Vestigial Structures

"A trait can be vestigial and functional at the same time.  It is vestigial not because it's functionless, but because it no longer performs the function for which it evolved." — Jerry Coyne, Why Evolution Is True, 2009¹

"Since it is not possible to unambiguously identify useless structures, and since the structure of the argument used is not scientifically valid, I conclude that 'vestigial organs' provide no special evidence for the theory of evolution." — S. R. Scadding, Evolutionary Theory, 1981²

Summary

The vestigial-organ argument has two versions.  The old one says that some organs are useless leftovers, which descent with modification predicts and creation does not.  The new one says that a vestige is any structure reduced or changed relative to its counterpart in related animals, working or not, and that the pattern of such reductions fits a family tree.  Creationist writers have spent forty years refuting the old version, mostly correctly.  Evolutionary biologists conceded it by redefining the word, and the new version is what they now defend.  Finding a function for the coccyx answers the old version and leaves the new one untouched.

• Darwin argued the old version and knew its weakness. In 1859 he called rudimentary organs "imperfect and useless" and said that on "the ordinary doctrine of creation" they present "a strange difficulty," ³ but on the same pages he allowed that an organ "might be retained for one alone of its former functions."³ Robert Wiedersheim's 1895 The Structure of Man is the source of the long lists. Read from the scan, its conspectus names 86 organs "which may be rightly termed Vestigial" and a separate 16 that still perform "clearly recognisable Functions."⁴ The "180" on creationist sites is a Scopes-trial witness's number, and Answers in Genesis's own chapter marks it "[sic]."⁵
• The redefinition was made in the open. S. R. Scadding argued in 1981 that vestigial organs "provide no special evidence for the theory of evolution" because uselessness cannot be proved.² Bruce Naylor replied in 1982 that "it is incorrect to state that to be vestigial an organ must be non-function[al]" and offered the coccyx as "a functional, vestigial organ."⁶ Jerry Coyne's 2009 statement is Naylor's, not Darwin's.
• A genuine vestige argues for common ancestry, not for bad design. A vestige claim has to show a homologue with a fuller role elsewhere, a real reduction, and an ancestral explanation that fits the tree; a refutation has to show that one of the three fails, and a list of functions touches none of them. The creationist reply that does engage the right claim is Sarfati and Batten's: loss "would prove devolution not evolution."⁷ That concedes reduction and moves the dispute to where the ancestral boundary lies.
• The cases sort unevenly. The appendix has a plausible protective role, and Heather Smith and colleagues find it appeared independently at least 29 to 32 times across mammals and was rarely lost, which undercuts Darwin's shrunken-cecum story on evolution's own terms.⁸ The coccyx anchors the pelvic floor and is a reduced tail on any reading of the bones. Wisdom teeth never form in about one person in four or five worldwide.⁹ Palmaris longus is absent in 1.5 to 63.9 percent of people by population.¹⁰ Ear muscles fire during hard listening and to sounds from behind.¹¹ Goosebump muscles maintain the nerve supply of hair-follicle stem cells.¹² Whale pelvic bones anchor the penis muscles and track testis size across 29 species, and are still small, detached bones.¹³ The panda's thumb is not a vestige at all.
• The shrinking list proves less than either side says. It shows that "useless" was a bad premise. It does not show that reduced homologues do not exist.

Verdict: the argument from useless organs is over, and creationists were right to end it.  The argument from reduced homologues is a common-ancestry argument that stands or falls with homology, embryology and the fossil sequence, and the creationist literature has not answered it by listing functions.  Whale and snake pelves and horse splint bones remain its strongest cases.  The appendix, goosebumps and palmaris longus are weak for either side.

Part 1: What the word has meant, from Darwin to Coyne

Darwin's rudimentary organs

Darwin opened the topic in chapter 13 of the first edition of On the Origin of Species.  His examples were rudimentary mammae in male mammals, the "bastard-wing" of birds, "rudiments of the pelvis and hind limbs" in snakes, teeth in fetal whales, and beetle wings "so reduced in size as to be utterly incapable of flight."³  The argument turned on uselessness.  The same reasoning that shows most organs to be exquisitely adapted, he wrote, "tells us with equal plainness that these rudimentary or atrophied organs, are imperfect and useless," and the creationist explanations of his day, that such organs exist "for the sake of symmetry" or "to complete the scheme of nature," were "merely a restatement of the fact."³

