GENERAL DISCUSSION OF RESULTS. 129 



a stereotyped result with ever-increasing precision and constancy. 

 Again, to take another example, it is obvious, as Jennings has pointed 

 out, that ' 'the operations of this law are seen on a vast scale in higher 

 organisms, where they constitute what we commonly call memory, 

 association, habit, and the basis of intelligence," If we consider for a 

 moment the case of memory in man, it will be still further clear that 

 there is objectively a fundamental similarity between one characteristic 

 of this psychological phenomenon and such facts of morphogenesis as 

 we have detailed for Ceratophyllum. Suppose a man sets to work to 

 memorize a number of lines of poetry, and tests his acquirements by 

 attempting to repeat the lines after each successive reading. The 

 result will be something like this: When he attempts to repeat the lines 

 after the first reading he will make a number of mistakes, or ' 'devia- 

 tions from the type" which is given by the exact cext. On repeating 

 the extract after the second reading the number of "errors" or "devia- 

 tions" will tend to be fewer; after a third reading still fewer, and 

 so on until finally there are no "deviations, " or, in other words, the 

 "type" is reproduced exactly at each successive repetition. Now, what 

 do we find in Ceratophyllum? When the first whorl on an axis is pro- 

 duced we see, if we examine a large number of such whorls, that devi- 

 ations from the type are produced relatively very frequently; second 

 whorls exhibit a smaller number of such deviations; third whorls a still 

 smaller number, and so on until we reach a condition of minimum 

 variability, or in other words, a condition in which the type is produced 

 each time with great precision and constancy. What goes on in the 

 case of the memorizing and in the case of the growing plant may be 

 objectively described in the same terms, the principal difference being 

 that in the former example the type is absolutely fixed and constant, 

 while in the latter it changes slowly. These illustrations will suffice to 

 show that v/hat we have called the law of diminishing variability 

 operates in psychological as well as morphological phenomena. 



In attempting to determine how generally this law of diminishing 

 variability holds in respect to processes of growth, one is met as before 

 with the difficulty that there have been but few investigations which 

 have brought to light direct evidence on this point. It should be kept 

 in mind that the conditions under which we should expect this law to 

 show its operation in the clearest and most unequivocal form are those 

 in which we have the production of a series of not greatly differentiated 

 parts or characters. Obviously these conditions are best realized on the 

 botanical side in plants having a type of structure similar to that of 

 Ceratophyllum, and on the zoological side in animals built up on a simple 

 metameric plan. So far as the writer knows, no systematic investiga- 



