But the disagreement broke up the organization. Gross stopped attending meetings. She had spent much of three years away from home for legislative hearings and media appearances. Through it all, she worked full-time as an office manager. She realized the toll it was taking one morning when her daughter missed the school bus. “After everything we’d been through and coming up empty-handed, I just had to let go and focus on my family.”
Phelan, who left Fallon in 2004, said something similar: “I think I lost focus on what was most important, which was my child and getting her better, because it became all about finding out what was going on.”
* * *
Todd had not intended the search for a cause to end with the official investigation. He described to me the thin line between “making sure there’s nothing obvious lurking in the community”—government investigators’ role—and doing research. “We said, ‘We’re going to go up to this point, and others can go beyond that.’ We had the sense that even though this was a highly significant cluster, it wasn’t large enough to find the answers, and we hoped that researchers would be able to cast a larger net.” In other words, scientists would have to combine Fallon’s samples with those from leukemia patients in the larger population and look for things that set Fallon’s apart.
For years after the investigation closed, several scientists did carry on. In 2004, Senator Reid secured $250,000 for Jill James, a professor of pediatrics at the University of Arkansas, to investigate a difference in the way Fallon children with and without leukemia metabolized toxic substances. James found that sick children had less glutathione, an antioxidant, in their blood. This could mean that children who developed leukemia were less able to rid their bodies of arsenic, tungsten, and jet fuel. Over time, these lingering contaminants may have damaged their DNA, making them susceptible to cancer. James never published the findings. She had only 20 blood samples—not enough to reach a conclusive result.
In 2006, Senator Reid secured another $700,000. Remaining FIST members, with help from the University of Nevada-Reno, solicited proposals. Three research teams won grants, including Witten to study the role of tungsten in leukemia development. After detecting naphthalene in the trees by the school, Witten and Sheppard had measured tungsten using a similar technique. Among their most compelling discoveries is that tungsten levels are significantly higher in tree rings since the mid-to-late 1990s, right before the cluster began. They have published more than a dozen scientific papers on the topic, many in reputable journals. But their work has drawn criticism, first from Todd, who recounted a phone call in which he accused Witten of bad methodology. Witten did not recall the conversation, and when I asked about Todd’s other criticism—that Witten shared his results with the press before publishing them—he said, “That’s just a cheap shot.”
Other critics are Seiler, the USGS hydrologist who maintains that the tungsten is natural, and scientists employed by Kennametal, who accuse Witten and Sheppard of misrepresenting data. Both have been deposed for ongoing litigation against Kennametal, and lawyers have subpoenaed their data and emails. Witten does not hide his opinions. “The story in all of this is how big companies go out to these little towns, and they’re the only employer, and they get away with murder, essentially, in the way they pollute,” he told me. Sheppard is more cautious: “I cannot say that tungsten causes leukemia. It may appear that one side wants me to survive and one side doesn’t, but that’s not my problem.”
There is only one team of scientists still actively researching the cluster, led by Joseph Wiemels, a professor of epidemiology and biostatistics at the University of California-San Francisco. In October, I met Wiemels and his collaborator, Stephen Francis, a doctoral candidate at the University of California-Berkeley Department of Epidemiology, at the Childhood Leukemia Research Center. Wiemels wore wire-rimmed glasses and the studied look of a laboratory scientist. Francis had sun-bleached hair to his shoulders. Wiemels received one of Reid’s grants in 2006 to investigate the possibility that a virus triggered the cluster. According to a respected theory, leukemia clusters may arise when waves of transient people spread infection to an otherwise isolated population. Fallon’s population shifts annually due to the Navy base; at the time of the cluster, there was a spike in childhood leukemia among families in the armed forces at large.
For his research, Wiemels needed Fallon children’s blood samples taken before treatment began. (Treatment obscures signs of prior infection.) He also needed data indicating children’s histories of infection. In 2007, he met with Todd, who informed him that the data were gone; his hard drive had been erased when he left the state office. Wiemels contacted the CDC, which had collected similar data, but was told he could not see it without approval from the families. “The problem was that by the time we started working on it, half the families weren’t talking to anybody, not even other families,” he said. As for diagnostic blood samples, the organization that routinely collects them was in transition during the cluster, and only four from Fallon leukemia patients were saved.