152 



SCIENCE. 



[N. S. Vol. XX. No. 500. 



the value of this evidence it is necessary first 

 of all to disentangle the fact that the crab 

 has a mechanism to facilitate self -mutilation, 

 from the fact that the mutilated parts are 

 restored. This distinction is not only easy 

 to make, since the legs regenerate at other 

 levels, but it is also very important. One 

 who recognizes the independence of these two 

 facts does not hold the foolish opinions attrib- 

 uted to him any more than he accounts for 

 his ability to mend a broken clavicle by refer- 

 ring to gifted ancestors whose success in life 

 depended on the frequency and completeness 

 with which they broke their collar bones. 



The separation between the fact that there 

 is a mechanism for throwing off legs, and the 

 fact that the legs are regenerated from the 

 point at which they are thrown off, leaves for 

 consideration only the basis of the belief that 

 the breaking joint is not of use to the species. 

 The evidence for this belief is as follows : 

 ' At any one time only a small percentage, 

 about ten per cent., have regenerating legs, 

 and as the time required completely to re- 

 generate a leg even in the summer is quite 

 long, this percentage must give an approxi- 

 mate idea of the extent of exposure to injury.' 

 Thus the extent to which the mechanism is 

 used is held to be too slight to account for 

 its existence in the other ninety per cent, of 

 the crabs. However, as this determination is 

 only for ' any one time,' it falls into a class 

 of statistical evidence which shows, according 

 to Professor Brooks, that ' our subject matter 

 lies midway between those exact sciences in 

 which we are told that figures can not lie 

 * * * and those social and political sciences 

 which show us continually how easily one may 

 lie with figures.'* 



Granted that at any given time ten per 

 cent, of crabs show that they have made use 

 of the mechanism for throwing off their legs, 

 this percentage gives no idea of the extent 

 to which each crab uses the breaking joints 

 during its entire life. How long a crab lives 

 is not definitely known, but from analogy and 

 indirect evidence five years is within the limit 

 of life for some species. ' As the time re- 



* W. K. Brooks, ' The Intellectual Conditions 

 for Embryological Science,' Science, XV., p. 488. 



quired completely to regenerate a leg even in 

 the summer is quite long,' it follows that in 

 six months an appendage may regenerate com- 

 pletely. 



Five years represent ten periods of six 

 months. If in each period we were to count 

 the ten injured individuals of a given hun- 

 dred, then at the end of the full term we 

 should have counted one hundred injuries, 

 which, according to the doctrine of chance, 

 would have been distributed among sixty-five 

 individuals. Thus in five years, two out of 

 three crabs would have been injured one or 

 more times. 



Regeneration. — Professor Morgan's book is 

 one continuous protest that natural selection 

 does not account for the ability of organisms 

 to regenerate lost parts. Thus on the last 

 page of ' Regeneration ' he summarizes his 

 convictions in the following words : " It seems 

 highly probable that the regenerative process 

 is one of the fundamental attributes of living 

 things, and that we can find no explanation 

 of it as the outcome of the selective agency 

 of the environment. The phenomena of re- 

 generation appear to belong to the general 

 category of growth phenomena and as such are 

 characteristic of organisms." 



This demonstration, ' that the regenerative 

 process is one of the fundamental attributes 

 of living things,' seems valid; but to those 

 who believe that natural selection is a law of 

 nature, proof that the regenerative process is 

 fundamental is likewise proof that natural 

 selection has no bearing on this process. 

 Natural selection is not an explanation of 

 things ultimate any more than the law of 

 falling bodies is an explanation of the funda- 

 mental characteristics of matter. No one 

 holds that Newton's laws are invalidated be- 

 cause they do not explain the ultimate attri- 

 butes of materials that fall, or of the space in 

 which they fall, or why they fall in the order 

 that we observe, because every one knows, or 

 has known, that Newton's laws are merely 

 records of events. Natural selection is the 

 series of events which occurs in nature as the 

 outcome of individual differences, the high 

 rate of increase and the environment of liv- 

 ing things. The charge, therefore, that this 



