European Starling (Sturnus vulgaris)
Photo by PaulLomax (via Wikipedia)
One of the assumptions in Batesian mimicry is that maintenance of the mimic color pattern requires that the unpalatable “model” organism must be more abundant than its mimic. In “an oldie but a goodie” paper, J.V.Z. Brower (1960) described her experiments with European starlings (Sturnus vulgaris) used as predators, and painted mealworms (Tenebrio molitor) used as artificial models and mimics.
The starlings were wild-caught adults, divided into 5 groups that were exposed to different proportions of models and mimics (10%, 30%, 60%, or 90% mimics, and 100% mimics as control). Mealworms were classified as edibles (painted with an orange band, dipped in distilled water), models (painted with a green band, dipped in bitter quinine), and mimics (painted with a green band, dipped in distilled water). Following two series of experiments with different starlings, and careful statistical analyses, Brower found that if the color pattern mimicry was nearly perfect, the birds learned to associate the mimetic pattern with unpalatability, even if the models were relatively rare (40%). Moreover, mimicry was partially effective when the model was quite rare (10%), compared to the mimic. Another important finding in this paper was that the learned negative response to a visual stimulus could override a positive taste stimulus, as when the starlings pecked at, but nevertheless rejected, a mimic mealworm that had been dipped in distilled water.
Brower, J.V.Z. (1960). Experimental studies of mimicry. IV. The reactions of starlings to different proportions of models and mimics. Amer. Naturalist 94, 271-282.