THE POLLINATION OF THE PRIMROSE 65 
sufficiently long to reach the nectar. Butterflies are compara- 
tively so rarely to be seen at that early spring season — only 
hibernated specimens being then upon the wing — that their 
influence would be almost nil. So these long-tubed flowers, 
naturally excluding all but a few insects, must assuredly be very 
ill-adapted for insect visitors. 
We nosv pass to the insect visitors seen by Dr. Weiss in 
the “two large patches of primroses” in the “sheltered” 
position. 
In this situation Prof. Weiss saw five species of insects 
visiting the primroses: Andrena Gwyuana, the hive bee {Apis 
mellifica), Bombus ierrestris (proboscis 7 to 9 mm., H. Muller) ; 
Anthophora furcata, and Bombyliiis major. The three first men- 
tioned must be set aside at once as effecting pollination in 
these flowers. The proboscis of all three is too short to reach 
the nectar. Prof. Weiss says of Andrena Gwynana, whose pro- 
boscis is only 2^ mm., that “ they are very active agents of cross- 
pollination of the primrose.” This opinion is completely at 
variance with its observed habits. Darwin, following Muller, 
says : “ Bees possess acute powers of vision and discrimination, 
for those engaged in collecting pollen ” (as A. Gwynana with its 
very short tongue would alone do) from Primula elatior invariably 
passed by the flowers of the long-styled form. This exactly 
agrees with our own observation of A. Gwynana’ s visit to the 
cowslip. (We have never met with it upon the primrose : this 
rarer occurrence in the primrose arises from the earlier appear- 
ance of the primrose ; but even on the cowslip their appearance 
is very exceptional.) We have caught it on the short-styled cow- 
slip, but have never seen it upon the long-styled. It would 
consequently, even if numerous, not be an agent of cross- 
pollination at all. The same would apply to the one hive bee 
which Dr. Weiss saw upon the flowers, and the three specimens 
of Bombus terrestris. There is thus left one, Anthophora jurcata} 
Bombylius major. Dr. Weiss saw on an average one Antho- 
phora a day (seven in eight days). Only one such visitor seen 
daily is conclusive evidence that no measurable pollination could 
be effected by Anthophora “ in two large patches of primroses,” 
even supposing such a visitor deposited a grain or two on each 
visit to a flower, and that it alternately visited the different 
forms of the flower ; whereas, usually it would visit every 
flower on each root before it passed to another plant. It 
would thus rather, if anything at all, be an agent of self- 
fertilisation. Moreover, it would not be likely to convey much, 
if any, pollen on its head, even if it touched with it the anthers 
of the short-styled, as its anthers burst inwardly atid only 
‘ Dr. Weiss, we think, must have mistaken A. pilipes for A. furcata 
as furcata does not fly until July (Saunders’ “ Hymenoptera aculeata,” p. 350 ; Fred 
Smith’s “ Hym. acul.f' vol. ii. 193). The specimens which I have captured of 
■^urcata have been always in July and August. 
