Therefore，inseekingfortheoriginofprotoplasm，wemusteventuallyturntothevegetableworld.Afluidcontainingcarbonicacid，water，andnitrogenoussalts，whichofferssuchaBarmecidefeast[105]totheanimal，isatablerichlyspreadtomultitudesofplants;and，withaduesupplyofonlysuchmaterials，manyaplantwillnotonlymaintainitselfinvigour，butgrowandmultiplyuntilithasincreasedamillion—fold，oramillionmillion—fold，thequantityofprotoplasmwhichitoriginallypossessed;inthiswaybuildingupthematteroflife，toanindefiniteextent，fromthecommonmatteroftheuniverse.
　　Thus，theanimalcanonlyraisethecomplexsubstanceofdeadprotoplasmtothehigherpower，asonemaysay，oflivingprotoplasm;whiletheplantcanraisethelesscomplexsubstances——
　　carbonicacid，water，andnitrogenoussalts——tothesamestageoflivingprotoplasm，ifnottothesamelevel.Buttheplantalsohasitslimitations.Someofthefungi，forexample，appeartoneedhighercompoundstostartwith;andnoknownplantcanliveupontheuncompoundedelementsofprotoplasm.Aplantsuppliedwithpurecarbon，hydrogen，oxygen，andnitrogen，phosphorus，sulphur，andthelike，wouldasinfalliblydieastheanimalinhisbathofsmelling—salts，thoughitwouldbesurroundedbyalltheconstituentsofprotoplasm.Nor，indeed，needtheprocessofsimplificationofvegetablefoodbecarriedsofarasthis，inordertoarriveatthelimitoftheplant’sthaumaturgy.Letwater，carbonicacid，andalltheotherneedfulconstituentsbesuppliedexceptnitrogenoussalts，andanordinaryplantwillstillbeunabletomanufactureprotoplasm.
　　Thusthematteroflife，sofarasweknowit（andwehavenorighttospeculateonanyother），breaksup，inconsequenceofthatcontinualdeathwhichistheconditionofitsmanifestingvitality，intocarbonicacid，water，andnitrogenouscompounds，whichcertainlypossessnopropertiesbutthoseofordinarymatter.Andoutofthesesameformsofordinarymatter，andfromnonewhicharesimpler，thevegetableworldbuildsupalltheprotoplasmwhichkeepstheanimalworlda—going.Plantsaretheaccumulatorsofthepowerwhichanimalsdistributeanddisperse.
　　Butitwillbeobserved，thattheexistenceofthematteroflifedependsonthepre—existenceofcertaincompounds;namely，carbonicacid，water，andcertainnitrogenousbodies.Withdrawanyoneofthesethreefromtheworld，andallvitalphaenomenacometoanend.Theyareasnecessarytotheprotoplasmoftheplant，astheprotoplasmoftheplantistothatoftheanimal.Carbon，hydrogen，oxygen，andnitrogenarealllifelessbodies.Ofthese，carbonandoxygenuniteincertainproportionsandundercertainconditions，togiverisetocarbonicacid;hydrogenandoxygenproducewater;nitrogenandotherelementsgiverisetonitrogenoussalts.Thesenewcompounds，liketheelementarybodiesofwhichtheyarecomposed，arelifeless.Butwhentheyarebroughttogether，undercertainconditions，theygiverisetothestillmorecomplexbody，protoplasm，andthisprotoplasmexhibitsthephaenomenaoflife.
　　Iseenobreakinthisseriesofstepsinmolecularcomplication，andIamunabletounderstandwhythelanguagewhichisapplicabletoanyonetermoftheseriesmaynotbeusedtoanyoftheothers.
　　Wethinkfittocalldifferentkindsofmattercarbon，oxygen，hydrogen，andnitrogen，andtospeakofthevariouspowersandactivitiesofthesesubstancesasthepropertiesofthematterofwhichtheyarecomposed.
　　Whenhydrogenandoxygenaremixedinacertainproportion，andanelectricsparkispassedthroughthem，theydisappear，andaquantityofwater，equalinweighttothesumoftheirweights，appearsintheirplace.Thereisnottheslightestparitybetweenthepassiveandactivepowersofthewaterandthoseoftheoxygenandhydrogenwhichhavegivenrisetoit.At32degreesFahrenheit，andfarbelowthattemperature，oxygenandhydrogenareelasticgaseousbodies，whoseparticlestendtorushawayfromoneanotherwithgreatforce.Water，atthesametemperature，isastrongthoughbrittlesolidwhoseparticlestendtocohereintodefinitegeometricalshapes，andsometimesbuildupfrostyimitationsofthemostcomplexformsofvegetablefoliage.
　　Neverthelesswecallthese，andmanyotherstrangephaenomena，thepropertiesofthewater，andwedonothesitatetobelievethat，insomewayoranother，theyresultfromthepropertiesofthecomponentelementsofthewater.Wedonotassumethatasomethingcalled"aquosity"enteredintoandtookpossessionoftheoxidatedhydrogenassoonasitwasformed，andthenguidedtheaqueousparticlestotheirplacesinthefacetsofthecrystal，oramongsttheleafletsofthehoar—frost.Onthecontrary，weliveinthehopeandinthefaiththat，bytheadvanceofmolecularphysics，weshallbyandbybeabletoseeourwayasclearlyfromtheconstituentsofwatertothepropertiesofwater，aswearenowabletodeducetheoperationsofawatchfromtheformofitspartsandthemannerinwhichtheyareputtogether.
　　Isthecaseinanywaychangedwhencarbonicacid，water，andnitrogenoussaltsdisappear，andintheirplace，undertheinfluenceofpre—existinglivingprotoplasm，anequivalentweightofthematteroflifemakesitsappearance？
　　Itistruethatthereisnosortofparitybetweenthepropertiesofthecomponentsandthepropertiesoftheresultant，butneitherwasthereinthecaseofthewater.ItisalsotruethatwhatI
　　havespokenofastheinfluenceofpre—existinglivingmatterissomethingquiteunintelligible;butdoesanybodyquitecomprehendthemodusoperandi[106]ofanelectricspark，whichtraversesamixtureofoxygenandhydrogen？
　　Whatjustificationisthere，then，fortheassumptionoftheexistenceinthelivingmatterofasomethingwhichhasnorepresentative，orcorrelative，inthenotlivingmatterwhichgaverisetoit？Whatbetterphilosophicalstatushas"vitality"than"aquosity"？Andwhyshould"vitality"hopeforabetterfatethantheother"itys"whichhavedisappearedsinceMartinusScriblerus[107]
　　accountedfortheoperationofthemeat—jack[108]byitsinherent"meat—roastingquality，"andscornedthe"materialism"ofthosewhoexplainedtheturningofthespitbyacertainmechanismworkedbythedraughtofthechimney.
