Peppered Moths

"The industrial melanism of the Lepidoptera is the most striking evolutionary change ever actually witnessed in any organism, animal or plant." — Bernard Kettlewell, Heredity, 1955¹

"Kettlewell was not a fraud, just a naturalist who wasn’t that good at experimental design." — Jerry Coyne, 2012²

Summary

The peppered moth, Biston betularia, is the textbook case of natural selection.  Most of the textbook story holds up.  The fraud charge does not.  The genetics turned out differently from what both sides expected.

1. The rise and the fall were both watched. The black form, carbonaria, was first recorded near Manchester in 1848, when it was rare. Around Manchester it made up 98 to 99% of moths in the 1950s. After the clean-air laws it fell to 16% by 2002 and to 1 to 3% by 2011 to 2014. In Michigan and Pennsylvania it fell from over 90% in 1959 to 6% by 2001.
2. Birds do the selecting. Michael Majerus released 4,864 moths near Cambridge over six seasons, 2001 to 2006, at natural densities, and let them choose their own resting places. Birds took 29% of the black moths and 21% of the pale ones. The selection coefficient against black moths was 0.091 per day.
3. The critics were partly right. The textbook photographs were staged, and Kettlewell's 1950s releases used unnatural densities, times and resting places. Most wild moths do rest off the trunks, as they said. They were wrong that the moths never rest on trunks, that bird predation was never shown, and that Kettlewell committed fraud. Judith Hooper hinted at fraud in 2002. Jonathan Wells, in Icons of Evolution (2000), called the photographs staged but did not allege fraud.
4. The black allele was new. It is a single insertion of a 21,925-letter transposable element into the gene cortex. Its most likely date is 1819. The estimate is skewed, with a median of 1763 and a middle half from 1681 to 1806. The catalogue line "a shift between alleles that already existed" is wrong for carbonaria and right for the older intermediate forms.
5. It scores as a modification, not a gain. The insert is copied DNA, about one letter in 18,450 of the genome, that raises the output of a gene region already active in the wing. By the scoring of Measuring Biological Information it is modification, up, by copy, like the dog FGF4 retrogene. It adds no unique functional sequence.
6. Creationist groups diverge. Answers in Genesis accepted Majerus's result in 2012. Current articles from Creation Ministries International and the Institute for Creation Research ignore it and repeat claims it answered.
7. What it shows. Selection changed a colour ratio within decades and reversed it when the environment reversed. A new allele arose within recorded history. No new function arose, and the species ended where it started. The series argument in its per-lineage form is in Benchmarking Evolution.

Part 1: What was watched

The rise

The first black peppered moth on record was caught near Manchester in 1848.  The record's wording suggests it was rare but not unknown.³  Cook and Saccheri, who between them have worked on the moth for decades, sum up what followed:  "In the peppered moth, the increase was so great that in some industrial parts of England the original wild type was almost lost by the end of the century."⁴

The nineteenth-century numbers are thin.  The often-printed "about 98%" near Manchester by 1895 comes from Grant and Wiseman's 2002 review.⁵  The series compiled for the 2016 genetic study has no 1895 count.  Its first number after 1848 is Chester, 1900 to 1906:  149 black moths of 180.⁶

The black form is dominant and controlled by one locus.  Intermediate insularia forms are alleles at the same locus.  Kettlewell listed over 200 British moth species with dark forms, about a fifth of which became commoner in industrial areas.⁷

The fall

Britain passed Clean Air Acts in the 1950s.  At Caldy on the Wirral, sampled every year from 1959, the fall did not begin until 1969.⁷  After that it was fast.  The Manchester-area series compiled by van't Hof's team runs as follows.

