Click on image for list of participants

Tadrart & Tassili N'Ajjer, South Algeria 9th November - 5th December, 2018 False colour images processed with DStretch, a freely available software developed by Jon Harman

In the first part of this expedition we mostly followed the route of our March 2017 trip, but also taking the opportunity to visit sites and areas that were missed prevously. In the second part we returned to the Taserert (central Tassili) to visit some amazing sites that we by-passed during our November 2016 journey.

Day 1. – Djanet - Tin Aressou

Arriving from various parts of the world to Algiers airport, we all gathered at the cafeteria of the Domestic Terminal the day before. Our travels were mostly smooth, the only casualty was one of Mike's bags that was left in Paris by Air France, but in Algiers it was promised that it will be sent after us to Djanet the next day. We took the 10pm Air Algerie flight via Ouargla, reaching Djanet in the small hours. The first day was rather the same as on previous trips, after a decent rest at the Essendilene premises we went to the market to purchase the supplies for the trip while the paperwork was being prepared, and set out with our two cars after lunch towards the Tadrart with Abdallah and Hama, our two drivers. Much to our surprise the graffitti covered sign of two years ago at the junction where the pistes leading South diverge was freshly re-painted, giving clear directions to anyone who may pass here without knowing where the road is heading. As we continued South along the edge of the Tassili plateau we noted that the fresh green vegetaton was much denser than at any time previously, there were very heavy rains during the summer all accross the Tassili reaching down into the Tadrart.

We continued without stopping to our campsite at Tin Aressou in a little valley leading a short distance into the low plateau away from the road. With only an hour left till sunset we immediately set out to visit the two prominent shelters above a broad terrace flanking the valley, with a large round tumulus under them. With the sun shining directly onto the main panel we just enjoyed the place, leaving the bulk of photography for the next morning. However this direct light also helped to reveal some details which were not evident when we visited in 2017.

Day 2. – Tin Aressou - Oued Beridj

We could affort a lazy morning as in any case we had to wait for Mike's bag (which was already confirmed the day before to have reached Algiers) to arrive, so we had a good long photo session at the shelters, taking in all the finer details of the lovely Iheren style lion hunt scene and all the other surrounding figures.

Sufficiently saturated with the paintings, we packed camp and backtracked along the road to Djanet for ten kilometres to the low basalt mound of Tanaout, with almost every block on the top of the hill covered with engravings. This is a really strange site, as there is nothing remotely similar along the edge of the Tassili. Why this particular inconspicious hill was chosen, and by whom, remains a mystery.

While we explored the mound and the engravings, Salah arrived with the missing bag, much to our relief (especially of Mike, who had to do without a sleeping bag and tripod...). We immediately set out towards the Tadrart, driving pretty much non-stop till the checkpoint at the entrance of the Ouer Beridj, which we reached after about an hour. This time the military appeared very relaxed, after a casual glance at our papers we were quickly waved through. We continued a short distance into the oued to halt at the first site, the "shelter of the hands" (Faleschini's site B) which needed a lengthy search to find in 2017.

We continued to the mid-section of the valley, where a huge rock wall near the cluster of sites (Faleschini's site C) provides the best shaded lunch spot in the whole valley. In the shade of the rock we found many mushrooms (Podaxis pistillaris, superficially very similar to the shaggy inkcap), another sign of the heavy summer rains. After a quick meal we set out to see the sites, starting at the late pastoral paintings just around the corner (Faleschini's site C2).

The most interesting site of this cluster is certainly Faleschini's C5 along the southern side of the large sandy bay, with a number of giraffe and large white cattle, which were superimposed by later smaller cattle.

We completed the round of the sites with the row of engravings (Faleschini's C11-C16) along the western side of the bay, then walked back to the cars which by now were packed and ready to leave.

Our next stop was the large panel of engravings near a natural arch (Falschini's site H) with the numerous round vessels and some fine cattle in a confusing jumble of superpositons. For some reason both Faleschini and the Halliers consider the round vessels to be a part of the "Roundhead" artistic tradition (whatever they mean by that) and by implication archaic, though taking a close look it is clear that whle the large cattle are indeed superimposed, there is a smaller cattle figure under one of the round vessels.

Continuing downstream we made a short stop at a fine panel of historic engravings at the confluence of the main and southern branches of the Oued Beridj south of Marka Ouandi, which to my knowledge have not appeared in any publication, but were reported by other visitors some years ago.

The magnificent panel of engraved cattle of Marka Ouandi, executed in the finest Messak style, were just a short drive away. The light conditions were not the best with the sun behind the rock, we immediately decided that a re-visit at sunrise is definitely warranted. Despite the lack of good contrast, DStretch gave surprisingly good enhancements to some of the photos, revealing the two large human figures and the giraffe at the right of the scene which are almost completely invisible in natural light. It is curious that despite being the finest piece of rock art in the Oued Beridj, this panel is not mentioned by either Allard-Huard, Muzzolini or Hachid, apparently it was only discovered in the last decade by some tourist group, who probably assumed that it was a long-known site.

We made camp in a small side valley just North of the engravings, planning to use the remaining two hours of daylight to search the area for the Iheren style paintings reported by Gouarat in Sahara 5. We already searched in the vicinity of the engravings in 2017 without finding anything, this time we concentrated on the valley further North, where a large prominent dry playa is visible on satellite imagery. Despite a rather thorough search and several suitable-looking shelters, the East bank of the oued yielded nothing but magnificent scenery. With the sun already having disappeared behind the cliffs, we made a rapid search of the west bank on our way back to camp, and in a small hidden side valley we did come upon a long and deep shelter at the base of the cliff, with some rather weathered pastoralist paintings along the rear wall, but certainly not the fine scene we were looking for.

Just as we were about to leave, Hideharu spotted a series of dark paintings on the ceiling, which on close scrutiny turned out to be some rather bizarre, archaic-looking scenes, very hard to make any sense of. A few metres away there were more figures in the same style, but this time a discernible pair of animals, definitely in the archaic style known from esewhere in the Tadrart. While the site is definitely not published anywhere, we were not the first visitors, as a little cairn was built right in front of the main concentration of paintings. We had to return to camp with the approaching darkness, but we planned to return in the morning to take detailed photos.

Day 3. – Oued Beridj - Tin Uded

In the morning we went back with Hideharu & Mike to the panel of engravings, hoping that the rising sun will provide a good sidelight for a short period, however it was far too late in the season, probably this only happens between april and september when the sun rises further North. The light was even duller than in the afternoon, we lingered for some time then returned to pack camp, checking out some shelters along the way without any luck.

While we were away the others fared better, on the slope above our camp they found a shallow shelter with some very delicate engravings of animals and a human figure on a vertical rock section.

After packing and loading our two cars we moved to the site found the previos evening to take a thorough look. Reviewing the rear wall in better light revealed nothing more than the few pastoralist figures already noted, but the ceiling revealed an amazing detail of bizarre shapes and symbols, unlike anything I have seen before. There were also several drawn hands (not handprints) and a figure that looks like a child's drawing, with stick-like hair radiating from the head. While the animal figures have similarities with other archaic paintings, the more abstract elements have no ready analogies.

While we still did not find the site reported by Gouarat, our time allotted for the search was up and we had to move on. The information given by Abdallah (which invariably differs from guide to guide) was that the term "Marka Ouandi" refers to the entire lower section of the Oued Beridj from the big panel of engravings all the way to where it opend to the plain facing the dunes. If this is true, the paintings we sought could have been anywhere along a winding 7-8 kilometre stretch of the valley. It would need several days to thoroughly search the entire area.

Our next objectives were the co-joined valleys of Iberdjen Uan Tabarakat and Tin Uded, exiting the Tadrart plateau towards the East a mere 6 kilometres to the south of the mouth of the Oued Beridj. However the Eastern edge of the Tadrart is innudated in several parts by the golden dunes of the Erg Tin Merzouga, we had to cross a roughly 2 kilometre wide belt of low dunes to reach firm ground on the other side. Abdallah and Hama very sensibly hid our reserve fuel at a suitable spot before the dunes to lighten the load, then we set out for the roller-coaster ride over the dunes. The route was well travelled with many clear tracks, in about 30 minutes we cleared the obstacles with a few repeated attempts at the higher crossings.

Once clear of the dunes, we continued to the end of the Oued Iberdjen Uan Tabarakat, until we reached the grove of bright green tamarisks that give the valley the name. Along the western side we aimed for a small shelter that was known to contain a painting of a chariot. The scene turned out to be exceptionally fine, and much to my surprise after the trip I recognised it to be the subject of a 1995 article by Muzzolini et. al. (Sahara 7) that was described to be in the Oued Beridj, unknown to any of the guides there.

Just a couple of kilometres further, in a broad side valley we reached one of the principal objectives of this trip, tha large shelter of Ouan Tabarakat. As it was past midday, we went past the site to take a quick lunch at the only shaded spot in the vicinity, then returned to the enormous shelter half filled by a large sand dune. One only starts to appreciate the huge dimensions of the shelter after ascending the dune that extends deep into the overhang, which more resembles a cave than a typical shelter.

