Using output from a mesoscale model simulation, this paper describes the evolution of the three-dimensional temperature and humidity structures of an intense cyclone that developed over the eastern half of the United States during 14–16 December 1987. Some specific findings include the following. • The occlusion process at the surface appeared similar to the classical paradigm, with the cold front catching up with the warm front. Aloft, a warm-type occlusion structure formed as an upper baroclinic zone merged with the surface-based occluded front. • The frontogenesis associated with the cold and warm fronts was initially continuous, but as the system evolved, a break in the frontogenesis along the northern section of the cold front developed. A single frontogenesis feature appeared to support both the warm and occluded fronts. • A quasi-linear region of convective activity was observed in advance of the surface cold front. This convective line was associated with an upper-level humidity front, which originated from the confluence of descended, mid-, and upper-tropospheric trajectories and saturated, rising warm-sector trajectories. • Many of the structural elements of the storm can be explained by the differing air-parcel trajectories across these features. Although most trajectories can be meaningfully grouped into a limited number of families, they cannot be presented accurately in terms of only two or three conveyor belts or airstreams. • During the early part of the simulation, the upper and lower baroclinic zones were relatively distinct, with the upper baroclinic zone associated with an upper-level short wave-jet streak. The two baroclinic zones came together as the short wave overtook the low-level baroclinic zone. • The model simulation of the December 1987 cyclone, as well as the observed storm itself, suggests both similarities and differences with the “T-bone” conceptual model. An attempt is made to explain these differences based on the dithering environments in which the storms evolved.