EXPLANATION OF FOKM AND COLOUEING. 2l3 



continuous observation of individual unconfined birds and from special experi- 

 ments both on wild birds and on captives *. 



Experiments on wild birds are obviously quite indispensable, if only as a 

 check on the others, but on captive birds they can be carried out in far greater 

 numbers, with greater ease and precision, and with fewer insects than are 

 required for wild-bird experiments, so that it certainly seems well worth 

 enquirino- whether, conducted with care on suitable birds, suitably fed and 

 looked after, they need really be so very unreliable. The factors, apart from 

 lack of exercise and limited choice of food (with both of which I will deal 

 Liter), that would render them unreliable, seem likely to be : — Fretting-, ill- 

 health, fear, suspicion or (the opposite) over-confidence. 



An animal that mopes, whether from ill-health or rebellion at confinement, 

 and continues to mope, is useless for experiment of this kind — and often 

 unusable. As might be expected — for even with ourselves ease of mind is a 

 condition of good digestion, — its appetite is lacking or so capricious that it is 

 frequently quite impossible to assign a value to its acceptances or refusals. 

 The variation in individual animals, even of the same species, is sometimes great. 

 Thus of two kingfishers (both Halcyon cyanoleucus), one at once took most 

 kindly to captivity, showing no appreciable fear of me when I approached his 

 cage, and, on my offering him various insects with the forceps little more than 

 an honr after his capture, accepted and ate them most readily. The other 

 (captured at a later date) fluttered to get out, and when it finally gave this up, 

 simply pined away, refusing food or accepting it only in small quantities and 

 capriciously. I have had similar experiences with a few other birds, with a 

 lizard (Agama sp.), and with a toad. Needless to say, I claim no reliability 

 whatsoever for the " results " obtained from these animals. 



The restlessness that is sometimes present for a few days after capture (or — 

 at any rate, if not confined to the night-time — that which appears at migration- 

 time) is also a form of unsettlement that is inconducive to consecutive and 

 therefore reliable experimentation. Fear of the experimenter falls into the 

 same category so far as its result is concerned, but its continuance for more 

 than the first few days will in many cases (if I am safe in judging from my 

 own experience) probably be the fault of the experimenter. I have found 

 that the persistent quiet offering of pleasant insects through the bars by the 

 forceps to newly-captured birds has in all but very intractable cases sooner 

 or later evoked a response. This is usually at first a timid one, but, once it 

 has been obtained, a little more persistence on my part has almost invariably 

 soon begun to win the bird's confidence. Delay in obtaining this has usually 

 been simply the result of my being unable to devote the necessary time to 

 the bird. 



* These and certain other opinions offered in this introduction I hope to slate iu greater 

 detail and with such qualitications as may be necessary' at a future date. 



