On Sept. 23, Nevada Department of Wildlife habitat biologist Craig Stevenson, right, and game warden Mike Maynard examine the decomposed carcass of a bighorn sheep, one of 22 that died in July in the North McCullough Range. Photo by Gary Thompson. Water tanks made of heavy-duty plastic stand Sept. 23 in the North McCullough Range, about 200 yards east of Sloan Canyon National Conservation Area. Photo by Gary Thompson. Click image for enlargement. Graphic by Mike Johnson. Click image for enlargement. Graphic by Mike Johnson. Bighorn sheep images are depicted in a Sloan Canyon petroglyph etched in a rock covered with desert varnish. REVIEW-JOURNAL FILE PHOTO

When game warden Mike Maynard arrived at the death scene on the afternoon of July 24, the soil was still damp from intense rains that had pummeled the North McCullough Range in the early morning hours.

Carcasses of 22 desert bighorn sheep rotted in the 105-degree heat. Flies buzzed over them and the strong odor of decomposing flesh hung in the air.

It was the biggest die-off of desert bighorn sheep in Nevada on record.

In the incident report he filed for the Department of Wildlife's Law Enforcement Bureau, Maynard noted there were three to four adult sheep lying on their sides within 10 yards of Guzzler No. 3 -- a system of heavy-duty, rubberlike tubes and three 2,300-gallon polyethylene tanks that hold rainwater piped downhill from a small dam at the base of a rock-wall crevice.

Each tank is equipped with a shut-off valve that, when open, allows water to flow to a trough, or "sipper." A float lever made of a rod attached to a hollow, copper sphere the size of a grapefruit controls the water level in the sipper, much like the float mechanism inside bathroom toilet tanks.

"Walked to the guzzler and observed several sheep dead in the immediate area," Maynard stated in the incident report's narrative. "Guzzler sipper had lid removed, float ball was floating in murky water in the sipper. Valve was in open position and flowing freely. Water was overflowing the sipper."

These were the conditions when he responded to this remote location where a pair of all-terrain vehicle riders had found the cluster of dead bighorns strewn among teddy bear cholla and scattered in the shade of a tall rock cliff, 20 miles south of Boulder City and 200 yards east of Sloan Canyon National Conservation Area.

During a visit to the guzzler last month, Maynard and wildlife habitat biologist Craig Stevenson estimated the sheep had been dead about four days before the carcasses were discovered by the ATV riders. That means they would have died at the end of a record-setting heat wave in July, when temperatures there probably reached 115 to 120 degrees, they said.

A few days after the sheep were found, game wardens eliminated poaching as a cause. There were no signs of trauma from gunshot wounds and metal detector scans of the area turned up no evidence of bullets.

Then on Sept. 9, after an extensive battery of tests on the freshest specimen and an assessment from the University of California, Davis, state wildlife officials released a statement saying the cause of death was inconclusive.

The statement relied in part on the findings of state veterinarian Dan Crowell, who in an Aug. 31 memo to Nevada wildlife biologists, ruled out 13 factors including electrocution from lightning strikes, heavy metals, poisoning from anti-freeze, and anthrax.

Although biologists have observed bees swarming over the sipper numerous times including a recent visit there this month, none of the carcasses showed signs of prominent bumps from bee stings.

Two Gambel's quail and a pair of mourning doves were found nearby, but a link to West Nile virus -- a sometimes fatal virus carried by mosquitoes that affects birds and mammals including humans -- could not be established. Tests for West Nile virus were not conducted, but state wildlife biologists have said there are no previous accounts of the virus affecting bighorn sheep. If the virus was involved they would not have expected all 22 bighorns to die in such a short period so close to each other.

Exactly what caused the sheep to die remains a mystery that might not ever be solved, according to wildlife officials.

However, they did issue a statement noting that there were three factors that Crowell "couldn't rule out." Those were dehydration, botulism in the water supply, and toxins from blue-green algae.

There was no mention of any factors that might have contributed to the deaths.

