VEGETATIOXAL ANALYSIS 35 



ject to reconsideration by other investigators, who may reinter- 

 pret them in the light of additional experience or information. 



SAMPLING 



The need for quantitative records has made it necessary to give 

 serious consideration to methods of sampling. Usually the mem- 

 bers of an entire community cannot be counted or measured, and 

 even if this were done, the information would be no more useful 

 or significant than an adequate set of data acquired by proper sam- 

 pling. Since this is true, it becomes of prime importance to deter- 

 mine what constitutes an adequate sample in terms of the commu- 

 nity as a whole and how to obtain such a sample with a minimum of 

 effort. At best, sampling for vegetational data is tedious and time- 

 consuming; often it may be extremely hard work. Nevertheless, 

 sampling conserves both time and labor as compared with an at- 

 tempt to analyze a whole community, and its results are much 

 more significant than those obtained by mere observation. 



In this connection it should be emphasized that the early pro- 

 cedures of observation and reconnaissance are still of extreme im- 

 portance in determining where, how, and what to sample. These 

 activities are still a necessary part of community study although 

 they cannot be substituted for detailed analysis. They serve to 

 form a basis for theories or ideas that may in turn be substantiated 

 by quantitative evidence obtained by sampling. Preliminary recon- 

 naissance may likewise help to reduce the effort expended in sam- 

 pling. No sampling should be done without a thorough knowledge 

 of the history, physiography, and vegetation of the region as a 

 whole. Prior to sampling, the community should have been ob- 

 served repeatedly in different parts of its range and more particu- 

 larly under the varying local conditions where it exists. Finally, 

 the specific stand should be observed thoroughly to determine its 

 obvious variations, its extent, limits, and transitions to contiguous 

 communities. Then, knowing all this, together with the size of in- 

 dividual plants, the strata present, and the purposes for which the 

 sampling is to be done, one may plan his procedure in terms of the 

 desired results, the necessary degree of accuracy, and the time 

 available for doing the work. 



Ecologists call a sample area or plot a quadrat, and the method 



