14 DR. J. D. HOOKER ON THE STRUCTURE AND 



cophyte, in which they occur in compound spikes or panicles. However simple these 

 Ciipitula appear, they are invarial)ly found to he compound if examined at an early period 

 of oTowth, when the hracts or scales imbricate completely over them, and cover definite 

 masses of flowers, representing branches of the inflorescence. Sarcophyte presents 

 the most perfect inflorescence, and the only one with a fully branched panicle ; it haa 

 general bracts on the main axis below each ramification, but no partial ones. Lophophy- 

 'turn presents the next degree of perfection in inflorescence : each bract is a very highly 

 developed peltate organ, subtendmg a cylindrical branch of the mam axis, which is covered 

 with flowers :— a modification of this arrangement is found in aU the Selosklea, and ia 

 Cynomor'mm, where the bracts are peltate and unbricate in a young state, and either 

 peltate and attached by their margins, or scattered, in the older state. 



In Ombrophjtmn the flowers are whorled round the pedicel of a very complete peltate 

 bract, and in most Balanophorm the female flowers are similarly arranged round a very 

 ruduuentary clavate one. In Thonningia and Langsdorffia the female flowers have no 

 bracts whatever, and the male flowers very rudimentary ones. Mystropetalon, the most 

 liio'hly developed genus in many other respects, has a trifid bract under each flower, and 

 no general bracts on the capitulum. 



Ai'ticulated filaments occur abimdantly over the whole surface of the capitula of most 

 of the HeloskJece, and are probably rudimentary female flowers : their similarity to the 

 paraphyses of Musci has been dwelt upon by Griffith, who (with some other authors) 

 attaches great systematic value to this resemblance. These anomalous organs will be 

 descriljed imder the respective species : analogous ones may be seen in the capitulum of 

 Langsdorffia, and between the male flowers of some Balanophorce. Por further structural 

 particulars respecting the inflorescence, the individual genera must be consulted. 



The periods of inflorescence present some remarkable anomalies in Balanophorte, and 

 especially in the Helosidece with bisexual capitula ; a curious phenomenon, first observed 

 by Richard (fully described by him under H. guyanensis), which necessitates the agency 

 of dichogamy, or the fertilization of the ovaria of one capitulum or plant by the males of 

 another. 



Some genera are constantly dioecious ; as Langsdorffia, Thonningia, Bhopalocnemis, 

 Sarcophyte, Lophophytnm, and some BalanophorcB ; though in B. dioica, which is one 

 of the most constantly so of that genus, I have occasionally found male capitula on 

 the same rhizomes with female ones. The inflorescence is bisexual or moncecious, with 

 the male flowers below, in some BalanophorcB : the male flowers are above in Lophophy- 

 tum and Mystropetalon, and the two sexes are irregularly mixed in Helosidece and Cyno- 

 morium, which latter occasionally presents also hermaphrodite flowers. 



Flowers.- — -These present many gradations of perfection, both in the male and female. 

 They are most fully developed in Mystropetalon, and the least so in the female of Bala- 

 nophora and the male of Lophophytiim. 



The perianth, when present, is almost invariably dimorphous, and most perfect in the 

 male flowers : in those of Lophophytiim it is wholly wanting, or reduced to two opposite 

 mamilliB alternating with the stamina ; in Thonningia it consists of three minute scales 

 which at no period cover the stamens ; in Cynomorium it appears as six linear or clavate 



