20 



naturally, but freely appearing under artificial cross- 

 pollination. They fourtd stigmas of unopened flowers with 

 l^ollcn tubes freely developing upon them. Abraham 

 l-'latters, in a paper on " Microscopical Research," read in 

 lyoO before the Manchester Microscopical Societ}-, likewise 

 maintained, and illustrated this point with micro-prepara- 

 tions from unopened flowers. 



Rendle (1904), Vol. I. " Classification of Flowering 

 Plants," pp. 230-1, states- " Grasses arc self- or wind- 

 jjoUinated. . . . The species of wheat are generally self- 

 pollinated, and in some cultivated races of barley the 

 flowers ncxer open." But evidence is not advanced in 

 support. 1 have seen notes of lectures given in February. 

 1 90 1, by Professor M. C. Potter, Newcastle-on-Tyne, in 

 which self-fertilisation is recorded for barley. 



The number of flowers and their arrangement in the 

 spikelet make it difficult to devise means for field experi- 

 ments whereby all error in observation may be prevented. 

 It has been readily assumed that the self-received pollen 

 is also that which effects fertilisation, but I have not seen 

 an)' micro-preparations the histor}- of which can be called 

 conclusive on the matter. The fertilisation process is quite 

 normal, and there is no evidence of parthenogenetic or 

 other adventitious conditions of embryogeny. 



The pollen grains in wheat before being shed from 

 the anther contain the vegetative or tube nucleus and the 

 two generative or male nuclei (Figs. 11 and lia)," not 

 unlike the antherozoids of a fern or of chara," as Golinsky 

 suggests and figures for rye. This stage appears about 

 20th to 30th June, fertilisation becoming general about the 

 hitter date ; on the 3rd July, 1906, young embryos of six 

 or more cells were freely developing ; those in material of 

 July 14th, 1900, were much further advanced (Figs. 28-29). 



