in the Order of Cucurbitacece. 281 



Each receptaculum in these fruits consists (Fig. 1. c) of two 

 half membranes, uniting at the centre of the fruit ; so that each 

 membrane, embracing the inner end of a septum in its central 

 fold, has one half in one cell, and the other half in another cell. 

 As these receptacula, consisting thus of a double membrane, are 

 usually thicker than the real septa, and continue longer conspi- 

 cuous, being less surrounded by the pulpy or fleshy nature of the 

 fruit, they have often been considered as the true septa dividing 

 the fruit into three cells, having entirely that appearance, as in 

 Fig. 3. Still, however, when fully ripe, they often disappear, as 

 even in the fruit now mentioned, the two rows of seeds that ad- 

 hered to each septum in the half ripe fruit, when this ripens are 

 lodged in a cavity formed by a dry fibrous texture, which suc- 

 ceeds to the fleshy substance secreted from the true septa (Fig. 5.) 

 In other cases, as in the Momordica Balsamina (GARTNER, t. 32.), 

 all vestiges of both septa and receptacula disappear, and the loose 

 seeds remain floating in a gelatinous fluid, which squirts out when 

 the fruit bursts. 



Sometimes the receptacula do not extend to the parietes, but 

 terminate in the middle of the cell, in which case they sometimes 

 have only a single row of seeds annexed to their margin, as in 

 the species of Luffa called Ghoza by the Bengalese (Fig. 6.), 

 sometimes a double row (as in Fig. 3.) ; and rarely only one seed 

 of each row comes to maturity, as in several species of Bryonia, 

 and in the Trichosanthes called Theba by the Bengalese (Fig. 4.) 



In general, however, the receptacula reach to the outer pa- 

 rietes, to which they adhere ; and as the portion between the 

 centre and parietes often disappears before the fruit is ripe, what 

 remains adhering to the parietes is considered as a parietal re- 

 ceptaculum, and the seeds are considered as centrifugal (GJERTN. 

 torn. i. Praef. xlvii.), directly contrary to fact. This appearance 

 is very plain in the ripe fruit of the Trichosanthes anguina, or 

 Chichingya of the Bengalese (Fig. 7.), which, when ripe, is per- 



