Causes and Consequences of Skewed Sex Ratios at the Range Margins of a Dioecious Plant



  • Tom E.X. Miller, Ph.D. -  ( - Assistant Professor, Department of Ecology and Evolutionary Biology, Rice University
  • Aldo Compagnoni, Ph.D. -  ( - Postdoctoral Research Associate, Department of Ecology and Evolutionary Biology, Rice University

Understanding what limits the geographic ranges of species is one of the major ecological challenges of the 21st century. Global change will drastically modify the shape and size of species' ranges, potentially affecting patterns of epidemics, biodiversity conservation, and land resource management. Forecasting these global change effects requires knowledge of the determinants of species range limits. An oft overlooked mechanism of range boundary formation is mate limitation. In dioecious species (those with separate female and male individuals), skew in the sex ratio can decrease offspring production and result in negative population growth rates. Mate limitation might be common in dioecious plants whose sexes tend to segregate along environmental gradients (Bierzychudek & Eckhart 1988, Graff et al. 2013). This segregation might result from sex-specific habitat requirements or from one sex out-competing the other (Lloyd & Webb 1977, Obeso 2002). Thus, we hypothesize that skewed sex ratios near range margins can limit the geographic distribution of dioecious species. 


Our work addresses the causes and consequences of mate limitation and range boundary formation in the dioecious cool-season grass Poa arachnifera (Texas Bluegrass), a species endemic to the southern plains. Preliminary data suggest segregation of the sexes across the range of this species: western sites tend to be male-biased and eastern sites tend to be female-biased, and this gradient in sex ratio is strongest at higher latitudes (Figure 1). Our work at Lake Lewisville Environmental Learning Area (LLELA) will quantify the effects of sex ratio variation on population growth, an important step toward understanding the contribution of mate limitation to range boundaries. LLELA is an ideal site because it hosts natural populations of Texas Bluegrass and is located approximately in the middle of this species' range (Figure 2). Experiments conducted elsewhere will address the causes of sex ratio variation across the range.

Our experiment at LLELA will address two main questions: 1) Does the sex ratio (frequency of males relative to females) influence the population growth of Poa arachnifera? 2) Is one sex competitively superior to the other and, if so, by what mechanisms do the two sexes coexist? We

address these questions using an experimental design in which we vary both the frequency and density of females and males. Varying sex frequency addresses question 1, while varying both density and frequency allows to determine the competitive hierarchy and the likelihood of coexistence (Damgaard 1998).

Experimental Methods

We plan to establish 0.5 x 0.5 m plots with Texas Bluegrass densities of 4, 20, 60, 100, and 200 individuals per m2. At each density, we will impose 7 different male to female frequencies. The experiment will thus have 35 treatments, each replicated four times, for a total of 140 0.5 x 0.5 m plots. These will be spaced at a minimum distance of 8 m from each other in order to minimize pollination across them. The experiment will be contained within an area as small as 1 hectare (3 acres) on the eastern side of LLELA in a mixed prairie / mesquite savannah. We intend to carry out the experiment for two years (Fall 2013 โ€“ Spring 2015).

We will establish Texas Bluegrass individuals in September/October 2013 by transplanting tillers grown in Rice University's greenhouse facilities. These tillers derive from Texas Bluegrass individuals collected in Northern Texas. Before establishing plants, we will estimate the average tiller biomass by sub-sampling thirty male and thirty female tillers from the greenhouse propagation material. After raking plots to prepare the ground, we will plant tillers in a hexagonal pattern and we will water them to promote establishment. To distinguish female and male individuals, we will mark each tiller using colored craft sticks. We will cover all experimental plots with shade cloth (40% light reduction) fixed on top of four, 0.7-m long wooden sticks driven 30 cm into the ground. We plan to shade plots because Texas Bluegrass usually occurs below tree canopies. During the two growing seasons of the experiment, we plan to hand-weed plots and a 0.2 m buffer around them every 1-2 months.

In spring 2013 we will count the number of female and male inflorescences, and the number of female stalks successfully producing seed. At the end of spring 2014, we will harvest the surviving Texas Bluegrass individuals in the experimental plots to measure: 1) male and female above-ground biomass, 2) seed production by females and biomass of male and female reproductive structures, 3) the number of male and female tillers. We will also remove colored craft sticks, wooden sticks and shade cloth.

We request 24 hr / 7 d access to Lewisville Lake Environmental Learning Area, if possible. Field work will occur in short, intense pulses. Because we are located in Houston (~ 4 hrs away), flexible access for arrivals in the evenings or on the weekends would be greatly beneficial.


Bierzychudek, P. and V. Eckhart (1988) Spatial Segregation of the Sexes of Dioecious Plants. The American Naturalist, 132, 34-43.

Damgaard, C. (1998) Plant competition experiments: testing hypotheses and estimating the probability of coexistence. Ecology, 79, 1760-1767.

Graff, P., F. Rositano, and M. R. Aguiar (2013) Changes in sex ratios of a dioecious grass with grazing intensity: the interplay between gender traits, neighbour interactions and spatial patterns. Journal of Ecology, in press.

Lloyd, D, and C. Webb (1977) Secondary sex characters in plants. The Botanical Review, 43, 177โ€“216.

Obeso, J. R. (2002) The costs of reproduction in plants. New Phytologist, 155, 321โ€“348.