APPENDIX 8
Letter to the Clerk of the Committee from
Professor J N Perry, Plant & Invertebrate Ecology Division,
Rothamsted Research
I write to refute the three worst examples of
several erroneous statements made by Rt. Hon. Michael Meacher
MP in his evidence to you concerning the Farm Scale Evaluations
earlier this month.
(1) He said: "the actual scope, structuring
and design of the trials was left basically to SCIMAC"
My comment: This is completely untrue. The scope
of the FSE experiments were originally proposed by Defra, who
funded and invited bids to do the work. The structuring and the
design of the FSE experiments were developed by the Research Consortium
that tendered successfully and subsequently carried out the work,
guided by the independent Scientific Steering Committee appointed
by Defra. The design of the experiment was written up by members
of the Research Consortium, in a paper of which I was the senior
author. It was published in February 2003 in an international
peer-reviewed journal, the Journal of Applied Ecology.
The full reference is: Perry, J N, Rothery, P, Clark, S J, Heard,
M S & Hawes, C (2003). Design, analysis and power of the Farm-Scale
Evaluations of Genetically-Modified Herbicide-Tolerant crops.
Journal of Applied Ecology, 40, 17-31. A companion paper
dealt with the protocols for the biological indicators measured,
and was published in the same issue of that journal: Firbank,
L G, Heard, M S, Woiwod, I P, Hawes, C, Haughton, A, Champion,
G, Scott, R, Hill, M O, Dewar, A, Squire, G R, May, M, Brooks,
D R, Bohan, D, Daniels, R E, Osborne, J L, Roy, D & Black,
H I J, Rothery, P & Perry, J N, (2003) An introduction to
the Farm Scale Evaluations of genetically modified herbicide-tolerant
crops. Journal of Applied Ecology, 40, 2-16.
By contrast with the statement made by Michael
Meacher, SCIMAC had no part in the design of the trials, although
they played a restricted role confined mainly to the supply of
seed and herbicide as detailed in the two papers cited above.
(2) He said: "it was, in my view, very
much oriented towards minimizing environmental impact in a way
which is not real because that is not how farmers would behave
in the field where their prime concern, understandably, would
be about commercial yields. I therefore think to a degree these
farm scale trials, in the way they were devised by SCIMAC, did
not realistically assess what would actually happen in the field."
And later:
"I think SCIMAC were responsible for the
design of the trials and that was not something which SCIMAC would
have wished to include"
My comment: Again, this is errant nonsense.
The FSE was designed to require farmers to apply herbicide management
schemes to the conventional crops according to normal commercial
practice and for the GMHT crops according to the criterion of
"cost-effective weed control", which is essentially
the same principle, ie that used to guide commercial practice.
Again, these criteria were discussed in the two methodology papers
cited above, put into the public domain over nine months ago.
Further results were given on 16 October in
one of the eight published FSE papers in Phil Trans Roy Soc,
written by Dr Gillian Champion and colleagues entitled: "Crop
management and agronomic context of the Farm Scale Evaluations
of genetically modified herbicide-tolerant crops". That paper
emphasizes that no grower deliberately used a scheme to minimize
environmental impact. It details the records taken by the Research
Consortium and demonstrates conclusively that herbicide spray
rates and the timings of applications compared well with current
practice and that the audit that they carried out found no evidence
of bias.
(3) He said: "Bayer crop science—and
it was that company which was particularly concerned with maize—it
had certainly been reported to me that they advised or instructed
(whichever word you think appropriate) farmers who were sowing
GM maize to have only one spraying by glufosinate ammonium,"
My comment: This form of anecdotal innuendo
is impossible to refute because too few details of the claim are
given. Suffice it to say that:
(i)
as evidenced in the paper by Champion et al,
there was no evidence of bias; (ii) with the large degree of replication
achieved, whatever data was recorded from a single site would
in any case not have a noticeable effect on the overall results,
which I am confident are robust; and (iii) data in the form of
gross outliers would have been checked, the underlying reason
examined and the data removed if thought dubious, but no outliers
of the form claimed were noted.
The fact that I do not list several other errors
of fact made by Mr Meacher should not be taken to imply that there
were none; however, those addressed above were amongst those that
were most seriously wrong.
November 2003
|