The table is from Supplementary Table 3 of the 2016 paper, which cites Kettlewell 1973 and papers by Cook and others for each row.⁶  Later samples are small, but the trend is clear:  the allele "declined from ~99% to less than 5% in its industrial heartland since 1970."³  Cook and Turner found the same rate of decline at two Leeds sites 0.5 km apart, "probably the first confirmation for melanic moths of the course of natural selection through replication at a pair of adjacent sites".⁷

The same thing in America

The American peppered moth is the same species.  In south-eastern Michigan, black moths were more than 90% of the catch by 1959.  Nobody sampled again for over thirty years.  When Bruce Grant, Denis Owen and Cyril Clarke trapped there in 1994 and 1995, black moths had declined as they had in Britain, in step with falling sulphur dioxide and soot.⁸  By 2001 black moths in Michigan and Pennsylvania were down to 6%.⁵

The American data also corrected the textbook story.  The decline happened with no visible return of lichens to the trees.  Grant and his co-authors concluded that "the role of lichens has been inappropriately emphasized".⁸  Soot on the bark matters, and lichen may matter less than the textbooks say.

Why this counts as observed

The rise was recorded by collectors as it happened, and the fall by scientists looking for it, on two continents, each tracking its own clean-air laws.  Under the series' membership rule the case is observed.  Only the date of the mutation, in Part 4, is inferred.

Part 2: The experiments

Kettlewell, 1953 and 1955

Bernard Kettlewell, an Oxford physician turned geneticist, wrote in 1955 that in more than twenty-five years he knew of no case of anyone seeing a bird take a camouflaged moth from its correct background.¹  In 1953, in a polluted wood near Birmingham where about 85% of wild moths were black, he released marked moths onto trunks and boughs early in the day and trapped them at night.  Of 447 black males, 27.5% were recaptured.  Of 137 pale males, 13% were.¹  He concluded:  "This paper supplies it."¹

In 1955 he added a control in clean woodland in Dorset.  He released 473 black and 496 pale moths there⁹ and recaptured 6.3% of the black and 12.5% of the pale.^{9 10}  The advantage reversed between the dirty wood and the clean one, with the same method.  That reversal is the strongest part of his work.  Kettlewell saw robins and hedge sparrows take resting moths, and Niko Tinbergen watched redstarts do the same from a hide.^{1 9}

What was wrong with the method

The critics were right about the method.  The 2012 paper that published Majerus's data lists the problems itself:

"In experiments, moths were often placed on tree trunks, which were argued to be abnormal resting sites; pinned carcases seemed particularly unnatural. Moths were often released at greatly inflated densities, potentially increasing predation. Reared insects from geographically distant sources were often used to supplement wild individuals, and may not have behaved as naturally in recapture experiments as wild moths."¹¹

Each could inflate predation.  None explains why the same method gave opposite results in Birmingham and Dorset.  Still, they needed an experiment designed to meet them.

Majerus, 2001 to 2006

Michael Majerus, a Cambridge geneticist, ran that experiment in a one-hectare rural garden near Coton, Cambridgeshire.  The paper says six years, and its table lists releases for 2001 to 2006.¹¹  Each night one moth went into each of 12 netting sleeves on branches picked at random, fewer than 10 moths per hectare per night.¹¹  The moths settled where they chose, and the sleeves came off before dawn.  Black and pale moths were released in the local proportions.  A moth missing four hours after sunrise was counted as eaten.  About 26% of disappearances were watched through binoculars, and nine bird species were seen eating the moths.¹¹

Black moths lost a larger share in five of the six years.¹²  The authors' model, which allows overall losses to vary by year, gives a selection coefficient of 0.091 per day against black moths, with a 95% interval of 0.028 to 0.157.¹¹  The pooled totals give about 0.10.  Over a moth's life of several days, this is enough for selection of 0.1 to 0.2 per generation, the rate at which black moths were in fact disappearing.¹¹

Majerus also found that 48 of 135 wild resting moths were on trunks, about 35%.¹¹  In night releases of 419 moths, pipistrelle bats took black and pale alike.¹¹  He died in 2009.  Laurence Cook, Bruce Grant, Ilik Saccheri and James Mallet published his data in 2012 and concluded that "it is virtually impossible to escape the previously accepted conclusion that visual predation by birds is the major cause of rapid changes in frequency of melanic peppered moths".¹¹

The experiment has limits.  It tested only the clean-air half, since Britain no longer has sooty woods, as Coyne noted.²  The resting-site survey covered only branches Majerus could climb to.  And three-quarters of the disappearances were not seen.  But for flight to explain the result, black moths would have to fly off more often than pale ones, and nobody has proposed why.

