PRIMULA. 



that catalogues, however, try to call P. viscosa (on Villars' most 

 unfortunate name, which was, however, only bestowed in 1787, and 

 therefore is incapable of standing against Allioni's prior description of 

 two years before). So much for P. hirsuta : the question being per- 

 fectly simple, as will be seen, and by no means involved. We now 

 turn to the true P. viscosa, where again Allioni steps in first, by 

 applying this name to the big stinking-leaved glandular Primula 

 of the Southern granites — lax in texture, running woodily about in 

 wide masses, sending up luxuriant flopping tufts of very long oval- 

 pointed foliage, sometimes toothed, and of a dusky grey-green with 

 glands that smell rancidly of goat ; to be followed by tall stems of 

 10 inches or more, carrying serried and one-sided bunches of hanging 

 narrow trumpet-shaped flowers usually of a hot and vinous violet, 

 rather squeezed and cylindrical in outline, large indeed in themselves, 

 yet not by comparison with the lush and leafy plant. But then, Allioni 

 having got in first with his name, Villars came leaping in as before, two 

 years later, and once more exactly reversed matters, making his P. 

 hirsuta out of Allioni's P. viscosa, just as he had turned Allioni's P. 

 hirsuta into a new P. viscosa of his own (and Allioni's Campanula 

 alpestris into a complimentary but untenable C. Allionii). However, 

 the priority of date settles the matter finally and quite simply in favour 

 of Allioni's names of 1785, as against those so mischievous transposi- 

 tions attempted by Villars in 1787. There can be no further room 

 for doubt or error. Nor need we trouble about the catalogue-names 

 of the species, except as mere varieties. P. graveolens (Hegetschw.) 

 and P. latifolia (Lapeyr.) are invalid synonyms of P. viscosa, All., 

 which name has the prior right over all. The type, however, occupies 

 a very large range, from the Pyrenees through all the Southern 

 granites (never on limestone) to the Voraarlberg ; accordingly it takes 

 various local developments, of which three are fairly well marked, 

 though by no means to be insisted upon, since each district will 

 yield many shades of difference, and many a valley may be found 

 with special forms as beautiful as that rarity, for instance, of broad- 

 toothed leaves and round powdered eye to the flowers of pure blue- 

 violet, that is so precious by certain rocks of the Mont Cenis among 

 abundance of an uglier type. However, as some of these try to 

 figure as species sometimes in catalogues, it may be well to point 

 out the three main streams of variation as at present recognised — all 

 forms of the species being equally ready to grow, when once re- 

 established in warm deep and perfectly-drained light peaty and stony 

 soil, especially among roots or under the edges of granitic rocks, so that 

 they can send their fibres far in beneath into the dry shelter they 



195 



