The Evolution of Complex Organs

ORIGINAL SCIENTIFIC ARTICLE

T Ryan Gregory

Published online 10 October 2008 Springer Science + Business Media LLC 2008

Abstract The origin of complex biological structures haslong been a subject of interest and debate Two centuries agonatural explanations for their occurrence were consideredinconceivable However 150 years of scientific investigationhave yielded a conceptual framework abundant data and arange of analytical tools capable of addressing this questionThis article reviews the various direct and indirect evolution-ary processes that contribute to the origins of complex organsThe evolution of eyes is used as a case study to illustrate theseconcepts and several of the most common misconceptionsabout complex organ evolution are discussed

Keywords Adaptation Constraints Convergence

Co-option Crystallin Duplication Exaptation Eye

Functional shift Historical contingency Lens

Natural selection Opsin Photoreceptor

Introduction

As a career science would hold very little appeal if all itentailed were the confirmation of existing knowledge or thememorization of long lists of well-established facts Sciencethrives on what is not yet known the more vexing a problemthe more inspiring it is to investigate With millions ofspecies alive today (and orders of magnitude more thoughtto be extinct) not only describing but also explaining thediversity history and complexity of life is a challenge nearlywithout equal in all of science Nevertheless the diligent ac-cumulation of data punctuated by occasional empirical ortheoretical breakthroughs has over the past two centuries

yielded tremendous advances in the understanding of lifersquoscomplexities and the historical origins thereof

There was a time when natural processes capable ofproducing complex biological features were deemed incon-ceivable leading to the conclusion that these like humanartifacts must be the products of intelligent agency (egPaley 1802) Beginning with Darwinrsquos (1859) description ofnatural selection and expanding considerably upon it in the150 years since the science of evolutionary biology has as-sembled a theoretical framework capable of explaining howcomplex features can arise naturally over time Still whilefew nonspecialists have trouble acknowledging small-scaleevolutionary processes such as the evolution of antibioticresistance within populations of bacteria they often remainuncertain as to how similar mechanisms could account forcomplex structures such as eyes or wings (Ayala 2007 Scottand Matzke 2007)

This article provides a general overview of the variousprocesses that play a role in the evolution of complex bio-logical systems The classic exemplars of organ complexityeyes are then used as a case study to illustrate these generalmechanisms Although it is not possible to deliver a com-prehensive discussion of eye evolution within the confines ofthis paper an extensive (but by no means exhaustive) ref-erence list is provided in order to facilitate further study of thesubject as well as to highlight the rich scientific literature thatexists on this topic but which may be largely unknown outsideprofessional biological circles Finally some common mis-conceptions regarding the evolution of complex features arediscussed

How Complex Organs Evolve

The concept of ldquocomplexityrdquo is anything but simple In factmany technical definitions are in use in mathematics

Evo Edu Outreach (2008) 1358ndash389DOI 101007s12052-008-0076-1

T R Gregory ()Department of Integrative Biology University of GuelphGuelph Ontario N1G 2W1 Canadae-mail rgregoryuoguelphca

computer science and other disciplines Means of quantifyingcomplexity have been developed in evolutionary biology aswell as have methods for assessing trends involving changesin complexity through deep time (for review see Gregory2008a) This paper is not about complexity per se butabout the evolution of complex organs biological structureswith several intricately interacting parts that function in asophisticated manner the eye being the primary example dis-cussed Complexity as defined in this intuitive sense is notrestricted to organsmdashit also applies to biochemical subcellu-lar behavioral and other biological systems On the otherhand not all functional systems are complex nor is com-plexity inherently advantageous or inevitable as efficiency canreadily trump a convoluted arrangement

Fundamentally evolutionary explanations for the originof biological features complex or otherwise are based on theassumption of continuity That is to say there is an unbrokenchain of ancestry and descent linking modern organisms withearlier species that lived long ago In order to account for theexistence of complex organs as they are observed today it isnecessary to provide an account of how these could havearisen without breaks or inexplicable leaps from lesscomplex antecedents The following sections outline variousprocesses that taken together are considered by evolution-ary biologists to meet this requirement

Direct Adaptation by Natural Selection

Darwin famously noted in 1859 (p189) that ldquoif it could bedemonstrated that any complex organ existed which couldnot possibly have been formed by numerous successive slightmodifications my theory would absolutely break downrdquo Itshould be understood that the ldquotheoryrdquo in question is not thenotion that species are related by descent rather it is theproposal that the mechanism responsible for evolution isnatural selection acting gradually on minor heritable differ-ences (see Gregory 2008b) There is absolutely no scientificdisagreement as to whether natural selection occurs as it canbe observed in both experimental and natural populations thequestion is whether this process alone can result in theemergence of complex organs such as eyes

The process of gradual stepwise adaptive change empha-sized (though not exclusively so) by Darwin has been calledldquodirect evolutionrdquo and can be further subdivided into twomajor types (Thornhill and Ussery 2000)

1 Serial direct evolution The simplest form of adaptiveevolution which proceeds step by step from A1 rarr A2 rarrA3 rarr A4 and involves gradual change along a singleaxis (ie each step serves the same function but ef-fectiveness increases from A1 to A4) A second series ofchanges may take place in addition to the first but inthis scenario it would be only after the first series ofchanges ended

2 Parallel direct evolution A slightly more complicatedform of direct adaptation in which changes occur at thesame time in more than one component as in A1B1 rarrA2B2 rarr A3B3 rarr A4B4 In this case no single com-ponent becomes greatly modified before the others do

Under serial direct evolution each change that occurs issmall involves only one component of a particular systemand in principle is reversible As a result serial directevolution does not produce organs with indivisibly integratedcomponents Parallel direct evolution on the other hand canproduce a moderate interdependence of parts because thesechange in concert with one another Neither serial nor paralleldirect evolution necessarily leads to an increase in complexityand in terms of highly complex organs the most importantcontribution made by direct adaptive evolution probablyrelates to refinements of particular functions through themodification of a few components Thus direct evolution byitself is not sufficient to properly account for the evolution ofhighly complex and integrated organs and as such additionalprocesses are necessary

Indirect Evolution

Direct adaptive evolution by natural selection has since it wasfirst proposed been subject to criticism by those who disagreethat it is sufficient to explain the origin of complex biologicalfeatures In Darwinrsquos own time opponents began listingfeatures of organisms for which incipient or intermediatestages seem unlikely to have been functional and which there-fore could not have been be shaped incrementally by naturalselection (eg Mivart 1871) Darwin (1872) responded bypointing to examples of organs of ldquointermediaterdquo complexitythat did or indeed still do exist in other species1

Nevertheless there are legitimate reasons to expect directgradual evolution along a single axis to be incapable of pro-ducing complex adaptations composed of tightly interactingparts As a result evolutionary biologists (including Darwin)have long pointed out the importance of indirect routes bywhich complex organs and systems can evolve

A recent analysis of the evolution of two hormone-receptorpairs provides an illustration of the basic concept of indirectevolution (Adami 2006 Bridgham et al 2006) The closematch between a hormone and the receptor to which it bindshas been considered analogous to that between a ldquolock andkeyrdquo both of which are required for the system to function

1 Interestingly Mivart (1871) cited among his examples of featuresthat he presumed would be very unlikely to evolve gradually theappearance of two eyes on one side of the head among flounders andthe feeding structures of baleen whalesmdashboth of which have beendiscussed recently in light of evidence indicating that intermediateforms did indeed occur (Demeacutereacute et al 2008 Friedman 2008 Janvier2008 Zimmer 2008)

Evo Edu Outreach (2008) 1358ndash389 359359

The challenge as with more complex biological features isto explain how such a system evolved through interme-diates to its current integrated state given that it is not rea-sonable to hypothesize that the components arose togetherinstantaneously by mutation

In vertebrates the stress hormone cortisol activates theglucocorticoid receptor which is involved in regulatingmetabolism and immunity whereas a related receptor themineralocorticoid receptor is activated by aldosterone toregulate electrolyte homeostasis (in abstract terms A + BrarrXwhereas C + D rarr Y) This specificity is important as acti-vation of the wrong receptor would be very detrimental to theorganism (it would be a problem if A + DrarrY Adami 2006)Having the two receptors activated separately is alsobeneficial as it allows metabolism to be regulated indepen-dently of electrolytes for example (Bridgham et al 2006)

Phylogenetic analyses suggest that the two receptors are de-rived from an ancestral gene that duplicated about 450million years ago (Mya) Aldosterone by contrast is foundonly in tetrapods and therefore evolved long after the originof the two receptors (Fig 1A) The question is how couldthese receptors become specific for different hormones whenone of the hormones did not yet exist

In order to address this question Bridgham et al (2006)used phylogenetic approaches to reconstruct the ancestralcorticoid receptor and found that it would have been sen-sitive to cortisol aldosterone (had it existed) and anotherhormone known as 11-deoxycorticosterone The differencebetween this early receptor and the modern glucocorticoidreceptor which does not bind aldosterone lies in amino acidchanges caused by twomutationsmdasheither one of these changesalone makes the receptor insensitive to cortisol but both

Fig 1 Indirect evolution of two hormone-receptor pairs A Stepsinvolved in the origin and evolution of the glucocorticoid receptor(GR) which is sensitive to cortisol and the mineralocorticoid receptor(MR) which is sensitive to aldosterone The two receptors are derivedfrom a gene for a single ancestral receptor (AncCR) which duplicatedabout 450 Mya Later two mutations occurred in the GR receptor thatmade it insensitive to aldosterone Aldosterone (Aldo) did not evolveuntil much later (represented by a dotted circle before it arises) but wasable to bind to the MR which had retained a form close to the ancestralreceptor because a third hormone similar to aldosterone formerly

activated it From Bridgham et al (2006) reproduced by permission ofthe American Association for the Advancement of Science B The twomutations required to make the GR insensitive to aldosterone (L111Q andS106P) both make the hormone insensitive to cortisol if they occur bythemselves and it is unlikely that both arose simultaneously Thisapparent problem can be explained by the fact that a third hormone 11-deoxycorticosterone (DOC) can still bind therefore the receptor can stillbe functional (in a different manner) if the S106P mutation occurs firstFrom Adami (2006) reproduced by permission of the AmericanAssociation for the Advancement of Science

360 Evo Edu Outreach (2008) 1358ndash389

together make it sensitive to cortisol and insensitive to al-dosterone It is unlikely that both mutations would occursimultaneously but Bridgham et al (2006) found that one ofthe mutations retained sensitivity to 11-deoxycorticosteronemeaning that this intermediate step would still be functionalthough in a different way and could still have been favoredbefore the second mutation occurred (Fig 1B) Meanwhileselection acting to maintain specificity for a different hor-mone that was structurally similar to aldosterone meant thatonce aldosterone arose it could bind to the mineralocorticoidreceptor which remained similar to the ancestral form (aprocess Bridgham et al 2006 call ldquomolecular exploitationrdquoFig 1A)

In short even though they are now indivisible the twohormone-receptor pairs could have evolved through a step-wise series favored at each stage by natural selection but thesteps were not direct The process involved gene duplicationthe input of a third hormone and shifts in function Similarprocesses may operate generally in the evolution of complexproteins in a manner that is readily explained by modernevolutionary theory (see Lynch 2005 for a technical dis-cussion) These and other indirect evolutionary processes arealso involved in the evolution of complex organs and theircomponents and are discussed in more detail in the fol-lowing sections It is important to note that most complexorgan evolution probably involves processes from variousparts of the directndashindirect evolutionary continuum

Exaptation

The notion of functional shifts is an old one in evolutionarybiology having been considered ldquoan extremely importantmeans of transitionrdquo by Darwin (1872 p147) The generalprocess has been known by many names ldquoco-optionrdquoldquoadoption from a different functionrdquo ldquorecruitmentrdquo (mostlyin reference to genes) or in an outmoded term a little toosuggestive of teleology ldquopreadaptationrdquo (eg Gould and Vrba1982 Arnold 1994 Thornhill and Ussery 2000 McLennan2008) However the basic concept became much morebroadly appreciated when it was granted a specific name witha clearer definition ldquoexaptationrdquo (Gould and Vrba 1982)

The term ldquoexaptationrdquo derives from ex + aptus meaningfit (aptus) by reason of (ex) existing form as contrasted withadaptation which derives from ad (towards) + aptus (fit)So whereas an adaptation is the product of natural selectionfavoring variants on the basis of and gradually improvingthe current function an exaptation is a feature that arose forsome other reason and subsequently acquired its currentfunction Designating a feature as an exaptation presumessome knowledge regarding ancestral function (or lack thereof)such that it can be shown to have differed from the currentrole Data bearing on this issue can be obtained using fossilsand phylogenetic inference (eg Arnold 1994) The process-

es of adaptation and exaptation are not entirely separatehowever because once a functional shift occurs naturalselection may modify the feature with regard to its new rolein a process of ldquosecondary adaptationrdquo (Gould and Vrba1982 Fig 2) Most complex organs are likely to represent amixture of primary adaptations exaptations and secondaryadaptations

The evolution of wings provides one of the classicexamples of exaptation and secondary adaptation As hasbeen pointed out many times a rudimentary version of awing would not be useful in flight because it would be unableto generate sufficient lift (eg Mivart 1871) Only when thewing reached a sufficient size and strength could it be use-ful for enabling powered flight meaning that natural se-lection could not favor variants within a population on thebasis of flight ability during the early stages of wingevolution So how then could wings have evolved to servetheir current function in flight The answer is that earlywings did not function in flight but served a differentfunction (primary adaptation) Bird feathers for exampleprobably originated for thermoregulation and rudimentarywings may have been useful in capturing prey or assistingwith running uphill or one or more other functions (egDial 2003) In bats early skin flaps probably would havebeen functional for gliding but not in powered flight (egBishop 2008) In insects it has been hypothesized that earlyldquowingsrdquo were used for skimming across the surface of water(Marden and Kramer 1994 Marden and Thomas 2003)Natural selection enhancing early forms of the structuremdashwhich initially may not have been considered a ldquowingrdquo atall had biologists examined it at the timemdashwould have atsome point brought it to a stage that could be useful in anew function (exaptation) resembling a rudimentary versionof flight (for example controlled descent from trees in birdsand bats or skimming with less contact with the water orpowered jumps in insects) Further modification for this newsemiflight function (secondary adaptation) would eventuallyrender the structure suitable for yet another functional shiftnamely to weak powered flight (exaptation again) withfurther modifications leading to new improvements specificto flight (secondary adaptation again)

Given that exaptations are defined largely by what they arenotmdashnamely the products of natural selection strictly for theircurrent functionmdashthere are several possible routes by whichan organ components of an organ or genes can becomeexaptations (Gould and Vrba 1982 Arnold 1994 Gould2002 McLennan 2008)

1 One organ (or gene) has an existing function but takes onor switches to a new function as a result of selectivepressures experienced after the organism moves into anew environment or adopts a new ecological lifestyleArnold (1994) distinguishes between ldquoaddition exapta-

tionsrdquo in which a second function is acquired in additionto the initial function and ldquotransfer exaptationsrdquo in whichthe shift to a new function involves the loss of the pre-vious function Example the middle ear bones of mam-mals are derived from former jaw bones (Shubin 2007)

2 One organ (or gene) has an existing function but at somestage modification of the feature for the initial functionmakes it amenable to modification in a new role and thisallows the organism to move into a new environment oradopt a new ecological lifestyle Example early tetrapodlimbs were modified from lobe-fins and probably func-tioned in pushing through aquatic vegetation at somepoint they became sufficiently modified to allowmovement on to land (Shubin et al 2006)

3 One organ (or gene) has two functions and is modified as itbecomes increasingly specialized for one of them Some-times the organ is specialized for one of the initialfunctions in one lineage and for the other initial function ina different lineage Example an early gas bladder thatserved functions in both respiration and buoyancy in anearly fish became specialized as the buoyancy-regulatingswim bladder in ray-finned fishes but evolved into anexclusively respiratory organ in lobe-finned fishes (andeventually lungs in tetrapods Darwin 1859 McLennan2008)

4 Two organs (or genes) perform the same function andthen one becomes more specialized for the originalfunction while the other takes on a different role This is

particularly significant when duplication generates mul-tiple copies that subsequently diverge (see below) Ex-ample some of the repeated limbs in lobsters arespecialized for walking some for swimming and othersfor feeding

5 A feature that had become vestigial2 in terms of itsoriginal function takes on a new function in its reducedstate Example the vestigial hind limbs of boid snakesare now used in mating (Hall 2003)

6 A feature that formerly had no function and was presentfor non-adaptive reasons (a ldquospandrelrdquo Gould andLewontin 1979 Gould 1997 2002) takes on a functionand may become specialized for that function This toocan occur at both the genetic level and the organ levelArnold (1994) considers these ldquofirst-use exaptationsrdquobecause the first function they fulfill is the exaptive oneExamples the sutures in infant mammal skulls are usefulin assisting live birth but were already present in non-mammalian ancestors where they were simply byprod-ucts of skull development (Darwin 1859) someformerly parasitic transposable elements in the genomewhich had no function at the organism level have been co-

2 ldquoVestigialrdquo does not necessarily mean non-functional it meansreduced in form and function in a particular species relative to othersin which the organ still performs the original function Thus findingsome function for a reduced organ which is often different from thefunction of the fully formed organ in other species does not affect itsstatus as being vestigial

Fig 2 A simple example of exaptation and secondary adaptation A Theoriginal and still primary adaptive function of coins is as currency B Acoin co-opted into a new exaptive role as an instant lottery ticket scraperCoins would always have been capable of scraping tickets but thisfunction did not become apparent until an environment arose in whichinstant lottery tickets were abundant Though functional as scrapers coins

are somewhat difficult to hold and may not reliably be on hand whenneeded C A secondary adaptation that enhances the novel function of acoin as a ticket scraper by incorporating it into a keychain that is easier togrip (US Patent 6009590 ldquoLottery ticket scraper incorporating coinrdquo byKM Stanford 2000) In this case a second preexisting structure (keyring) was co-opted into a function as a carrier for a lottery ticket scraper

opted into a variety of other roles such as in the vertebrateadaptive immune system (eg Zhou et al 2004)

