PKEFACE TO THE SECOND EDITION. 



SINCE the publication of the first edition of this 

 book in 1877, several articles have appeared on the sub- 

 jects therein discussed, and many letters have been re- 

 ceived by me. I will here briefly state their nature, 

 as an aid to any one who may afterwards pursue the 

 same subjects. The text has been left as it originally 

 appeared, excepting that a few errors have been cor- 

 rected. 



Dr. A. Ernst has proved in the clearest manner 

 '(' Nature/ Jan. 1, 1880, p. 217) that Melochia parvi- 

 folia, which is a common plant near Caracas, is hetero- 

 styled. The pollen-grains differ in the usual manner 

 in size in the two forms, as do the papillae on their stig- 

 mas. The illegitimate unions, especially when pollen 

 from the same flower was employed, were much less fer- 

 tile than the legitimate ones. A new family, the Bytt- 

 neriaceae, is thus added to those including heterostyled 

 plants. 



Errara and Gevaert have published a paper on the 

 heterostylism of Primula elatior, in ' Bull. Soc. E. Bot. 

 Belg.,' torn, xvii., 1879. 



I have quoted (p. 100) Dr. Alefeld's statement that 

 none of the American species of Linum are heterostyled. 

 This statement was disputed by Kuhn ('Bot. Zeit./ 

 1866, p. 201), but has since been confirmed by Dr. Ign. 



xiii 