This cave presents the best example of the strange engraved "Kel Essouf" (evil spirits in Tamashek) figures, found and published by Najib Ferhat, Karl Heinz Striedter & Michel Tauveron quite recently, in 2000. Reaching the top of the dune, after one's eyes become accustomed to the low light, they suddenly appear on the vertical wall facing the dine crest, much larger than one imagines them based on published photographs. They are totally different than anything encountered in the Tassili and the Tadrart, and only appear at a handful of sites, including the Acacus in Libya and the Djado in Niger. Some authors (Ulrich & Brigitte Hallier, Jitka Soupokova) consider these figures to be among the most ancient rock art in the Sahara, with some attributes suggesting they could be ancestral to the classic Roundhead paintings. While this latter assertion is still up to debate, they certainly do appear ancient, and are quite clearly superimposed by pastoralist paintings.

The Halliers and Soukopova saw here "Roundhead" style figures superimposed on some of the Kel Essouf figures, a key argument in their chronological assessment. Taking closeup photos of the left part of the panel, and making good use of DStretch, it is possible to discern many paintings, including a number of handprints and cattle, and several indistinct red lines which on close scrutiny prove to be several giraffe. Two giraffe are clearly superimposed over one of the Kel Essouf figures, however the giraffe, executed in a fine line drawing, are quite clearly of the Iheren style. There is another giraffe with a white infill and red contours superimposed over several of the engravings, it is very indistnct howeve several similar pastoralist giraffe are known throughout the Tadrart. There is one archaic-looking, hard-to-discern painted animal figure at the right of the shelter, which could be partially obliterated by a Kel Essouf figure, but equally it could have partially weathered away, the evidence is not conclusive.

While the Kel Essouf are certainly the stars of the show, there are several other interesting details in the cave, including a group of unusual (and probably late period) archers, a group of faint but fine hippotragus (closely resebling those at Tamrit), and two strange creatures which appear to be barbary sheep on account of the prominent mane, however strangely they lack the characteristic horns.

Just past the col leading into the next lateral valley to the south, there is another smaller shelter, which on first look only contains some very fresh looking, fine late pastoralist paintings, including a very nice couple. However on the left wall below the couple there are some very faint pastoralist paintings, DStretch reveals two distinct layers. A fine herd of older cattle is superimposed by a lovely scene of an Uan Amil style cow, giving birth, with a human figure with the typical head knot approaching to assist.

There is another smaller shelter with pastoralist figures at the rear of the hill with the large cave, where we stopped on our return.

This list site concluded the list I was aware of, we retraced our inward journey in the valley, stopping at a high vantage point overlooking the cluster of tamarisks and the encroaching dunes.

Earlier in the day as we entered the valley, we encountered another tourist group who halted at a low shelter near the valley entrance, where none of us were aware of any paintings. We stopped to investigate on the return, and found a number of crude but very well preserved late pastoralist paintings inside, not mentoned in any publication.

Not far from the last visited site the Oued Iberdjen Uan Tabarakat joins the Oued Tin Uded. It is hard to say which is really the principal watercourse, while Ouan Tabarakat is broader and appears to be the main valley, the water channel in Tin Uded is clearly larger on satellite imagery, and the latter also penetrates deeper into the Tadrart. We turned left into Tin Uded, and drove to a cluster of sites first reported by the Halliers in their 2014 book on the Tadrart. We made camp at the foot of a high rock wall opposite the site referred to as Tin Uded I.

With still plenty of daylight left, we immediately walked over to the large shelteralong the edge of a side valley. There are innumerable but unfortunately very weathered paintings for a good 20 metre section along the rear wall. The best preserved ones are at the leftmost extremity of the shelter, with a number of superpositions showing a long sequence of paintings, including some that are clearly archaic.

With the approaching sunset most of our party returned to make camp, but wth Hideharu we could not resist the temptation and climbed the steep dune leading high above the valley floor to an isolated deep shelter, Tin Uded II, containing some remarkable and well preserved archaic paintings. We spent a long time taking photos, only retreating to camp when it became too dark inside the shelter to discern the paintings.

Day 4. – Tin Uded

The Halliers reported an impressive panel with Kel Essouf type engravings and an elephant figure, Tin Uded III, on the large rock outcrop on the far side of the main valley opposite our campsite. The previous evening we were unable to locate it, in the better morning light we set out to find it, while some other members of the party ascended the steep dune to Tn Uded II. The engravings proved elusive, we checked out every suitable looking rock face, until finally we found them, almost completely invisible in the very bad light conditions. Possibly earier in the year with the sun rising further to the North the panel receives direct sunlight early morning, but at the time of our visit it remained in the shade throughout the day. It was just possible to make out the many Kel Essouf figures engraved high up on the rock face, identical to the ones seen at Ouan Tabarakat the day before.The elephant below them (and close to the present ground level) is probably much more recent.

After packing camp we continued upstream in Tin Uded. Just a short distance from camp we found the shelter published by the Halliers with some fine late pastoral figures.

A few hundred metres further there are a cluster of engravings form all ages, ranging from tifinagh inscriptions to a fine cattle, and a rock just around the bend of the valley with several engraved vessels of the same type as the ones seen in the Oued Beridj two days earlier.

The engraved vessels were the last site published by the Halliers, however there were a few more localities marked on the map of Striedter & Tauveron that accompanied their Kel Essouf article. One such site was marked about two kilometres further upstream, where we did encounter a triangular shelter with some crude giraffe engravings inside.

However the main site of this locality turned out to be just beyond, with a huge rock wall completely covered with large scale engravings on an elephant, several catle and two strange unrecognised creatures. These visibility of these engravings may be much imprved with DStretch.

While we were photographing the large panel, Abdallah searched the area and found a very fine panel of engraved cattle on the far side of the valley.

Continuing up the valley we found several other engravings, usually on prominent vertical rock faces recognisable from a distance.

We reached the end of the drivable section of Tin Uded just before a series of gueltas in the southern branch, with a large natural arch towering above them. Just upstream of the gueltas there is a huge shelter in the side of the valley.

Strangely while the shelter is very large, most of the rear wall is empty, one just finds a small but very fine panel of pastoraist paintings close to ground level at the right part of the shelter.

We had a quick lunch in the small sandy bay accessible wth the cars just before the gueltas, then set out on foot to explore the western branch of the valley (looking at the satellite imagery the main watercourse, penetratng deep into the plateau), forking a few hundred metres before the end of the track, blocked to vehicles by a number of gravel rigdes. Not far from the valley entrance we found a large shelter that looked promising, and indeed on close scrutiny we did find a large panel of weathered paintings at the right corner, including some faint dark archaic figures and some strange dotted patterns, to our knowledge not reported by anyone previously.

With the limited time we had, we could only explore the first kilometre of the valey, dotted with a number of gueltas and many promising shelters. After a sharp bend we came upon a large deep shelter, which however only contained a single faint white cattle figure.

Moving further upstream, at a bend in the distance we saw a row of large shelters, but they were all empty. However on the return there was a low shelter on the north side, missed on he inward leg, that did contain some very faint pastoralist paintings.

Returning to the cars, we started our drive back along the way we came, stopping at the fine large panel of engravings with the large elephant and cattle, which were now in a perfect light.

We looked at some more points marked on the Striedter-Tauveron map near the entrance of the valley, but found nothing other than an incredulous cluster of mushrooms poking out of the sand.

We drove out of the valley to the narrow plain bordering the Tadrart to the East, sandwiched between the rocks and the encroaching dunes. We visited the only known paintings in the area, a large shelter in the sde of an isolated rock tower with some fine late pastoralist figures.

We continued a little further south to a fine natural arch on a ridge, with a panel of engraved camels at the base of the opening. The location also provided a spectacular vista over the dunes and the plain in both directions, it must have been an excellent place to spot gam or to oversee the herds below.

With the day slowly coming to the end, we drove till the Northern end of the plain to make camp, leaving the crossing of the dunes on our way back for the following morning.

Day 5. – Moul Naga - Tin Tehaq

After packing camp we set out to cross the dunes in the lovely morning light, which we again accomplished without any mishaps. While the cars went to collect the stashed fuel, we took a little walk along the edge of the plateau, trying to locate an elephant engravng mentioned by Faleschini, but the only object of interest we found was a rather sluggish locust trying to warm itself in the hazy sunlight. Once the cars caught up, we continued to the fine late pastoralist paintings of Akaham Ouen Elbered.

I was now aware of another site less than a kilometre away which our guide did not know about on our previous trip, while our party was busy taking the photos and making a side visit to the nearby arch, I walked to the supposed location. The site was not hard to find in a large shelter, but it only contained some very faint late pastoralist figures.

We had another range of dunes to cross to reach Moul Naga, a lovely basin surrounded by huge dunes swelled up against the ramparts of the Tadrart. As we reached the last dune crest, we found to our amazement that the playa at the bottom of the basin was full of water after the rains of a couple of weeks ago, provding a unique vista not unlike the lake at Sossusvlei in Namibia after better rains.

This time we did not linger long, but continued to the upper reaches of the broad valley (that seems to carry no name on any map I know) leading in to the plateau towards the south west just a few kilometres to the north of the Oued Beridj. At the point where the valley pranches several unpublished sites have been reported by earlier travelers. The largest was a fine shelter along the east bank, containing a number of engravings and late pastoralist paintings, including a very nice barbary sheep huntng scene.

