I have been thinking about Randy's comments, and the area under the chart 
lines I imagine to be created under his suggestion, and the area under the 
actual data plot lines charts I present on my site at 
http://www.honeybeeworld.com/diary/files/drop.htm 
... 

Randy said, 
>When I look at your "average" chart, I see a baseline mite drop level typically 
>between 10 and 20 mites per day, punctuated by spikes after you treat. But 
>my point is, if your oxalic treatments were effective, I'd expect to see the 
>baseline drop permanently after each treatment, and to near zero after four 
>treatments! That doesn't appear to be the case. "

...and I think the simple answer is this: 

If we stayed at Randy's baseline of, "between 10 and 20 mites per day" for 
62 days straight, we get a total drop of 1240 mites dying naturally over that 
time, far fewer than I counted.

If use my data and we start at 14.5 mites.day and taper (linearly for the sake of 
convenience)  to the 4 mites we see dropping today, sans the oxalic effect
we get an expected (14.5+4)/2 x 62 = 574 naturally dead mites over the 
period of my observations, also far less than the real count with OA treatment.

The real cumulative average drop is 2279 -- by actual count, far more than the 
projected 574 (or 1240) natural mite deaths predicted.

In the first, artificially skeptical case, we would still calculate that the treatments
killed 2279-1240 = 1039 more mites than the background natural drop -- even 
that natural drop is exaggerated.

In the second case, using the actual data, then the treatments killed 
2279-574 = 1705 mites more than could have been predicted to die and drop 
naturally, or 3.97 times the expected natural drop over that time, (subject to 
the assumptions above). 

Quod erat demonstrandum.

             ***********************************************
The BEE-L mailing list is powered by L-Soft's renowned
LISTSERV(R) list management software.  For more information, go to:
http://www.lsoft.com/LISTSERV-powered.html

Guidelines for posting to BEE-L can be found at:
http://honeybeeworld.com/bee-l/guidelines.htm