The important point regarding exaptations then is that thecurrent function of a feature may not reflect the reasons for itsorigin Rather the feature may only have come to occupy itscurrent role comparatively recently

Duplication (or Furcation)

It has long been recognized that natural selection thoughcapable of producing directional change can also be a highlyconservative force If a biological feature currently serves afunction vital to survival then it is very likely that any de-viations from its current state will prove detrimental That is tosay individuals with a different form of the feature will leavefewer offspring than those with the original form such thatthere will not be change in the population from one generationto another with regard to this feature The most widely re-cognized escape from this constraint is through duplication atopic that has long been discussed in some detail with ref-erence to genes (eg Ohno 1970 Taylor and Raes 2004)Ohno (1970) in particular considered duplication anddivergence of genes a critical requirement for majorevolutionary diversification ldquoNatural selection merely mod-ified while redundancy createdrdquo he wrote

In general four different outcomes are possible following agene duplication event One the duplicate copy (or the ori-ginal) may simply be lost or rendered nonfunctional bymutation (to become a ldquopseudogenerdquo) Two multiple copiesmay prove to be beneficial such that their repetition is main-tained by selection thereafter (ldquoisofunctionalizationrdquo Oakleyet al 2006) As an interesting example the number of func-tional copies of the salivary amylase gene which is involvedin breaking down starch is higher in human populationswhere starchy foods are common in the diet (Novembreet al 2007 Perry et al 2007) Three mutations in differentparts of the two gene copies may mean that both copies mustbe retained in order to fulfill the original function or onecopy may come to serve the original function in one tissue orat one time during development while the second copy isactive in different places or at different times again such thatboth copies are needed to serve the role of the original gene(ldquosubfunctionalizationrdquo) Four as per Ohnorsquos (1970) dis-cussion one copy may take on a new function while theother fulfills the previous function (ldquoneofunctionalizationrdquo)

Repeated structures are common at the organism level aswellmdashbody segments teeth and flower petals are among themany examples In this case duplicated structuresmay remainin repeated series (eg identical body segments of myriapodsor annelids) one of the two repeats may retain the originalfunction while the other takes on a new function (eg thereduced hind wings of true flies called ldquohalteresrdquo which now

function as a kind of gyroscope) or the repeats may becomespecialized for different functions that are either new or arefunctions that were formerly carried out by a single structure(eg specialized appendages in crustaceans) Darwin (1859p437ndash438) himself recognized the importance of duplica-tion at the organism level when he wrote

We have formerly seen that parts many times repeatedare eminently liable to vary in number and structureconsequently it is quite probable that natural selectionduring a long-continued course of modification shouldhave seized on a certain number of the primordially si-milar elements many times repeated and have adaptedthem to the most diverse purposes

Although there clearly are similarities between the geneand organism levels in this regard it is important to note thatduplication of structures (eg organs or components thereof)may not necessarily be the result of the duplications of genesOrgan-level multiplication can also occur with regulatorymutations that cause the feature to appear in different placesat different times or in repeated series To clarify this issueOakley et al (2007) recently coined the term ldquofurcationrdquo(meaning ldquoformation of a fork or division into branchesrdquo) tocover the multiplication of existing structures more general-ly The important point for the present discussion is thatduplications whatever their cause and with or withoutdivergence can be an important mechanism for increasingcomplexity at both the genomic and organismal levels

Gene Sharing

Under the process of ldquoaddition exaptationrdquo a feature that isfunctional in one capacity assumes a second function withoutlosing its original function (Arnold 1994) At the molecularlevel it is becoming increasingly recognized that the sameprotein can carry out more than one function though it cansometimes be difficult to determine which if either was thesole original function In this sense a descriptor other thanldquoexaptationrdquo or ldquoco-optionrdquo is used in reference to the exis-tence of multifunctional proteins ldquogene sharingrdquo (Piatigorsky2007 2008)

As Piatigorsky (2007 pp 4ndash5) defines it

The term ldquogene sharingrdquomeans that one gene produces apolypeptide [protein] that has more than one molecularfunction Two or more entirely different functions of apolypeptide share the identical genehellipThe gene-sharingconcept postulates that protein function is determinednot only by primary amino acid sequence which remainsthe same in the multiple functions that are performed bythe protein but also by the microenvironment within thecell and by the expression of its gene Awareness of genesharing cautions against assuming that a protein will be

used in the same way wherever or whenever it is presentor that it has always done what it is doing at any givenmoment The functions of genes and proteins are contextdependent

There are several ways that a single protein can serve verydifferent functions such as by being expressed in differenttissues or at different times during development (ie due tochanges in regulatory genes) by undergoing changes in theamino acid sequence that enable a second function but do notcompromise the first by combining with another copy of thesame protein to form a ldquohomodimerrdquo with a different functionby combining with other proteins to form ldquoheterodimersrdquo or bybeing subject to different patterns of folding or other chemicalmodifications (True and Carroll 2002 Piatigorsky 2007)

The process of gene sharing can be important in theevolution of complex organs because it means that functionscan be enhanced or acquired without any change in the protein-coding gene itself if there is a change in the context in which itoccursmdashsay the emergence of a new type of tissue in which itmay be expressed Conversely an existing gene being expressedin a new place in the body may itself lead to the evolution of anew tissue This greatly facilitates the specialization of an organfor a new function because it does not compromise previousfunctions for the gene does not require gene duplication anddivergence (though this remains an important process in its ownright) and may involve little more than a quantitative change inthe amount or localization of the genersquos protein product

Bricolage (Tinkering) and Collage

In light of the processes described above it may seem anobvious point that the evolution of complex organs does notinvolve redesign from scratch at each stage whether by directadaptation or shifts in function the process builds upon andmodifies what is already present This was recognized byearly evolutionists including Darwin (see Jacob 1977 1982Laubichler 2007) but has often been overlooked when authorscharacterize natural selection as an optimization process Theclear exposition by Jacob (1977 1982) was therefore animportant reminder of this point from which it is worthquoting at length (Jacob 1982 pp 33 34)

The action of natural selection has often been compared tothat of an engineer This comparison however does notseem suitable First in contrast to what occurs duringevolution the engineer works according to a preconceivedplan Second an engineer who prepares a new structuredoes not necessarily work from older ones The electricbulb does not derive from the candlehellipTo producesomething new the engineer has at his disposal originalblueprints drawn for that particular occasion materialsand machines specially prepared for that task Finally theobjects thus produced de novo by the engineer at least by

the good engineer reach the level of perfection madepossible by the technology of the time

In contrast to the engineer evolution does not produceinnovations from scratch It works on what alreadyexists either transforming a system to give it a newfunction or combining several systems to produce a morecomplex one Natural selection has no analogy with anyaspect of human behavior If one wanted to use acomparison however one would have to say that thisprocess resembles not engineering but tinkering brico-lage we say in French While the engineerrsquos work relieson his having the raw materials and the tools thatexactly fit his project the tinkerer manages with oddsand ends Often without knowing what he is going toproduce he uses whatever he finds around himhellip noneof the materials at the tinkererrsquos disposal has a preciseand definite function Each can be used in differentways What the tinkerer ultimately produces is oftenrelated to no special project It merely results from aseries of contingent events from all the opportunitieshe has had to enrich his stock with leftovers In contrastwith the engineerrsquos tools those of the tinkerer cannot bedefined by a project What can be said about any ofthese objects is just that ldquoit could be of some userdquo Forwhat That depends on the circumstances

Though they describe more a principle than a process theterms ldquotinkeringrdquo and ldquobricolagerdquo include and are now mostoften used as substitutes for ldquoco-option of a gene or otherfeature into a new functionrdquo or simply ldquomutation and naturalselection leading to an alteration of preexisting traitsrdquo (egBock and Goode 2007) As Jacob (2001) noted one must becautious to avoid a potentially confusing anthropomorphismin which an actual tinkerer or bricoleur is imagined whoinvents through trial and error

What is perhaps missing and for the purposes of thisdiscussion is useful to emphasize is a particular processmentioned by Jacob (1977 1982) that differs from bothdirect adaptation and exaptation in which existing compo-nents be they functional for something else or nonfunctionalinitially are brought together or rearranged to form a newmore complex combination with a novel function Ratherthan ldquobricolagerdquo the term ldquocollagerdquo may more effectivelyencapsulate this concept3 Because a new function emerges

3 The terms ldquopatchworkrdquo and ldquojury-riggingrdquo also have been used in thisregard but the first may imply a pre-determined design that is achievedusing available materials whereas the latter often refers to the make-shiftrepair of damaged structures Unfortunately many terms currently in usesuch as ldquoadaptationrdquo ldquoexaptationrdquo ldquobricolagerdquo and most recentlyldquocollagerdquo suffer from potential confusion because they refer to both aprocess and its products Further terminological refinement clearly isrequired

through the combination of existing components and es-pecially once further modified by natural selection for thisnew function a feature produced through ldquocollagerdquo becomesmuch more than the sum of its parts As noted by Jacob(1977 1982) the feature is not assembled with a predefinedoutcome in mind rather its function depends on circum-stances and on which components are available and happento become linked Two important points bear mentioningabout the process of indirect evolution through ldquocollagerdquo (1)the linking of components is not an ldquoall or nothingrdquo processin which two or more already complex structures suddenlyare joinedmdashindividual parts which themselves may vari-ously be simple or relatively complex and functional forsomething else or nonfunctional can be added in series witheach new addition leading to a different function for thecombined structure and (2) the newly combined structuremay carry out its new function rather poorly at first withsubsequent direct adaptation leading to improvement along

this novel axis for example by enhancing the integration ofthe newly combined components (Fig 3)

Scaffolding

Many organs having been built up in overall complexity bydirect adaptation exaptation and collage and furtherspecialized through secondary adaptation exhibit a level ofintegration to the point that their components are interde-pendent on one another In these cases the removal of one ormore components may render the organ nonfunctionalmdashatleast with regard to its current integrated function (after allexaptation can also occur following a loss of parts) Shifts infunction help to explain how such a system could beassembled through less complex intermediate steps butanother process known as ldquoscaffoldingrdquo is sometimesinvolved in the evolution of such functionally indivisibleorgans In this case a component of the organ that is present

Fig 3 A diagram showing the processes of exaptation (shifts infunction) collage (assembling existing elements into new functionalcombinations) scaffolding (loss of a component that was formerlyrequired for the assembly of a complex arrangements of parts) and directadaptation by natural selection (including secondary adaptation) in theevolution of a complex feature The complex organ (J) includes manyparts all of which must be present for the organ to carry out its currentfunction Although it can now carry out this function only when all of itscomponents are present an organ such as this can evolve throughintermediates all of which have some functionmdashthough not necessarilythe function of the final complex organ (J) At an early stage two simplestructures each already present and performing its own distinct function(A) come together into a combined structure (B) that is capable ofcarrying out a function that neither component could before At first thecombined structure (B) may perform the new function rather poorly butif it nonetheless confers some advantage over alternatives lacking thestructure then the components may be modified by natural selectionfavoring improvements in this new function (C) Later a third

component (D) which was also already present and serving its ownfunction is co-opted and becomes associated with the simple two-partorgan to form a three-part organ (E) capable of performing yet anothernew function (though again not the one currently filled by J) Onceagain the various components may become modified due to selectionfavoring improvements to this new function (F) Later additionalcomponents (G)mdashwhich in this case are themselves built of combinedperhaps duplicated componentsmdashbecome associated with the modifiedthree-part organ (F) to form a complex but still not irreducibly complexorgan (H) that takes on the function of the final complex organ albeitnot very effectively This complex organ is once again modified for itsnew function and most of the components become more closely integrated(I) However one component that has become structurally unnecessary islost (eg because it is costly to produce and mutations that lead to lowerinvestment in its production are advantageous) leaving behind anirreducibly complex organ (J) whose ability to carry out its currentfunction is contingent on the presence of all its component parts

through some early stages has the effect of supporting theassembly of other components and when lost in later stagesleaves behind a complex structure that by all appearancescould not have been assembled one piece at a time (Fig 3)

Draper (2002) has provided an abstract description that isreadily applied to biological systems (including organs orbiochemical pathways) In this case a complex system withtwo required parts (AB) that performs a function (F) evolvesthrough a complicated but essentially direct path involvingboth the addition and loss of parts

Originally Z performs F though perhaps not very well(this is possible because from the fact that AB cannotperform F without A or B it does not follow that Zcannot perform F by itself) Then A is added to Zbecause it improves the function though it is notnecessary B is also added for this reason One now hasa reducibly complex system composed of three parts ZA and B [ie the system could still function if thenumber of parts is reduced] Then Z drops out leavingonly A and B [perhaps only after A and B have becomemodified to work in a more integrated fashion in theirnew joint arrangement] And without Z both A and Bare required for the system to function

This can perhaps be illustrated even more simply with astraightfoward architectural analogy involving the construc-tion of a stone arch (eg Cairns-Smith 1985 Dawkins 1986Schneider 2000 Thornhill and Ussery 2000) As Dawkins(1986 p149) put it

An arch of stoneshellipis a stable structure capable ofstanding for many years even if there is no cement tobind it Building a complex structure by evolution is liketrying to build a mortarless arch if you are allowed totouch only one stone at a time Think about the tasknaiumlvely and it canrsquot be done The arch will stand once thelast stone is in place but the intermediate stages areunstable Itrsquos quite easy to build the arch however if youare allowed to subtract stones as well as add them Startby building a solid heap of stones then build the archresting on top of this solid foundation Then when thearch is all in position including the vital keystone at thetop carefully remove the supporting stones and with amodicum of luck the arch will remain standing

A second issue is that a stone does not become a keystoneuntil it is added to the rest of the assembled arch and servingthis role cannot be the reason it is maintained until that pointThis is analogous to a component of a complex organ that canbe incorporated only relatively late in the process andtherefore cannot be maintained for this function earlier in theprocess One manner in which such components may remainpresent is if they serve a different function and are preserved

more or less in their current form on that basis In this regarddual functionality may itself be a form of scaffolding that inretrospect will have played a role in facilitating theproduction of a complex feature

Viewing a complex structuremdashbe it an arch (or for thatmatter the Great Pyramids or Stonehenge) an organ or abiochemical pathwaymdashonly as it appears in the presentwith no consideration of the scaffolding that may have beeninvolved in its construction can lead to undue pessimismregarding the plausibility of its assembly by comparativelyunremarkable processes

Non-adaptation Constraints Trade-offsand Historical Contingency

In the sixth and final edition of The Origin of SpeciesDarwin (1872 p 421) expressed the following frustration

[As] it has been stated that I attribute the modification ofspecies exclusively to natural selection I may bepermitted to remark that in the first edition of thiswork and subsequently I placed in a most conspicuouspositionmdashnamely at the close of the Introductionmdashthefollowing words ldquoI am convinced that natural selectionhas been the main but not the exclusive means ofmodificationrdquo This has been of no avail Great is thepower of steady misrepresentation

Natural selection is not the only mechanism involved inevolution It is not even the only process that may account forthe origin of biologically complex systems (eg Lynch 2007ab) It may be the creative process responsible for generatingadaptive complexity but this does not mean that its influenceis without limits constraints of various sorts (eg geneticdevelopmental physical energetic historical) also contributeto the observed form of complex features For examplealthough one may conceive of a change that would improvethe function of an organ it may be that the necessary muta-tions (which occur by accident and not in response to need)simply never arose It could be that changes in an organ wouldbe too disruptive to existing developmental programs to beviable could not physically be accommodated within themorphology of the organism carrying the organ or wouldcompromise the function of other organs through trade-offsOr it could be that there simply was no selective pressurein the environment that would favor increased complexityin a particular organ Moreover some characteristics of com-plex systems are not adaptive at all but represent theinevitable byproduct of other evolutionary changes (Gouldand Lewontin 1979 Gould 2002) or mechanisms (Lynch2007a b) Therefore an important point to bear in mind aboutcomplex organs is this not everything about them should beviewed in the light of adaptation

Case Study The Evolution of Eyes

The eye has long held a special place in discussions regardingthe origin of complex organs Paley (1802) famouslycompared the intricacies of an eye to those of a finely craftedwatch and concluded that both were the work of an intelligentdesigner4 Darwin (1859) offered a different explanation forthe origin of biological complexity and again used the eye asa prominent example Thus in a passage that is often(partially) quoted Darwin (1859 pp 186 187) remarked

To suppose that the eye with all its inimitable con-trivances for adjusting the focus to different distances

for admitting different amounts of light and for thecorrection of spherical and chromatic aberration couldhave been formed by natural selection seems I freelyconfess absurd in the highest possible degree Yet reasontells me that if numerous gradations from a perfect andcomplex eye to one very imperfect and simple eachgrade being useful to its possessor can be shown to existif further the eye does vary ever so slightly and thevariations be inherited which is certainly the case and ifany variation or modification in the organ be ever usefulto an animal under changing conditions of life then thedifficulty of believing that a perfect and complex eyecould be formed by natural selection though insuperableby our imagination can hardly be considered real

A considerable amount of research has illuminated manydetails of eye evolution in different groups of animals sinceDarwin penned these words The comparatively recent rise of

