This browser cannot draw the plates: they need WebGL. Another browser, or one with hardware acceleration turned on, should show them. The notes in Proof can still be read.
Where did this pollen come from?
Three beetles arrived carrying pollen from flowers no search had found.
Ahis print, on its own sheet
Bhim, drawn: his back
10 mm
Aaxonometric
Belevation · the height the print cannot keep
Cplan · the print
S
R
S
R
S
R
R and S apart
–
02468 Å
hatched: no rotation reaches it · a mirror would
05 Å
Afield sheet · one exposure
S
R
exposure
startend
10 mm
OBSERVED
the count, over two days
S0
R0
MODELED · THOUGHT EXPERIMENT
their prints, R turned as in step 4
SR
vial · not to scale
vial · not to scale
1 nmthe scale of the plate
ADisa forficaria, a flower from above · schematic
Ba later flower
about 13 mmthe flower, top sepal to tip of lip
9.88 mmthe beetle, head to abdomen tip
1
2
3
WHERE HE FITS
1hind legs on the lateral sepals
2antennae stroking the petals
3abdomen tip in the cleft of the lip
5 mm
Athe years
the one known plant
at scent lures
2016
2017
2018
2019
2020
visits filmed
not seen above ground from early 2019
a search nearby: no other plant
three males, fresh pollen
Ba male like the three, from below · one pollinarium, enlarged
5 mm
viscidium · very large
caudicle · short
pollen, in packets
EVIDENCE · ONE POLLINARIUM
unknown pollen
shape
—
ITS
—
time
—
Cthe barcode · one band a base
one pollinarium, 2020 · not published: drawn as reported
a leaf of Disa forficaria · MT703014.1
Dhow many plants?
?
?
?
?
at least 1other plant: one could have given all three their pollen
ONE OR MORE
Aan exposure, without a map
×1
10 mm of paper
ONE OR MORE
PROOF
What these plates claim, tested in this browser; and, in the notes, where each claim comes from.
PLATE I · THE PRINT
Four prints made with the plate’s own drawing: back up and belly up, with his pollen and without. The control draws the pollen beside his thorax, where a print would show it: the test must catch it.
PLATE II · SHEET 1 · THE PRINT
Lit control shows C as an ordinary lit render instead of the print: a picture that does know which atom is on top, so the test must catch it. It applies while the molecules are on the plate.
PLATE II · SHEET 2 · THE BEETLES
The plate stages the observed count, eleven and two. The model itself sends each beetle to S with probability 11/13; run it to see how the count varies from one exposure to the next.
PLATE III · THE MEASURES
The flower and the beetle, drawn with the plate’s own code and measured in pixels.
PLATE IV · THE BARCODE
At every offset from −80 to +80 bases, how many bands meet their band.
PLATE IV · HOW MANY PLANTS
Every way the three beetles’ pollen could be shared among plants, and the fewest plants that fits the evidence.
NOTES
Observed, Modeled, Interpreted
Each line under the plate carries one of three tags. Observed: what the paper reports, seen by its authors or concluded by them from what they saw. Modeled: made for these plates, to show what was observed (the drawings, the flights, the prints, the sheets). Interpreted: the project’s reading, not the paper’s. A line with no tag names something.
The beetle
A male Chorothyse hessei, a longhorn beetle: 9.88 mm from the head to the tip of the abdomen (mean of 19 males), antennae 12.33 mm. Drawn from those measurements and the paper’s description, not traced from a photograph. His underside follows the general plan of a longhorn beetle: Modeled.
The pollen
A pollinarium is a mass of pollen in small packets (massulae) on a short stalk (the caudicle), with a sticky base (the viscidium). In the paper’s photographs the pollinaria are bright yellow: the yellow on these plates is theirs, and means pollen and nothing else. The paper gives no measurements of them; their size here is drawn.
R and S
Look along the bond from C16, the starred atom, to its hydrogen, the hydrogen pointing away. C16’s other three neighbours, in order of priority (the ring oxygen, the ring carbon C15, the ethyl group), run clockwise in R and anticlockwise in S: the Cahn–Ingold–Prelog rule. The coil beyond the hydrogen draws that turn as a screw. It is read from the atoms, so turning a molecule never changes it, and a mirror always does. The paper found the flower’s own compound to be the S form.
Apart
The root mean square of the distances between matching atoms, hydrogens included. The molecule is flexible; this is one computed shape of it, held rigid. Over 300 computed shapes, the closest any rotation brings R to S lies between 1.6 and 3.0 Å, never zero.
The print
A thought experiment. A molecule is hundreds of times smaller than a wavelength of light and casts no shadow. The print here records only how much of the model lies along each vertical line: it keeps what blocks the light and no heights.
The vials
On sheet 1 the vials are not to scale: one is more than ten million times the size of the molecule. On sheet 2 vials and beetles are to scale (the bar is 10 mm); the distances between them are shortened. The paper gives the vials’ volume, 5 mL, not their size: they are drawn 20 mm across, as standard 5 mL vials are.
The count
Observed in the field: over two days, eleven beetles came to S and two to R, each form in an open glass vial with a dry reed, and no flower. The plate stages that count: one exposure shows exactly eleven and two, in a random order. Every flight, landing and moment is modeled, drawn the same way whichever vial a beetle reaches, and the two days are compressed into one exposure.
