462 



Journal of Agricultural Research 



Vol. XIX, No. 9 



The samples at Aberdeen consisted of at least two spikes. Just after 

 flowering, when the kernels were small, three spikes were used. In 

 Table IV only two of these are reported because the inclusion of the third 

 makes the table even more cumbersome. In this table the steady 

 growth of the kernel is apparent. Even when not averaged, the maxi- 

 mum kernel weights during the early part of the period constitute a very 

 uniform series. The difference between the clipped and undipped 

 spikes becomes increasingly apparent as growth progresses. 



The average weights and measurements in Table V are more easily 

 studied than are the unsummarized data in Table IV. Table V gives 

 the average by days. In some instances abnormal kernels have been 

 thrown out, because they introduce variations that may as well be 

 excluded. The kernels from the clipped spikes often exceed those of the 

 normal spikes in weight and dimensions during the first week after 

 flowering. As was the case in Minnesota, the normal spikes soon out- 

 strip the clipped ones and maintain their advantage until maturity. 

 The comparative development is illustrated in figures 7 and 8. 



Table V. — Average -wet -weight, length, lateral diameter, and dorsoventral diameter of 

 kernels from, normal and clipped spikes of Hannchen barley from flowering to m,aturity, 

 at Aberdeen, Idaho, in igi6 



UNCLIPPED SPIKES 



CLIPPED SPIKES 



