204 



NATURE 



[December 15, 1910 



Dr. Gaskell's programme of work at this immense problem 

 of the ancestry of vertebrates? 



Personally, were I concentrating in this way on the 

 study of Ammocoetes, I should be constantly oppressed 

 by an uneasy feeling of absolute uncertainty as to the 

 extent to which the c\clostomes are primitive in their 

 structure and to which specialised for their extra- 

 ordinary habits, unique amongst vertebrates. Certainly 

 we should expect a priori that vertebrates that took to 

 such a mode of life would tend to become extremely 

 specialised. We do not even know for certain whether 

 they are derived from jaw-possessing ancestors, though 

 we cannot help suspecting that a vertebrate taking 

 to the cyclostome mode of life would tend to lose its 

 hinged jaw and become, therefore, cyclostomatous. 



Another cherished principle of morphological research 

 receives a rude jar in Dr. Gaskell's sentence, " to me 

 and to all my friends who are accustomed to deal with 

 the vertebrate nervous system the explanation I have given 

 is so self-evident and natural that it is impossible to look 

 at the matter in any other way." Are myself and other 

 teachers of morphology talking nonsense when we urge 

 upon those commencing research work that the frame 

 of mind which they must endeavour to avoid at all cost, if 

 their work is to be of value, is that in which they come 

 to regard their working hypothesis as "so self^vident 

 and natural that it is impossible to look at the matter 

 in any other way "? 



An important principle is that enunciated bv Dr. 

 Gaskell when Tie says that " each higher group of animals 

 has arisen in succession from the highest race developed 

 up to that time, by highest meaning the group possessing 

 the best developed central nervous system." Apart from 

 the trivial point that many would differ from Dr. Gaskell 

 in their estimate of the trustworthiness of the palaeonto- 

 logical statements by which this principle is illustrated, 

 the principle itself seems to many a somewhat doubtful 

 one. I take it by " high development " we mean, roughlv, 

 " complexity of organisation." Now organisation is 

 upon the whole adaptive to functional activity. A highlv 

 organised animal' is, as a rule, one the details of the 

 structure of which are highly specialised in relation to 

 environmental conditions. Such adaptive specialisation of 

 the various organs renders their possessor peculiarly liable 

 to suffer from changes in environmental conditions. It 

 seems probable that changes of environmental conditions 

 form one of the chief factors in compelling evolutionarv 

 change. It is at such periods that natural selection is 

 most accentuated : the adaptable survive, the unadaptable 

 are exterminated. It appears quite credible that when 

 important geological or other environmental changes come 

 about it is those forms the organisation of which shows 

 the most complete linking up to the preceding set of 

 conditions which are exterminated, in other words, that it 

 is not the highest existing forms which proceed to evolve 

 into a state of adaptation to the new set of conditions. 



Apart from general principles, there are many important 

 statements as to fact which seem to me to call for com- 

 ment. Amongst these, for example, I read that " the 

 evidence of the rocks points to the Silurian age as the 

 time when the vertebrate first arose." To me it is simply 

 incomprehensible how any biologist can really believe that 

 the scales and teeth, comparable in complexity with those 

 of existing fishes, occurring in Silurian formations, can be 

 taken as having any bearing whatever upon the question 

 of the first appearance of vertebrates. There are indeed, 

 I think, many who feel compelled to admit that the 

 period of evolutionary time intervening between the first 

 appearance of the simplest chordates and the appearance 

 of these Silurian fishes may well have been as great or 

 greater than that which has intervened between the 

 Silurian times and the present day. 



Dr. Gaskell refers to zoologists " accepting as a 

 commonplace the manufacture Of a new organ for breath- 

 ing air instead of water in the transition from the fish 

 to the amphibian." He also says that the evidence seems 

 to him stronger that the vertebrate alimentary canal has 

 been formed from a pre-existing respiratory chamber 

 " than that an alimentary canal should have taken on a 

 respiratory function in its anterior end." I would onlv 

 remark that anyone who studies the evolution of t'le adult 

 vertebrate from the embryo ran see for himself, as a 



NO. 2146, VOL. 85] 



matter of fact, that the anterior end of the alimentar\ 

 canal does develop a set of gill clefts, i.e. a mechanism 

 which in fishes is respiratory. Further, the only morphu- 

 logist who bas had the opportunity of studying the develo])- 

 ment of the lung in the lowest lung-bearing vertebral' 

 asserts that the homology of the lung or swim-bladder 

 fishes with the lung of the higher vertebrates appears : 

 him to be beyond question, and he takes the view th. 

 the lung, instead of developing " in the transition fro) 

 fish to amphibian," was in all probability already presci 

 in the ancient common ancestral form from which tel' ■ 

 stomatous fish, lung-fish, and all the higher vertebral, 

 have been derived. 



