OF BALANOPHORE^. 59 



their roots intertwine with those of other plants : my specimens imdoubtedly contract 

 broad organic adhesions with the roots they encounter, and in the young state receive 

 woody bundles from them. Richard's admii-able abcount of the epoch of fecundation 

 agrees with what I have observed in the monoecious Balanopliorce of India. The capitula 

 are never self-fertilized ; the styles of the female flowers are protruded immediately after 

 the fall of the scales, and fertilized by the pollen of a neighbom-ing capitulum ; the styles 

 then fall away, and dm-ing the matiu-ation of the fruit, the male flowers are protruded 

 and shed their pollen to fertilize another capitulum ; by the time that the latter opera- 

 tion is performed the fruits have ripened, are shed, and the peduncle and capitulum 

 perish, though the latter still contains an abundant crop of young male flowers, apparently 

 destined never to perform then' functions. This apparent superfluity of male blossom is 

 a very remarkable phajnomenon, and not at all comparable Avith the common one of 

 nimierous male flowers on one inflorescence never becoming perfected except under 

 favourable conditions, for in this case there appears to be a second crop of males after the 

 first have performed their office, and after the females of the same and all the other 

 capitula are fertilized, and it is difficult to conceive any circumstances arising at all likely 

 to call for the operation of these complementary males. 



Martins mentions that a beetle of the family of Curculionidce, or its larva, possibly 

 assists in the fecundation, as it is found nidulating in the capitula ; judging however from 

 the fact of one capitvdum being fecundated by another, the larvae could be of little use, 

 nor can the beetles themselves be of much, under ordinary cu'cumstances. 



Martins mentions delicate thi-ead-like radicles as proceeding sometimes from the base 

 of the seed (embryo) : that author also states that the disposition of the vascular system, 

 both in its nature and arrangement, is monocotyledonous, an error to which I have else- 

 where alluded in my general remarks on the Order. 



Schott and Endlicher (Meletem. p. 8) observe, that in their S. Brasiliensis there are 

 sometimes two and even three cavities in the ovarium, accompanied in the latter case by 

 three styles. I have never seen such an arrangement in any specimens of this species, 

 but indications of it will be shown to occvu." in the lobed young flowers of S. Mexicana. 

 Swartz (Fl. Ind. Occ.) describes the styles as sometimes solitary, probably from one 

 having fallen away, as he did not examine living specimens. 



In the variety y, for which I am indebted to the late P. B. Webb, Esq., the cellular 

 tissue of the periphery consists of vertically elongated and much more delicate utricles, 

 often filled with starch and chlorophyll grains : there is also a slender central column of 

 true cellular pith surrounded by those woody tubes that are often seen to be the only pith 

 of the varieties a. and j3. 



2. Helosis Mexicana, liebmann, Proceedings of the Scandinavian Meeting of Natural- 

 ists, p. 181. 



H. aquaticu, Mutis MSS. in Herb. Hook. 



Hab. Mexico montibus ditionis Vera Cruz et Oajaca, alt. 3-5000 ped., Liebmann (v. ic. pict. a cl. auct.). 

 Mirador {Linden), Jul. Convallibus humidis Novae Granadae ad Melgar {Pur die), Febr. 1846. 



Less variable in form and more so in robust or slender habit than the preceding. The 



I 2 



