Additionally, CO is a metabolite for some methanogenic bacteria, specifically in production of acetyl CoA .
My point is, as long as enough diversity has survived in the ecosystem, it is quite possible that not only CO2/H20/O2 are able to maintain a balance, but _most_ common molecules would be able to remain at nontoxic levels for the plant.
By that logic if you had sufficient biodiversity to start there would be some limiting nutrient that would fix the stable state biomass of the whole system.
To me, metabolism is one of the most interesting biological topics. It is fascinating how intricate of a system has evolved and it shows how difficult it truly is to know the end consequences of changing literally any chemical concentration in the body.
Here is one of the more intricate maps for solely human metabolism: http://www.cc.gatech.edu/~turk/bio_sim/articles/metabolic_pa.... Crazy stuff.
They can be pretty small and made to fit in nicely. The cheapest way to power them would probably be to use a simple coated electrical cable (through a layer of wax, or through something less leaky), I suppose.
Having them relay data in some fashion while having all the electronics not be exposed to erosion/etc might be a bit tricky.. ideally, just the actual sensor surfaces themselves would reside within; coated wires would all go through the seal, etc. Or, the devices could be (not very easily) shielded, and just the sensor surfaces be exposed..
Of course the most fun way to do this would be to have a whole self-sustaining arduino/whatever running inside, relaying data wirelessly, and having it be powered by small solar panels (would still need to take care of erosion/etc.)
Collect all data, plot it (including the light levels), and see if you can make anything out of it. :)
edit actually, I can now imagine a really large aquarium, with not much light passing into the very middle; for mostly entertainment purposes, it would be nice to also be able to place a cheapo camera there which could take (infrequent, like once every 5min or 20min) timelapses. Another camera with IR lens if needed, for neat composite images.. etc etc. Not sure of possible scientific value re: the latter, but hey!
That said, even if it falls short of a 100% perfect hermetic seal it's still darn impressive.
I've looked all over the google and can't find anything that points to this being false.
There's an interesting discussion on the bottled plant over at skeptics.stackexchange.com  that covers many of these reasons, but in more detail. The crux of it is that we can't really prove whether or not David Latimer did or did not water this plant in the 40-odd years since bottling it, so we're left with taking his word for it.
But at the end of the day I don't think it really matters. He looks like a nice guy I wouldn't mind having for a neighbor and really appreciates his plants. Maybe he's spinning a yarn, maybe he's not. Does it really matter? No.
Edit: Someone already linked this 5 hours ago (I should've ctrl+f'd it), so toss 'em some upvotes .
This story for example. Don't you wonder how the interior of the glass has stayed so spotless for 40 years? It has to be cleaned, but they don't explain how.
Maybe it's some kind of magnetic cleaner, or maybe he's simply opening the bottle up to clean it. It's an obvious question, but it's typical of the Daily Mail's cynical and patronizing attitude that they don't care to answer it.
Post science not speculation.
...clear glass encourages the growth of algae...
The bottle garden should be cleaned inside and out once a month.
Algae growth tends to be rapid in bottle gardens... This monthly scrubbing and cleaning out is the key to the whole success story.
It seems the story is indeed disingenuous. Thanks for posting.
Whichever types it contains now, surely it started out with :)
By mass, that vast majority of living material in that bottle is almost certainly bacteria, some nitrifying, others denitrifying. All otherwise decomposing, consuming, and recycling the various chemistry within.
Think about the earth as a sealed globe within space, and you start to understand how good a job bacteria does at balancing the chemistry of life.
I've actually seen something similar to this at a local museum called an "Ecosphere", which includes tiny shrimp living inside. They've been known to last for over ten years.
Carl Sagan's review of an Ecosphere: http://www.eco-sphere.com/sagan.html
Or like mikeash replied, you can replace most of the nitrogen with helium.
I don't suppose my parents have any of them left though... they tended to get kind of crusty with something growing on the glass and blotting out the light.
I had one; amusing, though seemed finicky about lighting.
Very interesting, I had thought they were sustainable for much longer than that. I suppose even the most tightly controlled starting conditions will lead to unpredictable results when you have organisms reproducing.
I think these are more about the incredible toughness of the shrimp involved. Their native habitat consists of rainwater puddles held in volcanic rock, so the shrimposphere is probably fairly luxurious.
