ThespeculationsofDeMailletinthebeginningoftheeighteenthcenturyturnuponfossils;andBuffonfollowshimverycloselyinthosetworemarkableworks，the\"TheoriedelaTerre\"andthe\"EpoquesdelaNature\"withwhichhecommencedandendedhiscareerasanaturalist。
Theopeningsentencesofthe\"EpoquesdelaNature\"showushowfullyBuffonrecognisedtheanalogyofgeologicalwitharchaeologicalinquiries。\"Asincivilhistoryweconsultdeeds，seekforcoins，ordecipherantiqueinscriptionsinordertodeterminetheepochsofhumanrevolutionsandfixthedateofmoralevents;so，innaturalhistory，wemustsearchthearchivesoftheworld，recoveroldmonumentsfromthebowelsoftheearth，collecttheirfragmentaryremains，andgatherintoonebodyofevidenceallthesignsofphysicalchangewhichmayenableustolookbackuponthedifferentagesofnature。Itisouronlymeansoffixingsomepointsintheimmensityofspace，andofsettingacertainnumberofwaymarksalongtheeternalpathoftime。\"
Buffonenumeratesfiveclassesofthesemonumentsofthepasthistoryoftheearth，andtheyareallfactsofpalaeontology。
Inthefirstplace，hesays，shellsandothermarineproductionsarefoundalloverthesurfaceandintheinteriorofthedryland;andallcalcareousrocksaremadeupoftheirremains。
Secondly，agreatmanyoftheseshellswhicharefoundinEuropearenotnowtobemetwithintheadjacentseas;and，intheslatesandotherdeep—seateddeposits，thereareremainsoffishesandofplantsofwhichnospeciesnowexistinourlatitudes，andwhichareeitherextinct，orexistonlyinmorenorthernclimates。Thirdly，inSiberiaandinothernorthernregionsofEuropeandofAsia，bonesandteethofelephants，rhinoceroses，andhippopotamusesoccurinsuchnumbersthattheseanimalsmustoncehavelivedandmultipliedinthoseregions，althoughatthepresentdaytheyareconfinedtosouthernclimates。Thedepositsinwhichtheseremainsarefoundaresuperficial，whilethosewhichcontainshellsandothermarineremainsliemuchdeeper。Fourthly，tusksandbonesofelephantsandhippopotamusesarefoundnotonlyinthenorthernregionsoftheoldworld，butalsointhoseofthenewworld，although，atpresent，neitherelephantsnorhippopotamusesoccurinAmerica。Fifthly，inthemiddleofthecontinents，inregionsmostremotefromthesea，wefindaninfinitenumberofshells，ofwhichthemostpartbelongtoanimalsofthosekindswhichstillexistinsouthernseas，butofwhichmanyothershavenolivinganalogues;sothatthesespeciesappeartobelost，destroyedbysomeunknowncause。Itisneedlesstoinquirehowfarthesestatementsarestrictlyaccurate;theyaresufficientlysotojustifyBuffon’sconclusionsthatthedrylandwasoncebeneaththesea;thattheformationofthefossiliferousrocksmusthaveoccupiedavastlygreaterlapseoftimethanthattraditionallyascribedtotheageoftheearth;
thatfossilremainsindicatedifferentclimatalconditionstohaveobtainedinformertimes，andespeciallythatthepolarregionswereoncewarmer;thatmanyspeciesofanimalsandplantshavebecomeextinct;andthatgeologicalchangehashadsomethingtodowithgeographicaldistribution。
Butthesepropositionsalmostconstitutetheframe—workofpalaeontology。Inordertocompleteitbutoneadditionwasneeded，andthatwasmade，inthelastyearsoftheeighteenthcentury，byWilliamSmith，whoseworkcomessonearourowntimesthatmanylivingmenmayhavebeenpersonallyacquaintedwithhim。Thismodestland—surveyor，whosebusinesstookhimintomanypartsofEngland，profitedbythepeculiarlyfavourableconditionsofferedbythearrangementofoursecondarystratatomakeacarefulexaminationandcomparisonoftheirfossilcontentsatdifferentpointsofthelargeareaoverwhichtheyextend。Theresultofhisaccurateandwidely—
