with some time spent searching. As Gaora Hallagana was on the path of our onward journey, we decided to leave the remaining sites till after we have completed our round at Sivré.

The Sivré locality is one huge block of sandstone jutting from the plain just two kilometres North of Gaora Hallagana, with near-vertical smooth sides all around. There is a very long deep shelter visible from afar, about 20 metres above the level of the plain, running for half the length of the hill along the left of the southern side. It is only accessible on a steep sandstone ramp near the middle, elsewhere the rock drops off near vertically from the flat ledge forming the shelter floor.

The shelter was an obvious place to look for the paintings, it did not take too long to find two of the main panels copied by Bailloud (Sivré I). The first is a panel on a section of vertical wall behind a large detached block, with numerous Hohou style human figures and cattle painted around two larger dark outlined white archaic figures, one of which had its interior re-filled with red paint at the time the Hohou figures were added.

The other panel was on the ceiling above the fallen block, a strange tree-like object with utensils hanging from it (like the similar smaller one at Elikeo III). With the sun illuminating the top of the block and creating strong shadows on the irregular ceiling, it was left for the following day. We found no more paintings to the left of these panels, so we explored the eastern side of the shelter to the right of the ramp. We only found some very faint, incomprehensible traces of paint, which when photographed revealed itself to be a scene copied by Bailloud, a rather damaged human figure in red outline. Ursula and Els found some other traces of paint on the ceiling nearby, all very hard to see in the strong contrast, with the floor of the shelter illuminated by the low sun. We left these too for a more detailed investigation the next day.

With all of the long shelter searched, we were still missing the most important scene with the two archaic figures, and also an elephant which Bailloud traced. We drove around the hill to the Northern side, where another series of shelters looked promising. Here the side of the hill is less steep, we could investigate easier, but found nothing until reaching the shelter near the western top of the rock, which was full of cattle pastoralist and camel period paintings that were not published in Bailloud's book.

The elephant was found at the very right edge of the shelter - unfortunately someone recently chalked its outline. dStretch reveals a human figure in front of the elephant which was not included on Bailloud's tracing.

The shelter with the elephant was Sivré IV, thus the site with the two figures (Sivré III) had to be at the western tip of the rock somewhere between the two seen shelters. There was nothing in the shelters on the upper level, and it was getting close to sunset so we were ready to call it a day, when very close to the plain level I have come upon a low shelter, and found the figures straight away on the ceiling at the most likely spot. They are very hard to make out, appearing as yellowish blobs on the dark ceiling, however with dStretch they become perfectly visible with distinct heads, and all the banded and dotted body decoration is revealed.

As the sun was just about to disappear, we left detailed photography for the next day, with the location of all the important paintings now known. There was a cool northern breeze all day which was strengthening for the evening, we drove around the hill to the eastern side to seek out a campsite with some protection from the wind.

We found a reasonably sheltered spot at the eastern tip of the Sivré hill with a couple of small shelters along a vertical rock wall, overlooking the plain towards the rocks of Gaora Hallagana.

Day 12. – Sivré - Gaora Hallagana - Ga Korou - Archeï

Overnight the wind picked up and the temperature dropped considerably, we rose to a rather chilly morning with a crisp breeze and plenty of dust in the air. We set up our breakfast table in one of the shelters with a few paintings for backdrop, and had a good fight with the tents before we could pack them away.

By the time we packed camp and drove around to the Sivré I shelter the wind strengthened to a howling gale with true sandstorm conditions. The only bearable spot - fortunately - was the shelter itself, with the hill blocking the wind coming from the North.

We started at Ursula's find near the incomplete figure traced by Bailloud. It was very faint, but the photos taken with flash revealed it to be a very nice archaic running figure of the same type as in the yet to be found archaic panel at Gaora Hallagana.

Els' find was just a few metres away on the low ceiling. We could not make much sense out of it on the spot, there were just patches of lines and decorated bands over a 1 by 2 metre area, seemingly without any coherent feature. It was only back home, after looking at the photos and stitching them together that the true nature of this amazing painting was revealed: it is a huge (2m long) headless animal with the body decorated with bands, somewhat reminiscent of both the "headless beasts" of Wadi Sora and the "fantastic beasts" of Korossom & Karnasahi.

We made a thorough search of the entire ceiling in this section of the shelter, which is quite deep (~6-8m) and the ceiling is lower than human height. Near Bailloud's figure which is located right at the rear wall, we found another running figure and some incoherent blobs of paint.

It was a fairly lengthy photo session in the cramped conditions, mostly lying on one's back. After finishing we moved on to the left side of the shelter, to photograph the strange tree-like object painted on the ceiling above the Hohou style panel. There are some further traces of red and white paint next to the object, but even with dStretch it is not possible to discern anything meaningful.

A few dozen metres further along the shelter one reaches a spot where the ceiling rises to create a comfortable deep living space. The terrace here is littered with cupules, and there are some very damaged, mostly indiscernible remains of paint in patches along the rear wall. The best preserved one is a human figure with decorated body.

West of here the shelter ceases, but a small ledge remains where one may walk along the same level in the side of the hill, though with a rather precipitous drop. The ledge follows the side of the hill, and after about 80 metres a shallow shelter appears again. There are several cupules here too, including one which was bored right through a fairly thick (~30cm) rock slab jutting from the floor of the shelter. At the western end one may descend fairly easily towards Sivré III through a small field of rough tumuli on a terrace above plain level.

There are several archaic-looking paintings on the slanted ceiling and the rear wall, none of which were published by Bailloud, though by inference this must be Sivré II.

We finished off at Sivré III, spending a long time photographing the two figures. The resulting high resolution composite images combined with dStretch reveal hitherto unseen details.