The flower
A schematic, from the paper’s description: about 13 mm from the apex of the dorsal sepal to the apex of the lip; it faces up; narrow petals curl back like an insect’s antennae; the lip is densely hairy and kidney-shaped, with a cleft at its tip. In the paper’s photographs it is wine-red, with a yellow-green base to the lip; here it stays in white. How its parts are arranged is Modeled. A plant makes up to thirteen inconspicuous flowers, usually one open at a time.
The visits
Eleven visits to the one known plant were recorded over 12 days in 2016 and 2018; the beetles’ behaviour was photographed and filmed. The visits lasted 221 seconds on average (± 67.2, standard error). A male stands with his hind legs on the lateral sepals, strokes the petals with his antennae, buries the tip of his abdomen in the hairs of the lip’s cleft, extends his aedeagus into the cleft and ejaculates. A lip sampled after a visit held more than ten thousand sperm.
The fit
The authors write that the flower’s size closely matches the male’s length, which lines his genitalia up with the cleft. The plate sets the two numbers side by side rather than call them equal.
The pollen’s way
His sternum touches the viscidia, so the pollinaria come away fastened to the underside of his thorax. On a later visit, packets of pollen are left on a stigma.
The flight
Modeled. The scent is never drawn. Beetles coming to the flowers flew slow zig-zag paths, as insects do that follow a scent; the path here borrows only that.
Sound
Synthesised, not recorded in the field, and off until turned on. In Plate II, a faint buzz for each beetle in the air, placed by where it is over the sheet; a tick when one settles; water when the sheet is washed: the same for S and R. In Plate III, a faint buzz while the beetle is in the air, placed by where he is over the plate, and a tick when he settles on a flower. It follows the modeled flights. Nothing else makes a sound: Plates I, IV and V are silent.
The reading
“The molecule makes the beetle arrive; the flower’s body makes the mistake continue” is the project’s reading. The authors find scent key to attraction, and think the flower’s shape plays an important role once a beetle has found it. On scented model flowers (a dark bead with wire “antennae” and a tuft of hair like the lip’s), the first beetles on three of four days already showed prolonged mating behaviour, bending the abdomen; to the same models without the scent, none came.
The years
The one known plant had not appeared above ground since early 2019, the paper reports (2021); a search nearby found no other. In 2020, three males arriving at the experiments with lures carried fresh pollinaria. Beetles visiting the known plant were filmed in 2016 and 2018.
The shape
The pollinaria of Disa forficaria are distinctive for their very large viscidium and short caudicle. The paper gives no measurements of them; none are drawn as measures.
The barcode
The reference is GenBank MT703014.1, the ITS of Disa forficaria from fresh leaf material (the paper does not name the plant; the known plant was then the only one known): ITS1, the 5.8S gene and ITS2, complete, with parts of the genes on either side; 833 bases. The record does not say where each part begins and ends, so none is marked. Each band is one base, its tone set by the base (A palest, then T, G, C). The ITS of one pollinarium, from a beetle caught in 2020, was a 100% match. That sequence is not published with the paper (only the reference is), and how much of the ITS it covered is not given in it, so the plate draws the reference twice: the pollinarium’s strip is the reference, as reported. Sliding shows the bands out of place and in place; it measures nothing.
Why a barcode cannot count
Here ITS, with the pollinarium’s shape, names the species. Plants of one species usually carry the same sequence, so a match cannot tell one plant from another, or how many plants gave the pollen.
“At least two”
The university’s press release (27 May 2021) spoke of “at least two other individuals”, “confirmed by genetic barcoding”, and a newspaper, the Mail & Guardian (8 June 2021), wrote that two more plants must exist. The paper says one or more: one pollinarium was barcoded, and ITS names the species, not the plant.
The area
“In the area” is the paper’s phrase. Plate V is an exposure, not a map: no edge, no ground, no coordinates; looked at closer it shows only paper. Its bar measures the paper (the sheet is taken to be 240 mm wide). Where the pollen came from, the evidence does not say; where the known plant grew is not given here.
No place
No place, landscape or month of the field work is shown or implied.
Darwin
In his book of 1862, Darwin described moths sent to him with one orchid’s pollinia on their tongues: seven pairs on one, eleven on another. Those two moths, he wrote, “must have sucked many more than the seven and eleven flowers, of which they bore the trophies”: a lower bound, read from what an insect carried. From another orchid’s flower he inferred a moth no one had yet seen. These plates run that reasoning the other way, from what three beetles carried to plants no search had found. He knew another Disa, from a correspondent’s notes: its flowers, he wrote, are “rarely visited by insects”. He took the bee orchid to be self-fertilised, and doubted Robert Brown’s idea that its likeness to a bee kept insects away.
Sources
Cohen, C., Liltved, W. R., Colville, J. F., Shuttleworth, A., Weissflog, J., Svatoš, A., Bytebier, B. & Johnson, S. D. (2021). Sexual deception of a beetle pollinator through floral mimicry. Current Biology 31, 1962–1969.e6. doi:10.1016/j.cub.2021.03.037. The facts on these plates come from it, checked against its text and its photographs. The sequence is GenBank MT703014.1, the authors’ own. The press is quoted once, in “At least two”, set against the paper. The note Darwin comes from his book: Darwin, C. (1862). On the Various Contrivances by which British and Foreign Orchids are Fertilised by Insects, and on the Good Effects of Intercrossing. London: John Murray, pp. 35–37, 65–69 and 197–198; Disa from its second edition (1877), chapter 2.
Made with
three.js (MIT licence); EB Garamond and Courier Prime (SIL Open Font License). Every figure is drawn from the paper’s measurements and descriptions; none is traced from a photograph.