538 STEVENS: DIOECISM IN THE TRAILING ARBUTUS 
and small-stigma flowers. If flowers of the two forms growing 
in the same locality are compared, the pistillate flowers seem to 
be uniformly smaller than the apparently perfect flowers. Darwin 
(loc. cit. 307) considers that this is generally the case in gyno- 
dioecious plants, and names a dozen such species in which the 
pistillate flowers are smaller than the staminate flowers. He 
notes also the fact that no such difference in the size of the corolla 
has been found in the two forms of androdioecious species. 
The writer has been unable to demonstrate, however, any 
correlation of color of the corolla with the two forms of stigma. 
Wilson (loc. cit. 59) states that the corollas of the large stigma 
flowers are pink, while those of the small stigma flowers are white. 
This is apparently true in some localities; but in one lot of flowers 
sent from a locality in Maine nearly all the flowers of both forms 
were pink. Moreover, a small lot of flowers from N ew Hampshire 
showed the large stigma flowers small and white, while the small 
stigma flowers were larger and very deep pink. 
MORPHOLOGY OF THE SEED 
In the sections made for the purpose of determining the rate 
of growth of the pollen tubes several points were noted which 
seemed worthy of record, and the development of the seed has 
AL 
Sete) 
G. I. Longitudinal section through ovule of small stigma form of Epigaea 
repens showing ‘‘tapetum” 