Having successfully ticked off all our to-do list items at Sivré, it was time to gather all stray members of our party from various parts of the hill and the surrounding plain. Els was considerate enough to provide a perfect scale for the painting locations in Sivré I as she walked back along the edge towards the cars.

We returned to Gaora Hallagana to look for site XIV with the archaic paintings. Bailloud's book provided no clue, but in his 1960 l'Anthropologie article it was mentioned that the small shelter lies 6 metres above the level of the plain. Having looked at ground level previously, it was not too difficult to identify the few potential shelters lying on a higher level, the second one we checked contained the paintings we were looking for. Unfortunately they are very difficult to make out, but clearly they are the same type of running figures with cross-hatched body decoration as the new find at Sivré I. As the paint used was white, dStretch is of limited use, but it does reveal that there are several older figures underneath. At the rear of the shelter Andrea found a lion (or other large feline) with the same type of cross-hatched body pattern, I'm not aware that it was noted by anyone before.

We had an hour remaining before we had to move on, we used the time to make a brief survey of the Western group of rocks. Not far beyond site XIV Ursula found a double shelter (Gaora Hallagana 10) with some curious white figures, and a very large Tamada stile figure with the lower parts covered by sand drift.

At the base of the vertical cliff on the far side of the central "plaza" there was a prominent small shelter (Gaora Hallagana 9) that was bound to contain paintings. It was inhabited by a pair of small house buntings (Emberiza sahari), and indeed on the left wall there was a fine panel of late pastoralist paintings.

Moving along the edge of the same rock island back towards the principal site we son came upon a large shelter, with fine panels of paintings on both the left and right walls.

The left panel contained several fine cattle, a row of Tamada style human figures and a scene that included two dogs, a fairly uncommon theme.

The right panel contains numerous scenes of several styles, often superimposed over each other, presenting a good relative chronology of the period. The Girbi style camel riders superimposed over cattle at the right of the panel have been copied and published by Bailloud, positively identifying the shelter as Gaora Hallagana VII.

Less than a hundred metres further along the edge of the rock island we found the last shelter with paintings (Gaora Hallagana 8). It was a rather strange site, with a two storied interior, and the paintings were on the ceiling of the second level, to which we could find no access. Either the shelter was filled with sand to that level, or there were some rock stairs which have long since collapsed. The preservation of the paintings is excellent due to the protected location. The herd of miniature cattle under the larger ones in the central scene is highly unusual.

Our plan was to reach Archeï by the evening, about 50 kilometres away in a straight line distance, but probably twice as much along the winding tracks. We had to move on, leaving any remaining sites at Gaora Hallagana for some future visit. We returned to our cars and set out in a south easterly direction, aiming for a pass leading across the row of hills that offered a significant short-cut compared to the main track.

After the pass we soon entered the Béchiké canyon containing the well of the same name, and a number of small settlements dispersed among the magnificent landscape. While there was still some haze in the air the morning sandstorm has died down, and the chill of the morning was replaced by the midday heat.

We were aiming for a rocky rampart called Ga Korou, where Jacques and Brigitte Choppy with Sergio & Adriana Scarpa Falce found a fine shelter with archaic paintings which they called "L'abri des rhinocéros" and published in Les Cahiers de l'AARS 5 (1999). We did not expect to have time to visit this site so we had no precise location information, but based on satellite imagery I could identify a valley which could be a potential location. We approached the cliffs of Ga Korou at lunch time, and planned to spend the afternoon break searching.

The target valley had a soft sandy bed and dense vegetation impassable for the cars, we stopped in the shade of the southern side at the entrance. It was rather hot, in the end only Jeff and myself had sufficient curiosity to walk up the kilometre long canyon. It proved to be a very scenic walk, with more vegetation and green grass in the upper reaches, but no shelters with paintings.

Continuing, we stopped briefly at a natural arch a short distance from our lunch spot, then continued southbound to Deli, one of the farthest ramparts of the Ennedi. Bailloud in his diary reported finding 12 sites here, but none were published. We only had time to stop at one site right beside the track leading towards Archeï.

The panel on the rear wall of a shallow shelter contains a large number of figures, with several styles and many superpositions. The most striking and unique are a pair of ostrich, a subject that does not occur elsewhere among the paintings of the Ennedi.

From Deli we had a clear route towards Archeï some 30 kilometres further to the west, among the isolated outliers of the Ennedi eroded into some fantastic shapes.

By mid-afternoon we were in sight of a huge rock with two window-like natural openings, which marked the entrance to the Archeï region.

We stopped near the natural arch to see the first site in the area, Archeï 14 with a couple of crude cattle and human figures, their only distinction being the closeness to the road requiring no search or detour.

After another ten kilometres we reached an enormous rock with a huge hollow in its side, with the green belt of the Oued Archeï just beyond. This was the "Grande riparo" (Archeï 07), the most important rock art site of the region. At this time the paintings were fully exposed to the sun, our plan was to return first thing in the morning to photograph the site.

We followed the sandy bed of the Oued Archeï upstream towards the guelta, searching for a suitable campsite. After the mostly barren landscape of the Northern Ennedi here we were seeing an amazing transformation, with a dense acacia forest bordering both sides of the valley. We left the kitchen car at the chosen spot about two kilometres before the guelta, and as we still had an hour of daylight left, continued with the other two to see Archeï 02, the huge cave with paintings at the entrance of the canyon.

The site is located just 500 metres before the guelta, at the elbow of a sharp left turn in the valley as it emerges from the ramparts of the plateau to the plains below. From a distance it does not appear very impressive, it is just a dark opening in the cliff face. However as one approaches it soon becomes evident that the shelter is very large indeed, with a steep sand ramp leading up to the entrance a good 20 metres above the valley floor, with the cliff over 150 metres high. The ledge to the left of the cave was occupied by a large family of dassies (rock hyrax, Procavia capensis) with all the usual chatter, whistles and commotion as we approached.

