5
OurconceptionofthenatureofOdoursmustbeanalogoustothatofSavours；inasmuchastheSapidDryeffectsinairandwateralike，butinadifferentprovinceofsense，preciselywhattheDryeffectsintheMoistofwateronly。WecustomarilypredicateTranslucencyofbothairandwaterincommon；butitisnotquatranslucentthateitherisavehicleofodour，butquapossessedofapowerofwashingorrinsing[andsoimbibing]theSapidDryness。
FortheobjectofSmellexistsnotinaironly：italsoexistsinwater。Thisisprovedbythecaseoffishesandtestacea，whichareseentopossessthefacultyofsmell，althoughwatercontainsnoair（forwheneverairisgeneratedwithinwateritrisestothesurface），andthesecreaturesdonotrespire。Hence，ifoneweretoassumethatairandwaterarebothmoist，itwouldfollowthatOdouristhenaturalsubstanceconsistingoftheSapidDrydiffusedintheMoist，andwhateverisofthiskindwouldbeanobjectofSmell。
ThatthepropertyofodorousnessisbasedupontheSapidmaybeseenbycomparingthethingswhichpossesswiththosewhichdonotpossessodour。Theelements，viz。Fire，Air，Earth，Water，areinodorous，becauseboththedryandthemoistamongthemarewithoutsapidity，unlesssomeaddedingredientproducesit。Thisexplainswhysea—waterpossessesodour，for[unlike’elemental’water]itcontainssavouranddryness。Salt，too，ismoreodorousthannatron，astheoilwhichexudesfromtheformerproves，fornatronisalliedto[’elemental’]earthmorenearlythansalt。Again，astoneisinodorous，justbecauseitistasteless，while，onthecontrary，woodisodorous，becauseitissapid。Thekindsofwood，too，whichcontainmore[’elemental’]waterarelessodorousthanothers。
Moreover，totakethecaseofmetals，goldisinodorousbecauseitiswithouttaste，butbronzeandironareodorous；andwhenthe[sapid]moisturehasbeenburntoutofthem，theirslagis，inallcases，lessodorousthemetals[thanthemetalsthemselves]。Silverandtinaremoreodorousthantheoneclassofmetals，lesssothantheother，inasmuchastheyarewater[toagreaterdegreethantheformer，toalessdegreethanthelatter]。