PRESENT PROBLEMS OF PALEONTOLOGY. 235 



more or less central starting point in a smaller and older group which 

 contains a large number of primitive or generalized characters. 



The search for the primitive central form is always made by the 

 same method of reasoning, a method which was first clearly outlined by 

 Huxley, namely, by the more or less ideal reconstruction of the 

 primitive central form from which radiation has occurred. This is a 

 very difficult matter where the primitive central form is not preserved 

 either living or as a fossil. In such instances we may by analysis of 

 all the existing forms prophesy the structure of the primitive central 

 form, as Huxley, Kowalevsky and Cope did in the case of the hoofed 

 animals, a prophecy which was nearly fulfilled by the discovery in 

 northern Wyoming of Phenacodus. In other more fortunate cases the 

 primitive central form survives both living and fossil, as in the re- 

 markable instance of Paleohatteria of the Permian and the Tuatera 

 lizard (Hatteria) of New Zealand, which gave rise to the grand 

 adaptive radiation of the lizards, mosasaurs, dinosaurs, crocodiles, 

 phytosaurs and probably of the ichthyosaurs. 



In the reconstruction of these primitive central forms, we must 

 naturally discriminate between analogy and homogeny, and paleon- 

 tologists are not agreed in all cases on such discrimination. On the 

 border region, in fact, where the primitive central forms are still 

 unknown, where analogy has reached its most perfect climaxes and 

 imitations, are found the great paleontological controversies of to-day. 

 For example, among the paleozoic fishes, the armored ostracoderms 

 (Pteraspis, Cephalaspis, Pterichthys) and the arthognaths (Coccos- 

 teus, Dinichthys) by some authors (Hay, Eegan, Jaekel) are placed 

 in the single group of placoderms, while by other authors (Smith 

 Woodward and Dean) they are regarded as entirely independent and 

 superficially analogous groups. The dipnoi or lung fishes (Ceratodus, 

 Protopterus) present so many analogies with the Amphibians (sala- 

 manders and frogs) that they were long regarded as ancestors of the 

 latter ; but more searching anatomical and paleontological analyses and 

 recent embryological discoveries have proved that the dipnoi and 

 amphibia are parallel analogous groups descended alike from the 

 crossopterygian fishes, fishes which are now represented only by the 

 bichir (Polypterus) of Africa. It is interesting to recall paren- 

 thetically that two naturalists, Harrington, an American, and Budgett, 

 an Englishman, have given their lives to the solution of this problem 

 in searching for the embryology of Polypterus. The latter explorer 

 only was successful. 



Missing Links between the Great Classes of Vertebrates. 

 Among the varied fins of the crossopterygians we have nearly, but 

 not actually, discovered the prototype of the hand and the foot, the 

 fingers and toes of the primordial amphibian. Volumes upon volumes 



