STEVENS: DIOECcIsM IN THE TRAILING ARBUTUS 533 
anthers. In the kind with pollen-bearing stamens he found the 
same variations in length of pistil and an equal variation in length 
of stamen. The two organs do not show any correspondence 
in length, however, but all lengths of stamens are found associ- 
ated with each length of pistil. Wilson concludes that the pistil 
in this form is functionless because of its imperfect stigma and 
that the species is really dioecious; and he makes the interesting 
observation that the flowers of the two forms differ, the corolla 
being from one third to one seventh smaller and generally more 
pink in the large stigma form than in the small stigma form. He 
also considers that Epigaea must have been trimorphic before 
becoming dioecious. 
Miss Langdon? examined a large number of flowers and found 
all the types described above. She notes that several arrangements 
of anthers may occur with each length of stigma, and that while 
usually all the flowers of a cluster are of the same form occasional 
exceptions occur. For example, on one spray there was a cluster 
of medium-styled flowers and one of short-styled; and in one short- 
styled cluster was one long-styled flower. Bastin! in a paper on 
the structure of Epigaea, discusses the flowers but all his observa- 
tions on this point are taken from Wilson’s article. 
The peculiar condition of the flowers of Epigaea was brought 
to the writer’s attention, in connection with some work on heter- 
ostylous plants, by Gray’s suggestion that Epigaea might be an 
example of a heterostylous species that either was approaching 
or had attained a dioecious condition. As will be noted from the 
papers cited above, there is considerable uncertainty as to whether 
Epigaea repens is really dioecious, that is, whether in the form 
having perfect stamens the pistil is incapable of producing seed: 
and as to whether there is any real evidence of a heterostylous 
condition. The writer has accordingly investigated these two 
points. In connection with this work, flowers from several 
localities in New England and New York have been examined, 
and a number of plants taken up and cultivated for experimental 
purposes. The writer wishes to acknowledge his indebtedness 
to friends who have sent him specimens and to Mr. F. V. Coville 
for generously sending information as to his methods of cultivating 
the plant. 
