The last day of July looked OK. I had been a bit unwell over a long period and had less than ten hours in the air this year, flown no UK cross country for fifteen months and done precious little abroad. A good fly about would be something.

In a light westerly the lift was coming and going making people were wait until they were sure of staying up before launching. I figured there was enough and lobbed off at about two fifteen to find it quite hard work but good fun, with small cores popping up here and there. It's always interesting trying to work small cores. If they are strong enough, very small cores can be sometimes be worked by flying a figure of eight with only the 'cross' of the 'eight' in the core. Such a core may not produce a climb, but may prevent sinking and maintain contact while it develops into something better. Excellent practice for co-ordination too, but be very careful. There can be very large surges as you hit the core on each pass and turns need to be tight. When does a steep turn become a wing-over? Shortly before you wake up in hospital!

After enjoying the agility of my Whisper a while, I managed a heavily banked climb which gave me a few hundred feet. The thermals became wider and it was possible to circle more. Core edges were nicely defined enabling me to form a good mental 'map' of their shape. It is curious that some days I tend to fall out of thermals upwind and some days downwind. That day I seemed to be doing both, but could usually re-locate them without difficulty.

A number of people were airborne when one of the cores kept going and I was away. Around 700 feet or so there was some to-ing and fro-ing as bits seemed to break off the main body of the thermal. Others soon joined me making it easier to sort out, and a small group of us were soon outclimbing some sailplanes nearby. My brain is not what it was (no comments please!). I cannot remember how the sky looked before that time, but the area behind the edge was now filling up quickly with a murky and confused cloud base. OK, it was me that was confused, but cloudbase was murky and not particularly high for almost the whole flight. In what must have been a corker of a thermal, Liz Sampson, who had been quite low, suddenly shot up to join us. We neared cloud base, I think about 1500 feet ATO, when the UV protective visor I have to wear misted up. This was to happen whenever I approached cloudbase. Clearly (or perhaps foggily!) my visor requires further development.

If we stayed put, we would probably just have dithered around or ended up in cloud together. I headed downwind towards the thermal with the sailplanes, now below us, and booted the speed bar almost fully on. The speedbar served several useful purposes. Flying towards a thermal I believed was almost bound to work, I wanted to be there as quickly as possible making some height loss of no real concern. I was also headed directly downwind, with the possibility that streeting between thermals might push me up into the murk. Extra sink from the speed bar would help to keep me lower. It feels good too!

As it was, I lost only a little height and the lift increased steadily as I approached the thermal. Arriving not far below cloud base, I made a few circles to top up and look around. The horizontal visibility was poor making it impossible to see the clouds themselves, but a shadow on the ground suggested there was a cloud downwind. Still concerned about keeping low enough to see OK, I booted off again. The cloud did not work very well and I was soon looking round for something else. Checking behind me, the others had moved downwind, but did not seem to be doing much better. There was no point in going back. Figuring it best to stay over high ground, I opted to move south towards a marginally better looking cloud near the mast over Eyam edge. It was hard work. The others seemed to be still struggling behind me so, when I eventually found a fairly stable core, I made a radio call. No-one answered so I left them to it. Having heard nothing on my radio for the whole flight, I wondered if it had failed.

In the slower thermal I drifted towards Froggatt and a pattern began to emerge. The air to the south was clear, giving me a side on view of a street of nice pyramidal looking clouds. Unfortunately the air between me and them was probably clear because it was sinking. This obvious temptation had to be ignored and I had to concentrate on the stuff I was in. Close by, the rising air was murky higher up, obscuring nearer clouds. However, cloud shadows on the ground downwind suggested I was in a street running east. Although I could not tell the vertical shape of the clouds from their shadows, hopefully my street would be similar to the nice one I could see to the south. On the whole, the south side of my street was working better than the north side, probably because the street itself was shading the ground under its north side.