One only appreciates the true size of the cave from the inside, with the domed ceiling rising 20 metres above the floor, and the cave extending for a good 50-60 metres into the rock. The in the deeper part the floor is thickly covered with bat guano (with the associated smell), what appear to be stalagmites are in fact heaps of bat dung concentrated below the most favored spots.

The most conspicuous panel of paintings is a row of bright white figures painted on the upper right wall of the shelter. They are an interesting mix of camel riders, giraffe and cattle, all apparently from the late pastoral period.

To the right, towards the entrance of the cave there is a panel with some partially preserved white cattle, and several human figures wielding long metal tipped spears and shields, another late pastoral scene.

Yet further right, down and along the sand slope leading up to the inner parts of the cave, there are red human figures and several characteristic riders with horses in full gallop. A single better preserved panel with four mounted horses in gallop is on the opposite left side of the cave.

Until recently these were the only paintings known in the shelter, but recently Alessandro Menardi-Noguera discovered some archaic paintings high up on the inner right wall, 4-5 metres above present floor height (published in Les Cahiers de l'AARS 17, 2014). The paintings consist of a group of smaller human figures in red & yellow, and further left a large human figure exceeding a metre in length. They are very faint and extremely difficult to find on the dark ceiling even when one has a general idea on where to look. They were found by taking point-blank photos of the ceiling with flash until they showed up.

As we finished photographing the archaic scenes, there was a lovely sight below in the riverbed: a herd of cattle walked slowly to take their evening drink from the guelta just a few hundred metres upstream. These cattle at Archeï are presently the northernmost ones in the Sahara, the environment is a very good analogy to what the central Saharan massifs must have looked like when they too supported cattle in prehistoric times. By the time the cattle disappeared the sun was almost gone, only illuminating the very top of the cliffs, we made a hasty return to our campsite where there was already a good fire going, and our cook Ismail was busy preparing the dinner.

Day 13. – Archeï - Chéïré - Terkey

We were up at first light taking a hurried breakfast and quickly packing camp, in order to make it to the Archeï 07 shelter in time to take good photographs before the rising sun emerges from behind the hill.

On our way we passed the small village about 3 kilometres south of the guelta to report our presence to the local chief. We did not have much more to go, Archeï 07 is on the far side of the hill bordering the village to the South.

We were at the site by 7:30am, still it was in the shade but we had to hurry, we had approximately an hour before the sun would rise sufficiently to directly illuminate the left part of the shelter with the most important paintings. The shelter itself is immense, the floor is located about 20 metres above the level of the plain, and aside the single access from the right there is a precipitous drop at the edge of the shelter terrace, especially at the left edge, where one can barely find a foothold to take photographs.

This shelter was discovered by an Italian expedition in 1993 (who named it "Grande riparo"), it is one of the largest and best preserved rock painting sites in the entire Ennedi. Despite the hundreds of well preserved figures on the main panel at the left of the shelter, the most intriguing ones are the two at the western extremity of the site, quite high on the rock wall and rather difficult to photograph with the near-vertical drop one step back. One is clear, the other much larger one is very faint and can only be appreciated properly with dStretch. Already Roberta Simonis noted their similarity to the Niola Doa engravings (in her 1994 article on Niola Doa in Sahara 6), however the provided small black & white photo did not really reveal too many details. Now with good digital photographs it is possible to establish without doubt not only that they look similar, but in fact they are a complete match with some of the engravings at Niola Doa. Not only do the human figures appear identical, but dStretch reveals a very faint red cattle with forward curving horns next to the larger human, in about the same degree of weathering and colour tone. This cattle is practically identical to those associated with Niola Doa type figures on the Fuchs Panel. Unknown to me at the time Suzanne and Gérard Lachaud visited this site a year prior to us, and have come to the same conclusion - their article on the subject will appear in the forthcoming issue of Les Cahiers de l'AARS.

When comparing the better preserved left skirted (female ?) figure to similar examples at Niola Doa, there remains no question. Form, posture and body decoration are practically identical (as is the relative scale to the larger figures). To have practically identical figures executed in paint and engravings is very rare in Saharan rock art, the makers of most other hitherto known styles have chosen to express themselves through one medium or the other, but not both.

The large figure is more difficult to assess, as on the spot only the white parts are discernible. However with dStretch the entire body is revealed, together with a dotted body decoration. Most significantly, it is possible to discern (best on LRE filter) a very faint but clear stick held across the shoulder, in a posture identical to those of the large Niola Doa type figures. While there is no single figure match, the head protrusions, belt and cuffs all have analogies among the large Niola Doa figures.

To the right of the two figures, mixed among other style paintings, one may recognise two more of the small skirted Niola Doa type figures. They both appear to be fainter than the surrounding paintings, but unfortunately neither are in any superimposition. They are much less distinct than the one at extreme left, but are clearly of the same type on account of the body posture and proportions.

The remainder of the panel consists of hundreds of individual paintings composed of practically all the main pastoralist styles of the Ennedi, with numerous superpositions. They are excellently preserved, most of the scenes can be easily made out without the need for any enhancement.

Strangely the rear wall of the huge central part of the shelter, with the comfortable large platform in front of it, is almost completely void of any paintings, and the few that survive are mostly unrecognisable blobs. In the right side of the central part, one may recognise a white head that resembles that of the Niola Doa type large figure at the extreme left. dStretch reveals that the head is in fact red and white, and the faintest traces of a white body can just about be made out with the YRD filter. More surprisingly, there are three of the skirted small Niola Doa type figures to the right of this figure.

Once we finished at Archeï 07 we returned to the entrance of the gorge to visit the famous guelta during the time the camels were brought to water. Jack wanted to get up close and personal with the camels, while the rest of us, guided by the daughter of the local chief, took the circuitous path that starts about a kilometre to the East of the gorge entrance. The trail steeply ascends the plateau and crosses into the Oued Archeï about a kilometre upstream of the guelta.