Fig 4 Very simple but nonetheless functional light-sensing systemsA A single light-sensitive cell (ocellus) as found in the larva of thebox jellyfish Tripedalia cystophora In this case the light-sensitivecell is not connected to a nervous system of any kind but insteadincludes a cilium that can be stimulated to move the larva in responseto light The pigments (dark spots) within the cell are arranged in asimple cup meaning that some measure of the directionality of light isprovided to the cell From Nordstroumlm et al (2003) reprinted bypermission of The Royal Society B In stark contrast the adult of thesame box jellyfish species T cystophora has complex upper andlower eyes with retinas lenses and irises at the end of a sensory clubcalled a rhopalium From Nilsson et al (2005) reproduced bypermission of Nature Publishing Group C A simple eye spot found

in the larva of the trematode flatworm Multicotyle purvisi whichconsists of one pigment cell and one photoreceptor cell This organitself provides no information about the direction of a light source butthis can be achieved by comparing the input from two of these organsRedrawn by Land and Nilsson (2002) based on Rohde and Watson(1991) reproduced by permission of Oxford University Press D Aslightly more complex visual organ involving a single pigment cell butmultiple receptor cells found in the turbellarian flatworm Bdellocephalabrunnea In this species the pigment cell is cup-shaped such thatinformation about the direction of light can be obtained by comparinginput from the different receptors Redrawn by Land and Nilsson (2002)based on Kuchiiwa et al (1991) reproduced by permission of OxfordUniversity Press Note that these images are not drawn to the same scale

4 It bears noting that Darwin read Natural Theology (Paley 1802) as atheology student at the University of Cambridge between 1827 and1831 and noted in his autobiography that ldquoI did not at that time troublemyself about Paleyrsquos premises and taking these on trust I was charmedand convinced by the long line of argumentationrdquo (Darwin 1958 p59)

disciplines including molecular biology phylogenetics andevolutionary developmental biology (ldquoevondashdevordquo) in particu-lar has generated a great many insights regarding this subjectSo much in fact that any more than a cursory review of theavailable information must be considered well beyond thescope of this article (however references to papers containingthis information are provided whenever possible) Instead theeye is used as a case study to illustrate the various generalprinciples of complex organ evolution outlined above and inparticular to demonstrate that the multifaceted nature of thetopic requires that it be examined from a variety of perspectives

Eyes Definition and Diversity

According to some authors an eye is defined at minimum as aphotoreceptor shielded on one side by nearby pigment whichallows the detection of the direction of a source of light Thesimplest eyes then may consist of just one photoreceptor andone pigment cell or even a single cell that includes both photo-and shading pigments as found in some flatworms and algae(eg Arendt and Wittbrodt 2001 Oakley 2003) Someexamples of relatively simple ldquoeyesrdquo are shown in Fig 4Others argue for a more restrictive definition under which aneye is an organ that can produce an image however crudeand not simply detect light (eg Land and Nilsson 2002Piatigorsky 2008 Serb and Eernisse 2008) Even under thisstricter definition there are at least eight different types ofeyes5 prominent examples of which variously employ cupspinholes camera-type lenses arrays of lenses concavemirrors or telescope-like arrangements for image formation(Land and Fernald 1992 Land and Nilsson 2002 Serb andEernisse 2008) These are illustrated in Figs 5 and 6 To theextent that the eyes of each species are at least slightlydifferent from each other and given that many species havemore than one type of eyes there are probably millions ofdifferent kinds of eyes peering at the world around them atthis very moment

There is no doubt that access to visual information has beenimportant in a great many groups of organisms Eyes can befound in about one third of the worldrsquos animal phyla whileanother one third has light-sensing organs but not eyes

Roughly one third of the worldrsquos animal phyla have no lightsensitive organs but these tend to be groups exhibiting lowdiversity by contrast the phyla with eyes include more than95 of all animal species (Land and Nilsson 2002 Fernald2004a but see de Quieroz 1999) The extraordinary benefitsprovided by the ability to see are also shown by the fact thateyes appeared very early in animal evolution In fact most ofthe major types of eyes are recognizable in fossils from theCambrian some 530 Mya (Land and Nilsson 2002 Nilsson

5 For more detailed discussions of the form function and evolution ofdiverse animal eyes see Land (1988) Nilsson (1989) Land andFernald (1992) Arendt and Wittbrodt (2001) and Land and Nilsson(2002) or consult recent reviews on the eyes of specific groupsincluding annelids (Purschke et al 2006) crustaceans (Gaten 1998Elofsson 2006 Reimann and Richter 2007 Marshall et al 2007Cronin and Porter 2008) tardigrades (Greven 2007) insects (Land1997 Buschbeck and Friedrich 2008) velvet worms (Mayer 2006)millipedes (Muumlller et al 2007) jellyfishes (Nilsson et al 2005Kozmik et al 2008a b) mollusks (Serb and Eernisse 2008) trilobites(Clarkson et al 2006) horseshoe crabs (Battelle 2006) andvertebrates (Lamb et al 2007 2008)

Fig 5 Eight major types of complex eyes found in living animalsdivided into two major categories chambered eyes (top) andcompound eyes (bottom) (A) and (B) form images using shadows(C) to (F) use refraction and (G) and (H) use reflection The paths oflight rays entering the eyes are indicated by dark lines Thephotoreceptive structures are shown in shaded gray (A) A simple piteye as found in Nautilus as well as many flatworms and annelids (B)A basic compound eye in which each receptor is shielded from itsneighbor by a simple pigment tube as found in sea fans and a fewbivalve mollusks (C) A complex camera-type eye in which the lensdoes most of the focusing as found in fishes and cephalopodmollusks (D) A complex camera-type eye in which the cornea doesmost of the focusing as found in terrestrial vertebrates and spiders (E)An apposition compound eye found in diurnal insects and manycrustaceans (F) A refracting superposition compound eye as found ininvertebrates in dim environments such as krill and moths (G) Asingle-chambered eye in which an image is formed using a concavemirror as found in some scallops (H) A reflecting superposition eyesimilar to (F) but with lenses replaced by mirrors as found in lobstersand shrimps Modified from Land and Nilsson (2002) and Fernald(2006) reproduced by permission of Oxford University Press and theAmerican Association for the Advancement of Science

2004) Interestingly the phylum Chordata (of which humansare members) may have been among the last groups toevolve discernable eyes as these were not present in theknown Cambrian chordates (eg Pikaia) chordate eyes firstappear in the fossil record of conodonts from 30 millionyears later (Land and Nilsson 2002) On the other hand theextensive physical and molecular similarities between the eyesof lampreys and other vertebrates indicate that complexcamera-type eyes were already present in their last commonancestor 500 Mya (Lamb et al 2007)

There are several ways of categorizing complex image-forming eyes such as by the type of photoreceptor cells thearrangement of photoreceptors relative to pigment cells (ieinverted or everted) or by the mechanism of image formation(via shadows refraction or reflection eg Land and Nilsson2002 Fernald 2004a b) Perhaps the best-known distinctionis between chambered (or simple) and compound eyes (Nilsson1989 Land and Nilsson 2002 Fernald 2006) Although basedon a very narrow sampling of animal diversity it is clear thatmost references to the evolution of ldquotherdquo eye relate to thechambered camera-type lens eyes found in humans and othervertebrates as well as in cephalopod mollusks some annelidworms and various arthropods including spiders Certainlythese are the most effective at image formation and are themost familiar and they will form the basis of most of theremaining discussion However the evolution of compoundeyes is no less interesting than that of chambered eyesmdashandgiven the extraordinary diversity of groups exhibiting them

(most notably arthropods) this is an important question inbiology (Land 1997 Nilsson and Kelber 2007 Buschbeckand Friedrich 2008 Cronin and Porter 2008)

Direct Adaptive Evolution From Eyespot to Eyeball

The simplest hypothesis for how a complex feature arose isone involving direct adaptive evolution with incrementalimprovements in function favored at each stage by naturalselection Not surprisingly this has been the starting pointfor many discussions of eye evolution which is often depictedas a linear series of small changes each of which adds veryslightly to the organismrsquos ability to process visual information(eg Salvini-Plawen and Mayr 1977 Miller 1994 Nilssonand Pelger 1994 Osorio 1994 Dawkins 1996 Bahar 2002Kutschera and Niklas 2004)

The major question under these linear scenarios is whetherindeed each step along the path not only is functional but infact increases some aspect of visual ability In order to test thisand moreover to investigate how much time such a processmight require Nilsson and Pelger (1994) created a theoreticalmodel that began with nothing more than ldquoa flat patch oflight-sensitive cells sandwiched between a transparent pro-tective layer and a layer of dark pigmentrdquo In the model theyused incremental changes of 1 in one parameter at a time(length width or protein density) that improved visualacuity as calculated based on established optical principlesTheir model proceeded through a series of changes including

Fig 6 The distribution of eye types among major taxa of animalsSingle-chambered eyes are outlined with rectangles and compoundeyes are outlined with ovals See Fig 5 for more information about the

different types From Treisman (2004) reproduced by permission ofThe Company of Biologists and Oxford University Press

an inward folding of the flat patch to form a pit and then acup and when resolution could no longer be improved alongthis trajectory a very simple lens was added (as they noteldquoeven the weakest lens is better than no lens at allrdquo) whichthen changed incrementally to become spherical and then todevelop a gradient of refractive indices (Fig 7) with the

visual organ finally becoming similar in basic form to theeye of an aquatic animal like a fish or octopus (Fig 8)

Overall Nilsson and Pelger (1994) found that smallincremental changes that improve vision by a quantifiabledegree could connect both ends of the continuum from asimple patch of cells to a complex camera-type eye More-

Fig 7 The impacts of refinements in lens organization (a) Imageformed by a spherical glass bead with a single refractive index showingthe blurring resulting from spherical aberration (b) The same imagethrough the lens from a fish eye which has a graded refractive indexresulting from a higher crystallin protein concentration in the centre thanat the edges Although the first ldquolensrdquo may have functioned relatively

poorly it is only a matter of incremental adaptive changes to improve itsfunctioning to the level seen in modern camera-type lens eyes Forimages of eye lenses focusing light from a laser see Piatigorsky (20072008) From Sweeney et al (2007) reproduced by permission of TheRoyal Society and AM Sweeney

Fig 8 The results of a theoretical model (not as it is sometimesdescribed a ldquosimulationrdquo) developed byNilsson and Pelger (1994) to testthe time required for a complex camera-type eye to evolve through aseries of gradual steps from a simple patch of light-sensitive tissueconsisting of an outer protective layer a layer of receptor cells and abottom layer of pigment cells The number of generations passingbetween each step is indicated based on a change of only 0005 insome parameter (length width or protein density) per generation withchanges resulting in an improved calculated image formation retained

each time Although this model assumes a strictly gradualistic linearmodel that is not necessarily the route that camera-type eye evolutionactually took (Fig 10 Table 1) it does show two important things (1)that even very minor changes can improve image formation graduallyand (2) that the time taken for this process to occur less than 400000generations even under rather conservative assumptions is remarkablyfast in an evolutionary sense From Land and Nilsson (2002) based onNilsson and Pelger (1994) reproduced by permission of OxfordUniversity Press and The Royal Society

over only 1829 steps of 1 improvements were needed tocomplete this transition Even assuming a change of only0005 per generation the model suggests that the entiresequence could be completed in about 360000 generations(Fig 8) Given that many fishes and aquatic invertebrateshave at least one generation per year this would mean thatthe entire sequence in the model could be completed toinvoke an appropriate clicheacute in an evolutionary blink of aneye and well within the tens of millions of years availableduring the Cambrian

However while the intermediate stages used by Nilssonand Pelger (1994) are functional in an abstract model themost important question is whether organisms possessingthem could actually survive in nature It is in this regardthat the approach of comparing living species becomesuseful even though they are not ancestors and descendantsof one another In this way it can be shown that each of thehypothetical intermediates depicted in Fig 8 does still existand clearly is functional for the organism in which it occursAccording to Land and Nilsson (2002 p4) ldquonearly everyimaginable intermediate exists between the acute vision ofan eagle and the simple light sensitivity of an earthwormrdquo

In fact it has been recognized for over a century that such adiversity of eye types still exists (Darwin 1859 Conn1900) as seen in Fig 9

It is important not to take linear models and comparisonsof living species too far Although they demonstrate howdirect adaptive evolution could play a major role in theevolution of visual organs their emphasis on lineartransformation tends to obscure the complexity of theactual process In the simplest terms it is clear that a singlelinear path of eye evolution is too simplistic even forcomparisons among mollusks which appear to havefollowed several distinct routes leading to divergent eyetypes (Fig 10) In vertebrates there is ample evidence forthe gradual evolution of eyes but this does not follow thelinear model given in Fig 8 Indeed the current hypothesisof vertebrate eye evolution involves at least one functionalshift even at the organ level from an early photoreceptiveorgan performing nonvisual circadian (dayndashnight cycle)functions to a primitive eye capable of sight (Lamb et al2007 2008 Table 1) As will be shown in the followingsections the influence of indirect evolutionary processes iseven more pronounced at the level of the components of

Fig 9 Varying levels of complexity in the visual organs of livingspecies as illustrated by H W Conn in 1900 A a simple flat patch ofpigmented cells connected to nerve fibers (see also Fig 4C) B aslightly more complex pigment cup as found in the limpet Patellawhich does not form an image but provides information about thedirection of incoming light C a pinhole camera-type eye filled withwater as found in Nautilus D a camera-type eye with a large lensfilling the cavity E a camera-type eye with a basic lens and cornea asfound in the marine snail Murex F a complex camera-type eye with acornea lens iris and retina as in a cuttlefish This shows visual organs

as they occur in contemporary speciesmdashnone is ancestral to anotherand this does not necessarily reflect a historical series of steps in theevolution of complex eyes (see Fig 10) It does however indicatethat eyes of varying complexity such as would have been found inintermediate steps during complex eye evolution could have beenmdashand still aremdashfunctional for organisms living in different conditionsNote that these images are not drawn to the same scale Modified fromConn (1900) For another classic example of this kind of diagram seeSalvini-Plawen and Mayr (1977)

camera-type eyes specifically involved in image formationeither in receiving light (photopigments and photorecep-tors) or focusing it (lenses and corneas Table 2)

Photopigments and Photoreceptors6

Vision is not the only function for molecules capable ofreacting with light and it should be no surprise thatphotosensitive molecules can be found not only in eyesbut in animal tissues unrelated to vision as well as in plantsbacteria and other types of organisms that do not see Thephotopigments involved in vision in particular are madeup of two components (1) a light-sensitive molecule(chromophore) which changes physical conformation whenit interacts with light in all eyes studied to date this consists ofa photosensitive molecule known as retinal that is derived

from vitamin A (2) a membrane-bound opsin protein that isinvolved in the chemical cascade that transduces the incominglight to an electrical signal Opsins are members of a broadcategory known as G protein-coupled receptor proteins thatalso serve a range of nonvisual functions including chemore-ception (Nilsson 2004) Both component molecules predatethe origin of vision and their merger and subsequentspecialization in visual systems represents an importantexample of evolution through collage exaptation andsecondary adaptation

More than 1000 distinct opsin molecules have been iden-tified since the first example (bovine rhodopsin) was se-quenced in 1982 (Terakita 2005) Those that occur in animalsare divided into seven subfamilies all of which appear tohave originated before the split between the protostomes(most invertebrates) and deuterostomes (chordates and rela-tives including echinoderms) (Terakita 2005 Larusso et al2008) The extraordinary diversity of opsin molecules islikely a product of extensive gene duplication and subse-quent divergence (Arendt 2003 Plachetzki and Oakley 2007Oakley and Pankey 2008) Importantly the duplication ofopsin genes and their divergence in becoming reactive todifferent wavelengths of light forms the basis of color vision(eg Dulai et al 1999 Spady et al 2006 Briscoe 2008 Gerland Morris 2008)

Fig 10 Two of the several paths of eye evolution followed in mollusksContrary to some representations of the eyes in Fig 9 there is no simplelinear series from eye patch to complex camera type eye Rather eyesmay evolve in a variety of ways becoming specialized as pinhole-typelens-type or other types of eyes from an early beginning These areexamples of eyes from modern species and not actual ancestorndash

descendant transitions Note that these images are not drawn to the samescale (Nautilus eyes are about 10 mm across whereas the others areabout 1 mm) From Land and Nilsson (2002) based on drawings byHesse (1908 Patella Haliotis and Helix) Young (1964 Nautilus) andNewell (1965 Littorina) reproduced by permission of OxfordUniversity Press and Blackwell

6 For reviews of photoreceptor and photopigment function andevolution see Salvini-Plawen and Mayr (1977) Spudich et al(2000) Ebrey and Koutalos (2001) Hisatomi and Tokunaga (2002)Arendt (2003) Gehring (2004 2005) Terakita (2005) Plachetzki etal (2005 2007) Bowmaker and Hunt (2006) Purschke et al (2006)Santillo et al (2006) Lamb et al (2007) Kawamura and Tachibanaki(2008) and Oakley and Pankey (2008)

Table 1 Summary of the steps in the evolution of vertebrate eyes as proposed by Lamb et al (2007)

Time (Mya) Characteristicschanges Comments Functional

gt580 Bilaterally symmetrical animals evolve Some components may have functions otherthan in light detection and may worktogether in a simple light response systemin the absence of any visual organ

Yes Similar to the light sensing capabilityof soil-dwelling nematodes that lack anytype of eyesa

Various G-protein-coupled signalingcascades evolve and initially may functiononly in sensory systems other than vision(eg chemoreception)Early opsins (G-protein-coupled receptorproteins) evolveEarly rhabdomeric and ciliary photoreceptorsevolve

580ndash550 Ciliary photoreceptors and opsins continueto be modified

Organ serves as a simple light detector Yes Similar level of complexity as foundin some modern non-vertebrate chordates

550ndash530 Ciliary photoreceptor gains more complexsignal transmission capabilities

No image-forming capabilities but the organcan detect shadows or serve a circadianfunction

Yes Similar level of complexity as foundin modern hagfishes and larval lampreys

Eye-field region of brain bulges to formlateral ldquoeye vesiclesrdquo outside of the newlyevolved skullLateral vesicles invaginate bringing theproto-retina next to the proto-retinalpigment cell layerA transluscent layer of cells (a primordiallens placode) evolves and preventspigmentation of skin over the light-sensing organ

530ndash500 Photoreceptors develop cone-like features Eye has image-forming capabilities and canoperate over a relatively broad spectrumof light and range of light intensities

