Alouishasbeenoverlookedinacornerofthedrawer。Icanspenditwithoutseriouslyjeopardizingthedomesticbalance。Letmemakethisgifttoscience，who，Ifear，willbenonetoomuchobligedtome。Agorgeousequipmentmaybeallverywellforlaboratorieswhereinthecellsandfibersofthedeadareconsultedatgreatexpense;butsuchmagnificenceisofdoubtfulutilitywhenwehavetostudytheactionsoftheliving。Itisthehumblemakeshift，ofnovalue，thatstumblesonthesecretsoflife。
　　Whatdidthebestresultsofmystudiesofinstinctcostme？
　　Nothingbuttimeand，aboveall，patience。Myextravagantexpenditureoftwentyfrancs，therefore，willbeariskyspeculationifdevotedtothepurchaseofanapparatusofstudy。
　　Itwillbringmeinnothinginthewayoffreshviews，ofthatIamconvinced。However，letustry。
　　Theblacksmithmakesmetheframeworkofacageoutofafewironrods。Thejoiner，whoisalsoaglazieronoccasion——for，inmyvillage，youhavetobeaJack—of—all—tradesifyouwouldmakebothendsmeet——setstheframeworkonawoodenbaseandsuppliesitwithamovableboardasalid;hefixesthickpanesofglassinthefoursides。Beholdtheapparatus，complete，withabottomoftarredsheetironandatraptoletthewaterout。
　　Themakersexpressthemselvessatisfiedwiththeirwork，asingularnoveltyintheirrespectiveshops，wheremanyaninquisitivecallerhaswonderedwhatuseIintendtomakeofmylittleglasstrough。
　　Thethingcreatesacertainstir。Someinsistthatitismeanttoholdmysuppliesofoilandtotaketheplaceofthereceptacleingeneraluseinourparts，theurndugoutofablockofstone。
　　Whatwouldthoseutilitarianshavethoughtofmycrazymind，hadtheyknownthatmycostlygearwouldmerelyservetoletmewatchsomewretchedanimalskickingaboutinthewater！
　　Smithandglazierarecontentwiththeirwork。Imyselfampleased。Forallitsrusticair，theapparatusdoesnotlackelegance。Itlooksverywell，standingonalittletableinfrontofawindowvisitedbythesunforthegreaterpartoftheday。
　　Itsholdingcapacityissometenorelevengallons。Whatshallwecallit？Anaquarium？No，thatwouldbetoopretentiousandwould，veryunjustly，suggesttheaquatictoyfilledwithrockwork，waterfallsandgoldfishbelovedofthedwellersinsuburbia。
　　Letuspreservethegravityofseriousthingsandnottreatmylearnedtroughasthoughitwereadrawingroomfutility。Wewillcallittheglasspond。
　　Ifurnishitwithaheapofthoselimyincrustationswherewithcertainspringsintheneighborhoodcoverthedeadclumpofrushes。
　　Itislight，fullofholesandgivesafaintsuggestionofacoralreef。Moreover，itiscoveredwithashort，green，velvetymoss，adownyswardofinfinitesimalpondweed。Icountonthismodestvegetationtokeepthewaterinareasonablywholesomestate，withoutdrivingmetofrequentrenewalswhichwoulddisturbtheworkofmycolonies。Sanitationandquietarethefirstconditionsofsuccess。Nowthestockedpondwillnotbelonginfillingitselfwithgasesunfittobreathe，withputrideffluviaandotheranimalrefuse;itwillbecomeasinkinwhichlifewillhavekilledlife。Thosedregsmustdisappearassoonastheyareformed，mustbeburntandpurified;andfromtheiroxidizedruinstheremustevenriseaperfectlife—givinggas，sothatthewatermayretainanunchangeablestoreofthebreathableelement。Theplanteffectsthispurificationinitssewagefarmofgreencells。
　　Whenthesunbeatsupontheglasspond，theworkofthewaterweedsisasighttobehold。Thegreen—carpetedreefislitupwithaninfinityofscintillatingpointsandassumestheappearanceofafairylawnofvelvet，studdedwiththousandsofdiamondpin’sheads。Fromthisexquisitejewelrypearlsbreakloosecontinuouslyandareatoncereplacedbyothersinthegeneratingcasket;slowlytheyrise，liketinyglobesoflight。Theyspreadoneveryside。
　　Itisaconstantdisplayoffireworksinthedepthsofthewater。
