The report was conducted by SARtest Ltd., a consultancy that specializes in the measurement of radio frequency fields. A report, "SAR [specific absorption rate] Tests on Mobile Phones Used With and Without Personal Hands-Free Kits," can be downloaded from the Web site of the U.K. government's department of trade and industry, which commissioned the report.

The department commissioned the latest report after an earlier report by the Consumers Association, published in April, said that an earpiece, which enables hands-free phone usage, could increase a mobile phone user's exposure to electromagnetic radiation.

The Consumers Association had conducted laboratory tests that it claimed showed that an earpiece cable could act as an antenna and channel three times the level of radiation to a user's head as compared with normal handheld use.

Mike Manning, managing director of SARtest (Newdigate, England), said that there were experimental factors that distinguished the SARtest report from that of the Consumer Association. Manning said his methodology took an average of many readings from a probe moved through a liquid-filled phantom head while the Consumer Association report took just one reading near the ear. But Manning said he did not believe that alone could explain how the Consumer Association could show increased SAR with an earpiece while his results showed the SAR reduced in most cases to 20 percent or less of that experienced when using a mobile phone held next to the head. No evidence has yet been found that electromagnetic radiation, at the levels normally emitted by mobile phones, is harmful to human beings.

SARtest investigated worst cases, including a situation where a mobile phone antenna cable is taped to the cheek of a phantom head and the cable is wrapped around an external antenna, or draped over the case of the phone in cases of an internal antenna, to maximize coupling.

In a couple of instances SARtest was able to increase the SAR from mobile phone earpieces compared to normal hand-held units, but even in these cases absolute levels remained well within exposure guidelines of the National Radiation Protection Board (NRPB) and the International Commission on Non-Ionizing Radiation Protection (ICNIRP).

Tests were conducted on a range of Ericsson, Nokia and Maxon mobile phones operating at highest power ratings at 900 MHz and 1,800 MHz. Uncertainties of 40 percent were ascribed to absolute figures but for comparative figures they are reduced to 5 percent.

However, Manning said the SAR measurement involved a simplification of what in real life is a complex situation, and that his company's reports should be taken as guidelines. The tests' use of homogeneous phantom heads, rather than non-homogeneous heads that attempt to mimic bone structure and musculature, was a simplification, albeit a generally accepted one.