A deadly new virus could be on the rampage in south India. Since June, a mysterious "brain fever" claimed the lives of 163 children in Andhra Pradesh and spread to nearby Maharashtra. The killer was swift and lethal-most children died within 24 hours. Nobody could explain why. Now scientists at the National Institute of Virology (NIV), Pune, believe they have zeroed in on the culprit-a rare, mosquito-borne virus not generally known to kill. The treatment for the fever, however, continues unchanged.

MISERY: Parents mourning the death of their child at a hospital in Hyderabad

There were suspects galore for the epidemic-the common one being Japanese encephalitis (JE), a familiar mosquito-borne viral disease that claimed more than 600 lives in 1986. A laboratory in Hyderabad found only three JE cases. But a team of experts from Delhi that visited the affected areas concluded the disease was most likely JE. Their premise was based on clinical evidence. The confusion only deepened the mystery. Samples were then sent to NIV. The true identity of the killer added a further twist to the tale. The culprit, revealed NIV in a confidential report to the Union Health Ministry, is the Chandipura virus. It belongs to the same family as the rabies virus and is rarely found in human beings.

The revelation raises more questions than it answers. For the Chandipura virus is nowhere near the top of the list of killer viruses. "It is normally not pathogenic," says Dhrubajyoti Chattopadhyay, an expert on the Chandipura virus. First isolated from two dengue victims in Chandipura, Madhya Pradesh, the virus was identified in 1967. Since then it made an appearance in Delhi in 1980 and in Africa in 1988. But it had not been reported to be fatal in India.

How the virus spread rapidly can be easily explained. The virus is carried by the Aedes mosquito and is found in a range of hosts, from cattle to birds. Symptoms of Chandipura infection include high temperature, vomiting and diarrhoea, all of which were reported in most of the affected children. But the Chandipura virus generally does not affect the brain. Yet the children died as their brain was affected. Tests done on mice at NIV confirmed that this variant does indeed affect the brain-the infected mice died within 24 hours.

Could a virus change so much? Is this a new mutant? Like the common-cold variety of viruses that mutated to cause SARS, the genetic material of the Chandipura virus consists of RNA, not DNA. The single-strand RNA is more susceptible to changes than the double-strand DNA. "It is possible that as with the SARS virus, the Chandipura virus could mutate to become a more lethal variety," says Chattopadhyay. It is a problem NIV is working hard to answer. Its Director A.C. Mishra, who was reluctant to talk about the report to the Health Ministry, says, "Our knowledge of the Chandipura virus is based on a few episodes. Nobody knows much about it and its disease patterns." Union Health Ministry officials are also cautious. "The Chandipura virus has been reported, but you cannot rule out JE. Some symptoms tally with those of JE," says B.M. Das, director (emergency services), Ministry of Health and Family Welfare.

As officials debate, the clock ticks for the victims. Two months after the epidemic began, there is still no consensus on the cause. Hyderabad, cybercity and biotech hub, does not even have the Biosafety Level-3 laboratory facilities required to detect a lethal virus. "We badly need to improve facilities across India," says a public health official. It is the only way to quell problems before they turn into disasters. It is a common-sense solution-but it hits the authorities only after the disaster strikes.