kofoid: mutations in ceratium. 217 



upon the collections of the " Albatross " in the tropical Pacific and else- 

 where, many instances of chain formation have come under my obser- 

 vation. They have always been of interest and have received close 

 inspection, and many are recorded in my unpublished sketches, especially 

 those having any bearing on the question of variation in Ceratium. 

 Hundreds of such chains have been inspected by myself or my assistant, 

 Miss Rigden. With the exception of three instances noted later these 

 chains have been of the normal type described above. The material 

 examined from San Diego covers all seasons of the year and depths from 

 0-500 fathoms, and that from the "Albatross" includes collections from 

 Alaska waters, the coast of California, and from the Expedition to the 

 Eastern Tropical Pacific extending southward to Easter Island in the 

 South Pacific, and from depths of 0-800 fathoms. My material is there- 

 fore fairly representative of Ceratium in its oceanic and neritic environ- 

 ments within the range of conditions of the normal distribution of the 

 genus. 



The literature of the subject prior to 1908 contains but few references 

 to chain formation beyond those Qf Michaelis (1830), Allman (1855), 

 Murray (1882), Pouchet (1883, 1885, and 1894), Btitschli (1885), Bergh 

 (1886), Schtitt (1895, 1896), Karsten (1905, 1906, 1907), and Entz 

 (1907). In all these cases the type of chain formation is the normal 

 one above described. 



The Genus Ceratium. 



The genus Ceratium is a large one of wide distribution in the plankton 

 of fresh water and the sea. Most of the species are also of wide distri- 

 bution and some are cosmopolitan. They exhibit a wide range of 

 variation in many cases, especially in length and differentiation of the 

 horns, their principal organs of flotation, in adaptation to varying 

 conditions of life. No less than 290 different names have been given in 

 this genus in recognition of the species, subspecies, and other subordinate 

 categories in which the forms have been classified. In the opinion of 

 the writer probably not less than two thirds of these are based upon 

 growth, age, or temperature characters, while the remaining third are well 

 founded ; but even so, the degree to which the biological process of 

 speciation has progressed in this genus is, in comparison with other 

 dinoflagellates, except Peridinium, relatively great, a fact of possible 

 significance in connection with the phenomenon of mutation herein 

 described. All of the species of the genus, with the possible exception 



