Vlll 
with its lateral auriculae. Upon the supposition that the so called petals 
are really the external series of the andrcreeum, it becomes almost 
necessary to consider all the stamens ever present in Orchidaceae, as 
belonging to one and the same series, of which the central stamen is 
complete in the majority of the order, and the two lateral ones in Cypri- 
pediuni, and this seems to be confirmed by the triandrous monster of 
l’lalanthera already referred to, in which Dr. Brown describes the addi¬ 
tional stamens as opposite the lateral sepals. Nevertheless this acute ob¬ 
server is of opinion that the one perfect stamen in the monandrous genera, 
and the two of Cypripedium, belong to different series. This theory is 
founded partly upon the position of the stamina in Apostasia, a genus 
which cannot be considered actually to belong to the order, and in 
which of three stamens, two are described as being perfect and opposite 
the lateral petals, while one is sterile and opposite the intermediate sepal, 
and partly upon an appendage of the labellum found in Glossodia, Epi- 
blema, &c., to which Zygostates may now be added. But in the first place 
an argument relating to a question of this kind must lie regarded with 
much suspicion, if founded upon the condition of the parts in a genus not 
belonging to the order, and secondly, I do not see how it is possible to 
affirm with confidence to which division of the perianthium the perfect 
anthers are really opposite in Apostasia; had not so very careful a 
botanist as Dr. Brown asserted that they are opposite the sepals I should 
have described them as being opposite the lateral petals, in which case 
they would correspond with the supposed position of the absent stamina 
in Orchidaceae. With regard to the supposition that the appendage of 
the labellum, so common in many genera, and so remarkable in those 
already mentioned, is a sterile stamen, it may be observed that there is 
no proof of its being of such a nature; and it is quite conceivable 
that it may be a mere superficial dilatation of the labellum, the lamellae 
and varicose veins of which afford frequent evidence of a great tendency 
to the formation of superficial appendages, which I should rather 
regard as analogous to the corona of Silenaceae, or more correctly to 
the bearded crest of Iris. It may be added that the relation born by 
the carpels to the stamens is unfavourable to the supposition of there 
being more than one series of the latter within the so called petals, as 
will be explained hereafter. 
