KOFOID : MUTATIONS IN CERATIUM. 227 



fainter in the more distal parts. It thus appears that the ancestral 

 skeletal moiety, i. e. of G. tripos, of this posterior member of the chain 

 is gradually disappearing. In all other structural features and in the 

 process of cell division this member of the chain appears to be a normal 

 cell. The question naturally arises, will this process continue until the 

 old skeleton is entirely lost and will a new skeleton of the C. californiense 

 type be formed in its place 1 This cell IV 3 is a sister cell of a C. cali- 

 forniense III 3 . Is not its inherited skeleton of the C. tripos type, and 

 are its nucleus and plasma now, and perhaps since the first division, of 

 the C. californiense type 1 ? Only pedigree cultures can give a decisive 

 answer to these interesting questions. 



The Mutation between Ceratium californiense and C. ostenfeldi. 



Plate 4, Fig. 4. 



Another chain was taken in the intermediate haul from 800 fathoms 

 to the surface at 8 a. m. on December 31, 1904, at station 4711, 7° 47' 

 30" S., 94° 5' 30" W., on our line between Easter Island and the 

 Galapagos Islands. The chain consists of two individuals only. The 

 anterior member of the chain is a Ceratium, ostenfeldi, the posterior is 

 C. californiense. The anterior member has the antapical horns recurved 

 anteriorly with truncated open tips. Their bases are also projected 

 posteriorly, forming a deep postindentation and a long straight post- 

 margin. These are characters of the subgenus Macroceratium. The 

 posterior member has the antapical horns projected posteriorly, with 

 slight outward curvature, a characteristic of the species (C. californiense), 

 with tapering pointed tips and deeper postindentation, characteristics 

 of the subgenus Biceratium. The mutation here involves the two species 

 C. ostenfeldi and C. californiense, belonging to two of the important 

 subgenera of the genus, Macroceratium and Biceratium. The anterior 

 cell of this chain has the characteristically open tips of the subgenus 

 Macroceratium. The extruding plasma (plate 4, fig. 5) leaves no 

 doubt on this point. The typical proportions of the three horns and the 

 general habitus of the cell suggest that it is a normal cell, not an auto- 

 tomized one. Owing to the state of development in which the relatively 

 short horns of the anterior member of this chain appear, its specific 

 identity is somewhat obscured. I have referred the cell to the species 

 C. ostenfeldi rather than to C. macroceros, because of the distance to 

 which the major flexures in the bases of the antapicals are projected 

 posteriorly. This is much greater in C. macroceros and much less in 



