Michael Halliday’s argument for the value of ‘trinocular vision’ in linguistic research has particular relevance to the observation, exploration and description of register. Taking each semiotic dimension relevant to the characterisation of register by turn, I begin by discussing Halliday’s proposition. I then proceed, using the metaphor of cartography, to examine register variation via the intersection of three semiotic dimensions: stratification, instantiation and metafunction. I discuss how such examinations enable us to create description maps of register variation. From this basis, I discuss a long-term programme of systematically producing descriptive maps of registers, which I and colleagues have begun. Finally, I suggest that by using such maps we can better understand such important phenomena as aggregates of registers and personal register repertoires.