extendedobservationswastoestablishtheimportanttruththateachstratumcontainscertainfossilswhicharepeculiartoit;
andthattheorderinwhichthestrata，characterisedbythesefossils，aresuper—imposedoneupontheotherisalwaysthesame。Thismostimportantgeneralisationwasrapidlyverifiedandextendedtoallpartsoftheworldaccessibletogeologists;
andnowitrestsuponsuchanimmensemassofobservationsastobeoneofthebestestablishedtruthsofnaturalscience。Tothegeologistthediscoverywasofinfiniteimportanceasitenabledhimtoidentifyrocksofthesamerelativeage，howevertheircontinuitymightbeinterruptedortheircompositionaltered。
Buttothebiologistithadastilldeepermeaning，foritdemonstratedthat，throughouttheprodigiousdurationoftimeregisteredbythefossiliferousrocks，thelivingpopulationoftheearthhadundergonecontinualchanges，notmerelybytheextinctionofacertainnumberofthespecieswhichhadatfirstexisted，butbythecontinualgenerationofnewspecies，andthenolessconstantextinctionofoldones。
Thusthebroadoutlinesofpalaeontology，insofarasitisthecommonpropertyofboththegeologistandthebiologist，weremarkedoutatthecloseofthelastcentury。IntracingitssubsequentprogressImustconfinemyselftotheprovinceofbiology，and，indeed，totheinfluenceofpalaeontologyuponzoologicalmorphology。AndIacceptthislimitationthemorewillinglyasthenolessimportanttopicofthebearingofgeologyandofpalaeontologyupondistributionhasbeenluminouslytreatedintheaddressofthePresidentoftheGeographicalSection。<3>
Thesuccessionofthespeciesofanimalsandplantsintimebeingestablished，thefirstquestionwhichthezoologistorthebotanisthadtoaskhimselfwas，Whatistherelationofthesesuccessivespeciesonetoanother？AnditisacuriouscircumstancethatthemostimportanteventinthehistoryofpalaeontologywhichimmediatelysucceededWilliamSmith’sgeneralisationwasadiscoverywhich，couldithavebeenrightlyappreciatedatthetime，wouldhavegonefartowardssuggestingtheanswer，whichwasinfactdelayedformorethanhalfacentury。IrefertoCuvier’sinvestigationofthemammalianfossilsyieldedbythequarriesintheoldertertiaryrocksofMontmartre，amongthechiefresultsofwhichwasthebringingtolightoftwogeneraofextincthoofedquadrupeds，theAnoplotheriumandthePalaeotherium。TherichmaterialsatCuvier’sdispositionenabledhimtoobtainafullknowledgeoftheosteologyandofthedentitionofthesetwoforms，andconsequentlytocomparetheirstructurecriticallywiththatofexistinghoofedanimals。TheeffectofthiscomparisonwastoprovethattheAnoplotherium，thoughitpresentedmanypointsofresemblancewiththepigsontheonehandandwiththeruminantsontheother，differedfrombothtosuchanextentthatitcouldfindaplaceinneithergroup。
Infact，itheld，insomerespects，anintermediateposition，tendingtobridgeovertheintervalbetweenthesetwogroups，whichintheexistingfaunaaresodistinct。Inthesameway，thePalaeotheriumtendedtoconnectformssodifferentasthetapir，therhinoceros，andthehorse。Subsequentinvestigationshavebroughttolightavarietyoffactsofthesameorder，themostcuriousandstrikingofwhicharethosewhichprovetheexistence，inthemesozoicepoch，ofaseriesofformsintermediatebetweenbirdsandreptiles——twoclassesofvertebrateanimalswhichatpresentappeartobemorewidelyseparatedthananyothers。Yettheintervalbetweenthemiscompletelyfilled，inthemesozoicfauna，bybirdswhichhavereptiliancharacters，ontheoneside，andreptileswhichhaveornithiccharacters，ontheother。Soagain，whilethegroupoffishes，termedganoids，is，atthepresenttime，sodistinctfromthatofthedipnoi，ormudfishes，thattheyhavebeenreckonedasdistinctorders，theDevonianstratapresentuswithformsofwhichitisimpossibletosaywithcertaintywhethertheyaredipnoiorwhethertheyareganoids。