Shortly after those results were made public, a source close to the hunting community called the Review-Journal and said it might be productive to check into a trip made up to Guzzler No. 3 the previous November by retired hunting guide Ed Pribyl of Las Vegas. He and another volunteer had shut off the flow of water from the tanks, the source said.

Four days after wildlife officials announced that test results were inconclusive, Stevenson, the habitat biologist, confirmed that Pribyl and the other volunteer had indeed shut off the water flow from two of the three tanks.

A reference to the shut-off had been included in the rough draft of the Department of Wildlife's statement, but a biologist edited it out. Department spokesman Geoff Schneider declined to reveal who made the edit, but said it was an oversight not to reinsert the information into the statement that was released.

From June 11 until rains came on the night of July 23, the sipper and the southern tank apparently went dry from as many as 85 sheep using Guzzler No. 3.

One ram was apparently so thirsty that it used its curved horns to lift the heavy steel-plate lid from the float box to get at the last sips left in the chamber. The dented copper float ball apparently had been kicked by the ram.

Many sheep might have lingered around the guzzler because they could smell the 4,600 gallons of water that remained in the two full, shut-off tanks.

But such water shut-offs are routine, according to wildlife officials. They're done in the winter, Stevenson said, as a precautionary measure in case freezing conditions crack or rupture the pipes allowing water to spill after the plumbing thaws. Volunteers then return in the spring to turn the valves open so there will be plenty of water during the summer months.

"The theory is that you have one tank that's on through the wintertime to provide water for all the other wildlife and the other two are turned off just in case," Stevenson said.

But in the McCullough Range, the valves weren't turned back on.

"What happened here was the two guys that turned it off, one of them had hip-replacement surgery and the other one pulled an arm out of his socket, so it was just kind of an odd series of events," Stevenson said of why Pribyl and the other volunteer, Gary L. McDaniel, did not get back to the area.

But other hunters visited the guzzler during the summer months.

Stevenson said, however, that they only knocked on the tanks to see if they were full. They didn't realize they needed to be turned on.

Some off-duty game wardens also checked the tanks on June 11, before the heat wave, according to Maynard. Like the volunteers, they didn't realize the valves to the northern and middle tank were in the off position.

"One of them tapped on the two tanks and indicated they were brim full and the southern tank was half-full. They left everything as they found it," Maynard said.

Still, whether the failure to turn the tanks back on directly led to the die-off is an open question, according to officials.

As for dehydration, Crowell noted that assessment of brain samples for sodium levels, which are elevated in cases of dehydration were ''nondiagnostic."

"However, due to the lack of knowledge of the normal values for brain sodium levels in normal desert bighorn sheep, coupled with the history of the condition of the drinker and weather conditions, dehydration cannot be ruled out."

Crowell's memo also states that although tests for evidence of botulism and blue-green algae were negative in the water supply, "it is possible that the toxins could have been present and were subsequently flushed from the system with the rain incident."

After the rain showers on July 23-24, game warden Maynard noted in his incident report that the northern and middle tanks were ''brim full of clear, odor free water. ... The southernmost tank had water trickling in from the collection pipe. When I opened the tank hatch, a cloud of thick, warm, water vapor with a mild stale odor hit me in the face. The tank had a stale smell and approximately 4-6 inches of slightly murky water in it."

In a telephone interview on Sept. 19, Crowell said the three potential causes that couldn't be ruled out -- dehydration, botulism and blue-green algae -- or some combination of the three can all be caused by a dry sipper or low, stagnant water.

"I think those would be the most likely causes," Crowell said.

"You can pull out the stops and potentially blame it on anything, but at the end of the day those are the most likely" causes of the die-off.

There are three other guzzlers about five to seven miles away from Guzzler No. 3 where the sheep could have gotten water. It's possible, officials acknowledged, that the 22 that died might first have gotten sick from foul water and, too weak to walk to another guzzler or even a natural supply about a mile away, succumbed during the record heat wave.

If botulism were involved, their muscles could have become paralyzed. The bacterium, Clostridium botulinum, produces a toxin that affects muscular nerves.

