Sedgwick-rafter cell counts: a procedural analysis

Information in the existing literature on some aspects of the collection and statistical analysis of Sedgwick-Rafter cell data appears contradictory, confusing, or absent. Using data from an experimental phytoplankton population as a basis, an investigation of S-R cell procedure has been undertaken with the following conclusions: I) settling time depends upon the type of preservation and the composition of the sample; 2) the field counting technique gives more accurate data and is less time consuming than the strip counting technique; 3) making fewer counts on each of a greater number of S-R cells gives more accurate results than making a greater number of counts on one or several S-R cells; 4) nonparametric methods offer a more convenient and nearly as efficient a means of detecting statistically significant differences as compared with parametric methods. A method is presented for optimally allocating counts within and among S-R cells for getting an estimator with the greatest precision in the least time.