Agassiz’slongandelaborateresearchesuponfossilfishes，publishedbetween1833and1842，ledhimtosuggesttheexistenceofanotherkindofrelationbetweenancientandmodernformsoflife。Heobservedthattheoldestfishespresentmanycharacterswhichrecalltheembryonicconditionsofexistingfishes;andthat，notonlyamongfishes，butinseveralgroupsoftheinvertebratawhichhavealongpalaeontologicalhistory，thelatestformsaremoremodified，morespecialised，thantheearlier。Thefactthatthedentitionoftheoldertertiaryungulateandcarnivorousmammalsisalwayscomplete，noticedbyProfessorOwen，illustratedthesamegeneralisation。
AnothernolesssuggestiveobservationwasmadebyMr。Darwin，whosepersonalinvestigationsduringthevoyageoftheBeagleledhimtoremarkuponthesingularfact，thatthefauna，whichimmediatelyprecedesthatatpresentexistinginanygeographicalprovinceofdistribution，presentsthesamepeculiaritiesasitssuccessor。Thus，inSouthAmericaandinAustralia，thelatertertiaryorquaternaryfossilsshowthatthefaunawhichimmediatelyprecededthatofthepresentdaywas，intheonecase，asmuchcharacterisedbyedentatesand，intheother，bymarsupialsasitisnow，althoughthespeciesoftheolderarelargelydifferentfromthoseofthenewerfauna。
Howeverclearlytheseindicationsmightpointinonedirection，thequestionoftheexactrelationofthesuccessiveformsofanimalandvegetablelifecouldbesatisfactorilysettledonlyinoneway;namely，bycomparing，stagebystage，theseriesofformspresentedbyoneandthesametypethroughoutalongspaceoftime。Withinthelastfewyearsthishasbeendonefullyinthecaseofthehorse，lesscompletelyinthecaseoftheotherprincipaltypesoftheungulataandofthecarnivora;
andalltheseinvestigationstendtoonegeneralresult，namely，that，inanygivenseries，thesuccessivemembersofthatseriespresentagraduallyincreasingspecialisationofstructure。
Thatistosay，ifanysuchmammalatpresentexistinghasspeciallymodifiedandreducedlimbsordentitionandcomplicatedbrain，itspredecessorsintimeshowlessandlessmodificationandreductioninlimbsandteethandalesshighlydevelopedbrain。ThelaboursofGaudry，Marsh，andCopefurnishabundantillustrationsofthislawfromthemarvellousfossilwealthofPikermiandthevastuninterruptedseriesoftertiaryrocksintheterritoriesofNorthAmerica。
Iwillnowsumuptheresultsofthissketchoftheriseandprogressofpalaeontology。Thewholefabricofpalaeontologyisbasedupontwopropositions:thefirstis，thatfossilsaretheremainsofanimalsandplants;andthesecondis，thatthestratifiedrocksinwhichtheyarefoundaresedimentarydeposits;andeachofthesepropositionsisfoundeduponthesameaxiom，thatlikeeffectsimplylikecauses。Ifthereisanycausecompetenttoproduceafossilstem，orshell，orbone，exceptalivingbeing，thenpalaeontologyhasnofoundation;
ifthestratificationoftherocksisnottheeffectofsuchcausesasatpresentproducestratification，wehavenomeansofjudgingofthedurationofpasttime，oroftheorderinwhichtheformsoflifehavesucceededoneanother。Butifthesetwopropositionsaregranted，thereisnoescape，asitappearstome，fromthreeveryimportantconclusions。Thefirstisthatlivingmatterhasexistedupontheearthforavastlengthoftime，certainlyformillionsofyears。