Add ‘light bulbs growing on trees’ to ‘flying cars’ on your list of things Science didn’t bring you for Christmas

There once was a kickstarter that raised half a million dollars with the idea of genetically modifying trees to glow in the dark, thereby replacing the need for street lights. It failed. I’m not surprised. They are citing technical difficulties in getting the needed genes inserted (it looks like they were using the luciferase reaction, which is the enzyme used by fireflies).

To get the plant to glow well, the research team had to insert six genes. But they never could get all six in at once. At best, some plants glowed very dimly. (The photo above of the glowing plant is a long exposure, making it appear much brighter than it actually is.) Evans says that he realizes now trying to insert six genes into a complex organism like a plant—rather than single-celled bacteria or yeast—was premature.

That’s why TAXA, the company that Evans set up to work on glowing plants, eventually pivoted to creating genetically modified moss that smells like patchouli to subsidize continuing glowing-plant research. Moss is a simpler organism. They got the scented moss growing, but the last bunch was contaminated and could not be shipped to customers. Without the moss, there was no way to keep funding the company. That’s when Evans realized that glowing plants weren’t happening.

I don’t even…yes, this is harder than the popular press sometimes makes it sound. Keep that in mind when you hear some transhumanist wackaloon speak blithely of modifying human genes to increase intelligence or longevity or whatever. The tools keep getting better, so maybe someday it’ll be easy to spritz in any number of genes into any organism we want, but the hard part is always going to be figuring out what genes do what we want, and don’t do what we don’t want.

That’s not what made me instantly doubt the project, though. It was wondering what fantasy world they were living in to think bioluminescence would have adequate output to come even close to the illumination we can produce with street lights. Consider the amount of light you can get from a jar of fireflies, if you’ve ever caught them; they’re pretty, but it really isn’t enough to read by, or to compete with a cheap flashlight. I’ve seen seas lit up with swarms of bioluminescent organisms, but even there it was only bright in contrast to the total darkness of the night. Right away I’m questioning how effective even a “tree” that was a solid cylinder of bioluminescent molecules would be.

Another problem: your tree cells are busy making light, like a collection of glow sticks inside…but they’re surrounded by bark. Trees aren’t transparent, you know, so only the thin outer layer of living cells will count for light production.

So just make the bark the glowing part, you say. Can’t. The outer layer of bark is dead, the luciferase reaction requires constant consumption of ATP and O2. This is an energy-intensive reaction, even if the enzyme is remarkably efficient at converting chemical energy into photons. So you’re basically trying to engineer a plant with an alternative pathway to compete with the Calvin cycle, subverting the whole process of absorbing photons to produce chemical energy to instead throw away that energy by emitting photons. And you’re simultaneously expecting the plant to grow into a massive tree.

Yeah, it’s a hard problem alright.

You now may be thinking but wait — we’ve made glow-in-the-dark fish, and you can buy them for five bucks down at the pet store. That’s a whole different process. Glofish don’t make light, and don’t require internal energy to produce photons. They absorb light and re-emit it at a different wavelength, causing a color shift, a process called fluorescence. I’ve worked with a lot of fluorescent molecules (and even with luciferase), and it takes some non-trivial optics to separate out the shifted color signal. That light is also faint — it takes further non-trivial electronics to amplify it into a useful signal.

The dodge of producing patchouli-scented moss is a transparent fake-out, too. Nothing about that solves any of the problems of introducing a complex and energetically expensive set of genes into a plant. It’s also remarkably pointless. There’s an herb, Pogostemon, that grows naturally and already produces the scent, so why not just go with that? Also, I may have fond memories of patchouli perfume everywhere in the hippie culture of the 1970s, but there seems to be a lot of people who don’t care for it or its associations, so it’s an odd choice for a replacement. When I go to the store to buy a light bulb, I’m not going to be satisfied if the clerk “pivots” and sells me Axe body spray instead.

Bottom line: if you invested in this, you got taken.


  1. dhabecker says

    The most primitive of the human species figured it out many moons ago. You set the wood on fire.

  2. says

    Also, I may have fond memories of patchouli perfume everywhere in the hippie culture of the 1970s,

    Oh gods, I don’t. It was a patchouli overdose, everywhere. People pouring tons of it all over themselves, their clothes, burning patchouli incense by the packet, ugh. You couldn’t go anywhere without being assaulted by cheap patchouli.

  3. Randomly Generated Username says

    Well sure, it’s easy to say now that using firefly DNA wouldn’t work, because hindlight is 20/20.

  4. marcoli says

    Scientists fail to make street lights out of glowing plants? See? Science does not have all the answers! Therefore I choose to believe in special creation, and obviously climate scientists are all wet when it comes to their claims about global warming.

  5. says

    If I’d known about it at the time, my first question would have been, “have you calculated the energy requirements for a sheet of plant cells to produce a useful amount of light?” I checked their FAQ after I discovered this project, and strangely, that isn’t a question anybody asked. In fact, there are only two questions there:

    What’s the risk of cross-pollination? Am I going to infect my neighbors gardens?

    How do we grow and care for the seeds/plants?

