30 Coelenterata. 



12 stripes are attained or (less usually) owing to delay in division more than 

 12 stripes may be present. Triglyphic forms occur and tend to possess more 

 than 12 stripes. 



Roille( 4 ) finds the column wall of Stichopathes richardi n. remarkable for the 

 great thickness of the mesogloea, comparable to that of a young larva of Ceri- 

 anthus before the ectodermic musculature appears. The Antipatharia, which 

 are the most simply organised Anthozoa, represent the ancestral forms of the 

 Ceriantheae and also of all the Anthozoa. They have persisted owing to their 

 faculty of budding and producing fixed colonies. From such ancestral forms 

 one line of descent leads to the Cerianthids and the fossil Rugosa (grouped 

 together as Protanthozoa). The other Anthozoa (or Metanthozoa) have reached 

 a more complex histological structure especially in the musculature of the me- 

 senteries and column. 



Duerden( 5 ) states that Zoanthids are the nearest skeleton-less represen- 

 tatives of the stock from which sprung the skeleton-forming Rugosa, as is 

 shown by the method of mesenterial and septal formation [see Bericht f. 1902 

 Coel. p 20]. The Zoanthid type of coral flourished in Palaeozoic times, the 

 Hexactinian type of coral assumed predominance in Mesozoic times. In all pro- 

 bability the Ceriantheae date back further than these, but no skeleton-producing 

 allies, either living or fossil, have been met with. 



Bernard (^ holds that the parent calicle of a colony rises out of a basal 

 cup - the prototheca the rim of which can be bent up, flattened com- 

 pletely down and indefinitely expanded in any direction as a film the 

 epitheca from the upper surface of which, as originally from within the 

 cup, the coral skeleton arises. Those parts of the coral skeleton called epitheca 

 must, for the future, be referred to the rim of the prototheca. A succession 

 of epithecal cups or saucers occurs throughout the whole of the Madreporaria. 

 Wherever the epitheca occurs it represents the rim, or the coalesced rims, of 

 one or more epithecal cups the floors of which are represented by the tabulae. 

 The increasing instability of the polyps inhabiting a cup-shaped skeleton has 

 probably been an important factor in determining modification of the skeleton. 

 The falling over and recovery of the upright position explain some of the 

 protothecal and other structures found in Zaphrentis, Menophyllum, Aulopora, 

 Syringopora, Alveopora, Favosites and Halysites. The morphology of many of 

 the palaeozoic corals (e. g. Z. gigantea, Pal<zocyclus, Cyathophyllwn and Calo- 

 stylis) is explained on the hypothesis of an early flattening out of the proto- 

 theca. The chief difference between the palaeozoic and recent astrseiform corals 

 is due entirely to the more recent development of the radial or septal, as 

 compared with the concentric, protothecal foldings. In palaeozoic times the 

 former were not very pronounced so that the flattened or curved sides of the 

 protothecal cups with their tabulate floors formed the most characteristic portion 

 of the 'skeleton. In modern Astra?ids the tabulate character has become ob- 

 scured, but the septa have become prominent and the radial folds of the proto- 

 theca conspicuous. The author concludes that there is no generic difference 

 between Moseleya and Lithophyllia. 



Bernard ( 2 ) gives an historical account of the Poritidse and of Goniopora. 

 Litharcea, Rhodarcea and Tichopora belong to and are merged with G. He dis- 

 cusses the morphology of the skeleton. The primitive epithecal cup or proto- 

 theca of the Madreporaria has been flattened out and become vestigial; the 

 present thecae are entirely septal and internal. The epitheca takes the lead in 

 the formation of the free edge and is frequently seen projecting. The growth 

 forms of G. are very rich in free edges and consequently bands of wrinkled 