Walton and Stevens, 2018

Olivia Walton and Martin Stevens modelled the moths as a blue tit sees them.  Pale moths match lichen-covered bark;  black moths match bare bark.  Of 500 paper moths with pastry bodies pinned to lichen-covered trunks in Cornwall, 64% of pale and 43% of black survived 48 hours.¹³  The paper's "21% higher survival rates" is a difference of 21 percentage points.¹³  The pinned dummies repeat the old method.  The study's value is the bird-vision measurement.

Selection coefficients

The sources for the table are in the notes.^{7 4 4}  Selection of 10 to 30% per generation on a single locus is strong.  J. B. S. Haldane used this moth as his example of unusually intense selection.  The Haldane's Dilemma article prices what such intensity costs.

Part 3: The accusation

What the critics said, in their own words

The critics made different claims, and they should not be run together.

Michael Majerus, 1998.  The first serious criticism came from the moth's defender.  His book Melanism:  Evolution in Action set out the old experiments' weaknesses.  Cook and colleagues note that his caveats, "critiques by other biologists, as well as points made particularly forcefully in a review of the Majerus book [18], were soon exploited by non-scientists to promote an anti-evolution agenda".¹¹

Jerry Coyne, 1998.  That review was Coyne's, in Nature.  Coyne later summarised it:  "I suggested that Kettlewell’s experiments were so poorly designed that their results couldn’t be taken seriously."²  His verdict then:  "B. betularia shows the footprint of natural selection, but we have not yet seen the feet."²  He questioned the evidence for the cause, not whether selection happened.  The review is paywalled;  its lines are taken from Coyne's own reprint and from critics who quote it.

Jonathan Wells, 2000.  Wells, a Discovery Institute fellow with a doctorate in molecular and cell biology, wrote in Icons of Evolution:

"The most serious is that peppered moths in the wild don’t even rest on tree trunks. The textbook photographs, it turns out, have been staged."¹⁴

Later he wrote more cautiously that "peppered moths do not normally rest on tree trunks".¹⁴  That claim came from the moth workers themselves, and Wells quoted them.  The Finnish zoologist Kauri Mikkola, reporting on caged moths in 1984, concluded that "the species probably only exceptionally rests on tree trunks".  Cyril Clarke and his colleagues, Wells reports, found in twenty-five years of field work "only one peppered moth naturally perched on a tree trunk".¹⁴  Rory Howlett and Michael Majerus wrote in 1987 that "exposed areas of tree trunks are not an important resting site for any form of B. betularia".¹⁴  Wells quotes Coyne's 1998 review as saying that this "alone invalidates Kettlewell’s release-and-recapture experiments, as moths were released by placing them directly onto tree trunks."¹⁴  Hooper, writing in 2002, called it "now universally acknowledged" that Clarke "was right after all".¹⁵  Photographs, Wells explained, used "dead specimens that are glued or pinned to the trunk".¹⁴  He concluded:  "It seems that “Darwin’s missing evidence” for natural selection — at least in peppered moths — is still missing."¹⁴  Wells did not accuse Kettlewell of fraud.  In the whole book the word appears only in the passages on Haeckel, Archaeoraptor and Piltdown.¹⁴

Judith Hooper, 2002.  Hooper, a science journalist and not a creationist, went further in Of Moths and Men.  Looking at Kettlewell's recapture table she asked:  "First there are just a few recaptures — one or two of each type — and then all of a sudden from July first on, he is getting a whole bunch. What can he be doing?"¹⁵  She quoted the Massachusetts moth biologist Theodore Sargent:  "His stuff is too neat!"¹⁵  Then:  "The unspoken possibility of fraud hangs in the air."  In the next sentence Sargent declines the charge:  "It doesn’t have to be fraud."¹⁵  The fraud charge is insinuated, not documented.

