EXPLANATION OF FOEM AND COLOURING. 205 



Here I inolnde also experiments on unconfined Grround-Hornbills and, 

 for comparison of its tastes with those of wild individuals of the same 

 species, my experiments on a captive Bee-eater. 



2. As a standard for comparison, my perhaps most complete series of 

 preference experiments — that on my captive Rollers. 



3. Experiments on insect-eaters other than birds, experiments on 

 carnivorous and other animals, including my experiments in connection 

 with plants. 



In this introduction I will deal with the objections which, to my knowledge, 

 have been brought against the experimental method. There will remain the 

 objections to a selectionist view — as " Hypertely," " valuelessness of incipient 

 resemblance/' "instinctive knowledge of food-values,''' "night-hunting of 

 the larger Oarnivora," and so on. I have special work bearing on all of 

 them and I hope to describe it in detail in some future publication, but I will 

 meanwhile state its bearing, quite briefly, in giving my conclusions at the 

 end of the third section. This third section does, I think, suggest that the 

 conclusions we may draw from our results on insects are probably capable of 

 a very extended application. It will be followed by a brief preliminary state- 

 ment of thebeai'ingot' the general results on the theories of animal coloration. 



There will still remain for later and separate publication : 



4. The remaining experiments on captive insectivorous animals and 

 on eaters of birds' eggs and insects' eggs. 



5. Such observations, fairly extensive, as have been made on the 

 habits and usual backgrounds of the various species of prey used in the 

 experiments ; and some account of the habits, relatively to them, of 

 their enemies. Final conclusions from the whole of the work. 



Captive insectivorous animals experimented on. 



My best series of preference experiments in insects have been on four 

 species of birds — on Lanius coUaris, Linn., and IHcrurus afer, Licht., amongst 

 Passerines, and on Halcyon cyanoleucus, VieilL, and Coracias garrulus, Linn., 

 in the Picarise. 



Other useful series of experiments, " pref ei'ence ^' or otherwise, have been 

 on a monkey {Ceixopithecus albogularis heirensis, Poc.) ; on four Passerine 

 birds [Emberiza major orientalis, Shelley, Phyllastreplius milanjensis, Shelley, 

 Pijcnonotus layardi, Gurney, and Crateropus kirJci, Sharpe), two Picarians 

 (^Lophoceros melanoleucus, Licht., and Alerops apiaster, Linn.), an Owl 

 (^Syrnium woodfordi, Smith), two Falconidse (Aqidla ivahlberyi, Sundev., and 

 Asturinida nionogram'}nica, Temm.), and a Francolin (^Francoliniis shelleyi, 

 Grant) ; on a lizard (JSiahuia striata^ and a chameleon ; and on a few Arthropods, 

 notably a Solpugid (Solpuga sagittaria, Poc), a Locustid [Enyaliopsis petersi^ 



16* 



