204 DR. M. T. MASTERS ON THE DEVELOPMENT OF 
On the Development of the Andreecium in Cochliostema, Lem. 
By M. T. Mast¥rs, M.D., F.R.S., F.L.S,, &c. 
[Read March 7, 1872.] 
[Prarz IV.] 
In the ‘ Gardeners’ Chronicle’ for 1868, pp. 264, 323, I had occa- 
sion to describe that most remarkable and beautiful plant CocA- 
liostema Jacobianum, Lem., and to give the history of its intro- 
duction into this country. At that time I was only enabled to 
examine adult flowers, and, with some imprudence it must be con- 
fessed, I hazarded a theory as to the peculiarities of the staminal 
apparatus. I have since studied the course of development of the 
flower in question; and, as a result, I find that while my descrip- 
tion of the adult flower was in the main accurate, my theoretical ex- 
planation of its peculiarities was in many respects erroneous. It 
is with the view of correcting these errors and inducing others to 
investigate the peculiarly remarkable structure of this splendid 
plant, that I now beg to lay before the Society the following par- 
ticulars as to the development of the flower, and especially of the 
andreecium. 
The flower first appears in the form of a globular tubercle (Pl. IV. 
fig. 1), on one side of which shortly appears a thick semilunar pro- 
minence, which is the origin of the posterior sepal (fig. 2). The two 
remaining sepals are formed in succession, one after the other(fig. 3). 
I have never absolutely seen the flower with two sepals only ; but, 
from the small relative size of one of the three sepals (that placed 
next the bract) and from analogy with other flowers, I have no 
doubt that the three sepals are formed in succession. If the three 
sepals be turned back with the dissecting-needle, as shown in 
fig. 4, the central growing point may be seen to have lost its glo- 
bular form, to have assumed a triangular outline, and to be some- 
what depressed at the top. 
From the three corners of the central growing point protrude 
simultaneously the three petals (fig. 5), which only precede the ap- 
pearance of the stamens by avery short interval—so short, indeed, 
that one may examine a great many flowers before finding one 1n 
which the petals alone are formed without indications of the 
stamens. At this period the superiority in relative size of the 
posterior and first-formed sepal is very marked. 
As to the androcium, it is perfectly clear that this consists of 
six elements ; and all analogy would suggest that these six portions 
