An agar-based artificial diet containing carbohydrates, fats and twenty amino acids was constructed (Fig. 1). This diet is highly palatable and nutritionally complete forLimax maximus as demonstrated by significant ingestion on first encounter, consistent ingestion on subsequent days and good growth of young slugs fed this diet. Removing methionine, an essential amino acid, from the complete diet produces a food which is initially as palatable as the complete diet, but after one day's intake the amount of this deficient diet eaten is greatly reduced (Fig. 2). Removing alanine, a nonessential amino acid, does not produce any decrement in feeding relative to the complete diet (Fig. 5).
A single meal can be sufficient for establishing the aversion to the deficient diet (Fig. 5). Following seven days of feeding on the deficient diet the aversion is retained with little or no attenuation for at least 30 days (Fig. 3) and does not generalize to either a known ‘safe’ food (Fig. 3) or a novel food (Fig. 4). Evidence of a mild neophobia towards the artificial diet which attenuated after one or two meals was seen (Fig. 5).
The learned aversion to the deficient diet is reversible if slugs are repeatedly fed the complete diet following feeding on the deficient diet. Also, slugs initially fed the complete diet will develop an aversion to the methionine-deficient diet after sampling it (Fig. 7).
Slugs readily ate the artificial diets when these were offered 7 days post-hatch. The methionine-deficient diet however was not eaten in large amounts after the first meals and did not support growth (Fig. 9). Baby slugs fed the methionine-deficient diet for 10 days and then maintained on rat chow ate only small amounts when the deficient diet was presented again 126 days later, while baby slugs fed the complete diet or an alanine deficient diet for 10 days ate large amounts when these diets were presented 126 days later (Fig. 11 b).
Supplementing the methionine-deficient diet with an injection of methionine into the haemocoel one hour after the completion of a meal completely blocks the development of a learned aversion while injection ofLimax saline does not (Fig. 8).
These results are best explained by the hypothesis that the slugs acquire, post-ingestively, an aversion to the taste and probably the odor of the diet as the result of associative learning. The results of the experiments reported here indicate that there are substantial parallels at the behavioral level between mollusc and mammal with respect to post-ingestive feedback learning.