Two pages later he qualified the premise: an organ "rendered, during changed habits of life, useless or injurious for one purpose, might easily be modified and used for another purpose.  Or an organ might be retained for one alone of its former functions."³  His diagnostics were that rudimentary organs "are very liable to vary in degree of development" and are "of greater size relatively to the adjoining parts in the embryo, than in the adult," and his image was "the letters in a word, still retained in the spelling, but become useless in the pronunciation, but which serve as a clue in seeking for its derivation."³

The Descent of Man (1871) applied this to people.  The muscles that move the external ear "are in a rudimentary condition in man."¹⁴  The semilunar fold of the eye exists in man, "as is admitted by all anatomists, as a mere rudiment."¹⁴  Wisdom teeth "were tending to become rudimentary in the more civilised races of man."¹⁴  The appendix was "a rudiment of the shortened part" of a once long cecum: "Not only is it useless, but it is sometimes the cause of death."¹⁴  The os coccyx, "though functionless as a tail, plainly represents this part in other vertebrate animals."¹⁴  Note the last sentence: Darwin's coccyx claim was never that the bone does nothing, only that it does not work as a tail.

Wiedersheim's 86, and the 180 that never was

The long lists come from Robert Wiedersheim, professor of anatomy at Freiburg, whose Der Bau des Menschen reached a second German edition in 1893 and appeared in English, translated by H. and M. Bernard, as The Structure of Man: An Index to His Past History in 1895.  He defined vestigial organs on his second page as "those which were formerly of greater physiological significance than at present.  In the course of generations, in consequence of the adaptation of the body to special conditions of life, they have been, so to speak, put out of the running, subjected to reduction or degeneration, and now persist as mere vestiges."⁴  He did not say they do nothing; he said they do less than they did.

The famous list is the "Conspectus of the Organs Mentioned in the Text," pages 200 to 204.  Counted item by item in two independent scans, section A, "Retrogressively modified, the Organs still performing clearly recognisable Functions," has 16 entries, among them the fibula, the eleventh and twelfth ribs, the fifth toes and the cecum.  Section B, "Retrogressively modified, the Organs having become wholly or in part functionless ... For the greater part Organs which may be rightly termed Vestigial," has 86, from the "Os coccygis" through the wisdom teeth, the appendix, the pineal and pituitary bodies and the plica semilunaris to "the area scroti."  Six glands "Modified under Change of Function" and about 22 shifted structures follow.⁴  Eighty-six is the number Scadding read off the same pages in 1981.²

He was careful about the muscles most often attacked.  Palmaris longus and plantaris appear in section B only "when completely tendinous," and his closing note says they "are not infrequently so well developed that there can be no doubt of their being functional."  He then quoted Henry Fairfield Osborn: such organs "are mere pensioners of the body, drawing pay (i.e. nutrition) for past honourable services without performing any corresponding work."⁴

The number 180 comes from the Scopes trial, where the Chicago zoologist Horatio Hackett Newman testified in 1925 that "there are, according to Robert Wiedersheim, no less than 180 [sic] vestigial structures in the human body."  That is how David Menton quotes him in The New Answers Book 3, and Menton gives Wiedersheim's own number as 86 in the sentence before.⁵  Jerry Bergman's 2000 paper nevertheless opens with "The list of vestigial organs in humans has shrunk from 180 in 1890 to 0 in 1999,"¹⁵ and an undated Answers in Genesis page says "of the 180 or so organs that have been called vestigial in humans, all have been shown to have function."¹⁶  Neither figure survives a reading of the book.

The redefinition, made in public

The turn from "useless" to "reduced" happened in a two-paper exchange in Evolutionary Theory.  Scadding argued in 1981 that "as our knowledge has increased the list of vestigial structures has decreased," and that the argument fails twice over: "it is difficult if not impossible to unambiguously identify organs totally lacking in function," and the argument "has as a premise, either a statement of ignorance (I couldn't identify the function), or a scientifically invalid claim (it does not have a function)."²  The editors thanked Stephen Jay Gould as a referee.

