| 478 Charles Zeleny 
In a determination of the rates of successive regenerations it is 
necessary to bear in mind the essential difference between two 
points of view. On the one hand the observer may be concerned 
merely with the gross result of the second and later removals 
as compared with the first without regard to the essential factor 
involved. He will then merely determine whether, for instance, 
a salamander’s tail is replaced more slowly or more quickly the 
second time than the first. Even with this limited object in view, 
he is likely to get contradictory results in different sets of experi- 
ments, the results varying with the conditions. 
On the other hand, he may seek to determine the effect of the 
successive removal itself without special regard to the gross charac- 
ter and amount of the successive replacements. ‘The difficulty 
of such a determination is apparent and only an approximate 
value can be obtained. Most of the important sources of error 
may however be reached. Former results along this line are for 
the most part unreliable, because the co-ordinate accessory factors 
are neglected. For instance successive removals of an organ in an 
individual may show a change 1 in rate of replacement when as a 
matter of fact the change 1 is due not to the successive removals but 
to an age factor quite apart from removal. Or the experiment may 
reveal no change in rate, because the distinctly appreciable effect 
of successive removal is destroyed by the influence of accessory 
factors. 
II GENERAL METHOD 
The present paper is concerned as far as possible with the effect 
of successive removal alone. Some of the sets of experiments were, 
however, controlled with much more effectiveness than others. 
The principal disturbing factors that have been considered are 
the following: 
tH Age. 
2 Periodic physiological changes such as those due to the molt- 
ing process in the Crustacea. 
3. Physiological changes as a result of prolonged subjection to 
laboratory conditions. 