The path leads to a little ledge about 30 metres above the series of pools at the bottom of the gorge. By crawling out to the very edge, there is an amazing view over the guelta and all of a sudden the air becomes filled with the loud grunts of hundreds of camels, previously shielded by the rock walls of the canyon. This is the classic view of the Archeï guelta that appears in almost every coffee-table book on the Sahara.

The guelta extended up in the bend of the canyon beneath us, we were hoping to perhaps catch a glimpse of one of the famous pigmy crocodiles living in these waters (they actually feed on small fish, not on any animal coming to the water to drink), but one herd of camels have waded across the deep section cutting off the upper part of the valley, and have chased all crocodiles into hiding.

A little disappointed by the lack of crocodiles we started on our way back, passing a series of smaller pools where we did see the very fresh traces at the water's edge, the crocodiles must have only gone under into the murky water on our approach. It was only back home, looking at the photos that Els actually found a crocodile on one of the landscape shots taken as we were passing a pool. None of us ever saw it, it must have disappeared quietly under the water as we stumbled closer over the rocks.

We continued back along the sandy riverbed, then up the pass and down into the green savanna bordering the Oued Archeï as it emerges from the canyon. We drove back to the canyon entrance to see the guelta also from the lower level.

We stopped the cars opposite the Archeï 02 cave and walked in towards the guelta, with the noise of the camels audible already from a distance. The vegetation bordering the oued ceased and we walked on flat sand between vertical rock walls for a couple of hundred metres, until after a bend the guelta and the hundreds of camels came into full view. It is hard to decide which view is nicer, the one from above or the one from below, looking up at the towering walls of the canyon with the series of pools at their base.

We returned to the cars which have moved into the shade of the trees adjacent to the Archeï 01 shelter, opposite the larger cave of Archeï 02 on the far side of the oued. As we were having lunch, a juvenile white-crowned wheatear (Oenanthe leucopyga - confusingly only the adults have the white crown) joined us, hopping along a large branch. At first I did not realise what it was doing, but after a while it was clear that it too was having lunch, pulling maggots from the rotten parts of the wood.

Archeï 01 is a large hollowed-out shelter at the base of the cliff, unfortunately in a rather poor state of preservation. Being just adjacent to the path leading to the guelta, it must have been in continuous use ever since the paintings were made, and modern visitors have also left their very visible mark. Parts of the main panel of paintings along the right wall were covered by dassie urine, some of it fairly recent.

The site has the distinction of being among the first recorded in the Ennedi. It was visited by Henri de Saint-Floris in 1933 and published in 1935 in the first ever publication reporting rock art from the Ennedi (Passemard & Saint-Floris, Journal de la Société des Africanistes Vol. 5 No. 1), including several of the well known Ghirbi style meharists with their camels depicted in flying gallop. The camel scenes are very much faded, unfortunately even with dStretch they are quite hard to make out due to the faded white paint blending in with the crust and the colour of the rock. It is interesting to note that there appear to be several sequences of overpaintings, with camels and horses mixed in several layers.

The "dancers" further right are also much faded and are partly covered by a dassie urine crust, but because of the original colour they can be restored much easier.

There are several more panels depicting cattle, horses and humans from a range of periods, often one superimposed on another.

While we were having lunch and taking turns photographing the site, there was a constant traffic in the valley in front of us. Herds of camels who have taken their fill were moving out of the gorge, while a seemingly endless stream of newcomers were coming to the water, all accompanied by loud grunts.

Click image to see high resolution WMV video (0:57, 72MB). For codec problems or slow connection use links below:
High resolution MP4 (54 MB) | Low resolution MP4 (13.7 MB)

The Archeï region contains 18 sites which were catalogued and published by Choppy et. al. (2002), spread out over a fairly large area on both sides of the Oued Archeï. With our limited time available, we could only afford to visit one more, Archeï 04 (also known as Manda Gueli) a few kilometres to the East of the watercourse at a solitary rock outcrop. The site is rather unusual, the principal paintings are on the rear wall of a shelter located about 5 metres above ground level, with a vertical rock wall at its base, presently inaccessible. In prehistoric times there must have been some ascent from ground level, either a series of rock stairs that fell away, or possibly the dune in front of the shelter reached higher. Due to the protected location the paintings are in an excellent condition, with several very fine camel riders and other figures in what Bailloud referred to as the Keymena style.

Due to the curvature of the ceiling and the jutting ledge, it is only possible to photograph the paintings from a distance of 10-12 metres, which was beyond effective flash range in the strong glare of the sunlight reflected from the dune. The close-up photos taken with a long lens did not come out too well, nevertheless due to the very good preservation the details are well visible without any enhancement, at the left one can readily recognise some Hohou style figures underlying the camel riders. Probably night time is the best to take photos of details, but we did not have time for such luxuries. Another option is to take along a five metre high ladder, something that will seriously be considered on a future trip (would also come handy at Gaora Hallagana 8).

About a dozen metres to the right of the shelter there is a tunnel-like cave running through the rock, probably washed out by an ancient water flow along a fissure in the rock. It is possible to access it from the rear of the rock on a steep but negotiable ramp. Inside there are several panels of paintings from all periods of Ennedi rock art. The best preserved ones are on the ceiling of a "window" looking out over the dune.

There was one remaining shelter with paintings on the agenda for the day: in 1997 Andrea discovered a very fine panel a little over 5 kilometres to the East of Manda Gueli, near the prominent hill with a rock protrusion on the summit called Chéïré. It remains unpublished, but an article written by Alessandro Menardi-Noguera after a 2014 visit with Andrea will appear in the next issue of Les Cahiers de l'AARS. Technically it is the 19th site of the Archeï region, as sites spread to a similar distance on the other side of the Oued. From Archeï it can only be reached by a circuitous way of 20 kilometres, as it lies at an elevation 100 metres above the lower plain, and all the access passes to the higher level are filled with soft sand, may be used down but not up. We had to drive beyond the Chéïré hill, where turning north it was possible to ascend on a firm sand ramp, and make it back towards the site.

