Correlations between reported experimental factors and mean midpoints of activation (μ _a ) and inactivation (μ _i ), indicated by an orange linear regression line. The number of data points and the coefficient of determination are shown above each panel. Factors shown are the steepness of the activation curve (k_a); the approximate magnitude of a “representative” current, if one was shown; the holding potential in the activation protocol (V_hold,a) and inactivation protocol (V_hold,i); and external and internal concentrations of sodium, calcium, and chloride. The internal calcium concentrations shown were calculated using Maxchelator (Bers et al, 2010). This figure has two major caveats: (1) the variable on the x axis is not the only one varied between experiments, and since experimental design choices are often inherited from previous work, we can also expect them to show some correlation (e.g., copying both a holding potential and bath/pipette solutions from the same seminal work); (2) some choices are so common that the “groups” on the x axis are very small, making correlations more spurious. For example, only 27 experiments used a nonzero [Ca²⁺]_i.