220 ANGIOSPERMAEDICOTYLEDONES 



with hermaphrodite flowers there are rarely some with homogamous flowers, in 

 which automatic self-pollination is possible. In the South Tyrol Schulz observed 

 large-flowered male varieties, in which the branches of the style never spread out. 

 Unisexual flowers are gynodioecious or more rarely gynomonoecious, as well as 

 androdioecious and andromonoecious. In the Scandinavian highlands, Ekstam 

 observed female flowers with vestigial stamens, and small male flowers, in addition 

 to the protandrous hermaphrodite ones. 



Visitors. Schulz states that many insects visit the flowers in Central Germany. 



Herm. Muller saw 8 beetles, 21 Diptera, 24 Neuroptera, and 20 Lepidoptera, 

 in the Alps. 



In the Berlin Botanic Garden Loew observed two bees, skg., i. e. Apis mellifica 

 Z. 5 and Bombus hortorum L. 5 ; also Prosopis communis Nyl. $>, skg., in the variety 

 robuslum. He further noticed two Syrphids Platycheirus manicatus Mg. 5 and 

 Syrphus annulipes Zett, and a bee (Andrena sp.) in Switzerland. 



Schneider (Troms$ Mus. Aarsh., 1894) saw Bombus hypnorum L. and B. kirby- 

 ellus Curt. J and 5, in Arctic Norway, and Lindman observed Diptera and humble- 

 bees on the Dovrefjeld. 



In Dumfriesshire, Apis, freq., 2 humble-bees, 2 short-tongued bees, 3 Empidae, 

 5 Muscidae, and 2 hover-flies have been recorded (Scott-Elliot, ' Flora of Dumfries- 

 shire/ p. 34). 



Glandular hairs on the stem keep off creeping insects from below. 



559- G. pratense L. (Hildebrand, 'D. Geschlechts-Vert. b. d. Pfl.,' p. 27; 

 Herm. Muller, 'Fertilisation,' p. 150, ' Weit. Beob., II, p. 167 ; Schulz, ' Beitrage,' I, 

 pp. 27-8; Kirchner, 'Flora v. Stuttgart,' p. 336.) The mechanism of the markedly 

 protandrous flowers of this species agrees essentially with that of G. palustre. The 

 stamens at first lie on the petals, but become erect when the anthers dehisce, pass to 

 the middle of the flower, and after the pollen has been shed return to their former 

 position. Hildebrand has established by experiments that the stigmas are usually 

 still immature at the time when the anthers are covered with pollen, but become 

 mature when the dehisced stamens bend back. They cease to be receptive when 

 the petals fall away. 



Schulz says that the flowers vary considerably in size. In addition to herma- 

 phrodite flowers, he also observed female ones, distributed gynodioeciously or 

 gynomonoeciously. 



Warnstorf describes the pollen-grains as white, spherical, coarsely tuberculated, 

 and 100 fx. in diameter. 



This species also is protected against creeping insects by the sticky nature of 

 the stem. 



Visitors. The only one I have seen is the honey-bee, skg. Hermann Muller 

 observed the following in Westphalia and Thuringia. 



A. Coleoptera. Curculionidae: 1. Coeliodes geranii Payk., skg. (?); 2. Miarus 

 campanulae Z., do. (?). B. Diptera. (a) Siratiomyidae : 3. Nemotelus pantherinus Z. 

 (b) Syrphidae: 4. Melithreptus pictus Mg., po-dvg. C. Hymenoptera. Apidae: 

 5. Andrena coitana K. and J, skg. ; 6. A. gwynana K. $>, do. ; 7. Apis mellifica Z. 

 5, very freq., skg. ; 8. Chelostoma campanularum K. $ and J, freq., skg. ; 9. C. nigri- 

 corne Z. 5 and J, very freq., skg. ; 10. Coelioxys conoidea ///. $, skg. ; 1 1. C. elongata 

 Lep., do.; 12. C. quadridentata Z. S, do.; 13. C. rufescens Lep. 5 and J, do.; 14. 