    Apparently, the backers thought it was obvious that you could turn a tree into a giant light bulb, and only asked irrelevant questions.

  6. slithey tove (twas brillig (stevem)) says

    Also, I may have fond memories of patchouli perfume everywhere in the hippie culture of the 1970s,

    raising hand sheepishly

    I’ve recently been seeing ads for glow-in-the-dark pebbles to scatter in driveway pavement and walkways to make them visible at night. Sounds like a better solution than genetically modifying plant to bioluminate.

    I guess the OP was inspiraration for Moffat’s Sherlock rewrite of Hound of the Baskervilles, where the red herring was a doctor concocting glow-in-the-dark bunny-rabbits.

  7. says

    *raises hand*

    I was a backer on this Kickstarter. Wasn’t aware of the “trees as streetlights” non-idea — I was just backing it because glowing Arabidopsis would’ve make for a cool nightlight around the house, and I wanted to help counter some of the anti-GMO hysteria that’s out there in the general public. I knew the KS was a gamble when I funded it — “sure things” don’t exist in science research — but it wasn’t until I read their updates that I learned just how much basic research they had awaiting them. I’ll admit, it was a bit of an eye opener.

    All said, I’m disappointed with the outcome, of course, but I don’t regret helping to fund it. Through their updates I got to learn a bit of the nitty-gritty lab work on transplanting genes into higher organisms — stuff I otherwise never would’ve been exposed to — so yay me.

  8. Rich Woods says

    @Randomly Generated Username #5:

    because hindlight is 20/20.

    You are now on my jokey kill* list.


  9. Raucous Indignation says

    How are you going to control the level of light as the tree grows bigger and bigger? Dimmer switch? And what is it going to do to all the critters that, you know, live in trees?

  10. Pierce R. Butler says

    Back in the ’80s, I read a glowing (sorry!) puff piece about a scheme to inject luciferase genes into crop plants, set up so as to be activated by the hormones produced by stress conditions. Thus, farmers would need only to walk (this was pre-drones, kiddies!) around their fields at twilight to see which areas needed more irrigation/pesticide/whatev. That far back in pre-history, I took the picture of lit-up tobacco leaves to mean the lab people had already accomplished most of the relevant work, with only the commercializing hurdle left to leap, because back then nobody had invented Photoshop Phakery® to activate our cynicism genes.

    Now maybe I can stop wondering why I haven’t seen drought-stricken fields Shining in the absence of UFO floodlights or major supernatural entities … yet.

  11. leerudolph says

    Patchouli already got a bit of mockery back in Chekhov’s The Cherry Orchard, well before the hippies…

  12. says

    Another overhyped project – solar roadways. I thought it was a neat idea, and still do, but…. There are problems – 1) solar cells don’t generate the power to run them, so there is no instance of them “ever” actually powering themselves, 2) the tech relies on LEDs, so you have a trade off between how many LEDs you have on it, vs. how much is a solar panel, and then its not actually “visible” in the day, 3) the results is a defunct technology.

    Mind, my thought was that, as long as you limited your intended scope, you could use solar only panels, to power a few select panels with display capacity, but use the more limited “single color”, or some variation (maybe a pattern of two colors, so you have on, color 1, or color 2, e-ink style panels, for your display sections – benefit, these have “much lower” power requirements, they only need to be powered to make a state change, and they could be lit with much fewer LEDs, for night use. The original concept… requires “way” too many breakthroughs, not the least being power storage, which, like the solar panel part itself, just isn’t actually “available” yet, and won’t be for years/decades.A more limited, specialized, alternative.. may be possible though, using existing tech. But, the original project… not so much.

  13. says

    because hindlight is 20/20.

    Which gets me thinking – would it be possible to genetically engineer gastrointestinal microbiota to be luminescent, feed it to some cows and end up with glow in the dark bullshit?

  14. says

    If you just do the energy calculations, you can debunk a lot of pseudo-science. The thing is, it takes a fixed amount of Joules to make something happen (like moving 1 metre against a force of one Newton, which takes 1 Joule by definition; making 1 kg. of water 1° hotter, which takes about 4180 J.; or producing one photon of light of a wavelength λ, which takes exactly h * c / λ J.) and you have to account for all that energy somehow.

    There is just no arguing with Thermodynamics.

  15. says

    Another problem: Think about what you want a street lamp to do. You want it to project a relatively narrow cone of light toward the ground, because that’s what you’re trying to illuminate. Any light that goes sideways or up in the air not only represents wasted energy, it contributes to light pollution. Now think about what a tree is like. It has branches and leaves sprawling all over! Trees are meant to collect light from a source that is overhead and constantly moving about, which makes them possibly the worst design to use as a street lamp. Not only will there not be enough light, most of it will be wasted going up into space or into your neighbor’s window.

    The people who pushed this thing really didn’t think it through very well.

  16. says

    Actually, now I come to think of it, I am sure I remember an Arthur C. Clarke novel in which some species — an advanced race, but whose technology was based on biology as opposed to machinery — bred giant tame hovering fireflies to use as portable lamps.