Some observations on parental feeding in ring doves (Streptopelia roseogrisea) are reported. Lehrman (1955) advances an explanation of the development of parental feeding in ring doves by three experiments, in which the hormone prolactin was injected into ring doves. These birds, and appropriate control groups injected with distilled water, were then presented with seven day old squabs in a test situation. In experiment 1 Lehrman found that prolactin-injected birds fed young, and control birds did not. In experiment 2 he found that inexperienced birds, whether injected with prolactin or distilled water, did not feed seven day old squabs. In experiment 3, in which the crops of 12 birds were anesthetized with efocaine injected into the crop walls, and 12 birds were anesthetized elsewhere in the body, Lehrman found that 8 out of 12 birds whose crops were anesthetized did not feed young, while 2 out of 12 control birds injected elsewhere in the body did not feed the squabs.
From these results the conclusion is drawn that the engorged crop of ring doves serves as a drive stimulus for approach behavior to the squabs, and that the squabs, having by trial and error inserted their bills into the parents' mouth elicit vomiting (regurgitation). In this way the parent becomes conditioned to the squab as a tension-reducing stimulus, and will on subsequent occasions voluntarily approach the young to feed them.
Incidental observations collected during the course of another study demonstrate that Lehrman's explanation is not tenable. When mourning doves were hatched under ring dove foster parents, squabs were fed with crop milk or a clear watery liquid, irrespective of whether the foster parents had previous breeding experience, or at what stage of the parents' own incubation period these squabs hatched. If no cropmilk is in the foster parents' crops no increased sensitivity due to distension can be assumed; hence, the tension-reducing hypothesis becomes untenable. That regurgitation feeding is not the same as vomiting induced by the irritating bill of the squabs is shown by the observation that mechanical feeding devices used in hand-raising doves never elicit vomiting.
The use of seven day old partially feathered squabs as test objects rather than one or two day old down young, which alone may be the appropriate releaser for the feeding response, most likely accounts for the different results in Lehrman's study and the present one.
The conclusion is then that parental feeding in inexperienced ring doves is innate, and that an explanation in terms of trial and error learning is not acceptable.