When faced with the problem of surveying several thousand metres of new cave passage beyond a fairly long sump, in a limited number of trips, it became obvious that a BCRA grade 5b survey in time for this journal was out of the question. In fact, forcing oneself to take any survey notes at all requires a considerable degree of self-control, when there are still unexplored passages to be walked up ! Consequently, the technique decided on right from the beginning was a compromise between speed, accuracy and available manpower.

On the first survey trip, I surveyed through the sump using standard CDG methods and then down the streamway using a careful 'compass and pace' technique. In this way, about 1.5 km of passages were covered in about seven hours work. The method can be surprisingly accurate if the surveyor understands a little surveying theory and knows the likely sources of error.

This survey was drawn up to give us a rough idea of where everything was under Leck Fell before accurately locating six points along the streamway using Bob Mackin's 'Molefone'. The end of the survey was found to be out by about 80m, which is quite acceptable for a BCRA grade 2 survey over this distance. The coordinates of all the survey stations were then recalculated by Andy Waddington. Where a junction was reached by three distinct paths from the accurately located points, the position was determined by a least-squares analysis, then errors within traverses were distributed linearly. Thus the main line of the survey may be considered fairly accurate, yet having consumed the minimum number of man-dives.

Most of the other side-passages have been surveyed since, using a careful compass and pace method. Although the ends of most of these inlets have yet to be radiolocated, their positions as shown are not likely to be seriously out of place. A full grade 5b survey will have to wait until several other interesting sites in the cave have been given full attention, but will probably happen a lot sooner if the long sought-after dry entrance is found. Meanwhile, the possible link with Lost Johns' will probably need Molefones in both caves simultaneously in order to pinpoint exactly where the connection lies.

I would like to record my gratitude to Bob Hryndyj for his excellent final drawing made from my preliminary doodlings, when I had to dash off on the French cave diving holiday (oops ! 'expedition')