Tip

A seedling leans toward the window. Everyone has seen it. The question Charles and Francis Darwin asked in 1880 was narrower and much harder: which part of it can see?

They worked on canary grass, Phalaris canariensis, raised in the dark so the pale sheath around the first leaf — the cotyledon, or coleoptile — came up hungry for light and bent hard toward any it was given. Then they started covering it. A cap of blackened tinfoil over the top fifth of an inch, and the entire lower stem could stand in bright sun for a working day and stay dead straight. Take the cap off and put the same plant back in front of the lamp and it bends like anything. The part that moves is not the part that senses. Something travels down.

That something got a name in stages after Darwin's death: Boysen-Jensen slid mica into a cut coleoptile in 1911 and found the influence would cross gelatin but not a barrier on the shaded side; Paál swung a severed tip off-center in the dark in 1918 and got a bend with no light at all; Went caught it in a block of agar in 1926 and called it auxin. This room stops at 1880 on purpose — it is about the experiment, not the answer.

The part the diagrams leave out

Every textbook drawing of this experiment shows two seedlings: one capped and straight, one bare and bent. Clean. Darwin's own page is not clean. Twenty-one seedlings went under tinfoil caps; seventeen stood upright and four leaned toward the light anyway. Three more were capped and thrown out because the foil was too heavy. Five glass tubes were thrown out because the black paint cracked in the sunshine and let a little light in. He wrote all of that down, including the regret: after abandoning one line of cuts as too injurious, he says plainly, which we now regret.

So this room does not draw you the diagram. It deals you the seedlings. Every deck below holds exactly one entry per plant Darwin reported for that treatment, in his own counts. Grow them one at a time and watch a trial actually come out — unevenly, the way it did.

The strip of marks is the trial filling in: one mark per seedling set up, pale until it has been grown, then colored by how it came out. Empty a deck and you have Darwin's published totals back exactly, because the deck never held anything else.

Honest gap, and it is a real one: Darwin recorded outcomes in words, not degrees — quite upright, slightly bowed, considerably bowed, greatly curved. The bend angle drawn for each of those words is this room's, picked from a small range, not measured by anyone in 1880. Only the deepest range has a number behind it: he describes untouched seedlings as “almost rectangularly bent towards the light,” so the hardest bends here sit near a right angle. The order is shuffled, because his notes don't say which seedling was which — so you cannot ask this room whether the fourth plant under a cap was one of the four that leaned, and it will not pretend to know. Only the totals, the treatments, and the thrown-out plants are his. The untouched-control deck is the one place a count is invented: Darwin never numbers the free seedlings, he only ever writes that all the many others in the same pots bent, so the twelve in that deck are this room's and the verdict on each of them is his. The drawn seedling is schematic; his ranged from .04 to 1.15 inch tall, most about .75.

And a smaller one, about this site rather than Darwin: nothing here is seeded. Every other room that shows you something built from chance rebuilds it identically forever — that is the promise the whole garden runs on. This room breaks it deliberately and does not keep a seed, because an experiment you can only ever see come out one way is exactly the diagram it was built to argue with. Reload and you get a different order. The totals will not move.

Source: Charles Darwin, assisted by Francis Darwin, The Power of Movement in Plants (London: John Murray, 1880), ch. IX, Sensitiveness of Plants to Light: Its Transmitted Effects. Every count on this page is transcribed from that chapter. Also named, not built: Boysen-Jensen (1911); Paál (1918); Went (1926); and the blue-light receptor phototropin 1 behind all of it, identified in Arabidopsis by Huala and colleagues in 1997.