Solan ALES.] 



POLEMONIACEiE. 



635 



Order CCXLIII. POLEMONIACEiE.— Phloxworts. 



Polemonia, Juss. Ge7i. 136. (1789).— Polemonideae, DC. andDnhy, 329. (1828).— Polemoniaceae, Lindl. 

 Sytm-s- 168. (1829) ; Bentham in Bot. Rep. 1622; Endl. Gat. cxlv. ; Meisner, p. 273; Alph. DC. 

 Prodr. 8. 302.— Cobaeacese, Don in Edinb. Phil. Joum. 10. 111. (1824) ; Link. Handb. 1. 822. (1829)^ 



Diagnosis, — SolanaZ Fxogens, with 5 free stamens^ axile placentte, and straight plano- 

 convex cotyledons. 



Herbaceous plants, with opposite, or occasionally alternate, compound, or simple 

 leaves ; stem occasionally climbing. Calyx inferior, generally prismatical, S-parted, 

 persistent, sometimes irregular. Corolla re- 

 gvdar, or nearly so, 5-lobed. Stamens 5, 

 inserted into the middle of the tube of the 

 corolla, and alternate with its segments. 

 Ovary superior, 3-celled, ^^^th few or many 

 anatropal or amphitropal ovules ; style sim- 

 ple ; stigma trifid. Capsule 3-celled, 3-valved, 

 few or many-seeded, with a locuMcidal or 

 septicidal dehiscence ; the valves separating 

 from the axis. Seeds angular or oval, or 

 winged, often enveloped in mucus, in which 

 spiral threads are entangled, ascending ; em- 

 bryo sti'aight, in the axis of much horny albu- 

 men ; radicle inferior, very short ; cotyledons 

 eUiptical, foliaceous. 



The ternary division of the ovary connected 

 with the pentandrous corolla and 5-lobed ca- 

 lyx, bring this Order near Bindweeds, from 

 which the habit, embryo, and corolla, distin- 

 guish it, but Cobaea has the habit of a Bind- 

 weed without the leaves ; from Gentianworts, 

 to which it also approaches, the 3-celled ovary 

 divides it. To Hydi'ophyls it approaches very 

 nearly, but the placentation is cUfFerent ; and 

 therefore, Phloxworts are not placed in the 

 Cortusal AlUance, but on the borders of it; to 

 which the large embryo also persuades us. It is remarkable for the blue colour of the 

 pollen, which is usually of that hue, whatever may be the colom' of the corolla. In 

 Collomia linearis I have noticed (in Botanical Register, foho 1166) that the dilatation of 

 the mucous matter in which the seeds are enveloped, and which, when they are tlu-own 

 into water, forms aroimd them Hke a cloud, depends upon the presence of an infinite 

 multitude of exceedingly deUcate and minute spu'al vessels, lying coiled up, spire within 

 spire, on the outside of the testa ; when dry, these vessels are confined upon the surface 

 of the seed by its mucus, without beuag able to manifest themselves ; but the mstant 

 water is apphed, the mucus dissolves and ceases to comiteract the elasticity of the spiral 

 threads, wliich then dart forward at right angles with the testa, each caiTymg with it a 

 sheath of mucus, in which it for a long time remains enveloped as if in a membranous 

 case. This singiilar phenomenon appears to be not uncommon in the Order. Bronnia, 

 one of the plants referred to an imaginary Order, Fouquieraceoe, is very like a Cantua, 

 to which indeed it might have been referred if its calyx were not composed of distmct 

 imbricated sepals. Is it really so ? 



Mr. Bentham observes that Phloxworts had, perhaps, be better placed between Bo- 

 rageworts and Nightshades. They are, however, anomalous among CoroUifloi-al Orders 

 by the constancy (unless in accidentally abnormal flowers) of the tricarpellary ovary. 

 They possess the contorted rcstivation of Dogbanes and Gentianworts, the axile placcn- 

 tation of Nightshades and Figworts, with the inflorescence also, if not the habit, of some 

 groups in the two latter Orders. — London Journal of Botany, 4. 242. I think, however, 

 that the tricarpellary stinictm-e of the ovary forbids om- placmg Phloxworts m any Alh- 



Fig. CCCCXXVI. 



Fig. CCCCXXVI. 

 vessel. 



-Collomia gracilis. 1. caly.x and corolla ; 2. pistU ; 3. cross section of a ripe seed- 



