Catching a faint flicker of light from the corner of my eye, I shielded my powerful primary light with one hand and looked over to see my wife Donna about 50 feet away. Her primary light must have failed as she was holding a small backup light, and the 800 feet of cave line she had been reeling in was billowing off the reel and tangled around her fins. She was now hovering absolutely horizontal and motionless above the thick layer of fine silt covering the cave floor, which would destroy all visibility if disturbed. It was December 2004, and Donna and I had together been exploring a series of dry and underwater caves on Mexico's Yucatan Peninsula. We call these the Animal Caves, because every trip seems to introduce us to more of the local animals, ranging from tarantulas, boas and rattlesnakes, to the more benign characters such as tejones, jaguars, bats and tortoises. In this part of the world it isn't unusual to do your final decompression stop next to a boa or crocodile, as these seem to enjoy the warm clear waters in the cenotes, the surface openings leading to the underwater cave systems. In some cenotes which we know have a resident crocodile, it is more of a concern if you can't see the crocodile on your stop... We had spent several months completing surveys of our target zone, logging the location of clues such as depressions and breakdown areas, then plotting all of these over an aerial photograph to get a picture of what could be underneath the surface. We strongly suspected that there was an underwater cave system somewhere in the area, but up until a week ago all entrances to it had eluded us. The breakthrough came when we visited some local landowners to give them copies of our trip reports and maps - one of them announced himself so pleased with our work that he would take us to see a couple of small cenotes which he had found in the jungle. The first trip to these two cenotes, the Jaguar Cave and the Tortoise Cave, had been with just one small and very portable diving setup, consisting of a harness and BC and two side-mounted 5 litre tanks. At first glance the Tortoise Cave hadn't looked at all promising - a muddy puddle at the base of a small limestone bluff. The smell of decaying organic material when I entered the water and stirred up the silt didn't bode well for finding a big cave system, because cave systems here generally have clear flowing water which transports the decay products away. I was therefore somewhat surprised when I swum a short distance in and entered a large air dome with tortoises bobbing around on the surface, and indignant but curious bats looking down at me from their resting places on the ceiling. An even better surprise awaited when I submerged and tied off my line: a cave passage fully 100 feet wide, decorated with stalactites and stalagmites, and disappearing into the distance. This was our Christmas present for 2004, and it looked like it would take several months to unwrap fully. On that first dive, I only penetrated the cave system about 200 feet - enough to convince myself that there really was a cave (and that I wasn't dreaming), and that it merited returning as a team with proper backmounted cave diving equipment. I had also briefly dived a cave about 200 feet away on the other side of the breakdown area - named the Jaguar Cave because the smoothly polished shelf just inside the entrance to the dry section of it has obviously been home to a jaguar for many years. But the entrance to that cave was too narrow for backmounted gear and so we would leave it for another time. When Donna and I returned the next day, the first couple of hours were spent carrying double tanks and our other equipment half a mile into the jungle, and then assembling it. We were helped in this by Gabriel, the enthusiastic landowner, who cut the trail and carried more than his fair share of equipment (on a subsequent trip we were pleasantly surprised to find sturdy wooden stools standing by the water for us to set up on - luxury in the jungle). Ever optimistic, we had brought two large reels containing 800 feet and 1400 feet of cave line, knotted every 10 feet to measure distances for our survey. The dive plan was to follow the existing line to its end, then follow the cave as far as we could, in the expectation that it would trend in a NW or WNW direction towards another dry cave we had explored and mapped a few months prior. We also hoped that there would be another passage to the south, bypassing the breakdown area at the entrance, and connecting to the Jaguar Cave. If gas supplies allowed, we agreed to search for this when we returned from the far end of the cave. If we found a connecting passage, this would provide us with a way past the Jaguar Cave's restricted entrance, and enable us to explore it with backmounted doubles. To those unfamiliar with the Yucatan's underwater caves, this might all sound straightforward, especially as the water can be very clear with visibility exceeding 100 feet. But because of the way these caves have been formed (?diffuse recharge of groundwater by rainfall?), they are what the scientists call ?anastomosing maze networks?. To you and me, they are a maze with passages frequently dead-ending either at a solid wall or a crumbling breakdown pile. Although the water is initially very clear, exhaust bubbles hitting the ceiling dislodge fine silt particles which rapidly reduce visibility to an inch or two