8 VARIATION AND DIFFERENTIATION IN CERATOPHYLLUM. 



At the time when the work was begun the writer had not yet seen 

 the brilHant memoir by Professor Pearson on ' ' Homotyposis in the veg- 

 etable kingdom" (Pearson, :01). When later this memoir had been 

 read it was apparent that the problems which had been set for investi- 

 gation in Ceratophyllum were fundamentally similar to those which 

 Pearson had before him. There were, however, certain differences in 

 standpoint which seemed to make it desirable to go on with this work. 

 Pearson was concerned mainly with the determination in a wide series 

 of forms of the amount of the homotypic correlation. To quote his own 

 words (loc. cit., p. 294), the principle he was investigating was "the 

 principle that homotypes are correlated, i. e. , that the variation within 

 the individual is less than that of the race, or that undifferentiated like 

 organs have a certain degree of resemblance." From this standpoint 

 he very naturally dealt, so far as possible, with undifferentiated or 

 slightly differentiated like parts, not concerning himself at that time 

 with any special investigation of the factors which produce differentia- 

 tion in the repeated parts of plants. As will have been apparent 

 from what has gone before, it is with this latter problem that the present 

 work has to do. Our present aim is to examine as many as possible of 

 the factors concerned in producing differentiation of homologous parts, 

 and determine their effect on the intra-individual variability of the 

 differentiated parts. For such a study it seems best for practical 

 reasons to use at first only one organism, and make the investigation of 

 that as thorough and as detailed as possible. It will thus be seen that 

 this work, while not concerned with the determination of the degree of 

 homotyposis in a particular plant, yet deals with one of the fundamental 

 problems of homotyposis, and in so far may be considered supplementary 

 to Pearson's memoir on that subject. 



This work was begun early in 1902 by Miss Olive M. Pepper, at that 

 time a student in the Zoological Laboratory of the University of Michigan. 

 During the summer of 1902 Miss Pepper collected and nearly finished 

 the counting of the plants in Series I, II, and III (cf. p. 13, infra). 

 During the first half of the academic year 1902-03 she continued the 

 work, sorting the data into frequency distributions and making some 

 start on the computing. During this year she determined and recorded 

 the data for Series IV. As it was impossible for Miss Pepper to go on 

 further with the work, it was then carried forward as opportunity 

 offered by the writer, by whom practically the whole of the raw material 

 was reduced. When in the spring of 1905 it became possible, through a 

 grant from the Carnegie Institution of Washington, to get some aid for 

 the biometric work in hand, an assistant, Miss Florence J. Hagle, was 