Naylor, a vertebrate paleontologist at Alberta, conceded the shrinking list ("undoubtedly correctly") in 1982 and denied that it mattered: "it is incorrect to state that to be vestigial an organ must be non-function[al]."  Darwin's "retained for one alone of its former functions" showed "that, for Darwin, vestigial organs were largely, but not exclusively, useless to their possessors."  The coccyx is functional, "but just as certainly it is not functioning as an external tail for balance or grasping.  It is a functional, vestigial organ (contra Scadding, 1981)."⁶  To the logical point he replied that "no function" is "simply an hypothesis based on the interpretation of available data, and open to potential falsification," like "all crows are black."⁶

Coyne's 2009 formulation is Naylor's for a general reader: "Opponents of evolution always raise the same argument when vestigial traits are cited as evidence for evolution.  'The features are not useless,' they say."  But "evolutionary theory doesn't say that vestigial characteristics have no function.  A trait can be vestigial and functional at the same time."¹

Bergman saw the redefinition and objected that it "is vague and would allow almost every structure in humans to be labelled as vestigial," since the human jaw, nostrils and brow ridges are all smaller than in the alleged ancestors.¹⁵  He is half right.  The new definition has no content until a homologue and a phylogeny are specified.  But that is what it claims to be: a comparative claim, to be tested as one.

Part 2: What a vestige claim has to establish, and what defeats it

Three things to show

On the modern definition a vestige claim is a bundle of three claims, and each can fail on its own.

1. Homology. The structure corresponds, by position, construction and development, to one in related animals that does a fuller job. Scadding called vestiges "simply a special case of homologous organs."²
2. Reduction. The structure is smaller, simpler, more variable or less capable than its homologue. Darwin's diagnostics belong here: variability, occasional absence, and relatively larger size in the embryo.
3. An ancestral explanation that fits the tree. The reduced state must be the derived one, and its distribution across species must match the phylogeny that other characters give. This step does the evidential work, and a functional finding never reaches it.

A refutation has to show that the homology is false, that there is no reduction, or that the distribution does not fit the tree.  Bergman's complaint that a judgement of reduction "requires that the evolutionary history of an animal is known"¹⁵ is a fair statement of step 3.  It is circular only if the phylogeny used to call the coccyx reduced were built from the coccyx; it is built from hundreds of other characters and from DNA.  Whether that tree is right is the real question, and it is a question about common ancestry, not about vestiges.

Function is the wrong battlefield

A demonstrated function answers "this organ is useless," not "this organ is a reduced homologue with an ancestral explanation."  Menton's 2000 paper on the plantaris shows the mismatch.  It argues for a proprioceptive role and concludes that "any muscle in the body that actually exists in the present, serves some function."¹⁷  That is very likely true.  It does not bear on whether the human plantaris is reduced relative to the plantaris of animals in which it reaches the sole of the foot, which is the claim Cruveilhier made and which Menton himself describes.¹⁷

The reverse error is also common.  A genuine vestige, on the modern definition, is not a defect.  Coyne runs the two claims together when he writes that "tiny, nonfunctional wings, a dangerous appendix, eyes that can't see, and silly ear muscles simply don't make sense if you think that species were specially created."¹  The adjectives carry an argument from bad design; the nouns carry an argument from homology.  The parent article, Arguments from Bad Design, sets out why the two have to be judged separately.  This article judges only the second.

Where the creationist responses answer the wrong claim, and where the right one

Jerry Bergman and George Howe's 1990 book Vestigial Organs Are Fully Functional is the standard creationist reference, and its title states its method.  Bergman's 2000 summary says the appendix "is part of the immune system," gives the pineal and thymus their hormones and cells, and concludes that "the concept of evolutionary vestigial organs is useless, or largely speculative, and certainly is not good science."¹⁵  Menton's New Answers Book 3 chapter makes the same case: Wiedersheim's organs "have been subsequently shown to have useful functions and indeed some have functions essential for life."⁵  Lita Sanders's 2011 reply at Creation Ministries International finds a function or a possible one for each organ on a reader's list; for the ear muscles, "a CMI biologist suggested that one function of ear muscles could be to help remove wax from the ears."¹⁸  All of these answer Darwin 1871 and Wiedersheim 1895.  None answers Naylor 1982.