The replies

The 2012 paper calls the fraud charge groundless:  Kettlewell's experiments "were later argued to be fraudulent [19] (quite groundlessly: see [3,10,20])", citing reviews by Cook, Majerus and Grant.¹¹  Coyne, who had supplied the critics' best lines, called Hooper's book "execrable" and wrote that "Kettlewell was not a fraud".^{2 2}

The better answer is that the result was repeated.  Cook and Saccheri summarise the experiments that followed Kettlewell:  "Overall, these experiments suggested that the melanic peppered moth had an advantage of up to 2 to 1 over the typical form in industrial locations, where melanic frequency was 80 per cent or more, but a disadvantage, sometimes large, where the frequency was lower."⁴  Majerus's experiment found the same direction of selection.  The frequency records do not depend on Kettlewell at all.  After Majerus, Coyne judged that the data "show fairly conclusively that the Biston story is sound."²

Which criticisms were right

The critics deserve credit for the first five.  Those points came as much from evolutionary biologists as from Wells, and the field is better for them.  On resting sites both sides overstated.  Alan Gishlick of the National Center for Science Education replied to Wells that figures given by Majerus in 1998 "indicate that the moths do indeed rest on the trunks of trees 25% of the time", with 25% on branches and 50% where branch meets trunk, and that Wells had since retreated to "exposed" trunks.¹⁶  Majerus's later count settled it:  48 of 135 wild moths on trunks, 70 on branches and 17 among twigs, and "Of those that rest on trunks, the majority (87%) rest on the north, rather than the south half."¹¹  Wells's careful sentence, that the moths do not normally rest on trunks, matches the data.  His headline, that they "don’t even rest on tree trunks", does not.  The textbooks were wrong to show staged photographs without saying so.

What the creationist organisations said after 2012

Answers in Genesis accepted the Majerus paper within ten days.  Elizabeth Mitchell wrote that "Majerus and others have refuted" Hooper's allegations,¹⁷ and:

"And with Majerus’s observations now confirming moths really do rest on trunks before sunrise and get gobbled up by birds early in the day, we even know the agent of natural selection, at least in England. But the peppered moth is not and never has been proof of “Darwinian evolution” in the molecules-to-man sense."¹⁷

Jonathan Wells answered it in Zombie Science (2017), objecting that Majerus surveyed only branches he could reach and counted missing moths as eaten.  He wrote:  "In four of the six years, more dark moths disappeared than light ones."¹⁸  The paper's table shows five of six.¹²

Creation Ministries International still carries Carl Wieland's 1999 article, which calls the story "riddled with error, fraud and half-truths", marked by an editor's note as historic.¹⁹  Its current article, Joel Tay's "Reclaiming the peppered moth" (2020), does not mention Majerus.  It says "the videos and photos taken by Kettlewell were actually fraudulent" and "Peppered moths do not rest on tree trunks during the day."²⁰  Matthew Cserhati's 2020 CMI reply also omits Majerus.²¹

The Institute for Creation Research published Brian Thomas's "Peppered Moth Still Not Evolving" in 2016, four years after the Majerus paper.  It says "later researchers could not replicate the peppered moth results",²² which was untrue when written.  Randy Guliuzza's 2020 ICR article does not mention Majerus either.²³  AiG's 2012 article is the model the others should follow.

Part 4: The gene

One origin

In 2011 van't Hof and colleagues mapped the black allele and found "only one core sequence variant associated with the carbonaria morph, carrying a signature of recent strong selection."²⁴  Every British black moth carries the same stretch of chromosome around it.  Cook and Saccheri draw the consequence:  carbonaria "is descended from a single mutational event in the recent past", and it spread by migration.⁴

What the mutation is

In 2016 the group identified it.  Of 87 differences between black and pale versions of the region, only one is found in black moths and never in pale:  an insertion in the first intron of cortex.³

