114 PROF. T. H. HUXLEY ON THE GENTIANS. 
petals, at first, afforded an imperfect defence against useless 
visitors, and narrowing of the flower ensured the contact of the 
Bees with the stigmas as well as with the anthers (Crosso- 
petalum*). 
* Subsequently, in the one main branch, under the conditions 
already indicated, the Colanthe-form, adapted to Humble-bees, 
was bred into the Cyclostigma-typet, adapted to Butterflies; 
whilst in a particular offshoot of the other main branch, the de- 
velopment of the pectines and the narrowing of the corolla 
around the ovary brought Butterflies as well as Humble-bees into 
service as cross-fertilizers ( Endotricha 1). 
“The original yellow colour of the flowers (as in G. lutea) was 
gradually changed into blue by the selective action of Humble- 
bees; and instructive intermediate stages of this process are 
seen in the species of Celanthe. But after the blue coloration 
was once fixed, the colour maintained itself throughout the pro- 
cess by which Butterflies, by unconscious selection, bred Celanthe 
into Cyclostigma." $ 
I confess that, when I first read this uncompromising applica- 
tion of the theory of flowers, originated by Sprengel and Darwin, 
to the case of the Alpine Gentians, I thought it somewhat rash, 
and gave it a very cool and sceptical reception. It occurred to 
me that if it was a valid explanation of the morphological cha- 
racters of the flowers at all, it ought to be applicable not only 
to those of the Alpine Gentiane, but to the whole genus; and, if 
to the penus, then why not to the whole order? And the chicf 
motive which has led me to spend a good deal of time and trouble 
in working over a considerable proportion of the species of Gen- 
tiana and of the genera of Gentianes, was the obvious importance 
of arriving at some conclusion respecting this question. 
I do not profess to have settled it; but, so far as I have gone, 
I have found nothing that conflicts with Miiller’s extremely in- 
* J ophanthe (mihi). 
t 1r, chanthe (mihi). 
t Stc /ananthe (mihi). 
$ Sir Joseph Hooker (‘Flora of New Zealand ") observes that the pure blue 
ofthe European and Himalayan Gentians is unknown in the Arctic regions 
and the higher latitudes of the Andes. The widespread G. montana has pale 
whitish-yellow flowers streaked with red or blue. These facts are interesting 
in relation to the primitive character of the Novo-Zelanian, Australian, and 
Austro-Columbian Gentians, 