Two creationist responses do meet the modern claim.  David DeWitt's 2008 reply to a New Scientist list grants that the coccyx "is the anchor point for the muscles that form the entire pelvic diaphragm" and then puts the disagreement where it belongs: "the only reason to claim that the function has been modified is because of evolutionary assumptions.  If you believe that humans descended from animals that possessed tails, then there must have been a modification of the tailbone. ... Without the evolutionary presupposition, the evidence that the tailbone is vestigial evaporates."¹⁹  That is a correct description of step 3.  Its weakness is that the same sentence with "creationist" for "evolutionary" describes DeWitt's own position.  Both sides read the same bones through a prior on ancestry, and the argument has to be settled by whatever settles ancestry.

The second is Sarfati and Batten's 2006 answer: "even if the alleged vestigial organ were no longer needed, it would prove devolution not evolution.  The creation model allows for deterioration of a perfect creation."  On flightless birds they allow that ostriches and emus "derived from smaller birds that once could fly.  This is possible in the creationist model.  Loss of features is relatively easy by natural processes."⁷  This is the honest reply, and it concedes homology, reduction and a historical explanation within a lineage, withholding only the tree beyond the created kind.  So the vestige argument, once answered properly, is about whether the whale's hip, the snake's pelvis and the human coccyx have their fuller homologues inside or outside the boundary the model draws.

Part 3: The human cases

Each case gets three questions: what is claimed, what the sources say, and the verdict on the modern definition.  Planned child articles take each further.

The appendix

Darwin's claim was that the cecum "had become much shortened in various animals, the vermiform appendage being left as a rudiment of the shortened part."¹⁴  Coyne repeats it: "in humans, who don't eat leaves and can't digest cellulose, the appendix is nearly gone."  He allows "some small use," including that "it provides a refuge for useful gut bacteria," and still counts it vestigial because "it no longer performs the function for which it evolved."¹

The child article The Appendix traces both halves of Darwin's claim, finds the safe-house hypothesis resting on one donor's colon and a null pooled recurrence effect, and shows that on the mainstream tree the human appendix was gained rather than reduced.

The function is better than a suggestion and not yet proven.  R. Randal Bollinger and colleagues found in 2007 that biofilms in one unprepared human colon were densest in the appendix, and proposed that their shedding "would be expected to re-inoculate the large bowel with commensal organisms in the event that the large bowel became infected by a pathogen and was flushed out."²⁰  A 2013 review by the same group reports that appendectomy "does not apparently put patients at risk for an initial onset" of Clostridium difficile colitis, but that one study "found a 2.5-fold increased risk of recurrent C. difficile colitis in patients without an appendix."²¹

The comparative evidence changes the verdict, and it cuts against Darwin's story rather than against vestigiality.  Heather Smith, William Parker, Sanet Kotzé and Michel Laurin mapped the appendix onto a mammalian phylogeny in 2013: "The appendix has evolved minimally 32 times, but was lost fewer than seven times, indicating that it either has a positive fitness value or is closely associated with another character that does.  These results, together with immunological and medical evidence, refute some of Darwin's hypotheses."⁸  Their 2017 study of 533 species put the range at "a minimum of 29 times, possibly as many as 41 times, throughout mammalian evolution, while it has only been lost a maximum of 12 times."²²  On evolution's own tree the appendix is not a relic shed when lineages stop eating leaves; it keeps appearing and is rarely lost.  Verdict: the human appendix should come off vestige lists on both sides.

The coccyx

Gray's Anatomy of 1918 describes the coccyx as "usually formed of four rudimentary vertebrae," whose borders "give attachment on either side to the sacrotuberous and sacrospinous ligaments, to the Coccygeus in front of the ligaments, and to the Glutaeus maximus behind them," whose front surface "gives attachment to the anterior sacrococcygeal ligament and the Levatores ani, and supports part of the rectum," and whose apex carries "the tendon of the Sphincter ani externus."²³  Nobody on either side now says the coccyx does nothing.

The comparative claim is not disputed as anatomy either: the coccyx consists of the serial elements that continue as a tail in most other primates, fused and lacking pedicles, laminae and spinous processes, and the embryo's tail region is relatively larger than the adult's.  Coyne's diagnosis is that the coccyx "still has a function (some useful muscles attach to it), but remember that its vestigiality is diagnosed not by its usefulness but because it no longer has the function for which it originally evolved," and that some people carry a rudimentary extensor coccygis "identical to the one that moves the tails of monkeys."¹  DeWitt's answer is that reduction is visible only on an evolutionary presupposition.  Verdict: the coccyx passes steps 1 and 2 on any reading; step 3 is where the ancestry dispute lives.  It is the cleanest example of a functional vestige.  The embryonic tail and human "tails" at birth belong to the planned sibling on embryonic leftovers.