The dune we climbed was mostly anchored with grass, and there was a fabulous view from the top over the eroded sandstone hills. We also encountered a desert hare (Lepus capensis) which did not dash away at full speed when hearing the cars approaching, but crouched down among the vegetation trying to remain invisible. We could approach it to within a couple of metres with the cars before it bolted.

We rounded the Chéïré hill, and stopped at the entrance of a small valley sheltered by a big dune to the East. The valley was occupied by a single family living under an acacia, tending a large flock of baby lambs.

The Chéïré shelter is at the end of the small valley, overlooking the dune and the hill of the same name. It is facing south, when we reached it still half of the shelter was exposed to the sun, including the most important panel of paintings on the right. We had to wait a good hour until slowly the shade started moving in.

The left of the shelter which was already in the shade contained only a few scattered paintings, spanning the full spectrum of the late pastoralist Ennedi styles.

After the long wait the sun dipped below the summit of the hill to the south, and we could finally photograph the right side of the shelter. It was well worth the wait, this panel contains a multitude of very elaborate human figures (and cattle) in the Tamada style, the finest known ensemble of this particular school.

Waiting for the sun to move has eaten into our remaining daylight hours, we had just about an hour of sunlight left to make camp. We wanted to get to the hill of Terkey about 25 kilometres away for the night to be able to see the paintings there in the morning. It was a race against the clock down the steep dune then along the valley floor across the tufts of grass and small riverbeds which all conspired to slow us, we just barely made it to the hill to make camp before the last sunlight disappeared.

Day 14. – Terkey - Aloba - Bachikélé

In the evening we haven't really appreciated it as it was becoming dark, but we camped at a lovely spot at the foot of a huge cliff, facing a pleasant plain with plenty of grass and acacias. The multiple shelters of the Terkey-Bowdé site were right across the plain, at the base of the low cliffs a few hundred metres from camp.

Bailloud mentioned 10 sites at Terkey-Bowdé in his diary, but only illustrated site III with the well known horses in flying gallop. Marie-Anne Civrac published the sites more extensively in Les Cahiers de l'AARS 16 (2013), but referred to all the sites as Terkey-Bowdé III, which is probably incorrect in light of Bailloud's note and numbering practice elsewhere, probably each distinct shelter with paintings had its own number. As we have found seven shelters in a row along the south side of the valley, the numbering of Bailloud only works if starting from right (West) to left (East). There is a further shelter along the Northern side of the valley, and two more were marked on a map published by Pier-Paolo Rossi (Sahara 24, 2013) at the southern side of the massif with the row of sites, giving a matching count with Bailloud.

As we walked from the camp to the sites, the first one we came across was the long shelter (tentatively Terkey-Bowdé 8) along the northern side of the valley, on the far side of the rock bordering our camp to the south. Inside there are numerous panels of paintings from many styles and periods, including a large panel of the characteristic galloping horses, and more uniquely some of the same executed with white paint.

Crossing the valley to the southern side, the row of shelters at the base of the low cliff, previously obscured by the trees became visible. We started at the right end, with a crack following the strata at the base of the cliff, widening to a shallow shelter at one spot (Terkey-Bowdé 1). In the widening there were some conspicuous but fairly recent paintings of goats & sheep, however on the rear wall dStretch shows a large scene of older paintings.

The next site about 50 metres to the left (Terkey-Bowdé 2) was very similar in setting with a panel of recent paintings on the ceiling, and some fainter horses on the rear wall.

Another 50 metres further we reached the largest shelter in the row, even from afar it was clear that the ceiling and rear walls were full of paintings. This was Terkey-Bowdé III, the principal site with scenes copied and published by Bailloud.

The shelter has a deep cave-like hollow in its center, to the right of it the outer wall of the shallower part is covered by a continuous panel of paintings depicting a large variety of figures with superpositions.

The best known panels are at the right and left entrance of the deep hollow, both depicting several groups of horses with their riders in flying gallop. On the right panel, there is an earlier herd of fine giraffe, partially superimposed by the horses.

The panel on the left only shows horses, but those in great numbers, and in a very well preserved state. This panel is probably the most frequently photographed at this location, both because it is prominently featured by Bailloud, but also because it's presentation is much better than the average Ennedi paintings.

Continuing along the cliff face, there were several smaller panels along a very shallow shelter. It is unclear whether Bailloud would have catalogued them as a separate site or lumped them together with the shelter of the "flying" horses. The paintings here are mostly faint and damaged, as they are sufficiently distant from the other site I refer to them as Terkey-Bowdé 4.

At the end of the straight part of the cliff there are a pair of deep hollows and a series of shallower shelters between them, site Terkey-Bowdé 5, with paintings visible on practically all suitable surfaces.

All the smaller shelters between the two deeper caves were decorated, including scenes from the earlier cattle pastoralist styles which were mostly absent at the other sites n the vicinity. There were several areas covered with a pattern of red dots, without any recognisable shape.

The right hollow was mostly empty save for a few damaged paintings, but inside the left one the walls and part of the ceiling were covered with very well reserved scenes depicting numerous Fada style human figures and cattle. Some of the scenes were covered with the same pattern of red dots as were seen at the entrance of the cave.

The cave continued deep into the rock, narrowing to a tight corridor where one could just walk upright. After a dark turn one could unexpectedly see light again, the tunnel led out to the far side of the rock. Near the end, there was a large panel of paintings depicting humans and cattle, including a huge cow exceeding one metre in width accompanied by a Tamada style human figure and a smaller calf. Such dimensions are unknown among other Ennedi paintings, however are common among engravings of cattle in Borku, with a few examples also occurring in the Ennedi. This large cow was mentioned by Bailloud in his diary, but strangely it was either not copied, or the copy was not published for some reason, despite its uniqueness.

