Mid-Missouri experienced up to 2 min 40 s of totality at around solar noon during the total eclipse of 2017. We conducted the Mid-Missouri Eclipse Meteorology Experiment to examine land-atmosphere interactions during the eclipse. Here, research examining the eclipse responses in three contrasting ecosystems (forest, prairie, and soybeans) is described. There was variable cloudiness around first and fourth contacts (i.e., the start and end of partial solar obscuration) at the forest and prairie; however, solar irradiance (K ↓) signals during the eclipse were relatively clean. Unfortunately, the eclipse forcing at the soybean field was contaminated by convective activity, which decreased K ↓ beginning about an hour before first contact and exposed the field to cold outflow ~30 min before second contact. Turbulence was suppressed during the eclipse at all sites; however, there was also an amplified signal at the soybean field during the passage of a gust front. The standard deviations of the horizontal and vertical wind velocities and friction velocities decreased by ~75% at the forest (aerodynamically rough), and ~60% at the prairie (aerodynamically smooth). The eddy fluxes of energy were highly coherent with the solar forcing with the latent and sensible heat fluxes approaching 0 W/m2 and changing in direction, respectively. For the prairie site, we estimated a canopy-scale time constant for the surface conductance light response of 10 min. Although the eclipse imparted large forcings on surface energy balances, the air temperature response was relatively muted (1.5-2.5 °C decrease) due to the absence of topographic effects and the relatively moist land and atmosphere.