.-'O' 



NA TURE 



[Octoi;kr 8, 1896 



brain of the imago dillers from tliat of the larva only in its 

 increased growth and complexity. 



A striking instance of the same necessary law is seen in the 

 case of the transformation of the larval lamprey, or Ammocoetes, 

 into the adult lamprey, or Petroniyzon ; here also, by a process 

 of histolysis, most of Ihe organs of the head region of the 

 animal undergo dissolution and re-formation, while the brain 

 remains intact, increasing in size by the addition of new 

 elements, without any sign of preliminary dissolution. On the 

 other hand, when, as is the case in the Tunicates, the trans- 

 formation process is accompanied witli a degradation of the 

 central nervous system, we find the adult animal so hopelessly 

 degraded that it is impossible to imagine any upward evolution 

 from such a type. 



It is to my mind perfectly clear that, in searching among the 

 Invertebrata for the immediate ancestor of the Vertebrata, the 

 most important condition which such ancestor must fulfil is to 

 possess a central nervous system, the anterior part of which is 

 closely comparable with the brain region of the lowest verte- 

 brate. It is also clear on every principle of evolution that such 

 hypothetical ancestor must resemble the lowest vertebrate much 

 more closely than any of the higher vertebrates, and therefore 

 a complete study of the lowest true vertebrate must give the 

 best chance of discovering the homologous parts of the verte- 

 brate and the invertebrate. For this purpose I have chosen for 

 study the Ammocoetes, or larval form of the lamprey, rather 

 than Amphioxus or the Tunicates, for several reasons. 



In the first place, all the different organs and parts of the 

 higher vertebrates can be traced directly into the corresponding 

 parts of Petromyzon, and therefore of Ammocoetes. Thus, 

 every part of the brain and organs of special sen.se — all the 

 cranial nerves, the cranial skeleton, the muscular system, &c., 

 of the higher vertebrates can all be traced directly into the 

 corresponding parts of the lamprey. So direct a comparison 

 cannot be made in the case of Amphioxus or the Tunicates. 



Secondly, Petromyzon, together with its larval form, Ammo- 

 c<-"etes, constitutes an ideal animal for the tracing of the verte- 

 brate ancestry, in that in Amniocretes we have the most 

 favourable condition for such investigations, viz. a prolonged 

 larval stage, followed by a metamorphosis, and the consequent 

 production of the imago or Petromyzon — a transformation which 

 does not, as in the case of the Tunicates, lead to a degenerate 

 condition, but, on the contrary, leads to an animal of a distinctly 

 higher vertebrate type than the Ammoccetes form. As we shall 

 see, the Ammocostes is so full of invertebrate characteristics that 

 we can compare organ for organ, structure for structure, with the 

 corresponding parts of Limulus and its allies. Then comes that 

 marvellous transformation scene during which, liy a process of 

 histolysis, almost all the invertebrate characteristics are destroyed 

 or changed, and there emerges a higher animal, the Petromyzon, 

 which can now be compared organ for organ, structure for struc- 

 ture, with the larval form of the Amphibian ; and so through the 

 medium of these larval forms we can trace upwards without a 

 break the evolution of the vertebrate from the ancient king-crab 

 form. On the other hand, Amphioxus and the Tunicates are 

 distinctly degenerate ; it is easier to look upon either of them as 

 a degenerate Ammociete than as giving a clue to the ancestor of 

 the Ammoccete. It is to my mind surprising how difficult it 

 appears to be to get rid of preconceived opinions, for one still 

 hears, in the assertion that Petromyzon as well as Amphioxus is 

 degenerate, the echoes of the ancient myth that the Elasmo- 

 branchs are the lowest fishes, and the Cyclostomata their 

 degenerated descendants. 



The characteristic of the vertebrate central nervous system is 

 its tubular character : and it is this very fact of its formation as 

 a tube which has led to the disguising of its segmental character, 

 and to the whole difificulty of connecting vertebrates with other 

 groups of animals. The explanation of the tubular character of the 

 central nervous system is the keystone to the whole of my theory 

 of the origin of vertebrates. The explanation which I have given 

 differs from all others, in that I consider the nervous system to 

 be composed of two parts — an internal epithelial tube, surrounded 

 to a greater or less extent by a segmented nervous system ; and I 

 explain the existence of these two parts by the hypothesis that 

 the internal epithelial tube was originally the alimentary canal 

 of an arthropod animal, such as Limulus or Eurypterus, which 

 has become surrounded to a greater or less extent by the nervous 

 system. 



