APPENDIX. 



enveloping mat of grass and rootage equalising their moisture in summer, and 

 draining it uniformly away ; while in winter it dies down upon their dormant 

 crowns like a dry thatch, over which springy mattress lies the warm coverlid 

 of the winter's snow. I should, indeed, make an Alpenwiese on a raking but 

 well-watered slope, for nearly all the Nivalis Group, and especially for the forms 

 of P. nivalis itself. P. No. 8 (P. Woodwardii) is a joy to collect, with stalwart 

 oval pods of hearty brown, standing starkly up from the moorland on length- 

 ened scapes of a foot and more ; two lots have been sent, as 116a and 116b, 

 on the chance that the Monk Mountain form may perhaps prove in some way 

 different from that of Thundercrown. P. Woodwardii, however, despite fore- 

 bodings, grows very easily and vigorously and permanently with us, even 

 developing a white eye, which greatly enhances the beauty of the violet 

 flowers. It is one of the best Primulas from China. 



Primula Sp. No. 9 (F 121) is a unique occurrence, which yielded no seed, 

 and of which I have one sod here in Lanchow which may possibly get no further. 

 We found it only in the little mountain track ascending Thundercrown, between 

 9000 and 9500 feet, where, in clammy, limy loam it grew in wads and clusters 

 like seedling boxesful of groundsel (and by no means, in their earlier stages, 

 unlike). The majority seemed packed seedlings ; only here and there arose 

 the delicate 5-inch scapes in June, bearing flowers intermediate in appear- 

 ance between P. longiflora and P. farinosa, but much nearer the former, round- 

 faced, purple-tubed prettinesses of soft pink, above the minute leathern-grey 

 foliage huddled on the ground. Abounding as it does in its limited area, it 

 must seed and germinate copiously ; but not perhaps every season, since in 

 1914 not the trace of a capsule was anywhere discoverable. It is a dainty, 

 pleasant thing, with its remarkable long-tubed flowers swinging horizontal, 

 usually in pairs ; I suspect it of being very close indeed to F 168, from higher 

 up the mountain — and perhaps a mere microform. But F 168 is a larger, finer 

 plant in every way, with bigger, rounder flowers of milky pink ; it does not 

 grow in mats but in isolated crowns, and the shorter corolla-tubes are yellow 

 and not purple. (This is a minor form of P. gemmifera.) 



Primula Sp. No. 10 (F 122) is a most important and beautiful species of 

 the Nivalis Group, which, however, instead of a long, perilous neck with a few 

 roots at the end, breaks straight, in hearty crowns, from such a mat of stout 

 red fibres, ramified into such a mesh of white rootlets, that you can weed it up 

 in big sods like groundsel from the slopes of bare fine silt where it lives,' 

 between 12,000 and 14,000 feet on Thundercrown, occasionally flaunting from 

 the cliffs in big aged masses, but usually dotted about all by itself, over the 

 otherwise bare earth-pans, beck-shingles, and loamy patches of scree beneath 

 the crests, which it illuminates with its stout-pedicelled, domed (and often 

 2-tiered) heads of big lavender-blue stars in June, on stout powder-white 

 scapes of 3 to 10 inches, rapidly lengthening in flower and fruit. Its pods 

 are very long, straight, narrow-drainpipe-shaped, flat-ended, and pallid in 

 colour, going transparent at the top, as in P. Maximoxviczii ; the lovely flowers 

 have a strong scent of an old apple-cupboard haunted by mice. It should 



515 



