65 
seminal sport in the individual; but the plant from which 
the drawing had been made, was lost. Last spring I received, 
through the kindness of J. Dale, Esq. of Manchester, several 
similar bulbs, which he had just received from Java. These 
plants were in all respects like to Eleutherine anomala, 
excepting that, when they began to flower in March, they 
confined themselves to the legitimate number of three stamens. 
They produced a long succession of flowers, most of which I 
did not see; but yesterday (August 8) I observed one with 
four stamens, and from the same plant in a completely meta- 
morphic state, another with four perfect stamens, a fifth 
anther upon a supernumerary or seventh small petal in the 
position of a stamen, and an eighth petal by its side without 
an anther, From this it may be collected, that the hexan- 
drous plant might have returned to its proper habit, and the 
redundancy have proved to be only a sport of the season, 
tending to the production of a double flower. Hexapetalous 
triandrous plants are in fact in a defective state Jike the 
pentandrous Azaleas, amongst which Az. pontica (Rhodo- 
dendron in a defective state) frequently produces at Spofforth 
six, seven, or more stamens, and one seedling of that species 
flowered double the first season, but never again. Some 
purchased Ixia bulbs flowered also full double two seasons, 
and afterwards reverted to single flowers.—W. H. 
70. Tue Genus EPIDENDRUM. 
In the miscellaneous pages of the Botanical Register the 
whole of the genus Epidendrum has now undergone revision, 
except that which may be called Epidendrum proper, con- 
sisting of species which cannot be referred to any of the other 
subdivisions, and which all correspond in having a leafy (not 
pseudobulbous) stem, a short peduncle without any scales or 
sheaths at its base,.and the lip completely united to the 
column. They form an assemblage in which very different 
looking plants are collected ; and which may therefore here- 
after demand further subdivision. No means of separating 
them occurs however at present, and till a larger number 
shall have been critically examined, it will be as well they 
should remain in one group. The following is an entirely 
artificial distribution, without much reference to their affinity 
with one another. 
