April26th——Wethisdaymetwithamarkedchangeinthegeologicalstructureoftheplains。FromthefirststartingIhadcarefullyexaminedthegravelintheriver，andforthetwolastdayshadnoticedthepresenceofafewsmallpebblesofaverycellularbasalt。Thesegraduallyincreasedinnumberandinsize，butnonewereaslargeasaman’shead。Thismorning，however，pebblesofthesamerock，butmorecompact，suddenlybecameabundant，andinthecourseofhalfanhourwesaw，atthedistanceoffiveorsixmiles，theangularedgeofagreatbasalticplatform。Whenwearrivedatitsbasewefoundthestreambubblingamongthefallenblocks。Forthenexttwenty-eightmilestheriver-coursewasencumberedwiththesebasalticmasses。Abovethatlimitimmensefragmentsofprimitiverocks，derivedfromitssurroundingboulder-formation，wereequallynumerous。Noneofthefragmentsofanyconsiderablesizehadbeenwashedmorethanthreeorfourmilesdowntheriverbelowtheirparent-source：consideringthesingularrapidityofthegreatbodyofwaterintheSantaCruz，andthatnostillreachesoccurinanypart，thisexampleisamoststrikingone，oftheinefficiencyofriversintransportingevenmoderately-sizedfragments。
Thebasaltisonlylava，whichhasflowedbeneaththesea；buttheeruptionsmusthavebeenonthegrandestscale。Atthepointwherewefirstmetthisformationitwas120feetinthickness；followinguptherivercourse，thesurfaceimperceptiblyroseandthemassbecamethicker，sothatatfortymilesabovethefirststationitwas320feetthick。WhatthethicknessmaybeclosetotheCordillera，Ihavenomeansofknowing，buttheplatformthereattainsaheightofaboutthreethousandfeetabovethelevelofthesea：wemustthereforelooktothemountainsofthatgreatchainforitssource；andworthyofsuchasourcearestreamsthathaveflowedoverthegentlyinclinedbedoftheseatoadistanceofonehundredmiles。
Atthefirstglanceofthebasalticcliffsontheoppositesidesofthevalley，itwasevidentthatthestrataoncewereunited。Whatpower，then，hasremovedalongawholelineofcountry，asolidmassofveryhardrock，whichhadanaveragethicknessofnearlythreehundredfeet，andabreadthvaryingfromratherlessthantwomilestofourmiles？Theriver，thoughithassolittlepowerintransportingeveninconsiderablefragments，yetinthelapseofagesmightproducebyitsgradualerosionaneffectofwhichitisdifficulttojudgetheamount。Butinthiscase，independentlyoftheinsignificanceofsuchanagency，goodreasonscanbeassignedforbelievingthatthisvalleywasformerlyoccupiedbyanarmofthesea。
Itisneedlessinthisworktodetailtheargumentsleadingtothisconclusion，derivedfromtheformandthenatureofthestep-formedterracesonbothsidesofthevalley，fromthemannerinwhichthebottomofthevalleyneartheAndesexpandsintoagreatestuary-likeplainwithsand-hillocksonit，andfromtheoccurrenceofafewsea-shellslyinginthebedoftheriver。IfIhadspaceIcouldprovethatSouthAmericawasformerlyherecutoffbyastrait，joiningtheAtlanticandPacificoceans，likethatofMagellan。Butitmayyetbeasked，howhasthesolidbasaltbeenmoved？Geologistsformerlywouldhavebroughtintoplay，theviolentactionofsomeoverwhelmingdebacle；butinthiscasesuchasuppositionwouldhavebeenquiteinadmissible；because，thesamestep-likeplainswithexistingsea-shellslyingontheirsurface，whichfrontthelonglineofthePatagoniancoast，sweepuponeachsideofthevalleyofSantaCruz。Nopossibleactionofanyfloodcouldthushavemodelledtheland，eitherwithinthevalleyoralongtheopencoast；andbytheformationofsuchstep-likeplainsorterracesthevalleyitselfhadbeenhollowedout。Althoughweknowthattherearetides，whichrunwithintheNarrowsoftheStraitofMagellanattherateofeightknotsanhour，yetwemustconfessthatitmakestheheadalmostgiddytoreflectonthenumberofyears，centuryaftercentury，whichthetides，unaidedbyaheavysurf，musthaverequiredtohavecorrodedsovastanareaandthicknessofsolidbasalticlava。Nevertheless，wemustbelievethatthestrataunderminedbythewatersofthisancientstrait，werebrokenupintohugefragments，andtheselyingscatteredonthebeach，werereducedfirsttosmallerblocks，thentopebblesandlastlytothemostimpalpablemud，whichthetidesdriftedfarintotheEasternorWesternOcean。
Withthechangeinthegeologicalstructureoftheplainsthecharacterofthelandscapelikewisealtered。Whileramblingupsomeofthenarrowandrockydefiles，IcouldalmosthavefanciedmyselftransportedbackagaintothebarrenvalleysoftheislandofSt。Jago。Amongthebasalticcliffs，IfoundsomeplantswhichIhadseennowhereelse，butothersI