TRANSACTIONS OF SECTION D. 787 



I should have hesitated to difterentiate it. la the kestrel and the great spotted 

 woodpecker there are differences which suggest incipient species, while the forests 

 of the wooded western islands yield two very peculiar pigeons, differing entirely 

 from each other in their habits, both probably derived from our wood-pigeon, but 

 even further removed from it than the Columha trocaz of Madeira, and, by their dark 

 chestnut coloration, suggesting that peculiar food — in this case the berries of the tree 

 laurel — has its full share in the differentiation of isolated forms. If we remember 

 the variability of the pigments in the food of birds, and the amount absorbed and 

 transferred to the skin and plumage, the variability in the tints and patterns of 

 many animals can be more readily understood. 



One other bird deserves notice, the Caccabis, or red-legged partridge, for here, 

 and here alone, we have chronological data. The Spaniards introduced Caccabis 

 rufa into Canary, and C. 2)etrosa into Tenerife and Gomera, and they have never 

 spread from their respective localities. Now, both species, after a residence of only 

 400 years, have become distinctly modified. C. rufa was introduced into the Azores 

 also, and changed exactly in the same manner, so much so that Mr. Godman, some 

 years ago, would have described it as distinct, but that the only specimen he 

 procured was in moult and mutilated, and his specimen proves identical with the 

 Cauarian bird. Besides minor differences, the beak is one-fourth stouter and longer 

 than in the European bird, the tarsus very much stouter and longer, and the 

 back is grey rather than russet. The grey back harmonises with the volcanic dark 

 soil of the rocks of the Canaries, as the russet does with the clay of the plains of 

 England and France. In the Canaries the bird lives under ditferent conditions 

 from those of Europe. It is on the mountain sides and among rocks that the 

 stouter beak and stronger legs are indispensable to its vigorous existence. It is 

 needless to go into the details of many other species. We have here the effect of 

 changed conditions of life in 400 years. What may they not have been in 4000 

 centuries ? We have the result of peculiar food in the pigeons, and of isolation in 

 all the cases I have mentioned. Such facts can only be supplied to the generaliser 

 and the systematist through the accurate and minute observations of the field 

 naturalist. 



The character of the avifauna of the Comoro Islands, to take another insular 

 group, seems to stand midway in the differentiating process between the Canaries 

 and the Sandwich Islands. From the researches of M. Humblot, worked out by 

 MM. Milne-Edwards and Oustalet, we find that there are twenty-nine species 

 acknowledged as peculiar ; two species from South Africa and twenty-two from 

 Madagascar in process of specification, called by M. Milne-Edwards secondary or 

 derived species. 



The little Christmas Island, an isolated rock 200 miles south of Java, only 12 

 miles in length, has been shown by Mr. Lister to produce distinct and peculiar 

 forms of every class of life, vegetable and animal. Though the species are few 

 in number, yet every mammal and land bird is endemic ; but, as Darwin remarks, 

 to ascertain whether a small isolated area, or a large open area like a continent, 

 has been more favourable for the production of new organic forms, we ought to 

 make the comparison between equal times, and this we are incapable of doing. 

 My own attention was first directed to this subject when, in the year 1857-58, 1 

 spent many months in the Algerian Sahara, and noticed the remarkable variations 

 in different groups, according to elevation from thesea, and the difference of soil and 

 vegetation. The ' Origin of Species ' had not then appeared ; but on my return 

 my attention was called to the communication of Darwin and Wallace to the 

 ILinnean Society on the tendencies of species to form varieties, and on the perpetua- 

 tion of varieties and species by means of natural selection. I then wrote : ' ' It is 

 hardly possible, I should think, to illustrate this theory better than by the larks 

 and chats of North Africa. In all these, in the congeners of the wheatear, of the 

 rock chat, of the crested lark, we trace gradual modifications of coloration and 

 of anatomical structure, deflecting by very gentle gradations from the ordinary 

 type, but, when we take the extremes, presenting the most marked differences. . . . 

 In the desert, where neither trees, brushwood, nor even undulations of surface 



' Bm, 1859, pp. 429-433. 



3 E 2 