　　Ifscientificlanguageistopossessadefiniteandconstantsignificationwheneveritisemployed，itseemstomethatwearelogicallyboundtoapplytotheprotoplasm，orphysicalbasisoflife，thesameconceptionsasthosewhichareheldtobelegitimateelsewhere.Ifthephaenomenaexhibitedbywaterareitsproperties，soarethosepresentedbyprotoplasm，livingordead，itsproperties.
　　Ifthepropertiesofwatermaybeproperlysaidtoresultfromthenatureanddispositionofitscomponentmolecules，Icanfindnointelligiblegroundforrefusingtosaythatthepropertiesofprotoplasmresultfromthenatureanddispositionofitsmolecules.
　　ButIbidyoubewarethat，inacceptingtheseconclusions，youareplacingyourfeetonthefirstrungofaladderwhich，inmostpeople’sestimation，isthereverseofJacob’s，andleadstotheantipodesofheaven.Itmayseemasmallthingtoadmitthatthedullvitalactionsofafungus，oraforaminifer，arethepropertiesoftheirprotoplasm，andarethedirectresultsofthenatureofthematterofwhichtheyarecomposed.Butif，asIhaveendeavouredtoprovetoyou，theirprotoplasmisessentiallyidenticalwith，andmostreadilyconvertedinto，thatofanyanimal，Icandiscovernologicalhalting—placebetweentheadmissionthatsuchisthecase，andthefurtherconcessionthatallvitalactionmay，withequalpropriety，besaidtobetheresultofthemolecularforcesoftheprotoplasmwhichdisplaysit.
　　Andifso，itmustbetrue，inthesamesenseandtothesameextent，thatthethoughtstowhichIamnowgivingutterance，andyourthoughtsregardingthem，aretheexpressionofmolecularchangesinthatmatteroflifewhichisthesourceofourothervitalphaenomena.[109]
　　ONCORALANDCORALREEFS[110]
　　Themarineproductionswhicharecommonlyknownbythenamesof"Corals"and"Corallines，"werethoughtbytheancientstobesea—
　　weeds，whichhadthesingularpropertyofbecominghardandsolid，whentheywerefishedupfromtheirnativedepthsandcameintocontactwiththeair.
　　"Sicetcuralium，quoprimumcontigitaurasTemporedurescit：mollisfuitherbasubundis，"[111]
　　saysOvid（Metam.xv）;anditwasnotuntiltheseventeenthcenturythatBoccone[112]wasemboldened，bypersonalexperienceofthefacts，todeclarethattheholdersofthisbeliefwerenobetterthan"idiots，"whohadbeenmisledbythesoftnessoftheoutercoatofthelivingredcoraltoimaginethatitwassoftallthrough.
　　MesserBoccone’sstrongepithetisprobablyundeserved，asthenotionhecontroverts，inalllikelihood，arosemerelyfromthemisinterpretationofthestrictlytruestatementwhichanycoralfishermanwouldmaketoacuriousinquirer;namely，thattheoutsidecoatoftheredcoralisquitesoftwhenitistakenoutofthesea.Atanyrate，hedidgoodservicebyeliminatingthismucherrorfromthecurrentnotionsaboutcoral.Butthebeliefthatcoralsareplantsremained，notonlyinthepopular，butinthescientificmind;anditreceivedwhatappearedtobeastrikingconfirmationfromtheresearchesofMarsigli[113]in1706.Forthisnaturalist，havingtheopportunityofobservingfreshly—takenredcoral，sawthatitsbrancheswerebesetwithwhatlookedlikedelicateandbeautifulflowerseachhavingeightpetals.Itwastruethatthese"flowers"couldprotrudeandretractthemselves，buttheirmotionswerehardlymoreextensive，ormorevaried，thanthoseoftheleavesofthesensitiveplant;andthereforetheycouldnotbeheldtomilitateagainsttheconclusionsostronglysuggestedbytheirformandtheirgroupinguponthebranchesofatree—likestructure.
　　Twentyyearslater，apupilofMarsigli，theyoungMarseillesphysician，Peyssonel，conceivedthedesiretostudythesesingularsea—plants，andwassentbytheFrenchGovernmentonamissiontotheMediterraneanforthatpurpose.Thepupilundertooktheinvestigationfullofconfidenceintheideasofhismaster，butbeingabletoseeandthinkforhimself，hesoondiscoveredthatthoseideasbynomeansaltogethercorrespondedwithreality.Inanessayentitled"TraiteduCorail，"whichwascommunicatedtotheFrenchAcademyofScience，butwhichhasneverbeenpublished，Peyssonelwrites：——
　　"Jefisfleurirlecoraildansdesvasespleinsd’eaudemer，etj’observaiquecequenouscroyonsetrelafleurdecettepretendueplanten’etaitauvrai，qu’uninsectesemblableaunepetiteOrtieouPoulpe.J’avaisleplaisirdevoirremuerlespattes，oupieds，decetteOrtie，etayantmislevasepleind’eauoulecorailetaitaunedoucechaleuraupresdufeu，touslespetitsinsectess’epanouirent.——L’Ortiesortieetendlespieds，etformecequeM.
　　deMarsiglietmoiavionsprispourlespetalesdelafleur.Lecalicedecettepretenduefleurestlecorpsmemedel’animalavanceetsortihorsdelacellule."*[114]
　　*ThisextractfromPeyssonel’smanuscriptisgivenbyM.LacazeDuthiersinhisvaluableHistoireNaturelleduCorail（1866）.
　　Thecomparisonoftheflowersofthecoraltoa"petiteortie，"or"littlenettle，"isperfectlyjust，butneedsexplanation."Ortiedemer，"or"sea—nettle，"is，infact，theFrenchappellationforour"sea—anemone，"acreaturewithwhicheverybody，sincethegreataquariummania，musthavebecomefamiliar，eventothelimitsofboredom.In1710，thegreatnaturalist，Reaumur，[115]hadwrittenamemoirfortheexpresspurposeofdemonstratingthatthese"orties"
　　areanimals;andwiththisimportantpaperPeyssonelmustnecessarilyhavebeenfamiliar.Therefore，whenhedeclaredthe"flowers"oftheredcoraltobelittle"orties，"itwasthesamethingassayingthattheywereanimalsofthesamegeneralnatureassea—anemones.ButtoPeyssonel’scontemporariesthiswasanextremelystartlingannouncement.Itwashardtoimaginetheexistenceofsuchathingasanassociationofanimalsintoastructurewithstemandbranchesaltogetherlikeaplant，andfixedtothesoilasaplantisfixed;andthenaturalistsofthatdaypreferrednottoimagineit.EvenReaumurcouldnotbringhimselftoacceptthenotion，andFrancebeingblessedwithAcademicians，whosegreatfunction（asthelateBishopWilson[116]andaneminentmodernwriter[117]havesowellshown）istocausesweetnessandlighttoprevail，andtopreventsuchunmannerlyfellowsasPeyssonelfromblurtingoutunedifyingtruths，theysuppressedhim;and，asaforesaid，hisgreatworkremainedinmanuscript，andmayatthisdaybeconsultedbythecuriousinthatstate，intheBibliothequeduMuseumd’HistoireNaturelle.Peyssonel，whoevidentlywasapersonofsavageanduntameabledisposition，sofarfromappreciatingthekindnessoftheAcademiciansingivinghimtimetoreflectupontheunreasonableness，nottosayrudeness，ofmakingpublicstatementsinoppositiontotheviewsofsomeofthemostdistinguishedoftheirbody，seemsbitterlytohaveresentedthetreatmenthemetwith.ForhesentallfurthercommunicationstotheRoyalSocietyofLondon，whichneverhad，anditistobehopedneverwillhave，anythingofanacademicconstitution;andfinallyhetookhimselfofftoGuadaloupe，andbecamelosttosciencealtogether.
