After Hurricane Georges in 1998, a flush of basal sprouts developed on mature stems in tropical dry forests of Puerto Rico.  This flush of sprouts was not related to the severity of damage incurred by stems.  Previous studies of hurricane effects have shown sprouting related to structural damage to stems, but not on undamaged stems.  It was hypothesized that the sprout response was caused by increased ethylene production at the base of stems.  Ethylene is produced as a response to wounding or gravitational displacement and can stimulate sprout growth as a thigmomorphogenetic response.  A series of experimental treatments simulating hurricane effects were applied to mature stems to determine which effects would stimulate ethylene production. Experimental treatments were defoliation, static displacement, defoliation with static displacement, dynamic displacement, and control.  Displacement treatments simulated the force applied on a tree by sustained winds of a category 3 hurricane and dynamic displacement simulated sustained winds of a category 1 hurricane with gusts.  Four common dry forest tree species were used in the experiment.  
Results/Conclusions

Each species produced the greatest amount of ethylene with the dynamic displacement treatment.  Other treatments produced ~50% less ethylene than did dynamic displacement.  Species responses to the treatments varied both in the amount of ethylene produced and which treatments had elevated ethylene levels relative to control trees.  Across all treatments and species, the maximum ethylene production was about 8 times greater than the minimum.  This study begins to explain the mechanism for a post-hurricane sprouting response, and to our knowledge it is the first time ethylene production has been measured in mature stems in tropical forest field studies.  Furthermore, this study demonstrates how regional disturbances like hurricanes interact with physiological responses to determine ecosystem structure at a landscape level.