　　Chemistrytellsusthat，thankstoitsgreenmatterandthestimulusofthesun’srays，theweedsdecomposethecarbonicacidgaswherewiththewaterisimpregnatedbythebreathingofitsinhabitantsandthecorruptionoftheorganicrefuse;itretainsthecarbon，whichiswroughtintofreshtissues;itexhalestheoxygenintinybubbles。Thesepartlydissolveinthewaterandpartlyreachthesurface，wheretheirfrothsuppliestheatmospherewithanexcessofbreathablegas。Thedissolvedportionkeepsthecolonistsofthepondaliveandcausestheunhealthyproductstobeoxidizedanddisappear。
　　OldhandthoughIbe，Itakeaninterestinthistritemarvelofabundleofweedsperpetuatinghygienicprinciplesinastagnantpool;Ilookwithadelightedeyeupontheinexhaustiblesprayofspreadingbubbles;Iseeinimaginationtheprehistorictimeswhenseaweed，thefirst—bornofplants，producedthefirstatmosphereforlivingthingstobreatheatthetimewhenthesiltofthecontinentswasbeginningtoemerge。WhatIseebeforemyeyes，betweentheglasspanesofmytrough，tellsmethestoryoftheplanetsurroundingitselfwithpureair。
　　CHAPTERVIIITHECADDISWORM
　　WhomshallIlodgeinmyglasstrough，keptpermanentlywholesomebytheactionofthewaterweeds？Ishallkeepcaddisworms，thoseexpertdressers。Fewoftheself—clothinginsectssurpasstheminingeniousattire。Thepondsinmyneighborhoodsupplymewithfiveorsixspecies，eachpossessinganartofitsown。Today，butoneoftheseshallreceivehistoricalhonors。
　　Iobtainitfromthemuddybottomed，stagnantpoolscrammedwithsmallreeds。Asfarasonecanjudgefromthehabitationmerely，itshouldbe，accordingtothespecialists，Limnophilusflavicornis，whoseworkhasearnedforthewholecorporationtheprettynameofPhryganea，aGreektermmeaningabitofwood，astick。Inanolessexpressivefashion，theProvencalpeasantcallsitlouportofais，louporto—caneu。Thisisthelittlegrubthatcarriesthroughthestillwatersafaggotoftinyfragmentsfallenfromthereeds。
　　Itssheath，atravellinghouse，isacompositeandbarbaricpieceofwork，amegalithicpilewhereinart，retiresinfavorofamorphousstrength。Thematerialsaremanyandsundry，somuchsothatwemightimaginethatwehadtheworkofdissimilarbuildersbeforeoureyes，iffrequenttransitionsdidnottellusthecontrary。
　　Withtheyoungones，thenovices，itstartswithasortofdeepbasketinrusticwicker—work。Thetwigsemployedpresentnearlyalwaysthesamecharacteristicsandarenoneotherthanbitsofsmall，stiffroots，longsteepedandpeeledunderwater。Thegrubthathasmadeafindofthesefiberssawsthemwithitsmandiblesandcutsthemintolittlestraightsticks，whichitfixesonebyonetotheedgeofitsbasket，alwayscrosswise，perpendiculartotheaxisofthework。
　　Pictureacirclesurroundedbyabristlingmassoftangents，orratherapolygonwithitssidesextendedinalldirections。Onthisassemblageofstraightlinesweplacerepeatedlayersofothers，withouttroublingaboutsimilarityofposition，thusobtainingasortofraggedfascine，whosesticksprojectoneveryside。Suchisthebastionofthechildgrub，anexcellentsystemofdefense，withitscontinuouspileofspikes，butdifficulttosteerthroughthetangleofaquaticplants。
　　Soonerorlater，thewormforsakesthiskindofcaltropwhichcatchesontoeverything。Itwasabasketmaker，itnowturnscarpenter;itbuildswithlittlebeamsandjoists——thatistosay，withroundbitsofwood，brownedbythewater，oftenaswideasathickstrawandafinger’s—breadthlong，moreorless——takingthemaschancesuppliesthem。
　　Fortherest，thereissomethingofeverythinginthisragbag:
　　bitsofstubble，fagendsofrushes，scrapsofplants，fragmentsofsometinytwigorother，chipsofwood，shredsofbark，largishgrains，especiallytheseedsoftheyellowiris，whichwereredwhentheyfellfromtheircapsulesandarenowblackasjet。