　　FifteenorsixteenyearsafterthedateofPeyssonel’ssuppressedpaper，theAbbeTrembley[118]publishedhiswonderfulresearchesuponthefresh—waterHydra.BernarddeJussieu[119]andGuettard[120]
　　followedthemupbylikeinquiriesuponthemarinesea—anemonesandcorallines;Reaumur，convincedagainsthiswilloftheentirejusticeofPeyssonel’sviews，adoptedthem，andmadehimahalf—and—halfapologyintheprefacetothenextpublishedvolumeofthe"Memoirespourservirl’HistoiredesInsectes;"and，fromthistimeforth，Peyssonel’sdoctrinethatcoralsaretheworkofanimalorganismshasbeenpartofthebodyofestablishedscientifictruth.
　　Peyssonel，intheextractfromhismemoiralreadycited，comparestheflower—likeanimalofthecoraltoa"poulpe，"whichistheFrenchformofthename"polypus，"——"themany—footed，"——whichtheancientnaturalistsgavetothesoft—bodiedcuttlefishes，which，likethecoralanimal，haveeightarms，ortentacles，disposedaroundacentralmouth.Reaumur，admittingtheanalogyindicatedbyPeyssonel，gavethenameofpolypes，notonlytothesea—
　　anemone，thecoralanimal，andthefresh—waterHydra，buttowhatarenowknownasthePolyzoa，andhetermedtheskeletonwhichtheyfabricatea"polypier，"or"polypidom."
　　Theprogressofdiscovery，sinceReaumur’stime，hasmadeusverycompletelyacquaintedwiththestructureandhabitsofallthesepolypes.Weknowthat，amongthesea—anemonesandcoral—forminganimals，eachpoylpehasamouthleadingtoastomach，whichisopenatitsinnerend，andthuscommunicatesfreelywiththegeneralcavityofthebody;thatthetentaclesplacedroundthemoutharehollow，andthattheyperformthepartofarmsinseizingandcapturingprey.Itisknownthatmanyofthesecreaturesarecapableofbeingmultipliedbyartificialdivision，thedividedhalvesgrowing，afteratime，intocompleteandseparateanimals;
　　andthatmanyareabletoperformaverysimilarprocessnaturally，insuchamannerthatonepolypemay，byrepeatedincompletedivisions，giverisetoasortofsheet，orturf，formedbyinnumerableconnected，andyetindependent，descendants.Or，whatisstillmorecommon，apolypemaythrowoutbuds，whichareconvertedintopolypes，orbranchesbearingpolypes，untilatree—
　　likemass，sometimesofveryconsiderablesize，isformed.
　　Thisiswhathappensinthecaseoftheredcoralofcommerce.A
　　minutepolype，fixedtotherockybottomofthedeepsea，growsupintoabranchedtrunk.Theendofeverybranchandtwigisterminatedbyapolype;andallthepolypesareconnectedtogetherbyafleshysubstance，traversedbyinnumerablecanalswhichplaceeachpolypeincommunicationwitheveryother，andcarrynourishmenttothesubstanceofthesupportingstem.Itisasortofnaturalcooperativestore，everypolypehelpingthewhole，atthesametimeasithelpsitself.Theinteriorofthestem，likethatofthebranches，issolidifiedbythedepositionofcarbonateoflimeinitstissue，somewhatinthesamefashionasourownbonesareformedofanimalmatterimpregnatedwithlimesalts;anditisthisdenseskeleton（usuallyturnedredbyapeculiarcolouringmatter）clearedofthesoftanimalinvestment，asthehardwoodofatreemightbestrippedofitsbark，whichistheredcoral.
　　Inthecaseoftheredcoral，thehardskeletonbelongstotheinteriorofthestemandbranchesonly;butinthecommonerwhitecorals，eachpolypehasacompleteskeletonofitsown.Thesepolypesaresometimessolitary，inwhichcasethewholeskeletonisrepresentedbyasinglecup，withpartitionsradiatingfromitscentretoitscircumference.Whenthepolypesformedbybuddingordivisionremainassociated，thepolypidomissometimesmadeupofnothingbutanaggregationofthesecups，whileatothertimesthecupsareatonceseparatedandheldtogether，byanintermediatesubstance，whichrepresentsthebranchesoftheredcoral.Theredcoralpolypeagainisacomparativelyrareanimal，inhabitingalimitedarea，theskeletonofwhichhasbutaveryinsignificantmass;whilethewhitecoralsareverycommon，occurinalmostallseas，andformskeletonswhicharesometimesextremelymassive.
　　Withaveryfewexceptions，boththeredandthewhitecoralpolypesare，intheiradultstate，firmlyadherenttothesea—
　　bottom;nordotheirbudsnaturallybecomedetachedandlocomotive.
　　But，inadditiontobuddinganddivision，thesecreaturespossessthemoreordinarymethodsofmultiplication;and，atparticularseasons，theygiverisetonumerouseggsofminutesize.Withintheseeggstheyoungareformed，andtheyleavetheegginaconditionwhichhasnosortofresemblancetotheperfectanimal.