About 50 metres away on a rock forming a small promontory there was a small shelter with some more paintings (Terkey-Bowdé 6), the last in this nearly continuous row of sites. At first look the scenes appeared do depict the usual mix of horses and camels with a few Tamada style figures, but dStretch revealed two older large human figures underlying a scene of white running camels. These figures are not easy to classify, they appear different from what was seen elsewhere, and possibly their upper body was deliberately obliterated at some point.

The last site we found (Terkey-Bowdé 7) was on the far side of the rock forming the promontory, overlooking the valley with the dense growth of acacias. It was a rather unusual location, in a deep but very low fissure in the rock at about twice human height, rather difficult to get up to. The paintings can only be viewed by lying on one's back (and probably they were made this way too), and the lack of distance makes photographing them very difficult. Here too a number of different styles are represented, apparently all overlain by the polka dot pattern that seems to be peculiar to the Terkey locality, we did not observe it anywhere else.

The Terkey-Bowdé sites were much richer than anticipated, we spent several hours going through them, and I'm sure if we had more time, we would have found many more sites in the surrounding hills. However our time was up, we had to move on towards the East to reach our last targets before needing to start our return journey. We have finished with the main rock art sites, our last objectives were a series of spectacular natural arches in the Southern part of the Ennedi. As we left Terkey the landscape changed noticeably. We were no longer in pure desert, there was a dense grass cover everywhere, and animals were abundant, including herds of cattle and the first horses we encountered.

We also saw many patas monkeys, usually bolting at the sound of our engines so by the time we spotted them they were too far to make any decent photos.

Our first stop was at the Anoko arch, roughly 40 kilometres East of Terkey. It s a huge rectangular block of sandstone with a large central and two smaller side openings, giving the appearance of a classic Roman triumphal arch. From a distance it appears deceptively small, once being close up its true size becomes very apparent. The central arch is about 18 metres high (using Gábor's exact 2 metres as scale) and ~25 metres wide at its base

East of Aloba we reached a broad flat grass covered sand plain formed by the confluence of several larger oueds, which juts about 35 kilometres into the massif, with only a few remaining rock towers around its perimeter.

While there is good grazing on the plain, the wells are at the foot of the mountains around its perimeter, so we only saw a few grazing camels. However dorcas gazelle were abundant, we disturbed several small herds, including three animals which dashed right across the track ahead, permitting a very lucky shot of them.

As we progressed towards the East the landscape changed markedly. The sandstone towers were no longer rising from the plain with vertical walls, but were sitting on a skirt of eroded granite, the ancient continental basement over which the Palaeozoic sandstones were deposited. This also explained the scarcity of rock art sites in the Southern Ennedi, the sandstone no longer reached the level of the plain where water erosion could create the large shelters. As we approached the prominent Aloba mountain towering above the Northern edge of the plain, the country was once again filled by grazing camels and flocks of sheep.

We entered a valley leading in the general direction of the Aloba mountain that was densely filled with trees, with thousand of sheep and goats scattered among them. We soon encountered a village that was quite different from the ones seen before, the houses resembled those in the villages of the Sahel rather than the flimsy structures built by the Tibu. Aloba is the edge of the Zaghawa country, a tribe that inhabits the Eastern Ennedi and parts of the Darfur across the Sudan border.

We stopped the cars at the foot of a prominent sandstone tower, and continued on foot along the crest of a low dune towards the valley behind. After rounding a small rock all of a sudden we were facing the greatest natural spectacle of the Ennedi, the Aloba arch, the main purpose of our detour to the East. At some point in the past a river broke through a bend here, and the water channel was displaced to flow under the natural bridge, cutting an ever deeper opening. This process still continues, the result is the third largest natural arch in the world (and certainly one of the most difficult one to reach), the height of the opening is about 100 metres with a maximum width of about 60m.

The valley continues upstream beyond the arch towards the foot of Aloba mountain, and the old pre-arch channel is now covered by a large vegetation anchored dune which provides a perfect vantage point to photograph the arch from the sunlit side. As the stop was also our lunch break, we could afford a long leisurely walk in the valley to take photos from all angles.

While we were away a small group of women and children from the nearby village set up shop near our resting place, selling various pieces of handicraft and curios, the first place aside Archeï where we saw some signs of any tourist trade.

After lunch we drove another 10 kilometres further east to see the last rock art site of the trip, Kettara 02 in the lower reaches of the Oued Kettara. As we approached the level of the basement rocks rose, and one could clearly see that the sandstones were deposited over an uneven, eroded granite surface that had a considerable relief. The site itself is situated in a shallow and rather uncomfortable shelter at the junction of the sandstone beds and the underlying igneous rocks, accessible over a scree of large eroded basement blocks.

The paintings were on the rear wall of the shelter, partially exposed to the sun. After all the spectacular sites we have seen it was rather a let-down, with some faded and rather crude scenes of types which we have seen elsewhere in much better condition.

Mid-afternoon we returned along our route from Aloba, then turned south across the plain towards Bachikelé the locality with another spectacular natural arch where we intended to spend our last night in the Ennedi. Along the way we encountered a fennec (Vulpes zerda) peeking out of its den. We have seen a few before from a distance, this one finally remained curious long enough for a couple of photos before disappearing underground.

As we approached the southern part of the plain the country became visibly more populated, with herds and scattered houses everywhere. We reached the small village of Bachikélé in the late afternoon, from where the big arch clearly visible in the distance.

