Answer by Researcher davidsarokin on Sat 19 Jan 2008 - 11:38 pm UTC:

noclue,

Do I have a study for you!

An enterprising high school student won a national science contest with a
study titled:  "Bottled water: how safe is it?", which was eventually
published in 2005 as a formal scientific paper in the journal, Water
Environment Research:


http://www.ncbi.nlm.nih.gov/sites/entrez?db=pubmed&uid=16381148&cmd=showdetailview&indexed=google


As you can see from the above abstract, the goals of the study are right on
target:

"The objectives of this study were to examine the effects of time and
storage temperature on bacterial growth and characterize the types of
microorganisms contaminating bottled water after drinking once from the
bottle..."

and the findings show significant bacterial growth at all temperatures over
the course of 48 hours, though colder temperatures resulted in markedly
slower growth:

"The bacterial count in bottled water increased dramatically, from less
than 1 colony per milliliter (col/mL) to 38,000 col/mL over 48 hours of
storage at 37 degrees C. Bacterial growth was markedly reduced at cold
temperatures (refrigeration) compared with room temperature, with 50% fewer
bacterial colonies in 24 hours and 84% fewer colonies in 48 hours."


I was able to read the full study via a subscription database, and I would
encourage you to request a copy via interlibrary loan (your local public
librarian should be able to help you).  I can't quote at length due to
copyright restrictions, but the study involved the following:

--store-bought bottled water was tested after taking a single sip from the
bottle

--bottled water was initially free of bacteria

--after a single sip, bacterial counts were made at 2, 8, 24 and 48 hours

--with water kept at 37 degrees C (body temperature) bacterial counts
stayed low for approximately two hours, then began a rapid increase

--the increase was about a hundred-fold between 8 and 24 hours

--not surprisingly, the colder the temperature, the slower the growth,
although growth was signficiant even at refrigerator temperatures.

--after 24 hours, even refrigerated water had bacterial counts in excess of
limits recommended by some health authorites

--after 48 hours, refrigerated water had, on average, one-sixth the
bacteria counts of room temperature water



The study also tested tap water under the same conditions, and found it to
be markedly resistant to bacterial growth -- much more so than bottled
water -- probably due to chlorination of tap water.


=========================

You also asked about cap on/cap off as a factor.

The study I mentioned didn't examine this as a variable, nor did I see any
other studies that address this question.  However, as a microbiologist
myself, I can make an informed guess.  

The presence or absence of a cap on a half-filled water bottle probably
wouldn't make any difference to the amount of bacterial growth in the first
48 hours or so.  That's because the amount of air in the bottle would be
very large, relative to the amount of gasses either consumed or released by
the growing bacteria.  

Picture yourself alone in a movie theater.  As you inhale and exhale, your
breathing has very little effect on the composition of gasses in the
theater as a whole, since you are only one person in a very large space.

However, fill up the theater with a large crowd, and it's a different
story.  The same holds true with the bacteria in a bottle.  As bacterial
growth becomes so robust as to be clearly visible, then it will begin to
affect the composition of gasses in the bottle.  At that point, it would
make a difference whether the bottle is capped or not, but it would be long
past the point where anyone would consider actually drinking from the
bottle. 


Hope that's the information your were looking for.  If there's anything
more I can do for you, please let me know by posting a Clarification.

Cheers,

David