408 . DR. J. C. COSTERUS ON MALFORMATIONS 
The questions now arise: Which whorl is primary, and which 
has developed from the primary one? Do the petals belong to 
the original series, and did they give birth to the stamens, or are 
the antipetalous stamens the older elements which have subse- 
quently produced petals? This question is closely connected 
with another: Are the apetalous Fuchsias of South America and 
New Zealand the representatives of the prototype, from which 
the corolla-bearing Fuchsias have developed themselves, or are 
they to be regarded as the descendants? Both suppositions 
are in themselves possible, and both throw a strong light upon 
the cause of the otherwise inexplicable arrangement of either the 
petals or the antipetalous stamens with regard to the cells of the 
ovary. If, however, one of these whorls is suppressed, the irregu- 
larity disappears, and the law of alternation remains in full force. 
The following facts tend to render it probable that the petals 
are to be looked upon as the primary parts, from which the outer 
stamens have grown out, or, in other words, that the original dia- 
gram must have been as shown in Pl. LIX. fig. 375. In proof of 
this we would first adduce the Fuchsia of Simroth: this botanist 
possessed a flower which for convenience’ sake we represent by 
the diagram in fig. 37a. It shows that two of the petals have no 
stamen in front of them. Besides, the sepal at the top of the dia- 
gram was foliaceous, whereas the lowermost was red as usual ; the 
two lateral ones being half green, half red, so that the green half 
of each sepal was turned upward and the red half downward. 
According to Simroth, the flower may be conceived to consist of 
two parts, of which the undermost is quite normal, whereas the 
uppermost shows modification owing to the absenee of anti- 
petalous stamens aud the phyllody of the sepals. Now Simroth 
takeg the uppermost half to be a reversion to a former structure, 
and shows that the ancestral Fuchsia differs from the present 
form by two characters—1, the absence of an outer whorl of 
stamens; 2, the possession of a leafy calyx. We readily agree to 
this view and will iry to strengthen it by further arguments, 
which would certainly seem necessary to furnish a firm base to 
Simroth's opinion. 
Some years since the development of the flowers of some 
Onagrariacee was examined by Barcianu. His investigations 
brought to light that the outer or antipetalous stamens do not 
belong to the autonomous organs of the flower. It is not until 
the ealyx and the other whorls have been given off from the recep- 
tacle, that on the inner surface of each of the petals a small tubercle 
