CAMPANULA. 
These you must get for yourself. An hour in any alpine hayfield in 
August will bring you in half a dozen things for whose match in beauty 
you might easily range a Vincent Square Show in vain. And, finally, 
a “novelty ” sent out recently as Little Gem is surely nothing else 
but one of these forms of C. rhomboidalis, but not a favourable sample, 
nor deserving its distinction, being neither little, nor a gem, but a 
rather leafy C. rhomboidalis of sombre colour, differing chiefly from 
the type in a slightly dwarfer habit and yellowish-green foliage. 
C. Riverslea. See under C. carpatica. 
C. Rosani (Ten.)=C. versicolor, q.v. 
C. rotundifolia is best considered as a huge aggregate, of which 
our common Harebell, that specially bears that name, is only one. 
Every country and almost every alp produces a slightly different- 
seeming Harebell, and on the most subtle and minute and evanescent 
peculiarities of each have botanists built up an imposing list of species ; 
which, however, all tend to fade into Rotundifolia again after a year 
or two of cultivation, losing that precise hairiness of calyx or de- 
flexion of lobe, on the strength of which they had been promoted to 
specific rank, and published to the world as, let us say, C. Marchesettit, 
thus cruelly to delude innocent purchasers who, under that name, had 
thought to be buying some new and thrillmg Campanula. These, 
then, are the sub-species into which C. rotundifolia offers itself to pur- 
chasers by a dozen specifics, all meaning pretty much the same thing ; 
though some, indeed, of the forms are well-marked and valuable, 
C. rotundifolia of the North is always lovely with its pale-blue flowers 
loose and abundant; it has a beautiful white form, a ragged-robin 
development like a blue whirlwind called C. r. soldanellaeflora, a 
double pale-blue rosette which is paradoxically pretty, and the rather 
ambitiously named C. r. Warleyi, which appears to be another semi- 
double form or garden hybrid of extra-free growth. Then follow a 
crowd of dim and unmarked plants whose claim to recognition are of 
the slightest, as indeed is most properly conceded by C. inconcessa, 
if not by C. Marchesettii, C. solstitialis, and C. praesignis, all depend- 
ing upon some slight differentiation that the change from one side of 
the hill to the other would probably wipe away. More marked, though 
not easily distinguishable in the eye of the garden, are C. carnica, 
C. pseudo-carnica, C. Langsdorffii, C. arcuata, and C. linifolia—these 
being forms replacing typical C. rotund:folia in the Alps—handsome 
ordinary Harebells, but with flowers of much more violet tone than 
ours, and varying widely alike in size of bloom and depth of colour. 
There are pure whites, and delicate silver-pales, and from one pinch 
of seed a dozen different nameable forms. The thing called Oam- 
194 