　　Itis，infact，aminuteovalbody，manyhundredtimessmallerthanthefullgrowncreature，anditswimsaboutwithgreatactivitybythehelpofmultitudesoflittlehair—likefilaments，calledcilia，withwhichitsbodyiscovered.Theseciliaalllashthewaterinonedirection，andsodrivethelittlebodyalongasifitwerepropelledbythousandsofextremelyminutepaddles.Afterenjoyingitsfreedomforalongerorshortertime，andbeingcarriedeitherbytheforceofitsowncilia，orbycurrentswhichbearitalong，theembryocoralsettlesdowntothebottom，losesitscilia，andbecomesfixedtotherock，graduallyassumingthepolypeformandgrowinguptothesizeofitsparent.Astheinfantpolypesofthecoralmayretainthisfreeandactiveconditionformanyhours，orevendays，andasatidalorothercurrentintheseamayeasilyflowatthespeedoftwoorevenmoremilesinanhour，itisclearthattheembryomustoftenbetransportedtoveryconsiderabledistancesfromtheparent.Anditiseasilyunderstoodhowasinglepolype，whichmaygiverisetohundreds，orperhapsthousands，ofembryos，may，bythisprocessofpartlyactiveandpartlypassivemigration，coveranimmensesurfacewithitsoffspring.
　　Themassesofcoralwhichmaybeformedbytheassemblagesofpolypeswhichspringbybudding，orbydividing，fromasinglepolype，occasionallyattainveryconsiderabledimensions.Suchskeletonsaresometimesgreatplates，manyfeetlongandseveralfeetinthickness;ortheymayformhugehalfglobes，likethebrainstonecorals，ormayreachthemagnitudeofstoutshrubsorevensmalltrees.Thereisreasontobelievethatsuchmassesasthesetakealongtimetoform，andhencethattheageapolypetree，orpolypeturf，mayattain，maybeconsiderable.But，soonerorlater，thecoralpolypes，likeallotherthings，die;thesoftfleshdecays，whiletheskeletonisleftasastonymassatthebottomofthesea，whereitretainsitsintegrityforalongerorashortertime，accordingasitspositionaffordsmoreorlessprotectionfromthewearandtearofthewaves.
　　Thepolypeswhichgiverisetothewhitecoralarefound，ashasbeensaid，intheseasofallpartsoftheworld;butinthetemperateandcoldoceanstheyarescatteredandcomparativelysmallinsize，sothattheskeletonsofthosewhichdiedonotaccumulateinanyconsiderablequantity.Butitisotherwiseinthegreaterpartoftheoceanwhichliesinthewarmerpartsoftheworld，comprisedwithinadistanceofabouteighteenhundredmilesoneachsideoftheequator.Withinthezonethusbounded，byfarthegreaterpartoftheoceanisinhabitedbycoralpolypes，whichnotonlyformverystrongandlargeskeletons，butassociatetogetherintogreatmasses，likethethicketsandthemeadowturf，or，betterstill，theaccumulationsofpeat，towhichplantsgiveriseondryland.Thesemassesofstonymatter，heapedupbeneaththewatersoftheocean，becomeasdangeroustomarinersassomuchordinaryrock，andtothese，astothecommonrockridges，theseamangivesthenameof"reefs."
　　SuchcoralreefscovermanythousandsquaremilesinthePacificandintheIndianOceans.Thereisonereef，orrathergreatseriesofreefs，calledtheBarrierReef，whichstretches，almostcontinuously，formorethanelevenhundredmilesofftheeastcoastofAustralia.MultitudesoftheislandsinthePacificareeitherreefsthemselves，oraresurroundedbyreefs.TheRedSeaisinmanypartsalmostamazeofsuchreefs，andtheyaboundnolessintheWestIndies，alongthecoastofFlorida，andevenasfarnorthastheBahamaIslands.Butitisaveryremarkablecircumstancethat，withintheareaofwhatwemaycallthe"coralzone，"therearenocoralreefsuponthewestcoastofAmerica，noruponthewestcoastofAfrica;anditisageneralfactthatthereefsareinterrupted，orabsent，oppositethemouthsofgreatrivers.Thecausesofthisapparentcapriceinthedistributionofcoralreefsarenotfartoseek.Thepolypeswhichfabricatethemrequirefortheirvigorousgrowthatemperaturewhichmustnotfallbelow68
　　degreesFahrenheitalltheyearround，andthistemperatureisonlytobefoundwithinthedistanceoneachsideoftheequatorwhichhasbeenmentioned，orthereabouts.Butevenwithinthecoralzonethisdegreeofwarmthisnoteverywheretobehad.OnthewestcoastofAmerica，andonthecorrespondingcoastofAfrica，thecurrentsofcoldwaterfromtheicyregionswhichsurroundtheSouthPolesetnorthward，anditappearstobeduetotheircoolinginfluencethattheseaintheseregionsisfreefromthereefbuilders.Again，thecoralpolypescannotliveinwaterwhichisrenderedbrackishbyfloodsfromtheland，orwhichisperturbedbymudfromthesamesource，andhenceitisthattheyceasetoexistoppositethemouthsofrivers，whichdamagetheminboththeseways.
　　Suchisthegeneraldistributionofthereef—buildingcorals，buttherearesomeveryinterestingandsingularcircumstancestobeobservedintheconformationofthereefs，whenweconsiderthemindividually.Thereefs，infact，areofthreedifferentkinds;
　　someofthemstretchoutfromtheshore，almostlikeaprolongationofthebeach，coveredonlybyshallowwater，andinthecaseofanisland，surroundingitlikeafringeofnoconsiderablebreadth.
　　Thesearetermed"fringingreefs."Othersareseparatedbyachannelwhichmayattainawidthofmanymiles，andadepthoftwentyorthirtyfathomsormore，fromthenearestland;andwhenthislandisanisland，thereefsurroundsitlikealowwall，andtheseabetweenthereefandthelandis，asitwere，amoatinsidethiswall.Suchreefsasthesearecalled"encircling"whentheysurroundanisland;and"barrier"reefs，whentheystretchparallelwiththecoastofacontinent.Inboththesecasesthereisordinarydrylandinsidethereef，andseparatedfromitonlybyanarrowerorawider，ashalloweroradeeper，spaceofsea，whichiscalleda"lagoon，"or"innerpassage."Butthereisathirdkindofreef，ofverycommonoccurrenceinthePacificandIndianOceans，whichgoesbythenameof"atoll."Thisis，toallintentsandpurposes，anencirclingreef，withoutanythingtoencircle;or，inotherwords，withoutanislandinthemiddleofitslagoon.Theatollhasexactlytheappearanceofavast，irregularlyoval，orcircular，breakwater，enclosingsmoothwaterinitsmidst.Thedepthofthewaterinthelagoonrarelyexceedstwentyorthirtyfathoms，but，outsidethereef，itdeepenswithgreatrapiditytotwohundredorthreehundredfathoms.Thedepthimmediatelyoutsidethebarrier，orencircling，reefs，mayalsobeveryconsiderable;but，attheouteredgeofafringingreef，itdoesnotamountusuallytomorethantwentyortwenty—fivefathoms;inotherwords，fromonehundredandtwentytoonehundredandfiftyfeet.
