Hetero-and conspecific interactions, nutrient availability, climate, habitat heterogeneity and disturbances can generate variation and spatial patterns of femaleness in plants. We assessed whether year, site, plant size, plant density and canopy area of conspecific neighbors influenced the expression and spatial aggregation of femaleness in Croton aff. wagneri, a monoecious shrub from dry shrublands of the inter-Andean valleys in Ecuador. We georeferenced in two sites (1,700 and 1,400 m.a.s.l) in five 10x10 m plots, within each site, the position of each Croton reproductive plant during first part of flowering season in two years, and measured their height, length and width. The femaleness index of each plant was determined by the number of female and male buds and flowers. Plant density was determined for each plot, along with the number of neighbors and the summed canopy area of conspecific neighbors (at 1.0, 2.0, and 2.5 m radius, and the five closest plants) from each focal plant. Croton's femaleness at the lower elevation site was greater than at the higher elevation site and increased with plant size and with canopy of the closest five neighbors. Soil at the lower elevation site had higher temperatures and lower water content. Aggregate patterns of femaleness were found in more plots at the lower elevation site. Our results indicate that location, plant size and canopies of conspecific neighbors of Croton can affect femaleness and its aggregation and support the hypothesis that femaleness can be influenced by facilitative interactions.

One site was located at 1700 a.s.l. and identified as the higher elevation site and the other at 1400  a.s.l., identified as the lower elevation site. At each site and within a 70x50 m plot, five subplots of 10x10 m were established in 2013 and 2016. The subplots were distributed as follows: four in the corners and one in the center with respect to the larger plot. In each plot, the spatial location of the main stem of each reproductive plant was tagged and georeferenced. To estimate plant size, plant height as well as length and width of canopy of each sampled plant were measured. During the flowering season (November - December) in both 2013 and 2016, five inflorescences were randomly selected on each plant of each 10x10 m plot in both sites (1145 plants and 5725 total inflorescences). Then, number of female and male buds and number of female and male flowers of each inflorescence were counted in situ and an average of female and male flowers per inflorescence was obtained from each plant. The standardized phenotypic femaleness index (Gi) of each plant was used to estimate the sexual expression of Croton, according to Lloyd (1980).