On the far side of the valley a large shelter contained some more late pastoralist paintings, including some crude giraffe, and a number of recent handprints accompanied by tifinagh script.

We returned a few kilometres along the western side of the valley to another cluster of unpublished late pastoralist sites. The first proved to be a single engraved cattle, quite hard to find among a jumbled mass of boulders some way up the hillside.

The second site of the cluster was obvious, a large and deep cave facing a large dry playa, with sevelal panels of paintings inside.

The third site of this cluster proved to be the finest, again an easy to find site in a spacious shelter on the top of a small isolated hill on the far side of the dry playa, in plain sight of the revious cave.

By this time it was well past midday, we crossed the broad lower section of the valley covered with dense green vegetation, and set out to search for a shady spot along the corridor leading north to the Oued In Djerane. Finding the regular lunch spot taken by another tourist group, we moved to the East side where a spur offered some shade, and were rather pleased to find a series of fine camel period engravings scattered all over the rocks.

After lunch we continued north, leaving the sites of In Djerane for the return leg of our journey. We made a brief stop at the mouth of In Djerane, which was also densely covered with vegetation, a stark contrast to the barren playa a year earlier.

By the time we reached the huge pyramidial rock of Ouan Iska (Allard-Huard's site G13) it was fully in the shade, the best time to view the very fine herd of elephants and giraffe on the two sides of the rock. On these panels DStretch comes very handy, clearly revealing all the details that can only vaguely be seen with the naked eye.

We moved on to the isolated rock tower of Ouan Zawatan, containing arguably the finest single engraving in the Algerian Tadrart, a splendid bull executed in a raised relief, in the classic Messak style. There are some more engrafings around the rock, including a fine rhinoceros in a hard to reach crevasse Behind a small promontory we came to another high shelter in the side of the cliff, with a few rather meager paintings of cattle, humans and a single recognisable giraffe. Neither this one, not the shelter we saw before it, appear to have been published anywhere.

On the far side of the large playa of Ouan Zawatan (a dune confined basin filled by the runoff from the Oued In Djerane) there is another rock tower containing engravings of an elephant and an ithyphallic figure (Allard-Huard's site G14). Bernard Foullieux also noted some Kel Essouf figures on this rock, but we failed to find them. Now I know we looked in the right place, but they are so weathered that only a perfect contour light at the right time of the morning reveals them.

We reached Tin Tehaq well before sunset, allowing time to look for the sites which we failed to find in 2017. As we made camp a little further in the valley than last time, a rather gloomy looking thick cloud cover moved in from the west, dark enough to bring the prospect of some rain in the evening. Nevertheless we set out to locate the site with archaic paintings reported recently by Jitka Soukopova, a double shelter along the valley leading towards Bouhediane. The right shelter only contains the herd of archaic looking animals (oryx or perhaps ibex?), while the left shelter contains some pastoral scenes.

By the time we returned to camp the bulk of the clouds passed, and with them the threat of rain. It turned out to be a pleasant warm evening, and as we finished dinner a large dark beetle wandered into the circle of light. Surprisingly it was not one of the usual darkling beetles but a true scarab (dung beetle, Scarabeus cristatus), the first I have encountered anywhere in the Sahara proper.

Day 6. – Tin Tehaq - Tin Merzouga

During the elapsed time I have received some more precise information on the location of Allard-Huards site P4, for which we searched for in vain during our last visit. It is a smallish shelter in the side of the rock along the large playa, which again was full of green vegetation.

On the rear wall there are a series of rather faint late pastoralist paintings, including a human figure and a fine dama gazelle published by Allard-Huard. However apparently she failed to notice a group of ostrich, including a very fine facing pair, located on a smaller inward-facing panel along the side of a crevasse.

On a slightly higher level there is another smaller shelter with some white painted figures, and a fine cattle is engraved on a vertical rock wall some distance to the right of the main shelter.

Leaving Tin Tehaq we continued North accross the low pass to the Oued Bouhediane, stoping at the natural arch marking the southern entrance (inhabited by a very friendly agama) and visiting the guelta to replenish our water supplies.

Our next stop was the large shelter of Bouhediane (Allard-Huard's P5) with the numeros late pastoralist paintings, followed by the small nearby shelter with more lat period paintings, and the lage strange shape or object which appears to have some small human figure(s ?) associated with it.

We had lunch in the same small side valley as in March 2017

Not far from the shelter a large flowering Solenostemma argel plant was attracting a swarm of butterflies. Most were the yellow (female) and white (male) African migrants (Catopsilia florella) mixed with a few Painted ladies (Vanessa cardui), most of them with the annoying habit of refusing to sit still long enough for the camera to focus properly.

After our lunch break we continued North in the Oued Bouhediane, halting at the large shelter containing the pair of very fine giraffe. While we photographed the site, Abdallah explored the upper level, where he found another shelter with a few very faint paintings and many cupules and grinding basins on its floor.

Leaving the main valley we ascended the low bordering plateau to visit two more shelters with paintings, the one with the very fine Tin Abenhar style pastoralist paintings, including a pair of humans riding a cattle, and the nearby shelter with the exceptionally large late pastoralist human figure.

On our way to Tin Merzouga we passed by the peculiar eroded rock resembling a hedgehog (or pig?), then the mushroom shaped rock with the pair of fine engraved rhinoceros just in front of the huge dunes.

We reached our campsite at Tin Merzouga just in time to be able to climb the nearby ridge for the sunset spactacle, as the yellow dunes slowly turned orange then pale red before the sun disappeared behind the Tadrart.

Day 7. – Tin Merzouga - Oued Tidunadj

We were up at dawn, but the anticipated sunrise was not as spectacular with thin high altitude clouds creating a dull haze.

After packing camp we started our day at the fine paintings of Ouan Atan near the entrance of the Oued Sirik, a few kilometres to the west of Tin Merzouga.

The Oued Sirik (or rather Oued Tidunadj) is one of the five major valley systems of the Algerian Tadrart, wedged between the Oued In Djeran to the south, and the Oued Arrekine (which is now the main military access way to the border zone with Libya, hence off limits) to the North. Like the Oued Beridj, it was only explored in its entirety in 2007 by an Italian party led by Guido Faleschini, with their results published in Sahara 19. Abdallah was of the opinion that the term "Oued Sirik" only refers to the lower region of the valley system, downstream of the confluence of the two branches, and the upstream regions are called Oued Tidunadj, the toponym used by Ferhat et. al. in their 1997 article on a panel of engraved cattle found in the valley (but there is no mention of any of the other sites later reported by Faleschini).

Entering the valley we stopped at the large shelter with fine pastoralist paintings (Faleschini's Sirik 2), this time taking some better photos of the very faint but fine elephant we noted on our last visit.

While we were busy photographing the paintings, Abdallah - who certainly had a flair for exploration - scouted the far bank and came upon a very fine engraving of a rhinoceros (accompanied by a fainter elephant) which was not reported in any of the publications.

In March 2017 we did not plan to visit this valley (hence came unprepared), and our guide assured us that there was nothing of importance beyond the large shelter with paintings. At the time We only havd a couple of hours to explore further during the lunch break, and did find the site reported by Ferhat, Striedter & Tauveron, plus site 4 of Faleschini, clearly indicating that the valley has more to show. This time we started with these sites, but with the plan to continue beyond, where Faleschini reported a further ten sites, with a sketch map giving a fairly good idea of their location.

Site 5 of Faleschini was just opposite site 4 in a side valley. The brief description of 'some wild fauna' in the Sahara article certainly did no justice to the very fine panel of engravings, dominated by a large rhinoceros that was hard to photograph in its entirety in the confined space.

Further inside the side valley we were expecting to find shelter 6 with the engraving of a large cattle, which Faleschini considered their finest find. We did come upon a large shelter, which however contained several panels of late pastoralist paintings, plus a panel of engravings in the left corner. After consulting the article it became clear that site 3 and site 6 have been mixed up (however I did check out the locality marked as site 3 the year before, without finding anything).

Faleschini reported a further site (Riparo 7) with engravings in the same valley, however we failed to find anything at the marked location. A little beyond we did come upon a long shallow shelter with a single painted cattle, and on our return we found a large panel of engravings which the Italian party seems to have missed, as no site descriptions match the depicted scenes.

By this time our trust in the accuracy of the Faleschini paper was greatly diminished, and after coming to another shelter near the marked location 8 which proved to be empty except for some traces of paint, attention soon turned to an improptu game of boule promted by the abundant colocynths growing in the riverbed. We continued further upstream, trying to find a shady spot for lunch near site 9/10. We halted at the only possible spot at the foot of a low cliff facing a large dune, and some confidence in the Sahar article was confirmed by finding site 9 nearby, a small shelter with paintings right where the map suggested it should be.

After the lunch break we started out on foot towards the presumed location of sties 10 & 11, with the cars slowly following. Both were found at the expected locatons, being prominent shelters (more like caves) along the northern side of the valley, with only a few engravings and paintings.

Goind deeper into the wadi we reached the location of site 13, with a large shelter visible from a distance along the northern bank. Entering, we found to our surprise that it was not one of the usual shallow shelters, but a true cave with a domed roof at leat 25-30 metres high, with a huge mound of bat guano occupying the centre. However after an exhaustive search we found no rock art of any kind, this was clearly not the site with paintings we were looking for.