The child article The Coccyx gives the attachments, the sitting and childbirth mechanics, the coccygectomy record (84.6 percent good outcomes in 671 patients, pain scores only), the ape counts, and the 2024 TBXT Alu insertion that is the strongest evidence for the ancestral step.

Wisdom teeth

Two facts are measured.  Third molars are often never formed: Kathleen Carter and Steven Worthington's 2015 meta-analysis of 92 studies and 63,314 subjects put "the worldwide rate of agenesis" at 22.63 percent, with regional estimates from 5.32 to 56.0 percent.⁹  A Malaysian sample of 5,923 radiographs found 38.4 percent.²⁴  And crowding is partly environmental: Katz, Grote and Weaver's 2017 analysis of global skull shape found "modest directional differences between foragers and farmers," consistent "with softer diets in preindustrial farming groups."²⁵

DeWitt's reply is that agenesis "is an example of a loss of information, the opposite of what molecules-to-man evolution requires," and that "one of the reasons that they are problematic is the decreased jaw size of many people today which may be related to diet."¹⁹  Both halves are defensible.  Verdict: wisdom teeth pass the reduction test as variability and frequent absence, the diet finding weakens the impaction half of the argument, and whether the loss is ancestral or within the human line is step 3 again.

The child article Wisdom Teeth and the Modern Jaw shows that agenesis is ancient and inherited while impaction is a developmental effect of soft diet, so diet answers the bad-design charge and leaves the ancestral reading at step 3.

Palmaris longus

The 41st edition of Gray's Anatomy lists palmaris longus among the muscles that control the wrist joint and calls it "a weak wrist flexor" that "through its insertion in the palmar aponeurosis ... also acts as an anchor for the skin and fascia of the hand against horizontal shearing forces."²⁶  Its absence varies enormously by population; a 2015 review put the range at 1.5 to 63.9 percent.¹⁰  In 585 Turkish schoolchildren the right tendon was absent in 30.4 percent.²⁷

Within humans alone, a variable and dispensable muscle reads the same on both sides.  The comparative data say more.  Rui Diogo and Bernard Wood scored 166 muscle characters across the primates in 2011 and found that the palmaris longus "is always, or almost always, present in the specimens of taxa of CS-0 [0], but it is absent in more than 5% of the cases in specimens of Gorilla, Pan and Homo [1]," a character that fits their tree in a single step, "(L 1, CI 100, RI 100)."²⁸  Orangutans, gibbons and monkeys keep it.  Verdict: palmaris longus is good evidence for Darwin's variability criterion and for Wiedersheim's pensioner, and moderate evidence for ancestry, because its variable absence is confined to the one branch that common descent groups together;  a creationist has to read it as a shared variant across three created kinds.  Menton's comparative claim about the plantaris, that there is "no phylogenetically coherent story based on either the size or the location of the attachments,"¹⁷ is the kind of claim that would decide the plantaris case if it were documented across a full primate sample.

The child article Palmaris Longus and Plantaris runs both muscles through the three-step test and finds the plantaris weaker evidence:  on the evolutionary tree it lost its link to the sole in the ancestor of all Old World monkeys and apes, not when humans stood upright, and humans keep it more often than any great ape.

Ear muscles and goosebumps

A 2025 study in Frontiers in Neuroscience recorded the superior and posterior auricular muscles of 20 participants listening to a podcast against distractors.  Posterior auricular activity "was significantly larger when stimuli were presented from the back, as opposed to the front"; superior auricular activity "was significantly larger in the most difficult condition, which required the largest amount of effort."  The authors call this "the response of the vestigial pinna-orienting system to an effortful stream segregation task" and elsewhere a "neural fossil."¹¹  So the muscles do something measurable, and the people who measured it still call the system vestigial, because in a cat the same muscles turn the ear toward a sound and in people they cannot.

The child article Ear Muscles runs the three-step test in full: homology and reduction pass against horses, cats and macaques, and a direction-coded orienting reflex survives, but the ancestral step rests on the primate tree.

