434 University of California Publications in Zoology [Vol. 18 



the most highly specialized of the three divisions of the genus, Peridi- 

 nium, is further support for the suggestion that the modifications of 

 the ventral plate pattern are of more fundamental significance than 

 the interpolation of the dorsal accessory plates, because the develop- 

 ment of the ventral plate patterns seem to have been able, at least 

 in certain races of Orthoperidinium, to run through its full course 

 before the addition of the third accessory plate. 



Supporting also the view that the increase in number of the pre- 

 cingular plates comes from the addition of a new plate, at first small, 

 at one end of this row near the rhomboid plate is the case of P. mi)iu- 

 tum Kofoid var, tatihouensis Faure-Fremiet (1908, p. 227, fig. 13) in 

 which there is at the right hand end of the precingular row a very 

 small plate, the eighth in this row. In this species there are still but 

 two dorsal accessory plates, and these are arranged as in fig. 12. The 

 ventral plate pattern is, moreover, that of Orthoperidinium. 



The group, Orthoperidinium, presents itself to us, perhaps, as a 

 portion of the genus which because of its high specialization is in a 

 condition of depletion in a manner "running out" and groping 

 about in a desperate fashion for some combination of its parts which 

 may again render it stable. It seems to be a portion of the genus 

 which is "fraying out." This suggestion is all the more emphasized 

 by the large number of irregular and unusually as\Tnmetrical forms 

 which it contains. In addition to those already mentioned in which 

 the accessory plates are reduced to two in number, usually symmetri 

 cally placed, other forms such as P. marsonii Lemm. may be men- 

 tioned in which the structural system in this part of the genus seems 

 to have broken down even more completely and all traces of sym- 

 metry in the arrangement of the epithecal plates have been lost. 



Summary of Changes of Plate Pattern within a Species. — Among 

 the figures in literature and among the observations of the writer 

 several cases have come to light illustrating what is apparently 

 a change in plate pattern within a given species ; that is to say, the 

 plate patterns found in two individuals are different while all the 

 other characters agree to such an extent as to compel placing these 

 forms in the same species. 



1. P. ovatum Pouchet has been reported by Faure-Fremiet (1908, 

 p. 219, fig. 5) with an arrangement of ventral plates according to a 

 pattern which Joergensen (1912) later termed that of Paraperdi- 

 nium, and with the primary symmetrical dorsal plate pattern which 

 is illustrated in fig. 5 of this paper; and again by Broch (1910, pp. 40 