The true site was a little upstream of the cave, in a shallow shelter at the rear of a little bay, with the paintings visible from afar.

Faleschini published two scenes from this shelter, one including an elaborately enhanced negative handprint and a cow giving birth. It is interesting to note the apparently skirted human figures in the scene, apparently associated with the hand, which bear resemblance to the Niola Doa style female figures of the Ennedi, and are quite unlike other depictions of the Tassili-Tadrart region. The other scene was a group of fine and elaborate archers.

However to our surprise and delight there was more in the shelter than what the article presented. At the lef side we found a number of archaic looking yellow depictions that were quite hard to make sense of. At the extreme right, there were several patches of red dots, plus another of the large unclear shape/object the like of which we have already seen at the second Bouhediane shelter.

There was one further known site upstream (Riparo 12), another large shelter which contained just a single scene of a late pastoralist couple, but exceptionally fine and well preserved.

We re-traced our inward route for a few kilometres until a major fork, entering the southern pranch from where only a single engraving of elephants was reported (Riparo 14). It was fairly easy to find, right under a large promnent shelter visible from a distance. Faleschini did not note, but there were also some very faint late pastoralist paintings on the rear wall of the shelter, only proberly visible with the help of DStretch.

Having finally located the bulk of Faleschini's sites we returned to the mid-section of the valley, making camp next to a towering rock spire which also contained a pair of engraved giraffe which were apparently not recorded by the Italian team. We spent a leisurely hour in the fine light before sunset to set up camp and explore the area.

Some clouds moved in as sunset was appoaching, and for a few minutes we were treated to a wonderful display of Kelvin Helmholtz clouds forming above the plateau (as two layers of different celocity air were mixing), perfectly backlit by the setting sun.

Day 8. – Oued Tidunadj - Oued In Djerane

In the morning after packing camp we drove out of the Oued Tidunadj, stopping near the entrance where Hama vaguely recalled seeing a single engraving. We did find a rather curious nonkey-like creature, possibly modern, and certainly unlike any other rock art in the valley. Crossing the dry playa at the oued entrance we continued back along the Oued Bouhediane, taking the outer route via teh Tin Tehaq arch (where we met a fairly large group of French tourists, some with their wn vehicles).

We continued to the nearby site of Ouan Tikal, with the unique engravings of animal tracks. While such depictions are common in southern Africa, they are very rare in Saharan rock art, and such a concentration is without parallels anywhere in the region.

We returned to the entrance of the Oued In Djerane, to see a small shelter with paintings which I was made aware of before our trip, not appearing in any publication. The shelter overlooks the withe plain and playa at the mouth of In Djerane, and contains a few very weathered pastoralist paintings.

It was only a short drive to the well known little shelter with pastoralits paintings in the lower Oued In Djerane. Abdallah showed us a detail that most visitors (including ourselves the last time) miss, a series of very fine engravings high on the vertical rock face a fw metres to the left of the shelter.

Continuing down the row of sites dotting the Oued In Djerane, we stopped at the unique fish engravings and the adjacent large panel with elephants chasing a hunter (Allard-Huard's site G10).

The Oued In Djerane appeared very different compared to our last visit. The centre of the valley was now full of green vegetation, the result of the heavy rainfall a few months earlier. The bulk of the vegetation was Cassia italica intermingled with a few other species.

The next obligatory stop was the row of fine cattle (Allard-Huard's site G3) along the Northern bank. Nearby, Abdallah showed us another cluster of engravings which we missed on our previous visit.

We continued past the rather fresh looking giraffe and hunter site which is one of the most photographed site of In Djerane, then the huge panel of elephants (Allard-Huard's G4), both along the southern bank, then crossed to the northern side for the site with the half-submerged giraffes (Allard-Huard's G2). In the adjacent valley again Abdallah showed us some more engravings that our previous guide did not know about.

For lunch we stopped at the large shelter opposite Allard-Huards P3 site in the small northern tributary of In Djerane, with the fine white giraffe depictions on the rear wall, including one that is uniquely depicted in a full gallop.

The lunch break also offered plenty of time to visit the principal paintings on the far side of the small valley, with the rhinoceros and the fine Iheren style cattle and human figures. Taking some more detailed photos than last time, it was possible to identify two indistinct blobs as human figures, DStretch revealing the very fine red lines of their arms, legs and faces.

Near the shelter we encountered a very friendly female agama, which tolerated the cameras rather patiently, allowing multiple close-up shots without running away like the majority do.

After the break we continued a few kilometres west in the main valley, entering the side valley with the two large shelters with pastoralist paintings on the opposing sides of the valley (Allard-Huard referred to both as site P2), starting with the larger one on the northern side that contains many fine scenes, including what appears to be a butchery scene.

The second shelter on the south side contains fewer paintings, but these too are fine including the depiction of a round hut, a less common theme in the Tadrart.

It was appraching sunset by the time we reached the last of the rock art sites of the Oued In Djerane. Just before the large shelter with the fabulous painting of a crouching giraffe giving birth (Allard-Huard's P1) we found a rock with an engraved giraffe that was later matched to Allard Huard's site G1 which we were unable to find on the previous visit. Finished for the day, we made camp in the same small side valley nearby as on the last Tadrart trip.

Day 9. – In Djerane - Tin Hanakaten - Djnanet

Before returning to Djanet we made a detour to visit the sites of Tin Akaham and Tin Hanakaten, both located along the edge of the low inner Tassili, relatively close to the point where the route towards the Tadrart and Libya makes the ascent. Tin Hanakaten was the site of the only serious archaeoligical excavations u tertaken in the Tassili, led by Ginette Aumassip in the nineteen seventies. Despie the relative closeness to Djanet, possibly due to the obscurity of the resulting publications, very few people visit the site, and most guides do not know the locality. Abdallah visited Tin Akaham before, and vaguely recalled the location, but not Tin Hanakaten which we had to find ourselves, based on Aumassip's rather inexact position and our best guess on the likely location. Getting to Tin Akaham proved straight forward, a track leaving the main road led right towards it, and once we apprached the locality the huge shelter with the semicircular row of stones in front of it (unclear if ancient or modern) ws readily visible from the distance. Strangely there are no paintings inside the shelter proper, only a panel of giraffe and some superimposed darker pastoralist figures on a semi-exposed wall outside the right side of the shelter.

On the far side of the valley, in a small niche which is hardly big enough for someone to crawl into, Abdallah found the finest depiction of the locality, most likely a large aurochs (wild cattle) bull, a very unusual depiction (with the only parallel at the nearby Tin Hanakaten).

While we were busy taking photos, Abdallah found a third shelter which he was not aware of, with some faint pastoralist scenes, including a depiction of what appears to be a two-headed catle.

We took the well traveled track leading to the south east, and after ten kilometres reached the oued which was the most likely location. Driving in, we soon came upon the huge double shelter, with the excavation trenches clearly visible in the larger one.

Aumassip was mostly interested in the archaeology of the Tin Hanakaten, and spared little attention to the paintings which were mostly damaged and very faint, hard to decypher. However with DStretch many fine and unique details are revealed, including a tree surrounded by sitting women, with utensils hanging from the branches, and a very fine pair of muscular (Iheren style ?) male figures.

The smaller shelter to the left of the main shelter contains some better preserved late pastoralist paintings, including a chariot depiction, which received the bulk of attention in the Aumassip articles.

Like at Tin Akaham, the finest paintings are not in the main shelter, but in a smaller one about a kilometre away. Fortunately the large double semicircular rows of rock in front of the shelter are visible on Google Earth, and we had a reasonably good idea of the likely location. Inside the shelter there is just a very small but extremely fine and well preserved panel of a row of human figures and at the right another large bull (again possibly aurochs, rather than domesticated cattle), prompting Aumassip to refer to the site as "l'abri du Taureau". Again the previous publications do not do full justice to this delightful little scene, which is among the finest rock art anywhere in the Tassili n'Ajjer.

We expected the search for the sites to take much longer than it actually did, it was still early afternoon. With the prospect of a warm shower at the Essendilene premises, after a quick lunch we decided to make an early entry into Djanet after nine days of camping in the desert.

Day 10. – Djanet - Aït Talawaten - Djnanet

We had one day remaining before Hideharu had to leave us with the overnight flight to Algiers due to his limited vacation time, and Magdi was to join us on the same flight for the remainder of the trip. With a pair of different cars and two drivers both by the name of Mohammed we used this day to make the excursion to the lovely roundhead site of Aït Talwaten, anout 60 kilometres to the north west of Djanet, the only unquestionable classic roundhead site located at the foot of the plateau, a not on the top (there is a school of researchers who refer to all archaic sites, also in the central Tassili and the Tadrart as "roundheads", but I'm not comfortable with this blurring of the term, which I would only use for the classic Lhote sites near Djanet and anything broadly similar, like Aït Talwaten).

Having been tipped off by Jitka Soukopova, this time I also found the single human figure in the large shelter bordering the valley which I missed on both previous visits.