Coyne says the goosebump muscles "serve no useful function, at least in humans."¹  Cell biology has since found a second job.  Yulia Shwartz and colleagues reported in Cell in 2020 that the arrector pili muscle and the sympathetic nerve "form a dual component niche to modulate hair follicle stem cell activity," that "APMs maintain sympathetic innervation to HFSCs," and that "cold stimulates not only goosebumps but also hair growth."¹²  A 2014 review had tied loss of the muscle's attachment to androgenetic alopecia.²⁹  Verdict: the arrector pili is not vestigial as a muscle; whether the reflex is a reduced fur-raising reflex is a step-3 question the reflex cannot answer.

The child article Goosebumps runs the three-step test in full and adds one measurement: human hair-follicle density matches the chimpanzee's (Kamberov 2018), so the reflex is not reduced, the hair is.

The plica semilunaris

Gray 1918 describes "a slight semilunar fold of conjunctiva" at the inner corner of the eye and notes that "Müller found smooth muscular fibers in this fold; in some of the domesticated animals it contains a thin plate of cartilage."²³  Scadding's answer was that it is simply conjunctiva "and as such aids in the cleansing and lubrication of the eye ball," and that "the absence of a nictitating membrane in humans is of no significance to a discussion of the function of the semilunar fold."²  That second sentence is the wrong-claim error in miniature: the absence of a nictitating membrane is the whole of the comparative claim.  Verdict: a small, unstudied structure on which the argument rests entirely on homology with the third eyelid.

The child article The Plica Semilunaris finds that the fold intercepts debris and carries immune tissue, that its homology with the mammalian third eyelid is strong, and that the primate series fits need as well as ancestry, so the third step remains untested.

Part 4: Cases in other animals

The panda's thumb

Stephen Jay Gould's 1980 essay made the giant panda's enlarged radial sesamoid the emblem of improvised engineering: "The sesamoid thumb wins no prize in an engineer's derby.  It is, to use Michael Ghiselin's phrase, a contraption, not a lovely contrivance.  But it does its job."³⁰  Three later findings change the picture.  Hideki Endo and colleagues imaged a panda hand in 1999 and found that the radial sesamoid "cannot move independently of its articulated bones" but works with the accessory carpal bone as "a double pincer-like apparatus."³¹  Their CT study of the same hand, published the same year, concluded:  "In this refined system, the giant panda is equipped with the virtual but functional opposability between the phalanges and the double pincers."³²  Manuel Salesa and colleagues showed in 2006 that the red panda, which is not a bear, has a false thumb too, and that its Miocene relative Simocyon batalleri already had one, "producing one of the most dramatic cases of convergence among vertebrates."³³  Xiaoming Wang and colleagues reported in 2022 the oldest giant-panda false thumb, in Ailurarctos from the late Miocene, and argued that "since the late Miocene, the 'thumb' has not enlarged further because it must be balanced with the constraints of weight bearing while walking in a plantigrade posture."³⁴

Verdict: the panda's thumb is not a vestige on any definition; it is a bad-design argument that belongs to the parent article's tradeoff test.  A solution that arose at least twice is less of an accident than Gould implies, and a thumb held at its size by the demands of walking is a priced tradeoff, not a botch.

The child article The Panda's Thumb runs the tradeoff test in full: Gould's "clumsy" fails against the feeding studies and the weight-bearing constraint, but a wrist bone rebuilt into a thumb, with a useless twin in the foot, still counts as evidence of history.

Whale and snake pelvic bones

Coyne calls whales "treasure troves of vestigial organs" because their pelvic and leg bones "are not connected to the rest of the bones, but are simply imbedded in tissue."¹  The best recent study is by James Dines and colleagues in Evolution in 2014.  They describe cetaceans as having "lost external hind limbs and evolved a highly reduced pelvis which seems to serve no other function except to anchor muscles that maneuver the penis," note that "all 92 extant cetacean species except two" retain pelvic bones, and show from 130 specimens across 29 species that relative pelvic size rises with relative testis mass while rib size does not: "far from being mere relics of a terrestrial past, cetacean pelvic bones are targets of sexual selection."¹³

Creationist sites cite Dines as a function found.  It is one, in a paper whose first description of the bones is "highly reduced" and whose point is that selection kept a detached hip because it had a second use: Darwin's "retained for one alone of its former functions" with data.  Snakes run the same way.  A 2023 report on Boa constrictor summarises the literature: basal snakes "have pelvic girdles and vestigial hind limbs that are seen externally as pelvic spurs"; the spurs "have musculature, innervation, and vascularisation and can move autonomously"; and they "are known to be used by males during combat, courtship, and mating."³⁵  Verdict: both pass steps 1 and 2 without dispute, and the function findings do not touch them.  Step 3 for whales is the fossil series, treated in this site's whale evolution article; for snakes it is the relation to limbed lizards.  These are the strongest vestige cases in the literature, because the fuller homologue sits outside any kind boundary a creationist model has drawn.