　　Theheterogeneouscollectionispiledupanyhow。Somepiecesarefixedlengthwise，othersacross，othersaslant。Thereareanglesinthisdirectionandanglesintheother，resultinginsharplittleturnsandtwists;thebigismixedwiththelittle，thecorrectrubsshoulderswiththeshapeless。Itisnotanedifice，itisafrenziedconglomeration。Sometimes，afinedisorderisaneffectofart。Thisisnotsohere:theworkoftheCaddiswormisnotamasterpieceworthsigning。
　　Andthismadheapingupfollowsstraightupontheregularbasketworkofthestart。Theyounggrub’sfascinedidnotlackacertainelegance，withitsdaintylaths，allstackedcrosswise，methodically;and，loandbehold，thebuilder，grownlarger，moreexperiencedand，onewouldthink，moreskilful，abandonstheorderlyplantoadoptanotherwhichiswildandincoherent！Thereisnotransitionstagebetweenthetwosystems。Theextravagantpilerisesabruptlyfromtheoriginalbasket。Butthatweoftenfindthetwokindsofworkplacedoneabovetheother，wewouldnotdareascribetothemacommonorigin。Thefactoftheirbeingjoinedtogetheristheonlythingthatmakesthemone，inspiteoftheincongruity。
　　Butthetwostoreysdonotlastindefinitely。Whenthewormhasgrownslightlyandishousedtoitssatisfactioninaheapofjoists，itabandonsthebasketofitschildhood，whichhasbecometoonarrowandisnowatroublesomeburden。Itcutsthroughitssheath，lopsoffandletsgothestern，theoriginalwork。Whenmovingtoahigherandroomierflat，itunderstandshowtolightenitsportablehousebybreakingoffapartofit。Allthatremainsistheupperfloor，whichisenlargedattheaperture，asandwhenrequired，bythesamearchitectureofdisorderedbeams。
　　Sidebysidewiththesecases，whicharemereuglyfaggots，wefindothersjustasoftenofexquisitebeautyandcomposedentirelyoftinyshells。Dotheycomefromthesameworkshop？Ittakesveryconvincingproofstomakeusbelievethis。Hereisorderwithitscharm，theredisorderwithitshideousness;ontheonehandadaintymosaicofshells，ontheotheraclumsyheapofsticks。Andyetitisallproducedbythesamelaborer。
　　Proofsabound。Onsomecasewhichoffendstheeyewiththewantofarrangementinitsbitsofwood，patchesareapttoappearwhicharequiteregularandmadeofshells;inthesameway，itisnotunusualtoseeahorridtangleofjoistsbracedtoamasterpieceofshellwork。Onefeelsacertainannoyanceatseeingtheprettysheathsobarbarouslyspoilt。
　　Thismixedconstructiontellsusthattherusticstackerofwoodenbeamsexcels，whenoccasionoffers，inmakingelegantshellpavementsandthatitpracticesroughcarpentryanddelicatemosaicworkindifferently。Inthelatterinstance，thescabbardismade，aboveall，ofPlanorbes，selectedamongthesmallerofthesepondsnailsandlaidflat。Withoutbeingscrupulouslyregular，thework，atitsbest，doesnotlackmerit。Thepretty，close—whorledspirals，placedoneagainsttheotheronthesamelevel，haveaverypleasinggeneraleffect。NopilgrimreturningfromSantiagodeCompostellaeverslunghandsomertippetfromhisshoulders。
　　Butonlytoooftenthecaddiswormdashesahead，regardlessofproportion。Thebigisjoinedtothesmall，theexaggeratedsuddenlystandsout，tothegreatdetrimentoforder。SidebysidewithtinyPlanorbes，eachatmostthesizeofalentil，othersarefixedaslargeasone’sfingernail;andthesecannotpossiblybefittedincorrectly。Theyoverlaptheregularpartsandspoiltheirfinish。
　　Tocrownthedisorder，thecaddiswormaddstotheflatspiralsanydeadshellthatcomeshandy，withoutdistinctionofspecies，provideditbenotexcessivelylarge。Inotice，initscollectionofbric—a—brac，thePhysa，thePaludina，theLimnaea，theAmbersnail[allpondsnails]andeventhePisidium[abivalve]，thatlittletwin—valvedcasket。
