leaves visited after consuming a droplet decreased with 

 decreasing food concentration. Evidently a fly 

 "sensed" it was not worth the effort to spend a lot of 

 time looking for food when the most recent information 

 available to it indicated the food in the immediate 

 vicinity was likely to be of low quality, that is of low con- 

 centration per volume ingested. The results also indi- 

 cate thresholds of food concentration below which flies 

 were not stimulated to probe the leaf surface with the 

 proboscis to feed. Most flies accepted yeast at a concen- 

 tration about 10 times lower than they accepted su- 

 crose. Food acceptance thresholds can vary many-fold 

 depending on the general state of fly nutrition. Obvi- 

 ously, the limits of acceptance obtained correspond to 

 those of hungry immature flies. 



Detection and Ingestion Thresholds 



In our second test we used dry food and asked, what 

 are the smallest food quantities on a leaf surface a 

 hungry fly can detect? At the same time we asked, what 

 are the upper thresholds of food quantity that inhibit 

 further fly appetite and further food foraging behavior? 



We found that protein-deprived apple maggot flies 

 would regularly detect and ingest the smallest amount 

 of dry yeast treatment we offered (6^g, which is some- 



what less than about one thousandth 

 of the weight of an average fly). They 

 responded, provided that they walked 

 directly over the food so that their 

 tarsi came into contact with it. These 

 results confirm our observations 

 from nature (see first article in this 

 series) and from our tests using pot- 

 ted trees in field cages (see fourth 

 article in this series) that flies feed on 

 very small dry particles on leaf sur- 

 faces. They also indicate that in fu- 

 ture tests, we would need to present 

 flies with far smaller quantities of 

 food than we did here if we want to 

 approximate the true threshold of dry 

 food ingestion by hungry flies. We 

 find it remarkable that flies can detect 

 and eat such tiny particles of food, 

 which in future studies could possibly 

 turn out to be about equivalent to a 

 hungiy human feeding on grains of 

 rice one at a time. 



In relation to the largest amount 

 of food a fly can ingest in a meal, our 

 findingswere more definitive. Of flies 

 presented with concentrated dry 

 yeast particles of 0.25 mg, 83% were 

 able to consume them in a single meal lasting an aver- 

 age of 7.5 minutes. About 57% of flies consumed par- 

 ticles twice as large (0.5 mg), with the average feeding 

 time 11.9 minutes. Finally, only 20% of flies were able 

 to consume food particles of 1.0 mg, with an average 

 feeding time of 21.7 minutes. This would be about 

 equivalent to a human eating a meal of one sixth of his 

 weight. For liquid food, the amount that roughly 50% of 

 flies could injest was slightly higher (about 0.8 mg) 

 than in the case of dry food. Overall, flies could ingest 

 liquid food faster than dry food. Variation in fly size was 

 probably responsible for some of the differences ob- 

 served between flies. But the results show that like 

 Americans at Thanksgiving dinner, a fly can stuff itself 

 only so much (equal to about a 30 pound turkey, with 

 all the trimmings) before it is time to quit eating. 



Regurgitation Behavior 



We found that flies drinking a liquid food solution 

 were generally still hungry, even though fully en- 

 gorged. These flies regurgitated droplets of liquid food 

 and held them externally on their mouthparts for 

 varying periods of time after which they ingested again 

 some liquid food and regurgitated droplets once more. 

 We hypothesized that during this regurgitation or 



Fruit Notes, Summer, 1990 



