322 BOTANY 



recently shown that, at least in a particular stage of development, the 

 homozygous plants of Urtica pilulifera can be distinguished from the 

 heterozygotes. Nevertheless it may be said that two plants possessing 

 different determinants may be apparently similar, while on the other 

 hand two organisms possessing the same determinants may appear 

 distinct owing to diverse action of the environment. The nature of 

 the determinants which are contained in a plant can thus not be 

 discerned from its appearance but only by breeding experiments. 



3. AUTONOMY OF CHARACTERS. When the parents differ in two 

 characters instead of only one, monohybrids instead of dihybrids 

 result. It then appears that the several characters are independ- 

 ently transmitted and distributed in the descendants (autonomy of 

 characters). Thus new combinations of characters may come about, a 

 fact of great importance in plant-breeding. From the crossing of 

 peas with yellow, wrinkled seeds, and those with green, smooth seeds, 

 among other possible combinations of the characters the new ones 

 yellow-smooth and green- wrinkled appear. Many characters, however, 

 tend to remain associated together (coupled characters). 



It is not possible to enter in this place into the complicated 

 phenomena of the production and segregation of dihybrids and 

 polyhybrids. 



Validity of the Mendelian Rules. These rules are not limited 

 to hybrids in the narrow sense of the word, but have an extensive 

 application to inheritance in both the animal and vegetable kingdoms. 

 It cannot be said that there are not other laws followed in inheritance, 

 for there are already well-investigated cases which do not conform to 

 the Mendelian rules ( 84 ). On the other hand, it is noteworthy that 

 many phenomena which at first appeared to contradict these rules 

 have proved on further investigation to be consistent with them. 



Variability ( 85 ). By variability is understood the fact that the 

 individuals belonging to any species are not all alike. Frequently 

 the variability is only apparent, the species not having been properly 

 defined. Thus in Rosa, Bubus, Draba verna, etc., there are many 

 species that closely resemble one another. The impression given of a 

 " varying " species is in these cases a completely false one ; each 

 of the "ELEMENTARY SPECIES," of which the "COLLECTIVE SPECIES" 

 is composed, proves to be constant and does not exhibit transitions to 

 the other elementary species. 



Such cases are to be left out of consideration here. We are 

 concerned with the most strictly limited species, if possible with the 

 descendants of a single self-fertilised plant constituting what is known 

 as a pure line (JoHANNSEN). It is found that these also vary. The 

 process of variation and the varieties can be traced to two causes 

 and are therefore distinguished as MODIFICATIONS and MUTATIONS. 

 To these must be added the combinations originating from crossing. 



MODIFICATIONS. This name is given to variations which have 



