204 POLLINATION AND CROSS-FERTILIZATION IN THE JUAR PLANT 



were pendent. In a small number of flowers in the more compact type of 

 panicle the anthers did not assume the pendent position remaining partially 

 within the glumes close to the stigmas (Fig. 3) : while in a few of the last flowers 

 to open for the flowering period of that night there was no dehiscence of the 

 pendent anther. The opening of the pores of the anthers caused no scattering 

 of pollen and it was only when the anthers had fallen over that pollen 

 escaped. The flowers therefore are protogynous as is also admitted by 

 Koernicke^ and Kirchner.2 The details of the process related above vary 

 considerably, however, fiom those described by the two author?,. 



The whole process from the time of the opening of the glumes till the 

 anthers assumed the pendent position occupied 10 minutes. Instances were 

 not wanting in which the process took place in as short a time as 3 minutes, 

 while, on the contrary, the process may occupy 30 minutes. Finally as stated 

 above the stamens may never become pendent. 



The flowers of the juar opened only once, the glumes remaining open 

 for 2 — 3 hours, the closing of the glumes being a much more gradual process 

 than the opening. The stigmas remained outside after the glumes had closed 

 and appeared quite fresh for as long a period as 24 hours after the glumes had 

 closed. (Fig. 2.) The length of time taken for the whole panicle to complete 

 flowering varied with the size of the inflorescence and the number of flowers. 

 The average time was about seven days. 



From the observations it is evident that cross-pollinatioii between flowers 

 of the same panicle is the rule, the pollen from the higher and earlier opening 

 flowers falling on and pollinating the stigmas of lower and later flowers. This 

 is the conclusion also arrived at by Fruwirth^. Cross-pollination by foreign 

 pollen can only take place in the fiist flowers to open, and possibly in the case 

 of late opening flowers whose anthers do ngt dehisce. Self-pollination can 

 only take place in late opening flowers whose stigmas remain surrounded by 

 the anthers which do not fall out, but even in those cases the chances are 

 in favour of pollination from higher flowers in the same inflorescence. Though 

 typically anemophilous the flowers of the juar were visited in considerable 

 numbers by insects. In addition to small beetles which crawled over the 

 panicle, the flowers were regularly visited by bees. During the months of 

 March and April the first bee to arrive soon after day-light was a small 

 bee with white stripes {Apis florea). This was followed by the large wild bee 



1 Koornicke, F. Loc. cit. 



2 Kirchaor, D. Neue Beohachlungen, p. 11. 



8 Fruwirth, C. Die Zuchtung der landuirtscliaftlicticn Kulturpflanzen, Bd. V, 1912, p. 50. 