in caves which are not frequently dived - and the Tortoise Cave was probably formed up to 50,000 years ago and had certainly not been dived in the 20,000 or so years since sea levels rose after the last ice age. In practical terms this means that you are generally swimming into good visibility, but if you pause, or even worse backtrack, you will rapidly find yourself struggling to see even your primary light when shining it straight into your mask. Initial exploration is therefore a rather exhilarating but potentially frustrating process of choosing which direction to follow at each branch, in the sure knowledge that if you call it wrong and have to backtrack, you will no longer be able to see the ?leads? in other directions, and the exploration will be over for the day. Instead you will be following your cave line out by touch, to face the hungry mosquitoes and that long walk back through the jungle. After performing our standard bubble checks and safety drills, we recapped the dive plan, including minimum turn pressures and other conditions for calling the dive, and agreed who would be doing what for the survey on the way out. Swimming into the air dome, we descended a few feet over the start of the line I had left from the first dive. After a pause to let our bodies adjust to the environment while we checked our equipment one last time, we made the 5 minute swim to the end of the line, I tied on the 1400 foot reel, and we exchanged "OK" light signals. Hearts pounding with adrenalin, breathing under control but only just, we followed our noses and slowly advanced into the unknown. The biggest fear on a dive like this is that the cave will suddenly dead-end and it will all have been for nothing - there really is no way of knowing what is around the next corner without going there. But almost immediately we saw a group of cave-adapted animals: four blind albino cave fish, and numerous transparent and sightless shrimps and other oddities, then a set of beautiful white stalactites and columns and a number of cave leads in different directions. We had ourselves a going cave, and we slipped into that highly focussed trance-like state of cave diving, mesmerised by the cave unfolding before us where we knew no-one had gone before. The first time I looked down at the reel, I was pleased to see that perhaps 500 feet of line had been laid - a good amount for any exploration dive - and we both relaxed. Time seemed to stop, although a number on my bottom timer kept increasing, and my pressure gauge was no doubt creeping downward at the corresponding rate. As we progressed into this strange world we saw large passages heading invitingly to the south - who knew whether they dead-ended or would lead to a whole new area of the cave? The main passage became so wide that we lost sight of the north wall behind the clusters of stalactites and columns. Gradually the colouration of the cave and its features changed, from a warm yellow-pink to a greyer, deader looking shade, the cave passages became narrower and the branches became smaller but more frequent, until finally the passage we were following narrowed down to a stalactite-draped restriction just a couple of feet wide. Nervous that we had taken a wrong turn and were going to burrow ourselves into a silty dead-end, and concerned not to damage the fragile formations, I looked back at Donna for the first time in what my bottom timer said was half an hour. The expression in those big blue eyes said it all, but she gave me an OK signal with her light as well. Turning back round to face the restriction, I looked for a suitable belay point to avoid the line moving into a narrow ?line trap? where we would not be able to follow it by touch if the visibility was lost. Then we gently eased ourselves through the restriction, the cave opened out again, and we were back on the road. After another 10 minutes I felt that tug on the reel that indicated that it was empty. I reached behind me and pulled out the second 800 foot reel, and tied that in to the first line while hovering above the silt. With a grin and a wink at Donna, we headed on. Now close to 2000 feet in, the cave was becoming siltier, shallower, and narrower from floor to ceiling. We were more frequently passing under pools of yellow water trapped in large pockets on the ceiling. These pools (?freshwater cap layers?) take their colour from the dissolved tannic compounds produced by decaying vegetation, and although they can be colder than the main cave water, they tend to stay stratified above a certain level, probably something to do with the relative rates of convection, heat diffusion, and dissolved chemical diffusion. In particularly dense tannic water, you can actually get a taste of tea through your regulator - something I miss very much after too long in the jungle. Finally we came to a long stretch of floor gently rising up into an opaque layer of dark yellow-orange tannic water, with enormous smooth white columns descending out of it down to the floor. Most probably we were approaching another cenote - I had a pretty good idea of where we were from our surveys: close to a dry cave called Mil Columnas. But the visibility in the water would have been too bad for Donna and I to proceed safely, and we had a long way to survey out. The exploration part of the dive was over. With mixed feelings of jubilation and sadness, I tied off the cave line to one of the columns and