THE FERTILISATION OF SALVIA AND OTHER FLOWERS. 263 
introducing a pencil into the mouth of the flower, but it is a far 
more interesting sight to watch it when brought about by the 
action of the bee. It can be seen with the greatest ease, and 
no one who has once seen it will doubt that the peculiar form 
and arrangement of the stamens is not an accidental and in- 
different one, but stands in direct connection with the visits of 
the bee to the nectary ; and that the curious modifications in 
the structure of the whole flower have occurred in order to 
ensure the adherence of the pollen to the back and sides of the 
bee (fig. 7). 
It will be noticed that the lower anther cells, those against 
which the head of the bee strikes, are sterile or nearly so. This 
is an instance of the apparent occasional economy of nature. 
It would be, as will be seen presently, of little or no use that 
the bee should have pollen on its head. None, therefore, or 
little, is produced by the cells against which the head impinges ; 
and the economy thus practised is very probably one of the 
conditions which favour the abundant production of pollen in 
the upper anther cells. In these it can be of use, and thus 
the material saved from the low^er cells is expended here to 
greater advantage.* 
It will also be noticed, as a further illustration of the accuracy 
of adaptation, that the upper portion of connective is very much 
longer than is the lower. In some other species of salvia this 
difference of length is much greater even than here. The result 
of this is, that the bee produces a very considerable rotation in 
the upper limb of the lever, notwithstanding that the direct 
motion produced by its own pressure on the lower limb is com- 
paratively slight. 
The shape of the corolla is also adapted to facilitate the 
motion in question or rather to increase its range. It will be 
noticed in fig. 2 that the tube bulges out just behind the barren 
anther cells. This allows of a greater displacement of the lower 
arm of the lever in a backward direction than would be possible 
were the bulging not present. It is easy to convince oneself, by 
inspection, that after the bee has struck the lower anther cells 
Avith its head, it penetrates still deeper, and that its back forces 
the cells into this retreat. It thus happens that, though the 
connective is not nearly so long in this species of salvia as in 
many others yet that the amount of motion produced in the 
* I refer, of course, to the law of halancement of growth, which was thus 
expressed by Goethe : — order to spend on one side. Nature is forced to 
economise on the other side.” The ment of haA’ing first propounded this 
law is claimed for Geofffoy St.-IIilaire, and also for Goethe. It had, how- 
ever, been most distinctly enunciated by Aristotle. For instance, cf. his trea- 
tise ^^De Partibus Animalium,” ii. 9 : tVjua r/)v avTt)v vTrepoxf)i’ fk ’roWoi'*; 
ToTTOvg adwarii dtav^i-uiv <pvaig^ k.t.X. 
