People are getting hot-headed about mobile phones. No, not just when a shrill "Yellow Rose of Texas" breaks the silence of a concert hall; mobile phone users are complaining of hot ears and headaches.

However, when it comes to choosing a phone, its colour, shape, features and battery life are the main criteria, while minimising the exposure to radiation has barely become an issue.

Should it be? Well, given the public concerns over the location of GSM base stations, the fact that using a mobile phone is said to expose your head to up to 10,000 times stronger radiation, the silence on this issue is surprising. And according to one leading scientist, the design of the phone causes a large variation in the amount of radiation absorbed by your head.

There are two types of handset: analogue (088 numbers) and digital, also known as GSM. In Ireland, both types of phone transmit at nominally the same power level, 600 milliwatts, but GSM phones don't transmit continuously, and their average output power during a call is reduced to 250 milliwatts. On top of this, if you're near a base station the power of the signal being sent from your GSM handset may be lower again.

But is the radiation harmful? Well, there are two areas to consider here. The first is that mobile phone signals can heat matter, like extremely weak microwaves, while the second area concerns more serious biological effects, about which less is known.

One of the world's leading scientists in mobile phone radiation effects is Dr Mike Repacholi, an Australian who works at the World Health Organisation in Geneva. Famous for his 1995 statement that the worst aspect of mobile phones was that they often interrupted his meals in restaurants, Dr Repacholi insists any heating effect is negligible.

He cites a recent study by Motorola, Ericsson and Nokia, comparing heating effects of mobiles pressed against the head but switched off, and similar mobiles switched on. The results, to be presented in Florida next month, show that the temperature "rises a little bit, but not significantly", he says. At GSM frequencies (900MHz), he says, radio waves can penetrate some two to three centimetres into tissue, while at the newer PCS frequencies (1800MHz) this is reduced to one to two centimetres. Another leading scientist takes a different tack. Dr Niels Kuster, of the Swiss Federal Institute of Technology, in Zurich, is recognised as the leading researcher into the amount of mobile phone radiation actually absorbed by the head. Using dummies, he measures the specific absorption rates (SAR) of various phones, which is the rate at which energy absorbed by tissue in the area of the head near the antenna. The goal of any phone, he says, is to transmit as much energy as possible towards the base station, with the minimum amount absorbed by the head.

Dr Kuster's studies reveal wide differences between the SAR values of various mobile phones. In the only published set of results he found SAR values ranging from 0.28 watts per kilogram to 1.33 watts per kilogram. Thus the phone with the highest SAR made by Ascom which has won style awards for its design causes more than five times as much radiation to be absorbed by the head than the lowest valued phone made by German company, Hagenuk.

Dr Kuster confirms that extending a phone's antenna, when it has one, significantly reduces the radiation being absorbed by the head.

The Hagenuk phone features a special antenna which directs the radio signals away from the head. Dr Kuster says most mobile antennas cost about a dollar each to the manufacturer, while the Hagenuk antenna would only cost a few dollars, and says other tests indicate its efficiency is superior to many other GSM antennas.

Hagenuk last year used the slogan "Low radiation is better" in its ads, but following pressure from network operators it toned down the campaign. The German company has since come close to liquidation, and was partially taken over by Ratinger Tiptel last March.

Dr Kuster also performs SAR and antenna efficiency tests on behalf of the mobile phone manufacturers, but the results are confidential. However, there is evidence manufacturers are taking steps to reduce the radio exposure to the head.

Motorola's Star Tac, which features an antenna directed away from the head, recorded the second lowest SAR value in the publicised tests.

French company, Alcatel, which recently started marketing mobile phones in Ireland, says it too is angling the antenna away from the head. But none of this tells us if mobile phones are dangerous to use.

A study last year by Dr Repacholi did find the incidence of tumour growth was nearly doubled among mice exposed to two 30-minute doses of mobiletype radiation daily for 18 months, but other scientists have criticised this study, saying the dosage levels weren't sufficiently controlled and may have been much stronger than dosages associated with phones. Despite the findings, Dr Repacholi talks down any danger, saying: "I don't expect any biological effects."

Dr Kuster says tumours would not be related to any heating effects, and that studies into whether cancer development is promoted by mobile exposure are now being done. The International Agency for Research on Cancer, in Lyons, is currently planning a major epidemiological study, comparing people using phones with those who do not over three to five years.

A Swedish/Norwegian study, released last week, shows people complaining of heating and headaches associated with using both analogue and GSM phones.

This all makes uncomfortable reading for manufacturers and operators. While many more studies will be necessary to establish any risk to human health, a couple of steps can be immediately taken. The first is for manufacturers and operators to publicise SAR figures, so people can choose phones which minimise the absorbed radiation. ESAT, for example, admits that it currently does not even look at SAR figures.

Finally, given that radiation intensity drops dramatically with distance, concerned users can use hands-free accessories to maximise the distance between them and the mobile as often as is practical. Simply moving a mobile from your ear to a desk in front of you reduces the exposure by up to 1,000 times.