Muztagh Tower, northeast face. At 7,284m, Muztagh Tower is a prominent landmark on the way to most of the 8,000ers in Pakistan, yet there have not been many ascents, and there has never been an alpine-style ascent. Our goal was an alpine-style ascent of the unclimbed northeast face. Pavle Kozjek, Grega Kresal, and I set base camp on the Biange Glacier. During our approach trek, Grega had gotten increasingly ill, and after a couple of painful days, we decided that it wasbest for him to be airlifted from base camp to Skardu and then home.

With our friend safe, Pavle and I could concentrate on the mountain. We set our advanced base camp 30 minutes from the northeast face on the Younghusband Glacier. After acclimatizing, sleeping at 6,050m on a neighboring mountain, we prepared for the climb. Our goal was to be as fast as possible, so we climbed unroped. We carried only the bare minimum: a 110m, 5mm Kevlar rope, four pitons, four screws, a stove, a tent fly, gels, and drink mixes.

We started climbing at 2:30 a.m. on August 24. The first snow- field went smoothly, and by dawn we were over the first rockband (WI5 M5). Clouds enveloped us as we continued to the top of the second icefield (55°-75°) and reached a maze of thin ice, snow, and steep rock (sustained M5). Our pace had slowed a bit, partly because of the lack of oxygen and partly due to the more committing climbing. The last 150m were 80° snow, and it took some effort to dig ourselves a path to the ridge crest at 6,800-6,900m.

We found ourselves on an exposed ridge riddled with cornices, and buffeted by a cold wind. Because we topped out at 7 p.m, we decided to dig a snow hole on the ridge. After two cold hours, we were cooking inside our hole, feeling quite alright.

In the morning Pavle suggested going down, via the 1956 French route on the southeast ridge. As I was the rookie in the Himalaya, I went along with his decision. The day was cloudy with wind gusts up to 80 km/h, and we began moving quite carefully along the southeast ridge. Pavle was carrying the rope inside his pack. Not long after we started the descent, at 10 a.m. on August 25, as Pavle approached the edge of a cornice, it gave way and he fell to his death. All of a sudden I was alone, with only the top fly of a tent, a gas canister but no stove, and a satellite phone. I phoned Grega Kresal, who was already recovering in Slovenia; he was one of the few people who could understand the position I was in. I knew I was too high to be picked up by helicopter; I would have to descend on my own at least to 5,500m.

It took me 15 hours over two days to descend to ca 5,400m. The first 400m were the hardest—harder than the climbing had been on the way up the northeast face. Meanwhile, I waited for news about a possible rescue. My hope was that a helicopter could at least drop a rope for the icefall at the bottom of the mountain. If nothing could be arranged, I would have tried to descend on my own, but it would have been very hazardous in the steep, heavily crevassed icefall. The phone battery died in the evening of August 26, and the weather was poor on the 26th and 27th, but I knew the forecast was good for the 28th, so I waited. Over two days I created two liters of fluids, mixing snow, isotonic drinks, and urine. I was doing what I needed to survive, but I also knew there were friends in Slovenia and Pakistan doing everything they could—dealing with permits, raising money, and arranging for Tomaz Humar, as well as some friends climbing in the Charakusa Valley, to fly to my base camp as quickly as possible. Humar was very determined and would not take no for an answer when people tried to get in the way of the rescue effort.

A chopper picked me up at 11 a.m. on August 28. I survived without major consequences: I lost 12kg and had some mild frostbite.

The line I climbed in 17 hours with Pavle is named Magic Lost (1,800m, VIWI5 M5), after the loss of a great magician whose company it was my privilege to enjoy.

Dejan Miskovic, Slovenia