As we stopped to check-in with the village foreman, a sand snake (Lytorhynchus diadema, Gábor found the same species coiled under his sleeping bag during our 2012 Tassili trip) slithered out of the sand just adjacent to the car, twisting itself into a knot in the process. It is completely harmless, but that did not dissuade the villagers from beating it to death with large sticks as soon as we moved on, there was nothing we could do.

The Bachikélé arch is the product of a very different type of erosion than the river carved arches we have seen earlier in the day. It is the remnants of a collapsed sandstone mesa perched on the top of a conical granite hill, probably it has a few hundred years left at most before these last remnants also fall down due to weathering. Once the protective sandstone is gone, the remaining granite weathers rapidly, leaving only a sandy plain.

We made camp in the sandy bed of a small wadi below the arch, then all scattered to various vantage points to make the best use of the last hour of sunlight.

We remained at our various perches taking in the amazing view and snapping away until the last of the sun disappeared and it was time to return to camp to pitch the tents before it became pitch dark.

Day 15. – Bachikélé - camp before Kalait

The night was the coldest we experienced on the trip, and it remained chilly well into the morning, it was the first time that we had any use for the warmer gear.

We were in no particular rush, and as this was to be our last camp in the Ennedi we took some time to explore the surroundings in the morning light. A few hundred metres from our campsite, on a terrace formed at the junction of the granites and sandstone, there was a fairly large ancient settlement with tumuli and circular buildings, probably of medieval age.

After packing camp and taking our group photo we moved back to the village, where we picked up the foreman and his grandson to be our guide to visit the Bachikélé guelta, located at the end of a very picturesque valley with lush vegetation and a flowing stream. We left the cars at the valley entrance where the stream disappeared into the sand, and started walking upstream along the riverbank, passing flocks of goat and sheep taking their morning drink.

As the vertical walls of the valley closed in to form a narrow canyon we met a herd of polychrome cattle which could have just walked off one of the rock paintings. These, together with the ones at Archeï are presently the Northernmost surviving cattle in the Sahara, and the entire environment of the Bachikélé valley is a very good analogy to the mid-Holocene conditions in the central Saharan massifs.

We spotted some wildlife too: a few dassies (rock hyrax, Procavia capensis), and more surprisingly a band of olive baboons (Papio anubis) which climbed the near-vertical cliffs with amazing ease, getting out of camera range almost as soon as we spotted them.

The canyon continued to narrow, and some proper broad leaved bright green trees appeared along the banks of the river. The stream started to fill the sandy riverbed, and we had to wade across the ankle-deep icy cold water at several spots.

As we neared the guelta, we passed a large herd of camels taking a pause on their way to the water. They soon caught up with us again, making a loud splashing and gurgling spectacle as they passed along the stream.

Following the camels we soon reached the guelta wedged in between two vertical walls of rock at the end of the canyon. It is fairly large and deep fed by a permanent spring, the outflow creating the stream that runs the length of the canyon. With all the vegetation it is a much more pleasant place than the better known counterpart at Archeï.

The herd that passed us was the first to arrive at the guelta, but soon several more herds arrived in waves, each accompanied by only one or two herders, and the small space in front of the water was filled with jostling animals. Despite all the commotion and apparent chaos, each herd managed to take their turn at the clean water without mixing. Soon the firstly arrived started to leave, making room for the newcomers. It was clearly a well rehearsed exercise, the sight and accompanying sounds made by the hundreds of animals was an amazing spectacle.

Click image to see high resolution WMV video (2:08, 157 MB). For codec problems or slow connection use links below:
High resolution MP4 (132 MB) | Low resolution MP4 (54 MB)

We too had to leave as our time in the Ennedi was up, we had to start on our return journey towards N'Djamena. We followed the camels as they slowly marched out of the canyon, spreading out to graze among the trees once the valley opened up.

On our way out we met the cattle again, the herd has grown considerably since we passed, now counting more than a hundred animals. They exhibited the same variety of hide patterns and horn shapes that may me observed on the cattle paintings all over the Sahara.

We dropped of our guide and the little boy at the village, then continued out towards the plain. On our way we saw several more herds of cattle, already taking their midday rest in the shade of the meagre acacias.

For a while we were following the bed of the oued marked by a row of acacias, then turned West across the flat grassy plain, passing among the last major ramparts of the Ennedi plateau towards the featureless flat country beyond.

We stopped for lunch among the last outlying sandstone towers marking the edge of the Ennedi, beyond which the plain is only interrupted by a few low granite knolls. Unexpectedly we found that the surrounding low terraces were filled with hundreds, if not thousands of round tumuli. They ranged from two to 7-8 metres across, some just conical heaps of stone, but others with a well built low circular retaining wall. Clearly this was a major cemetery from some ancient period, however there were no clues to establish their age. I am not aware of any publication mentioning this major site.

Not far from our lunch spot there was a hill with a conspicuous vertical rock face. Forfeiting lunch I went to check it out for paintings, but it was completely empty. The walk was not entirely in vain though, as the terrace provided one of the best exposed sections of the boundary layer between the archaic basement and the overlying Palaeozoic sandstones. When these sandstones were deposited during the late Cambrian, the basement was already much eroded, covered with rounded quartz pebbles which formed a red cemented conglomerate as the sediments were deposited.

A kilometre beyond, at the base of a spectacular rock formation there was a large conspicuous shelter. It was getting late, but the shelter looked very inviting so I made a quick dash, dodging a herd of grazing camels along the way. It was certainly used for human habitation, attested by several small stone structures as well as some crude flakes littering the shelter floor and the surrounding area, however there were no paintings except for one barely recognisable blob which turned out to be a recent camel.

After our lunch spot we continued across the grassy plain towards Kalait. At first we mostly saw grazing camels, but as the Ennedi disappeared behind us we saw many dorcas gazelle. After about 70 kilometres we reached the edge of the shrubland and with it several Sahelian bird species appeared: a pair of Nubian bustards (Ardeotis nuba) and several flocks of helmeted guineafowl (Numidie meleagris) among the low acacias.

