APPENDIX. 



to my taste, a reversal of propriety. But under any name, no doubt, these 

 new Primulas will smell as sweet. As soon as we have got them, that is ; 

 while many are now fluttering into our reach, yet many more still exist only 

 as dry and often fragmentary specimens in a Herbarium, shadowy foundations 

 of new species that may ultimately be found impossible to maintain. 



P. acclamata. See my Chinese Appendix. 



Primula aemula is a very splendid species from open mountain pastures of 

 Yunnan ; and possibly, therefore, sharing in the " softness " of constitution 

 which is the one bane of the beautiful treasures that hail from such southerly 

 regions, when translated to the cooler, lachrymose conditions of England. 

 P. aemula stands nearest to P. rejlexa and P. orbicularis, and is a stout and stately 

 splendour, with several remote tiers of brilliant satin-sheening golden stars up 

 its tall stems, after the general style of P. japonica and P. helodoxa (though 

 in reality this species belongs to the cousinship of P. Maximowiczii, but lacking 

 the reflexed petals of P. szechuanica and P. rejlexa in that group). The basal 

 rosette is large and ample, with fleshy narrow-oblong foliage of a foot in length 

 or so, minutely toothed at the edge, and absolutely without powdering or down 

 of any kind. 



P. alsophila. See my Chinese Appendix. 



P. alia represents P. denticulata on the hills of Yunnan, and unfolds its 

 downy foliage, together with its heads of lavender-purple blossom, in February, 

 attaining a foot or more in height. 



P. annulata is a delicate little thing only existing hitherto in dried specimens. 

 It has a look of P. yunnansis, and yet more of P. bella; but from the first is 

 distinguished by a ring round the mouth of the flower, and from P. bella by 

 the lack of whiskers in its throat. 



P. Calderiana is yet another of the Primulas long sunk in the welter of 

 P. obtusifoliu. It is perhaps at best only a microform of P. Gammieana, with 

 fewer smaller flowers, with dark-purple calyx, and the corolla not velvety outside. 

 It is common on peaty hill-tops in Sikkim, and its purple blossoms are very 

 variable in colour. 



P. celsiaeformis exactly describes itself. It is a near relation of P. blattari- 

 formis in the rather coarse group of P. malvacea, and is just like a purple- 

 flowered Mullein with long-stalked foliage. But the variety Duclouxii of P. 

 blattariformis has now been promoted to specific rank as P. Tenana, q.v. 



P. cephalantha stands very close in prettiness, difficulty and impermanence 

 to P. pinnatifida, but quite distinct in the Muscarioid Group by its special 

 hairiness, which prevents it taking on the bright green look of P. pinnatifida. 

 Its flower-head, too, is more of a spike, and yellow meal develops, as in P. aerin- 

 antlia, on the 10-inch scape and the seed-head as it ripens. 



P. chionantha takes us back into the clan of P. nivalis, and is one of the grandest 

 in that grand group, a superb tall plant, hairless and yellow-powdered, with 

 tier over tier of large pure-white stars, ravishingly fragrant. It can attain 

 almost a foot and a half on the open alpine meadows of Western Yunnan, k 

 very eagerly to be desired, and thrives handsomely in rich soil. 



(i,oo6) 481 ii.— 2 h 



