Capsule shaped like the germen, of one cell, and three 

 valves, splitting between the fixed ribs, generally at both 

 sides of each valve; rarely pulpy internally; receptacles 

 three, attached longitudinally to the centre of each valve. 

 Seeds very numerous, minute, roundish, each enveloped in 

 a large, loose, membranous tunic^ wanting in the exotic 

 Vanilla only, where they are imbedded in pulp; albumen 

 the shape of the seed ; embrijo minute, simple, central, 

 near the scar. 

 Our British Oixhidea: are all herbaceous; with either tu- 

 berous or stoutly fibrous roots, seldom if ever parasiti- 

 cal. Their stem is simple, naked, or partly leafy; leaves 

 simple, undivided, entire, smooth, and sleek; sheathing 

 or sessile ; Jlouiers in a simple spike or cluster, which is 

 rarely downy, glandular, or viscid ; each flower accom- 

 panied, at the base of its partial stalk if there be any, by 

 a solitary bractea. Some are highly fragrant, particu- 

 larly in an evening, and many are very beautiful as well 

 as singular. The tuberous roots abound in glutinous mat- 

 ter, and are generally esteemed nourishing and stimula- 

 ting, making the Salep of the Turks. We must leave to 

 conjecture the origin of this opinion. 

 The integuments of \\\e Jlo'wer in this family have never 

 been understood as above described, by any preceding 

 writer. Linnaeus considered the whole as a corolla, con- 

 sisting of 5 petals and a nectary. Swartz calls the former 

 a calyx of 5 leaves, and the latter a corolla, of a single 

 petal, or lip. Jussieu and the botanists of his school, 

 denying the existence of a corolla in any monocotyledo- 

 nous plant, and rejecting the Linnaean term nectarixun, 

 call the whole a calyx of six divisions. But if a corolla be 

 allowed to exist in nature at all, it surely cannot be de- 

 nied to these plants ; any more than to the Scitaminecs, 

 far better known at present than when Jussieu's immor- 

 tal work appeared in 1789. These evidently settle the 

 question. See Grammar 19. f. 1. If we will not allow 

 them to have a corolla, we may, with equal propriety, 

 deny the existence of such an organ in all herbs, or in 

 all trees, or in any one natural order according to our 

 whim or fancy. The OrcliidecB, righdy understood, will 

 be found as conclusive on the same side as the Scitami- 

 nece, and they may possibly afford additional evidence of 

 a corolla in monocotyledonous genera. 

 Perhaps the most ingenious and important remark, in phy- 



