Experiments on the Hybridisation of Barleys. 307 



dense types. One other point has to be briefly noted here — that 

 is, that, knowing the range of the variability of the parents, one 

 should be able to calculate from the F 2 curve the number of indi- 

 viduals in the three groups of which it is composed. This assumes 

 that the fluctuating variability of the extracted lax and dense 

 forms is the same as that of the parents. 



This does not appear to be the case, and a considerable amount 

 of evidence, not only among the barleys but also among the wheats, 

 has accumulated, which shows that the range of variability of the 

 extracted characters is widened to an appreciable extent. Thus 

 amongst the lax individuals are some in which the ears show as 

 much as a 10 per cent, increase in this character. It is also doubt- 

 ful whether the average internode length of the extracted types is 

 the same as that of the lax or the dense parents. What little 

 evidence we have on this point leads me to suspect that the average 

 length itself is altered. To obtain the necessary statistics will be 

 tedious work, but, in view of the importance of the question to 

 those who are attempting to apply Mendelian principles to the 

 improvement of our crops, it will have to be undertaken. 



As a proof of the value of these observations to the plant- 

 breeder it may be stated that the extremes selected from the F 2 

 have in the cases tested bred true to this particular character. 



The crosses between varieties showing different degrees of 

 development of the lateral florets have given some interesting 

 results. In a previous communication I pointed out that where 

 a variety with no lateral florets was crossed with one possessing 

 male florets the F 1 resembled the former parent in this respect, 

 inasmuch as that, though very minute florets were produced, they 

 were altogether sexless. Since then the F 2 of a number of such 

 crosses has been grown, and also one F 3. In each case the F 2 

 consisted of individuals with either no lateral florets, or with 

 minute and sexless laterals, or with well-developed male laterals. 

 The three groups occur in the proportion of 1:2:1, and they are 

 perfectly distinct from one another. In the single F 3 so far 

 tested the individuals with the minute laterals have invariably 

 proved to be heterozygotes, whilst those with no laterals and those 

 with male laterals have proved to be homozygotes. The sexless 

 condition of the laterals may therefore be described as dominant. 

 Where (me parent has black paleae in such crosses the lateral 

 florets, both male and rudimentary, are white. 



A number of hybrids between varieties of this decipiens type 

 and varieties with fully fertile laterals have been raised. The F 1 

 in such cases bears small lateral florets containing either vestiges 

 of stamens or fully developed ones. The heterozygote is thus a 

 two-row barley in the popular sense of the word, or from the 

 systematic point of view a distichum or a zeocriton according to 



21—2 