Yes Similar to the eyes of modern adultlampreysDuplication of genome creates multiple

copies of phototransduction genesCell types of photoreceptors diverge in formand have distinct opsinsRetinal information processing capabilityincreases as neural changes take placeLens placode invaginates and forms into asimple lensIris develops and basic pupil constriction ispossibleExtraocular muscles evolve

500ndash430 Myelin evolves and increases efficacy ofneural transmission

Eye has strong image-forming capabilitiesincluding for adjusting the amount ofincoming light and accommodating the lensto focus at different distances

Yes Similar to the eyes of many modernfishes

Rhodopsin evolves from cone opsin and rodbipolar cells evolve (possibly from rodphotoreceptors)Highly contractile iris evolvesRefractive capabilities of the lens improvedIntraocular eye muscles evolve allowingaccommodation of the lensRetina contains both rods and cones and hasmore efficient processing capability

lt 430 In tetrapods the lens becomes elliptical tocompensate for added refractive power ofthe cornea in air

Eyes as found in modern amphibiansreptiles mammals and birds

Yes You are reading this page

Eyes become specialized in different groupsaccording to different conditions (egnocturnal vs diurnal predators vs prey etc)

This hypothesis is based on several independent lines of evidence including analyses of genes and proteins comparisons of living species withdiffering degrees of eye complexity and information regarding the development of eyes in embryonic vertebrates For a detailed discussion seeLamb et al (2007) and for a less technical review see Lamb et al (2008)Mya Million years agoa See Ward et al (2008)

As noted photopigments are membrane-bound proteinswhich means that maximizing the number of molecules thatcan be contained within specialized light-sensitive cellsinvolves increasing the surface area of the cell membranesThis has generally been accomplished in two ways therebydefining two distinct categories of photoreceptors (1) inrhabdomeric photoreceptor cells membrane area is increasedthrough the growth of projections of the upper end of themembrane (apical microvilli) (2) in ciliary photoreceptor cells

a fold of the ciliary membrane is used for increasing theamount of photopigment that can be held These two cell typesdiffer in other important ways including in the type of opsinthey contain (rhabdomeric or r-opsin and ciliary or c-opsinrespectively) and the mechanism by which interactions withphotons are transduced into electrical information (Arendt andWittbrodt 2001 Arendt 2003 Nilsson 2004 Fig 11)

It was traditionally thought that rhabdomeric photorecep-tors were found only in protostomes whereas the ciliary type

Table 2 Examples of some of the direct and indirect evolutionary processes that may be involved in the evolution of eyes

Process Examples from eye evolution

Direct adaptive evolution Gradual evolution of lens crystallin concentrations resulting in evolution of graded refractiveindex lenses in aquatic animals

ExaptationOne structure has one function and takes on orswitches to a new function in a newenvironment

The cornea which has no refractive capacity in water became the primary focusingstructure after tetrapods moved onto landThe lens became far less important in image formation in terrestrial vertebrates and becamespecialized for accommodation instead

One structure has one function but becomesmodified enough to allow a shift in function

Circadian organ in early chordates became modified sufficiently that it became capable ofvisual functionsAn early protective transparent layer of cells became sufficiently thickened and invaginatedthat it could begin serving as an early lens

Two structures perform the same function butbecome differently specialized

Though both cell types were probably found in the distant bilaterian ancestor ciliaryphotoreceptors became the dominant type in vertebrates whereas rhabdomericphotoreceptors came to predominate in most other animals (see also duplication anddivergence)

A vestigial structure takes on new function In vertebrates rhabdomeric photoreceptors lost their microvilli and became retinal ganglioncells that function in circadian entrainment rather than in vision

Duplication and maintenance of repetition The compound eyes of arthropods are composed hundreds or thousands of repeated lenseyes called ommatidia

Duplication and divergence Opsin genes duplicated and diverged to become r-opsins and c-opsins along withspecialization of rhabdomeric cells with r-opsins and ciliary cells with c-opsinsIn certain taxa duplications and diversification of opsins to respond to different wavelengthsof light allowed the evolution of color visionThe rod cells of vertebrates are derived from cone cells both of which are derived from asingle ancestral ciliary photoreceptor

Gene sharing Some lens crystallin proteins function both in the eye in light refraction and elsewhere in thebody for other functions (eg cellular stress response)

Collage The first photopigment was formed by the combination of a preexisting light sensitivemolecule derived from vitamin A (which became retinal) with a preexisting G protein-coupled receptor protein (which became the ancestral opsin)The first ldquoeyerdquo arose by the combination of a photoreceptor cell with a pigment cellDuring the evolution of complex camera-type eyes various types of tissue that alreadyexisted (eg blood vessels nerves muscles) were incorporated

Scaffolding May apply to the evolution of phototransduction pathways or other relevant biochemicalsystems but more data are required

Constraints trade-offs and historical contingency Trade-off between resolution versus brightness in pinhole camera eyesTrade-off between visual acuity versus size of compound eyesInverted retina in vertebratesA narrow range of available wavelengths of sunlight is perceived in most animals probablybecause eyes first evolved in water which filters most wavelengths

Convergence Lenses irises and various other components of camera-type eyes emerged independently invertebrates and cephalopods

Parallel evolution The same developmental or other genes may have been independently co-opted in differentlineages (though homology is also a possibility)

was restricted to deuterostomes However further researchhas identified species containing both types The polychaeteworm Platynereis dumerilii for example has rhabdomericphotoreceptors in both its larval eyes and in its two pairs ofadult eyes as would be expected for a protostome Howeverit was also recently discovered to have ciliary photoreceptorsin its brain complete with c-opsins and regulatory genes moresimilar to those of vertebrates than to those associated with itsrhabdomeric photoreceptors These ciliary cells are not in-volved in vision but apparently serve a circadian function(Arendt et al 2004)

The box jellyfish Tripedalia cystophora provides a secondexample It has 24 eyes including two pit-shaped and twoslit-shaped pigment cups and two camera-type lens eyes posi-tioned at right angles to one another on each of four spe-cialized sensory clubs called rhopalia (Nilsson et al 2005Kozmik et al 2008a b Fig 4) The jellyfish camera-typeeyes appear to use ciliary opsin as the photopigment andmelanin as the shielding pigmentmdashboth as in vertebratesmdashthough its lens protein is distinct from those of other animals(Kozmik et al 2008a b as an aside it is somewhat enig-matic that complex camera-type eyes complete with tinyspherical graded index lenses and in the lower eye an iriswould occur in these animals as they are not connected to abrain and are not arranged in such a way as to generate asharp image in any case Nilsson et al 2005 Wehner 2005)

Humans like other vertebrates make use of ciliary photo-receptors with c-opsins in their retinas for image processing

(see Kolb 2003 for a review of retina morphology and func-tion) However a second category of photoreceptor wasdiscovered in 1991 in the form of a subset of retinal ganglioncells containing a photopigment dubbed melanopsin It is nowrecognized that these cells are remarkably similar to therhabdomeric cells of invertebrates and that melanopsin issimilar to r-opsin (see Van Gelder 2007 for review) Thoughthey are located within the retina these cells do not functionin image formation instead they appear to serve a circadianfunction It is for this reason that some blind people lackingrods and cones can nonetheless respond to dayndashnight cycles(Van Gelder 2007 Zaidi et al 2007)

The cases of Platynereis Tripedalia and humans suggestthat most animals will turn out to exhibit both types of pho-toreceptor cells or at least that they had both at some stagein their ancestry (Plachetzki et al 2005) They also suggestthat both cell types were present in the common ancestor ofall bilaterally symmetrical animals (the Urbilaterian Arendtand Wittbrodt 2001 Arendt et al 2004) In particular thesecells are thought to derive from a common ancestral cell viaduplication (or furcation sensu Oakley et al 2007) and diver-gence (making them ldquosister cell typesrdquo Arendt 2003) withprotostomes eventually using the rhabdomeric type in theireyes and deuterostomes using the ciliary type The reason forthis distinction is not clear and may represent little morethan a quirk of history

There is also evidence that duplication and divergencehave occurred within cell types Not only are the rod cells

Fig 11 The two photoreceptor types found in animals A Rhabdomericphotoreceptors which use extensions of the membrane (apical microvilli)to increase the amount of photopigment (r-opsin) that they can containRhabdomeric photoreceptors are the predominant type found in proto-stomes (most invertebrates) B Ciliary photoreceptors which make use ofa modified ciliary membrane to increase the surface area available forstoring photopigment (c-opsin) Ciliary photoreceptors are the main type

observed in deuterostomes (vertebrates echinoderms and relatives) Thetwo cell types differ in their phototransduction pathways and in theelectrical responses that result It is thought that both cell types werealready present in the ancestor of all bilaterian animals having beenderived from one ancestral photoreceptor type (Fig 14) From Arendtand Wittbrodt (2001) reproduced by permission of The Royal Societyand D Arendt

of the vertebrate retina thought to be derived from conecells within the ciliary type (Okano et al 1992 Kawamuraand Tachibanaki 2008) but so too are the bipolar cells inthe retina (Lamb et al 2008) Similarly it has beensuggested that the descendants of rhabdomeric photo-receptors include not only the retinal ganglion cells butalso the horizontal and amacrine cells (Arendt 2003) Ifcorrect then this would mean that the retina at large isderived from two ancestral types of photoreceptors whichthemselves are derived from a single type that existed longbefore the evolution of vision

Lenses7

Sharp image formation requires that incoming light rays beredirected so that they converge on the photoreceptors of theeye There are various ways by which this can be accom-plished for example by either reflecting or refracting lightAnimals with camera-type eyes use the latter mechanism byplacing material of a refractive index different from theexternal medium in between the source of light and the retina(Fig 5)

It is not difficult to see how gradual changes (ie directadaptive evolution) could refine the function of lenses oncethey appeared in very rudimentary form As Land and Nilsson(2002 p 58) explained

a small blob of jelly or mucus with a refractive indexsomewhat higher than the surrounding water placed inthe pupil of the eye will converge entering rays slightlyand this in turn will reduce the width of the blur circle onthe retina without requiring a decrease in pupil diameterThe process of improvement could continue until thelsquolensrsquo converged the light to a point on the retina atwhich stage the transformation to an image-forming eyewould be complete

Any substance with an appropriately different refractiveindex can improve focus in an eye lacking a lens as Dawkins(1996) showed even a bag of water or glass of white winecan enhance the image formed by a pinhole camera That aldquoblob of jellyrdquo could be functional is demonstrated by thefact that such a simple lens which converges light but is notsufficiently effective to form an image exists in snails of thegenus Helix (Land and Nilsson 2002 Fig 10) The lenses ofcephalopod mollusks or vertebrates are far more sophisticat-ed and are capable of focusing light sharply They also ex-hibit graded refractive indices lower at the edges and higher

in the centre which correct for spherical aberration (Landand Nilsson 2002 Fig 7) Nevertheless the difference be-tween the lens ofHelix and that of Octopus is one of degreeseasily connected through incremental steps The origin of theearliest lenses on the other hand can only be understood asa result of indirect evolution

As Lamb et al (2007 2008) explain the tissues that formthe lens in vertebrate eyes did not always have this functionAt first they probably served a developmental role by trig-gering the invagination of the eye vesicle later thickeningand becoming transparent and eventually becoming suffi-ciently modified in this regard to provide a small amountof focusing power Certainly this series of changes doesoccur within the early development of vertebrate embryos(Cvekl and Piatigorsky 1996 see Fig 12) though of courseone must be very cautious not to overstate any similaritiesbetween ontogeny (development) and phylogeny (evolution-ary history) One of the major ways that such rudimentarylenses became specialized for their new imaging function wasby the accumulation of refractive proteins known as crystal-lins within the cells of which they are composed Allcrystallins have in common the fact that they are globularwater-soluble proteins capable of refracting light either alone(monomers) or in combinations of various numbers ofmolecules (dimers tetramers or more complex aggregates)can be densely packed and remain stable for the lifetime ofthe organism (Piatigorsky 2007)

Unlike opsins in the retina crystallins are not descendedfrom a common ancestral protein Rather these have been co-opted from a wide range of preexisting proteins in differentlineages especially from common stress proteins or metabolicenzymes (Fig 13) For this reason True and Carroll (2002)suggest that crystallins represent ldquoby far the classic and beststudied cases of co-option in animal evolutionrdquo Notablymany lens crystallins are not only similar but identical toproteins that serve other functions in the eye and elsewherein the body That is they provide not just an example of co-option but of gene sharing such that only comparativelyminor changesmdashsuch as in the amount and location of geneexpression of existing proteinsmdashwas required to produce thefirst simple layers of tissue capable of refracting light(Piatigorsky 2007 2008) Once this occurred gradualnatural selection could have refined them and the tissues inwhich they reside to produce increasingly efficacious lenses

Vertebrate lens crystallins consist predominantly of repre-sentatives from two protein groups the α-family and the βγ-superfamily The α-family crystallins in the lens are derivedfrom chaperone proteins and include two major forms thathave emerged through duplication and divergence αA islargely lens-specific while αB is expressed elsewhere in thebody such as in the heart muscles and brain Both stillfunction in a chaperone capacity in addition to their role in

7 Discussions of the structure andor evolution of lenses corneas andcrystallins are provided by Tomarev and Piatigorsky (1996) Piatigorsky(1998 2007 2008) Land and Nilsson (2002) Bloemendal et al (2004)and Jonasova and Kozmik (2008)

image formation (Bloemendal et al 2004) The β and γcrystallins are each also represented by multiple gene copiesonly some of which are expressed exclusively in the lensHowever in this case the functions of the copies expressedoutside the eye remain to be determined (Bloemendal et al2004 Piatigorsky 2007)

One of the key questions in the evolution of the vertebratelens relates to how these genes could first have come to beexpressed at a high level in the tissue that evolved into thelens such that there could then be selection on their ar-rangement abundance and other properties to enhance theirrefractive function This question was answered in part in a

recent study by Shimeld et al (2005) who showed that thesea squirt Ciona intestinalis (a chordate specifically a uro-chordate but not a vertebrate) possesses a version of a βγcrystallin gene that is expressed preferentially in larval light-sensing structures called ocelli even though these consist ofonly a single pigmented photoreceptor cell It therefore musthave a function that is associated with visual systems but notlight refraction because Ciona ocelli do not have lenses Theregulatory genes that produce this tissue-specific expressionare very similar to those that direct crystallin expression invertebrate lenses in fact one can transfer the gene fromCiona to a frog and frog lenses will be produced (Shimeld

Fig 12 A schematic representation of eye development in cephalopodmollusks (eg squid) and vertebrates these represent the series of stagesthat occur during the development of individual embryos (proceedingfrom top to bottom) and should not be confused with the historical pathfollowed during eye evolution in either mollusks or vertebrates The dif-ferent evolutionary and developmental pathways followed by cepha-

lopods versus vertebrates result in eyes that are superficially similar(both are complex camera-type lens eyes) but with some majordifferences Notably the optic nerve does not pass through the retina incephalopods meaning that they do not have blind spots From Harris(1997) reproduced by permission of the National Academy ofSciences of the USA

Turtle

Liz ard

Crocodile

BirdKangaroo Human

Guinea pig

Rabbit

Camel Octopus

Squid Fly Jellyfish

Embryonic

Crystallin Ancestral protein function

Small heat shock protein

Related to bacterial stress protein

NADPH-dependent reductase

Arginosuccinate lysase

α enolase

Lactate dehydrogenase

Similar to bacterial ornithine deaminase

Aldehyde dehydrogenase

Alcohol dehydrogenase

Hydroxyacyl-CoA dehydrogenase

Similar to yeast TSFI

Insect cuticle protein

Similar to chaperonin 60kd hsp

Glutathione-S-transferase

Glyceraldehyde phosphate dehydrogenase

Fig 13 Molecular co-option in the evolution of lens crystallin proteinsThe α β and γ crystallins are ubiquitous among vertebrates though theembryonic γ crystallin has been lost in birds Other crystallins are shownfor the specific taxa in which they have been found and this may notreflect their overall distribution among animals (A) Each of thesecrystallins is similar to another protein with a different function (B)suggesting that it was co-opted from an existing protein in a new role in

refracting light in the eye In fact many crystallin proteins are expressedin tissues besides eye lenses where they carry out additional functionsunrelated to vision (see Piatigorsky 2007) From True and Carroll(2002) reproduced by permission from the Annual Review of Cell andDevelopmental Biology Volume 18 copy2002 by Annual Reviews wwwannualreviewsorg

et al 2005) This suggests that both the ancestral βγ crys-tallin gene (perhaps in only one copy) and the regulatorysequences that influence where it is expressed already existedin a chordate ancestor prior to the evolution of lenses invertebrates and were co-opted into this new role

An independent example of this process was recently high-lighted in an elegant study of squid lenses by Sweeney et al(2007) In cephalopods the lenses include only one type ofcrystallin S-crystallin which was derived by gene duplicationand divergence from the liver enzyme glutathione S-transferaseThe major difference between the two copies is that the S-crystallin sequence includes an exon (protein-coding region)that the liver enzyme gene does not with the result that thecrystallin protein exhibits an extra loop in its three-dimensionalstructure This loop turns out to be functionally significant inthe lens and there is evidence that it began small and wasaccentuated by gradual selection as the crystallin took on itsnew visual function (Sweeney et al 2007) Thus the history ofthe squid lens includes gene duplication and divergence small-scale mutation and natural selection regulatory mutations andco-option (Oakley 2007) all of which left tell-tale traces thatare now being uncovered by careful scientific study

