Reply to Royal Society's Report of 18 May 1999:

1.  The original research proposal submitted to SOAEFD with details of the design of the
experiments, research tasks, milestones and deliverables were fully peer-reviewed by expert
refererees of BBSRC in 1995 prior to awarding us the contract.  Thus, the conclusion that
"the work from the Rowett is flawed in many aspects of design, execution and analysis and
that no conclusions should be drawn from it" makes a strange reading, particularly as the RS
referees had apparently not been given a copy of the original proposal.  This is even more
curious because the internal Rowett reports which had been put on the internet by the Rowett
and used for refereeing by the RS do not describe the experimental design and methodology
and contain no detailed discussion and were therefore unsuitable for peer-review. Thus, there
is no credible explanation is given as how the RS arrived at these conclusions.

The RS was offered in a letter by myself dated 19 March 99 that more updated data and
conclusions could be made available to them providing that these were kept confidential and
that, in cooperation with the Working Group to be set up by the RS, I was fully involved in
their scientific discussions and drawing up their conclusions.  Unfortunately, the RS's
decision was that my role only was to provide my data for the Working Group and they will
then with the help of six unnamed referees draw up the RS Final Report.  This, I am afraid,
I could not agree to and therefore I had no part whatsoever in this RS exercise.

2.  As the RS referees and the Working Group had no suitable data for refereeing it is not
clear how they arrived at their damning conclusions to be able to state with such conviction
that any differences between rats fed GM and non-GM-potatoes "were uninterpretable
because of the technical limitations of the experiments and the incorrect use of statistical
tests".  It is the more curious because previously 24 international experts declared the same
data to be of good quality and suitable as a basis for further research.  Moreover, these
experts, unlike the referees chosen by the RS, signed their declaration.

3.  Our work was a pioneering study as acknowledged by even some of the RS reviewers
although not by the RS Report.  The report had not a single reference to anything to indicate
that our work could have possibly had some value no matter how little that might have been
and that it was most unfair to expect our work to have been more than a start of the
biological testing programme of GM-crops.  Notwithstanding the statement in the RS Report
that in the work only one species of animal was fed one particular product modified by the
insertion of one particular gene by one particular method, the results of this work was still
one more than had been done before with GM-crops.  We have NEVER inferred from the
results of the work with GM-potatoes that GM-foods were harmful to human beings.  Indeed,
we never said that GM-potatoes were harmful to anything but rats.  However, we have to
point out that the genetic modification method we used is almost identical to the methods
used in most presently marketed GM-foods in the UK without any biological testing.

4.  Our work is the most extensively known and refereed research in the history of biological
science.  However, because of the Rowett's release of our data into the public domain
without my consent (12,000 hits on the internet in the first week alone) our attempt to
publish the data in a major scientific journal has been almost completely ruined.  Clearly,
the damning report of the RS was no great help in this respect either.


My comments to the RS Report's two specific conclusions on the Rowett work:

First:  Did the GM-potatoes have specific effect on organ development and metabolism of
the rats? 
As these were pioneering studies in these preliminary experiments we have explored different
aspects of the problem of GM testing.  Thus, it is most unreasonable for the RS to complain
that "the structure of the experiments was changed as they progressed which made
comparisons between rats fed GM and non-GM-potatoes more difficult".  I am afraid, it tells
something of the standard of the refereeing that the RS came to the conclusion that even
though some of the results were statistically significantly different, they "did not fall into  
a readily discernible pattern".  If I had been given a chance by the RS to interact with the
Working Group I could have helped them to see the pattern and consistency of the results. 
However, for that the referees and the Working Group should have realized that the four
experiments were different, carried out with two substantially not equivalent GM-potato lines
and that, what they called changes in the structure of the experiments meant that many other
variables (protein concentration, duration, etc) were also different when they tried to compare
the results.  Given the right approach, however, there is consistency as it will be obvious
when the results will be fully published.

A great deal has been said about the design of the experiments which is quite remarkable
because no design is described in the reports and apparently, our original project proposal
to SOAEFD has not been consulted by the RS.  True, no diets, methods, analytical
techniques, rat ages, methods of feeding and other animal experimentation methods are
described but as it has been pointed out on innumerable occasions these were reports (and
not scientific papers) compiled for people who were fully aware of all these and not written
for peer-reviewing.  The implication by the RS Report that we had bias in our measurements
is highly offensive and therefore fully rejected by us.

Second:  Did the GM-potatoes have a specific effect on the immune system of the rats?
The RS Report refers to inappropriate statistical tests.  Thus according to the RS, our claims
that the immune responsiveness of rats fed GM-potato diets was depressed do not stand up.
However, this is clearly based on the referees' and the Working Group's misunderstanding
of the statistical analysis used.  The results of the independent multivariant statistical analysis
so praised particularly by reviewer no. 4 show that the immune responsiveness of rats fed
diets containing GM-potatoes was always significantly and consistently depressed.

Conclusions:
In contrast to the conclusions in the RS Report which is based in most instances on poor
refereeing and using inappropriate internal reports for their peer-reviewing, our data reliably
and convincingly demonstrate that the inclusion of GM-potatoes in the rat diet has a number
of harmful effects on growth, organ development and immune responses.  Although these
experiments were "preliminary", they were well designed, expertly carried out and subjected
to correct statistical analysis.  Thus, the results could well serve as basis for further
developments.  Had the RS taken up my offer of cooperation with them we could have
arrived at the real significance of our work with GM-potatoes for GM-food testing generally. 
Sadly, speed was more important for the RS than my cooperation and therefore we have
missed a great opportunity for advance.  Thus, we now have to wait till the results of this
and any follow-up studies are fully published.