Reviewing manuscripts

[Marc Roussel]

In article <199403110032.QAA14818 at net.bio.net> CGE at CU.NIH.GOV writes:
>Most manuscripts for review are marked with some variation
>on: 'Not to be duplicated or used in any way'. However, what
>constitutes use? Consider the following situations:
>
>1. The manuscript is related to work you are doing and its
>content shows that an experiment you were planning to do
>would be a waste of time. Must you do it anyway?

     Definitely not.

>2. The manuscript is not directly related to your field but
>its content gives you insight into a problem you face in
>your own work. Do you have to wait until the paper is
>published (assuming it is accepted) before proceeding
>with the experiment it suggests?

     I would go ahead with the experiment, but then do two things:

	  1. Contact the author.  Tell him your situation and ask for a
	     reprint.  This is tantamount to asking the author for
	     permission to use his or her results prior to publication.
	     Furthermore, make sure you acknowledge this pre-publication
	     communication in your own paper.

          2. Refrain from publishing my own results until the first
	     author's paper is published.  This self-imposed waiting
	     period removes the temptation to do anything truly unethical
	     while eliminating any possible perception that anything
	     underhanded was done.

     The second question is much trickier if the paper you were asked to
review is directly related to your research.  I honestly don't know how I
would deal with such a case.  I suppose that it would depend on how
directly the paper bore on my own research; I would hesitate a lot less
if I thought of the point brought to mind by the paper as a minor one than
if it were the central focus of my own work.

				Marc R. Roussel
                                mroussel at alchemy.chem.utoronto.ca