DISTRIBUTION OF BIRDS IN TIME AND SPACE. 183 



If our notions concerning the evolution of 

 higher from lower groups, and of their gradual 

 transformation into species, be correct, we should 

 expect to find that the further back in time we 

 could trace any given animal, the more unlike 

 its present self, so to speak, would it become. 

 At first the change would be hardly perceptible, 

 but at the last it would be marked indeed, so 

 much so, that could wo compare the ancestral 

 form with its latest descendants, we should find 

 that they would have but very little in common. 



The history of some of the hoofed mammalia 

 very well bears out this hypothetical case. 



With the birds, unfortunately, the "missing 

 links" are numerous. The marvel is that we 

 have any record at all. As that distinguished 

 authority on fossils, Mr Arthur Smith Woodward, 

 of the British Museum, has reminded us. " We 

 may . . . without exaggeration, declare that every 

 item of knowledge we possess concerning extinct 

 plants and animals depends upon a chapter of 

 accidents. Firstly, the organism must find its 

 way into water where sediment is being deposited, 

 and there escape all the dangers of being eaten ; 

 or it must be accidentally entombed in blown sand 

 or a volcanic accumulation on land. Secondly, 

 this sediment, if it eventually happens to enter 

 into the composition of a land area, must escape 

 the all-prevalent denudation (or destruction and 

 removal by atmospheric and aqueons agencies) 

 continually in progress. Thirdly, the skeletons 

 of the buried organism must resist the solvent 

 action of any waters which may percolate through 

 the rock. Lastly, man must accidentally excavate 



