W. D. Lang — Bernard's Biological Theories of Fossils. 555 



universal presence and importance of radiating linin threads that 

 connect unit with unit and the periodicity of evolutionary processes. 

 In consideration of the first point, Bernard claims as one of the most 

 primitive and fundamental functions of the linin filaments the 

 conveyance of waste matter towards the periphery of the organism. 

 In some cases waste matter is deposited around the filaments as 

 a sheath, and so spicular skeletons are formed. In other cases the 

 waste matter is laid down at the ends of the filaments forming 

 cuticular and other exoskeletons. Under the former category fall 

 the Alcyonarian and under the latter the Madreporarian coral 

 skeletons. Now two diiSering theories have heen held regarding 

 the formation of the latter skeletons, namely (1) that ectoderm cells 

 secrete calcareous matter on their outer surface,' and (2) that ectoderm 

 cells secrete such matter internally and gradually convert themselves 

 into the granular elements of the skeleton.* The last great work on 

 the microscopic structure of the Madreporarian skeleton was that 

 of Ogilvie,^ who described the granular unit that composed the coral 

 skeleton as a cake-shaped mass built up of calcium-carbonate fibres 

 and produced by a calcified ectoderm cell (' calicoblast').^ Ogilvie 

 claims to have found organic matter (the remnant of the chalicoblast) 

 mixed up with the granule, so that the granule is not a mere 

 secretion on the outside of the ectoderm cell (as liourne has since 

 reaffirmed).* Whichever view is right, neither explains the minutely 

 fibrous structure of the granule. The simultaneous secretion of all 

 the chalicoblasts forms a pavement of granules, and beneath this 

 another laj^er, and so on, until the skeleton consists of many growth- 

 lamellae. Regarded in a direction at right angles to the plane of the 

 secreting surface, the skeleton consists of piles of granules heaped up 

 over the chalicoblasts, though, naturally, not so regularly as can be 

 imagined. 



Now, instead of a row of chalicoblasts laying down calcium carbonate 

 within or without themselves, let us, with Bernard, regard the 

 secreting surface as a cytoplasmic plane studded with the ends of 

 the bundles of linin filaments, each conveying calcium carbonate from 

 the interior and laying it down at their ends. The resulting skeleton 

 would be a succession of sheets, each a superficial layer of minute 

 specks of calcium carbonate far smaller than the granule secreted by 

 a chalicoblast. Regarded in a direction at right angles to the secreting 

 surface, the skeleton would be seen to consist of fibres, each composed 



' Von Koch, " Ueber die Entwicklung des Kalkskeletes von Asteroides 

 caZi/cztZaris, und dessen morphologischer Bedeutung " : Mittheilungen aus der 

 zoologischen Station zu Neapel, vol. iii, pp. 284-92, 1882 ; and Bourne, 

 ' ' Studies on the Structure and Formation of the Calcareous Skeleton of the 

 Anthozoa " : Quart. Journ. Micro. Sci., N.S., vol. xh, pp. 499-547, 1899. 



'•^ Von Heider, " Die Gattung Cladocera, Ehrenberg ": Sitzungsberichter der 

 Kaiserlichen Akademie der Wissenschaften, Wien, vol. Ixxxiv, pt. i, pp. 634-68, 

 1882; and Ogilvie, " Microscopic and Systematic Study of Madreporarian Types 

 of Corals" : Phil. Trans. Eoy. Soc.Lond., ser. B, vol. clxxxvii, pp. 83-345,1897. 



^ Op. cit., under (2). 



■* ' Chalicoblast,' von Heider, op. cit., under (2), p. 651, not ' calicoblast', 

 after Bourne and Ogilvie (x«^'l, ' a stone '). 



* Bourne, "Anthozoa," p. 80, in Lankester,^ Treatise on Zoology, 1900, -ptAi. 



