A BEACH YTIC VARIATION IN MAIZE. 25 



usual position are found. The intermediate stages between these 

 well-formed ears and branches which have only a few pistillate 

 flowers at the base closely resemble the ears that terminate in stami- 

 nate spikes. 



The pistillate branches, or ears, are clustered around the base of 

 the tassel, the internodes being so short as to be indistinguishable 

 from one another. (See PL XVII.) The transformation from male 

 branches to ears is accomplished by a shortening of the rachis which 

 is first noticeable at the base of the branch. In the advanced stages 

 this shortening results in a crowding of the spikelets, which fre- 

 quently is augmented further by a zigzagging of the rachis. In these 

 cases there often is evidence of twisting. The spikelets in the early 

 stages of the transformation of the branches are little altered, but 

 with the added shortening of the rachis the glumes become reduced 

 in length and gradually assume the characteristics of glumes on 

 normal ears (Pis. XVIII and XIX). 



While the tendency to have ears clustered at the base of the tassel 

 is found on plants with normal internodes as well as on brachytic 

 plants, the nature of the transformation from staminate to pistillate 

 branches strongly suggests brachysm. Thus, there is an abrupt short- 

 ening of the main axis at the point where the branching occurs and 

 also a marked shortening of the rachis of the branches, accom- 

 panied by a development of pistillate flowers. These changes are 

 essentially those which took place in the development of the normal 

 ear and are similar to those indicated in the brachytic plants of the 

 Chinese- Algeria hybrid. 



Taken in connection with the fact that half the brachytic plants 

 of the Hopi-brachytic hybrid produced ears terminating in staminate 

 spikes and that the brachytic plants had but few nodes above the ears, 

 the close resemblance of tassel ears to ears in the normal position with 

 staminate spikes suggests a possible origin of these staminate spikes. 



Viewed in this light, the ear of maize may have developed not 

 from a central spike of the terminal inflorescence of a lateral 

 branch but from the transformation of the lower staminate branch 

 of the terminal panicle into a pistillate branch. If this hypothesis 

 is adopted, the sterile nodes which intervene between the pistillate 

 and staminate inflorescences are not so difficult to explain and be- 

 come with the ear a part of the terminal inflorescence. 



The lower branches of the tassel occasionally develop into re- 

 plicas of normal plants and even more frequently are subtended by 

 bracts. These lower branches are relatively unstable, being the 

 first to develop pistillate flowers, and the number of rows of spikelets 

 often exceeds four. This increase in the number of rows may arise 

 through the reduction of secondary branches or perhaps from a zig- 



