A study tested the data-ink ratio theory, which holds that a reader's recall of quantitative data displayed in a graph containing a substantial amount of non-data-ink will be significantly less than recall from a graph containing little non-data-ink, as it might apply to graphics used in mass circulation newspapers. The experiment employed a between subjects, posttest only control group design. The stimuli were ten simple, horizontal bar graphs taken from the "Snapshot" section of "USA Today." All had familiar labels and no value was more than three digits long. The control stimuli were created to serve the role of identical graphs with much of the non-data-ink erased. The 120 observers were students in two sophomore level classes taught at the School of Journalism at Indiana University. The questions, which were constructed to measure the observer's accuracy in making comparisons between the data groups and accuracy in recalling the exact numbers associated with each data group, probed information such as number of bars displayed in the graph, labels associated with certain graphs, numerical value given to graphs, and relative length of bars in graphs. Observers were given 15 seconds to examine the data contained in the graphs. Statistical analysis of results showed that control groups and treatment groups made a nearly identical number of errors. Further examination of the results indicated that no single graph produced a significant difference between the control and treatment conditions. (One table of data is included and 10 pairs of graphs and 21 notes are attached.) (ARH)