The determination of frequency value (percentile limits) and the classification of the different variation factors allow us to define more and more homogeneous subpopulations as we use these factors for sorting. Using as our study population those persons coming to the Centre for Preventive Medicine, we were able to: (a) Describe and measure the significance and importance of physiological variations or of variations attributed to age--the latter largely related only to excessive weight, which it seems to us is often the case. (b) Establish a classification for variation factors; the recapitulatory table should be useful to clinical chemists in helping physicians interpret a laboratory test result that falls within the zone of incertitude. (c) Suggest a preliminary group of reference values for healthy subjects, to be used in interpreting a laboratory test in this way.