REPORT OF WORK IN 1915 IN KANSU AND TIBET. 341 
fading to grey by degrees from the ten-lobed vague white blur that 
radiates from the intense clarety black eye of the tube's mouth. It 
is always a striking, strange, and artificial-looking plant, local and 
scattered, never widespread, but sometimes occurring in impregnable 
colonies up on the cliff s (especially when calcareous), and often solitary 
in crevices that look incapable of containing it ; but invariably in 
corners that never get the sun, in cool, dank, moist vegetable soil, 
lodged in the chinks. From experience, though, I judge P. Farreri to 
have a strong and hearty temper, and hope good things of it in careful 
cultivation, if only the none too abundant seed will condescend to 
germinate and thrive. Meanwhile we have speaking likenesses, both 
plain and coloured. 
P. gemmifera is F 562, 563, 168 ! 121 ! In other words, it swallows 
up everything hitherto known as P. acclamata sp. nova. Little blame 
either to Professor Balfour or to me, seeing that the classic descrip- 
tion of P. gemmifera not only assigns it to a wildly impossible place in 
the family, but also, apart from other misleading details, declares the 
plant to be annual and produce bulbils. How should one, in this, 
recognize a Primula that is certainly perennial, and never produces 
bulbils of any sort ? However, the awful conclusion leapt on me, 
when in Petrograd I went through the Chinese collections, and under 
P. gemmifera found an old friend that I should have greatly preferred 
to meet in some other context. And at the same time Professor 
Balfour, in Edinburgh, was arriving at the same sad conclusion. So 
back into P. gemmifera go all my forms of P. acclamata. In any 
case, my own P. gemmifera, dug out from beneath this accumulation 
of errors, is no less lovely a thing than when we proclaimed it as 
P. acclamata ; infinitely variable, indeed, but never in the least like 
the P. gemmifera of Pax. It seems to have an enormous range, and 
in the Da-Tung abounds throughout the alpine zone, high up on the 
alps being specially stalwart and wholly powderless, while down in 
the beck-beds of Wolvesden it develops a slenderer form with very 
prognathous flowers and the scapes all white with meal. Of both 
I have paintings and photographs ; as of the minor form by the track- 
side on Thundercrown in 1914, and the Min S'an type of the same year. 
And, whatever you may call it, P. gemmifera is among the supreme 
treasures of the family, rivalled only by P. stenocalyx in the group of 
P. farinosa. 
P. Reginella (F 561) is the other novelty of 191 5. And yet no 
other Primula has been more often collected and mixed up with others. 
It was left for my specimens at last to determine this as a real new 
species, after it had for years been reduced to a subordinate part in 
others of the race. It is a great deal of the true P. Pumilio, on that 
plant's unique original sheet at Petrograd ; it has also been much of 
P. tibetica, of P. diantha, and even of P. sibirica. In point of fact, P. 
Reginella stands quite apart, and is indeed a Little Queen of loveliness 
in the Auriculata group, a tiny compressed wee thing, in strongest 
possible contrast to my other novelty, the gross and gorgeous P. 