We were in no particular hurry, after lunch we had some time left to explore the surrounding valley, going in to areas not visited previously, but we found nothing of interest aside the abundant fresh vegetation everywhere and some fine Cruziana trace fossils. On our return we also visited a large guelta in one of the side gorges along the access route, which too was teeming with life.

For the late afternoon we returned to the environs of Djanet to see the engravings of In Debirene and Terarart in the best light. Unfortunately some clouds moved in and there was only a dull pale light at the fabled crying cows instead of the crisp sunset we were hoping for.

Day 11. – Djanet - Tazerouk

Hideharu departed in the small hours, and Magdi too arrived safely after spending a night in Algiers. We could afford a lazy morning after the interrupted night as our two drivers (Ibrahim, the father of Salah, the Essendilene office manager) and Meimouni (Salah's father-in-law) both returned from the Tadrart the day before after a week with a large French group, so they too needed some rest and a few hours in the morning to sort out their affairs before we could leave. After re-stocking our supplies at the Djanet market we departed around midday, stopping for lunch at Tim Ras before continuing along the road towards Bords el Hawass (the old Fort Gardel). As we drove along the edge of the plateau (which was amazingly green everywhere), the thin clouds became darker and darker, and by mid-afternoon it appeared distinclty gloomy, with a good chance of rain.

As we were fueling in Bordj el Hawass a howling wind struck us, stirring up the dust, but there was no rain from the dark clouds. It did not appear to be a too inviting evening to camp, but by the time we ascended the plateau and reached Tazerouk the wind subsided and the clouds too dissipated a bit, and we found a reasonably sheltered campsite at the foot of a rock a few hundred metres off the road, just out of sight.

Day 12. – Tazerouk - Tin Terirt - Tikadiouine

In March 2017 we spent a couple of hours at lunchtime to find the fine Iheren style paintings reported at Tazerouk by Alfred Muzzolini and Giancarlo Negro. We searched the valley adjacent to the large well known shelter with late pastoralist paintings, but found nothing. This time I was better prepared, knowing that there are two more shelters in the valleys immediately to the north. Our campsite was next to the first one, which did contain the group of human figures noted by Muzzolini, but in the high ceiling to the rght of the shelter we found also some very fine Iheren style scenes on the ceiling, including a superb hippotragus and an incomplete but very fine donkey, both of which were only visible proberly with DStretch.

After packing camp we drove back a short distance to approach the other shelter along the side of the adjacent valley, which contained most of the scenes noted by Muzzolini & Negro, including some strange late pastoralist figures, and several Iheren style figures, including a human battling a therianthrope.

Having found all the scenes we were looking for, we continued to the nearby engravings of Tin Terirt on the southern end of the Dider plain. The clouds cleared just long enough to get a few good photographs of some of the panels in the still low morning light.

Continuing along the road towards Illizi, our next stop was at the magnificent Tikadiouine shelter a few kilometres off the road in the similarly named wadi. Previously there was a fairly easy track leadind into the valley, but the rains of the late summer have completely rearranged the boulders in the riverbed, now we had to stop with the cars a good kilometre before the site and continue on foot among the abundant fresh vegetation. The site itself never ceases to amaze, despite numerous prevous visits I still found several small details which I have not seen or photographed before.

To catch up on our schedule, after a quick lunch in the Oued Tikadouine we continued non-stop past the village of Tasset and Afara towards the Ifedaniouene mountains.

We reached our destination a good hour before sunset, allowing a quick visit to the Halliers' "Cascade shelter" with the very fine Iheren style paintings, including the unique mother with child, and a scene that could possibly depict milking (but it is rather insistinct, even with DStretch. I also spotted a group of faint figures which I did not see previously, and were apparently not recorded by the Halliers either.

With the wind of the previous day it remained hazy for most of the day, but by the evening the dust settled and the clouds too broke up, with a lovely calm evening at our Ifedaniouene campsite.

Day 13. – Ifedaniouene - Tin Batoulete

In the morning we made a quick repeated visit to the Cascade shelter to photograph it in different light.

In March 2017 we spent nearly a half day searching for the shelter with archaic paintings reported by the Halliers' without success. Having realised subsequently that we were searching on the wrong side of the dry lakebed, this time we continued beyond the cascade shelter, and in the next larger valley we finally managed to locate the site, a huge shelter at the base of a cliff following the rising terrain. While there were several panels of faint pastoralist paintings along the entire lenght of the shelter, the archaic paintings were concentrated at the upper left, dominated by a huge animal (possibly hippopotamus) that is clearly superimposed by better preserved cattle. This animal has a number of analogies on the Tamrit plateau, and is normally associated with the Roundhead style. While in other instances I have reservations about referring to the archaic paintings of the central tassili as 'Roundheads' this animal is fairly convincing, and could represent the westernmost occurrence of true Roundhead art (however the normally associated human figures are lacking here). There are some more archaic looking animals to the right of it, however they are superimposed over Tin Abenhar style cattle, adding some further confusion to the already muddled chronology of the region.

With our objectives met, we packed camp and departed for the western part of Ifedaniouene, taking a more northerly route this time to see the "Shelter of the three dancing women", a late pasoralist site also reported by the Halliers' which we bypassed on our ast visit. As we drove along the valleys, in places we found an incredible amount of white Androcymbium wyssianum flowers and other fresh green vegetation, in one valley a veritable green field. After successfully tackling some sandy areas we reached the big shelter along the side of a solitary hill by midday.

The huge shelter contains surprisingly few paintings, most of them dating from the late pastoralist period, the finest being the group of three women which give the shelter the name. There is also a panel of historic paintings with tifinagh inscriptions and a rather bizarre human figure (of a type common in the rock art of the Aïr mountains in Niger).

We had lunch at the shaded base of a rock tower nearby, then continued towards the sites reported by Jürgen Kunz. Not far from our lunch spot we passed a large shelter with a large dune in front where no nown rock art was reported to our knowledge. Climbing up, there was a depression behind the blocking dune with much fresh vegetation, and an immense deep shelter. At first it appeared completely empty, but after close scrutiny I found two very faint iheren style cattle. Reviewing the literature it became apparent that the scene was recorded by the Halliers', but assigned to the "Shelter of the three dancing women" by mistake.

We continued to the large shelter of Uan Erasen (Weiresen) reported by Kunz a few kilometres away, one of the finest late pastoralist sites with many well preserved scenes, including several horse-drawn chariots.

The site of Intemeilt, also reported by Kunz was just a short distance away, with a cluster of archaic looking animals and the very fine barbary sheep at the right of the shelter. We reached the site a bit oo late, the right side was already half-exposed to the lowering afternoon sun.

For the night we continued to Tin Batoulete, near the Western end of the Ifedaniouene region. Along the way we stopped at the large open shelter and terrace of In Abtal, a site we did not see in 2017. It is another of the Halliers' sites with a mix of archaic and pastoralist paintings in a rather poor state of preservation.

Due to its sheer size we spent a long time at In Abtal, only reaching our campsite at Tin Batoulete at sunset.

Day 14. – Tin Batoulete - Oued Tasset - Baidakoré

In the morning we made a quick visit to the principal site of Tin Batoulete located quite high up among the surrounding rocks, overlook a deep valley draining to the north. Here too there was much green vegetation, and we also saw many caterpillars of the Painted Lady butterflies.

We returned east mostly the way we came, passing the Ifedaniouene mountains and Afara to the south, then making te steep ascent to the Lower Tassili which here appears more like a row of low hills, rather than a vertical escarpment.

We reached Ikadnouchére by midday, visiting the huge but very poorly preserved site after a guick lunch. The famous quadriga that has no real parallels was partially defaced by a red tifinagh inscription sometime during 2016. I'm not sure if the Park service did some cleaning, or simply the organic pigments fade quickly in the strong sunlight that hits the panel every day, but the recent graffitti was much fainter on this visit, enabling a somewhat better viewing of the chariot and the horses, especially with DStretch.

Our next stop was the immense Tedar shelter, best known for the fine Iheren style lion hunt scene. Unfortunately this shelter was in almost continuous use from prehistoric to modern times, and the fine Iheren style paintings are mostly covered with scenes of more recent periods. However on every visit I catch a couple of fine details which I have not noticed before.

It was late afternoon by the time we reached the Route National leading north towards Illizi. We just had time for a quick visit to the magnificent site of Tissebouk before continuing to Baidakoré for the night.

We found a very pleasant secluded stretch in the nearby riverbed just as the sun went down and the full moon appeared on the opposite horizon. As we were having our sundowners, we were treated to a rare spectacle as a huge meteorite passed low along the southern horizon, creating a bright streak that actually cast shadows despite the bright moon. It lasted about 10-15 seconds, too swift and unexpected for any of us to catch it on the camera.

Day 15. – Baidakoré - Illizi

We has our first really cold, cloudy morning with a lot of humidity, everyone took turns around the campfire to warm up a bit.

From our campsite it was just a shot walk to the Baidakoré shelter, a short way into a side valley. It is agan one of those sites which reveals tself slowly, with new details emerging on every visit. Most of the paintings are very faint, requiring DStretch to truly appreciate them.

Returning towards the cars we also found a small shelter reported by Giancarlo Negro with some late period paintings, which we failed to see on our prevous visit.

