C— GEOLOGY 6 1 



structure.' This is not, however, always the case. It is true for such 

 organisms as belemnites and trilobites in which only aduh features are 

 visible in the fully grown specimen. But it is not quite true for organisms, 

 like ammonites and gastropods, for which in any given specimen large 

 portions of the ontogenetic record are exposed to view, and provide 

 juvenile features which become automatically incorporated into the 

 diagnosis of the species and consequently into the construction of the 

 corresponding phyletic series. 



Some of the shades of meaning that attach to the term ' recapitulation ' 

 have already been discussed. Another must now be mentioned. For 

 Haeckel himself this term implied the idea of causation as well as of 

 repetition. I doubt whether this causal relation of ontogeny to phylogeny 

 has any place in the thinking of the average palasontologist. For him 

 recapitulation is merely a descriptive term for the observed fact that there 

 exists a striking resemblance between some stages in individual develop- 

 ment and the ancestral types. But even as a descriptive term it requires 

 the addition of certain qualifying terms such as ' general ' and ' specific,' 

 ' adult ' and 'juvenile,' which have been already suggested. 



A marked divergence occurs also between various writers in the use 

 of the terms ' ancestor ' and ' ancestral.' Thus Gars tang, referring to 

 recapitulation as enunciated by Haeckel, says (1921), ' The only way I 

 can see of establishing this theory by purely embryological methods, is 

 to show that the penultimate stage of the ontogeny of a given type of 

 adult resembles the final (adult) stages of the ontogeny of some 

 theoretically ancestral type more closely than it resembles the corre- 

 sponding penultimate stage of the same.' He then repeats the same 

 kind of requirement for the antepenultimate stages also. 



Discussing various examples brought forward by other workers, he 

 declares that they all fail to survive the imposition of this test. In 

 coming to this conclusion he apparently does not realise that they fail 

 for the simple reason that the material used in the examples quoted did 

 not conform to that stipulated in his test. His test requires that the 

 comparison should be between the ontogeny of a given type and that of 

 an ancestral type. The examples he quotes deal almost exclusively 

 with the ontogeny of forms co-existing at the present day, and therefore 

 cannot possibly have the relationship to one another of ancestor and 

 descendant. They are in fact merely collateral descendants. Had he 

 selected his material from the works of his palasontological colleagues 

 he would have found much which closely fulfilled the conditions required 

 by his test. Some of these will be considered presently. 



Another term that calls for consideration is ' race,' for by using it in 

 quite different senses writers have fallen into much unnecessary contra- 

 diction. In seeking to discover the relationship between development 

 and evolution it has always been the custom to think of development in 

 terms of a specific individual, and of evolution in terms of the ' race.' 

 Now about the development of the individual there is no ambiguity, for 

 the individual is a single being having a range in time from the cradle 

 to the grave. On the other hand the phrase ' evolution of the race ' is 

 used in almost as many different senses as there are examples quoted. 



