CAPRIFOLIACEAE 527 



not consider the swelling at the base of the style as a nectary; but Hermann 

 Rluller, on the other hand, describes it as such. The swelling in question appears 

 to me to possess the characters of a nectary in so marked a form that I concur with 

 the latter view. 



1219. S. racemosus IMichx. (Herm. Muller, 'Fertilisation,' pp. 292-3, 'Weit. 

 Beob.,' Ill, p. 78; Knuth, ' Bl. u. Insekt. a. d. nordfr. Ins.,' p. 81, ' Bliitenbesucher,' 

 I, p. 16, 'Blutenbiol. Beob. in Thiiringen,' ' Bloemenbiol. Bijdragen'; MacLeod, 

 Bot. Centralbl, Cassel, xxix, 1887, P- 119; Loew, 'Blutenbiol. Floristik,' p. 250, and 

 bibliography.) — The slightly odorous pendulous flowers of this species obviously 

 belong to class Hw. They are 7-8 mm. long and 5 mm. broad, and therefore, 

 according to Hermann Miiller, conveniently receive the head of a wasp (5 mm. 

 broad, 2-2^ mm. thick). Wasps actually appear to be the most frequent visitors 

 and pollinators, but, in my opinion, the size of the heads of other guests, such as 

 hover-flies and bees, corresponds equally well to the internal dimensions of the 

 flowers. 



Where the pendulous bell-shaped flower is broadest, its inner surface is closely 

 beset with numerous long hairs. These stretch 

 from the five lobes of the corolla to the middle of 

 the bell, and thus not only eff'ectively protect the 

 abundant nectar from the rain, but also prevent 

 it from running out. 



The five epipetalous stamens spring from 

 about the middle of the bell. They converge so 

 that the introrse anthers are placed at the entrance 

 of the flower, in the lowest part of the lining of Fig. 172. Svmf>/ioricarpos racetnosa, 



hr J- .. 1 u L .!_ 1 .1 M{c/ix.(3iherHerm.MiiUeT). (i) Flower 



airs. Immediately above the anthers, where the ^^en from the side. (2) Ditto, in longu 



bell narrows, the simultaneously maturing stigma is tudinai section ( x 2I). 

 situated. 



When a nectar-seeking insect, with a head corresponding to the size of the bell, 

 pushes this into the flower, it first of all comes into contact with the five anthers, 

 getting covered with pollen, and then one side of it touches the stigma. Hermann 

 Miiller says, however, that little or no pollen remains on the insect's head by the time 

 the stigma is reached, partly because this is but slightly sticky, partly because any 

 grains that happen to adhere are brushed off again by the thick coating of hairs which 

 lines the corolla. It is only when withdrawing from the bell that the head of the 

 insect, now wet with nectar, becomes thoroughly covered with pollen, part of which 

 is transferred to the stigma of the next flower visited. Cross-pollination is thus 

 effected. In the absence of insect visitors, automatic self-polhnation is probably 

 always prevented, owing to the pendulous character of the flower, and the relative 

 positions of anthers and stigma. 



Visitors. — I have seen the humble-bee Bombus agrorum F., and the true wasp 

 Vespa saxonica F., skg., in Thuringia : in Schleswig-Holstein, Mecklenburg, and 

 Pomerania, in spite of careful observation of the flowers, I have never seen a wasp 

 visiting them, but only nect-skg. bees (Apis, Bombus terrester L.) and hover-flies 

 (Eristalis sp., Syrphus ribesii £., Syritta pipiens Z., po-dvg.): also (20. 7. '97), at 