Any hypothesis which deals with the origin of one group of 

 animals from another must satisfy three conditions : — 



(i) It must be in accordance with the phylogenetic history oi 

 each group. It must therefore give a consistent explan.ation of 

 all the organs and li.ssues of the higher group which can be 

 clearly shown not to have originated within the group itself At 

 the .same lime, the variations which have occurred on the 

 hypothesis must be in harmony with the direction of variation in 

 the lower group, if not actually foreshadowed in that group. 



This condition may be called the Phylogenetic test. 



(2) The anatomical relation of parts must be the same in the 

 two groups, not only with respect to coincidence of topographical 

 arrangement, but also with respect to similarity of structure, 

 and, to a large extent, also of function. 



This condition may be called the Anatomical test. 



(3) The peculiarities of the ontogeny or embryological develop- 

 ment of the higher group must receive an adequate explanation 

 by means of the hypothesis, while at the same time they must 

 help to illustrate the truth of the hypothesis. 



This condition may be called the Ontogenetic test. 



I hope to convince you that all these three condition^ are 

 satisfied by my hypothesis as far as tlie head region of the 

 vertebrate is concerned. I speak onl> uf the head region at 

 present, because that is the part which I liave especially studied 

 up to the present time, and also because it is natural and con- 

 venient to consider the cranial and spinal nerves separately ; and 

 I hope to demonstrate to you that not only the nervous system 

 and alimentary canal of such a group of animals as the Giganto- 

 straca — i.e. Limulus and its allied forms — is to be found in the 

 head region of Ammocoetes, but also, as must logically follow, 

 that every part of the head region of Ammoccetes has its 

 homologous part in the prosomatic and mesosomatic regions 

 of Limulus and its allies. I hope to convmce you that 

 our brain is hollow because it has grown round the old 

 cephalic stomach ; that our skeleton arose from the modifica- 

 tions of chitinous ingrowths ; that the nerves of the medulla 

 oblongata — i.e. the facial, glosso-pharyngeal, and vagus nerves — 

 arose from the mesosomatic nerves to the branchial and opercular 

 appendages of Limulus, while the nerves of the hind brain are 

 derived from the nerves of the prosomatic region of Limulus ; 

 that our cerebral hemispheres are but modifications of the supra- 

 resophageal ganglia of a scorpion, while our eyes and nose are 

 the direct descendants of its eyes and olfactory organs. 



In the first place, I will give you shortly the reasons why the 

 central nervous .s.ystem of the vertebrate must be considered as 

 derived from the conjoined central nervous .system and alimentary 

 canal of an arthropod. 



Comparison of the Central Nei-ootis SysUni of Ammoca-tes with 

 the Conjoined Central Nen-ous System and Alimentary 

 Canal of an Arthropod Animal suih as Limulus. 



I. The phylogenetic test proves that the tube of the central 

 nervous system was originally an epithelial tube, surrounded to 

 a certain extent by nervous material. 



The anatomical test then proves that this epithelial tube 

 corresponds in its topographical relations to the nervous material 

 exactly with the alimentary canal of an arthropod in its relations 

 to the central nervous system ; and, further, that the topo- 

 graphical relations, structure, and function of the corresponding 

 parts of this nervous material are identical in the Ammocretes 

 and in the arthropod. 



We see from these diagrams, taken from Edinger, how the 

 greater simplicity of the brain region as we descend the vertebrate 

 phylum is attained by the reduction of the nervous material more 

 and more to the ventral side of the central tube, with the result 

 that the dorsal side becomes more and more epithelial, until at 

 last, as is seen in .\mmocretes, the roof of the epichordal portion 

 of the brain consists entirely of fold upon fold of a simple 

 epithelial membrane, interrupted only in one place by the cross- 

 ing of the I\'th nerve and commencement of the cerebellum. 

 In the prechordal part of the brain this simple epithelial portion 

 of the tube is continued on in the middle line as the first choroid 

 plexus of Ahlborn, and the lamina terminalis round to the ventral 

 side ; where, again, in the infundibular region, the epithelial 

 saccus vasculosus, which has been bediming more and more 

 conspicuous in the lower vertebrates, together with the median 

 tube of the infundibulum, testifies to the withdrawal of the 

 nervous material from this part of the brain, as well as from the 

 dorsal region. Further, as already mentioned in my previous 

 papers, the invasion of this epithelial tube by nervous material 

 during the upward development of the vertebrate is beautifully 

 shown by the commencing development of the cerebellar hemi- 



NO. 1406, VOL. 54] 