• Structure. "The carbonaria insert is 21,925 nt in length, composed of a ~9 kb essentially non-repetitive sequence (except for ~370 nt at the repeat unit junctions) tandemly repeated approximately two and one-third times".³ It is a cut-and-paste DNA transposon, with short inverted repeats and a duplicated four-letter target site.³
• Origin. The genome holds about 255 copies of its repeat unit. No match is known outside Biston.³
• Carriers. 105 of 110 wild black moths caught since 2002; none of 283 pale moths and none of 31 insularia.³ The authors read the five black moths without the insert as carriers of the darkest insularia alleles, older variants that "occasionally produce carbonaria-like phenotypes".³ The moth could already make a near-black form before carbonaria arose, though rarely.
• Effect. None of the insert enters the gene's messenger RNA. It raises one full-length cortex transcript in the developing wing, most in moths with two copies. "How the carb-TE promotes cortex expression is unknown".³

Cortex was not a suspected pigment gene.  In fruit flies it regulates cell division in egg formation.  The authors suggest that more of it changes the timing of scale development in the wing, and note parallel findings for cortex in Heliconius butterfly wing patterns.³

Work published in 2024 suggests that cortex is not the gene that does the colouring.  Three studies of butterflies found that the colour switch in this stretch of chromosome is a long non-coding RNA named ivory, whose transcript is processed into a microRNA, mir-193, that turns down pigment genes.  Shen Tian and colleagues report that "none of the four candidate protein-coding genes within this locus, including cortex, serve as major effectors".²⁵  Luca Livraghi and colleagues say the case for cortex in peppered moths and Heliconius rested on its expression, and that later gene-editing of cortex anchored "the incorrect impression that cortex itself is necessary for color specification during development."²⁶  None of these studies tested the peppered moth, and no published study yet places the carbonaria insert relative to ivory.  The 2016 measurement, more of one cortex transcript in black moths' wings, still stands.  Which gene the insert acts through is open again.

When it happened

The date does not rest on the 1848 record.  Recombination slowly breaks up the stretch of chromosome that travels with a new mutation, so a shorter shared stretch means an older mutation.  The team simulated one million frequency histories fitted to the Manchester counts, simulated recombination along each at the measured rate of 2.9 centimorgans per million letters, and compared the results with the break points seen in 105 black moths.³  The result:

"The probability density for the date of the carb-TE mutation event (Fig. 2d) is highly skewed (median = 1763, interquartile range = 1681-1806) with a maximum likelihood at 1819, a date highly consistent with a detectable frequency being achieved in the mid-1840s."³

So the single most likely year is 1819, 29 years before the first record, but half the probability lies before 1763.  The model allowed only dates from 1000 to 1845, so the curve is cut off at 1845 by construction.^{27 27}  It treated the allele as neutral before about 1750,²⁷ and drew the population size from 10,000 to 150,000.  The effective size estimated from genetic markers is 11,760, with 95% limits of 5,400 to 112,730.²⁷  The paper's summary, "approximately 200 years (generations) old", uses the peak.³  The fair reading is that it most likely happened early in the nineteenth century, and possibly a century or more earlier.  Either way it is a mutation of the last few centuries.

Other dark forms are older

In 2019 van't Hof's group mapped the dark forms of two other moths that rose with coal smoke, the scalloped hazel and the pale brindled beauty.  Both map close to cortex.²⁸  But their dark forms show no link to a single stretch of chromosome, so they "are likely to be much older alleles, recombined onto a diversity of haplotype backgrounds, before spreading with coal pollution", or else several recent mutations.²⁸  The peppered moth's insularia forms look the same, and the authors think they were present when carbonaria arose.²⁸  Carbonaria "stands out as one of the forms convincingly not observed prior to the 1840s".²⁸

Correcting the catalogue line

The catalogue said the change "was a shift between alleles that already existed."  That is wrong for carbonaria, which was a new mutation, and right for insularia and the other moths' dark forms.  A corrected line:  "The textbook case, and the fraud charge against it.  Majerus's experiment confirmed bird predation, and the black form was a single new transposon insertion of around 1819 that raised an existing gene's output."

Part 5: How the change scores

The procedure

Measuring Biological Information asks four questions in order and stops at the first yes.

