populations. Most studies were done purposefully where 

 environmental variation was minimal, or in the case of 

 laboratory populations, the environment was held stable. 



Ecologists, especially these emphasizing the concept of 

 equilibrium (Caughley 1981a) , generally follow the tenet that 

 experimental perturbation of the population is necessary to 

 determine the effects of various factors on population 

 dynamics and their importance in regulation. Laboratory or 

 penned populations are often used because it is easier to hold 

 most factors stable while experimentally varying a single 

 factor. The problem with this approach is that it sacrifices 

 general applicability and reality to obtain strength in 

 precision. This problem has been recognized (Levins 1966, 

 Smith and Fowler 1981) but the complexities of dealing with 

 the multi-factorial "real world" have been daunting and the 

 current fad is precision. 



Van Ballenberghe (1980) characterized much of the work on 

 population regulation as ... "people trying to force 

 ecological events into the conceptual framework of classical 

 physics and systems theory." He also indicated that most work 

 stresses stability and employs models. Further, the model is 

 usually constructed first and the theorist scrambles to look 

 for field data that will confirm the model's predictions. 



Because our studies were supported by a management 

 oriented agency, our research contained an inherent bias 

 toward applicability in the "real world". Attempting to track 

 the many varying factors inherent in "real" populations was no 

 less a challenging task for us than for anyone else. We 

 found, however, that while uncontrolled variation ("chaos") is 

 intimidating and incomprehensible when followed for short 

 periods, patterns of population behavior begin to emerge after 

 longer periods of time. Conclusions may suffer some degree of 

 reduction in precision, but they are applicable to the real 

 situations in which management must occur. 



The most recent phase of our study (1975-1987) began with 

 an established "natural" population and the beginnings of 

 long-term data. As opposed to many previous study situations, 

 the environment on our study area was extremely variable. 

 Perhaps that, in itself, resulted in our somewhat different 

 conclusions about and interpretations of population 

 "regulation" in deer. For mule deer, at least, a variable 

 environment is not necessarily atypical. Thus, our results 

 and conclusions should have some general applicability. 



Because existing theories of population regulation and 

 management philosophy did not appear to explain mule deer 

 population trends, we collected and analyzed data with the 

 suggestion of Chitty (1967b) in mind: "Perhaps ecologists of 



15 