The whale pelvic bones article takes this structure in depth: the muscles that attach to the bones, Dines 2014 in full, the embryonic limb bud that forms and dies, the five whales described with hind limbs or hind fins, and the creationist replies from Wieland 1998 to Surtees 2024.

The Snake Pelvic Spurs article goes deeper on this case and finds that, unlike the whale, the creationist literature concedes the spurs are reduced legs and disputes only whether the legged ancestor was a lizard or a created snake.

Horse splint bones and dewclaws

Septimus Sisson's 1914 veterinary anatomy is plain: three metacarpal bones are present in the horse, "only one, the third or large metacarpal bone, is fully developed and carries a digit; the other two, the second and fourth, are much reduced, and are commonly called the small metacarpal or 'splint' bones."  The medial one "bears two facets above which support the second and third carpal bones" and the lateral one articulates with the fourth carpal, so they carry load, and their lower ends are usually "a mere point" but in large draft horses "may present a prolongation which is regarded as the vestige of the digital skeleton."³⁶  Coyne calls splint bones "true vestigial features" and rests the ancestral claim on the fossil record of toe loss and on rare extra-toed horses.¹  Verdict: reduced second and fourth digits that still bear the carpus; step 3 is the equid fossil series, which the planned child on other animals must examine rather than assume.

Dewclaws are two cases, not one.  The front dewclaw of dogs and cats is the thumb, and Phil Senter and John Moch's 2015 survey of vestigial structures in mammal skeletons rejects it on all three of their criteria:  "it is not reduced enough to satisfy the first criterion for vestigiality," it "retains a claw and therefore does not satisfy the second criterion," and it "retains the typical function of a finger with a sharp, curved ungual and claw: prehension."³⁷  The reduction is in the hind foot.  Sisson notes for the dog's hind foot that "the first digit is often absent.  When present, its development varies and it contains one or two phalanges."³⁶  Senter and Moch count the canid first toe as vestigial:  "Its metatarsal is tiny, and its toe has only one phalanx, which is reduced to a small spike or nubbin."³⁷  The full hind dewclaw that some breeds carry is not that remnant kept.  Kyoungtag Park and colleagues traced it to mutations in a limb enhancer and call it "a developmental trait that restores the first digit lost during canine evolution."³⁸  Verdict: the front dewclaw is not a vestige, the tiny hind first metatarsal is one, and the variable hind dewclaw is a regained digit, so its variability is not evidence for Darwin's criterion.

The child article Dewclaws and Carpal Pads separates the front dewclaw, a working thumb that fails the reduction step, from the hind first metatarsal, a true vestige traced through fossil canids, and shows that a dog's hind dewclaw is a toe regained by mutation, not a remnant kept.

The child article Horse Splint Bones examines the splints in depth: the carpal load that makes the medial one non-disposable, the 2020 finding that horse embryos form five digit condensations which fuse into the two splints, and the extra-toed horses that regrow a second digit with its own carpal bone.

Flightless birds are the clearest case of all: the Flightless Bird Wings article finds that ostrich, kiwi, cormorant and rail wings pass all three steps, that creationist authors concede descent from fliers, and that the case proves repeated loss of flight within families and nothing beyond them.

Part 5: The shrinking list, and what it proves

Scadding's sentence is the one every creationist summary quotes: "Wiedersheim could list about one hundred in humans; recent authors usually list four or five."²  The trend is real and the reason is simple.  Wiedersheim put the pineal body, the pituitary and the thymus in the vestigial column, and they turned out to run the endocrine and immune systems.  "Useless" was a claim about the physiology of 1893, and physiology moved.

What the shrinking list does not show is that reduced homologues disappeared with it.  Naylor's reply lists more than twenty vestiges, from whale-embryo teeth to caecilian eyes, without claiming that any is useless,⁶ and the modern argument was never a count.  The creationist side won the argument being made in 1925 and has spent the decades since answering it again, while the other side changed the argument and has not always admitted how much the change conceded.