We made camp at the base of a low ridge just off the track about 20 kilometres before Kalait, the last feature on the plain before the settlement. The ridge was composed of almost pure white fused quartz crystals, the fragments of which littered the ground about our camp.

Day 16. – Kalait - camp after Abéche

In the morning we quickly covered the remaining distance to Kalait, where we had no intentions to linger long. We only needed to uplift fuel for the roughly 230 kilometres remaining till Abéche, a much quicker affair than filling all our available tanks and canisters on the onward journey. In the town we had a surprise encounter with a French party led by Yves Gauthier also returning from the Ennedi, including Suzie and Gérard Lachaud, our AARS friends. Once fueling was complete we continued along the track south, passing several of the big all terrain trucks and the row of wells surrounded by animals as we entered the permanently inhabited zone.

We drove along the track for the rest of the morning, stopping for lunch in the shade of some very meagre acacias among a cluster of granite hills after the town of Biltine. We were in cattle country, there was nothing unusual about having herds all around us while having our lunch, but the behaviour of one small herd was quite exceptional: unlike their brethren on the plain below, they climbed the slippery and rolling granite blocks of the steep hillside with surprising agility, more like a flock of goats. They were the proof that cattle could actually tackle some of the steep rocky paths leading to water holes at Jebel Uweinat.

On the hill adjacent to our rest stop Andrea spotted a flash of colour in the otherwise dull grey and barren landscape. It turned out to be a low shrub with some big bright pink flowers, totally out of place in this environment. It was later identified as "baobab du chacal" (Adenium obesum), a species that is much in favor among bonsai enthusiasts.

We continued to Abéche for a quick stop at the gas station, then we set out immediately along the road to reach a camping site outside the densely inhabited areas before sunset. There were still some cattle about, but we mostly passed harvested millet fields full of scavenging storks. Finally we reached open country and made camp some distance from the road at the foot of a granite ridge.

Day 17. – Abéche - camp after Bitkine

We had a lovely crisp morning, the surroundings of the camp were surprisingly pleasant given that we were well within the inhabited zone. We could afford a relaxed start as we were well within our schedule to reach N'Djamena by the afternoon of the following day.

In the sharp morning light we had another good look at the strange dense black rock which Ursula found at Nola Doa. As much as we wished it to be a meteorite, on close look there were several things that did not look right. While one surface showed signs of melting, even with a naked eye it was possible to see small air bubbles in the molten crust - something that never occurs in meteorites. Also there were clear crystals inside an internal void which do not occur in meteorites. On the other hand it did not appear to be slag, which would show signs of melting throughout, while this rock was clearly composed of well formed crystals, something that can only form during a slow cooling. The exact nature of this rock still remains a mystery at time of writing (March 2017). It is weakly magnetic (a strong magnet sticks to it without falling, but immediately jumps to an approaching iron object), most likely it is related to iron smelting (Pier Paolo Rossi confirmed that there are traces of ancient iron workings in the area North of the Fuchs panel at Niola Doa). If anyone has any ideas, please let me know.

As we left camp, we realized that hundreds if not thousands of storks (Ciconia ciconia) hid all around us in the harvested millet fields. They were everywhere we looked, this area of Chad is one of their northernmost wintering grounds, but many of them remain here year round.

The rest of the day was spent with the long and rather unexciting drive back towards N'Djamena. We passed the town of Oum Hadjer, then turned south west following the asphalt road that makes a big detour from the direct route. By lunchtime we reached the Abou Terfan mountains, we had a quick stop in a bay along the Eastern side of the hills. There were many big granite boulders at the base of the mountain bordering the little plain, but there was no trace of any prehistoric habitation at this spot.

Soon after our lunch stop we spotted another Adenium obesum by the roadside, bearing a much denser cluster of flowers than the one we saw near Biltine the day before. Subsequently we passed many more growing among rocks in the low granite hills, but this one was by far the finest example.

Passing Bitkine, the granite mountains came into view in the best afternoon light. We stopped at several spots to take photos, but it was too early to make camp, we had a good two hours left till sunset which we could use to allow more time for showers in N'Djamena before the flights the next day.

Our last camp was in the bush about 80 kilometres beyond Bitkine, as far as we could tell away from the immediate vicinity of any village.

Day 18. – Camp after Bitkine - N'Djamena

Other than the drive back to N'Djamena, not much happened that is worth mentioning. The biggest excitement of the day happened soon after leaving camp, when we spotted a big lappet-faced vulture (Torgos tracheliotos) perched on a tree just by the road. Unusually it did not alight immediately on our approach but waited long enough for us to get close for good photos, then it made a broad circle with a low pass before disappearing over the bush.

As we were on the main road now, we also passed at regular intervals some customary monuments to human ingenuity on how to wreck vehicles. One truck in particular surpassed all imagination, leaving us wondering how on earth did they manage to achieve this.

Approaching N'Djamena we passed a very fine herd of long horned cattle, a breed not seen anywhere along our route (or on other trips to Chad). Again, they could have walked right off some Saharan paintings, especially the scenes of the Iheren shelter in the central Tassili N'Ajjer spring to mind.

We reached N'Djamena by mid-afternoon, and settled into the lobby of the Novotel, taking turns to shower and re-pack our gear before taking the overnight flights back home.


On this voyage we have just received a small glimpse of the wealth of rock art sites littering the landscape in the Fada region. Many of these sites, recorded by Gérard Bailloud and others, have never been re-visited since the nineteen fifties. Another expedition is planned to the Ennedi for January 2018, focusing on re-locating the most important "lost" sites, plus of course to visit all the well known localities. Please visit the planned itinerary page for more details (or "like" the FJ Expeditions FaceBook page to receive notices of news and updates).