294 THE EVOLUTIONARY CAREER 



different way inasmuch as it gives a scale of the actual importance 

 of the various divergences. To this matter we return. 



And it will be seen that paleontology, being an imperfect 

 historical record, and incapable of being reconstructed where 

 the data have been destroyed, cannot of itself give us complete 

 phylogenies. But this may not be so with the data of morphology 

 — ^which may be sought in the intensive study of ontogenies and 

 anatomical comparisons. 



///. THE PALEONTOLOGICAL RECORDS 



The remains of animals and plants are found in the sedimentary 

 rocks and these remains are very often so perfectly preserved 

 that much of the coarse and fine anatomy, and even of the embryo- 

 logical phases of extinct organisms, can be elucidated. Paleonto- 

 logical records, when considered along with the morphology 

 of living organisms, may greatly extend the scopes of classifica- 

 tions ; thus some fossil animal may exhibit the anatomical 

 characters a b c d e (where the characters b c d e are those con- 

 sidered in the last sections. Obviously a may be a more 

 generalized character than b c for this reason — there are other 

 animals which exhibit the characters a^yd ; therefore the char- 

 acter a is common to the two categories abed and a^yd. The 

 fossil animal that exhibited a in addition to b c d therefore extends 

 the phylogeny of the category a b c d in that it links this up with 

 the category a^yd : it is therefore said to be an annectant form. 

 Thus the fossil bird Archeopteryx had teeth — and no living birds 

 have teeth. Reptiles have teeth and there are other anatomical 

 characters in respect of which Birds are akin to Reptiles. Arche- 

 opteryx is therefore an " annectant form," linking together Birds 

 and Reptiles. 



Also it is usually the case that the more generalized character 

 is also the more ancient one. Thus all living fishes have a 

 notochord — either in the adult, or in the embryonic phases, and 

 only some (though by far most) of living fishes have distinct, 

 segmental vertebras : therefore the notochordal, axial skeleton is 

 a more generalized character than the vertebral axial skeleton. 

 Now the axial skeleton of the most ancient (Silurian) fishes was 

 a notochordal one. 



And so the rational classification, based on a logical arrangement 



