In late May, George Burgess, director of the Florida Museum of Natural History's International Shark Attack File and a noted shark researcher, was quoted as saying, "Falling coconuts kill 150 people worldwide each year, 15 times the number of fatalities attributable to sharks."When I called Burgess, he told me he had gotten this statistic off the Internet--specifically, from a widely reported press release from the British travel-insurance firm Club Direct, saying that "holidaymakers hit by falling coconuts will be guaranteed full cover under their travel insurance policy. The news follows reports from Queensland, Australia, that coconut trees are being uprooted by local councils fearful of being sued for damages by people injured by coconuts. . . . 'Coconuts kill around 150 people worldwide each year, which makes them about ten times more dangerous than sharks,' says Brent Escott, managing director of Club Direct."So, Brent, do coconuts kill ten times as many people as sharks, or fifteen? No response yet from the UK. However, Club Direct's release also cites an article by Dr. Peter Barss in the Journal of Trauma entitled "Injuries Due to Falling Coconuts." (The article received an Ig Nobel Prize, given annually at Harvard by the editors of the Annals of Improbable Research in recognition of research that "cannot or should not be replicated." The award was presented in 2001, notwithstanding that the paper had been published in 1984. Apparently news takes a while to filter through to Cambridge.) The article soberly reported on nine injuries in Papua New Guinea due to falling coconuts, none fatal. Barss notes that a coconut palm tree commonly reaches 25 meters in height, that a coconut can weigh two kilograms or more, and that a two-kilogram coconut falling 25 meters would have a velocity of 80 kilometers per hour on impact and a force of as much as 1,000 kilograms. Several victims suffered fractured skulls, were rendered comatose, etc.OK, getting hit by a coconut is no laughing matter. But nowhere does Barss say that 150 people get killed by coconuts each year. He provides an anecdotal account of one such death and in a separate paper estimates that over a four-year period five deaths in his hospital's service area were related to coconut palm trees (including climbers falling out of them). A recent report (Mulford et al, "Coconut Palm-Related Injuries in the Pacific Islands," ANZ Journal of Surgery, January 2001), which describes itself as "the largest review of coconut-palm related injuries," also reports no deaths and on the question of mortality merely cites Barss. Given that Barss' hospital in Papua New Guinea served a population of 130,000, one conceivably could project 150 deaths over that portion of the world population living in proximity to coconut palm trees, but I'm not aware of any systematic attempt to do so. Noting that death reports in tropical countries are limited, Barss tells me, "I am surprised that someone has come up with an actual number for such injuries. It must be a crude estimate, and you would have to ask them what methodology they used to verify whether it has any validity." Conclusion: Somebody pulled the figure about 150 deaths due to coconuts out of thin air. Take that, shark lovers.Barss, incidentally, wrote numerous frightening reports while stationed in the tropics. His subjects included injuries by pigs in Papua New Guinea, penetrating wounds caused by needlefish in Oceania, scombroid fish poisoning at Ala Tau, grass-skirt burns, wound necrosis caused by the venom of stingrays, and inhalation hazards of tropical "pea shooters." He's now teaching at United Arab Emirates University, in a desert city built around an ancient date oasis. Can't blame him for making the switch--who ever heard of getting KO'd by a falling date?