400 



A NGIOSPERMA E—DICOTYLEDONES 



Visitors. — Small flies and ants were seen in Nova Zemlia, but no visitors in 

 Spitzbergen. 



Herm. Miiller observed in the Alps no fewer than 85 Diptera (mostly Muscidae). 

 besides 8 beetles, 20 Hymenoptera, and 13 Lepidoptera ; Loew saw a hover-fl)' 

 in the same region (' Blutenbiol. Floristik,' p. 397). Lindman noticed flies, Hymeno- 

 ptera, and a beetle on the Dovrefjeld. MacLeod saw 8 short-tongued Hymenoptera, 

 a Phryganid, a beetle, 4 Syrphids, and 19 other Diptera in the Pyrenees (Bot. Jaarb. 

 Dodonaea, Ghent, iii, 1891, pp. 420-2). 



945. S. Hirculus L. (Warming, Bot. Tids., Kjobenhavn, xvi, 1888, p. 25.) — 

 Warming says that the flowers of this species are markedly protandrous in Spitz- 

 bergen. Andersson and Hesselman (' Bidrag till Kanned. cm Spetsbergens o. Beeren 

 Eil. Karlvaxtflora,' pp. 27-8) say that this species flowers in Spitzbergen during 

 the second half of July and in August ; feebly developed flower-buds were noticed in 

 Beeren Island on June 18, 1898. Fruits appear to be ripened with difficulty at 

 the end of August or the beginning of September. The petals are bright yellow. 



Fig. 128- Saxifraga rotuttdi/olia^ L. (after Herm. Mi. Her). A. Flower at the beg^inning of t:he 

 first (male) stage. B. Flower towards the end of the same stage. C. Flower in the second (female) 

 stage ( X \%^. a-a-, anthers ; Ji, filament ; /;, drop of nectar ; ov^ ovary ; /, petal ; .y, sepal ; si, stigma. 



with a basal swelling on either side, and citron-yellow spots. There is marked 

 protandry, according to Andersson and Hesselman, confirmed by Ekstam. Stocks 

 with purely female flowers were seen, and also gynomonoecious transition forms. 

 Ekstam thinks that the basal folds of the petals, the roots of the filaments, and 

 the base of the ovary possibly secrete nectar. In Nova Zemlia, according to Ekstam, 

 the diameter of the odourless, nectarless (?), slightly protandrous flowers is 1 2- 

 25 mm. 



Visitors. — On seven days of July and August (1897) Ekstam noticed an 

 extremely large number of flies. These insects have also been seen in Nova Zemlia. 



946. S. rotundifolia L. (Herm. MuUer, ' Fertilisation,' p. 245 ; ' Alpenblumen,' 

 pp. 89, 90.) — The flowers of this species are white, spotted with purple-red, and 

 secreting half-concealed nectar. They are so markedly protandrous that automatic 

 self-pollination cannot take place. They belong to the class of fly flowers. Insect- 

 visits necessarily bring about crossing, for even the smallest nectar-seeking guest 

 touches the anthers of younger flowers, and one or both stigmas in those which 

 are older. 



