Among fog oasis conservationists in Peru, coastal fog is envisaged as a potential water source for reforestation purposes. Enacting it as water involves field studies intended to detect the locations where ground touching clouds are such that they can be captured and contained. Along the process, fog emerges disparately through the devices deployed, each form by way of a particular relation. Rather than offering mutually exclusive views on this atmospheric phenomenon, one relation becomes the ground for the next, each adopted successively and heuristically for the purpose of rendering fog catchable. This process helps elucidate the sometimes contradictory forms that atmospheric phenomena take in social theory, where in recent years the atmospheric is increasingly being used as theory machine for thinking about other things. With the help of visuals documenting the fog capture process, this talk posits empirical and conceptual approaches to the atmospheric as analogous, thus suggesting that the atmospheric as theory machine is equally relational. Akin to how fog is enacted relationally and differently by my interlocutors, atmospheric thinking in social theory depends on the technopolitical apparatuses used to think the atmospheric with. It follows that atmospheric thinking is an operation whereby one aspect of the empirical is held still for the purpose of describing something else. Understanding “the elemental” in this way is crucial for allowing situated environmental imaginaries to properly inflect upon our own quests for novel approaches to rapid ecological change.