88 INHERITANCE IN ANIMALS 



this implies that they are incompetent, I advise you to 

 measure a few angles so that you can be sure of the 

 minutes ; you will then begin to have some idea what 

 measuring an angle to a hundredth of a second means. 

 But while some of the uncertainty about any particular 

 determination of latitude is due to the slight inaccuracy 

 of the instruments, and to the fact that no one can read 

 them quite exactly, yet by carefully comparing the obser- 

 vations of latitude made at different places at the same 

 time, it has been shown that the position of the axis 

 about which the earth rotates, and therefore the position 

 of the equator, does in fact change to some small extent 

 from time to time, so that when we replace the record of 

 variable experience by a constant, we are attributing to 

 the latitude of the Radcliffe telescope a constancy it does 

 not really possess. 



Now, all experimental constants whatever are deter- 

 mined in essentially the same way as the latitude of 

 the Radcliffe telescope is determined. Men measure a 

 certain thing, and find that up to a certain point their 

 measurements agree with each other, and their experience 

 is uniform ; but beyond that point, their experience is 

 contradictory. Having made this contradictory record 

 with as much care as they can, they substitute for it one 

 constant value of the thing measured. But the question 

 always arises, how far this proceeding is justified, how 

 far the variability of the actual experience depends upon 

 imperfect observation, and how far it is a true record of 

 differences in the thing measured. It is just this region 

 of uncertainty, fringing our experience with a haze of 

 doubt, which fascinates the real experimenter. He is 

 always striving to reduce it within narrower limits, and to 

 enlarge the region of experimental certainty as much as 

 he can. I want to show you one case in which such an 



