We analyze the structure of diameter, richness, and diversity of the forests in the upper limit of the great Amazon basin located in the Ecuadorian territory of the Cordilleras del Cóndor and Cutucú. Our hypothesis was that the forests of the eastern mountain ranges are not homogeneous, but rather present differences in their structure, richness, and floristic diversity. Our main objective was to classify the types of forests based on the characteristics of the diameter structure and the species composition of the Amazonian forests of the eastern mountain ranges in southern Ecuador, and we determined the influence of critical edaphic, environmental, and geomorphological factors, For this we installed eight permanent plots of one hectare in homogeneous and well preserved forest stands, four plots in the province of Zamora Chinchipe and four in the province of Morona Santiago. We identified and measured all trees > 10 cm at chest height and for each plot, soil samples, as well as environmental and slope data were taken. We performed an non-metric multidimensional scaling analysis (NMDS) analysis to evaluate changes in climatic and geomorphological gradients, and used the CCA analysis to assess the relationship between the composition of the species at the plot level and the edapho-climatic variables. Finally, we modeled the change in diversity ad species (Fisher's alpha) in relation to climatic, altitudinal, and geomorphological gradients using a GLM. We determined the existence of two different types of forest, the first called Terra Firme, characterized by the presence of a greater number of species and individuals per plot as compared to the second type of forest called Tepuy or Sandstone forest. Species richness was negatively correlated with the phosphorus content of the soil and the pH, annual average temperature, annual rainfall, and altitude. Terra Firme forests, settled in more stable and nutrient-rich climatic areas, were more diverse and Sandstone forests are poor in nutrients and develop in areas with greater seasonality.