225 
of Westralian Plants . 
flower contains a fair-sized drop of nectar at the base of the corolla tube, 
yet hive-bees always pass the flowers and visit those of other plants, 
though they contain less nectar. I was puzzled by this strange behaviour 
till recently when I saw a hive-bee visit several flowers. Her head was 
blocked by the sturdy style, and I do not think her tongue reached the 
nectar-drop about 7 mm. distant. So far as I could see the pouch did not 
open. I discovered by accident not long ago that the flowers so blue by 
day appear brilliantly white in the moonlight. This appears to be due to 
the structure of the cuticle of the petals, making them excellent reflectors 
of light—so excellent, indeed, that the flowers appear brighter than many 
pure white blossoms. This fact suggests night-flying insects as pollinators. 
I have not yet been able to make any definite observations in this direction. 
I cannot think that Lepidoptera would prove efficient pollinators. The 
pollen is somewhat moist or waxy, and does not readily fall from the 
pouch, which, moreover, is broad-mouthed and seems intended for the back 
of a pollinator. This species is a very erratic fruiter. I have made some 
attempt to discover the reason. The plants commonly occur in groups or 
colonies. Careful digging reveals that the rhizome spreads widely under¬ 
ground and sends up aerial shoots at intervals. What seem to be several 
separate plants growing near together are often in reality simply branches 
of the same plant. I have found that the flowers are infertile with pollen 
from other flowers of the same plant. I have observed that when two 
distinct individuals grow in close proximity, both usually fruit freely; 
groups, on the contrary, seldom produce many fruits. This seems to show 
that the pollinator does not as a rule travel far between visits. In all other 
genera of Goodeniaceae known to me, viz. Goodenia , Scaevola , Darnpiera y 
and Brunonia , the indusium surrounds the style apex, which pushes the 
pollen before it as it grows ; and the stigma is not mature till the style tip 
has extended beyond the indusium and all pollen has gone. 
Species of Stylidium , Sw., bear flowers in some respects more highly 
specialized than those of Goodeniaceae. Stamens and style are closely 
welded together into a highly irritable column, which in the ‘set’ position 
leans upon the much reduced fifth petal (‘ labellum ’) with its apex below 
the plane of the corolla limb. The labellum is morphologically anterior, 
but torsion of the corolla tube carries it to the side or back of the flower. 
The column is bowed forwards a little just as it emerges from the corolla, so 
that it almost occludes the entrance. Slight pressure (a mere touch is not 
enough) against this bowed part causes the column to move with great 
rapidity till it leans over on the opposite side of the flower. It soon returns 
to its former position, but does not regain irritability till after a short rest 
there. I have seen various hymenopts visit the flowers and apparently 
obtain nectar without causing the column to act. Once I saw a largish 
‘ bee ’ visit several flowers on a raceme in rapid succession, working from 
Q 
