ON A MONANDKOUS CYPRIPEDIUM. 3 



terminal flower of C. candidum, L,, which had no labenum, but 

 two sterile ' shield' stamens, and two fertile stamens oi^posite the 

 petals, and therefore normal in position. Unfortmiately the two 

 lateral sepals are, in this note, considered as one ; and as it is said 

 that the sterile stamens were oj)posite the sepals, it seems scarcely- 

 possible to conclude otherwise than that they represent the organs 

 marked A^ and a^ in the diagram. The surprise which is 

 naturally felt at the appearance of the usually-absent a^ is lessened 

 by the fact of the absence of the labellum,-'' and by the existence 

 of the former organ in the closely allied genus, if not monstrous 

 form known as Uropedium. Then Masters! figures a monster 

 which seems to have been modified in a somewhat similar way to 

 ours. In this the lateral sepals are wanting, and the central one 

 is divided into two ; the labellum is quite normal, except for a 

 slight lateral disarrangement ; the petals are placed in a median 

 or nearly median position, and the androecium is regular, except 

 that the shield is suppressed. The same author says : — " A tetran- 

 drous flower of Cypripedium has also been recorded." This I 

 presume to refer to Asa Gray's case above-mentioned, though 

 possibly I may be mistaken. To these must be added Uropedium, \ 

 which has a flat petal-like labellum, three complete stamens 

 opposite the petals, and therefore in the position of a\ a^ and 

 a^^ as well as a median sterile one (Ai ), something like the Cypri- 

 pedium ' shield,' but fi-ee from the style, and united to the lateral 

 stamens. We see then that in Cyprijjediece every stamen may be 

 antheriferous with the exception of A^ and A^, which by 

 adherents to the Brown-Lindley-Darwin morphology are supposed 

 to be united with the labellum. On the other hand, if we turn to 

 Monandi'ese, we find that in Pogonia ophioglossoides all the stamens 

 have been seen, and in the well-known case of Arimdina pentandra, 

 figured by Eeichenbach in ' Xenia Orchidacea,' t. 105, all with the 

 exception of a^ . Many instances have also been recorded of 

 diandrous and triandrous monsters in several other genera. § 



Whether we incline to the conclusion of Brown, Lindley,|| 

 Darwin, and their followers, according to which the position of the 



* May not the labellum have been present in the form of the sterile stamen ? 

 Cases of pollen being borne by petals in Orchideai are on record. Perhaps an 

 intermediate condition may be that of a flat labellum, mentioned by Reichenbach 

 as occurring in the case of Selenipedium Warsczeioiczii. 



+ * Vegetable Teratology,' p. 93, fig. 44. 



+ Brongniart, 'Ann. Sc. Nat.' III. Ser., Botanique, vol. xiii., p.ll3, tab. 2. The 

 question as to the monstrous condition or generic validity of this form has been 

 answered by Reichenbach (' Bot. Zeituug,' 187G, p. 41) in the latter sense. This 

 conclusion is founded on two facts : first, that it bears seed capable of repro- 

 duction ; and secondly, that for the most part Uropedium and Selenipedium 

 inbabit diff'erent countries, and that where they are compatriots they are not 

 neighbours. It matters little to my present purpose which view be adoi^ted ; it 

 ought, however, to be said that Reichenbach speaks of having seen a tripetalous 

 flower of S. Warsczewiczii. 



§ See Masters, I.e. p. 380, for a number of these, 



II Every appeal to theoretical structure indicates, in my opinion, a belief in 

 Evolution so far as relates to the diff'erentiations from that structure, so that 

 we may fairly claim tbe two greatest uinetc-enUi -century English botanists as 



