Crop and Food Research is being accused of tunnel vision on genetic engineering (GE) by one of its former scientists.

Biotechnologist Dr Elvira Dommisse, who worked on the early stages of Crop and Food's GE onion experiments before the current field trials began, says GE crops have not lived up to their initial promise and the Crown research institute should invest more in conventional plant breeding.

The institute's application to carry out a 10-year Lincoln field trial of broccoli, cabbage, cauliflower and forage kale genetically engineered to contain a natural pesticide to kill caterpillars was heard by the Environmental Risk Management Authority (Erma) in Christchurch last week.

Dommisse worked for the old Department of Scientific and Industrial Research and then Crop and Food from 1985 to 1993. She left because she found the work unrewarding and could not see it solving the problems people said it would.

She was critical of the "lack of precaution and lack of thinking" in Crop and Food's application to Erma, and said scientists working in the GE area were under pressure to develop lines that would become commercially viable.

Some scientists were not keen on GE work but were afraid to talk out about it for fear of losing funding on which jobs depended.

"New Zealand has invested quite heavily in it. As a scientist, once you narrow down into GE your skills are very much in that area. You can't just say, `I don't like this area any more, I'll zip over to plant breeding instead'.

"You have to try to push it - 'we have got this GE stuff, what are we going to do with it now? We have to keep getting our salaries for the next 10 years, get funding that will keep this project going'.

"If you can get a 10-year bloc of funding, you are home and hosed," Dommisse said.

She doubted the field trial would be a useful exercise if it were approved. Most people ate broccoli, cabbage and cauliflower for health benefits and would be unimpressed by GE brassicas.

By the time the 10-year trial ended, that would mark 30 years since the experiment began.

"They could have been spending that time and money to develop new commercial lines. What they could do instead is put a bit more money into conventional brassica breeding, using hand pollination and selection to look for good traits without tweaking the genes."

Scientists were "theoretical people, not growers", she said.

"They have done this under very strict conditions in the glasshouse but not in the field. You can't just transfer that to the fields, it's completely different."

On the last day of the Erma hearing on Friday, Crop and Food project leader Dr Mary Christey admitted there were no cast-iron guarantees all the GE material could be contained on the site.

"I don't think you can give an absolute to anything, but we would have a high level of probability of detecting things," she told the hearing.

Christey said she would not engage in any research that would compromise the environment her children inherited.

"I'm interested in ensuring the environment is preserved for them. I wouldn't engage in this research if I didn't think it wouldn't be damaging the environment.

"At the same time, I can see GE plants growing overseas and I can see the benefits that can accrue."

BioAg New Zealand founder Phyllis Tichinin said if the trial went ahead it was important for New Zealand's "social cohesion" that it was scientifically robust and advanced the country's international "scientific mana".