1918] Barrows: Skeletal Variations in the Genus Peridinium 441 



suggests that these two accessory plates may have been budded olf 

 from the anterior corners of that plate (4"). The subsequent bud- 

 ding off of a third and median plate from the now single and median 

 corner of the mid-dorsal accessory plate, 4", or from the posterior 

 corner of the mid-dorsal apical plate, 3', would provide the third 

 accessory plate, found characteristically in this genus. 



On the other hand, the occurrence of such a species as P. qiiadri- 

 dens, in which there are apparently six apical plates, two of which may, 

 however, be interpreted as accessories, suggests that the two lateral 

 accessories may have arisen from buds from the two lateral apical 

 plates, 2' and 4'. These plates would have forced what was originally 

 the mid-dorsal apical plate, 3', into the future position of the middle 

 accessory plate, and a bud from the anterior corner of this plate may 

 have formed another plate to take the place of the mid-dorsal apical 

 plate just removed. The varied articulations of the mid-dorsal acces- 

 sory plate, 2a, do not make clear, however, the manner by which these 

 variations can be connected with this method of origin, and we are 

 inclined, on the whole, to favor the hypothesis that the two lateral 

 accessory plates arose in the region of the anterior lateral corners of 

 the mid-dorsal precingular plate, 4'^ 



The middle accessory plate may have originated in one of at least 

 two ways; either as a bud from the anterior corner of the mid-dorsal 

 precingular plate, now projecting more or less between the two lateral 

 accessory plates already formed, or the middle accessory plate may 

 have been formed from a bud from the posterior corner of the middle 

 apical plate, 3'. This latter suggestion seems most readily to fit in 

 with the subsequent development of the middle accessory plate and 

 the readjustment of the lateral accessory and other plates due to its 

 enlargement. In the absence of definite evidence to suggest a contrary 

 view, we are inclined to favor the origin of the middle accessory plate 

 from a different source than that from which the lateral accessory 

 plates may have come, and probably from the region of the anterior 

 median corner of the mid-dorsal precingular plate secondarily after 

 the formation of the lateral accessory plates (la and 3a). 



Evidently the forms produced in these early stages, if this hypothe- 

 sis be correct, were not stable, and evolution seems to have proceeded 

 at once to the formation of more fully specialized forms having three 

 well-developed accessory plates. In this condition the genus seems to 

 have settled down into approximate equilibrium and durability. Spe- 

 cies illustrating fully all of the early stages through which this pro- 



