We evaluate the influences of elevation and climate on the spatio-temporal distribution of wetland and dryland biomes during the Pennsylvanian and early Permian in tropical Pangea. The longstanding “upland model” places drought-tolerant vegetation in elevated habitats, where slope and drainage created moisture-limited substrates under a humid climate that simultaneously promoted peat accumulation in contemporaneous lowlands. Upland plants were periodically transported to, and buried in, lowlands. Rare preservation of dryland vegetation thus reflects its general absence in basins, and taphonomic vagaries of long-distance transport. The alternative “climate model” proposes that drought-tolerant plants dominated tropical habitats when climate was seasonally dry, with wetland vegetation reduced to scattered refugia. Environmental changes attending glacial-interglacial cycles caused alternating wetter-drier conditions, and the relative abundance of wetland versus dryland biomes in basinal lowlands thus varied with climatic oscillations. The paucity of drought-tolerant plants reflects a preservational megabias against habitats with seasonal moisture deficits.The environmental signal of “mixed” plant-fossil assemblages, comprising taxa characteristic of both wetland and dryland biomes, may help resolve these debates. We review key Pennsylvanian and lower Permian mixed assemblages from tropical Euramerican Pangea, and interpret their original habitats and climatic contexts based on multidisciplinary lines of evidence, including sedimentology, taphonomy, physiology, and paleoecology. Evaluations also consider patterns of vegetational distribution and taphonomy in modern tropical environments. We suggest that even a cursory view of current tropical plant distribution exposes flaws in the upland model. Where tropical climate is sufficiently humid to support peat swamps, slopes and elevated habitats do not host drought-tolerant vegetation, but are occupied by plants similar to those in lowland settings. This occurs because equable, high precipitation strongly dampens water-table variation across entire landscapes. Furthermore, taphonomic studies indicate that most plant-fossil assemblages record vegetation living near the burial site. Fossil floras thus reflect environmental conditions near their growth site, excluding an upland origin for most occurrences of drought-tolerant taxa. Conversely, the climate model is consistent with modern tropical vegetational distribution and soundly explains late Paleozoic floristic patterns. When Pangean tropical lowlands experienced seasonally dry conditions, plants tolerant of moisture deficits dominated most habitats, whereas wetland vegetation was restricted to wetter sites with greater preservation potential. This occurred because topographic variations are magnified under seasonal precipitation regimes, creating a complex habitat mosaic with wetland patches in a landscape subject to seasonal drought. Accordingly, we propose that a macrofloral assemblage with even rare drought-tolerant taxa indicates seasonality in the broader landscape.At larger spatio-temporal scales, disagreement also persists about whether tectonic uplift or long-term climatic drying was the primary driver of changes in late Paleozoic floristic patterns and areal extent of tropical peat swamps. We argue that tectonic activity alone cannot explain the drastic reduction in peat swamps or coincident changes in dominance-diversity of wetland vegetation. Rates of plant dispersal and evolution far outpace that of mountain building, and peat-forming wetlands persisted in elevated habitats well into the Late Pennsylvanian. Therefore, progressive late Paleozoic aridification was the most probable driver of changing floral patterns and the distribution of wetland and dryland biomes in tropical Pangea.