Figure 1 Red blood cell counts in mice as a function of the dose of chloramphenicol showing counts for individual mice with a line connecting the mean count at each dose level. Note that this example provides a better impression of the variability of the data than Figure 2 . Raw data from Festing MFW, Diamanti P, Turton JA. 2001 . Strain differences in haematological response to chloramphenicol succinate in mice: Implications for toxicological research. Food Chem Toxicol 39:375-383. From: Guidelines for the Design and Statistical Analysis of Experiments Using Laboratory Animals ILAR J. 2002;43(4):244-258. doi:10.1093/ilar.43.4.244 ILAR J | © Institute for Laboratory Animal Research

Figure 2 Same data as in Figure 1 , but just showing the group means and error bars of one standard error about each mean. This type of presentation is not recommended as it tends to obscure individual variability. From: Guidelines for the Design and Statistical Analysis of Experiments Using Laboratory Animals ILAR J. 2002;43(4):244-258. doi:10.1093/ilar.43.4.244 ILAR J | © Institute for Laboratory Animal Research

Figure 3 Control chart of micronuclei counts per 1000 polychromatic erythrocytes in 47 batches of five control mice and two batches of mice treated with 1,2-dimethylhydrazine. Such a chart provides one method of making use of relatively homogeneous sets of historical control data collected within a single laboratory over a long period of time. (Data used to illustrate the point, although the real time sequence was not available, raw data was extracted from: Morrison V, Ashby J. 1995 . High resolution rodent bone marrow micronucleus assays of 1,2-dimethylhydrazine: Implications of systemic toxicity and individual responders. Mutagenesis 10:129-135.) From: Guidelines for the Design and Statistical Analysis of Experiments Using Laboratory Animals ILAR J. 2002;43(4):244-258. doi:10.1093/ilar.43.4.244 ILAR J | © Institute for Laboratory Animal Research