　　Thus，ifthewateroftheoceanshouldbesuddenlydrainedaway，weshouldseetheatollsrisingfromthesea—bedlikevasttruncatedcones，andresemblingsomanyvolcaniccraters，exceptthattheirsideswouldbesteeperthanthoseofanordinaryvolcano.Inthecaseoftheencirclingreefs，thecone，withtheenclosedisland，wouldlooklikeVesuviuswithMonteNuovowithintheoldcraterofSomma;[121]while，finally，theislandwithafringingreefwouldhavetheappearanceofanordinaryhill，ormountain，girdedbyavastparapet，withinwhichwouldlieashallowmoat.AndthedrybedofthePacificmightaffordgroundsforaninhabitantofthemoontospeculateupontheextraordinarysubterraneanactivitytowhichthesevastandnumerous"craters"borewitness！
　　Whenthestructureofafringingreefisinvestigated，thebottomofthelagoonisfoundtobecoveredwithfinewhitishmud，whichresultsfromthebreakingupofthedeadcorals.Uponthismuddyfloortherelie，hereandthere，growingcorals，oroccasionallygreatblocksofdeadcoral，whichhavebeentornbystormsfromtheouteredgeofthereef，andwashedintothelagoon.Shellfishandwormsofvariouskindsabound;andfish，someofwhichpreyuponthecoral，sportinthedeeperpools.Butthecoralswhicharetobeseengrowingintheshallowwatersofthelagoonareofadifferentkindfromthosewhichaboundontheouteredgeofthereef，andofwhichthereefisbuiltup.Closetotheseawardedgeofthereef，overwhich，evenincalmweather，asurfalmostalwaysbreaks，thecoralrockisencrustedwithathickcoatofasingularvegetableorganism，whichcontainsagreatdealoflime——theso—
　　calledNullipora.Beyondthis，inthepartoftheedgeofthereefwhichisalwayscoveredbythebreakingwaves，theliving，true，reef—polypesmaketheirappearance;and，indifferentforms，coatthesteepseawardfaceofthereeftoadepthofonehundredorevenonehundredandfiftyfeet.Beyondthisdepththesounding—
　　leadrests，notuponthewall—likefaceofthereef，butontheordinaryshelvingsea—bottom.Andthedistancetowhichafringingreefextendsfromthelandcorrespondswiththatatwhichtheseahasadepthoftwentyorfive—and—twentyfathoms.
　　If，aswehavesupposed，theseacouldbesuddenlywithdrawnfromaroundanislandprovidedwithafringingreef，suchastheMauritius，[122]thereefwouldpresenttheaspectofaterrace，itsseawardface，onehundredfeetormorehigh，bloomingwiththeanimalflowersofthecoral，whileitssurfacewouldbehollowedoutintoashallowandirregularmoat—likeexcavation.
　　Thecoralmud，whichoccupiesthebottomofthelagoon，andwithwhichalltheintersticesofthecoralskeletonswhichaccumulatetoformthereefarefilledup，doesnotproceedfromthewashingactionofthewavesalone;innumerablefishes，andothercreatureswhichpreyuponthecoral，addaveryimportantcontributionoffinely—trituratedcalcareousmatter;andthecoralsandmudbecomingincorporatedtogether，graduallyhardenandgiverisetoasortoflimestonerock，whichmayvaryagooddealintexture.
　　Sometimesitremainsfriableandchalky，but，moreoften，theinfiltrationofwater，chargedwithcarbonicacid，dissolvessomeofthecalcareousmatter，anddepositsitelsewhereintheintersticesofthenascentrock，thusglueingandcementingtheparticlestogetherintoahardmass;oritmayevendissolvethecarbonateoflimemoreextensively，andre—deposititinacrystallineform.Onthebeachofthelagoon，wherethecoralsandiswashedintolayersbytheactionofthewaves，itsgrainsbecomethusfusedtogetherintostrataofalimestone，sohardthattheyringwhenstruckwithahammer，andinclinedatagentleangle，correspondingwiththatofthesurfaceofthebeach.Thehardpartsofthemanyanimalswhichliveuponthereefbecomeimbeddedinthiscorallimestone，sothatablockmaybefullofshellsofbivalvesandunivalves，orofseaurchins;andevensometimesenclosestheeggsofturtlesinastateofpetrification.Theactiveandvigorousgrowthofthereefgoesononlyattheseawardmargins，wherethepolypesareexposedtothewashofthesurf，andaretherebyprovidedwithanabundantsupplyofairandoffood.
　　Theinteriorportionofthereefmayberegardedasalmostwhollyanaccumulationofdeadskeletons.Wherearivercomesdownfromthelandthereisabreakinthereef，forthereasonswhichhavebeenalreadymentioned.
　　Theoriginandmodeofformationofafringingreef，suchasthatjustdescribed，areplainenough.Theembryosofthecoralpolypeshavefixedthemselvesuponthesubmergedshoreoftheisland，asfaroutastheycouldlive，namely，toadepthoftwentyortwenty—
　　fivefathoms.Onegenerationhassucceededanother，buildingitselfupuponthedeadskeletonsofitspredecessor.Themasshasbeenconsolidatedbytheinfiltrationofcoralmud，andhardenedbypartialsolutionandredeposition，untilagreatrampartofcoralrockonehundredoronehundredandfiftyfeethighonitsseawardfacehasbeenformedallroundtheisland，withonlysuchgapsasresultfromtheoutflowofrivers，intheplaceofsally—ports.
　　Thestructureoftherockyaccumulationintheencirclingreefsandintheatollsisessentiallythesameasinthefringingreef.
　　But，inadditiontothedifferencesofdepthinsideandout，theypresentsomeotherpeculiarities.Thesereefs，andespeciallytheatolls，areusuallyinterruptedatonepartoftheircircumference，andthispartisalwayssituatedontheleewardsideofthereef，orthatwhichisthemoreshelteredside.Now，asallthesereefsaresituatedwithintheregioninwhichthetradewindsprevail，itfollowsthat，onthenorthsideoftheequator，wherethetrade—
　　windisanortheasterlywind，theopeningofthereefisonthesouthwestside：whileinthesouthernhemisphere，wherethetrade—
　　windsblowfromthesoutheast，theopeningliestothenorthwest.
　　Thecuriouspracticalresultfollowsfromthisstructure，thatthelagoonstothesereefsreallyformadmirableharbours，ifashipcanonlygetinsidethem.Butthemaindifferencebetweentheencirclingreefsandtheatolls，ontheonehand，andthefringingreefsontheother，liesinthefactofthemuchgreaterdepthofwaterontheseawardfacesoftheformer.Asaconsequenceofthisfact，thewholeofthisfaceisnot，asitisinthecaseofthefringingreef，coveredwithlivingcoralpolypes.For，aswehaveseen，thesepolypescannotliveatagreaterdepththanabouttwenty—fivefathoms;andactualobservationhasshownthatwhile，downtothisdepth，thesounding—leadwillbringupbranchesoflivecoralfromtheouterwallofsuchareef，atagreaterdepthitfetchestothesurfacenothingbutdeadcoralandcoralsand.
