254 



ANGIOSPERMAE—DICOTYLEDONES 



fact that the iirst flowers of an inflorescence to open are all purely male (with 

 vestigial pistils), while it usually happens that in the lower part of the inflorescence, 

 some flowers are actually female, since their anthers fall oflF without dehiscing, 

 although the pollen-sacs are full of pollen-grains (Miiller). Martelli states (op. cit.) 

 that the lowest flowers are the only fertile ones, and that not more than two to 

 four of them produce seeds, these being the fourth (rarely the third) to the seventh 

 in regular succession, counting from the base of the cyme. 



Warnstorf (Schr. natw. Ver., Wernigerode, xi, 1896, pp. 1-12) gives a similar 

 account of the distribution of the sexes. The lower flowers of the panicle are male, 

 and are the first to open ; towards the middle isolated pseudo-hermaphrodite pollen- 

 flowers, with no style and a sessile stigma are often to be found. The upper flowers 

 are hermaphrodite and protogynous, with projecting styles ; their ovaries are beset 

 with stalked red glands of large size. The smooth pollen-grains are vermilion in 

 colour and ellipsoidal, with several longitudinal furrows ; size about 20 /«, broad 

 and 37-40 /i, long. 



The dimensions of the flowers correspond with those of the chief visitors, 

 humble-bees, which when alighting at once settle in the most convenient position 



Fig. 78. Aesculus Hippocastanum^ L. (After Herm. Miiller.) i. Male flower in section. 2, Herm- 

 aphrodite flower in the first (male) condition, seen obliquely from the front. 3. The same in the second 

 (female) condition, in section, a, anthers; «, nectary; (Jz/, ovary ; oz'' , vestigial ovary ; ^.petals; j, sepals; 

 St. stigma. 



for sucking, and in so doing touch either the stigma or the anthers with the under- 

 side of the abdomen, so that cross-pollination is always effected. The other bees 

 observed by Herm. Miiller (Apis, Eucera, Osmia rufa L., Halictus rubicundus C^r., 

 Andrena) do not correspond with the flowers in dimension. 



Visitors. — The height of the trees makes it very difficult to observe visitors, 

 but I was able to carefully note the following actively engaged on the flowers. — 



A. Diptera. (a) Muscidae: i. Musca domestica Z., skg. from the side without 

 touching stigma or stamens; 2. Scatophaga merdariaZ., do.; 3. S. stercoraria Z., do. 

 {b) Syrphidae : 4. Syritta pipiens Z., skg. from the side as before, also po-dvg., but 

 not touching the stigma ; 5. Syrphus balteatus Deg., as Musca. B. Hymenoptera. 

 Apidae: 6. Apis mellifica Z. 5, skg. while hanging on the filaments from below, 

 and therefore touching neither stigma nor anthers ; sometimes also po-cltg. ; 7. Bombus 

 lapidarius L. 5, skg. legitimately and effecting pollination; 8. B. terrester Z. 5 

 and 5, do. 



Loew saw the honey-bee, skg. and po-cltg., in the Berlin Botanic Garden. 



Alfken and Hopper (H.) noticed the following Apidae at Bremen: — i. Apis 

 mellifica Z. g ; 2. Bombus terrester Z. 5 ; 3. Podalirius retusus Z., var. obscurus 

 Friese 5 (H). All freq., skg.; i and 2 also po-cltg. 



