1918] Barrows: Skeletal Variations in the Genus Peridinimn 453 



of the fourth variable region, the right hand side of the ventral sur- 

 face of the epitheca, seems, however, to be unstable and to present occa- 

 sionally the alternate pattern for this region, unusual in this species. 

 This particular region seems, therefore, to be in a state of greater 

 plasticity than any one of the other potentially variable regions. 



The essential point in this portion of the discussion of plate varia- 

 tion in Peridinium is that there is no complete series of variations 

 between the usual pattern and the exceptional pattern found in this 

 part of the shell of this species. In such a complete series of imper 

 ceptible variations the suture between plates 4' and 7" would become 

 shorter and shorter and finally disappear, after which disappearance 

 a new suture between plates 1' and 6" would gradually take its place. 

 In the series there would be a stage in which four plates and four 

 sutures would come together. 



Extremes of such a possible series are known, presenting both 

 alternate plate patterns. Within this species this region is, there- 

 fore, known to be unusually variable and to present the full gamut 

 of possibilities of plate arrangement possible, though the plate pat- 

 tern is usually confined to one type. However, the variation of the 

 critical suture in this usual plate pattern never approaches beyond a 

 certain amount the point of obliteration, which must be passed before 

 the alternate plate pattern can be produced. In a few instances 

 found of the presence of the alternate plate pattern, the new critical 

 suture is shown to be of definite and considerable value. No authentic 

 instance is known of the merging of the two patterns by the meeting 

 of four plates at one point with the obliteration of the critical sutures 

 of both patterns. Drawings which seem to show such a condition 

 ma}^ be rejected as not sufficiently accurate to be trusted in a matter 

 where unprecise observation may make so great a difference as here, 

 in view of the large number of first hand cases which have been very 

 critically examined in this regard. 



It seems probable, therefore, that when the plate pattern at a 

 variable region changes within a species or between two different 

 species, it does so with a jump ; that each pattern is characterized 

 by a certain suture, which varies about its own norm; that these 

 norms do not merge into one another in each pair of alternate pat- 

 terns possible at a given point, but that the two patterns are separate ; 

 and that when the plates do readjust themselves in response to internal 

 stresses they do so with a certain amount of abruptness without pass- 

 ing in the process through all of the intervening stages between the 

 normal proportions of the two patterns. 



