"Martin Brown" wrote in message
...
In message , Franz Heymann
writes

wrote in message
...
Franz Heymann wrote:

"dave @ stejonda" wrote in message
...
There's a news story at

http://www.newscientist.com/news/news.jsp?id=ns99994051

which reports that laboratory studies have suggested that
glyphosate
increases the risk of fungal infections.

I have read the URL. The item is interesting, but it was quite
incorrect to
publish it in New Scientist at this stage, since the work is quite
inconclusive so far, and the article contains zero quantitative
information
to help the reader decide whether the effect is statistically
significant or
not.. The folk involved are not making any claims yet, they are only
making
suggestions.

New Scientist isn't a 'learned journal' where scientists publish
papers, it's a magazine that tries to popularise science and publishes
news to that end. This sounds like science news of a sort so it seems
reasonable for New Scentist to publish it. It is a news item rather
than an article so doesn't imply much regarding the truth of the
claims.

That the matter is discussed in New Scientist is in fact my main
objection.
The results have not been subjected to peer review and are therefore to
be
regarded purely as gossip from publicity-seeking scientists.

Why? The article is genuine science news and the version in the printed
magazine is actually well balanced and informative about live research.

The headline "Weedkiller may encourage blight" is a bit sensationalist
but then journalists are paid to make mundane news sound interesting.
The article is otherwise pretty well written and covers what has been
observed already, plans for future work and the following caveats:

"But the investigators warn against jumping to conclusions."
[snip]
"Hanson stresses that the real issue is whether the fungi leave more
spores in the soil. It is also possible that the effect is simply due to
herbicides leaving more dead plant matter in the soil for fungi to grow
on and is not directly caused by glyphosate"

Seems to me like quite a good article on a potentially controversial
topic.

The fact that it is gossip does not detract from ite interest. The
problem,
however, is that some readers might be misled by it.

It is impossible to prevent this.

Not so. The problem can be avoided by delaying publication in any form
until the work is completed.

Franz