1882.] 



The Hetcr agony of Oxalis viol ace a. 

 TABLE IT.— Oxalis violacf.a. 



Flower Numbers. 







Short Stamens. 





2.5 " 

 2.5 •« 



3- ;; 



5- ;; 



5- '* 

 5- " 



5-4 " 

 5-4 " 

 5- " 



51 ;; 



5- ;; 



5- " 



5- ;; 





















































;s 



U :: 













3-5 ;; 









|EE::::::::::::: 



3.8 « 











5:::::::::::::::::::::::: 



I ■■ 



Though both stamens and pistils vary in length, as might, in- 

 deed, be expected from the fact that the flowers are by no means 

 of uniform size, a glance at the tables and appended diagram 

 shows that, as a rule, the styles of either form are intermediate in 

 length between the two sets of stamens belonging to the other 

 form ; while the difference between the stigmas and the nearest 

 set of anthers is, in cither, greater than that between the stamens 

 themselves, both differences being nearly constant for both long 

 and short-styled flowers. 



This is slightly different from the usual arrangement of the 

 parts in trimorphic species, as may be seen by comparing Figs. I 

 and 2, representing the species under consideration, with Figs. 3, 

 4 and 5, after Hildebrand, representing the trimorphic 0. gracilis. 

 That the long stamens of the long-styled flowers, amfthe short 

 stamens of the short- styled flowers stand at different heights— 

 as may be most clearly seen by comparing the lines a and />' in 

 the diagram— and not at the same height, is, it seems to me, oi 