Observing this pattern, I applied the 'energy line' concept I learnt about when flying sailplanes. Essentially this involves visualising a line across the sky along which you think the air will be rising and trying to fly along it. I flew east, following cloud shadows and darker bits of cloudbase where I could see them, thermalling now and again to top up. Soon I was over the north edge of Chesterfield. Here things were less encouraging with the street breaking up and forming stratus which shaded the ground. I slowed down and looked around from a thermal. There was simply no good stuff to go for and no clues in the cloud shadows. Time to go into 'stick with it, drift with it' mode, hang on to what lift I could find and change gear mentally. Instead of romping down a street, I had to slow down, have a good look round and, perhaps, enjoy the view a bit. With much of the ground now shaded, I was reduced to sticking with a core, drifting towards the motorway and gradually losing height.

'Stick with it, drift with it' is all very well, but not always so easy to apply. I was sitting in a wide weak thermal beneath an overcast and it did not seem a good idea to leave to try for a chancy ground source. Nevertheless, I was still going down, things on the ground were getting very big, and the looming motorway had to be taken seriously. I had been picking landing areas some time when, gradually, the thermal started picking up a little. My sink rate slowed. Almost imperceptibly, I began to climb. Barely daring to hope, I concentrated only on milking the thermal. Painfully slowly, the ground pulled away and I cleared the motorway. To the south I could see a paraglider in a rather final looking glide, I think Captain Jean Luc Boudin of the starship Mac, who soon disappeared in the poor visibility. Another appeared and disappeared to the north, probably Dave Martin. My climb was still very slow, but I seemed to be in the right spot.

The stratus was clearing a bit from the south. With a little more height I worked my way towards the clearing sky and was briefly rewarded with a better climb before it all went dead. Clearly on my way down, I headed for some sunlit ground and found a tilled field with trees on the downwind side. A short search revealed the trigger spot, a clump of trees on a slight hump. I managed a gradual climb by repeatedly moving upwind to my trigger as each weak core died. Hard work and slow again, but it was sunny and things were looking up. A reasonable core developed and connected with a street taking me north of Mansfield. A nice save.

The sky was filling with cloud. I toyed with the idea of trying to fly south where it looked a bit clearer, but as earlier, this was probably a red herring. Resolving to stick with my easterly track, I looked for signs of further streeting downwind. Near cloudbase, I went for a likely looking shadow and arrived to find the cloud dying. Trying another to the south, I got a good climb and found myself under a street again. The sky further ahead was covered with stratus which would end my flight but it was filling up all around me anyway. I stuck with the street which seemed to extend into the stratus with a sunny slot on its south side. I passed the end of the sunny slot expecting the lift to end, but was pleasantly surprised to find the street continued to work well into the fully overcast area. No cloud shadows to help here, just plodding east trying to 'feel' the line of lift as reduced sink, looking for areas of darker of cloudbase and thermalling where I could get a climb. I have never flown so far under complete cloud cover before.

Eventually the street had to end, and finding what was to be my last climb, I made a few turns in cloud. Making a glide east, I emerged to see that continuing east would bring me down somewhere just after the Trent. To my south was the possibility of Caunton microlight field where, perhaps, Andy Buchan might find me a lift back. I settled on flying southeast, getting as near to Newark as I could safely land and going for the train. After a long, quiet glide I landed in a cut field at South Muskham at five to six.

An ex motorcycle racer, whose friend flew microlights, had seen me land and kindly took me to the station. My previous rail excursion from Newark involved several examinations of the same bit of the 'Route of the Flying Scotsman'. Doncaster, Peterborough, Doncaster again (overnight on the station forecourt!), before getting the train to Sheffield and Hope. This time I was careful not to get the wrong train at Doncaster, and arrived in Hathersage just after nine, where Liz picked me up. There was a talk that night at Camphill about GPS. I just caught the end. A pity really. Maybe I could have done with GPS in all that clag. Nevertheless my trusty map and magnetic compass had got me a good flight. Difficult and hard work perhaps, with most of it less than 2000 feet above take-off, but thoroughly enjoyable and very satisfying. And all without help from my magic cross country pullover.