COURESPONDENCE. 27 
late bractsj broadest at tlie base, with scarioua raargius. Flowers white ; pe- 
tals small, round, and equal. Fruit smooth, tipped with the spreading styles ; 
carj^els with prominent ridges. 
I have been in correspondence with Mr. Syme, from whom I leam that it is 
a very variable plant, and like FlmpineUa Sascifraga is very changeable in the 
forms of its leaves. Owing to these yariationSj the older botanists, in dividing 
it into three species, viz. A, maJns^A.'intermedimn, and A, glaitcifolium^ described 
as species the two latter, from mere forms of the first ; and as Mr. Syme re- 
marks, modern authors consider them all one and the same species. 
Yours, etc, 
Gloucester, Bee. 21, 1864, O. St. Beodx, Ph.D., F.L.S. 
Oil Plants producing Jjoxihle Floivers, 
Permit me to add to your Hst of plants producing double flowers Viola tri- 
color, a handsome variety of which I saw in bloom last September at the 
Liverpool Botanic Grardens. Allow me also to remind you of the "VYliite Water 
Lily. 
In your list, I perceive that you class double flowers according to their Na- 
tural Orders. Would it not be preferable to do so according to their phy- 
siological pecidiarities ? Some, for example, as Chelidoninm majiis, multiply 
their petals without losing any of their stamens, and produce abundance of ripe 
seed. Others, such as Bihes sangiiineum^ not only multiply their petals, but 
the stamens also, every flower becoming, as it were, two or three rolled into 
one. In others, again, we have a gradual transition, few or no perfect stamens 
being retained, though the number of floral pieces, intermediate in character 
between stamen and petals is increased, often very greatly so, as in the double 
Tuhp. This latter circumstance shows that double flowers become so, not as 
supposed by some authors, simply through the portions that would naturally 
have appeared as stamens presenting themselves as petals, but, in the large 
number which have less than ten stamens in the single state, by a considerable 
extra development. Double flowers of this description should be carefully dis- 
tinguished from such as are polyandrous in the single state. Among those 
that show the gradual transition, some retain the pistil unchanged, while in 
others it is corrupted, and in others again it becomes an irregular cluster of 
expanded carpels bearing imperfect ovules, and reminding us of the fully ripe 
and expanded carpels of a StercuUa or a Firmiana. This last is remarkably 
exemplified in occasional states of the double Tulip. I have had them every 
season for many years past. In others, again, both the stamens and the pistil 
are almost or entirely obliterated, as in the common DaflPodiL It vrould be 
■well further to distinguish those which lose the whole of their stamens, 
retaining not a trace of them, yet preserve the pistils intact, and, like the 
double Chelidonimn, produce ripe seed, insects conveying the pollen to them 
from single flowers somewhere in the neighbourhood. This occm-s in the 
double Pseony, and appears to me to be one of the most interesting pheno- 