By the time we returned to the cars the sun was well up and everything was dry, we packed camp and set out towards the North, climbing the winding road to the top of the Fadnoun plateau. Our first target was a unique panel of paintings first reported by Alain Sebéé in the Oued Tissalatine, somewhere in the general vicinity of Kunz's In Akkarene. In the elapsed time I have received information on its location, in the little side valley of In Edoune (Hachid 2000) not far from a vehicle track continuing to a small village in the main valley. Leaving the cars we continued in the side valley, passing a small keyhole monument, reaching a large guelta with teeming green vegetation along its banks.

The shelter itself was along the near side of the guelta at the base of a low rock outcrop, with the panel of paintings hidden (and well protected) behind a palm tree growing in front of it. The paintings cover a small area, but are exquisitely fine, with a row of very elaborate Iheren style human figures (there are some much less well preserved examples in the Oued Telmest shelter on the Tadjelahine plateau) accompanied by cattle and sheep.

We drove north along the track as far as we could in the main valley of Tissalatine to get as close as possible to In Akkarene, located in a northern tributary. We stopped for lunch near a cluster of acacias offering some shade, and I took the opportunity to walk in t another tributary leading west, with a single known site (In Adjer Tighas) about a kilometre upstream, with apair of well preserved cattle and some very faint human figures.

Returning to the cars I noticed a low shelter along the north side of the valley close to its mouth which looked inviting. On first look it appeared empty, but deep inside I found some very faint patches of paint which turned out to be some cattle and human figures.

After lunch we set out on foot towards the fine elephant hunting scene at In Akkarene first reprted bu Kunz, which we visited with Magdi two years earlier walking directly from the main road. Now the distance was considerably less, only two kilometres along the bottom of the valley.

I was not aware, but near the junction of the track leading to the Oued Tissalatine, right at the side of the main road (In Djaran, according to the roadside sign) there are a series of fine engravings in two distinct clusters about a hundred metres apart, inclding a nice pair of hippopotamus (a rather rare theme), a lion and many other cruder animal figures.

Our last stop before Illizi was the shelter of Tin Amasekni, located just adjacent to the main road, yet apparently unknown to most passers-by. It is known to some guides, but I could not find any reference to it in any of the obvious publications (Kunz, Muzzolini, Hachid). There are some fine scenes including an elephant, a pack of horses and several other animals.

With all the sites to see, it was well int the evening by the time we reached Illizi and settled in to the Bou Naga hotel for the night.

Day 16. – Illizi

The flight of Christa & Christa back to Ajgiers was in the early afternoon, we spent the morning re-stocking our supplies for the third part of our trip, then going to the airport around midday to check-in for the flight, which was reasonably on time. With Ibrahim we stayed at the airport until the plane took off to make sure it is really gone. Once the flight boarded things turned out very cozy at the airport, with all the staff and police valking out to the apron to watch he plane take off (they too know that it is really gone when it is well into the air), inviting mee too to watch, bringing pack some fine memories from my aviation career.

Originally we planned to leave after the flight and camp somewhere near the road, but poor Michael picked up a bad stomach upset, and we decided to stay another night in the Bou Naga to help him recover, the few hours not really making much difference in our plans.

Day 17. – Illizi - Oued Imirhou

Our original plans for the third part of our trip was to return south, drive along the base of the Fadnoun to Imirhou, then take the old piste built in the nineteen thirties (the first 'motrable track' to traverse the Tassili) to the Dider plain, trekking to a number of rock art sites from the closest point accessible with the cars. However the summer rains have caused havoc with the unpaved tracks, Ibrahim received the information that the Imirhou-Dider piste was cut about one third of the way north from Dider and impassable at that point. As the sites adjacent to the track were our most important targets (and seemed to be reachable from the North) we decided to drive in, visit the sites, then drive back via Imirhou to the main road, flexibly adjusting our remaining plans to fit the remaining available days.

We completed the re-stocking of our supplies the previous evening, we only needed to buy fresh bread (was not so easy in the rather sleepy early morning Illizi) then set out along the lengthy drive southwards. All was fine till the bottom of the pass descending from the Fadnoun, but the track to Imirhou was in a dismal state, with deep ruts at every water crossing, many of the concrete passageways in the larger oueds covered with boulders or partially washed away, our progress was at a snailpace. After stopping for lunch we bumped along the track (at least there were work crews repairing it at several places, an encouraging sign for the future), sometimes leaving the new piste bult from soft shale (excellent going when dry and compacted, but easily washed away) to the old stony piste, which now was better going. At this pace, we approached Imirhou by late afternoon instead of the expected lunchtime. We stopped about ten kilometres before the village, where the track approached the broad Oued Imirhou near the location where Muzzolini reported a fine site with paintings along the riverbank.

When we passed here with Magdi in November 2016 I was vaguely aware of this site, but we did not have time to search with more important tasks ahead. Now I knew that in fact there were four shelters, the westernmost being the main site reported by Muzzolini. We set out in the valley, expecting to find the sites after a few hundred metres, but we had to go a good kilometre before we encountered the first two large boulders with some rather meager pastoralist paintings in the shelters underneath.

Perhaps a hundred metres further a small shelter under large rock presented a very impressive and well preserved panel of apparently Iheren style, but rather crude paintings, mainly small red human figures surrounding a lion and a sheep (published by Muzzolini).

The main shelter was another couple of hundred metres upstream, well over a kilometre farther than where I expected the site to be. It is a large impressive shelter at the base of the cliff bordering the riverbed, with numerous late pastoralist scenes, including three distinct two horse chariots. Unfortunately the unique lion hunt scene published by muzzolini, which was already damaged on the photo published in 1995 (but probably taken earlier) had been mostly destroyed about a decade ago, apparently the whole right part of the panel fell away (hard to say whether naturally or someone intentionally broke it off, but the cracking was already evident on Muzzolini's photo).

At the extreme right of the shelter there is a very fine Iheren style homestead scene, quite far from the other later paintings.

It was past sunset by the time we finished at the shelters, we made a hasty retreat to the cars and in the last light of the day searched for a suitable sandy patch along the riverbank for our tents, trying to be away from the riverbed gueltas and the sprawning mosquitoes (unsuccessfully, as it turned out).

Day 18. – Oued Imirhou - Tin Mousa

After days of cloudy and hazy weather we finally had a crisp clear morning. We packed up camp and crossed the Oued Imirhou along the track leading towards Tarat, then turned south along the old piste.

The piste itself did not appear any worse than it was two years earlier, but that was only because there was not much room left for becoming any worse. We crawled along at snailpace, in some places having to guide the wheels of the cars over the boulders. It took us the whole morning to cover the roughly 20 kilometres till the large shallow basin with abundant vegetation abeam of Tin Mousa, one of our princpal targets. On Google Earth it was possible to see a faint set of tyretracks leading from the basin till about half-way to the site, but as the total distance was only four kilometres it did not seem sensible to search for the start of the trail and struggle on with the cars. We chose a suitable flat spot for the camp, and after lunch we set out on foot towards the paintings.

For an hour we walked on a completely featureless plateau, then suddenly without much warning a huge chasm appeared ahead of us, the western tributary of the Oued Tizzeine, which is the main drainage of the Taserert, the section of the lower Tassili wedged between the Oued Imirhou (to the west) and the Oued Zarzawa (to the east). We reached the river about a kilometre upstream of the location of the site, and there was no obvious way down the vertical cliffs. However from satellite imagery we knew that a small tributary joins just a few hundred metres downstream, and indee there among the boulders and dry waterfalls we found a difficult but manageable way down to the bottom of the canyon, whick was dotted with large gueltas full of water, a leftover from the summer rains.

The large shelter of Tin Mouse was a litle beyond where we expected it to be, near the confluence of the tributary and the Oued Tizzeine (at least this name is used on the IGN maps). It is a long deep shelter, with the paintings concentrated in the deepest central part, visible from afar. If was first reported by Alfred Muzzolini in 1989 (Sahara 2), and it is one of the least known and visited sites of the Tassili.

The most intriguing part of the site is the large complex panel at the left of the main painted area of the site. Muzzolini observed at least four distinct styles appearing, including several large archaic figures and unclear symbols which he considered "Roundhad" (including the large and clearly visible mouflon in dark paint), superimposed by intermixed (and sometimes indistinguishable) Iheren and Tin Abenhar style pastoralist scenes and figures, with a number of late pastoralist figures completing the big picture. It is the Iheren style figures that are the strangest. While they convey the familiar elements of the style, there are a number of figures which are outright strange, like the two headed human figure or the "matryoshkas", and several of the figures are clumsily drawn, almost like a caricature, very unlike the usual refined and artistic figures one is accustomed too. There is also a cleal giant buffalo (Pelorovis antiquus), however it is hard to say whether it is to be assigned to the Iheren or Tin Abenhar style paintings.

At the right of the main panel there is a group of plumed figures which is hard to assign to any style, but by their state of preservation they seem to be late pastoralist, likely associated with the giraffe made out of white dots only (or the rest disappeared). There is also a curious figure in thin outline, and a group of fish, a very rare element among paintings n the Tassili.

The right of the shelter contains mostly the more usual late pastoralist paintings, intermingled with a few Iheren and Tin Abenhar style scenes. There is one older and hard-to-discern hippo looking animal under one of the group of Iheren style cattle.

