204 GENERAL BIOLOGY 



were, about the median dimension, whereas in the 

 flatter curve they are distributed more evenly along 

 the whole base-line. In other words, the bulk of 

 the individuals of the first lot are much alike, or, 

 as we say, are not so variable as in the other lot. 

 It is evident that the flatter the curve, the farther 

 away from the modal axis (M) moves the point of 

 inflection. The measure of the line connecting this 

 point and the modal axis (designated by the Greek 

 letter <r) thus becomes the measure of variability, 

 being greater in proportion as the curve is flatter, 

 i.e. in proportion to the greater variability. 



Asymmetrical Variation. In actual observations 

 the variation curve for organisms is rarely quite 

 symmetrical ; that is, the mode is somewhat nearer 

 one extreme than the other. The index of varia- 

 bility may be calculated for such curves as for 

 symmetrical ones. If the mode is much nearer one 

 extreme than another, the curve is spoken of as 

 " skew." Such a curve frequently indicates that 

 selection has taken place in the material examined, 

 and the variates have been eliminated dispropor- 

 tionately. 



Discontinuous Variation. In measuring the di- 

 mensions of an organ or in estimating weights, the 

 number of classes obtainable is limited only by the 

 accuracy of our own observations, and the gradations 

 of variation are continuous, as is indicated by the 

 symmetrical curves which we usually obtain. Such 

 variations as the above are due to the variability 



