One of the great problems of systematics concerns the 

 equating of hving forms with primitive types. We are all 

 inclined to represent the modern amphibians, reptiles, and 

 mammals as progressively more complex stages. This type 

 of staging is, however, extremely artificial and inaccurate. 

 One must necessarily move back in time and carefully fol- 

 low structural changes in each of these groups until a con- 

 cept of the actual or hypothetical ancestral type is achieved. 

 The ancestral tetrapod is not like a living amphibian. One 

 may then ask whether the use of "amphibian" to identify 

 the primitive tetrapod is proper. If the basis for defining an 

 amphibian is only its habit of laying eggs in water, eggs 

 which develop to a larval form which metamorphoses to an 

 adult, then perhaps the original tetrapod was an amphibian. 

 Again, the question of definition arises in the use of the term 

 metamorphosis. In the case of the salamander this term in- 

 fers only that the early stages have gills, while the adult 

 utilizes lungs, without undergoing any sudden or drastic 

 change in its way of life. In terms of the structure of 

 the adult modern amphibian, the primitive tetrapod was as 

 distinct from it as is the reptile or the mammal. 



Let us consider specific cases in which problems of defi- 

 nition or interpretation crop up. 



INTERRELATIONSHIPS OF AGNATHS 

 AND GNATHOSTOMES 



Perhaps the best marked division we have observed is 

 that between agnath and gnathostome. In spite of the ap- 

 parent difference of these two categories, there is consider- 

 able doubt as to how they should be identified. Generally 

 speaking, it is assumed that the agnaths preceded in time, 

 and represented a structural stage leading to, the gnatho- 

 stomes. If it is assumed that some sort of skeleton was 

 characteristic of the ancestral vertebrate, whether or not 

 this is of a procartilaginous material, cartilage or bone, then 

 this general type of skeleton should be common to these two 

 groups. The skeleton observed in the agnath should be of a 

 more primitive (or simpler) type than that of the gnatho- 

 stome and the skeleton of the latter should show evidence of 

 having been derived from the former. Looking at the cranial 

 structure of these two groups, one cannot visualize such an 

 interrelationship. It might be assumed that the.se two lines 

 became distinct before the development of the complex 

 skeletal system and, therefore, that the cranial skeleton 

 cannot be utilized for comparison in this respect. Such a 

 negative approach merely shifts the burden of proof to 

 some other area of information. 



Attempts to compare the various cartilaginous rods ob- 

 served in the living agnaths with the various arches of the 

 gnathostome suggest that, in fact, the gnathostome type is 

 the less modified and, in some ways, the more primitive. 

 However, the lack of articulation in the agnath branchial 

 rods would seem to be a less modified condition than the 

 articulated rods observed in the gnathostomes, although 



this form could also be achieved by carrying over the 

 embryonic form into the adult (paedomorphosis or neoteny). 

 This conflict of interpretations is not in itself a strange 

 situation, but rather is typical of what one encounters in 

 trying to analyze the features of living forms in terms of 

 ancestral or primitive types. 



Regarding the separation of the agnath and gnathostome 

 lines, one can search for evidence from other areas. The 

 lack of a premandibular aortic arch and the breaking up 

 of the mandibular arch in both gnathostome and agnath, 

 or the complexity of the cranial nerves, suggest that a com- 

 plex mouth, like that of the gnathostome, was a basic verte- 

 brate feature and that the first radiation of the vertebrates 

 involved loss of this type of mouth by the agnaths. Such a 

 feeding adaptation is matched by the several basic changes 

 in the jaws of the gnathostomes in the course of their 

 evolution. 



In terms of the various systems of soft parts, there is noth- 

 ing striking to support the direct derivation of the agnath 

 from the gnathostome, or the reverse. However, from analyz- 

 ing these systems, one does arrive at an image of the 

 primitive vertebrate, an organism from which both agnath 

 and gnathostome lines might have been derived with a 

 minimum of change (this ancestral type was partly de- 

 scribed at the end of Chapter 2). In comparisons with this 

 hypothetical primitive type the agnath may appear to be 

 more markedly specialized and modified than the gnatho- 

 stome. For example, in their venous systems the lamprey 

 and hagfish appear to have diverged from the basic verte- 

 brate pattern further than the gnathostome. The same might 

 be said of the digestive tract (loss of pancreas) and the 

 excretory system (development of a peculiar type of holo- 

 nephros). 



In viewing the agnath-gnathostome dichotomy, one is 

 impressed not by the positive nature of the evidence, but 

 rather by the number of individually inconclusive clews on 

 which one sort of interrelationship or another might be 

 founded. No feature positively delimits these two groups. 

 Each bit of evidence when added to others becomes more 

 convincing but only in terms of probability or possibility. 

 This taxonomic division, like most others, must be based on 

 a number of characteristics, some of which may be true 

 only of the majority of members. These features cannot 

 generally be identified as primitive or advanced. 



The theory of the agnath to gnathostome sequence is 

 founded in part on the observation that the agnaths occur 

 earlier in the fossil record. However, the first agnath fishes 

 were already a rather heterogeneous array when first en- 

 countered. It is certainly true that we do not know a great 

 deal about the structure of the first fishes, but there is 

 sufficient evidence to support the view that at least three 

 distinct types were present. These types differed in such 

 fundamental features as the histology of their bones; that is, 

 osteostracans had bone cells as opposed to the heterostracans 

 which had acellular, aspidin bones. These three distinct 

 types indicate that a great deal of vertebrate evolution had 



INTERRELATIONSHIPS OF AGNATHS AND GNATHOSTOMES • 443 



