Although the primary function of fleshy fruits is to attract seed dispersers, many ripe fruits contain toxic secondary compounds. A number of hypotheses have been proposed to explain this evolutionary paradox, most of which describe the potential adaptive role secondary compounds may play in seed dispersal. However, some authors have argued that fruit secondary compounds may be non-adaptive, and instead explain their occurrence as a pleiotropic consequence of selection for defense of leaves and other tissues. We address these alternative evolutionary hypotheses through a comparative examination of iridoid glycosides in leaves, unripe fruits, and ripe fruits of Lonicera x bella (Belle's bush honeysuckle), combined with an examination of fruit damage and removal in natural populations. We provide several lines of evidence that fruit secondary compounds cannot be explained solely as a consequence of foliar defense, including higher concentrations and more individual compounds in fruits compared to leaves and a negative relationship between iridoid glycoside concentration and fruit damage. However, we also show that the composition and concentrations of secondary compounds in leaves and fruits are not entirely independent, emphasizing that selection in different plant parts is intrinsically linked. We conclude that the adaptive significance of chemical traits is best considered in a whole-plant context that includes fruit-frugivore interactions.