Corneas

Refraction occurs when the velocity of a light wave changesas it travels through different media The camera-type eyes ofcephalopods and vertebrates evolved in water which has arefractive index about the same as the fluid inside the eye Asa consequence there is very little refraction of light as it entersthe eye from the surrounding aqueous medium The situationbecame quite different for the vertebrates who moved ontoland because the refractive index of air is very different fromthat of eye fluid meaning that passage into the eye through thecornea itself generated significant refraction In fact whereasthe cornea has little or no focusing function in aquaticanimals it may contribute as much as 70 of the lightrefraction in terrestrial vertebrate eyes (Land and Nilsson2002) Thus movement into a new environment led to amajor functional shift in the cornea from a largely protectiverole to a visual one for which it then became secondarilyadapted (eg by becoming curved8) This in turn meantthat the lens was largely obsolete for the major refractivefunctionmdashin fact at first it would have been redundant to thepoint of overfocusingmdashand in terrestrial vertebrates it becameco-opted for a role in accommodation (focusing at differentdistances) which it accomplished by becoming flattened anddeformable (or in the case of nocturnal animals especiallylarge Land and Nilsson 2002) Thus indirect evolution

especially exaptation and secondary adaptation has played asignificant role in the evolution of corneas and lenses as theyexist in modern vertebrates

Given that corneas rather than lenses now serve the primaryrole of refraction in many terrestrial vertebrates it is interestingto consider whether they share any similarities to lenses at themolecular levelmdashin particular whether corneal crystallins havea similar evolutionary history to crystallins found in lensesAlthough the situation is not as clear (in part because a visualfunction of corneal crystallins has not yet been establishedPiatigorsky 2007) there is evidence that corneal crystallinevolution does indeed exhibit a history of duplication co-option and gene sharing For example the corneal crystallinsof mammals are the same as an active aldehyde dehydroge-nase class 3 enzyme (which functions in detoxification andmay protect against UV damage) whereas those of birds aresimilar to peptidyl prolyl cis-trans isomerase (Cuthbertson et al1992 Piatigorsky 2007) Some corneal crystallins turned outto be the same as those in the lens including a glutathione S-transferase derivative in the squid cornea Recent studies offishes and frogs have confirmed similar patterns of geneco-option In fishes a series of proteins derived by gene dupli-cation from an ancestral gelsolin gene has become differentiallyexpressed in different tissues some primarily in the corneasome in the cornea lens brain and heart and others in tissuesother than these (Jia et al 2007 Piatigorsky 2007) In frogswhich are amphibious and therefore have eyes adapted tovision both on land and in the water the same proteins havebeen recruited in the cornea as in the lens (Krishnan et al2007) Overall then the evolution of the cornea appears tohave been similar to (and perhaps predated Piatigorsky 2007)the evolution of lenses in terms of the taxon-specific co-optionandor gene sharing of crystallin genes

How Many Times Have Eyes Evolved

The term ldquohomologyrdquo can be defined as ldquothe presence of thesame feature in two organisms whose most recent commonancestor also possessed the featurerdquo (Hall 2007) The coun-terpoint to ldquohomologyrdquo is ldquohomoplasyrdquo (often also taught asldquoanalogyrdquo) which refers to similarities among species thatare not due to the inheritance of the same trait from a com-mon ancestor9 Obviously the question of how many times aparticular feature has evolved is predicated on a clear dis-tinction between homology and homoplasy10 As Hall (2007)

8 Interestingly aquatic mammals such as seals have secondarilyreverted to possessing flat corneas and spherical lenses like those offishes (Land and Nilsson 2002)

9 For historical reviews of the concepts of ldquohomologyrdquo ldquoanalogyrdquo andldquohomoplasyrdquo see Hall (1994 2003) and Panchen (1994 1999) For aless technical discussion of homology and homoplasy see Thanukos(2008)10 Alternatively the question can be framed in a phylogenetic sense asbeing between ldquomonophylyrdquo (all species with the feature descend from acommon ancestor with the feature) or ldquopolyphylyrdquo (species with the featurehave evolved independently in separate lineages Piatigorsky 2008)

noted ldquohomology is the persistence of similarityrdquo whereasldquohomoplasy is the recurrence of similarityrdquo Yet as is typicalfor most questions in the complex world of biology anunambiguous line between homology and homoplasy is notso easy to draw

First one must consider the possible causes of homoplasytwo of which are convergence which reflects the independentevolution of a similar feature using different genetic anddevelopmental systems and parallel evolution the evolutionof similar features that occurs independently but uses the sameunderlying systems11 Thus already the question is compli-cated by the fact that a homoplasious organ can evolve inde-pendently from the same starting point and using the samegenes Some authors consider the sharing of the same devel-opmental genetic systems as a form of ldquodeep homologyrdquo (egShubin and Marshall 2000 Rutishauser and Moline 2005)Nielsen andMartinez (2003) proposed the term ldquohomocracyrdquofor organs whose development is governed by homologousgenes Second the designation of a feature as homologousversus homoplasious is dependent on the scale at which thecomparison is made (Hall 2003 Rutishauser and Moline2005) For example the last ancestor shared by all modernbirds was endothermic (ldquowarm-bloodedrdquo) as was the lastshared ancestor of all modern mammals meaning thatendothermy is homologous within both birds and mammalsHowever the most recent ancestor common to both birdsand mammals was not endothermic which means thatendothermy is homoplasious when these groups are com-pared with each other Third it is clear that a distinction canbe drawn between a complex feature as a whole and thecomponents of which it is made with regard to homologyversus homoplasy As an example it is often correctly notedthat the wings of birds and bats are homoplasious given thatthe distant ancestor from which they are both descended(which was neither a mammal nor a bird) did not fly On theother hand the limbs of all tetrapods share a fundamentalbone structure that has been inherited from an early commonancestor meaning that the limbs of which bird and bat wingsare constructed are homologous (Rutishauser and Moline2005) These are also homologous with the bones of humanarms dog legs and whale flippers because homology isdefined on the basis of shared inheritance not function(Wagner 2007)

Overall it is debatable whether a dichotomy exists betweenhomology and homoplasy some authors have argued thatinstead these represent extremes on a continuum of degrees ofmodification (eg Hall 2003 2007) In this sense the ques-tion of whether a complex organ as a whole has evolved onceor many times may be unanswerable Rather it is often muchmore useful to discuss the components of the organs

themselves or the genetic biochemical cellular and devel-opmental systems that underlie their production and functionand to explore the extent to which these rather than the com-plex organs per se are shared via common descent amonggroups It is within this context that one must consider thequestion of how many times ldquoeyesrdquo have evolved

In their classic review of eye diversity and evolutionSalvini-Plawen and Mayr (1977) noted that ldquoit requires littlepersuasion to become convinced that the lens eye of avertebrate and the compound eye of an insect are indepen-dent evolutionary developmentsrdquo Even within a single eyetype for example the camera-type lens eyes of vertebratesversus those of cephalopods it is obvious that there has beensubstantial convergent evolution at the level of the organ as awhole Human and squid eyes are constructed from differenttissues exhibit different morphological organization of theretina and other features and use different photoreceptorsopsins and crystallins (Land and Nilsson 2002 Sweeney etal 2007 Fig 12) Moreover the distribution of eyes acrossthe one third of phyla that have them makes it unlikely thatthey existed in any complex form in the ancestor of allanimals because this would require that they have been lostin an improbably large number of lineages (eg Land andNilsson 2002) Similar phylogenetic analyses even suggestthat eyes have evolved independently within some taxa (egostracod crustaceans Oakley and Cunningham 2002) Thusconsiderations of the morphological differences in eyes andphotoreceptors and their occurrence in divergent phyla speakstrongly in favor of homoplasy This led Salvini-Plawen andMayr (1977) to provide an oft-cited estimate in which eyeshave evolved independently at least 40 times and perhaps 65times or more

Modern information derived from many lines of evidenceincluding comparative morphology molecular biology phy-logenetics and developmental biology clearly shows that eyesare the product of a complex evolutionary history At the sametime the combination of data from these divergent lines ofinquiry has tended to blur rather than resolve the long-standingpuzzle of howmany times eyes have evolved (eg Arendt andWittbrodt 2001 Arendt 2003 Oakley 2003 Fernald 2004aNilsson 2004) Whereas ldquoeyesrdquo strictly defined as complexvisual organs clearly have arisen independently in differentlineages the ancestral opsins and photoreceptors from whichall modern eyes are derived are thought to have been presentin the last common ancestor of all bilaterians thereby makingthem homologous (Arendt and Wittbrodt 2001 Arendt 2003Fig 14) It is also possible that vertebrates and jellyfishesinherited their similar opsins and photoreceptors from an evenmore distant common ancestor though it has been argued thatthe intriguing similarities between them reflect independentrecruitment of the same genes (ie parallel evolution) ratherthan homology (Kozmik et al 2008a) The possibility thatvertebrate retinal ganglion cells and invertebrate rhabdomeric

11 A third cause of homoplasy reversal is not covered here See Hall(2003 2007)

photoreceptors represent examples of parallel evolution ratherthan homology also has been raised though this is not acommon interpretation (Nilsson 2004 2005)

In any case most current arguments in favor of eyehomology are based not on comparisons of eyes or evenphotoreceptors but of genesmdashspecifically those involved inregulating eye development In 1994 it was discovered thatgenes that had been known to cause a loss of eyes whendefective in flies (eyeless) mice (small eye) and humans(Aniridia) are in fact highly conserved versions of the sameregulatory gene Pax6 (Quiring et al 1994) The followingyear Halder et al (1995a) manipulated the expression of theeyeless gene in flies so that it was active in tissues other thanwhere it normally functions with the result that eyes formedon the wings the legs and the antennae Not only this butHalder et al (1995a) showed that the Pax6 gene from micecould also induce the generation of eyes in fliesmdashbut notmouse eyes fly eyes (Fig 15) Similarly it was shown thatectopic expression of Pax6 would induce the production ofeye lenses in non-eye locations in frogs (Altmann et al1997 Chow et al 1999) Observations such as these led tothe conclusion that Pax6 is a ldquomaster control generdquoresponsible for eye formation and that it has served this role

at least since the last common ancestor of insects andvertebrates Combined with the notion that photoreceptorsalso were present in the urbilaterian it was suggested thatldquoeyesrdquo defined minimally as a pigment cell and an opsin-containing photoreceptor cell regulated by a version of Pax6evolved once with all subsequent eyes arising throughtinkering and elaboration of this ancestral system (Halderet al 1995b Gehring and Ikeo 1999 Gehring 2001 20042005 see Kozmik 2008 for discussion)

It may seem reasonable to conclude therefore that eyessimply exhibit deep homology with regard to their underlyingphotoreceptors photopigments and regulatory genes but arehomoplasious in terms of their crystallins and overallorganization Life however is rarely ever this simple Thusit has also been discovered that Pax6 is not the only majorgene involved in eye development (in flies there is anetwork of seven such genes Kumar 2001 Gehring 20042005) Moreover Pax6 is involved in many functionsbesides eye development including in brain nose andpancreas development in mice It even regulates someaspects of development in nematodes which lack visualorgans altogether (Hodin 2000 Piatigorsky 2008) There-fore the gene itself may be homologous but this would not

Fig 14 The current distribution and origin of rhabdomeric (darkgray) and ciliary (white) photoreceptor cells among animals (see alsoFig 11) These are thought to have descended from cells alreadypresent in the last common ancestor of all bilaterally symmetricalanimals (the Urbilaterian) It is thought that ciliary and rhabdomericphotoreceptor cells are descended from a single ancestral cell type

though there is some ongoing debate as to whether the Urbilaterianpossessed (a) a ciliary precursor type from which both are derived (b)a dual ciliary or rhabdomeric precursor that later diverged into twotypes or (c) both ciliary and rhabdomeric cells that had alreadydiverged earlier in animal evolution From Arendt and Wittbrodt(2001) reproduced by permission of The Royal Society and D Arendt

make eyes homologous if its ancestral function was unre-lated to the development of visual organs As Kumar (2001)noted ldquoseveral of the eye-specification genes are expressedand are involved in the development of non-eye tissue butwe do not say that the eye is homologous to the brain or tomusclesrdquo It is even possible that Pax6 has been indepen-dently recruited for a role in guiding visual organ devel-opment in different lineages (ie by parallel evolution)because it is one of the few regulatory genes expressed ldquoatthe right place at the right timerdquo during development thatcould be co-opted into this role in each lineage (Hodin 2000)

Overall the question of whether ldquoeyesrdquo evolved once ormany times remains an open one though the availableanswers depend more than anything on definitions and levelsof analysis In fact it may not be useful to consider complexorgans in this way at all Instead it is more productive to focuson the components of eyes which have evolved and beencombined and modified in a variety of ways in differentgroups

Constraints and Historical Contingency

Eyes represented one of the main examples of what Darwin(1859) called ldquoorgans of extreme perfection and complica-tionrdquo To be sure many features of eyes appear to have beenoptimized within the limits imposed by optical principles(Goldsmith and 1990 Land and Nilsson 2002) Howeverwhile the eyes of many species are complex and effectiveperfect they are not In fact they provide as strong an ex-ample of imperfectionmdashin particular imperfection due tohistorymdashas anything else There are numerous suboptimalfeatures of eyes at the best of times and various ailmentsmay arise because of the way eyes have been tinkered intoexistence by direct and indirect evolutionary processes De-tached retinas angle closure glaucoma macular degenera-

tion and other breakdowns of eye function all fall within thiscategory (see Novella 2008 for a review)

All eyes not just those of humans are influenced byhistorical contingency and constraints and moreover onceadaptation has begun down a particular pathmdashfor example inthe formation of a camera-type compound or pinhole eyemdashitcan be very unlikely for evolution to reverse course andchange to a different avenue Natural selection has no fore-sight or objectivesmdashit is simply the greater reproductivesuccess of certain individuals relative to others in the popu-lation based on heritable traits that they happen to possess Ifindividuals with slight improvements in a particular kind ofeye leave more offspring than individuals lacking thisimprovement then this change will be favored regardless ofwhether it means another step up an alley that is partly blind

Put another way optimization is local rather than idealisticThe metaphor most often used in this sense is one of amountain range consisting of ldquoadaptive peaksrdquo of varyingheights Insects provide a notable example of what can beconsidered climbing a ldquolocal adaptive peakrdquo which may notbe the highest but any movement off of which means areduction in fitness No doubt eyes (alongwith other complexstructures like wings) have contributed greatly to theirenormous success and it is not difficult to imagine even amoderately less effective eye being quickly eliminated from apopulation in any environment in which light is available Yetas well tuned as they now are the compound eyes of insectsare subject to fundamental physical constraints that preventthem from ever matching the visual acuity provided bycamera-type eyes (Fig 16) Whereas human eyes can resolvepoints that are 0008deg apart flies can distinguish points noless than 4deg apart (Kirschfeld 1976) as Nilsson (1989) put itldquoone fruit fly would need to be closer than 30 mm to seeanother as anything other than a single spotrdquo Similarly thecephalopod mollusk Nautilus possesses a pinhole eye of

Fig 15 Eyes induced to develop in unusual locations in a fly(Drosophila) such as on the head near the normal eye on the antennaand next to the wing (arrows) by experimentally expressing theeyeless (Pax6) gene Using the Pax6 gene from a mouse has a similar

result indicating that the gene in both species is very similar and wasprobably inherited in both lineages from a distant common ancestorFrom Halder et al (1995a) reproduced by permission of the AmericanAssociation for the Advancement of Science

which further improvement is greatly restricted by a tradeoffbetween the resolution of an image and its brightness (Landand Nilsson 2002)

Likewise the human eye is limited in many regards ascompared to the eyes of other animals Squid for examplehave retinas that are not inverted and are without a blindspot The resolution of hawk eyes is about twice as high asthat of humans and there are many nocturnal animalscapable of vision in far dimmer light than is conducive tohuman sight Stomatopod crustaceans (mantis shrimps)which exhibit the most complex eyes thus far discoveredpossess 16 morphologically distinct photoreceptor typesand can see both UV and circular polarized light (Marshallet al 2007 Chiou et al 2008 Cronin and Porter 2008)

Misconceptions About Complex Organ Evolution

Although the exact history of any particular complex bio-logical feature may never be fully worked out biologists areconfident that the mechanisms by which they arise can bemdashand to a growing extent aremdashunderstood Yet despite majoraccomplishments on both theoretical and empirical fronts

misconceptions regarding the evolution of complex organsremain common (including among those who accept evolu-tion as historical reality) and prevent this understanding fromextending more widely Some of the most common of theseare outlined in the following sections (see also Le Page 2008)

Misconception 1 Supposed intermediate stages in theevolution of complex organs could notbe functional

This misconception is often expressed in the form of aquestion ldquoWhat good is half an Xrdquo where X is any complexfeature such as a wing or an eye12 More generally it is basedon the assumption that a complex system missing some of itsparts would be completely nonfunctional and therefore couldnot have evolved through less complex ancestral stagesSuch an assumption is faulty for several reasons First theassumption that complex organs must be either perfectlyfunctional or totally functionless is demonstrably false Forexample various components of the eye can be misshapenmisaligned or even missing and yet vision is not totallylost13 The same often holds in biochemical systems forexample a heritable deficiency of Factor XII (Hagemanfactor) of the blood clotting cascade may have little effect oncoagulation ability because other components of the systemcan compensate for its absence (Ratnoff andMargolius 1955)Second whereas removal of some components may indeedrender a biological system nonfunctional for its current roleit may still be functional in a different way For example theeyes of individuals missing rod and cone cells may still servea circadian function (Zaidi et al 2007) Again this alsoapplies to subcellular systems as with the observation that abacterial flagellum missing many of its parts is no longeruseful for locomotion but can be used as a toxin injectionapparatus (Pallen and Matzke 2006) Third species exhibit-ing less complex but obviously functional versions of thesame system often can be found With eyes there is a clearrange in complexity from single photoreceptors to complexeyes of various types all of which function in the organismspossessing them (Salvini-Plawen and Mayr 1977 Land andNilsson 2002) What all of this shows is that biologicalfeatures can exist function and be beneficial relative toalternativesmdashand therefore evolve by natural selectionmdasheven when some components are only marginally effectiveor indeed lacking altogether Again the only requirement is

