342 bulletin: museum of comparative zoology. 



The fact that there is no "longitudinal furrow" extending upon 

 the epitheca to the apex as figured in nearly all of the described spe- 

 cies of Gonyaulax and presented as a generic character by Butschli 

 (1885) and Schiitt (1896), is in my opinion without significance. As 

 we have shown in a recent paper this reputed furrow upon the epi- 

 theca is merely the narrow mid ventral apical plate 1', often guarded 

 by lateral ridges so that it appears to form a trough. It is in no sense 

 a part of the furrow occupied by the longitudinal flagellum and is 

 not structurally differentiated from the rest of the thecal wall. It is, 

 moreover, not present as a trough or furrow in some species of Gonyau- 

 lax in which apical plate 1' has a greater width than usual and has been 

 unduly emphasized in the figures of species in which it is narrow. To 

 treat this region as a longitudinal "furrow" is to obscure its function 

 and make still more difficult a correct diagnosis of this exceedingly per- 

 plexing genus. 



It is at once evident upon inspection of Levander's (1894) figures 

 and description of Peridinium catenation that it and the form here 

 described are closely related. In Levander's material the apical and 

 antapical regions are revealed on isolated individuals and also in 

 specimens in chain. He gives an analysis of these regions but assigns 

 the species to the genus Peridinium, possibly because it shows a slightly 

 expanded ventral apical (Rautenplatte) and has plates which might 

 be interpreted as dorsal intercalaries in the terminology which I have 

 elsewhere applied (Kofoid, 1909) to Peridinium. The hypotheca 

 has, however, a wholly different composition as to plates from that 

 in Peridinium and one perfectly typical of Gonyaulax. It has but a 

 single antapical while Peridinium has two. Levander specifically 

 mentions the single antapical but fails to differentiate the posterior 

 intercalary the existence of which, however, is plainly suggested in 

 his text figure 4 and figures 1 and 3 of his plate. The distribution of 

 the four or more spines along the angulate margin of the antapical 

 plate is similar to that in G. triacantha and G. ceratocoroides. 



This type of plate structure in the hypotheca is so characteristic 

 of Gonyaulax that it seems best to transfer Levander's species from 

 Peridinium and to refer to it as Gonyaulax catentata (Levander). 

 See Kofoid "On the skeletal morphology of Gonyaulax catenata 

 (Levander)" in Univ. Calif. Publ. Zool. v. 8, 1911. Should it be- 

 come necessary in the future to break up the genus Gonyaulax into 

 smaller genera Levander's species and the one here described might be 

 kept together. The peculiar arrangement and number of apical 

 plates in Levander's species is without precise parallel among the 



