70 



A NGIOSPERMA E—DICO TYL ED ONES 



proved by numerous experiments that the flowers are quite infertile when pollinated 

 with their own pollen, and very largely so when geitonogamously pollinated. 

 Xenogamy is necessary for complete fertility. 



Visitors. — As, according to Hermann Muller, the spur extends backwards for 

 12 mm. from its point of attachment to the peduncle, and the nectar fills only 

 its last 4-5 mm., there is — among the bees which are active when this plant is 

 in flower — only one species (Anthophora pilipes F. 5 and 5, with a proboscis 

 19-21 mm. long) that can reach the nectar in the legitimate way. This insect — 

 according to Hermann MuUer's observations at Lippstadt and my awn at Kiel — visits 

 the flowers so diligently, and in such numbers, that probably none of them remain 

 unfertilized. 



Bombus terrester Z. here makes its appearance as a nectar-thief. It would 

 be just able to reach the beginning of the nectar with its proboscis (7-9 mm. long). 



D. 



mm 



Fig. 24. Corydalis cava, Schweigg. et Korl. (from nature). A. Flower with elosed hood, seen from 

 the side ; the spur has been perforated by Bombus terrester. B. Flower seen from the front. C, Flower 

 with hood turned down so that the pollen-covered stigma is visible : seen from the side ; the nectary 

 is seen shining through the spur (enlarged). D. Lobed stigma covered ^vith pollen : beneath it are 

 the withered empty anthers (more highly n>flgnified). 



and to get some of it, but instead of doing so it perforates the upper side of the 

 spur in the neighbourhood of the nectar, either at the point of curvature or nearer 

 the base. It then inserts its proboscis into the hole and sucks the sweet fluid. 

 (Cf. Fig. 24 ^.) By using the holes made by Bombus terrester, other short-tongued 

 bees are also able to get the nectar, e.g. the honey-bee (proboscis 6 mm. long), 

 as well as species of Andrena, Sphecodes, and Nomada. The honey-bee certainly 

 sometimes attempts to reach the nectar legitimately, but always fails, owing to 

 the shortness of its proboscis. In making these occasional vain attempts it — and 

 also Anthophora pilipes — effects cross-pollination. Crossing is also brought about 

 if — abandoning the search for nectar — such a bee remains in the flowers to collect 

 pollen. Hermann Muller also observed a few hover-flies (Bombylius major Z., and 

 B. discolor Mi'kan, with proboscis of 10 and 11-12 mm. respectively), hovering after 



