224 Rhodora [NOVEMBER 
Our plant differs in many ways from this. The calyx is normal, 
and the corolla, although much modified, is still two-lipped, while, as 
stated, the most import- 
ant variation is in the 
anthers. Furthermore, 
, no flowers of the normal 
form were found on the 
same plants.  Masters,' 
in treating of phyllody, 
states that this occurs less 
frequently in the stamens 
than in the neighboring 
organs; that sometimes 
the whole stamen is af- 
fected, at other times only a part ; that the change of the anther from 
its ordinary condition to that of a leaf, indicates a great degree 
of perverted development. He gives a list of species in which 
phyllody of the stamens has been noted, but does not mention any 
species of the order Zadiazae. 
In all the flowers of our plant the anthers are changed to leaves, 
varying somewhat in form, and in some cases the leafy growth ex- 
tends downward on the filament. The style and stigma of these 
flowers seemed normal and perfect, but were, so far as observed, abor- 
tive, and did not produce seed, although surrounded by an abundance 
of flowers with perfect anthers. Unfortunately, notes as to this point 
were not as extended as desired, owing to the untimely mowing off of 
the colony under observation. 
The accompanying figure of one of the flowers is from a drawing 
kindly prepared by Mr. Charles E. Faxon, The modified stamens 
are shown at s7. 
SOUTHINGTON, CONNECTICUT. 
AUTUMNAL FLOWERING OF VACCINIUM PENNSYLVANICUM. — It was 
an interesting experience in the middle of September to find in a de- 
serted overgrown timber-road near Westford, Massachusetts, a quantity 
of the early low blueberry, Vaccinium pennsylvanicum, covered with 
blossoms. ‘The unusual effect of the buds and blossoms, among the 
old leaves, suggested the southern evergreen species, Vaccinium Jor- 
! Vegetable Teratology, 253-256. 
