SCHWEINITZIA ODORATA. 17 



Several such simple stems spring in a cluster, from a coralloid- 

 fibrous and matted root, to the height of two to four inches. They 

 are purplish in color, and are thicklj beset with the rather fleshy 

 brownish scales which take the place of foliage ; these are ovate, 

 acute, one-nerved, spirally alternate, about three lines in length ; 

 the upper becoming rather larger and more crowded, forming the 

 bracts of the spike, and partly enveloping the blossoms. The 

 flowers, usually six or eight in number, are borne on very short 

 pedicels, and are subtended by a pair of opposite bracteoles, which 

 resemble the bracts, and are intermediate in size and shape between 

 them and the sepals. The calyx consists of five imbricative sepals, 

 as long as the corolla ; these are scarious in texture in the dried 

 state, ovate-oblong or lanceolate-oblong, acute or acutish, more 

 or less concave, and slightly gibbous at the very base. The co- 

 rolla is about one fourth of an inch long, thickish, of a firm fleshy 

 texture, imbricated in aestivation, and with five rather strong gib- 

 bosities at the base, corresponding with the lobes. The stamens 

 are a little shorter than the corolla ; the anthers are nearly as broad 

 as long ; the two short saccate cells are somewhat enlarged down- 

 wards, and are united by their contiguous faces without any con- 

 nective ; they are attached to the filament by a point at or near 

 their summit on the outside, so that they are introrse. They are 

 not retroverted before anthesis, like those of Pyrola,* but are turned 

 inwards from the first. I notice, however, that, in the young 



* By almost every writer, from Wahlenberg and Don to Koch, De Candolle, and 

 Endlicher, the anthers of Pyrola are said to open by basal pores, and to be inverted. 

 during flowering. The pores, are, however, really apical, as in Ericacea? proper ; 

 the anthers are retroverted in the flower-bud, as is common in the order, and re- 

 sume the truly normal position soon after the flower opens. The correct view was 

 adopted by Dr. Torrey, in his Flora of the Northern and Middle States, p. 432. 

 3 