　　Landshells，sweptintotheditchesbytherainsaftertheinmate’sdeath，areacceptedquiteasreadily。IntheworkmadeoftheMollusk’scast—offclothing，IfindencrustedthespindleshelloftheClausilium，thekeyshellofthepupa，thespiralofthesmallerHelix，theyawningvoluteoftheVitrina，orglasssnail，theturretshelloftheBulimus[alllandsnails]，denizensallofthefields。Inshort，thecaddiswormbuildswithmoreorlesseverythingthatcomesfromtheplantorthedeadmollusk。Amongthediversifiedrefuseofthepond，theonlymaterialsrejectedarethoseofagravellynature。Stoneandpebbleareexcludedfromthebuildingwithacarethatisveryrarelyabsent。Thisisaquestionofhydrostaticstowhichwewillreturnpresently。Forthemoment，letustrytofollowtheconstructionofthescabbard。
　　Inatumblersmallenoughtoallowofeasyandpreciseobservation，Iinstallthreeorfourcaddisworms，extractedthismomentfromtheirsheathswitheverypossibleprecaution。Afteranumberofattemptswhichhaveatlastshownmetherightroad，Iplaceattheirdisposaltwokindsofmaterials，possessingoppositequalities;thesuppleandthefirm，thesoftandthehard。Ontheonehand，wehavealiveaquaticplant，suchaswatercress，forinstance，orombrelled’eau，havingatitsbaseatuftybunchoffinewhiterootsaboutasthickasahorsehair。Inthesesofttresses，thecaddisworm，whichobservesavegetariandiet，willfindatoneandthesametimethewherewithaltobuildandeat。Ontheotherhand，wehavealittlefaggotofbitsofwood，verydry，equalinlengthandeachpossessingthethicknessofagoodsizedpin。Thetwosortsofbuildingmaterialliesidebyside，minglingtheirthreadsandsticks。Theanimalcanmakeitschoicefromthelump。
　　Afewhourslater，havingrecoveredfromtheshockoflosingitssheath，thecaddiswormsetstoworktomanufactureanewone。Itsettlesacrossabunchoftangledrootlets，whicharebroughttogetherbythebuilder’slegsandmoreorlessarrangedbytheundulatingmovementofthehinderpart。Thisgivesakindofincoherentandilldefinedsuspendedbelt，anarrowhammockwithanumberofloosecatches;forthevariousbitsofwhichitismadeuparerespectedbytheteethandextendedfromplacetoplacebeyondthemaincordsoftheroots。Here，withoutmuchtrouble，isthesupport，suitablyfixedbynaturalmoorings。Afewthreadsofsilk，casuallydistributed，makethefrailcombinationatriflemoresecure。
　　Andnowtotheworkofbuilding。Supportedbythesuspendedbelt，thecaddiswormstretchesitselfandthrustsoutitsmiddlelegs，which，beinglongerthantheothers，arethegrapnelsintendedtoseizethingsatadistance。Itmeetsabitofroot，fastensontoit，climbsabovethepointgripped，asthoughitweremeasuringthepiecetoarequisitelength，andthen，withthefinescissorsofitsmandibles，cutsthestring。
　　Thereisatonceabriefrecoil，whichbringstheanimalbacktothelevelofthehammock。Thebitdetachedliesacrosstheworm’schest，heldinitsforelegs，whichturnit，twistit，waveitabout，layitdown，liftitup，asthoughtryingforthebestposition。Thoseforelegsmakeadmirablydexterousarms。Beinglesslongthantheothertwopairs，theyarebroughtintoimmediatecontactwiththoseprimordialimplements，themandiblesandthespinneret。Theirdelicateterminaljointing，withamovableandcrookedfinger，isthecaddisworm’sequivalentofourhand。Theyaretheworkinglegs。Thesecondpair，whichareexceptionallylong，servetospeardistantmaterialsandtogivetheworkerafirmfootingwhenmeasuringapieceandcuttingitwiththepliers。
　　Lastly，thehindlegs，ofmediumlength，affordasupportwhentheothersarebusy。