1. Did the sequence arrive from another lineage? No. Its repeat unit has hundreds of copies in the moth's own genome and no known match elsewhere.³
2. Did a working element stop working? Not shown. The insert sits in an intron, none of it enters the messenger RNA, and the gene makes more of one transcript, not less.³
3. Does something now work that did not exist before? Not shown. Pale moths already make the same cortex transcript in the wing. Black moths make more. If the colour switch proves to be ivory, as in butterflies, the same test applies. Pale peppered moths are speckled with black scales, so a melanin programme already runs in their wings; the question is whether the insert spreads it or starts something new.
4. None of the above: modification, up, by copy.

New sequence and copied sequence

The insertion added 21,925 letters, all copied from a mobile element already in the genome.  The only new letters are the joins at its ends, including a four-letter duplication of the landing site.³  By the series' conversion rule a change by copy counts zero unique functional sequence, with the join noted.  The most famous adaptation in the textbooks made a new allele and a new-looking moth out of copied DNA.

Against the whole genome the change is small.  The peppered moth's genome is about 404.5 million letters long and carries 12,251 protein-coding genes.²⁹  The insert is about one letter in 18,450 of it.³⁰  It sits inside one gene and changes the sequence of no protein.  A small change can still have a large effect:  this one turned a white moth black.  But the size of a visible change is no guide to how much new genetic material it took.

Two comparisons

The dog FGF4 retrogene is a copy of a gene that landed at a new site and raises the output of a growth factor already active in the growing bones.  Dog Breeds scores it modification, up, by copy.  The moth insertion matches it on every point:  copied DNA, a gene already active in the right tissue, more of its product, a large visible effect.

The citrate mutation in E. coli put a copy of the citrate transporter gene under a control region active in oxygen, where the gene had been silent.  That is a new control relationship, so the series scores it a gain despite its zero count.  The moth is not like this.  On the 2016 evidence cortex was already on in the wing and was turned up, not switched on somewhere new.

Does the taxonomy handle this badly?

There is a fair case that it does.  The visible result is new:  no solid black peppered moth was recorded before 1848, and the insertion's authors call transposons a possible "source of major phenotypic novelty".³  A skeptic can say a scheme that calls this "modification" is built to avoid admitting gains.

The answer is that the taxonomy scores the molecule and reports the effect on the animal beside it.  Here the effect is a new colour state produced by pushing an existing regulator past a threshold, as the authors suggest.³  That is a modification in the molecule and a new appearance in the moth, and both are stated.  The score is provisional.  If the insert proves to switch cortex or ivory on where it was silent, it becomes a gain by copy, like citrate.  If it works by breaking a repressor in the intron, part of it is a loss.

Observed or inferred

The spread and the reversal were observed.  The mutation was not.  Its single origin and date are inferred from moths caught after 2002.  Genotyping pinned moths from before 1848, of the kind Kettlewell looked for in museums,⁴ would test the date directly.  No such study was found.

Part 6: What it shows

The mainstream claim

Birds hunting by sight changed the ratio of two colour forms within decades, in step with soot on the trees, and changed it back when the soot went.  Cook and Saccheri call the moth "the type example of rapid response to human-induced environmental change, driven by selective predation".⁴  That is accurate.  It is also true, and less often said, that a new mutation arose and was selected within recorded history.

The creationist reply

The creationist reply is that the moths stayed moths.  AiG:  "A simple reshuffling of existing genetic information within a created kind (so-called microevolution) is not logically connected to the evolution of new kinds of organisms from lower forms".¹⁷  CMI's Cserhati:  "The moths have always remained moths, shifting back and forth in response to the environment."²¹

This side is right that nothing new in function arose, that the species ended where it started, and that natural selection itself was never in dispute.  AiG writes that its authors "fully affirm the observable reality and the importance of natural selection".¹⁷  The textbooks did slide from a change in colour ratio to the origin of new kinds.

It is wrong on the variant.  Cserhati writes that "No new gene variants are being created".²¹  ICR's Thomas writes that "what began as a population of light and dark moth varieties existed through the industrial revolution as a population of light and dark moth varieties."²²  The black variety was not there at the beginning.  It arose by mutation, most likely around 1819.