12 The literal answer is that half an ldquoXrdquo is not a non-functional half-letter it is either a forward slash () or backslash () or a ldquoVrdquo or caret(^) all of which work just fine for different functions Of course theseare all modern characters and are not ancestors of the letter ldquoXrdquo whichwas probably co-opted from the Greek chi13 Case in point the present author suffers from a trifecta of myopiaastigmatism and strabismus all only partially corrected with eye-glasses and rather unpleasant surgery but he notes that this is stillvastly preferable to no vision whatsoever

Fig 16 Physical constraints on eye function Although compound eyeshave proven remarkably successful they are fundamentally limited intheir resolution by physical constraints In order to exhibit the sameresolution as is typical with camera-type lens eyes a human being wouldrequire compound eyes a meter across and comprised of a million facets(only about 100 are shown for simplicity) From Kirschfeld (1976)

that less complex stages be functional for something and notnecessarily for their current function and that they be at leastslightly better than existing alternatives in enhancing survi-val and reproductionMisconception 2 Ancestral intermediates either missing

or still presentThe concept of ancestral intermediates (or transitional forms)

is subject to misunderstanding in a variety of ways On the onehand it is sometimes suggested that stepwise evolution shouldleave behind millions of fossil intermediates that illustrate theevolution of a feature along a smooth gradation The fact thatonly a miniscule fraction of individual organisms is preservedas fossils makes this an unreasonable expectation especially incases where evolutionary changes are relatively rapid Notablyvarious complex eyes appear to have evolved during thecomparatively short period of the Cambrian more than500 Mya On the other hand there is sometimes a tendencyto see ldquoancestral intermediatesrdquo in the wrong place namely incomparisons of living species Whereas some species do retainmore primitive (ancestor-like) versions of particular organs andtherefore provide information about how such simpler config-urations could function it is incorrect to compare modernspecies as though they represent actual steps in an evolutionarysequence of ancestors and descendants In both cases a betterunderstanding of ldquotree thinkingrdquo can help to correct themisunderstanding (eg Gregory 2008c)Misconception 3 Biologists propose that complex organs

arise ldquoby chancerdquoThe variation upon which natural selection acts is gene-

rated by mutation and it is certainly the case that theseaccidents of inheritance occur randomly with respect to anyconsequences that they may have It is also true that naturalselection possesses no capacity for foresight and has no finalobjectives as it alters population characteristics from onegeneration to the next It does not follow however that theevolution of complex organs occurs ldquoby chancerdquo By defini-tion natural selection is the non-random differential successof individuals on the basis of heritable variation and there-fore the cumulative outcome of this processmdashadaptationmdashisthe opposite of random chanceMisconception 4 All features of a complex organ are

optimalNatural selection can lead to remarkably effective func-

tional capabilities in complex organs That it will do so in allcases is far from guaranteed however For a start any opti-mization that does occur will proceed only as far as thelimitations of the system will allow Furthermore functionalimprovement in one organ must take place within the contextof tradeoffs with other systems Finally there must be anappropriate imposition of selective pressures by the environ-ment the necessary genetic variation and a sufficiently flexi-ble developmental program if significant change is to occur inthe direction of improved function Because organs are built

by tinkering rather than design their features are impacted byhistorical contingency and inevitably reflect holdovers of paststates (Shubin 2007) The net result is that all complex or-gans represent a mixture of optimizations and imperfectionsboth of which are accounted for by their evolutionary historyMisconception 5 Indirect evolution implies that various

preexisting structures are assembled in-stantaneously into a new organ

Indirect evolutionary processes such as shifts in function(exaptation) and the combining of existing structures (collage)play important roles in the origin of complex organs How-ever this is sometimes misunderstood to imply that all com-ponents are (1) already in their final form when assembledtogether and (2) brought together simultaneously to form anew complex organ In actuality the addition of componentsto an evolving organ may occur in series (ie one at a time)and may involve a relatively poor fit at first Again the onlycriteria for such additions to be preserved by natural selectionare that they must serve some function and they must conferan advantage relative to alternatives within the populationOnce combined secondary adaptation may improve the inte-gration of parts and may in the process enable other parts to beadded once again possibly resulting in a shift in function andbeginning as a poor fit that is enhanced only later (Fig 3)Thus even indirect evolution is gradual in the sense that itproceeds in a stepwise fashion and requires no prohibitivelyimprobable leapsMisconception 6 Hypotheses about complex organ evo-

lution cannot be testedBecause they relate to unique events postulated to have

occurred in the deep past it is sometimes argued that hypo-theses about the evolution of complex organs cannot be testedIt is a truism that the exact details of distant historical eventscannot be known with absolute certainty However this doesnot mean that no conclusions can be drawn about them on thebasis of supporting data (Cleland 2001 2002) Interestinglythe fact that hypotheses about the evolution of complexorgans can be tested is tacitly acknowledged by those whoclaim that a lack of a great many transitional fossils calls fora rejection of the hypothesis that complex organs evolvedgradually ndash erroneous and based on false premises to besure but a prediction nonetheless Of course biologists domake predictions regarding the characteristics of as-yet-unknown fossil species in order to test hypotheses aboutevolutionary transitions As a prime example the transitionalspecies Tiktaalik roseae which possessed characteristics ofboth fishes and tetrapods was recently discovered in rocksof the anticipated age as part of an explicit effort to test ideasregarding the evolutionary transition from water to land(Daeschler et al 2006 Shubin et al 2006) Aside from fos-sils biologists also make use of molecular genetic anato-mical developmental and cytological data to elucidate theevolution of complex organs in some cases making very ex-

plicit predictions in order to test their explanatory hypotheses(see for example the thorough list of testable predictionsregarding vertebrate eye evolution by Lamb et al 2007)Some of the extensive information resulting from theseindependent types of study has been reviewed here though itshould be noted that this relates to just one organ and coversbut a fraction of the scientific literature on complex organevolutionMisconception 7 An inability to explain every detail of a

complex organrsquos history challenges thevalidity of evolutionary science

Following in the tradition of Paley (1802) from twocenturies ago it is sometimes asserted that if a naturalexplanation is unavailable to account for an observation thenthe only alternative is to assume a supernatural one Such anassumption misses the obvious third option and the one thatdrives scientific inquiry that there is a natural explanationthat is not yet known Thanks to work carried out in pursuitof natural explanations a great deal is now understood aboutthe form function and probable origins of many complexbiological organs and systems That said it should come asno surprise that these investigations do not provide detailedunambiguous step by incremental step reconstructions ofevents that in many cases occurred hundreds of millions ofyears ago Of course the same is true even of studies dealingwith much more recent events as in archeology history orforensicsmdashin all cases the important requirement is aconvergence of several independent lines of evidence thatpoint to the same conclusion As discussed in this article datafrom genetics molecular biology developmental biologycomparative anatomy paleontology and even medicine allpoint to a conclusion that eyes are the product of naturalevolutionary mechanisms

In any case there is a broader misconception inherent inthe claim that an incomplete recounting of a particularevolutionary outcome challenges the validity of evolutiongenerally This involves the common confusion surround-ing the meanings of ldquofactrdquo and ldquotheoryrdquo in science and interms of historical sciences in particular of ldquopathrdquo (Gregory2008b) That species are related by common ancestry issupported by an overwhelming body of evidence frommany disciplines In fact no reliable observation has beenmade to refute this conclusion in more than 150 years As aresult evolutionmdashdefined as a historical process of descentwith modificationmdashhas long been accepted as a fact by thescientific community Facts require explanations howeverand this is the job of evolutionary theory which includesamong other things the mechanism of natural selectionQuestions regarding a particular organrsquos unique evolution-ary path (When did it appear Did it evolve more thanonce) or mechanism responsible for its origin (AdaptationExaptation) do not bear on the fact that species are relatedby common descent which has long been established using

other evidence As such a current or even permanentinability to explain the origin of a specific feature in greatdetail with regard to path or even mechanism does not posea challenge to the accepted fact that they were driven downsome evolutionary path by some evolutionary processesmdashitmerely poses a challenge to researchers to expand theirefforts in the field in the lab and in the development oftestable ideas

Concluding Remarks

The origin of complex features is quite obviously a complexsubject The precise details of how when and how manytimes a particular biological organ has evolved may never beknown with absolute certainty but the great conceptual andempirical advances made over the past 150 years haveprovided a solid understanding of the processes involvedComplex organs do not come into being fully formed butneither do they necessarily evolve through a direct series ofincremental steps each of which improves their efficacy fora single function Theirs may be a complicated historyinvolving not only direct adaptive evolution but also shifts toor additions of new functions and the juxtaposition ofcomponents that themselves may have been built up bythese same historically contingent processes The genesinvolved in the production of complex organ componentsmay be co-opted from or shared with other systems and maybe inherited in common or recruited independently indifferent lineages Finally although natural selection mayresult in impressive levels of optimization for a givenfunction it is not the only mechanism at workmdashmeaningthat complex organs no matter how intricate exhibit thesigns of having traveled a circuitous evolutionary path

Acknowledgments I am grateful to Russell Fernald Bob GregoryTrevor Lamb Todd Oakley and Joram Piatigorsky for providinghelpful comments that improved the manuscript

References

Adami C Reducible complexity Science 200631261ndash3doi101126science1126559

Altmann CR Chow RL Lang RA Hemmati-Brivanlou A Lensinduction by Pax-6 in Xenopus laevis Dev Biol 1997185119ndash23 doi101006dbio19978573

Arendt D Evolution of eyes and photoreceptor cell types Int J DevBiol 200347563ndash71

Arendt D Wittbrodt J Reconstructing the eyes of Urbilataria PhilosTrans R Soc Lond B Biol Sci 20013561545ndash63 doi101098rstb20010971

Arendt D Tessmar-Raible K Snyman H Dorresteijn AW Wittbrodt JCiliary photoreceptors with a vertebrate-type opsin in an invertebratebrain Science 2004306869ndash71 doi101126science1099955

Arnold EN Investigating the origins of performance advantageadaptation exaptation and lineage effects In Eggleton PVane-Wright RI editors Phylogenetics and ecology (Linneansociety symposia volume 17) London Academic Press 1994p 123ndash68

Ayala FJ Darwinrsquos greatest discovery design without designer ProcNatl Acad Sci U S A 20071048567ndash73 doi101073pnas0701072104

Bahar S Evolution of the eye lessons from freshman physics andRichard Dawkins Biol Physicist 200222ndash5

Battelle BA The eyes of Limulus polyphemus (Xiphosura Chelicer-ata) and their afferent and efferent projections Arthropod StructDev 200635261ndash74 doi101016jasd200607002

Bishop KL The evolution of flight in bats narrowing the field of plausiblehypotheses Q Rev Biol 200883153ndash69 doi101086587825

Bloemendal H de Jong W Jaenicke R Lubsen NH Slingsby C TardieuA Ageing and vision structure stability and function of lenscrystallins Prog BiophysMol Biol 200486407ndash85 doi101016jpbiomolbio200311012

Bock G Goode J Tinkering the microevolution of developmentHoboken Wiley 2007 Novartis Foundation Symposium 284

Bowmaker JK Hunt DM Evolution of vertebrate visual pigmentsCurr Biol 200616R484ndash9 doi101016jcub200606016

Bridgham JT Carroll SM Thornton JW Evolution of hormone-receptor complexity by molecular exploitation Science200631297ndash101 doi101126science1123348

Briscoe AD Reconstructing the ancestral butterfly eye focus on theopsins J Exp Biol 20082111805ndash13 doi101242jeb013045

Buschbeck E Friedrich M Evolution of insect eyes tales of ancientheritage deconstruction reconstruction remodeling and recy-cling Evo Edu Outreach 20081(4) doi101007s12052-008-0086-z

Cairns-Smith AG Seven clues to the origin of life Cambridge CambridgeUniversity Press 1985

Chiou TH Kleinlogel S Cronin T Caldwell R Loeffler B Siddiqi Aet al Circular polarization vision in a stomatopod crustaceanCurr Biol 200818429ndash34 doi101016jcub200802066

Chow RL Altmann CR Lang RA Hemmati-Brivanlou A Pax6 inducesectopic eyes in a vertebrate Development 19991264213ndash22

Clarkson E Levi-Setti R Horvath G The eyes of trilobites the oldestpreserved visual system Arthropod Struct Dev 200635247ndash59doi101016jasd200608002

Cleland CE Historical science experimental science and thescientific method Geology 200129987ndash90 doi1011300091-7613(2001)029lt0987HSESATgt20CO2

Cleland CE Methodological and epistemic differences betweenhistorical science and experimental science Philos Sci200269474ndash96 doi101086342455

Conn HW The method of evolution New York G P Putnamrsquos Sons1900

Cronin TW Porter ML Exceptional variation on a common themethe evolution of crustacean compound eyes Evo Edu Outreach20081(4) doi101007s12052-008-0085-0

Cuthbertson RA Tomarev SI Piatigorsky J Taxon-specific recruit-ment of enzymes as major soluble proteins in the cornealepithelium of three mammals chicken and squid Proc NatlAcad Sci USA 1992894004ndash8 doi101073pnas8994004

Cvekl A Piatigorsky J Lens development and crystallin gene expressionmany roles for Pax-6 Bioessays 199618621ndash30 doi101002bies950180805

Daeschler EB Shubin NH Jenkins FA A Devonian tetrapod-like fishand the evolution of the tetrapod body plan Nature 2006440757ndash63 doi101038nature04639

Darwin C On the origin of species by means of natural selection orthe preservation of favoured races in the struggle for life 1st edLondon Murray 1859

Darwin C The origin of species by means of natural selection or thepreservation of favoured races in the struggle for life 6th edLondon Murray 1872

Darwin C The autobiography of Charles Darwin 1809ndash1882 (Editedby Nora Barlow) London Collins 1958

Dawkins R The blind watchmaker New York Penguin 1986Dawkins R Climbing mount improbable New York WW Norton amp

Co 1996Demeacutereacute TA McGowen MR Berta A Gatesy J Morphological and

molecular evidence for a stepwise evolutionary transition fromteeth to baleen in mysticete whales Syst Biol 20085715ndash37doi10108010635150701884632

de Quieroz A Do image-forming eyes promote evolutionary diversifi-cation Evolution Int J Org Evolution 1999531654ndash64 doi1023072640429

Dial KP Wing-assisted incline running and the evolution of flightScience 2003299402ndash4 doi101126science1078237

Draper P Irreducible complexity and Darwinian gradualism a reply toMichael J Behe Faith Philos 2002193ndash21

Dulai KS von Dornum M Mollon JD Hunt DM The evolution oftrichromatic color vision by opsin gene duplication in New Worldand Old World primates Genome Res 19999629ndash38

Ebrey T Koutalos Y Vertebrate photoreceptors Prog Retin Eye Res20012049ndash94 doi101016S1350-9462(00)00014-8

Elofsson R The frontal eyes of crustaceans Arthropod Struct Dev200635275ndash91 doi101016jasd200608004

Fernald RD Eyes variety development and evolution Brain BehavEvol 2004a64141ndash7 doi101159000079743

Fernald RD Evolving eyes Int J Dev Biol 2004b48701ndash5 doi101387ijdb041888rf

Fernald RD Casting a genetic light on the evolution of eyes Science20063131914ndash8 doi101126science1127889

Friedman M The evolutionary origin of flatfish asymmetry Nature2008454209ndash12 doi101038nature07108

Gaten E Optics and phylogeny is there an insight The evolution ofsuperposition eyes in the Decapoda (Crustacea) Contrib Zool199867223ndash36

GehringWJ The genetic control of eye development and its implicationsfor the evolution of various eye types Zoology 2001104171ndash83doi1010780944-2006-00022

GehringWJ Historical perspectives on the development and evolution ofeyes and photoreceptors Int J Dev Biol 200448707ndash17 doi101387ijdb041900wg

Gehring WJ New perspectives on eye development and the evolution ofeyes and photoreceptors J Hered 200596171ndash84 doi101093jheredesi027

Gehring WJ Ikeo K Pax 6 mastering eye morphogenesis and eyeevolution Trends Genet 199915371ndash7 doi101016S0168-9525(99)01776-X

Gerl EJ Morris MR The causes and consequences of color visionEvo Edu Outreach 20081(4) doi101007s12052-008-0088-x

Goldsmith TH Optimization constraint and history in the evolutionof eyes Q Rev Biol 199065281ndash322 doi101086416840

Gould SJ The structure of evolutionary theory Cambridge HarvardUniversity Press 2002

Gould SJ Lewontin RC The spandrels of San Marco and thePanglossian paradigm a critique of the adaptationist programmeProc R Soc Lond B Biol Sci 1979205581ndash98

Gould SJ Vrba ES Exaptationmdasha missing term in the science ofform Paleobiology 198284ndash15

Gregory TR Evolutionary trends Evo Edu Outreach 2008a1259ndash73

Gregory TR Evolution as fact theory and path Evo Edu Outreach2008b146ndash52 doi101007s12052-007-0001-z

Gregory TR Understanding evolutionary trees Evo Edu Outreach2008c1121ndash37 doi101007s12052-008-0035-x

Greven H Comments on the eyes of tardigrades Arthropod StructDev 200736401ndash7 doi101016jasd200706003

Halder G Callaerts P Gehring WJ Induction of ectopic eyes bytargeted expression of the eyeless gene in Drosophila Science1995a2671788ndash92 doi101126science7892602

Halder G Callaerts P Gehring WJ New perspectives on eye evolutionCurr Opin Genet Dev 1995b5602ndash9 doi1010160959-437X(95)80029-8

Hall BK Introduction In Hall BK editor Homology the hierarchical basisof comparative biology New York Academic Press 1994 p 1ndash19

Hall BK Descent with modification the unity underlying homology andhomoplasy as seen through an analysis of development and evo-lution Biol Rev Camb Philos Soc 200378409ndash33 doi101017S1464793102006097

Hall BK Homoplasy and homology dichotomy or continuum J HumEvol 200752473ndash9 doi101016jjhevol200611010

