IN FUCHSIA GLOBOSA. 431 
Regarded by itself, this hypothesis does not lack plausibility, 
since researches by Dr. S. Stadler* have proved that honey- 
glands are always connected with vascular bundles, whether 
these are specially adapted to the honey-gland or modify their 
course in their behalf, or, finally, the honey-glands happen to be 
placed just where vascular bundles abound. According to Stadler, 
in (Enothera, an Onagrariad that is in not a few respects like 
Fuchsia, there exist vascular bundles exclusively destined for 
the supply of the nectary, though in a limited number. It is 
therefore by no means impossible that in our case the nectary 
supplied by vascular bundles attained a degree of development so 
as to form the tube in question. 
On the other hand, it must be observed that the nectary, 
though irregular in shape, sometimes (especially when the ovary 
is superior) never tends to change into a petal, sepal, or stamen, 
so far as our observations go; it is therefore not probable that 
it would, as it were, all of a sudden develop into a tube with 
sepals, petals, and stamens. Besides, it is not known whether 
or not the honey-gland is actually absent; if the nectary were 
present, the supposition of its metamorphosis would at once lose 
all ground. 
In the third place, the supposition that the flower with the long 
tube is the primary one, and the parts beneath are of secondary 
origin is much more in accordance with the flowers of Liebe, 
Masters, and so on. 
I beg leave to subjoin a few more remarks based on observa- 
tions by Mr. H. W. Heinsius, Assistant in the Phytophysiological 
Laboratory in Amsterdam. Mr. Heinsius has obligingly lent me 
his notes and sketches of monstrous Fuchsias, from which I deduce 
the following conclusions :— 
1. Often two sepals have become united in a very broad whole, 
the composition of this whole is always proved by two cireum- 
stances: (a) the presence of two ribs, (b) the occurrence of two 
stamens opposite to it. 
2. That in one flower two typical numbers are possible; the 
antisepalous (inner) stamens are then in accordance with the 
number of sepals, the antipetalous ones with that of the petals. 
This fully agrees with our statements in § 7 on this point. The 
way in which this difference in number in one flower comes about 
cannot as yet be satisfactorily determined. There may be 
* 'Beitráge zur Kenntniss der Nectarien und Biologie der Blüthen, von 
Dr. S. Stadler (Berlin, 1886). 