We spent as much time as we could at the shelter, but the deepening of the shadws signalled the time to retreat. We made our way back in the gorge to the same pass where we descended, and continued on the plateau top till our camp, which we reached just around sunset.

Day 19. – Tin Mousa - Ihetsen

With the uncertainty of the condition of the track, we decided to go to our most distant point first to test the route, enabling a more leisrely return. We packed camp and set out on the 25 kilometres remaining till the closest point along the track to the site of Ihetsen. With the dismal piste, it again took the whole marning to complete this distance, we reached the starting point of our trek by midday. After a very quick lunch we set out towards Ihetsen a good 7 kilometres away, crossing the mostly featureless plateau.

We reached Ihetsen mid-afternoon after a good three hour walk, where we found the vallyey completely changes since our last visit. Now there was a huge deep guelta occupying the riverbed adjacent to the shelter, with a veritable meadow along the banks with a carpet of bright green grass dotted with innumerable white Androcymbium wyssianum flowers.

We spent the rest of the afternoon taking turns in photographing the amazing elephant hunt panel with the dozens of exquisite small figures, many of which are only revealed in full detail when looking at the photos with DStretch.

We stayed at the shelter with the paintings till dusk, only retreating after the light started to fade. We made camp in a sandy stretch of the riverbed some dstance from the guelta (anticipating mosquitoes, but there were none here).

Day 20. – Ihetsen - camp beside Imirhou-Dider piste

In the morning before starting out on the return journey we made another quick visit to the shelter, taking another series of photos under different light conditons.

On our walk back to the cars we stopped to see the sites we found in November 2016. At the one neared to Ihetsen we found no new details, but at the row of sites about a kilometre from the road we did find a fine Iheren style panel that I missed earlier.

We reached the cars by midday, and after some rest we loaded the vehicles and drove back north along the piste till a big bend which was the closest point to our target of the next day, Ouan Azawa. There was a conspicious rock outcrop nearby with a flat plain nearby suitable for our tents, we made camp there and spent the afternoonexploring the adjacent rocks.

At first we found nothing, but then near the centre of the rock outcrop we came upon a double shelter partially blocked by a stone wall, with a few broken metal utensils scattered about, signs of recent occupation. In the right shelter there were some barely distinguishable traces of paint, which when processed with DStretch revealed a row of archaic looking animals and some strange human figures.

On first look the left shelter looked empty save for a number of recent charcoal graffitti, however on close look it was possible to make out some faint traces of red figures on a patch of unweathered rock under the charcoal drawings and inscrtions, which turned out to be a very fine Iheren style scene. The posture of the humans suggest that perhaps this too cound have been a hunting scene invloving a lion, elephant or some similar dangerous animal, however the rock surface to the right is now completely deflated, erasing any traces of paint if they ever existed.

Day 21. – Camp beside piste - Ouan Azawa - Tan Timzar

At sunrise we set out on our longest trek of the trip, towards the site of Ouan Azawa near the Oued Zarzawa, about 14 kilometres from our camp. Our starting point was about two kilometres further south than on our 2016 trip, which not only made the walk slightly shorter, but enabled a much easier crossing of the Oued Tizzeine in the shallower upstream section. After the crossing we followed mostly the same monotonous route as previously along the rocky but generally flat plateau, only interrupted by a few shallow oueds and some rock outcrops with promising shelters, but all were found to be empty. We trudged along for about four hours, passing a number of prehistoric tumuli along the way, and by midday we were approaching our main destination.

In 2016 we searched for the sites of Ouan Azawa in vain, and only after returning did we find out that the location information was off by one minute, meaning that we searched a good one kilometre from the true position. This time we had the correct information, and it was not difficul to find the first shelter (named Ouan Azawa 2 in the 2010 article of Civrac. et. al. in Les Cahiers de l'AARS 14.) with a fine panel of Iheren style paintings.

We knew that the principal site, Ouan Azawa 3 had to be nearby. As we set out to search for it in the same 'street' among the rocks, we came upon a small shelter with a single archaic-looking painting of an indiscernible large animal. This site was not mentioned in the Civrac article, looking at the photos with DStretch suggest either a hippopotamus or possibly a rhinoceros.

The main shelter of the locality was indeed not far, practically behind the first shelter in the adjacent 'street' among the rock outcrops. It was a lond deep shelter, with a small depression in the middle of the ceiling hiding one of the finest Iheren style scenes anywhere, depicting a pair of elaborately dressed women, both holding a (largish) child both of whom uniquely are depicted with blond hair. The central figures are surrounded by several men (one also with blond hair) and a black dog with red eyes that could be a perfect illustration for the Hound of the Baskervilles.

Above this scene there are a number of very faint figures, the most discernible being a spear-wielding man at the left. The centre is just a very indistinct blob, but with DStretch it is just about possible to make out a very faint yellow lion, with two spears embedded in its neck (with streaking blood at the imact points), mauling a fallen hunter who is also bleeding profusely.

The shelter contains a number of other Iheren style scenes in varying state of preservation, including another lion hunt scene and a large ithyphallic figure of an unclear stylistic affinity.

Mid-afternoon it was time to move on, we collected our packs and covered the remaining one and a half kilometres to the Ouer Zarzawa, descending in a narrow ravine that was full of fresh green vegetation. The bottom of the valley was also filled with a near continuous row of gueltas, all containing ice-cold clear water.

We reached the huge shelter of Tan Timzar just before the sun passed below the western cliffs of the valley. Once the sun was gone I focused on a small scene at the left of the shelter which I completely missed on our last visit. Here, under some rather roughly executed baboon-like animals there is a very fine but almost invisible Iheren style scene, suggesting that the bulk of the other paintings in the shelter, despite their archaic appearance, post-date the Iheren pastoralists (this is something also suggested by their very fresh appearance and good preservation).

By the time we finished taking photographs it was almost dark. We made our camp in a patch of soft sand in the riverbed opposite the shelter, and with the aboundant driftwood about we soon had a nice warm fire going as the temperature dropped rapitly from a warm afternoon to a distinctly chilly evening.

Day 22. – Tan Timzar - Camp beside piste

In the morning we immediately set out on the good 16 kilometre return journey, climbing out of the Oued Zarzawa on a stepped rock wall right behind our bivouac, then seting course towards the cars.

We set aside time to explore the sand filled rock city about two kilometres to the suth east of Ouan Azawa, parts of which we already searched without any luck on our onward trek the previous day. Again we found nothing, despite a number of shelters with suitable surfaces for painting, for some reason the ancient inhabitants did not favor this very picturesque area.

By midday we reached the Oued Tizzeine with the large pools in the shallow gorge. This was a perfect place for a quick lunch and a longer rest, plus if one felt so disposed, a bath in the icy waters. Interestingly the larger pools were full of small shrimp-like creatures, which were later identified as a fairy shrimp species, Streptocephalus torvicornis.

With only about an hour left till the cars we could afford a lazy spell at this lovely spot, then gathered our remaining strength to cover the remaining three kilometres and easy flat terrain, reaching camp well before sunset.

Day 23. – Camp beside piste - Imirhou

We have visited all the sites we planned to to the east of the old Dider piste, this day was all about returning north alng the same route. After a good night's rest we packed camp and started driving at snail-pace along the dismal piste towards Imirhou.

By midday we reached the basin with abundant green vegetation to the west of Tin Mouse where we camped five days earlier. We stopped for lunch and spent some time to explore the surroundings. The most interesting find were the large black and green caterpillars, a few at first but then once knowing what to look for they appeared everywhere, feeding on the abundant Asphodelus tenuifolius plants. They were later identified as the caterpillars of the large sphinx moth, Hyles livornica.

It was late afternoon by the time we reached the crossing of the Oued Imirhou, marking the end of the old piste. It says all about the condition of the track that we needed the best part of a full day to cover a distance of forty kilometres. We filled our water cannisters in the clean and still flowing water in the oued, then continued some distance west along the Imirhou track, stopping at a suitable campsite at sunset.

Day 24. – Imirhou - In Tahadouft - Dider Plain

Along the piste we passed several road repair crews rebuilding the concrete passes in the oueds washed away by the summer floods. It gave a good chuckle to see the 50kph speed limit signs, as the condition of the piste permitted a maximum speed of 20-30, one would have needed to speed up considerably to comply with the signs.

Mid-morning we finally reached the asphalt road, from where it was just a quick short drive to In Tahadouft, the "rock city" close to the Tasset junction. With the local guardian we set out to see the Hallier's Tasset shelter, but along the way the guardian showed us two more sites, one which Koen found during his wanderings the year before but the other was new to us, in a rather unusual location on the wall of a true cave formed under a huge angular boulder, and to my knowledge never published.

Continuing among the maze of rocks we reached the Tasset shelter with the small but magnificent Iheren style panel of paintings, certainly one of the ultimate highlights of the central Tassili.

Returning to the cars we passed the wall with a number of archaic-looking and much weathered human figures, which the Halliers called the "gallery of ancestors". While they do look ancient, their overall style is very different from the classical Tassili roundheads.