Answers in Genesis's own older line is half right.  Tommy Mitchell wrote in 2008:  "The variation seen is simply the result of sorting and resorting of the genetic material present in the original moths.  At no time has there been any new information introduced into the genome of the moth".³¹  Georgia Purdom repeated the passage in 2012, in a post that thanked Majerus for his experiment.³²  The first sentence is wrong for carbonaria:  the black allele was not present in the original moths.  The second holds if information means new functional sequence.  The insert is a copy of DNA the moth already carried, about one letter in 18,450 of its genome, and it changed the sequence of no protein.  A skeptic will answer that a copy in a new place is a new instruction, because it changed what the moth looks like.  Part 5 weighs that objection and grants part of it.

Was the insertion directed?

Tay suggests "there may be an environmental trigger, such that more dark trunks means more dark moths are born".²⁰  Thomas suggests the element was pasted in "as part of an internal variation-generating protocol."²²  Guliuzza points to "The repetitive insertion of the transposable element in the same location".²³

The data test this.  A response triggered by soot would insert the element many times, in many districts, on many chromosome backgrounds.  Instead every British black moth descends from one insertion, and the form spread by migration.⁴  There was no repetitive insertion.  Guliuzza also miscounts:  he says 103 of 105 black moths carried it, when the paper reports 105 of 110.  And the other moths' dark forms are old alleles, not fresh responses to soot.  This does not show that mobile elements are never part of a designed capacity for variation.  It shows that this insertion behaved like an ordinary rare mutation, arising once and then selected.

Guliuzza's stronger point is about the place, not the event.  Among the findings that should make a scientist ask whether the change was regulated, he lists "genetic hotspots for wing coloration, coloration master switches".²³  The hotspot is real, and mainstream biologists use the same words for it.  Livraghi and colleagues note that the region "has been independently mapped as the locus controlling color polymorphisms in 15 lepidopteran species":  industrial melanism in three geometrid moths, mimicry in Heliconius and a swallowtail, the dead-leaf disguise of the oakleaf butterfly, seasonal forms in buckeyes, and more.²⁶  Tian and colleagues call it a "hotspot" locus, "repeatedly implicated in generating intraspecific melanic wing color polymorphisms across 100 million years of evolution".²⁵  Colour change in moths and butterflies does not come from anywhere in the genome.  It comes, again and again, through one old switch.

That pattern fits a design view in which animals carry built-in, adjustable traits.  The mainstream reply is that it is also what unguided evolution predicts.  A switch that controls one trait with few side effects is where mutations can help without harming, so selection keeps finding it.  Livraghi's team found exactly that low cost:  knocking out ivory changed scale colour with "no detectable effects other than in epithelial scales".²⁶  The hotspot fits both views, so it is fair evidence that variation in these insects runs through a small number of pre-existing switches.  It is no evidence that the 1819 insertion was triggered or aimed, which the single origin still argues against.

The reversal

The fall matters as much as the rise.  Selection favoured black when the bark was black and pale when it was pale, kept the variation while it paid, and is now removing it.  It built nothing.  It chose between two levels of one gene's output.  The mainstream side can say the classic case of selection has been confirmed against its critics.  The creationist side can say it shows selection sorting variation and leaving the species where it began.  Both are true, and neither settles the origin of new functions, which is the question Benchmarking Evolution asks.

Part 7: The ledger

The series runs the same ledger for every organism:  population size times generations times mutation rate gives the mutations explored, set against every adaptive change found.  For the peppered moth only some cells can be filled.

The empty cells have not been measured for this moth, and no figure from another insect is put in their place.  The population figures are for the population studied;  no census of British peppered moths exists.

The change was one step.  One insertion produced the whole effect, with no second mutation needed before selection could act.  The series' price list says one-step changes are found quickly, and this one was found once in a population of tens of thousands.

Unpriced changes are few.  The insularia forms were selected too, from older variation with unknown mutations.  The moth has not been searched genome-wide for other adaptive changes, so the ledger is a lower bound.

