Bindeez suspected

"At this point, the suspicion initially was that perhaps there was some GHB in the family somewhere, someone using it as a recreational drug and he'd got hold of it somehow," Carpenter says.

The child's mother denied the family had anything of the sort in the house, but she did mention that the child had eaten a substantial amount of Bindeez beads, a toy that involved creating pictures from little coloured beads that was named last year's Toy of the Year. Beads had been found in the child's vomit and in a stool sample.

"Very astutely, the doctor, Olivia Chan, said perhaps it was something on the beads," Carpenter says. "She got hold of some of the beads that he'd been playing with and sent them to me and asked if I'd have a look for GHB.

"We had a look to see if there was anything there and there wasn't, but I did see a big peak on the chromatograph, which was something I hadn't seen before."

Using an Agilent gas chromatography-mass spectrometry instrument, Carpenter investigated the molecular weight of the chemical causing the peak. He then went to the literature to look for precursors of GHB.

"The second one I came up with was 1,4-butanediol. I was very suspicious of what I was actually dealing with - I wanted to be sure it wasn't something that was just in that particular batch of beads.

"So I went and bought some more beads from Toys 'R' Us and put them through our system and found the same thing. Now I was sure that what was coming off the beads was consistent and I was pretty sure it was 1,4-butanediol, but I was not so sure that I was going to ask them to withdraw the Toy of the Year. I was fairly cautious about that."

He then contacted the distributor of the toy in Australia, Moose Enterprise, who put him in touch with the manufacturer, who provided him with a list of ingredients in the beads.

1,4-butanediol wasn't in there, but another chemical, 1,5-pentanediol, was. Still sure it was the former he was looking at due to the difference in molecular weights, he ordered a batch of 1,4-butanediol from laboratory chemical company Sigma (delicately explaining to the sales person that he did not want to use it as a date-rape drug), ran it through the Agilent system and got the same peak.