THE IDENTIFICATION OF VARIETIES OF BARLEY. 9 



of the problem to be solved. When it is realized that a great number 

 of these forms are known and that there is no reason why all of them 

 can not be produced, or indeed may not be found already in exist- 

 ence in some part of the world, the futility of continuing the policy 

 of describing each minor variation becomes apparent. In fact, there 

 is more to be done than merely resolving not to add to the confusion. 

 Over 200 forms have been named and described. Some semblance 

 of order must be brought about in the work already published and 

 some constructive plan suggested whereby future additions may be 

 limited to variations of real importance. When the variable char- 

 acters are studied carefully with this object in view, it soon becomes 

 apparent that the factors are not all of the same value. 



The fertility and the adherence or nonadherence of the lemmas are 

 major distinctions. They seldom present any difficulties. The 

 nature of the appendages of the flowering glume are major distinc- 

 tions so far as the distinctions between awns and hoods are con- 

 cerned. The presence or absence of hoods is obvious. In the sepa- 

 ration of varieties the term " awned " has been made to include awn- 

 less and all intergradations, as a complete series from awnless to fully 

 awned seems likely to exist. White and black kernels are also major 

 distinctions. White has been used to include blue and purple, because 

 of the frequently imperfect production of the latter pigments and the 

 difficulty of determining doubtful varieties. By using only these 

 four groups, the varieties are founded upon distinct and readily sepa- 

 rable characters. Each variety is then a specific, tangible group. 



As lesser but still usable distinctions there are the awnless and 

 smooth-awned variations of the awned form, the blue and purple 

 variations of white, and the character of the outer glumes. 



Besides these, a consider"able number of characters have been used 

 in establishing varieties which in the light of present information 

 seem unsuited as bases for taxonomic groups. In many instances 

 they are very useful in agronomic work in the description of biotypes 

 and in the identification of farm varieties, but they are not here recog- 

 nized. Varieties which have been established upon these characters 

 will be found only in the alphabetical list of rejected forms. These 

 characters will not be discussed individually, but will be found 

 listed in the summary below. 



To summarize the variations, the three grades of distinction are as 

 follows : 



Major distinctions useful in establishing species and varieties: 

 Spikes 6-rowed, intermediate, 2-rowed, and deficient. 

 Kernels bulled and naked. 



Lemmas awned (including awnless) and hooded. 

 Kernels white (including blue and purple) and black. 

 14681°— 18— Bull. 622 2 