　　Wemust，therefore，picturetoourselvesanatoll，oranencirclingreef，asfringedforonehundredfeet，ormore，fromitssummit，withcoralpolypesbusilyengagedinfabricatingcoral;while，belowthiscomparativelynarrowbelt，itssurfaceisabareandsmoothexpanseofcoralsand，supporteduponandwithinacoreofcorallimestone.Thus，ifthebedofthePacificweresuddenlylaidbare，aswasjustnowsupposed，theappearanceofthereef—
　　mountainswouldbeexactlythereverseofthatpresentedbymanyhighmountainsonland.Forthesearewhitewithsnowatthetop，whiletheirbasesareclothedwithanabundantandgaudily—colouredvegetation.Butthecoralconeswouldlookgreyandbarrenbelow，whiletheirsummitswouldbegaywitharichly—colouredparterreofflowerlikecoralpolypes.
　　Thepracticaldifficultiesofsoundingupon，andofbringingupportionsof，theseawardfaceofanatollorofanencirclingreef，aresogreat，inconsequenceoftheconstantanddangerousswellwhichsetstowardsit，thatnoexactinformationconcerningthedepthtowhichthereefsarecomposedofcoralhasyetbeenobtained.Thereisnoreasontodoubt，however，thatthereef—conehasthesamestructurefromitssummittoitsbase，andthatitssea—wallisthroughoutmainlycomposedofdeadcoral.
　　Andnowarisesaseriousdifficulty.Ifthecoralpolypescannotliveatagreaterdepththanonehundredoronehundredandfiftyfeet，howcantheyhavebuiltupthebaseofthereef—cone，whichmaybetwothousandfeet，ormore，belowthesurfaceofthesea？
　　Inordertogetoverthisobjection，itwasatonetimesupposedthatthereef—buildingpolypeshadsettleduponthesummitsofachainofsubmarinemountains.Butwhatisthereinphysicalgeographytojustifytheassumptionoftheexistenceofachainofmountainsstretchingforonethousandmilesormore，andsonearlyofthesameheight，thatnoneshouldriseabovethelevelofthesea，norfallonehundredandfiftyfeetbelowthatlevel？
　　How，again，onthishypothesis，areatollstobeaccountedfor，unless，assomehavedone，wetakerefugeinthewildsuppositionthateveryatollcorrespondswiththecraterofasubmarinevolcano？Andwhatexplanationdoesitaffordofthefactthat，insomepartsoftheocean，onlyatollsandencirclingreefsoccur，whileotherspresentnonebutfringingreefs？
　　Theseandotherpuzzlingfactsremainedinsolubleuntilthepublication，intheyear1840，ofMr.Darwin’sfamousworkoncoralreefs;[123]inwhichakeywasgiventoallthedifficultproblemsconnectedwiththesubject，andeverydifficultywasshowntobecapableofsolutionbydeductivereasoningfromahappycombinationofcertainwell—establishedgeologicalandbiologicaltruths.Mr.
　　Darwin，infact，showedthat，solongasthelevelofthesearemainsunalteredinanyareainwhichcoralreefsarebeingformed，orifthelevelofthesearelativelytothatofthelandisfalling，theonlyreefswhichcanbeformedarefringingreefs.
　　Whileif，onthecontrary，theleveloftheseaisrisingrelativelytothatoftheland，ataratenotfasterthanthatatwhichtheupwardgrowthofthecoralcankeeppacewithit，thereefwillgraduallypassfromtheconditionofafringing，intothatofanencirclingorbarrierreef.And，finally，thatiftherelativelevelofthesearisesomuchthattheencircledlandiscompletelysubmerged，thereefmustnecessarilypassintotheconditionofanatoll.
　　For，supposetherelativeleveloftheseatoremainstationary，afterafringingreefhasreachedthatdistancefromthelandatwhichthedepthofwateramountstoonehundredandfiftyfeet.
　　Thenthereefcannotextendseawardbythemigrationofcoralgerms，becausethesecoralgermswouldfindthebottomoftheseatobetoodeepforthemtolivein.Andtheonlymannerinwhichthereefcouldextendoutwards，wouldbebythegradualaccumulation，atthefootofitsseawardface，ofatalusofcoralfragmentstornoffbytheviolenceofthewaves，whichtalusmight，incourseoftime，becomehighenoughtobringitsuppersurfacewithinthelimitsofcoralgrowth，andinthatmannerprovideasortoffactitioussea—bottomuponwhichthecoralembryosmightperch.If，ontheotherhand，theleveloftheseawereslowlyandgraduallylowered，itisclearthatthepartsofitsbottomoriginallybeyondthelimitofcoralgrowthwouldgraduallybebroughtwithintherequireddistanceofthesurface，andthusthereefmightbeindefinitelyextended.Butthisprocesswouldgiveriseneithertoanencirclingreefnortoanatoll，buttoabroadbeltofupheavedcoralrock，increasingthedimensionsofthedryland，andcontinuousseawardswiththefreshfringingreef.
　　Suppose，however，thatthesea—levelroseinsteadoffalling，atthesameslowandgradualrateatwhichweknowittoberisinginsomepartsoftheworld，——notmore，infact，thanafewinches，or，atmost，afootortwo，inahundredyears.Then，whilethereefwouldbeunabletoextenditselfseaward，thesea—bottomoutsideitbeinggraduallymoreandmoreremovedfromthedepthatwhichthelifeofthecoralpolypesispossible，itwouldbeabletogrowupwardsasfastasthesearose.Butthegrowthwouldtakeplacealmostexclusivelyaroundthecircumferenceofthereef，thisbeingtheonlyregioninwhichthecoralpolypeswouldfindtheconditionsfavourablefortheirexistence.Thebottomofthelagoonwouldberaised，inthemain，onlybythecoraldebrisandcoralmud，formedinthemanneralreadydescribed;consequently，themarginsofthereefwouldrisefasterthanthebottom，or，inotherwords，thelagoonwouldconstantlybecomedeeper.And，atthesametime，itwouldgraduallyincreaseinbreadth;astherisingsea，coveringmoreoftheland，wouldoccupyawiderspacebetweentheedgeofthereefandwhatremainedoftheland.Thustherisingseawouldeventuallyconvertalargeislandwithafringingreefintoasmallislandsurroundedbyanencirclingreef.
