1918] Barrows: Skeletal Variations in tlie Genus Peridinium 449 



On the opposite side of the rhomboid plate the suture between 

 plates 1' and 2" was found to range from .0620 to .2389 and in one 

 instance to .3221 of the transdiameter, the summation of the measure- 

 ments of this suture giving- a triple crested curve with maxima at .13, 

 .15, and .17. The lowest maximum is so slightly separated from the 

 rest of the curve, however, as to be probably of no significance. The 

 occurrence of two maxima consistently in this analysis seems to be 

 accidental and is doubtless correlated with the derivation of these 

 specimens as geographic races from several widely separated sources, 

 tliough the cause for the double crested curve is not directly apparent 

 even upon examining suture measurements for the specimens from 

 each region. The crests are separated in a given curve by only two 

 units of measurement on the ordinate axis, and such a segregation 

 may be due to the reduction of the measurements for plotting to the 

 nearest integral number of hundredths of the length of the transdiam- 

 eter of the organism. 



It is to be noted that the length of the suture between plates 4' 

 and 7" and plates 1' and 2" in no case fell below .027 of the trans- 

 diameter and that the length of the suture between plates 1' and 2" 

 did not fall below .0617. In this connection a further observation is 

 of peculiar significance. In a longer series of specimens of P. diver- 

 gens, sixty-three in number, observed by the writer, two specimens 

 were found which presented not the usual ventral patterns for 

 P. divergens but the pattern for the subgenus Paraperidinium. These 

 specimens were, however, characteristic of P. divergetis in all other 

 respects. Mangin (1911, pi. 7, figs. 10 and 13) also figures two speci- 

 mens in which the same modification of plate pattern is found in 

 specimens which must still be assigned to the species P. divergens. 



We cannot but infer from these records, therefore, that occasion- 

 ally the plates of the ventral side of the epitheca of P. divergens 

 present a plate pattern which, upon the basis proposed for the deriva- 

 tion of these plates, must be regarded as more primitive than the pat- 

 tern usually presented by this species in this region. It seems, there- 

 fore, that occasionally, as if inhibited by some factor, or perhaps if 

 not stimulated as much as usual, plate 1" in some specimens does not 

 grow to its usual size, and does not quite reach plate 4', nor fully 

 separate plates 1' and 6". Here, then, appears to be a case of certain 

 instability in respect to this one potentially variable region, involving 

 not only the minor variations in lengths of sutures, which is to be 

 expected, but involving also variations in the size of certain plates 