After lunch and some exploring of the other sites at In Tahadouft we set out towards the Dider plain. We had enough time to make a little detour to anothe site not fart to the North of the main road, referred to as Tabarakat by Giancarlo Negro, and Tin Edjedjele II by the Halliers. It was readily accessible decaes ago, but since the construction of the gas pipeline along the road and the associated continuous barrier of the trench rubble, it is now only possible to reach the site on foot accross a dark featureless plain.

Approaching the "rock city" a number of shelters were readily visible along the western facing edge, and we did find several panels of rather weathered paintings in them, but none matching the fine panel reproduced by the Halliers.

It took quite a bit of searching to find the main panel among the maze of rocks, on the side of a rather unassuming shallow shelter facing east. The paintings are rather faint and damaged, but with DStretch all the finer details are revealed.

Not far from the main shelter we found a smaller one with only a single but rather curious animal figure on the ceiling, executed half in red, half in white paint. The hear is indistinct, the short tail might suggest it is a sheep, but the overall body prportions more suggest cattle.

The walk combined with the searching and photography took up most of the afternoon, it was fairly late by the time we returned to the cars to cover the remaining distance to the Dider plain, which we found densely inhabited by nomadic families and their herds (including some cattle, the first I have seen in the Tassili). It was quite hard to find a secluded campsite, finally we stopped in a small sandy patch not far from where the Imirhou piste reaches the plain. Our turning-back point was a mere 25 kilometre to the north, the destruction caused by the rains forced us to make a detour of almost 200 kilometres, and cost us a good day and a half.

Day 25. – Imesseridjen - upper Oued Zarzawa

In 1973 Ginette Aumssip published three sites in the central Tassili (Libyca XXI), the most interesting of which appeared to be the site of Imesseridjen with two shelters containing numerous paintings located somewhere to the south east of the enfravings of Tin Terirt, further upstream in the oued. The site seems to have faded from memory, the guards at Tin Terirt had no idea about the paintings, however recently I have received information on their whereabouts. On google earth it is even possible to make out a track leading upstream towards the location, thus in the morning we set out to get as far as possible. However soon after passing Tin Terirt the track became almost impassable, the rainwater having carved deep ruts in the place of the once easily negotiable track, so we decided to walk the last five kilometres in a spectacular landscape.

Approaching the indicated location we did come upon a very large prominent shelter, but our excitement was a bit premature. This was just a minor site with a few hard to decypher blobs of paint on the rear wall.

The principal shelter was just behind the corner, but we decided to continue to the more distant location about a kilometre further upstream in the oued. At the indicated spot we did find a prominent shelter at the base of a high cliff, which on close look was almost completely filled along its entire width by fine paintings of the late pastoral period (Aumassip's Imesseridjen 2.).

On our onward way we passed a couple of promising shelters aside the two published by Aumassip. On our way back we investigated them, and unsurprisingly we did find a panel of paintings right in the first one.

The next one, with a dune providing a convenient ascent, only had a single small panel of paintings, but those very fine: a pair of antelopes faced by a single hunter with a bow.

The main shelter (Aumassip's Imesseridjen 1.) only reveals it's true nature once one makes the climb to the shelter floor, high above the valley. From a distance the shelter looks sizeable, but once inside it appears immense, with a very deep recess in the central part, hence even with the south-facing orientation it provides a good shade at midday, expecially in the summer when the sun is high. It must have been an excellent living space, as attested by the paintings from numerous periods along its entire length.

On first look most of the paintings appear to be relatively young, but on close look one can find several Iheren style scenes scattered about the shelter, sometimes covered by later paintings. There is a particularly fine elephant, apparently by itself and not as a part of a hunting scene.

There are several chariot representations in the central part of shelter, including one of the most bizarre ones anywhere: there are a pair of harnessed giraffe, drawn in the same style as the horses of the regular chariots, being held by a human standing in front of a chariot. Whether this is an actual representation of an attemt to harness giraffe in front of chariots, or the scene has some deeper meaning, is anybody's wild guess.

The left of the shelter is shallower, but reaches for several dozens of metres, with a number of mosly late pastoralist scenes, and several more recent camel period depictions and tifinagh script.

We returned the way we came mid-afternoon, this time following a fresh looking footpath on the terrace adjacent to the oued, a much easier going than the densely vegetated riverbed. We passed a circular stone structure of an uncertain age, it looked similer to, but much more massive and better built than the common recent animal pens used to keep baby goats safe in the night.

Our plan was to have three days at the end of our trip to explore the sites around the small village of Tin Akn some 25 kilometres to the north east of the Dider plain. However with the necessary backtracking we only had one day left, which we intended to use to explore the upper reaches of the Oued Zarzawa, adjacent to the Tin Akn piste, where we hoped to find the site of Tin Leh-Leh (Boccazzi, Sahara 3, 1990). We drove accross the plain to the piste, and continued on it till sunset, making camp along the bank of the Oued Zarzawa fromwhere we intended to set out on foot the next day.

Day 27. – Upper Oued Zarzawa

It was the 4th December, we had our first really cold night, with a layer of frost on our black bags that were left outside the tent in the riverbed. Fortunately the sun made its mark rather quickly, bu the time we had our first morning coffee the frost was gone, with a markedly better outlook for the coming day.

Our plan was to search the riverbed upstream of our camp for a stretch of about 2-3 kilometres. When Aldo & Donatella Boccazzi visited Tin Leh-Leh decades ago, they started from the Dider plain with camels, and came upon the site after approximately 20 kilometres (in pre-GPS days). Looking at Google Earth a bend in the Oued Zarzawa seemed to match the description (and also Aumassip published a single bad photo in the same 1973 article, refferring to the site as Oued Zarzawa). After completing breakfast we set out, passing a large quelta not far from camp.

After about one and a half kilometre we came upon a large shallow shelter along the south bank that looked promising, and indeed we did find some very faint traces of paintings. Later DStretch revealed a very nice elephant and a group of Iheren style figures. However these were not the paintings we were looking for, we seem to have stumbled upon an unpublished site.

We continued upstream for another kilometre in the riverbed which here appeared as if it were paved, with smooth polished rock along its bottom. We found nothing on either side, and after a while the country opened up into a large basin with no suitable cliffs for any shelter.

There was a patch of eroded rocky country north of the riverbed, exploring this area was our plan B in case we would not find the sites in the valley. Sure enough, soon after enterng the rocks we found a shelter with some interesting paintings, but again not the ones we were looking for.

We continued systematically exploring the maze, splitting up to explore the numerous transverse valleys. We found a number of very promising shelters, but all turned out empty.

Finally Magdi had some luck, coming upon a very fine panel of archaic looking gazelle in a small shelter at the corner of one of the many crossroads.

By this time we reached the far side of the rocky area, in full view of our campsite. There only remained one small valley to explore, on entering we were immediately faced a large shelter, which again proved to be empty. However on the opposite side Koen found some very curious paintings in a small shelter.

The style of these paintings is really difficult to classify. Mostly human figures are depicted in a rather crude fashion, resembling some archaic paintings, which is also suggested by their state of preservation. In any case, it is hard to provide good analogies elsewhere.

With all of the target area fully explored all we could say with certainty was that Tin Leh-Leh was somewhere else, so we returned to camp for lunch and some head scratching.

The only remaining potential area appeared a small valley about 4 kilometres west of camp. Magdi and Mike elected to have a lazy afternoon in camp, Koen wished to explore the area around the quelta for wildlife tracks, so I set out alone for the last search. As I made a shortcut on the low plateau bordering the riverbed I unexpectedly came upon a very fine example of an "antenna monument", a tumulus with two long straight protrusions at an obtuse angle resembling an old fashioned TV antenna. Such monuments are quite rare in the Tassili n'Ajjer, but are quite common further west in he Sahara all the way to the Atlantic coast.

At the end of the shortcut I came upon a very large guelta in the riverbed, with thriving green vegetation along its banks, but strangely no other life was to be seen anywhere.

Continuing upstream I pased several more gueltas. Ever since we first saw them at Sefar in 2010 I have been on the lookout for tadpole shrimp (Triops granarius) but have not seen any. Now in the dried out bed of a sandy guelta I once more found their traces, though not live ones but the dry carapaces, nevertheless clearly recognisable.

In little over an hour I reached the side valley visible on satellite imagery, but it turner out to be much shallower than thought, with no cliffs along the sides where shelters could form. The only item of interest was a small patch of the bedrock in the riverbed, where glacial striations were exposed from the time of the Ordovican glaciation when present NNorthern Africa was located close to the geographical south pole.

On the way back to camp I explored a few rock outcrops without any success. On the horizon there was a conspicious "rock castle" about 8 kilometres away, too far to be considered - some other time perhaps.

Day 28. – Oued Zarzawa - Dider plain - Djanet

After another cold night at the same campsite we packed up to start our return journey. On the way we stopped to check out a few shelters, but found nothing more. We crossed the Dider plain again, then followed the asphalt road with a brief lunch-stop after Bordj Hawass. We reached Djanet mid-afternoon, with plenty of time to pack and get ready for the overnight flight to Algiers, then the connections home.

With the apparent ease of restrictions we plan to return to the Tassili n'Ajjer again in November 2019 for a trek along the length of the Oued Djerat and if possible, a visit to the classic Lhote sites on the plateau above Djanet. Please check the News page periodically or "like" the FJ Expeditions FaceBook page to receive notices of news and updates.