March of Dimes Funds Useless Animal Torture -->> Give to PETAinstead

Brian L. O'Connor kate at
Sat Aug 5 14:16:39 EST 2000

In article <398B87E2.BCFBC093 at>, Rick Bogle <rbogle at> wrote:

> Tatiana wrote:
> > Very few animals used in scientific research are "random source,"
> > which is the correct terminology for animals derived from unknown 
> > backgrounds.

          Correct.  The majority of dogs and cats used are
          "random source" animals.  But dogs and cats are
          in the distinct minority of all animals used.

          So ... I guess I don't see anything erroneous in Tatiana's
          comment to this point ...

> > Almost ALL animals used in research are bred in commercial facilities
> > solely for this purpose.  This serves two purposes: to control the 
> > genetic background of the population, and to ascertain any diseases 
> > animals may have.

          Correct - I would add the minor qualifier "try" to change
          the wording to:  "... and try to ascertain ..." (like
          everything "of this world" errors sometimes occur, new
          discoveries are made, etc.).

          And the commercial facilities are remarkably successful 
          in doing just this.

          How long could they remain in business were they
          distributing animals right and left that produced
          consistently inconsistent results??????

> This is misleading.

          What is misleading? 

> Carefully controlled and documented genetic backgrounds are the
> rule for the burgeoning genetically manipulated and engineered 
> mouse market being promoted by facilities such as Jackson Labs, 
> and in a few dog and rat strains as well, but, with other 
> experimental animals genetic backgrounds are unknown to the
> actual researcher. This is certainly the case with primates.

          Yeah ...

          So ... perhaps we could agree that Tatiana's comment
          "Almost ALL animals ... is correct and properly qualified.
          And we could also agree that Tatiana's comment "Very
          few animals ... are 'random source'," is completely
          correct and properly qualified ...

> > For instance, a  random source animal may have a virus that 
> > could nullify the scientific results.

          Tatiana is absolutely correct.  Still, it is possible for 
          animals to carry a virus load without that affecting 
          the results.

          And it is also possible that random source animals
          might actually be preferable for some experiments
          specifically because they "rule out" the influence
          of genetic similarity on the end-point.

> A common and growing criticism of animal studies
> is that animals used in experiments are frequently infected
> with known and unknown viruses or are otherwise weakened due
> to various stressors.

          The animals are frequently infected with unknown
          viruses ...

          I guess the obvious question is this ... if the
          virus is unknown, how can we know that the animals
          are infected by it?

> These problems confound many results.

          Can we go back to the bit about "misleading" again?
          You know - the word you used to characterize a statement
          of Tatiana's?
          Because your comment here is grossly misleading.

          What you should have said was: "These problems *can*
          confound results in important ways."  This would
          be an entirely true statement, while drawing attention
          to the real limits of your claim.

          Don't you think my phraseing is a little less ...
          misleading than yours?

          Having said this,if you know which results don't 
          "fit" with the rest of the scientific literature 
          you know that the results themselves are misleading 
          and then you need to account for the lack of "fit."

          "Oh shit!  Did I screw up?  Probably!  But where?  
          How can I design a study that will tell me?  But
          what if I *didn't* screw up?  What does that mean?" 

          And what might that mean for a scientist?  Grant 
          proposals!  A dream come true!

          "We have identified a very important discrepency in
          how Bd/++ mice respond to therapeutic doses of  ..."
          On the other hand, if the results *are* consistent with 
          the literature, how on earth can you know that they are 

          Do you *really* think that scientists are unmindful
          of the need to distinguish between "true and false
          positives" and "true and false negatives"?

          What is the purpose of control groups in an experimental
          design, if not to make this distinction?

          I don't know what I might find more amazing - your 
          ignorance of the process, or your contempt for it!

> Acknowledging this design weakness, NIH has begun funding 
> Specific Pathogen Free (SPF) colonies of  monkeys for research use. 
> These are currently the minority of monkeys available to
> vivisectors.

          Which validates the notion that the commercial
          suppliers of animls are doing it right ...

          By the way.  How are pathogen free animals created?

> >  All conditions must be carefully controlled for interpretation to be
> > valid,
> Whole live animal models confound the notion of careful controll 
> for all conditions.

          This is true, and Tatiana's comment should have been

          But it is not necessary to control *all* conditions
          in order for a study to be valid.

          It is only necessary to control the conditions that
          are known to affect results, and reasonable conditions
          that seem likely to affect the results.

          This is where "repeatability" comes into the equation.
          AR folks frequently bash science because scientists are
          perceived as doing basically the same thing for the
          20th time, the 20th showing only that there was no difference
          in results between it and the first 19.

          But most of those 20 studies are undertaken to "rule out"
          exactly the sorts of subtle and confounding variables 
          you are suggesting invalidates animal studies.

          If the results are unchanged through all 20 studies, how
          important are the 20 variables studied?  Do those studies
          as a group strengthen or weaken the original results?

          Sorry ARA's - you can't have it both ways ...
> Factors contributing to individuality are poorly understood and 
> the frequently small number of experimental animals used in any 
> particular study make the control problem very difficult. Trends 
> are hard to spot.

          I don't know of any IACUC that would pass a study in
          which there wasn't a proper justification for the
          number of animals proposed (I've seen proposals kicked
          back because of too few numbers, and I've seen 'em
          kicked back because the number seemed unjustifiably

          Funding agencies look at such things as "sample size
          calculations" you know (or maybe you don't).  When
          submitting a grant proposal, you need to justify the
          number of animals you plan on using, not to mention
          the species and strain.  And statisticians have come
          up with a way to calculate what a proper sample size
          is for a given study.

          A single study is just that.  The results of a single
          study, when published, are measured against data and
          interpretations already in the literature, and are
          subsequently challanged by future studies.

          Science is incremental.  For better or worse.  Far more
          often than not, big changes come very slowly (sensational
          reports on the news notwithstanding ...).
> > and most research protocols call for animals that are free from any
> > number of pathogens and virusses.

> This is generally incorrect.

          I agree and disagree.  There is often a blanket statement
          having to do with animal health.  At least in the proposals
          I've reviewed.

          The statements were seldom specific, except in cases
          where results could be influenced.

          Then I was burried with details ...

> The many protocols I have reviewed  have never commented on pathogen
> burdens in the the animals to be experimented upon unless the protocol
> is specific to a particular organism,

          Which is as it should be. 

> in which case the protocol might mention that the animals  would be 
> tested prior to the procedures to demonstrate that they were
> not infected. But even this is rare.  Generally the animal is assumed 
> to be disease free.

          Yeah ... but the person doing the "assuming" is the
          guy whose grant proposal would be turned down, and/or
          whose reputation and future would be trashed in a flash 
          if he screwed up something major.

          It happens, of course.  But it is rare.  Very rare.
          You only have to see someone pilloried publicly by
          their peers once to know you don't want it to happen
          to you ...
> > Some companies conducting animal testing may use random source 
> > animals, etc.,  but even then, most of these do not come from 
> > the pound.  Most shelters do NOT dispose of animals by selling 
> > them to companies.  It is truly rare to come across those that do.
> This is incorrect. Most states regulate animal shelters on a county
> level. I have never heard of a direct sale to a supply company, but 
> sale to a university for research purposes is not unusual.

          Could you provide evidence for that?  AFAIK, 
          Universities are required by law to buy experimental 
          animals from federally licensed dealers.

          Has that changed?

          [ ... ]


More information about the Bioforum mailing list