marked it with a plastic line arrow, pointing back towards the exit, while Donna hovered patiently over the line about 20 feet back, to stay clear of the reduced visibility resulting from bubble percolation. I took out my survey slate, and no doubt inspired by the taste of tea in the water and my long absence from England, wrote down the identifying number on the arrow and the name I had decided on for this magical place: ?Satannic Mills?. The numbers on the bottom timer started to mean something now. The second, frequently longer, part of any exploration dive is the survey out, and this is what distinguishes the cave explorer from the tourist cave diver. A quick check on gas pressures and the margin of gas remaining for our exit, and then we turned to leave. As agreed at the start of the dive, Donna swam along the line one tie-off ahead, to check that the line was running straight and was not caught on stalagmites or other protrusions, so I could get an accurate azimuth. She was also looking for and marking leads with her own supply of numbered line arrows. I followed on, noting the depth at each tie off, and the distance and azimuth between tie-offs. As is usual on such surveys, much of my time was spent in a cloud of low visibility, which is the reason that Donna had the responsibility of looking for and marking the leads - by the time I got there I could just about see the line and the markers she had placed on it. Because it is easy to lose track of time while surveying, I had noted a number of key points on the way in, and the time (and hence distance and gas consumption) to the exit, and I relaxed somewhat when we passed the first couple of these points more or less on schedule. Soon we were back at the end of the original 200 feet of line, taking slightly longer than the time we had taken to enter, but using slightly less gas, and I re-surveyed this back to the air dome to check on the accuracy of the survey from the first dive. By this time we had been in the water close to two hours, but we still had more than half of our breathing gas left, plenty of line on the second reel, and the possible connection to the Jaguar Cave to look for. Heading south from the air dome, we immediately spotted a large passage going where we wanted. The floor was covered in silt at least three feet thick (silt tends to accumulate in downstream passages near cenotes), and the walls were decorated with strange ?sculpted? features which we had seen in the dry caves in the area, but not in an underwater cave so far. Following the passage, we were disappointed to note that it soon turned west and then north, heading back up to join into the main upstream cave passage. Still with plenty of gas, we followed the southern wall a little further, and then found another passage heading off south. We followed this through a series of beautifully decorated rooms for a couple of hundred feet, but then the realities of safe gas management required us to turn around and survey out. The two of us had been in the water for about two and a half hours now, we were a little cold as we had used wetsuits rather than carry drysuits and additional weights all the way into the jungle, and we were a little tired. Perfect conditions for something to go wrong. The survey of the new passage back to the upstream main line went without incident, but now we had a slight mess of lines to deal with. We didn't want to leave the line in the silty downstream passage - it just looped around and wasn't very well laid because there were few good tie-off points due to the thick silt floor. So we decided instead to cut the end of the new line, tie it into the upstream main line, then take out the downstream loop. I cut the line, then indicated to Donna to start reeling in the downstream line, while I made the tie-in, placed the necessary markers, and made a survey shot back to a marked point on the main line. The two lines were parallel for about a hundred feet, so Donna could safely reel in while I was doing this, without us losing sight of each other. It was while I was making the tie-in that I saw that little flicker of light in my peripheral vision which indicated that there was a problem. Donna's primary light had failed, then her first backup light as well, and while she was in the dark deploying her second backup light, the line had come slack from the reel and tangled itself around her fins. Kicking her fins and struggling, or even worse panicking, would have increased the entanglement and destroyed all visibility. But Donna knew better than that, and kept herself under control, hovering motionless until she could get my attention and I could come to her assistance. But only after I had completed that final survey shot, of course...  

	Since 1961, more than 400 open water divers have died in cave diving accidents in Florida, Mexico and the Caribbean. All of them thought they were good divers, and many of them were open water instructors. Interviews with surviving dive buddies often indicated that they were “just going to take a peek inside the entrance”.
Underwater caves have specific and unique hazards. These require training, equipment, and techniques which are very different from open water diving. NO AMOUNT OF PREVIOUS OPEN WATER DIVING EXPERIENCE OR TRAINING CAN ADEQUATELY PREPARE YOU FOR CAVE DIVING. Be smart and obtain proper training before entering any underwater cave or other overhead environment.