Carol,

sorry, I thought I was merely repeating a previous point. You should not take the weight/length (or whatever you measure) of the individual plant as the datum to start with but the average value of the plants in each condition. That means one experimental run only gives you these two data to start with.

Amos,

you're right with your point (3). But please tell me what the opposite to 'right field' would be. Wrong field? Little Hawk might not agree.

Clint,

If something would work consistently (or well above chance) not knowing why it works wouldn't be a good reason not to use it. Though to know why something works is always better, for it gives us more control about the situation.
The problem scientists have with many alternative healing methods is not that they do work but we don't know why (that would be great adn a good start for research), but that they do not seem to work better than any placebo cure (and sometimes even worse).

Your example shows beautifully why alternative cures are believed to be valid: and was finally cured of it --instantly-- by a chicken-sacrifice ceremony (my emphasis). That is the post hoc ergo propter hoc thinking leading to a feeling of validity. The correct word would be after (in the eyes of a scientist) because we can never be sure (from one case alone) that the cure was in any way responsible for the healing. The doctor you cite tries his best to give an explanation in terms of placebo effect (and he does it well).

I think that evidence based medicine is working with the placebo effect as well, up to 90% of all times. That's why I seldom go to a doctor when I think a tea and a bit of bedrest will do enough for me (and cost less), but in the 10% or more cases I go to a doctor doing evidence based medicine.

Wolfgang