Aliphatic ketones are widely present in body-borne and food odors of primates. Therefore, we used an operant conditioning paradigm and determined olfactory detection thresholds in four spider monkeys for a homologous series of aliphatic 2-ketones (2-butanone to 2-nonanone) and two of their isomers (3- and 4-heptanone). We found that, with the exception of the two shortest-chained ketones, all animals detected concentrations <1ppm (parts per million), and with five odorants individual animals even reached threshold values <0.1ppm. Further, we found a significant correlation between olfactory sensitivity of the spider monkeys and carbon chain length of the 2-ketones which can best be described as a U-shaped function. In contrast, no significant correlation was found between olfactory sensitivity and position of the functional carbonyl group. Across-odorant and across-species comparisons revealed the following: spider monkeys are significantly less sensitive to the odors of aliphatic ketones than to the odor of other classes of aliphatic compounds (1-alcohols, n-aldehydes, n-acetic esters, and n-carboxylic acids) sharing the same carbon length. Spider monkeys do not differ significantly in their olfactory sensitivity for aliphatic ketones from squirrel monkeys and pigtail macaques, but are significantly less sensitive to these odorants compared to human subjects and mice. These findings support the notion that neuroanatomical and genetic properties do not allow for reliable predictions with regard to a species' olfactory sensitivity. Further, we conclude that the frequency of occurrence of a class of odorants in a species' chemical environment does not allow for reliable predictions of the species' olfactory sensitivity.