　　Thecaddisworm，Iwassaying，withthepiecewhichithasremovedheldcrosswisetoitschest，retreatsalittlewayalongitssuspendedhammockuntilthespinneretislevelwiththesupportfurnishedbytheclosetangleofrootlets。Withaquickmovement，itshiftsitsburden，getsitasnearlybythemiddleasitcan，sothatthetwoendsstickoutequallyoneitherside，andchoosesthespottoplaceit，whereuponthespinneretsetstoworkatonce，whilethelittleforelegsholdthescrapofrootmotionlessinitstransversalposition。Thesolderingiseffectedwithatouchofsilkinthemiddleofthebitandalongacertaindistancetotherightandleft，asfarasthebendingoftheheadpermits。
　　Withoutdelay，othersticksarespearedinlikemanneratadistance，cutoffandplacedinposition。Astheimmediateneighborhoodisstripped，thematerialisgatheredatayetgreaterdistanceandthecaddiswormbendsevenfartherfromitssupport，whichnowholdsonlyitslastfewsegments。Itisacuriousgymnasticdisplay，thatofthissoft，hangingspineturningandswaying，whilethegrapnelsfeelineverydirectionforathread。
　　Allthislaborresultsinasortofcasingoflittlewhitecords。
　　Theworklacksfirmnessandregularity。Nevertheless，judgingbythebuilder’smethods，Icanseethatthebuildingwouldnotbedevoidofmeritifthematerialsgaveitabetterchance。Thecaddiswormestimatesthesizeofitspiecesveryfairly;itcutsthemalltonearlythesamelength;italwaysarrangesthemcrosswiseonthemarginofthecase;itfixesthembythemiddle。
　　Noristhisall:themannerofworkinghelpsthegeneralarrangementconsiderably。Whenthebricklayerisbuildingthenarrowshaftofafactorychimney，hestandsinthecenterofhisturretandturnsroundandroundwhilegraduallylayingnewrows。
　　Thecaddiswormactsinthesameway。Ittwistsroundinitssheath;itadoptswithoutinconveniencewhateverpositionitpleases，soastobringitsspinneretfullfacewiththepointtobegummed。Thereisnostrainingofthenecktoleftorright，nothrowingbackoftheheadtoreachpointsbehind。Theanimalhasconstantlybeforeit，withintheexactrangeofitsimplements，theplaceatwhichthebitistobefixed。Whenthepieceissoldered，thewormturnsalittleaside，toalengthequaltothatofthelastsoldering，andhere，alonganextentwhichhardlyevervaries，anextentdeterminedbytheswingwhichitsheadisabletogive，itfixesthenextpiece。
　　Theseseveralconditionsoughttoresultinageometricallyordereddwelling，havingaregularpolygonasanopening。Thenhowcomesitthatthecylinderofbitsofrootissoconfused，soclumsilyfashioned？Thereasonisthis:theworkerpossessestalent，butthematerialsdonotlendthemselvestoaccuratework。Therootletssupplystumpsofveryunevenshapeandthickness。Theyincludebigandsmallones，straightandbent，simpleandramified。
　　Tocombineallthesedissimilarpiecesintoanorderlywholeishardlypossible，allthemoresoasthecaddiswormdoesnotappeartoattachverymuchimportancetoitscylinder，whichisatemporarywork，hurriedlyconstructedtoaffordaspeedyshelter。
　　Mattersareurgent;andverysoftfibers，clippedwithabiteofthemandibles，aremorequicklygatheredandmoreeasilyputtogetherthanjoists，whichrequirethepatientworkofthesaw。
　　Theinaccuratecylinder，inshort，heldinpositionbynumerousguyropes，isabaseuponwhichasolidanddefinitestructurewillrisebeforelong。Soon，theoriginalworkwillcrumbletoruinsanddisappear，whereasthenewone，apermanentstructure，willevenoutlasttheowner。
　　Theinsectsrearedinatumblershowyetanothermethodofbuildingthefirstdwelling。Thistime，thecaddiswormisgivenafewveryleafystalksofpondweed（Potamogetondensum）andabundleofsmalldrytwigs。Itperchesonaleaf，whichthenippersofthemandiblescuthalfacross。Theportionleftuntouchedwillactasalanyardandgivethenecessarysteadinesstotheearlyoperations。