Harris WA Pax-6 where to be conserved is not conservative ProcNatl Acad Sci USA 1997942098ndash100

Hesse R Das Sehen der niederen Tiere Jenna Fischer 1908Hisatomi O Tokunaga F Molecular evolution of proteins involved in

vertebrate phototransduction Comp Biochem Physiol B 2002133509ndash22 doi101016S1096-4959(02)00127-6

Hodin J Plasticity and constraints in development and evolution JExp Zool (Mol Dev Evol) 20002881ndash20 doi101002(SICI)1097-010X(20000415)2881lt1AID-JEZ1gt30CO2ndash7

Jacob F Evolution and tinkering Science 19771961161ndash6 doi101126science860134

Jacob F The possible and the actual Seattle University of WashingtonPress 1982

Jacob F Complexity and tinkering Ann N YAcad Sci 200192971ndash3Janvier P Squint of the fossil flatfish Nature 2008454169ndash70

doi101038454169aJia S Omelchenko M Garland D Vasiliou V Kanungo J Spencer M et

al Duplicated gelsolin family genes in zebrafish a novel scinderin-like gene (scinla) encodes the major corneal crystallin FASEB J2007213318ndash28 doi101096fj07-8172com

Jonasova K Kozmik Z Eye evolution lens and cornea as an upgradeof animal visual system Semin Cell Dev Biol 20081971ndash81doi101016jsemcdb200710005

Kawamura S Tachibanaki S Rod and cone photoreceptors molecularbasis of the difference in their physiology Comp Biochem PhysiolA 2008150369ndash77 doi101016jcbpa200804600

Kirschfeld K The resolution of lens and compound eyes In Zettler FWeiler R editors Neural principles in vision Berlin Springer1976 p 354ndash70

Kolb H How the retina works Am Sci 20039128ndash35Kozmik Z The role of Pax genes in eye evolution Brain Res Bull

200875335ndash9 doi101016jbrainresbull200710046Kozmik Z Ruzickova J Jonasova K Matsumoto Y Vopalensky P

Kozmikova I et al Assembly of the cnidarian camera-type eye fromvertebrate-like components Proc Natl Acad Sci U S A2008a1058989ndash93 doi101073pnas0800388105

Kozmik Z Swamynathan SK Ruzickova J Jonasova K Paces V Vlcek Cet al Cubozoan crystallins evidence for convergent evolution of Paxregulatory sequences Evol Dev 2008b1052ndash61

Krishnan K Kathiresan T Raman R Rajini B Dhople VM AggrawalRK et al Ubiquitous lens α- β- and γ-crystallins accumulated inanuran cornea as corneal crystallins J Biol Chem 200728218953ndash9 doi101074jbcM609275200

Kuchiiwa T Kuchiiwa S Teshirogi W Comparative morphologicalstudies on the visual systems in a binocular and multi-ocular speciesof freshwater planarian Hydrobiologia 1991227241ndash9 doi101007BF00027608

Kumar JP Signalling pathways in Drosophila and vertebrate retinaldevelopment Nat Rev Genet 20012846ndash57 doi10103835098564

Kutschera U Niklas KJ The modern theory of biological evolution anexpanded synthesis Naturwissenschaften 200491255ndash76 doi101007s00114-004-0515-y

Lamb TD Collin SP Pugh EN Evolution of the vertebrate eye opsinsphotoreceptors retina and eye cup Nat Rev Neurosci 20078960ndash75 doidoi101038nrn2283

Lamb TD Pugh EN Collin SP The origin of the vertebrate eye EvoEdu Outreach 20081(4) doi101007s12052-008-0091-2

Land MF The optics of animal eyes Contemp Phys 198829435ndash55doi10108000107518808222601

Land MF Visual acuity in insects Annu Rev Entomol 199742147ndash77 doi101146annurevento421147

Land MF Fernald RD The evolution of eyes Annu Rev Neurosci1992151ndash29 doi101146annurevne15030192000245

Land MF Nilsson DE Animal eyes Oxford Oxford University Press2002

Larusso ND Ruttenberg BE Singh AK Oakley TH Type II opsinsevolutionary origin by internal domain duplication J Mol Evol200866417ndash23 doi101007s00239-008-9076-6

Laubichler MD Tinkering a conceptual and historical evaluation InBock G Goode J editors Tinkering the microevolution ofdevelopment Chichester Wiley 2007 p 20ndash34 NovartisFoundation Symposium 284

Le Page M Evolution a guide for the not-yet perplexed New Sci200819 April25ndash33

Lynch M Simple evolutionary pathways to complex proteins ProteinSci 2005142217ndash25 doi101110ps041171805

Lynch M The evolution of genetic networks by non-adaptive processesNat Rev Genet 2007a8803ndash13 doi101038nrg2192

LynchM The frailty of adaptive hypotheses for the origins of organismalcomplexity Proc Natl Acad Sci USA 2007b104(Suppl 1)8597ndash604 doi101073pnas0702207104

Marden JH Kramer MG Surface-skimming stoneflies a possibleintermediate stage in insect flight evolution Science 1994266427ndash30 doi101126science2665184427

Marden JH Thomas MA Rowing locomotion by a stonefly that pos-sesses the ancestral pterygote condition of co-occurring wings andabdominal gills Biol J Linn Soc Lond 200379341ndash9 doi101046j1095ndash8312200300192x

Marshall J Cronin TW Kleinlogel S Stomatopod eye structure andfunction a review Arthropod Struct Dev 200736420ndash48 doi101016jasd200701006

Mayer G Structure and development of onychophoran eyes what isthe ancestral visual organ in arthropods Arthropod Struct Dev200635231ndash45 doi101016jasd200606003

McLennan DA The concept of co-option why evolution often looksmiraculous Evo Edu Outreach 20081247ndash58 doi101007s12052-008-0053-8

Miller KR Lifersquos grand design Technol Rev 19949724ndash32Mivart SG On the genesis of species London MacMillan 1871Muumlller MCM Sombke A Rosenbreg J The fine structure of the eyes

of some bristly millipedes (Penicillata Diplopoda) additionalsupport for the homology of mandibulate ommatidia ArthropodStruct Dev 200736463ndash76 doi101016jasd200709002

Newell GE The eye of Littorina littorea Proc Zool Soc Lond 196514475ndash86

Nielsen C Martinez P Patterns of gene expression homology orhomocracy Dev Genes Evol 2003213149ndash54

Nilsson DE Vision optics and evolution Bioscience 198939298ndash307 doi1023071311112

Nilsson DE Eye evolution a question of genetic promiscuity CurrOpin Neurobiol 200414407ndash14 doi101016jconb200407004

Nilsson DE Kelber A A functional analysis of compound eyeevolution Arthropod Struct Dev 200736373ndash85 doi101016jasd200707003

Nilsson DE Pelger S A pessimistic estimate of the time required foran eye to evolve Proc R Soc Lond B Biol Sci 199425653ndash8doi101098rspb19940048

Nilsson DE Gislen L Coates MM Skogh C Garm A Advancedoptics in a jellyfish eye Nature 2005435201ndash5 doi101038nature03484

Nordstroumlm K Walleacuten R Seymour J Nilsson D A simple visualsystem without neurons in jellyfish larvae Proc R Soc Lond BBiol Sci 20032702349ndash54 doi101098rspb20032504

Novella S Suboptimal optics vision problems as scars of evolution-ary history Evo Edu Outreach 20081(4) doi101007s12052-008-0092-1

Novembre J Pritchard JK Coop G Adaptive drool in the gene poolNat Genet 2007391188ndash90 doi101038ng1007-1188

Oakley TH The eye as a replicating and diverging modular devel-opmental unit Trends Ecol Evol 200318623ndash7 doi101016jtree200309005

Oakley TH Todayrsquos multiple choice exam (a) gene duplication (b)structural mutation (c) co-option (d) regulatory mutation (e) allof the above Evol Dev 20079523ndash4

Oakley TH Cunningham CW Molecular phylogenetic evidence for theindependent evolutionary origin of an arthropod compound eyeProc Natl Acad Sci U S A 2002991426ndash30 doi101073pnas032483599

Oakley TH Pankey MS Opening the ldquoblack boxrdquo the genetic andbiochemical basis of eye evolution Evo Edu Outreach 20081(4) doi101007s12052-008-0090-3

Oakley TH Oslashstman B Wilson ACV Repression and loss of geneexpression outpaces activation and gain in recently duplicated flygenes Proc Natl Acad Sci U SA 200610311637ndash41 doi101073pnas0600750103

Oakley TH Plachetzki DC Rivera AS Furcation field-splitting andthe evolutionary origins of novelty in arthropod photoreceptorsArthropod Struct Dev 200736386ndash400 doi101016jasd200708002

Ohno S Evolution by gene duplication New York Springer 1970Okano T Kojima D Fukada Y Shichida Y Yoshizawa T Primary

structures of chicken cone visual pigments vertebraterhodopsins have evolved out of cone visual pigments ProcNatl Acad Sci USA 1992895932ndash6 doi101073pnas89135932

Osorio D Eye evolution Darwinrsquos shudder stilled Trends Ecol Evol19949241ndash2 doi1010160169-5347(94)90286-0

Paley W Natural theology or evidences of the existence and attributesof the deity collected from the appearances of nature LondonGlazier 1802

Pallen MJ Matzke NJ From The origin of species to the origin ofbacterial flagella Nat Rev Microbiol 20064784ndash90 doi101038nrmicro1493

Panchen A Richard Owen and the concept of homology In Hall BKeditor Homology the hierarchical basis of comparative biologyNew York Academic 1994 p 21ndash62

Panchen A Homologymdashhistory of a concept In Bock GR CardewG editors Homology Chichester Wiley 1999 p 5ndash23Novartis Foundation Symposium 222

Perry GH Dominy NJ Claw KG Lee AS Fiegler H Redon R et alDiet and the evolution of human amylase gene copy numbervariation Nat Genet 2007391256ndash60 doi101038ng2123

Piatigorsky J Multifunctional lens crystallins and corneal enzymesmore than meets the eye Ann N Y Acad Sci 19988427ndash15doi101111j1749-66321998tb09626x

Piatigorsky J Gene sharing and evolution Cambridge HarvardUniversity Press 2007

Piatigorsky J A genetic perspective on eye evolution gene sharingconvergence and parallelism Evo Edu Outreach 20081(4)doi101007s12052-008-0077-0

Plachetzki DC Oakley TH Key transitions during the evolution of animalphototransduction novelty ldquotree-thinkingrdquo co-option and co-dupli-cation Integr Comp Biol 200747759ndash69 doi101093icbicm050

Plachetzki DC Serb JM Oakley TH New insights into the evolutionaryhistory of photoreceptor cells Trends Ecol Evol 200520465ndash7doi101016jtree200507001

Purschke G Arendt D Hausen H Muller MCM Photoreceptor cellsand eyes in Annelida Arthropod Struct Dev 200635211ndash30doi101016jasd200607005

Quiring R Waldorf U Kloter U Gehring WJ Homology of theeyeless gene of Drosophila to the Small eye gene in mice andAniridia in humans Science 1994265785ndash9 doi101126science7914031

Ratnoff OD Margolius A Hageman trait an asymptomatic disorder ofblood coagulation Trans Assoc Am Physicians 195568149ndash54

Reimann A Richter S The nauplius eye complex in lsquoconchostracansrsquo(Crustacea Branchiopoda Laevicaudata Spinicaudata Cycles-therida) and its phylogenetic implications Arthropod Struct Dev200736408ndash19 doi101016jasd200708005

Rohde K Watson NA Ultrastructure of pigmented photoreceptors oflarval Multicotyle purvisi (Trematoda Aspidogastrea) ParasitolRes 199177485ndash90 doi101007BF00928415

Rutishauser R Moline P Evo-devo and the search for homology(ldquosamenessrdquo) in biological systems Theory Biosci 2005124213ndash41

Salvini-Plawen LV Mayr E On the evolution of photoreceptors andeyes Evol Biol 197710207ndash63

Santillo S Orlando P De Petrocellis L Cristino L Guglielmotti VMusio C Evolving visual pigments hints from the opsin-basedproteins in a phylogenetically old ldquoeyelessrdquo invertebrate Biosys-tems 2006863ndash17 doi101016jbiosystems200603008

Schneider TD Evolution of biological information Nucleic AcidsRes 2000282794ndash9 doi101093nar28142794

Scott EC Matzke NJ Biological design in science classrooms Proc NatlAcad Sci U S A 20071048669ndash76 doi101073pnas0701505104

Serb JM Eernisse DJ Charting evolutionrsquos trajectory using molluscaneye diversity to understand parallel and convergent evolution EvoEdu Outreach 20081(4) doi101007s12052-008-0084-1

Shimeld SM Purkiss AG Dirks RPH Bateman OA Slingsby C LubsenNH Urochordate βγ-crystallin and the evolutionary origin of thevertebrate eye lens Curr Biol 2005151684ndash9 doi101016jcub200508046

Shubin NH Your inner fish New York Pantheon 2007Shubin NH Marshall CR Fossils genes and the origin of novelty

Paleobiology 200026(Suppl)324ndash40 doi1016660094-8373(2000)26[324FGATOO]20CO2

Shubin NH Daeschler EB Jenkins FA The pectoral fin of Tiktaalikroseae and the origin of the tetrapod limb Nature 2006440764ndash71 doi101038nature04637

Spady TC Parry JWL Robinson PR Hunt DM Bowmaker JKCarleton KL Evolution of the cichlid visual palette throughontogenetic subfunctionalization of the opsin gene arrays MolBiol Evol 2006231538ndash47 doi101093molbevmsl014

Spudich JL Yang CS Jung KH Spudich EN Retinylidene proteinsstructures and functions from Archaea to humans Annu Rev CellDev Biol 200016365ndash95 doi101146annurevcellbio161365

Sweeney AM Des Marais DL Ban YEA Johnsen S Evolution ofgraded refractive index in squid lenses J R Soc Interface20074685ndash98 doi101098rsif20060210

Taylor JS Raes J Duplication and divergence the evolution of newgenes and old ideas Annu Rev Genet 200438615ndash43 doi101146annurevgenet38072902092831

Terakita A The opsins Genome Biol 20056213 doi101186gb-2005-6-3-213

Thanukos A Bringing homologies into focus Evo Edu Outreach20081(4) doi101007s12052-008-0080-5

Thornhill RH Ussery DW A classification of possible routes ofDarwinian evolution J Theor Biol 2000203111ndash6 doi101006jtbi20001070

Tomarev SI Piatigorsky J Lens crystallins of invertebrates diversity andrecruitment from detoxification enzymes and novel proteins Eur JBiochem 1996235449ndash65 doi101111j1432-1033199600449x

Treisman JE How to make an eye Development 20041313823ndash7doi101242dev01319

True JR Carroll SB Gene co-option in physiological and morphologicalevolution Annu Rev Cell Dev Biol 20021853ndash80 doi101146annurevcellbio18020402140619

Van Gelder RN Non-visual photoreception sensing light withoutsight Curr Biol 200718R38ndash9 doi101016jcub200711027

Wagner GP The developmental genetics of homology Nat Rev Genet20078473ndash9 doi101038nrg2099

Ward A Liu J Feng Z Xu XZS Light-sensitive neurons and channelsmediate phototaxis in C elegans Nat Neurosci 200811916ndash22doi101038nn2155

Wehner R Brainless eyes Nature 2005435157ndash8 doi101038435157aYoung JZ A model of the brain Oxford Oxford University Press 1964Zaidi FH Hull JT Peirson SN Wulff K Aeschbach D Gooley JJ et

al Short-wavelength light sensitivity of circadian pupillary andvisual awareness in humans lacking an outer retina Curr Biol2007172122ndash8 doi101016jcub200711034

Zhou L Mitra R Atkinson PW Burgess Hickman A Dyda F CraigNL Transposition of hAT elements links transposable elementsand V(D)J recombination Nature 2004432995ndash1001 doi101038nature03157

Zimmer C The evolution of extraordinary eyes the cases of floundersand stalk-eyed flies Evo Edu Outreach 20081(4) doi101007s12052-008-0089-9

Online Resources

In the Light of Evolution I Adaptation and Complex Design (ArthurM Sackler Colloquium of the National Academy of Sciences)

httpwwwnasonlineorgadaptation_and_complex_designUnderstanding Evolution

httpevolutionberkeleyeduevolibraryarticleeyes_01httpevolutionberkeleyeduevolibraryarticle_0_0evo_53httpevolutionberkeleyeduevositeevo101IIIENaturalSelectionshtml

Evolution 24 Myths and Misconceptions (New Scientist)h t t p wwwnewsc i en t i s t c om channe l l i f e dn13620DCMP = NLC-nletterampnsref = dn13620

PBS Evolutionhttpwwwpbsorgwgbhevolutionlibrary011l_011_01html

Animation of vertebrate eye development (Lamb et al 2007)httpwwwnaturecomnrnjournalv8n12suppinfonrn2283html

Videos of eyeless nematode detecting light (Ward et al 2008)ht tp wwwnaturecomneuro journal v11n8suppinfonn2155_S1html