Pretty much; this guy claims that basically, the shrimp slowly starve over the course of a couple years, shrinking after each molt.
Your shrimp are new shrimp made out of the same molecules as their ancestor shrimp.
Wait, what's this? What happened?
Fungus and other pests destroyed some important plants and the eco system collapsed.
It had been built in the middle of a desert but urban expansion means the site is reasonably near housing now.
There were a bunch of photos of the dusty derelict interiors floating around a few years ago.
I kind of wish people would try again, just becuase.
Ah, here's a summary of the various things that went wrong, including details about the concrete: http://biology.kenyon.edu/slonc/bio3/2000projects/carroll_d_... .
It doesn't mention fungus, but does talk about other ecosystem problems (the pollinating insects died out, the morning glories took over, etc.).
My first thought was also, "Where does the CO2 come from?"
I suspect this is a very fine balance - if you're off by just a little, one or the other would die. There's not much room for error in a jar.
This is an illusion; it's a dialectic between the two. Problems lead to a shift in balance, not a death.
Why would all bacteria die if there is less oxygen? I imagine the weakest bacteria would die, thereby increasing the amount of O2 per bacteria, and thus reaching equilibrium again.
Article mentions that most of the jars failed.
It would be nice to hear an explanation.
It's an easy experiment to reproduce and works great if you have enough time to wait around.
(Just assuming it's true)
I am curious how one might go about reproducing this exact ecosystem in other bottles by cloning the original, without adding contaminants.
Under these assumptions selling clones could become a commercial proposition.
Can a plant survive bottled in its own ecosystem for 50 years?
The stopper doesn't look like its particularly well fastened and could potentially pop up and allow leaks. Impressive but without closer inspection it looks like a slightly flawed execution to me. Maybe I'm just being pedantic. A picture of the starting point would be nice.
"[he] has only watered the plant twice, the last time in 1972 when he oiled the plastic stopper so that it wedged so tightly it hasn't been out since."
Some kind of breakdown here: http://www.andrew-drummond.com/2014/05/daily-mail-and-mail-o...
-- John C Hardy
Second, the plants chosen for bottle gardens are very sturdy and tolerant of non-optimal environments. Humans are a lot more sensitive.
If you're interested in the topic, read "The Martian" by Any Weird, a very well-researched SF story about an astronaut who gets left behind during a Mars mission and has to survive for several years using equipment that was only specced to sustain the expedition for a month.
There's also the issue that it will cost tens or hundreds of billions of dollars to put people on another planet for an extended period of time.
Elon Musk wants to put thousands of people on mars  and he's one of the very few individuals with the resources to actually do it successfully. He has had some highly ambitious goals in the past (rocket ship company, electric car company) and has been wildly successful.
NASA supposedly had some funds appropriated to them recently for missions such as researching Mars colonization, they have had plans and discussions of what would be necessary for basic Mars colonization for decades. 
Future of clean energy ?
For that article it's probably the Daily Mail: http://www.dailymail.co.uk/sciencetech/article-2267504/The-s...
He started it in 1960 with 'about a quarter of a pint of water', then added water in 1972.
You can hear David asking at 35:20 in the BBC Radio 4 program Gardeners Question Time first broadcasted Friday 18 January 2013.
The original copy text is under all these 10 orginal photographs taken by Phil Yeomans/BNPS published 21 Jan 2013: http://bnps.photoshelter.com/gallery/52-year-old-sealed-gard...
There is also a close up of the seal:
I see this more and more even in edited publications. It's especially irksome here since the article contains almost no actual writing.
It's working in the same constraints the origins ed evolved for... Just smaller
I mean how can we actually test that this guy isn't just making things up, and wasn't opening the bottle to water the ecosystem?
A potentially crazier idea: measure the trace amounts of uranium, plutonium, and other long-lived fallout products. Atmospheric nuclear testing continued on a semi-regular basis until 1980, and so the makeup and concentration of the fallout would be different if it had been unsealed after 1972.
I'm not sure about tritium dating for the water. Reading the USGS website on it, it seems to indicate that the way tritium dating works is that they seal some water in a copper tube and see how much helium magically appears as the tritium decays. They ask for two 500mL (two 1 pint) samples.
So if tritium has been decaying into a sealed glass container for 40 years, should there be a measurable amount of helium in the container?