　　Fromanadjoiningleafasectioniscutoutentirely，anangularandgoodsizedpiece。Thereisplentyofmaterialandnoneedforeconomy。Thepieceissolderedwithsilktothestripwhichwasnotwhollycutoff。TheresultofthreeorfoursimilaroperationsistosurroundtheCaddiswormwithaconicalbag，whosewidemouthisscallopedwithpointedandveryirregularnotches。Theworkofthenipperscontinues;freshpiecesarefixed，fromonetoanother，insidethefunnel，notfarfromtheedge，sothatthebaglengthens，tapersandendsbywrappingtheanimalinalightandfloatingdrapery。
　　Thuscladforthetimebeing，eitherinthefinesilkofthepondweedorinthelinsey—woolseysuppliedbytherootsofthewatercress，thecaddiswormbeginstothinkofbuildingamoresolidsheath。Thepresentcasingwillserveasafoundationforthestrongerbuilding。Butthenecessarymaterialsareseldomnearathand:youhavetogoandfetchthem，youhavetomoveyourposition，aneffortwhichhasbeenavoideduntilnow。Withthisobject，thecaddiswormcutsitsmoorings，thatistosay，therootletswhichkeepthecylinderfixed，orelsethehalf—severedleafofpondweedonwhichthecone—shapedbaghascomeintobeing。
　　Thewormisnowfree。Thesmallnessoftheartificialpond，thetumbler，soonbringsitintotouchwithwhatitisseeking。Thisisalittlefaggotofdrytwigs，whichIhaveselectedofequallengthandofslightthickness。Displayinggreatercarethanitdidwhentreatingtheslenderroots，thecarpentermeasuresouttherequisitelengthonthejoist。Thedistancetowhichithastoextenditsbodyinordertoreachthepointwherethebreakwillbemadetellsitprettyaccuratelywhatlengthofstickitwants。
　　Thepieceispatientlysawnoffwiththemandibles;itisnexttakenintheforelegsandheldcrosswisebelowtheneck。Thebackwardmovementwhichbringsthecaddiswormhomealsobringsthebitoftwigtotheedgeofthetube。Thereupon，themethodsemployedinworkingwiththescrapsofrootarerenewedinpreciselythesamemanner。Thesticksarescaffoldedtotheregulationheight，allalikeinlength，amplysolderedinthemiddleandfreeateitherend。
　　Withthepickedmaterialsprovided，thecarpenterhasturnedoutaworkofsomeelegance。Thejoistsareallarrangedcrosswise，becausethiswayisthehandiestforcarryingthesticksandputtingtheminposition;theyarefixedbythemiddle，becausethetwoarmsthatholdthestickwhilethespinneretdoesitsworkrequireanequalgrasponeitherside;eachsolderingcoversalengthwhichisseentobepracticallyinvariable，becauseitisequaltothewidthdescribedbytheheadinbendingfirsttothissideandthentothatwhenthesilkisemitted;thewholeassumesapolygonalshape，notfarremovedfromarectilinearpentagon，because，betweenlayingonepieceandthenext，thecaddiswormturnsbythewidthofanarccorrespondingwiththelengthofasoldering。Theregularityofthemethodproducestheregularityofthework;butitisessential，ofcourse，thatthematerialsshouldlendthemselvestoprecisecoordination。
　　Initsnaturalpond，thecaddiswormdoesnotoftenhaveatitsdisposalthepickedjoistswhichIgiveitinthetumbler。Itcomesacrosssomethingofeverything;andthatsomethingofeverythingitemploysasitfindsit。Bitsofwood，largeseeds，emptyshells，stubblestalks，shapelessfragmentsareusedinthebuildingforbetterorforworse，justastheyoccur，withoutbeingtrimmedbythesaw;andthisjumble，theresultofchance，resultsinashockinglyfaultystructure。
　　Thecaddiswormdoesnotforgetitstalents;butitlackschoicepieces。Giveitapropertimberyardanditatoncerevertstocorrectarchitecture，ofwhichitcarriestheplanswithinitself。
　　Withsmall，deadpondsnails，allofthesamesize，itfashionsasplendidpatchworkscabbard;withaclusterofslenderroots，reducedbyrottingtotheirstiff，straight，woodyaxis，itmanufacturesprettyspecimensofwickerworkwhichcouldserveasmodelstoourbasketmakers。
