ALISMACEAE 503 



blossoms of the inflorescence open first, and then the pseudo-hermaphrodite pollen- 

 flowers above them, so that the species is protogynous in the neighbourhood of 

 Neu-Ruppin. The anthers are of a beautiful dark-brown colour, and dehisce 

 laterally. The pollen-grains are yellow, rounded polyhedral, closely beset with short 

 spines, and therefore clinging to the dehisced anther-lobes, about 27-31 fx in 

 diameter. 



Warnstorf distinguished the following variations in the inflorescences at Ruppin. — 

 (i) The lowest of the trimerous whorls contains female flowers only, and the one 

 above this 1-2 such flowers, while the rest are entirely composed of pseudo- 

 hermaphrodite pollen-flowers. 



(2) One of the female flowers of the lowest whorl is replaced by a branch bearing 



pseudo-hermaphrodite pollen-flowers; otherwise as (i). 



(3) There is only one female flower in the lowest whorl, the other two being replaced 



by branches bearing pseudo-hermaphrodite pollen-flowers only. This is rare. 



(4) The lowest whorl possesses one female flower, a single pseudo-hermaphrodite 



pollen-flower, and two branches bearing pollen-flowers only. 



(5) Instead of the one female flower in the lowest whorl there is a pseudo-herma- 



phrodite male flower with a branch in the axil of its bract bearing flowers of 

 the same kind. 



(6) The lower whorl consists of one female flower, three male flowers, and two 



flower-branches bearing male flowers. Very rare at Ruppin. 

 Visitors. — MacLeod observed 3 Muscids in Flanders (Bot. Jaarb. Dodonaea, 

 Ghent, V, 1893, p. 288). 



961. Butomus Tourn. 



Flowers homogamous to protandrous, with half-concealed nectar secreted at the 

 base of the ovary. 



2943. B. umbellatus L. (Sprengel, 'Entd. Geh.,' p. 234; Herm. Miiller, 

 ' Weit. Beob.,' I, p. 293; Kirchner, 'Flora v. Stuttgart,' pp. 182-3; Schuiz, 

 'Beitrage,' I; Knuth, 'Bl. u. Insekt. a. d. nordfr. Ins.,' pp. 133-5.) — The flower 

 mechanism of plants belonging to this species seems to vary in different districts, for 

 the descriptions of Sprengel, Hermann Miiller, and A. Schuiz diff"er considerably. 

 I will therefore first give an account of the mechanism as I have observed it in 

 Schleswig-Holstein. About twenty flowers form a terminal umbel, and the diameter 

 of each flower is about 4 cm. Both sepals and petals are pale-pink in colour, darker 

 in the middle and towards the base. The stamens and, in the second stage of 

 anthesis, the carpels also, are dark-red ; they thus help to make the flowers con- 

 spicuous. In the island of Fohr and near Kiel I observed rather marked protandry. 

 There are nine stamens, and the anthers of the six which are not superposed on the 

 petals dehisce first, their filaments at the same time bending backwards and outwards. 

 When these six have withered, the anthers of the other three dehisce, but their 

 filaments remain fairly erect. The carpels now mature. The ovaries, hitherto pink 

 on the outer margin only, become dark-red, and the stigmas appear in the form of 

 slits, which expand gradually to such an extent that they are brought below the still 

 pollen-covered anthers of the three stamens which matured last. In the first part of 



