512 THE FERTILISATION OF FLOWERS. [PART III. 
Volucella bombylans, L., and Syrphus ribesii, L. All three settled 
upon the spikes, thrusting their tongues into one flower after 
another, and so dusting their ventral surfaces with pollen. 
In my garden I have often seen Sarcophaga carnaria trying to 
insert its proboscis into the flowers, and generally failing in the 
attempt. Syritta pipiens, L., always failed in the same attempt ; 
if the flower happened to be in the second stage, it flew away 
after a few failures, but stayed to eat pollen if the flower was in its 
first or male stage. 
I have often seen Andrena albicans, L. ¢, climbing up the 
spikes. Now and then I have watched an individual first failing 
continually on several spikes to insert its proboscis into the flowers, 
and then gradually acquiring the knack, and ending by performing 
the operation regularly. 
The honey-bee visits the plant in considerable numbers, and 
never fails to insert its proboscis successfully. 
Visitors in Low Germany: A. Hymenoptera—Apide : (1) Apis mellifica, 
L. %; (2) Andrena albicans, lL. 9. B. Diptera—(a) Syrphide : (3) Volucella 
bombylans, L. ; (4) Syrphus ribesii, L. ; (5) Syritta pipiens, L. ; (6) Rhingia 
rostrata, L. ; (6) Muscide: (7) Sarcophaga carnaria, L. See also No. 590, m1. 
A list of Alpine visitors (one beetle, ten flies, five Hymenoptera, twenty-one - 
Lepidoptera, and Panorpa communis) is given in No. 609. 
Polygonum viviparum, L., is remarkable for the polymorphism 
of its flowers. In Sweden, Axell found it occurring in two forms, 
some plants being hermaphrodite and proterandrous, others female 
only (17). I have also found hermaphrodite and female forms at 
Franzenshoh, but the former were homogamous. At Madulein I 
found hermaphrodite stocks showing all stages of suppression of 
the pistil down to purely male flowers and plants (609, fig. 169). 
371. Potyconum Perrsicaria, L.—Like the two foregoing 
species, the flowers have white or red perianths; they are 
aggregated in a spike 20 to 30 mm. long and 6 to 10 mm. broad. 
The smaller size of the flower and of the spike, the want of 
odour and much smaller supply of honey, render insect-visits very 
much rarer. It is, therefore, of more importance for self-fertilis- 
ation to be possible in default of insect-visits, than for cross- 
fertilisation to be absolutely insured when insect-visits do take 
place. Accordingly, the structure of the flower differs considerably 
from that of the two preceding species. 
Of the eight stamens three at most are much reduced or have 
entirely disappeared, though flowers often occur with six, seven, or 
