PROF. T. H. HUXLEY ON THE GENTIANS. 105 
they have been described as appendages of the filaments of the 
stamens. I have looked into this point with considerable care; 
and, in all cases, I find that the fimbrie cohere with the corolla 
and not with the filaments, if sufficient care is taken in sepa- 
rating the two. If the filaments are roughly torn away, on the 
other hand, some of the fimbrie may come with them, attached 
by flakes of the corolline epidermis. In orderto remove all pos- 
sible doubt, it would of course be necessary to study the develop- 
ment of the fimbrie. 
In other exemplifications of the Lophanthe type the fimbris are 
perinectarial ; that is to say, they are developed, not in longitu- 
dinal series parallel with and close to the stemono-petaline sym- 
physes, but in close relation with the margins of the nectarial 
arem. 
The species in which I have met with the Lophanthe modifica- 
lion are :— 
1. With parastemonal fimbrie. 
A. B. D. E. 
Gentiana Gentiana Gentiana Gentiana 
7 ciliata. “foliosa. montana. montana. 
crinita. rupicola. diemensis. concinna. 
barbatula, Hookeri. pleurogynoides. 
fastigiata. 
cerastioides. 
cernua. 
Jamesonii. 
radicata. 
limoselloides. 
diffusa. 
2. With perinectarial fimbrie. 
Swertia (Ophelia). 
t rasera. 
The fimbriæ of Swertia and Frasera are well known. In the 
Gentiane proper the only distinct account of them which I have 
met with is that given by Müller (‘Alpenblumen,’ P. 343, G. 
ciliata). The disposition of the fimbriæ in the various species in 
which they occur is well worthy of a more careful and detailed 
study than it has yet received. In some species (e. g. Gentiana 
montana, pleurogynoides, and rupicola) the parastemonal fimbriæ 
are so minute as to suggest that they may be absent in some 
varieties, which would thus effect a transition between Lophanthe 
and Actinanthe. Sir J. Hooker (‘Flora of New Zealand’) has 
