Progressive, Retrogressive, Degressive Mutations. 375 



oiis or syncotylous individuals. The seeds, li<nve\er, of 

 both types give the same proportion of tricot}'ls or syn- 

 cotyls even when self-fertilization has been insured. 



If we wish to elaborate terminologies still further, 

 the term semi-latent may be limited to the anomalous 

 character of the half race, and the character of the middle 

 race may be described as semi-active (see p. 21). \.\'e 

 can then distinguish four conditions of one and the same 

 factor; the active, the latent, the 

 semi-active and the semi-latent. 

 This classification may suffice for 

 the time ; and at any rate it can 

 be said that it is in accord, so far 

 as my experience goes, with the 

 facts known at present. 



Such a factor cannot be trans- 

 ferred at will from one of these 

 conditions to the other, either 

 by selection or by any other 

 means, at any rate in the present 

 state of our knowledge. Such a 

 transition is onlv effected bv com- 

 binations of causes of which we 

 still know nothing, or as we say 

 by chance. Moreover the tran- 

 sitions, so far as we can observe them, arc not slow or 

 gradual, but take ])lace suddenly. The new race apix\'irs 

 on the scene at once and unexpectedly, as in the case <>f 

 the peloric Linaria. Sudden transitions of this kind arc 

 exactly what we call mutations; and to distinguish thcni 

 from the progressi\'e and retrogressive t}-pes. we may 

 refer to them as degressive nuitations. 



Every mutation therefore consists fundamentally \n 



^ig. 135. Papavcr coin- 

 iiiutattmi polycct^lialum. 

 The same anomaly wliicli 

 occurs in P. soninifcntm 

 as a middle race (Vol. 1, 

 Fig. 27, p. 138) occurs 

 here as a half race, mani- 

 festing the character very 

 rarely and only to a small 

 extent. 