The unit of account gives zero.  The only new allele was made of copied sequence:  a real, large, selected change the unit cannot count, like the dog retrogene and the citrate mutation.  That is a limit of the unit as well as a finding about the moth.

None of these numbers should be carried over to mammals by head count.  Benchmarking Evolution shows why that comparison fails and why only the per-lineage form of the argument survives.

Objections and responses

"The photographs were faked, so the story is a fraud."  The photographs were staged, and textbooks should have said so.  But they illustrated the claim.  The evidence is the frequency records and the predation experiments, including one built to avoid the old flaws.

"Kettlewell committed fraud."  Hooper hinted at it without evidence.  Later workers found the same direction of selection, and Majerus found it with a better design.  Coyne rejects the charge, and Answers in Genesis says it has been refuted.

"Peppered moths don't rest on tree trunks, so the experiments were worthless."  Most don't.  About two in three rest on branches or twigs, as Mikkola, Clarke, Howlett and Majerus reported before Wells repeated it.  But about a third do, and Majerus's experiment let moths pick their own resting places and still found black moths eaten more often.

"Majerus assumed every missing moth was eaten."  He did, and only about a quarter of disappearances were seen.  But the finding is a difference between black and pale moths released side by side.  Any non-bird loss would have to hit black moths harder to produce it.

"The black form was always there."  For insularia and for two other moths, that looks right.  For carbonaria the genetic data say it arose once, most likely around 1819.

"It's just microevolution."  The site agrees that nothing new in function arose.  But a new allele did arise by mutation within recorded history.  The accurate summary is a new allele of an old gene, selected, with no new function.

"The same gene region changes again and again, so the change was designed or directed."  The hotspot is real:  one region has been mapped as the colour switch in 15 species of moths and butterflies.  That fits a built-in, adjustable switch.  It also fits unguided evolution, because a switch with few side effects is where helpful mutations are found.  Either way, this insertion arose once and spread by migration, so it was not triggered by soot.

"A new colour pattern is a new trait, so 'modification' is special pleading."  Part 5 reports both the new appearance and the molecular class, and says what evidence would change the score.

Issues and unanswered questions

• Mechanism. How the insert works is unknown, and the score depends on it. Butterfly studies of 2024 point to the ivory RNA and its microRNA mir-193, not cortex, as the colour switch in this region. Nobody has yet tested ivory in the peppered moth or placed the insert relative to it.
• The date is a model estimate, bounded by a prior of 1000 to 1845 and by neutrality before 1750. Genotyping pre-1848 museum moths would test it.
• Sources not read: Coyne's 1998 and 2002 Nature reviews, Cook 2003, Majerus 1998 and 2005, Grant's and Rudge's reviews of Hooper, Tutt 1896, and the full texts of Grant 1996 and Grant and Wiseman 2002. Coyne's 1998 words come from his own 2012 reprint and from Wells's quotation. Majerus's 1998 resting-site figures are Gishlick's. Wells 2017 was read in a copy without page numbers.
• "98% by 1895" is untraced to a count.
• No mutation rate or transposon insertion rate has been measured for Biston, so the ledger cannot compute mutations explored.

Verdict

The peppered moth is a sound case of natural selection, observed in both directions.  Birds hunting by sight changed the ratio of black to pale moths within decades as soot darkened the trees and changed it back when the air was cleaned.  Kettlewell's methods were flawed and the textbook photographs were staged, as the critics said.  Kettlewell was not a fraud, and Majerus's 4,864-moth experiment, published in 2012, confirmed the predation with a design that answers the old objections.  Creationist articles that still say the moths never rest on trunks, or that nobody repeated Kettlewell, are wrong and should be corrected.

The black form was a new mutation, not old variation:  one insertion of a 21,925-letter transposable element into cortex, most likely around 1819, with a median estimate of 1763.  It copied existing DNA and turned up an existing gene region, so it scores as a modification, up, by copy, and adds no unique functional sequence.  The moth shows selection sorting a new allele quickly and reversibly.  It does not show the origin of a new function, and the species ended where it began.

Sources