- 1McAtee has tried to show that protective adaptations in insects are non-effective against the predations of birds, and his corollary is that predation in birds is proportional to availability. His methods and deductions are shown to be faulty.
- 2Predation is clearly influenced by availability, but there is no proof that it is exactly proportional to availability. Given the extent to which predation is proportional to availability, one would still not be in a position to deny the selective value of alleged protective adaptations, since the percentage protection necessary for selective value is not known. It is very difficult to establish the numbers of available food animals in a district accurately because the presence of the observer himself upsets the numbers which existed before his arrival. The best chance would appear to be in a desert or polar region.
- 3The stomachs from Jidda form a nucleus for such a study, and, pending an ecological survey of the Jidda region, an attempt has been made to see how far McAtee's assertions may be proved or disproved from internal evidence alone, i. e., from considerations of the stomach-contents. The results do not show anything conclusive, but from the abundance of “protected” forms it appears the percentage protection referred to in 2 must be very small.The following insects, usually considered to be protected, formed the bulk of the food: —Ants.—Eaten as a staple food by 47 per cent. of the birds (30 out of 44 species). These were nearly all workers and soldiers, and over 50 per cent, of the ants were Myrmecines (which have a sting).Coleoptera.—The chief families found were Curculionidæ, Tenebrionidæ, and Histeridæ, all of them considered distasteful to birds on account of their hardness.Hymenoptera (other than ants).—Were very plentiful, stings being found frequently.
- 4The individual tastes of birds of the same species were considered. Most species exhibit a basic preference which may be overruled by special abundance of other prey, but this other prey is not necessarily of a kind considered unpalatable, merely not so usual in the diet. Hence the feeding, though to a certain extent influenced by availability, is not entirely dictated by it.
- 5The work of sorting the débris from a stomach was facilitated by having a trough whose width was the same as the diameter of the field of the binocular microscope. In this way the stomach-contents could be examined in alcohol with the certainty that nothing was missed and that no part of the contents was looked at more than once.Lepidoptera fared worse than other insects in the stomachs, and often the only trace of lepidopterous food was a few wing-scales floating on the surfàce of the alcohol. These could often be detected only with a good binocular, and since they are out of the focal plane of the rest of the stomach-contents (which do not float), they might easily be missed. This may account for the fact that McAtee only found Lepidoptera in 87 out of 80,000 stomachs, whereas they occurred in nearly 121/2 per cent. of the Jidda stomachs.