APPENDIX. 



distinct in the gardener's eye (especially as he possesses none of them), though 

 very much so in the botanist's. 



P. Harrissii stands in the very shadow of P. rosea, but is smaller, and with 

 leaves and flowers unfolding together. 



P. helvenacea flaunts its tall and stately stems of thin-textured violet flowers 

 in the cliffs of the Mekong-Salween Divide, and is allied to gorgeous P. calliantha, 

 though distinguished at once by its long-stalked small-bladed leaves. 



P. indo-bella explains itself. It replaces P. bella in Bhutan, where it forms 

 dense cushions of inter-ramifying stolens and wee frail rosettes, close over which 

 are constellated the fat and fleshy stars of blue blossom. 



P. leimonophila loves the pastures in Hunan, at Yo-jo, but has not yet come 

 into ours. It is a remarkably lovely plant, in the general clan of PP. Gageana, 

 Kingii, argutidens, and amethystina. It swings out a one-sided head of droop- 

 ing bells in a rich tone of Prussian blue. 



P. lhasaensis may be no more than a larger form of P. Jaffrayana, q.v. 



P. meiantha is actually annual or monocarpic. It is a small-flowered Bur- 

 mese plant in the Malacoeides Group, but its annual habit and squinny minute 

 blossoms sink it below further consideration. 



P. melichlora is known only from a set of specimens, and is a cushiony, woody- 

 rooted yellow-mealed cousin to P. spathulifolia. 



P. minor is a charming tufted alpine, which may best be described as a 

 miniature of unparalleled P. pulchella, but with leaves mealy on the upper 

 surface, and rounded undivided lobes to the round-faced lovely lavender flowers, 

 borne few in a head. It loves open and rather dry exposures on the mountain 

 slopes of Adundz' and under the rocks. 



P. nemoralis has violet flowers and soft leaves in the forests of Yunnan, 

 and stands near P. sinnata in the chain of P. sonchifolia. 



P. oresbia, apart from its dense vesture of white meal on the underside of 

 the leaves and on the inflorescence, might almost be P. incisa. And to the 

 aggregate of P. incisa it more certainly belongs than as yet to any earthly garden. 

 It seems to be very pretty, but specimens and descriptions of P. incisa itself 

 are not satisfactory or unanimous. The group links on the Soldanelloids with 

 the Bellae. 



P. petrophyes, though with structural differences, may by us be taken (if 

 only we could get it) as virtually a diminutive and alpine state of P. leimono- 

 phila, q.v. 



P. philoresia is a little matted loveliness from near Adundz' (A-tuntzu), and 

 is so exactly like a smaller version of P. dryadifolia that it might be taken for 

 a mere variety, were it not that its wee leaves are hairy above, while each flower, 

 long-tubed from its calyx, stares straight up from its nest in the mat of foliage, 

 instead of arising several on a stem. 



P. prionotes only exists in specimens for us, and is so far the only Indian 

 Sikkimensis-cousin with purple flowers. 



P. pseudo-malacoeides explains at once its uses and affinities. It is slenderer 

 and smaller in flower than the type. 



483 