　　Anditwillbeobviousthatwhentherisingoftheseahasgonesofarascompletelytocoverthehighestpointsoftheisland，thereefwillhavepassedintotheconditionofanatoll.
　　Buthowisitpossiblethattherelativelevelofthelandandseashouldbealteredtothisextent？Clearly，onlyinoneoftwoways：eithertheseamusthaverisenoverthoseareaswhicharenowcoveredbyatollsandencirclingreefs;or，thelanduponwhichthesearestsmusthavebeendepressedtoacorrespondingextent.
　　Iftheseahasrisen，itsrisemusthavetakenplaceoverthewholeworldsimultaneously，anditmusthaverisentothesameheightoverallpartsofthecoralzone.Groundshavebeenshownforthebeliefthatthegeneralleveloftheseamayhavebeendifferentatdifferenttimes;ithasbeensuggested，forexample，thattheaccumulationoficeaboutthepolesduringoneofthecoldperiodsoftheearth’shistorynecessarilyimpliesadiminutioninthevolumeoftheseaproportionedtotheamountofitswaterthuspermanentlylockedupintheArcticandAntarcticice—cellars;
　　while，inthewarmperiods，thegreaterorlessdisappearanceofthepolarice—capimpliesacorrespondingadditionofwatertotheocean.Andnodoubtthisreasoningmustbeadmittedtobesoundinprinciple;thoughitisveryhardtosaywhatpracticaleffecttheadditionsandsubtractionsthusmadehavehadontheleveloftheocean;inasmuchassuchadditionsandsubtractionsmightbeeitherintensifiedornullified，bycontemporaneouschangesintheleveloftheland.Andnoonehasyetshownthatanysuchgreatmeltingofpolarice，andconsequentraisingofthelevelofthewateroftheocean，hastakenplacesincetheexistingatollsbegantobeformed.
　　Intheabsenceofanyevidencethattheseahaseverrisentotheextentrequiredtogiverisetotheencirclingreefsandtheatolls，Mr.Darwinadoptedtheoppositehypothesis，viz.，thatthelandhasundergoneextensiveandslowdepressioninthoselocalitiesinwhichthesestructuresexist.
　　Itseems，atfirst，astartlingparadox，tosupposethatthelandislessfixedthanthesea;butthatsuchisthecaseistheuniformtestimonyofgeology.Bedsofsandstoneorlimestone，thousandsoffeetthick，andallfullofmarineremains，occurinvariouspartsoftheearth’ssurface，andprove，beyondadoubt，thatwhenthesebedswereformed，thatportionofthesea—bottomwhichtheythenoccupiedunderwentaslowandgradualdepressiontoadistancewhichcannothavebeenlessthanthethicknessofthosebeds，andmayhavebeenverymuchgreater.Insupposing，therefore，thatthegreatareasofthePacificandoftheIndianOcean，overwhichatollsandencirclingreefsarefoundscattered，haveundergoneadepressionofsomehundreds，or，itmaybe，thousandsoffeet，Mr.Darwinmadeasuppositionwhichhadnothingforcedorimprobable，butwasentirelyinaccordancewithwhatweknowtohavetakenplaceoversimilarlyextensiveareas，inotherperiodsoftheworld’shistory.ButMr.Darwinsubjectedhishypothesistoaningeniousindirecttest.Ifhisviewbecorrect，itisclearthatneitheratolls，norencirclingreefs，shouldbefoundinthoseportionsoftheoceaninwhichwehavereasontobelieve，onindependentgrounds，thatthesea—bottomhaslongbeeneitherstationary，orslowlyrising.Nowitisknownthat，asageneralrule，thelevelofthelandiseitherstationary，orisundergoingaslowupheaval，intheneighborhoodofactivevolcanoes;and，therefore，neitheratollsnorencirclingreefsoughttobefoundinregionsinwhichvolcanoesarenumerousandactive.Andthisturnsouttobethecase.AppendedtoMr.
　　Darwin’sgreatworkoncoralreefs，thereisamaponwhichatollsandencirclingreefsareindicatedbyonecolour，fringingreefsbyanother，andactivevolcanoesbyathird.Anditisatonceobviousthatthelinesofactivevolcanoesliearoundthemarginsoftheareasoccupiedbytheatollsandtheencirclingreefs.Itisexactlyasiftheupheavingvolcanicagencieshadlifteduptheedgesofthesegreatareas，whiletheircentreshadundergoneacorrespondingdepression.Anatollareamay，inshort，bepicturedasakindofbasin，themarginsofwhichhavebeenpushedupbythesubterraneanforces，towhichthecratersofthevolcanoeshave，atintervals，givenvent.
　　ThuswemustimaginetheareaofthePacificnowcoveredbythePolynesianArchipelago，ashavingbeen，atsomeformertime，occupiedbylargeislands，or，maybe，byagreatcontinent，withtheordinarilydiversifiedsurfaceofplain，andhill，andmountainchain.Theshoresofthisgreatlandweredoubtlessfringedbycoralreefs;and，asitslowlyunderwentdepression，thehillyregions，convertedintoislands，became，atfirst，surroundedbyfringingreefs，andthen，asdepressionwenton，thesebecameconvertedintoencirclingreefs，andthese，finally，intoatolls，untilamazeofreefsandcoral—girdledisletstooktheplaceoftheoriginallandmasses.
　　Thustheatollsandtheencirclingreefsfurnishuswithclear，thoughindirect，evidenceofchangesinthephysicalgeographyoflargepartsoftheearth’ssurface;andeven，asmylamentedfriend，thelateProfessorJukes，[124]hassuggested，giveusindicationsofthemannerinwhichsomeofthemostpuzzlingfactsconnectedwiththedistributionofanimalshavebeenbroughtabout.
　　Forexample，AustraliaandNewGuineaareseparatedbyTorresStraits，abroadbeltofseaonehundredoronehundredandtwentymileswide.Nevertheless，thereisinmanyrespectsacuriousresemblancebetweenthelandanimalswhichinhabitNewGuineaandthelandanimalswhichinhabitAustralia.But，atthesametime，themarineshellfishwhicharefoundintheshallowwatersoftheshoresofNewGuineaarequitedifferentfromthosewhicharemetwithuponthecoastsofAustralia.Now，theeasternendofTorresStraitsisfullofatolls，which，infact，formthenorthernterminationoftheGreatBarrierReefwhichskirtstheeasterncoastofAustralia.Itfollows，therefore，thattheeasternendofTorresStraitsisanareaofdepression，anditisverypossible，andonmanygroundshighlyprobable，that，informertimes，AustraliaandNewGuineaweredirectlyconnectedtogether，andthatTorresStraitsdidnotexist.Ifthiswerethecase，theexistenceofcassowariesandofmarsupialquadrupeds，bothinNewGuineaandinAustralia，becomesintelligible;whilethedifferencebetweenthelittoralmolluscsofthenorthandthesouthshoresofTorresStraitsisreadilyexplainedbythegreatprobabilitythat，whenthedepressioninquestiontookplace，andwhatwas，atfirst，anarmoftheseabecameconvertedintoastraitseparatingAustraliafromNewGuinea，thenorthernshoreofthisnewseabecametenantedwithmarineanimalsfromthenorth，whilethesouthernshorewaspeopledbyimmigrantsfromthealreadyexistingmarineAustralianfauna.
