Hyatt.] 124 [March 5, 



two distinct ancestral types for Hydrozoa, and that either the 

 planula or gastrula, therefore, must have been a secondary or 

 derivative form. By far the best and most complete analysis of, 

 this difficult problem is given by Dr. Whitman (Embryol. Clepsine 

 p. 300), where the author concludes that delamination is the second- 

 ary, and gastrulation the primitive mode of developing the endo- 

 derm, and also in large part anticipates the views advocated in 

 this paper with regard to the origin of gastrular invaginations. 

 When the gastrula and planula are found together, as in the 

 Acraspeda, the gastrula procedes in development the planula 

 (Claus, Polyp. Quail. Denk. Akad. Wien., vol. xxxvin, 1877 

 Balfour, Embryol., vol. i, ~p. 137 ; and Alex. Agassiz, Embryol. 

 Monog. Mus. Comp. Zool., vol. ix, no. 3, pi. 7) ; and this j^lanula is 

 really a three-layered mouthless sac in Chrysaora, but, through 

 the suppression of the primitive mesenchyme of the paren- 

 chymula, it subsequently has the aspect of being only a two- 

 layered hydroplanula. Among the Actinozoa it seems evident that 

 the gastrula, often composed of primitive and amoeboidal cells, is 

 an earlier condition of the ovum and leads to the formation of the 

 hydroplanula which retains the blastopore as its mouth, as shown 

 by Kowalevsky in Actinia? (Embry. Monog. Mus. Comp. Zool., 

 vol. ix, no. 3, pi. 11, f. 27-29). * The tacit assumption that the 

 planula was necessarily a mouthless form and preceded the gas- 

 trula, was due to the resemblance between the hydroplanula and 

 the parenchynula which are often considered identical and con- 

 fused with the true planula. The figures of Cerianthus mem- 

 branaceus, also from Kowalevsky (ibid., pi. xn, f. 1-6), is pre- 

 sented as the next stage after the blastula, and compare closely 

 with the similar period of the parenchymula in the sponges, 

 which also in Halisarca precedes in development the permanent 

 gastrula. The gastrula in Cerianthus, also, immediately follows 

 this stage and becomes an elongated hydroplanula with a persis- 

 tent blastopore. From our point of view the single layered 

 embryo (L 1) is a parenchymula in which the mesenchyme is 

 absent or has not been obsep^ed (a frequent occurrence among 

 sponges on account of the transparency of the layer or some 



1 The originals of these and other figures, quoted from this author are to be found in 

 Memoirs, vol. x, Imperial Soc. of Friends of Nat. Sci., Anthrop., and Ethn. Moscow, 

 n ). 2, 1874; and the text has been translated for the author by the kindness of Miss B. 

 F. Hapgood. 



