456 Universitij of California Fuhlications in Zoology [Vol. 18 



cited — P. oceanimim, P. claudicans, P. granii, and P. ovatum — in 

 which two plate patterns occur in specimens which must still be re- 

 garded as belonging to the same species. These rare changes of pat- 

 tern are always well isolated from the normal pattern characteristic 

 of the given species. No series of gradations has been observed be- 

 tween such an unusual pattern and the normal pattern; on the con- 

 trary, the length of the critical suture for the normal pattern seems to 

 vary between definitely recognizable extremes, the minimum limit of 

 which is still well above the vanishing point for the suture. The gen- 

 ral conclusion from all of these observations is, therefore, that a cer- 

 tain pattern is characteristic for a given species ; that this pattern may 

 occasionally be changed; but that such changes occur abruptly as 

 mutations and are without a connecting series of intergradations. 



Mutations 



We have pointed out here several definite and different arrange- 

 ments of the plates of the shell of species of Peridinium. These 

 arrangements presumably represent states of equilibrium between 

 potentially variable elements. A change in pattern involves the short- 

 ening of one suture and the introduction of another suture at right 

 angles to it and between two different plates. However, each suture 

 maintains its own norm for the fluctuating variations which it may 

 be expected to undergo, and transitional stages between the two 

 possible patterns which a given group of plates may assume are 

 unknown. 



Hence, in at least this superficial character of plate pattern, these 

 organisms undergo nnitations which must be recognized as such if we 

 are to admit any phylogentic sequence at all among these species ; that 

 is, the change from one plate pattern to another requires a certain 

 saltation to bridge the gap. That such saltations when they have 

 occurred have remained permanent at least for a time is shown by 

 the presence of species exhibiting nearly all of the arrangements 

 of a given group of plates which are geometrically possible. This 

 character of the plate pattern seems, therefore, to undergo mutations 

 in the generally accepted meaning of abrupt, discontinuous, morph- 

 ological changes which reappear in successive generations after their 

 first occurrence. 



The pattern of the sutures is, however, not the real character, but 

 is dependent upon the size of the plates which the sutures bound. The 