That's what biologists believe happened in August 1995 to a herd of bighorn sheep near Old Dad Peak, about 19 miles south of Baker in San Bernardino County, Calif. According to a report in "Journal of Wildlife Diseases," 45 bighorn sheep had died near two guzzlers, including 13 that fell into an open guzzler tank and drowned.

"Two of the three storage tanks had insufficient water to feed the drinker basin. The third tank was full but the valve connecting it to the system was closed, leaving the drinker basin dry," the Journal's January 2000 issue reported.

"The hatch cover on top of this tank was dislodged, allowing access to the interior of the tank. The top portion of this tank also was indented and supported a pool of algae-covered water containing hundreds of dead fly larvae," the report stated.

Blood from a sheep carcass tested positive for type "C" botulism toxin as did a fly larvae sample. Biologists concluded that type "C" botulism most likely caused the California bighorn sheep die-off.

Long before Guzzler No. 3 was built in the McCullough Range in 1986 and upgraded in 1991, bighorn sheep had survived in the scorching Mojave Desert heat by drinking from "tenajas," natural water catchments formed by rock surfaces in what is now Sloan Canyon National Conservation Area.

There, in the canyon's core area, generations of Mojave, Chemehuevi and Paiute Indians left galleries of petroglyphs, or rock art. Some of the panels contain images of bighorn sheep that were etched in stone going back a thousand years or more.

These etchings in refrigerator-size boulders depict the importance of desert bighorn sheep to the ancient people. Hunting blinds sit above the wash that lines the canyon floor. Even today, bighorns wander up the dry wash to paw at the gravel to suck water from the mud beneath it.

Stevenson said he relies on volunteers from the Fraternity of the Desert Bighorn, Nevada Bighorns Unlimited, Reno and five other wildlife organizations to monitor 61 guzzlers in four Southern Nevada counties. The Bureau of Land Management maintains another 25, the U.S. Fish and Wildlife Service has 30 and the U.S. Forest Service has two guzzlers.

He said guzzlers are typically designed to provide enough water for 50 bighorn sheep plus extra for emergencies and dry years. There are about 300 to 350 adult bighorn sheep in the North and South McCulloughs and the Highland Range. Water had to be hauled to Guzzler No. 3 during droughts in 1996 and 2002.

"We've done a pretty good job of monitoring these projects. It's the mistakes everyone wants to know about," Stevenson said.

Rob Buonamici, Nevada's chief game warden, said it was "just an unfortunate oversight" that the two tanks weren't turned back on in the spring.

He said the Game Bureau is re-evaluating what happened and is trying to put "checks and balances" on how guzzlers are monitored and controlled.

"If we did something wrong, we need to fix it," Buonamici said Sept. 20. "If we didn't, we need to know how to prevent it."

Stevenson said his records show that Pribyl, who he said turned off the tanks, had inspected the guzzler 28 times since 1986, including the last time in November with McDaniel.

On Tuesday in a telephone interview, Pribyl, 67, said he couldn't remember visiting Guzzler No. 3 or shutting off the tanks in November.

"That would have been normal procedure," said McDaniel, who also didn't have a specific recollection of the event.

"I could have been (there). I'm not going to deny it. I check so many projects," Pribyl said, estimating the number to be 105.

Two days after the carcasses were discovered, Pribyl and Bill Vasconi, president of the Fraternity of the Desert Bighorn, helped wildlife officials remove some of the dead sheep, according to Maynard's report.

"If you're looking for a smoking gun, you're not going to find one," said Pribyl, who has hunted big game in Nevada since 1950.

"The die-off certainly didn't occur because the project ran out of water," he said. "They just move to the next one. ... I don't think there will ever be a definite reason for why it happened. ... It could be a freak of nature."

Stevenson said he intends to talk to volunteers and new wildlife employees to make sure they know how to check guzzlers and turn valves on and off. Putting locks on the valves would be impractical he said because about 200 keys would have to be distributed and tracking them would be a nightmare.

"Ultimately, it's not the volunteer's fault who turned off the water. If anybody's to blame, it's me," Stevenson said.