C— GEOLOGY 6s 



process of elongation by which each septum in early phases both of 

 development and evolution attained the centre of the coral. Thus it 

 comes about that in the later members of the gens there is as it were a 

 conflict between these earlier and later discordant tendencies, with the 

 result that the antepenultimate stages exhibit a mixed combination of 

 features made up of the long cardinal of Z. delanouei {s. str.), the 

 elongated septa of Z. constricta and the radial arrangement of Z. disjuncta. 

 In these stages, therefore, there is merely a recapitulation of some of the 

 individual features, but not a recapitulation of the combination of features 

 of the adult of any preceding generation. It becomes advisable, therefore, 

 to distinguish between complete recapitulation of the whole or part of the 

 adult combination and the limited recapitulation of only isolated adult 

 features. 



Up to this point the earliest developmental stage which has been 

 considered has always been one in which the individual was already 

 sufficiently advanced to have attained a diameter of about half that of 

 the adult. Obviously, therefore, important earlier stages still remain to 

 be considered. Unfortunately very little information for these is forth- 

 coming, for the pointed end in the specimens of this coral is rarely 

 preserved. My re-examination of Mr. Carruthers' material, however, 

 has enabled me to see several earlier stages than those which were figured 

 by him. Of these there were three which only slightly preceded the 

 earliest figured by him for Z. delanouei {s. str.) and for typical and 

 advanced forms of Z. disjuncta. The two latter exhibit the same absence 

 of resemblance, except in isolated features, to the adult of any previous 

 species. On the other hand they did closely resemble the correspondingly 

 young stage of Z. delanouei {s. str.). They provided, therefore, as excellent 

 an example of the recapitulation of juvenile conditions described by Von 

 Baer, and emphasised by modern biologists, as the later growth stages 

 provided for the recapitulation of adult conditions reiterated above. 



In the development of the typical Z. disjuncta a much earlier stage was 

 fortunately preserved. In this there were only six septa, but these were 

 arranged in an almost perfectly radial manner. Whether they were equal 

 to one another in length was uncertain, for the section may have been 

 slightly oblique to the organic axis of the coral at this level. Though the 

 corresponding stage in the other members of this gens was not forth- 

 coming in the material discussed above, it may be noted that it had been 

 recognised by Duerden (1906), Carruthers (1906), and Butler (1935) in 

 the earliest stages of development not only of other species of Zaphrentis 

 but also in other palaeozoic genera, viz. Lophophyllum, Cyathaxonia, 

 Dibunophyllum, Cyclophyllum, Streptelasma, Syringaxon. Duerden sums 

 up his investigations by saying, ' The rugose corals and the zoanthid 

 actinians have both a primary hexamerism.' 



The embryo in this case appears therefore to retain features 

 characteristic only of the embryonic stages in the development of other 

 members of the phylum, for as yet no adult coral of earlier date is known 

 to possess them. The examination of this very young stage in the develop- 

 ment of Z. disjuncta therefore furnishes further confirmation of Von 

 Baer's principle. The careful consideration of the example before us, 