Dr. Gaskell refers to the old idea that the infundibuhn 

 of the vertebrate brain represents the oesophagus of tli 

 invertebrate, and that the suprainfundibular part of t'l 

 brain represents the supracesophageal ganglia. Such an 

 idea would only be of value as a scientific hypothesis it 

 based upon the facts of the earlier stages of brain develop- 

 ment in the more primitive groups of vertebrates. I hav 

 personally studied the development of the brain in elasmo- 

 branchs, crossopterygians (Polypterus), actinopterygian 

 ganoids (Amia, Lepidosteus), lung-fishes (Ceratodus, 

 Lepidosiren, Protopterus), and urodele amphibians. The 

 phenomena seen in these forms do suggest certain con- 

 clusions as to the general morphology of the vertebrate 

 brain, as that the hemispheres are primitively paired or 

 that the primary subdivision of the brain is into two 

 rather than into three parts {" vesicles "). They do not 

 suggest, however, any such view as that referred to. On 

 the contrary, they appear to me to indicate that the part 

 of the central nervous system which in the annelid or 

 arthropod has become the supracesophageal ganglia has 

 in the vertebrates completely disappeared. 



Dr. Gaskell refers to those views which " turn the 

 animal topsy turvy, making the back of the invertebrate 

 correspond to the ventral surface of the vertebrate." As 

 a matter of fact, the only two stages of vertebrate ancestry 

 which may be regarded as established with a fair degree 

 of probability are : — (i) a protozoan stage recapitulated 

 in the unicellular zygote, and (2) a coelenterate stage re- 

 peated in the diploblastic stage of Amphioxus and in the 

 corresponding stages of lampreys, elasmobranchs, crosso- 

 pterygians: lung-fishes, and amphibians. We know, of 

 course, nothing of the details of structure of the diplo- 

 blastic ancestor, but it has been suggested by Sedgwick 

 and others that these ancestral forms passed through a 

 stage resembling in its general features the existing 

 actinozoan. This idea, which in my opinion is still a 

 perfectly reasonable working hypothesis, affords an 

 adequate explanation of developmental phenomena other- 

 wise extraordinary and ununderstandable, such as the 

 occasional occurrence of a mid-dorsal slit dividing the 

 central nervous rudiment and notochord into lateral 

 halves. It would clearly be unfair to state that such a 

 view is " doomed to failure " because it makes the back 

 of an invertebrate correspond to the ventral surface of 

 a vertebrate. The view does not assume the reversal of 

 the body. It merely refers back vertebrates and coelomate 

 invertebrates to a common ancestor in which there was 

 no ventral or dorsal surface — a form in which there was 

 a certain amount of concentration of the nervous system 

 in the region round the primitive -mouth or protostoma — 

 and suggests that in certain of the descendents of this 

 ancestral form the normal position of the bod}' is with 

 the neural surface beneath or ventral (annelids, arthro- 

 f>ods, molluscs), while in others (vertebrates) it is such 

 that the neural surface is uppermost or dorsal. 



A strong point about this view is that it suggests a 

 possible origin of the segmented character of the meso- 

 derm of the vertebrates. It is generally agreed that the 

 mesoderm is of enteroccelic nature, and it was emphasised 

 long ago by Sedgwick that just such a segmentation of 

 enteroccelic pouches is already present within the phylum 

 Ccelenterata in the Actinozoa. 



Prof. Starling in the discussion said : — " To an onlooker 

 like myself the striking resemblance between the earliest 

 fishes and the Arthropoda ... is striking evidence in 

 favour of Gaskell's theory." Morohologists. unfortu- 

 nately, have as yet no knowledge of the earliest fishes. 

 They are acouainted only with certain dermal skeletal 

 structures and a few impressions which give a vague 



