There is almost always some temptation to cheat on the basic principles of experimental design. You will likely be happier in the long run if you avoid this temptation.
All field trials should have untreated controls, replication and randomization. Trials testing the efficacy of a new nematicide or anything designed to replace the use of a nematicide will receive more acceptance if a standard chemical treatment is included.
With experience, one will learn the number of replications needed for a trial involving nematodes. Perusing work available in the literature will also provide some indications. Although as few as three replicates may be suitable in some cropping situations, four to five are more likely to be needed in many instances.
Some form of randomization of treatments is needed. A totally randomized design is acceptable, although with the variability possible within a field, a randomized complete block, or a design with blocking in two directions may provide better results.
The plot size needed to evaluate a nematode population is often smaller than that needed to obtain useful information on yields. Crop such as melons in which a single plant may spread 10 feet or more require larger plots than those with more compact plants. One should always consult an expert in the crop of interest to determine this information ahead of time. If commercial harvesting equipment is to be utilized, the procedures should be reviewed with the grower or operator ahead of time. One key question to ask is how many rows the equipment or pickers will harvest at one time. Having a plot one or two rows wide, will cause problems if the commercial equipment harvests three or four rows at a time.
Often, one can set up field trials more rapidly than they can be harvested and the samples processed and this should be considered when establishing plots.
Statistical analysis programs developed for computers can perform in a few seconds, analyses which would literally require weeks to perform by hand or with a calculator. For PC computers, SAS is the program in most frequent use by nematologists. For Macintosh computers, SuperAnova and JMP are two commonly used programs. Perusing the literature will acquaint one with the types of statistical analyses common for nematology experiments. Analysis of Variance followed by Duncan's New Multiple Range Test or Fisher's Least Significant Difference Test (LSD) are common procedures as is regression analysis. Nematode counts are often transformed by log or square root transformations to stabilize variances.