Mr. J. Miers on the Bignoniacese. 115 



breaks away from them, along the line of their inflexion, close 

 to the sutural margins of the capsule ; so that the seeds remain as 

 if parietally attached to the valves, while the main body of the 

 dissepiment, thus detached from its seminiferous portions, ap- 

 pears flat, smooth, and naked. The fruit of T. pralongum is of 

 the same shape as that of T. albiflorum, only somewhat smaller; 

 the ligneous valves are not quite so thick in substance, and they 

 are covered with a closely adherent, coriaceous, rough epidermis, 

 which can be scraped ofi" the more ligneous shell. A similar 

 coating exists upon the ligneous valves of Adenocalymna, Me- 

 mm-a, and some other genera. In both these species of Tanaecium 

 the valves split down the middle, as in Distictis and Haplolo- 

 phium, and a similar replum detaches itself on each side along 

 the main sutural line of dehiscence. The hilum in T. prcelonyum 

 is considerably curved, and broader, owing to the more thickened 

 margin of the seed on that side, and it lies at a right angle with 

 the axis of the fruit ; in T. albiflorum, on the contrary, the hilum 

 lies parallel with the axis, when the radicle is therefore centri- 

 fugal, while in T. prcelongum it points a little outwards to the 

 base of the fruit. There is, however, another essential difference 

 to be noticed in these two developments : in the one, the seeds, 

 from their mode of attachment, are pressed and twisted half 

 round, in several series on each side beyond the dissepiment ; in 

 the other they are 2-serial, appearing almost uniserial, owing to 

 their different direction and to the greater length and breadth of 

 the hilum. 



Tanaecium. therefore, fully corresponds with the conditions of 

 the Eubignoniea ; and from the peculiar shape of its large hand- 

 some flowers, as well as of its capsule and seeds, it forms one of 

 the most interesting genera of the family. Its place is certainly 

 close to Adenocalymna. The generic diagnosis, in accordance 

 with the facts stated, is therefore remodelled, entirely from my 

 own observations, in the following manner : — 



Tanaecium, Sw. (^n parte). — Calyx tubulosus, margine integro, 

 ciliolato, nervis extus in denticulos 5 excurrentibus. Corolla 

 valde elongata, anguste tubulosa, apicem versus paulo am- 

 pliata et infundibuliformis, limbo expanso 5-fido, laciniis sub- 

 sequalibus, 2 superioribus erectioribus, 3 inferioribus reflexis, 

 aestivatione valde imbricatis. Stamina 4 didynama cum 5*" 

 sterili; filamenta filiformia, subulata, glabra, summum versus 

 tubi inserta, inclusa; antherce 2-lobse, lobis linearibus, sagit- 

 tato-distensis, aut divaricatissimis. Ovarium conico-oblongum, 

 glabrum, disco carnoso cylindrico sulcato insidens, 2-loculare, 

 pluri-ovulatum ; ovula in extremis marginibus dissepimenti 

 utrinque pluriseriata. Stylus filiformis, longitudine tubi co- 



8* 