　　Letuswatchitatworkwhenitisunabletouseitsfavoritejoist。Thereisnopointingivingitclumsybuildingstones;thatwouldonlybringusbacktotheuncouthsheaths。Itspropensitytomakeuseofsoakedseeds，thoseoftheiris，forinstance，suggeststhatImighttrygrains。Iselectrice，which，becauseofitshardness，willbetantamounttowoodand，becauseofitscleanwhitenessanditsovalshape，willlenditselftoartisticmasonry。
　　Obviously，mydenudedcaddiswormscannotstarttheirworkwithbricksofthiskind。Wherewouldtheyfixtheirfirstlayer？Theymusthaveafoundation，quickandeasytobuild。Thisisoncemoresuppliedbyatemporarycylinderofwatercressroots。Onthissupportfollowthegrainsofrice，which，groupedoneatoptheother，straightorslanting，endbygivingamagnificentturretofivory。Nexttothesheathsmadeoftinysnailshells，thisistheprettiestthingwithwhichthecaddisworm’sindustryhasfurnishedme。Afinesenseoforderhasreturned，becausethematerials，regularandofidenticalcharacter，havecooperatedwiththecorrectmethodoftheworker。
　　Thetwodemonstrationsareenough。Sticksandgrainsofricemakeitplainthatthecaddiswormisnotthebunglerthatonewouldexpectfromthemonstrousbuildingsinthepond。ThoseCyclopeanpiles，thosemadconglomerations，aretheinevitableresultsofchancefinds，whichareusedforthebestbecausethereisnochoice。Thewatercarpenterhasanartofitsown，hasmethodandrulesofsymmetry。Whenwellservedbyfortune，itisquiteabletoturnoutgoodwork;whenill—served，itactslikeothers:theworkwhichitturnsoutisbad。Povertymakesforugliness。
　　Thereisanothermatterwhereinthecaddiswormdeservesourattention。Withaperseverancewhichrepeatedtrialsdonottire，itmakesitselfanewtubewhenIstripit。Thisisopposedtothehabitsofthegeneralityofinsects，whichdonotrecommencethethingoncedone，butsimplycontinueitaccordingtotheusualrules，takingnoaccountoftheruinedorvanishedportions。Thecaddiswormisastrikingexception:itstartsagain。Whencedoesitderivethiscapacity？
　　Ibeginbylearningthat，givenasuddenalarm，itreadilyleavesitsscabbard。WhenIgofishingforcaddisworms，Iputthemintinboxes，containingnoothermoisturethanthatwherewithmycatchesaresoaked。Iheapthemuploosely，toavoidanygrievoustumultandtofillthespaceatmydisposalasbestImay。Itakenofurtherprecaution。ThisisenoughtokeepthecaddiswormsingoodconditionduringthetwoorthreehourswhichIdevotetofishingandtowalkinghome。
　　Onmyreturn，Ifindthatanumberofthemhavelefttheirhouses。
　　Theyareswarmingnakedamongtheemptyscabbardsandthosestilloccupiedbytheirinhabitants。Itisapitifulsighttoseetheseevictedonesdraggingtheirbareabdomensandtheirfrailrespiratorythreadsoverthebristlingsticks。Thereisnogreatharmdone，however;andIemptythewholelotintotheglasspond。
　　Notoneresumespossessionofanunoccupiedsheath。Perhapsitwouldtakethemtoolongtofindoneoftheexactsize。Theythinkitbettertoabandontheoldcloutsandtomanufacturecasesnewfromtoptobottom。Theprocessisarapidone。Bythenextday，withthematerialswhereintheglasstroughabounds——bundlesoftwigsandtuftsofwatercress——allthedenudedwormshavemadethemselvesatleastatemporaryhomeintheformofatubeofrootlets。
　　Thelackofwater，combinedwiththeexcitementofthecrowdingintheboxes，hasupsetmycaptivesgreatly;and，scentingagraveperil，theyhavemadeoffhurriedly，doffingthecumbersomejacket，whichisdifficulttocarry。Theyhavestrippedthemselvessoastofleewithgreaterease。Thealarmcannothavebeenduetome:
　　therearenotmanysimpletonslikemyselfwhoareinterestedintheaffairsofthepond;andthecaddiswormhasnotbeencautionedagainsttheirtricks。Thesuddendesertionofthecribhascertainlysomeotherreasonthanman’smolestations。
