Many mothers, particularly first-time ones, say they only really start to believe they are pregnant when they see the mysterious moving form of their baby on the ultrasound screen. Others crave the reassurance the image can bring. "I had a scan every two weeks," says one woman, pregnant again after three miscarriages. "I just needed to see that he was there, and growing."
Some 95 per cent of women are scanned during pregnancy, according to the Audit Commission. Most have at least one scan, but some have as many as 10.
Ultrasound uses high-frequency sound waves to make a picture of the baby, so that medical practitioners can check for abnormalities, or monitor the growth rate of the foetus and the position of the placenta. New techniques are still being developed: one, being pioneered at King's College Hospital, London, uses scans to assess the risk of Down's syndrome in the developing foetus from as early as 11 weeks. This could eventually lead to the introduction of a routine early scan for many women, in addition to the one usually given at around 20 weeks to check for other "anomalies".
But questions over the safety of ultrasound, and its possible effect on the delicate tissue of the developing embryo, refuse to go away. The campaigning group Aims (Association for Improvements in the Maternity Services), which has long been critical of the routine use of scanning, now suggests that the rising incidence of language delay in children might be the result of ultrasound exposure in the womb.
A long-term study of more than 1,000 children, reported widely earlier this year, found that a startling one in five had speech problems. Dr Sally Ward, the speech therapist who carried out the study, said that too many parents were sitting their toddlers in front of the television for lengthy periods rather than talking to them.
But Aims says that there is evidence to suggest a connection between this type of problem and the routine use of ultrasound scanning. "Studies indicate that there could be a problem with ultrasound," says Jean Robinson, Aims researcher, author, and vice-chair of the Patients' Association. "If there is, it will affect the way the brain works and the effects will be subtle and may not show up unless you seek them out. If cases of language delay are going through the roof, there could be a reason other than parental neglect."
Ms Robinson cites a Canadian study, published in 1993, which found that children with delayed speech were twice as likely as their non-delayed peers to have been exposed to ante-natal ultrasound waves. Two hundred and fifty Albertan children aged between two and eight years were studied; the authors, from the University of Calgary, found "an association" between ultrasound exposure and delayed speech, although they did not claim to have proved cause and effect.
In addition, a group of researchers in Norway, engaged in a long-term study of scanned and unscanned children, found that children were less likely to be right-handed if they had been scanned. They, too, were tentative about their results, suggesting merely that their finding that scanning might affect handedness "may have some merit and should be tested".
Aims points out that if scanning is linked to handedness, this means it affects the developing brain of the foetus - a suggestion that infuriates some members of the British medical profession. "I believe it is totally and utterly safe," says Stuart Campbell, professor of obstetrics and gynaecology at St George's Hospital, London, who edits a medical journal on ultrasound. "It has been widely used since 1970 and has never been proven to harm the baby. If it had, my four children would not have had repeated scans in utero."
Professor Campbell - who estimates he has scanned some 30,000 babies - dismisses the Norwegian findings on handedness. "If you look at one hundred variables," he says, "one is bound to be significant by chance."
The heated debate on ultrasound - which extends not only to whether the technique is safe but also to whether it is useful - can be seen partly as a culture clash. A schism has developed between the mainly male medical establishment - with its respect for high technology - and a mainly female pressure group upholding the value of traditional midwifery skills and minimum intervention. Aims stands accused by one doctor of "hatred of male gynaecologists".
But the participants in the debate do not line up altogether neatly. Dr Hylton Meire is consultant radiologist at King's and a respected member of the medical establishment. He too expresses disquiet over some aspects of the use of ultrasound. "The evidence shows that you can damage tissue if you turn the power up high enough," he says. "A 'dose relationship' exists."
Dr Meire questions the use, not of regular scanning machines on pregnant bellies but of the more powerful Doppler ultrasound, which can track the movement of red blood cells, and which in pregnant women is sometimes used to check the functioning of the placenta and flow of blood through the cord. "It might be all right in adults," says Dr Meire, "but the developing embryo is, I think, more sensitive to ultrasound. Levels which are being used on human embryos are the same as the ones that have been shown to damage rats and mice."
Britain follows guidelines issued by the European Federation of Societies for Ultrasound in Medicine and Biology which advises against the use of pulsed Doppler devices in the first three months of pregnancy, the crucial period when the major organs are being formed, although, according to Dr Meire, "people are using Doppler in early pregnancy". The federation also said Doppler use should be minimised in the second and third trimesters, "particularly when foetal bone structures lying within the Doppler beam may be preferentially heated".
Dr Meire and others also question the so-called "transvaginal" scanning method. In this technique, a probe is inserted into the woman's vagina. Because the beam doesn't have to pass through the flesh and muscle of the belly, a higher frequency can be used and a sharper picture obtained. Transvaginal scanning is used if a woman is very big, and the ordinary scan fails to give a clear picture of the foetus. "The probe is very close to the target organ - the uterus," explains Dr Henry Irving, president of the British Medical Ultrasound Society. Dr Irving and his society do believe in the safety of ultrasound - but recommends the alara (as low as reasonably achievable) principle which, he admits, has not always been adhered to. "New techniques always bring a burst of enthusiasm," he says. "But it's important to bear in mind that you're putting into the body energy which it is not normally exposed to. We're constantly conscious of the potential to do harm if the technique is not used sensibly."
Laurel Fennell, chair of the Royal College of Nursing's midwives in ultrasound group, warns that the more powerful transvaginal and Doppler scans may become the norm. "Thesemachines enable you to see more, earlier," she says. "If people want perfection, and early diagnosis of abnormalities, that's the way it's going to go."
Now the mother of an eight-month-old baby, Laurel Fennell elected not to have scans during the first 12 weeks of her own pregnancy. Early scans, although not routine, are increasingly being offered, either as a method of detecting the risk of Down's syndrome, or as a response to bleeding or pain. "While the foetus is developing, I'm not sure that we ought to be interfering," says Ms Fennell.
In the absence of conclusive research, the debate is likely to continue. But with ultrasound an almost universal practice, research becomes more difficult to carry out: Professor Stuart Campbell is now scanning the pregnant daughters of women he scanned 25 years ago. In England at least, there is no "unscanned" group of children against which to test hypotheses.
'It was a toy. They loved switching it on'
However slight the evidence against ultrasound, some parents are making a connection. Pat McCarthy, 46, has an eight-year-old daughter called Sarah. Apparently perfectly healthy at birth, Sarah was slow to begin to talk: at two, she was barely saying a word. At four, says her mother, "she was walking around as if she'd just arrived off Mars - not quite getting it".
Sarah now attends a special school for moderate learning difficulties. The speech and language areas of Sarah's brain, tests have established, did not develop normally in the uterus. A brain scan suggested that something interfered with the formation of the temporal lobe at between 10-20 weeks gestation.
Mrs McCarthy had five scans while pregnant, although she cannot recall what they were all for. "It was a toy - they loved switching it on," she says.
"Now we know that something happened to Sarah's brain between 10 and 20 weeks. I wonder if the five scans might have caused some damage and I'm not sure they were all necessary."
Join our new commenting forum
Join thought-provoking conversations, follow other Independent readers and see their replies