Wilson. — Observations on Epigeca re pais, L. 59 



two forms do not occur on the same, but on different indi- 

 vidual plants. 



Epigaa npens is therefore, not becoming dioecious, but is 



already so. 



If a large number of flowers be examined in any given 

 locality, a few will be found with long styles that carry the 

 fertile or infertile stigma, as the case may be, considerably 

 beyond the throat of the corolla (Plate VIII, Figs. 3, 6). In a 

 much larger number of flowers will be found an extremely 

 short form with stigmas only half elevated in the tubular 

 corolla (Figs. 1, 7). 



All intermediate forms may be found if one examines a 

 sufficient number of plants (see Plate VIII, Figs. 2, 3, 4, 8). 

 Two of the most common forms are Figs. 2, or 8 and 5. One 

 can make out a tri-morphic condition of this plant, and quite 

 as easily show, with a still farther examination of flowers, a 

 polymorphic condition. In some of the flower sections on 

 the Plate bare rudiments of stamens exist, as in Figs. 2 and 

 3. In Fig. 4 the anthers are present without pollen. Only 

 one plant with flowers in this stage of development was found 

 while examining a large mass of material from North Caro- 

 lina, Pennsylvania, and New Jersey. 



In Fig. 5 the stamens are short ; in Fig. 7 they are long. 

 In others they occupy an intermediate position. 



Figs. 5 and 7 show the predominating forms. At the 

 present time cross-pollination is secured with absolute cer- 

 tainty through dicecism. 



The rudimentary stamens and pistils in the different kinds 

 of flower show that this has not always existed, but that the 

 flower was once perfect. In this early and perfect condition 

 of Epigcza it is evident that it developed, as many other 

 flowers have, these varying lengths of styles and stamens to 

 aid in securing cross-pollination. 1 



There are differences between these two kinds of flower 

 other than those which relate to the essential organs, the 

 stamens and pistils. These differences indicate that while 



1 Any one not familiar with di-and tri-morphic flowers and the way in which they 

 aid cross-pollination, can consult Lubbock's British Wild Flowers. 