　　Inasmuchasthegrowthofthereefdependsuponthatofsuccessivegenerationsofcoralpolypes，andaseachgenerationtakesacertaintimetogrowtoitsfullsize，andcanonlyseparateitscalcareousskeletonfromthewaterinwhichitlivesatacertainrate，itisclearthatthereefsarerecordsnotonlyofchangesinphysicalgeography，butofthelapseoftime.Itisbynomeanseasy，however，toestimatetheexactvalueofreefchronology，andtheattemptswhichhavebeenmadetodeterminetherateatwhichareefgrowsverticallyhaveyieldedanythingbutpreciseresults.A
　　cautiouswriter，Mr.Dana，[125]whoseextensivestudyofcoralsandcoralreefsmakeshimaneminentlycompetentjudge，stateshisconclusioninthefollowingterms：——
　　"Therateofgrowthofthecommonbranchingmadreporeisnotoveroneandahalfinchesayear.Asthebranchesareopen，thiswouldnotbeequivalenttomorethanhalfaninchinheightofsolidcoralforthewholesurfacecoveredbythemadrepore;and，astheyarealsoporous，tonotoverthree—eighthsofaninchofsolidlimestone.Butacoralplantationhaslargebarepatcheswithoutcorals，andthecoralsandsarewidelydistributedbycurrents，partofthemtodepthsoveronehundredfeetwheretherearenolivingcorals;notmorethanone—sixthofthesurfaceofareefregionis，infact，coveredwithgrowingspecies.Thisreducesthethree—eighthstoONE—SIXTEENTH.Shellsandotherorganicrelicsmaycontributeone—fourthasmuchascorals.Attheoutside，theaverageupwardincreaseofthewholereef—groundperyearwouldnotexceedONE—EIGHTHofaninch.
　　"Nowsomereefsareatleasttwothousandfeetthick，whichatone—
　　eighthofaninchayear，correspondstoonehundredandninety—twothousandyears."*
　　*Dana，ManualofGeology，p.591.
　　Halve，orquarter，thisestimateifyouwill，inordertobecertainoferringupontherightside，andstillthereremainsaprodigiousperiodduringwhichtheancestorsofexistingcoralpolypeshavebeenundisturbedlyatwork;andduringwhich，therefore，theclimatalconditionsoverthecoralareamusthavebeenmuchwhattheyarenow.
　　Andallthislapseoftimehasoccurredwithinthemostrecentperiodofthehistoryoftheearth.Theremainsofreefsformedbycoralpolypesofdifferentkindsfromthosewhichexistnow，enterlargelyintothecompositionofthelimestonesoftheJurassicperiod;[126]andstillmorewidelydifferentcoralpolypeshavecontributedtheirquotatothevastthicknessofthecarboniferousandDevonianstrata.ThenasregardsthelattergroupofrocksinAmerica，thehighauthorityalreadyquotedtellsus：——
　　"TheUpperHelderbergperiodiseminentlythecoralreefperiodofthepalaeozoicages.Manyoftherocksaboundincoral，andareastrulycoralreefsasthemodernreefsofthePacific.Thecoralsaresometimesstandingontherocksinthepositiontheyhadwhengrowing：othersarelyinginfragments，astheywerebrokenandheapedbythewaves;andotherswerereducedtoacompactlimestonebythefinertriturationbeforeconsolidationintorock.Thiscompactvarietyisthemostcommonkindamongthecoralreefrocksofthepresentseas;anditoftencontainsbutfewdistinctfossils，althoughformedinwaterthataboundedinlife.AtthefalloftheOhio，nearLouisville，thereisamagnificentdisplayoftheoldreef.HemisphericalFavosites，fiveorsixfeetindiameter，lietherenearlyasperfectaswhentheywerecoveredbytheirflowerlikepolypes;andbesidesthese，therearevariousbranchingcorals，andaprofusionofCyathophyllia，orcup—
　　corals."*
　　*Dana，ManualofGeology，p.272.
　　Thus，inallthegreatperiodsoftheearth’shistoryofwhichweknowanything，apartofthethenlivingmatterhashadtheformofpolypes，competenttoseparatefromthewateroftheseathecarbonateoflimenecessaryfortheirownskeletons.Grainbygrain，andparticlebyparticle，theyhavebuiltupvastmassesofrock，thethicknessofwhichismeasuredbyhundredsoffeet，andtheirareabythousandsofsquaremiles.Theslowoscillationsofthecrustoftheearth，producinggreatchangesinthedistributionoflandandwater，haveoftenobligedthelivingmatterofthecoral—builderstoshiftthelocalityofitsoperations;and，byvariationandadaptationtothesemodificationsofcondition，itsformshaveasoftenchanged.Theworkithasdoneinthepastis，forthemostpart，sweptaway，butfragmentsremain，and，iftherewerenootherevidence，sufficetoprovethegeneralconstancyoftheoperationsofNatureinthisworld，throughperiodsofalmostinconceivableduration.
　　NOTES
　　AUTOBIOGRAPHY
　　[1]
　　Autobiography：Huxley’saccountofthissketch，writtenin1889，isasfollows："Amanwhoisbringingoutaseriesofportraitsofcelebrities，withasketchoftheircareerattached，hasbotheredmeoutofmylifeforsomethingtogowithmyportrait，andtoescapetheabominablebadtasteofsomeofthenotices，Ihavedonethat."
　　[2]
　　pre—Boswellianepoch：thetimebeforeBoswell.JamesBoswell（1740—1795）wrotethefamousLifeofSamuelJohnson.Mr.LeslieStephendeclaresthatthisbook"becamethefirstspecimenofanewliterarytype.""Itisafull—lengthportraitofaman’sdomesticlifewithenoughpicturesquedetailtoenableustoseehimthroughtheeyesofprivatefriendship"AnumberofbiographerssinceBoswellhaveimitatedhismethod;andLeslieStephenbelievesthat"weoweitinsomedegreetohisexamplethatwehavesuchdelightfulbooksasLockhart’sLifeofScottorMr.Trevelyan’sLifeofMacaulay."
　　[3]
　　"Benequilatuit，benevixit"：fromOvid.Hewhohaskepthimselfwellhidden，haslivedwell.