Thesecondisthat，duringthislapseoftime，theformsoflivingmatterhaveundergonerepeatedchanges，theeffectofwhichhasbeenthattheanimalandvegetablepopulation，atanyperiodoftheearth’shistory，containscertainspecieswhichdidnotexistatsomeantecedentperiod，andotherswhichceasedtoexistatsomesubsequentperiod。Thethirdisthat，inthecaseofmanygroupsofmammalsandsomeofreptiles，inwhichonetypecanbefollowedthroughaconsiderableextentofgeologicaltime，theseriesofdifferentformsbywhichthetypeisrepresented，atsuccessiveintervalsofthistime，isexactlysuchasitwouldbe，iftheyhadbeenproducedbythegradualmodificationoftheearliestformsoftheseries。Thesearefactsofthehistoryoftheearthguaranteedbyasgoodevidenceasanyfactsincivilhistory。
HithertoIhavekeptcarefullyclearofallthehypothesestowhichmenhaveatvarioustimesendeavouredtofitthefactsofpalaeontology，orbywhichtheyhaveendeavouredtoconnectasmanyofthesefactsastheyhappenedtobeacquaintedwith。
Idonotthinkitwouldbeaprofitableemploymentofourtimetodiscussconceptionswhichdoubtlesshavehadtheirjustificationandeventheiruse，butwhicharenowobviouslyincompatiblewiththewell—ascertainedtruthsofpalaeontology。
Atpresentthesetruthsleaveroomforonlytwohypotheses。
Thefirstisthat，inthecourseofthehistoryoftheearth，innumerablespeciesofanimalsandplantshavecomeintoexistence，independentlyofoneanother，innumerabletimes。
This，ofcourse，implieseitherthatspontaneousgenerationonthemostastoundingscale，andofanimalssuchashorsesandelephants，hasbeengoingon，asanaturalprocess，throughallthetimerecordedbythefossiliferousrocks;oritnecessitatesthebeliefininnumerableactsofcreationrepeatedinnumerabletimes。Theotherhypothesisis，thatthesuccessivespeciesofanimalsandplantshavearisen，thelaterbythegradualmodificationoftheearlier。Thisisthehypothesisofevolution;andthepalaeontologicaldiscoveriesofthelastdecadearesocompletelyinaccordancewiththerequirementsofthishypothesisthat，ifithadnotexisted，thepalaeontologistwouldhavehadtoinventit。
Ihavealwayshadacertainhorrorofpresumingtosetalimituponthepossibilitiesofthings。ThereforeIwillnotventuretosaythatitisimpossiblethatthemultitudinousspeciesofanimalsandplantsmayhavebeenproduced，oneseparatelyfromtheother，byspontaneousgeneration;northatitisimpossiblethattheyshouldhavebeenindependentlyoriginatedbyanendlesssuccessionofmiraculouscreativeacts。ButImustconfessthatboththesehypothesesstrikemeassoastoundinglyimprobable，sodevoidofashredofeitherscientificortraditionalsupport，thateveniftherewerenootherevidencethanthatofpalaeontologyinitsfavour，Ishouldfeelcompelledtoadoptthehypothesisofevolution。Happily，thefutureofpalaeontologyisindependentofallhypotheticalconsiderations。Fiftyyearshence，whoeverundertakestorecordtheprogressofpalaeontologywillnotethepresenttimeastheepochinwhichthelawofsuccessionoftheformsofthehigheranimalswasdeterminedbytheobservationofpalaeontologicalfacts。Hewillpointoutthat，justasStenoandasCuvierwereenabledfromtheirknowledgeoftheempiricallawsofco—
existenceofthepartsofanimalstoconcludefromaparttothewhole，sotheknowledgeofthelawofsuccessionofformsempoweredtheirsuccessorstoconclude，fromoneortwotermsofsuchasuccession，tothewholeseries;andthustodivinetheexistenceofformsoflife，ofwhich，perhaps，notraceremains，atepochsofinconceivableremotenessinthepast。
FOOTNOTES
（1）DeSolidoiintraSolidum，p。5——\"Datocorporecertafigurapraeditoetjuxtalegesnaturaeproducto，inipsocorporeargumentainvenirelocumetmodumproductionisdetegentia。\"
（2）\"Corporasibiinvicemomninosimiliasimilietiammodoproductasunt。\"
（3）SirJ。D。Hooker。