38 Mr. J. Miers on the Winteracea. 



them and the ovaries, upon a very short stipitiform g\"nobase. 

 The ovaries, which are variable in number, from five to eight, 

 are always free, nniserial, and erect ; they have no sensible style, 

 the short imibilicated stigma being sessile upon the ventral side 

 of the conical and somewhat gibbous ovan,-, just below its apex, 

 from which point a ridge extends to the base : they are always 

 unilocular, with a single placentiferous line upon the ventral 

 face, corresponding with the ridge just mentioned. Upon the 

 edges of this band are arranged about sixteen ovules, in eight 

 collateral series, extending from the stigma to the base of the 

 cell, each reniform ovule being suspended from its sinus by a 

 short funicle. The ovaries ripen into as many small pear-shaped 

 berries, each containing ten or fewer seeds, closely packed in a 

 thin pulp, which possesses a very aromatic taste and smell. The 

 seeds are densely black, very polished, obtusely rostellated above, 

 swelling below in a somewhat reniform or cochleate shape, with 

 a small concave hilum beneath the summit, to which the short 

 funicle is attached. The outer shell is thin, hard, and brittle, 

 formed entirely of short, transverse, ciystalline cylinders, with- 

 out vessels of any kind : the next tunic beneath this brittle shell 

 is of a spongy texture, and of cellular structure, the cells being 

 filled with coloured aromatic oily matter ; it is covered by a thin 

 pellicular reticulated membrane, inside of which, and adhering 

 intimately to it upon its ventral side, there is seen a very 

 distinct thick cord of some length (a raphe), consisting of a 

 bundle enclosing spiral vessels, which cord extends from the 

 hilar point to a dark spot (the chalaza) situated just below the 

 deep sinus : again, within this coating, there are two very di- 

 stinct reticulated membranes, which, by the medium of a small 

 quantity of intervening glandular matter, become somewhat 

 adherent together ; these integuments are homogeneous in all 

 parts, except where they are thickened at the chalazal disk just 

 mentioned. The enclosed nucleus consists of a mass of tleshy 

 albumen, within which, close to the hdum, is found a very minute 

 embr% o, of a short cylindrical form, rounded obtusely at each 

 end, that directed towards the centre of the albumen basing a 

 very short but distinct cleft, indicating two minute cotyledons. 

 St. Hilaire states that this embr\"0 is entire, and that, after care- 

 ful examination, he could not detect in it any tiace of coty- 

 ledons*. I have invariably found, on the contrar), in the seeds 

 of Drirmjs Chilensis, the verj' distinct cleft above mentioned. 

 St. Hilaire also describes the radicle as protruding beyond the 

 albumen : I have not found this to be the case; although there 

 is an appearance of such an occurrence, when the thin portion 

 of the albumen that covers its extremity breaks away by its 

 * Loc. cit. p. "• 



I 