Objections and responses

"A function has been found, so the organ is not vestigial."  This answers the 1893 claim, not the 1982 one.  Naylor and Coyne define a vestige by reduction relative to a homologue, and the coccyx is their standard example of a functional vestige.

"The reduced-function definition makes every organ vestigial."  Bergman's objection has force against loose use of the word, and the remedy is the three-part test.  A jaw smaller than an australopith's is a reduced homologue only if the tree is accepted, which is the point: the definition depends on common ancestry, and that dependence is what it claims.

"Uselessness cannot be proved, so the argument is unscientific."  Right about uselessness and irrelevant to reduction.  Naylor's answer stands: "no function" is a falsifiable hypothesis, and the modern argument does not need it.

"Loss proves devolution, not evolution."  Correct as far as it goes.  The vestige argument never claimed that reduction shows gain.  It claims that the pattern of reductions matches a tree, and the creationist model accepts that pattern within kinds.  The disagreement is only about where the fuller homologue sits, and that has to be argued from fossils, embryos and genomes, as the whale evolution and shared genetic errors articles do.  The cleanest test of that reply is the blind Mexican cave fish, where the loss is within one interfertile species, the mechanism has been watched in the embryo, and the only thing separating the models is the 20,000 to 190,000-year age of the caves.

"Vestiges show bad design."  Not on the modern definition.  A reduced part that does its remaining job is not a defect.  The parent article gives the tests a bad-design claim must pass; the recurrent laryngeal nerve and inverted retina articles apply them.

Skeptic: "You accept every step of the anatomy and then refuse the conclusion."  Yes, and the other side does the same in reverse.  The vestige cases are not independent evidence for the tree; they are its predictions, confirmed to the extent the tree is.  For whales and snakes that confirmation is strong; for the small human muscles it is weak because the comparative data have not been collected.

Skeptic: "If the appendix arose 32 times, that is worse for creation."  It is worse for Darwin's specific story, which is the story creationists were answering.  For a design view, a structure that appears wherever the cecum and colon are large is a recurrent solution, the kind of pattern the convergence literature already documents.  Neither side should keep the appendix on its list.

Creationist: "Dines showed whale pelvic bones have a function, so they are not vestigial."  Dines and colleagues describe the pelvis as "highly reduced" and detached from the spine, and their finding is that selection kept a reduced hip for a new job.  That is a functional vestige, not a refutation of one.

Issues and unanswered questions

• No experimental study of the plica semilunaris's function was found. The tear-drainage claim is repeated in anatomy texts without a measurement.
• The appendix's reseeding role has not been followed directly in patients. The clinical evidence is one study's higher recurrence of C. difficile colitis after appendectomy, with no effect on first onset.
• Menton's claim that plantaris attachments show "no phylogenetically coherent story" is not tested here on a full primate sample; the Palmaris Longus and Plantaris child article takes it up. For palmaris longus, Diogo and Wood's 2011 survey supplies that sample, and it scores the upper limb only.
• The creationist literature has no published treatment of the whale pelvis or the snake spurs as reduced homologues; every treatment found answers the function claim.
• The ancestry step for horse splint bones rests on the equid fossil series and on extra-toed atavisms, neither examined here.
• The auricular-muscle results are from one laboratory and 20 participants, and their "neural fossil" reading assumes the ancestry it is used to support.

Verdict

The argument from useless organs is finished.  Wiedersheim's 86 vestigial organs included the pituitary and the pineal body, Scadding was right in 1981 that the list had collapsed, and creationist writers have been right to say so since.  But the number they give is wrong, the redefinition they complain of was made in public in 1982, and the redefinition was a concession.  The argument now made is that vestiges are reduced homologues whose distribution fits a family tree.  That is an argument for common ancestry and not for bad design, it cannot be answered by finding a function, and it has to be met at the level of homology, embryology and fossils.  On that level the whale and snake pelves and the horse's splint bones are strong cases, the coccyx and ear muscles are real reductions whose ancestral reading depends on the tree, and the appendix, goosebumps, palmaris longus and the panda's thumb do not belong on the list.  Creationists who reply that loss within a kind is expected have conceded the mechanism and moved the question to where the kind boundary lies, which is the question this site's ancestry articles exist to answer.

Sources