D.— ZOOLOGY. 69 



are of the cubic opal. 8 ' 2 I propose to call these Orthogonida, with the four 

 orders Clavulida, Axinellida (?), Desmacidonida and Renierida. 



In biology a new secretion seems one of the most difficult things to be 

 produced in evolution, so that biochemists tabulate long series of allied 

 respiratory pigments and chlorophyll derivatives : we inherit our visual 

 purple from so far back that we still denominate as ' light ' the limited 

 range of vibrations which the sea transmitted to our flagellate ancestors, 33 

 and prove our descent from them by considering water perfectly trans- 

 parent. Therefore it seems most probable that opal crystallising on the 

 cubic system was only evolved once. 



This principle indicates that the Orthogonida are probably descendants 

 of Porifera nuda, and there is some evidence that in histology and re- 

 production they retain some resemblances to the Hexactinellida. 34 I 

 suggest that they are descended (probably the four orders independently) 

 from hexactinellids which found themselves in waters where the permanent 

 current was inadequate. Thus they found it necessary to develop a 

 hydraulic engine generally similar to that developed, probably inde- 

 pendently of each other, by each of the three orders of the Porifera 



vera. 35 



It is agreed by all that the main groups of sponges were developed 

 before the Cambrian. And since Porifera nuda and Porifera vera arose 

 from differing forms of Choanoflagellate colony, we may suppose that in 

 the Flagellate Sea we had. at least these two strains of sponges, and possibly 

 the four main forms of skeleton ; though possibly the slime of Halisarca 

 and the horny sponges, and the three forms of crystallising secretion, may 

 have only been developed for defence, when Metazoa arose and began to 

 feed on sponges. 



For of metazoan animals in the Flagellate Sea there were none. All 

 the choanoflagellates, all the sponges, and all the intermediate ancestors 

 were microphagous, that is, evolved to supply their component monads 

 with food in the form of minute particles, such as spores, very small pro- 

 tista, and minute fragments of decayed protista. So they fed in the 

 Flagellate Ocean of the early Pre-Cambrian age, and so they feed to-day. 



32 Cf. Nature, 1925, vol. cxv, p. 299; and G. C. J. Vosrnaer, 1886: Bronn's' Porifera,' 

 p. 473. The chelae of Media and Melonanchora have each three planes of symmetry 

 at right angles to each other. The spicules of Reniera and Chalina are absolutely 

 symmetrical end for end, as are the birotulae of Ephydatia. This is extremely unlikely 

 in spicules formed of tetraxon opal, where one end corresponds to bases of tetrahedra, 

 and the other to their apices. 



33 Church, I.e. p. 6. 



34 H. V. Wilson, 1891 : Journ. Morph. Boston, v. p. 511 (Esperella) ; and 1894, ix. 

 p. 277. Kirkpatrick, 1911 : Q.J. M.S. lvi, pi. xxxii. (Media normmi— note collar- 

 cells) ; J. Cotte, 1903 : Theses presentees a la Faculte des Sciences de Paris. Lille ; 

 p. 428, ' chez les Clionides une disposition rappelant la structure trabeculaire des 

 Hexactinellides ' ; Topsent (on the flagellate chambers of Cliona), Arch. Zool. Exp. 

 (2) V bis. 



:,s As to the relationship to each other of the Porifera vera, I have published remarks 

 on the similarity of the outer cells of horny sponges with those of Calcarea (P.R.S. vol. 

 52, p. 134), Dendy and Row have noted the resemblance between the canal system 

 and collar-cells of Oscarella and those of a calcareous sponge (Lcucettusa, P.Z.S. 1913, 

 p. 738), and I may add that the collar-cells of Oscarella and Clathrina are strangely 

 alike. 



