Visual datasets have long been used to support teaching and learning in biology. The Oh Phlox! dataset emphasizes visual learning opportunities presented by individual plants encountered in the field. Not only can datasets provide visual practice before students go to the field, but they can encourage reflection about visual investigation and methodology including the use of applications like NIH Image or "graphics" packages like Adobe Photoshop as potential research tools. Oh Phlox! includes image files of individual leaves and "whole" views of 34 mature garden phlox plants. The images are labeled so that each leaf image can be linked to a specific plant image. Many are rather "unlovely" with medium resolution that limits visual examinations to macroscopic features. Students can sample any of the myriad features of this population, develop hypotheses, and use statistical programs to support their ideas. Investigations centered on this dataset might include some standard measures of physical traits such as number of leaves per plant, percent of leaves showing leaf miner damage, average surface area of leaves, or leaf damage per individual leaf miner as an estimate of feeding required by developing larva. Behavior could be studied by determining directionality of leaf miner trails or plant growth responses after endoparasite activity. Since the plant axis provides a sort of "timesect" - a transect through space and time for these leaves, students could describe the interval nature of leaf miner "attacks." Students could measure "green" or pixel density in new versus "old" leaves, analyze leaf shape as evidence of abiotic interactions such as nutritional deficiency, estimate variance of leaf form from "idealized" leaf or build a model of the "average" leaf from this population to compare with leaves shown in standard phlox images from identification keys. Student directed activities could extend well beyond this list.

Key words: education, field biology, image analysis, Phlox, visual dataset