OF BALANOPHORE^. 37 



III. Sarcophyte, Sparrmann. 



(Tab. I. C.) 



1. Sarcophyte sanguinea, Sparrm. in Act. Holm, xxxvii. p. 300. t. 7 ; Scliott & Endl. 

 Melet. Bot. p. 11 ; Grifiitli, Linn. Soc. Trans, xix. p. 339 ; linger, Ann. Wien. Mus. 

 ii. t. 5. f. 28 ; WeddeU in Ann. Sc. Nat. s4r. 3. p. 14. 1. 10. f. 34-38. 



Ichthyosma Wehdemanni, Schlecht. in Linnfea, ii. 671. t. 8. 



Hab. Africa Australi, ad ruAicss Ekeba-gi(e prope Grahams-town (?FeArfeOTaMW,&c.). Ad radices .^c«c/« 

 Capensis, Quagga's Flat, Uitenhage {Zeyher !). 



My observations on this remarkable plant chiefly refer to two points : the structure of 

 the anther, and the relation of the genus to other Balanophoreee, in both which I differ 

 from Mr. Griffith. 



The anther is rightly described by Endlicher as consisting of a solid capitate body, 

 containing many loctili filled with pollen. The contracted persistent sej^ta between these 

 loculi have been mistaken by Griffith for pedicelled anthers, which he describes as forming 

 together a " caput antherarum," crowning a common peduncle, which rises from the axil 

 of a bract. On the contrary, I fhid that the anther* contains about fifteen to twenty cells 

 of very variable size, radiating from a cellular axis : a transverse section shows about twelve 

 such cells symmetrically disposed round the apex of the filament; and a vertical one 

 exhibits about eight, which radiate from the blunt stimmit of the tilament, and of which 

 the outermost are very small. At a very early period the septa consist of three tissues : 

 an inner cellular not materially different from that of the filanient, from which, however, 

 it is separated by a broad dark line ; this is lined by a delicate hyaline endothecial coat, 

 and upon this is a mass of matted filaments of excessive tenuity and pulpy natiu'e pre- 

 cisely similar to the anther-Uning of Phyllocoryne, &c., amongst which the poUen-grains 

 nestle. Each cell is distended with spherical pollen-grains. 



I have not observed the circumscissal dehiscence of the outer membrane described by 

 Endlicher, and which was apparently suggested to that author by the appearance of an 

 annulus at the base of the dehisced anther ; as on the contrary the anther dehisces by 

 the disruption of the membrane over each loculus, in a manner quite analogous to that of 

 the apices of the anthers of Hhopalocneniis, JSelosis, and CoryncBa, on the one hand, and 

 of Balanophora polyandra, &c., on the other ; and is an instance of the general tendency 

 to dehiscence by irregular disruption of the anther-wall, which prevails throughout the 

 Order. I have not been able at any period of its growth to reduce the anther to the 

 ordinary 4-locular type; the pollen being developed, as in Viscum and various Hhizo- 

 phor(B, simultaneously in many independent points of the epithehum : that these points 

 originated along detuiite lines, answering to the position of anther-ceUs of the ordinaiy 

 tyjje, can therefore only be assumed. 



Erom the above it is evident that Griffith is perfectly correct in insisting {I. c. p. 339) 

 that a continuous solid tissue must exist between the cells of the anther, if it be assumed 

 that these cells are not separate anthers. 



In assuming that the filaments of Sarcophyte are axillary to the lobes of the perianth 



* A correct section of the anther is given in Unger's paper ; Ann. Wien. Mus. /. c. t. 7. fig. 48. 



