GRADATION 205 



(has reached its final state of equilibrium) with reference to the 

 primordium length when the latter was 36 mm., whereas the 

 growth of the length of internode VII. has been stopped when 

 the value 126 mm. was reached. Therefore a gradation curve 

 may be compared to a curve of development. Especially in 

 those numerous gradation-curves in which the increase is con- 

 tinuous from one extremity to the other (without waves or 

 secondary humps), the successive ordinates represent, as it were, 

 a series of individuals (segments) measured in successive states 

 of development {example : internodes I. -VIII., first stem, p. 200). 



The above comparison is, of course, only applicable when each 

 primordium is taken separately, as an independent something. 

 The aspect of the question is completely changed if we consider 

 each individual (specimen or segment) as a whole. When we 

 compare, for instance, in Holcus mollis (p. 202), leaf VIII. with 

 leaf XIII. , we see that XIII. is comparatively juvenile (in an 

 embryonal state) in respect of the length of the sheath, whereas the 

 same leaf XIII. is further advanced in its development (com- 

 paratively senile) in respect of the length and the breadth of the 

 blade. (Compare colour and length of the corolla in Centaurea 

 cyanus, p. 58 ; see also p. 52.) One may find numerous 

 similar examples in my paper on Mnium. 



In the study of gradation, as well as in the study of embry- 

 ology, each primordium must be measured at first separately in 

 order to obtain EXACT information. The gradation curves of 

 several primordia being plotted down into the form of one dia- 

 gram (Fig. 23, p. 158, and Fig. 27, p. 203), exact notions about 

 each individual taken as a whole become attainable. 



The notion of pcedogenesis and the terms pcadogenetic form, 

 juvenile state, embryonal form, etc., when applied on individuals, 

 have been often the expression of vague conceptions.^ When 

 applied on species, genera, families, etc., those conceptions have 

 been the cause of erroneous conclusions and many contradic- 

 tions. The fruit of laborious phylogenetic researches, carried 

 out for half-a-century, has been partly spoiled because we 

 have overlooked the FUNDAMENTAL PRINCIPLE OF THE 

 INDEPENDENCE OF THE PRIMORDIA. (See § 33 : 

 MENDEL.) 



§ 142. — In the present book attention has been repeatedly 

 called to the importance of gradation, and several curious and 

 even unexpected consequences of gradation have been men- 

 tioned. In this paragraph I wish to recall the most important 

 passages in which gradation is alluded to : 



(i) § 38 (p. 43) • Segregation of properties produced by 



1 For instance, with reference to the comparison between male and female 

 individuals of the same species (sexual and secondary sexual characters). 