　　Icatchaglimpseofthisreason，therealone。TheglasspondwasoriginallyoccupiedbyadozenDytisci，orwaterbeetles，whosedivingperformancesaresocurioustowatch。Oneday，meaningnoharmandforwantofabetterreceptacle，Iflingamongthemacoupleofhandfulsofcaddisworms。BlundererthatIam，whathaveIdone！Thecorsairs，hidingintheruggedcornersoftherockwork，atonceperceivethewindfall。Theyrisetothesurfacewithgreatstrokesoftheiroars;theyhastenandflingthemselvesuponthecrowdofcarpenters。Eachpirategrabsasheathbythemiddleandstrivestoripitopenbytearingoffshellsandsticks。Whilethisferociousenucleationcontinueswiththeobjectofreachingthedaintymorselcontainedwithin，thecaddisworm，closepressed，appearsatthemouthofthesheath，slipsoutandquicklydecampsundertheeyesoftheDytiscus，whoappearstonoticenothing。
　　Ihavesaidbeforethatthetradeofkillingcandispensewithintelligence。Thebrutalripperofsheathsdoesnotseethelittlewhitesausagethatslipsbetweenhislegs，passesunderhisfangsandmadlyflees。Hecontinuestotearawaytheoutercaseandtotugatthesilkenlining。Whenthebreachismade，heisquitecrestfallenatnotfindingwhatheexpected。
　　Poorfool！Yourvictimwentoutunderyournoseandyouneversawit。Thewormhassunktothebottomandtakenrefugeinthemysteriesoftherockwork。Ifthingswerehappeninginthelargeexpanseofapond，itisclearthat，withtheirsystemofexpeditiousremovals，mostofthelodgerswouldescapescot—free。
　　Fleeingtoadistanceandrecoveringfromthesharpalarm，theywouldbuildthemselvesanewscabbardandallwouldbeoveruntilthenextattack，whichwouldbebaffledafreshbytheselfsametrick。
　　Inmynarrowtrough，thingstakeamoretragicturn。Whenthesheathsaredonefor，whenthecaddiswormsthataretooslowinmakingoffhavebeeneatenup，theWaterbeetlesreturntotherockeryatthebottom。Here，soonerorlater，therearelamentablehappenings。Thenakedfugitivesarediscoveredand，succulentmorselsthattheyare，areforthwithtorntopiecesanddevoured。
　　Withintwenty—fourhours，notoneofmybandofcaddiswormsisleftalive。Inordertocontinuemystudies，Ihadtolodgethewaterbeetleselsewhere。
　　Undernaturalconditions，thecaddiswormhasitspersecutors，themostformidableofwhomappearstobetheWaterbeetle。Whenweconsiderthat，tothwartthebrigand’sattacks，ithasinventedtheideaofquittingitsscabbardwithallspeed，itstacticsarecertainlymostappropriate;but，inthatcase，anexceptionalconditionbecomesobligatory，namely，thecapacityforrecommencingthework。Thismostunusualgiftofrecommencingitpossessesinahighmeasure。IamreadytoseeitsorigininthepersecutionsoftheDytiscusandotherpirates。Necessityisthemotherofindustry。
　　Certaincaddisworms，oftheSericostomaandLeptocerusspecies，clothethemselvesingrainsofsandanddonotleavethebedofthestream。Onaclearbottom，sweptbythecurrent，theywalkaboutfromonebankofverduretotheotheranddonotthinkofcomingtothesurfacetofloatandsailinthesunlight。Thecollectorsofsticksandshellsaremorehighlyprivileged。Theycanremainonthelevelofthewaterindefinitely，withnoothersupportthantheirskiff，canrestinunsubmersibleflotillasandcanevenshifttheirplacebyworkingtherudder。
　　Towhatdotheyowethisprivilege？Arewetolookuponthebundleofsticksasasortofraftwhosedensityislessthanthatofthewater？Cantheshells，whicharealwaysemptyandabletocontainafewbubblesofairintheirspiral，befloats？Canthebigjoists，whichbreakinsouglyafashionthenonetoogreatregularityofthework，servetobuoyuptheover—heavyraft？Inshort，isthecaddiswormversedinthelawsofequilibriumanddoesitchooseitspieces，nowlighterandnowheavierasthecasemaybe，soastoconstituteawholethatiscapableoffloating？
　　Thefollowingfactsarearefutationofanysuchhydrostaticcalculationsintheanimal。