Abstract

Introduction



Indirect Evolution

Exaptation


Gene Sharing


Scaffolding

Non-adaptation Constraints Trade-offs and Historical Contingency




Photopigments and Photoreceptors

Lenses

Corneas




Concluding Remarks

References

ltlt ASCII85EncodePages false AllowTransparency false AutoPositionEPSFiles true AutoRotatePages None Binding Left CalGrayProfile (None) CalRGBProfile (sRGB IEC61966-21) CalCMYKProfile (ISO Coated v2 300 050ECI051) sRGBProfile (sRGB IEC61966-21) CannotEmbedFontPolicy Error CompatibilityLevel 13 CompressObjects Off CompressPages true ConvertImagesToIndexed true PassThroughJPEGImages true CreateJDFFile false CreateJobTicket false DefaultRenderingIntent Perceptual DetectBlends true ColorConversionStrategy sRGB DoThumbnails true EmbedAllFonts true EmbedJobOptions true DSCReportingLevel 0 EmitDSCWarnings false EndPage -1 ImageMemory 524288 LockDistillerParams true MaxSubsetPct 100 Optimize true OPM 1 ParseDSCComments true ParseDSCCommentsForDocInfo true PreserveCopyPage true PreserveEPSInfo true PreserveHalftoneInfo false PreserveOPIComments false PreserveOverprintSettings true StartPage 1 SubsetFonts false TransferFunctionInfo Apply UCRandBGInfo Preserve UsePrologue false ColorSettingsFile () AlwaysEmbed [ true ] NeverEmbed [ true ] AntiAliasColorImages false DownsampleColorImages true ColorImageDownsampleType Bicubic ColorImageResolution 150 ColorImageDepth -1 ColorImageDownsampleThreshold 150000 EncodeColorImages true ColorImageFilter DCTEncode AutoFilterColorImages false ColorImageAutoFilterStrategy JPEG ColorACSImageDict ltlt QFactor 076 HSamples [2 1 1 2] VSamples [2 1 1 2] gtgt ColorImageDict ltlt QFactor 076 HSamples [2 1 1 2] VSamples [2 1 1 2] gtgt JPEG2000ColorACSImageDict ltlt TileWidth 256 TileHeight 256 Quality 30 gtgt JPEG2000ColorImageDict ltlt TileWidth 256 TileHeight 256 Quality 30 gtgt AntiAliasGrayImages false DownsampleGrayImages true GrayImageDownsampleType Bicubic GrayImageResolution 150 GrayImageDepth -1 GrayImageDownsampleThreshold 150000 EncodeGrayImages true GrayImageFilter DCTEncode AutoFilterGrayImages true GrayImageAutoFilterStrategy JPEG GrayACSImageDict ltlt QFactor 076 HSamples [2 1 1 2] VSamples [2 1 1 2] gtgt GrayImageDict ltlt QFactor 015 HSamples [1 1 1 1] VSamples [1 1 1 1] gtgt JPEG2000GrayACSImageDict ltlt TileWidth 256 TileHeight 256 Quality 30 gtgt JPEG2000GrayImageDict ltlt TileWidth 256 TileHeight 256 Quality 30 gtgt AntiAliasMonoImages false DownsampleMonoImages true MonoImageDownsampleType Bicubic MonoImageResolution 600 MonoImageDepth -1 MonoImageDownsampleThreshold 150000 EncodeMonoImages true MonoImageFilter CCITTFaxEncode MonoImageDict ltlt K -1 gtgt AllowPSXObjects false PDFX1aCheck false PDFX3Check false PDFXCompliantPDFOnly false PDFXNoTrimBoxError true PDFXTrimBoxToMediaBoxOffset [ 000000 000000 000000 000000 ] PDFXSetBleedBoxToMediaBox true PDFXBleedBoxToTrimBoxOffset [ 000000 000000 000000 000000 ] PDFXOutputIntentProfile (None) PDFXOutputCondition () PDFXRegistryName (httpwwwcolororg) PDFXTrapped False SyntheticBoldness 1000000 Description ltlt ENU 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 DEU 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 gtgtgtgt setdistillerparamsltlt HWResolution [2400 2400] PageSize [5952756 8418897]gtgt setpagedevice

computer science and other disciplines Means of quantifyingcomplexity have been developed in evolutionary biology aswell as have methods for assessing trends involving changesin complexity through deep time (for review see Gregory2008a) This paper is not about complexity per se butabout the evolution of complex organs biological structureswith several intricately interacting parts that function in asophisticated manner the eye being the primary example dis-cussed Complexity as defined in this intuitive sense is notrestricted to organsmdashit also applies to biochemical subcellu-lar behavioral and other biological systems On the otherhand not all functional systems are complex nor is com-plexity inherently advantageous or inevitable as efficiency canreadily trump a convoluted arrangement

Fundamentally evolutionary explanations for the originof biological features complex or otherwise are based on theassumption of continuity That is to say there is an unbrokenchain of ancestry and descent linking modern organisms withearlier species that lived long ago In order to account for theexistence of complex organs as they are observed today it isnecessary to provide an account of how these could havearisen without breaks or inexplicable leaps from lesscomplex antecedents The following sections outline variousprocesses that taken together are considered by evolution-ary biologists to meet this requirement

Darwin famously noted in 1859 (p189) that ldquoif it could bedemonstrated that any complex organ existed which couldnot possibly have been formed by numerous successive slightmodifications my theory would absolutely break downrdquo Itshould be understood that the ldquotheoryrdquo in question is not thenotion that species are related by descent rather it is theproposal that the mechanism responsible for evolution isnatural selection acting gradually on minor heritable differ-ences (see Gregory 2008b) There is absolutely no scientificdisagreement as to whether natural selection occurs as it canbe observed in both experimental and natural populations thequestion is whether this process alone can result in theemergence of complex organs such as eyes

The process of gradual stepwise adaptive change empha-sized (though not exclusively so) by Darwin has been calledldquodirect evolutionrdquo and can be further subdivided into twomajor types (Thornhill and Ussery 2000)

1 Serial direct evolution The simplest form of adaptiveevolution which proceeds step by step from A1 rarr A2 rarrA3 rarr A4 and involves gradual change along a singleaxis (ie each step serves the same function but ef-fectiveness increases from A1 to A4) A second series ofchanges may take place in addition to the first but inthis scenario it would be only after the first series ofchanges ended

2 Parallel direct evolution A slightly more complicatedform of direct adaptation in which changes occur at thesame time in more than one component as in A1B1 rarrA2B2 rarr A3B3 rarr A4B4 In this case no single com-ponent becomes greatly modified before the others do

Under serial direct evolution each change that occurs issmall involves only one component of a particular systemand in principle is reversible As a result serial directevolution does not produce organs with indivisibly integratedcomponents Parallel direct evolution on the other hand canproduce a moderate interdependence of parts because thesechange in concert with one another Neither serial nor paralleldirect evolution necessarily leads to an increase in complexityand in terms of highly complex organs the most importantcontribution made by direct adaptive evolution probablyrelates to refinements of particular functions through themodification of a few components Thus direct evolution byitself is not sufficient to properly account for the evolution ofhighly complex and integrated organs and as such additionalprocesses are necessary

Indirect Evolution

Direct adaptive evolution by natural selection has since it wasfirst proposed been subject to criticism by those who disagreethat it is sufficient to explain the origin of complex biologicalfeatures In Darwinrsquos own time opponents began listingfeatures of organisms for which incipient or intermediatestages seem unlikely to have been functional and which there-fore could not have been be shaped incrementally by naturalselection (eg Mivart 1871) Darwin (1872) responded bypointing to examples of organs of ldquointermediaterdquo complexitythat did or indeed still do exist in other species1

Nevertheless there are legitimate reasons to expect directgradual evolution along a single axis to be incapable of pro-ducing complex adaptations composed of tightly interactingparts As a result evolutionary biologists (including Darwin)have long pointed out the importance of indirect routes bywhich complex organs and systems can evolve

A recent analysis of the evolution of two hormone-receptorpairs provides an illustration of the basic concept of indirectevolution (Adami 2006 Bridgham et al 2006) The closematch between a hormone and the receptor to which it bindshas been considered analogous to that between a ldquolock andkeyrdquo both of which are required for the system to function

1 Interestingly Mivart (1871) cited among his examples of featuresthat he presumed would be very unlikely to evolve gradually theappearance of two eyes on one side of the head among flounders andthe feeding structures of baleen whalesmdashboth of which have beendiscussed recently in light of evidence indicating that intermediateforms did indeed occur (Demeacutereacute et al 2008 Friedman 2008 Janvier2008 Zimmer 2008)

Exaptation

Gene Sharing

Scaffolding

Lenses7

Turtle

Liz ard

Crocodile

BirdKangaroo Human

Guinea pig

Rabbit

Camel Octopus

Squid Fly Jellyfish

Embryonic

α enolase

Corneas

Concluding Remarks

References

Online Resources

Abstract

Introduction



Indirect Evolution

Exaptation


Gene Sharing


Scaffolding






Lenses

Corneas




Concluding Remarks

References


Exaptation

Gene Sharing

Scaffolding

Lenses7

Turtle

Liz ard

Crocodile

BirdKangaroo Human

Guinea pig

Rabbit

Camel Octopus

Squid Fly Jellyfish

Embryonic

α enolase

Corneas

Concluding Remarks

References

Online Resources

Abstract

Introduction



Indirect Evolution

Exaptation


Gene Sharing


Scaffolding






Lenses

Corneas




Concluding Remarks

References


Exaptation

Gene Sharing

Scaffolding

Lenses7

Turtle

Liz ard

Crocodile

BirdKangaroo Human

Guinea pig

Rabbit

Camel Octopus

Squid Fly Jellyfish

Embryonic

α enolase

Corneas

Concluding Remarks

References

Online Resources

Abstract

Introduction



Indirect Evolution

Exaptation


Gene Sharing


Scaffolding






Lenses

Corneas




Concluding Remarks

References


Gene Sharing

Scaffolding

Lenses7

Turtle

Liz ard

Crocodile

BirdKangaroo Human

Guinea pig

Rabbit

Camel Octopus

Squid Fly Jellyfish

Embryonic

α enolase

Corneas

Concluding Remarks

References

Online Resources

Abstract

Introduction



Indirect Evolution

Exaptation


Gene Sharing


Scaffolding






Lenses

Corneas




Concluding Remarks

References


Gene Sharing

Scaffolding

Lenses7

Turtle

Liz ard

Crocodile

BirdKangaroo Human

Guinea pig

Rabbit

Camel Octopus

Squid Fly Jellyfish

Embryonic

α enolase

Corneas

Concluding Remarks

References

Online Resources

Abstract

Introduction



Indirect Evolution

Exaptation


Gene Sharing


Scaffolding






Lenses

Corneas




Concluding Remarks

References


Scaffolding

Lenses7

Turtle

Liz ard

Crocodile

BirdKangaroo Human

Guinea pig

Rabbit

Camel Octopus

Squid Fly Jellyfish

Embryonic

α enolase

Corneas

Concluding Remarks

References

Online Resources

Abstract

Introduction



Indirect Evolution

Exaptation


Gene Sharing


Scaffolding






Lenses

Corneas




Concluding Remarks

References


Scaffolding

Lenses7

Turtle

Liz ard

Crocodile

BirdKangaroo Human

Guinea pig

Rabbit

Camel Octopus

Squid Fly Jellyfish

Embryonic

α enolase

Corneas

Concluding Remarks

References

Online Resources

Abstract

Introduction



Indirect Evolution

Exaptation


Gene Sharing


Scaffolding






Lenses

Corneas




Concluding Remarks

References


Lenses7

Turtle

Liz ard

Crocodile

BirdKangaroo Human

Guinea pig

Rabbit

Camel Octopus

Squid Fly Jellyfish

Embryonic

α enolase

Corneas

Concluding Remarks

References

Online Resources

Abstract

Introduction



Indirect Evolution

Exaptation


Gene Sharing


Scaffolding






Lenses

Corneas




Concluding Remarks

References


Lenses7

Turtle

Liz ard

Crocodile

BirdKangaroo Human

Guinea pig

Rabbit

Camel Octopus

Squid Fly Jellyfish

Embryonic

α enolase

Corneas

Concluding Remarks

References

Online Resources

Abstract

Introduction



Indirect Evolution

Exaptation


Gene Sharing


Scaffolding






Lenses

Corneas




Concluding Remarks

References


Lenses7

Turtle

Liz ard

Crocodile

BirdKangaroo Human

Guinea pig

Rabbit

Camel Octopus

Squid Fly Jellyfish

Embryonic

α enolase

Corneas

Concluding Remarks

References

Online Resources

Abstract

Introduction



Indirect Evolution

Exaptation


Gene Sharing


Scaffolding






Lenses

Corneas




Concluding Remarks

References


Lenses7

Turtle

Liz ard

Crocodile

BirdKangaroo Human

Guinea pig

Rabbit

Camel Octopus

Squid Fly Jellyfish

Embryonic

α enolase

Corneas

Concluding Remarks

References

Online Resources

Abstract

Introduction



Indirect Evolution

Exaptation


Gene Sharing


Scaffolding






Lenses

Corneas




Concluding Remarks

References


Lenses7

Turtle

Liz ard

Crocodile

BirdKangaroo Human

Guinea pig

Rabbit

Camel Octopus

Squid Fly Jellyfish

Embryonic

α enolase

Corneas

Concluding Remarks

References

Online Resources

Abstract

Introduction



Indirect Evolution

Exaptation


Gene Sharing


Scaffolding






Lenses

Corneas




Concluding Remarks

References


Lenses7

Turtle

Liz ard

Crocodile

BirdKangaroo Human

Guinea pig

Rabbit

Camel Octopus

Squid Fly Jellyfish

Embryonic

α enolase

Corneas

Concluding Remarks

References

Online Resources

Abstract

Introduction



Indirect Evolution

Exaptation


Gene Sharing


Scaffolding






Lenses

Corneas




Concluding Remarks

References


Lenses7

Turtle

Liz ard

Crocodile

BirdKangaroo Human

Guinea pig

Rabbit

Camel Octopus

Squid Fly Jellyfish

Embryonic

α enolase

Corneas

Concluding Remarks

References

Online Resources

Abstract

Introduction



Indirect Evolution

Exaptation


Gene Sharing


Scaffolding






Lenses

Corneas




Concluding Remarks

References


Lenses7

Turtle

Liz ard

Crocodile

BirdKangaroo Human

Guinea pig

Rabbit

Camel Octopus

Squid Fly Jellyfish

Embryonic

α enolase

Corneas

Concluding Remarks

References

Online Resources

Abstract

Introduction



Indirect Evolution

Exaptation


Gene Sharing


Scaffolding






Lenses

Corneas




Concluding Remarks

References


Lenses7

Turtle

Liz ard

Crocodile

BirdKangaroo Human

Guinea pig

Rabbit

Camel Octopus

Squid Fly Jellyfish

Embryonic

α enolase

Corneas

Concluding Remarks

References

Online Resources

Abstract

Introduction



Indirect Evolution

Exaptation


Gene Sharing


Scaffolding






Lenses

Corneas




Concluding Remarks

References


Lenses7

Turtle

Liz ard

Crocodile

BirdKangaroo Human

Guinea pig

Rabbit

Camel Octopus

Squid Fly Jellyfish

Embryonic

α enolase

Corneas

Concluding Remarks

References

Online Resources

Abstract

Introduction



Indirect Evolution

Exaptation


Gene Sharing


Scaffolding






Lenses

Corneas




Concluding Remarks

References


Lenses7

Turtle

Liz ard

Crocodile

BirdKangaroo Human

Guinea pig

Rabbit

Camel Octopus

Squid Fly Jellyfish

Embryonic

α enolase

Corneas

Concluding Remarks

References

Online Resources

Abstract

Introduction



Indirect Evolution

Exaptation


Gene Sharing


Scaffolding






Lenses

Corneas




Concluding Remarks

References


Turtle

Liz ard

Crocodile

BirdKangaroo Human

Guinea pig

Rabbit

Camel Octopus

Squid Fly Jellyfish

Embryonic

α enolase

Corneas

Concluding Remarks

References

Online Resources

Abstract

Introduction



Indirect Evolution

Exaptation


Gene Sharing


Scaffolding






Lenses

Corneas




Concluding Remarks

References


Turtle

Liz ard

Crocodile

BirdKangaroo Human

Guinea pig

Rabbit

Camel Octopus

Squid Fly Jellyfish

Embryonic

α enolase

Corneas

Concluding Remarks

References

Online Resources

Abstract

Introduction



Indirect Evolution

Exaptation


Gene Sharing


Scaffolding






Lenses

Corneas




Concluding Remarks

References


Corneas

Concluding Remarks

References

Online Resources

Abstract

Introduction



Indirect Evolution

Exaptation


Gene Sharing


Scaffolding






Lenses

Corneas




Concluding Remarks

References


Concluding Remarks

References

Online Resources

Abstract

Introduction



Indirect Evolution

Exaptation


Gene Sharing


Scaffolding






Lenses

Corneas




Concluding Remarks

References


Concluding Remarks

References

Online Resources

Abstract

Introduction



Indirect Evolution

Exaptation


Gene Sharing


Scaffolding






Lenses

Corneas




Concluding Remarks

References


Concluding Remarks

References

Online Resources

Abstract

Introduction



Indirect Evolution

Exaptation


Gene Sharing


Scaffolding






Lenses

Corneas




Concluding Remarks

References


Concluding Remarks

References

Online Resources

Abstract

Introduction



Indirect Evolution

Exaptation


Gene Sharing


Scaffolding






Lenses

Corneas




Concluding Remarks

References


Concluding Remarks

References

Online Resources

Abstract

Introduction



Indirect Evolution

Exaptation


Gene Sharing


Scaffolding






Lenses

Corneas




Concluding Remarks

References


Concluding Remarks

References

Online Resources

Abstract

Introduction



Indirect Evolution

Exaptation


Gene Sharing


Scaffolding






Lenses

Corneas




Concluding Remarks

References


Online Resources

Abstract

Introduction



Indirect Evolution

Exaptation


Gene Sharing


Scaffolding






Lenses

Corneas




Concluding Remarks

References


Online Resources

Abstract

Introduction



Indirect Evolution

Exaptation


Gene Sharing


Scaffolding






Lenses

Corneas




Concluding Remarks

References


Online Resources

Abstract

Introduction



Indirect Evolution

Exaptation


Gene Sharing


Scaffolding






Lenses

Corneas




Concluding Remarks

References


Online Resources

Abstract

Introduction



Indirect Evolution

Exaptation


Gene Sharing


Scaffolding






Lenses

Corneas




Concluding Remarks

References


Documents

The Evolution of Complex Organs