RELATIONSHIPS AMONG THE GENERA 



The Girdle 



The number of girdle plates has not been generally 

 determined by previous workers, and has frequently 

 been omitted from generic plate formulas. Therefore, 

 this feature is known for only a limited number of ge- 

 nera. This is the more deplorable since there are in- 

 dications from the present investigations that the num- 

 ber of girdle plates may be important from a taxonomic 

 standpoint. 



All the genera treated in this report have six gir- 

 dle plates except Peridinium and Ceratium, which have 

 four. Six is also recorded for certain other genera not 

 treated in this report. Other numbers have been ques- 

 tionably reported (see p. 3), but these aberrant num- 

 bers must be checked by careful dissection before they 

 can be included in evolutionary discussions. The num- 

 ber is thus apparently constant and it is probable that 

 six is the primitive number. Any deviation from this 

 number should, at least for the present, be interpreted 

 as a divergent feature. 



In regard to the relationships among the genera 

 treated here, the number of girdle plates fits into the 

 scheme devised by a comparison of the ventral areas 

 and, at the same time, indicates a greater unity in the 

 order as a whole. The six girdle plates in Goniodoma , 

 Ceratocorys . Gonyaulax , Acanthogonyaulax , and Spirau - 

 lax bring these genera more closely together. The di- 

 vergent number, four, in Peridinium and Ceratium cor- 

 roborates the previous conclusion that these two genera 

 have been separated from the ancestral form for a long 

 period of time. 



The Hypotheca 



When considering the hypotheca, we have to deal not 

 only with the number of plates, but also with the pattern 

 which these plates assume. For intergeneric studies it 

 is sufficient to limit the discussion of patterns to a con- 

 sideration of the number of plates in the individual ser- 

 ies, inasmuch as the relative positions of the plates and 

 the course of the sutures are usually of specific value 

 only. Thus, the expression of plate pattern is itself re- 

 duced to numbers. 



Table 1. Comparison of number of plates in the four 

 parts of the theca of the genera treated in 

 this report 



Genera 



Epi- 

 theca 



Hypo- 

 theca 



Girdle 



Vent, 

 area 



Total 



Goniodoma 11 



Ceratocorys 10 



Gonyaulax 12 



Acanthogonyaulax . . 13 



Spiraulax 12 



Peridinium 16 



Ceratium 10 



In regard to the total number of plates in the hypo- 

 theca, the genera treated here are divided into two 

 groups (see table 1). The first group has eight plates 

 and includes all the genera except Peridinium and Cera - 

 tium . The second group has seven plates and Includes 

 these two genera. There is a striking parallelism here 

 with the number of girdle plates, which substantiates the 

 assumptions based on the comparison of ventral areas 

 alone that Peridinium and Ceratium are divergent from 

 the other genera. 



In regard to me tabulation of the hypotheca a simi- 

 lar grouping is indicated, except that Goniodoma is sep- 

 arated from related genera (see table 2). There are, 

 thus, three groups: Goniodoma with hypothecal formu- 

 la 5po, Op, 3ant; Ceratocorys. Gonyaulax . Acanthogon - 

 yaulax . and Spiraulax with formula 6po, Ip, lant; and 

 Ceratium and Peridinium with formula 5po, Op, 2ant. 

 This corroborates the conclusions derived from a 

 study of the ventral areas, but emphasizes the fact 

 brought forth by the position of the posterior sulcal 

 plate that Goniodoma represents a short, independent 

 side branch, even though close to Ceratocorys . The 

 approximate similarity of the pattern of Goniodoma 

 with that of Peridinium and Ceratium must be consid- 

 ered a result of convergence, since the three equally 

 sized antapicals of Goniodoma present a fundamentally 

 different plan from that of Peridinium or Ceratium. 



Table 2. Comparison of plate formulas of genera 

 treated in this report 



Genera 



Pl 



ap 



pr g 



po 



ant 



Goniodoma 1 



Ceratocorys 1 



Gonyaulax 1 



Acanthogonyaulax . 1 



Spiraulax 1? 



Peridinium 2 



Ceratium 1 



3 7 6 5 5 



6 5 



6 6-7 

 6 7 

 6 6 



4 5-6 

 4 ? 



The Epitheca 



In considering the epitheca we have to deal with 

 both the number and the arrangement of the plates, as 

 in the case of the hypotheca. The total number of plates 

 in the genera investigated ranges from ten to sixteen, 

 and it shows no correlation with the grouping of the ge- 

 nera indicated by the other parts of the theca (see ta- 

 bles 1 and 2). In the other genera of the Peridiniales, 

 not included in this report, the number of epithecal 

 plates is also extremely variable, as is shown by the 

 above list of generic plate formulas (p. 3). When we 

 turn to the arrangement of these plates we find that 

 here, too, no generic grouping can be made which will 

 parallel that made on the basis of the hypotheca, girdle, 

 and ventral area. In fact, of the genera reported here, 

 no two have the same epithecal formula (table 2). It is 

 thus evident that the evolution of the epitheca has been 

 in many directions and that it has been decidedly inde- 

 pendent of the evolution of the other parts of the theca. 



Summary 



On the basis of the above comparisons of the vari- 

 ous parts of the peridinian skeleton, it is thus possible 

 to draw some conclusions regarding the relationships 

 of the genera studied and to postulate certain aspects of 

 the course of evolution in the group. When the term 

 "relationships" is used here, it is clearly understood 

 that the true relationship is not always indicated by 

 morphology. Results obtained in the field of genetics 

 have clearly demonstrated that the closest morphologi- 

 cal similarities do not always represent the closest re- 

 lationships. For the time being, however, we must 

 build our genetic discussion of the peridinians on the 

 assumption that morphological similarities express de- 

 grees of relationship, unless evidence clearly demon- 

 strates that the similarities are due to convereence. 



