ABSTRACT
In this chapter, the merit or quality of a single, specific hypothesis is critically evaluated using the stress-gradient hypothesis. While this is clearly not the most precise hypothesis, it is commonly used as an heuristic to interpret changes in plant-plant interactions at various points on gradients/continuums and, sometimes, on contiguous ranges within populations or communities. In it simplest form, the stress-gradient hypothesis predicts that the relative frequency of positive interactions between plants will increase with increasing consumer pressure or environmental stress (Bertness and Callaway 1994). This initial formulation is a hypothesis that generated (directly and indirectly) an extensive set of testable predictions, including the following:
7.1 Introduction .................................................................................................. 125 7.2 Purpose ......................................................................................................... 127 7.3 A Brief History of Gradients in Ecology: All the World Is a Gradient ........ 127
7.3.1 Definition of an Ecological Gradient ................................................ 127 7.3.2 Degree of Extraction ......................................................................... 128
7.4 How Gradients Are Used, Tested, and Quantified ....................................... 132 7.4.1 Gradients as a Hypothesis-Testing Tool ............................................ 132 7.4.2 The Concept of Gradient Length ...................................................... 132 7.4.3 The Importance of Species-Specific Responses to Gradients .......... 134
7.5 Stress as Shorthand for Environmental Variation: Should We Be Stressed about Using Stress? ........................................................................ 136
7.6 A Meta-Analytical Test of the Top 10 Published Studies of the Stress-Gradient Hypothesis ............................................................... 140
7.7 Why Use Gradients and How To Do So: Recommendations for Structuring Gradient Studies................................................................... 143
References .............................................................................................................. 144
The sign of net interactions will change with the environment, typically gradients (Brooker and Callaghan 1998; Callaway et al. 2002; Holmgren, Scheffer, and Huston 1997; Holzapfel and Mahall 1999; Pugnaire and Luque 2001).
