96 



INTRODUCTION 



this. These will be thoroughly described later on (cf. also p. 67). In this place 

 the various forms of nectary will not be fully considered, but one example may serve 

 to indicate in what variety these organs may occur even within a single family, 

 the Ranunculaceae (Fig. 12). 



The form of the nectaries is not always the same, even in one and the same 

 species of plant. Herm. Miiller gives in his works two particularly striking examples 

 of the kind, i.e. Ranunculus auricomus ('Fertilisation,' p. 79) and R. pyrenaeus 

 (* Alpenblumen,' p. 133) (see Fig. 13). 



Fig. 13. (1-8) Different petals of Ranunculus anyicomus with diversely-formed nectaries (). 

 (g) Petal oi Eranthis hyemalis. (After Herm. Miiller.) 



Nedar-covers, which also were frequently mentioned by Sprengel, and which are 

 formed by projecting parts of flowers, processes, hairs, and so forth, serve as a means 

 of protecting nectar from rain. Owing to such shelter nectar is not only saved from 

 dilution, or even from being washed away, but more abundant secretion and 

 accumulation are also rendered possible, with the result that insect visits are more 

 numerous. On the other hand, since the nectar is sheltered deeper down, numerous 

 of the less industrious insects are debarred from visiting and pollinating the flowers. 



In order to render more easy the finding of nectar by insects that have been 

 attracted by colour or odour there are, as Sprengel pointed out (see pp. 4-6), many 

 spots or lines on the flower, which by their position or direction indicate the place 

 where the honey is concealed (cf. Fig. 14). These Nectar-guides, however, are 

 naturally only present in such flowers as are visited by insects during the day. They 

 are wanting (see p. 67) in moth flowers, where they would be useless (cf. also 

 Sprengel, ' Entd. Geh.,' p. 16). 



The nectar-guides of one and the same species of plant have not always the 

 same form, but may be developed in various ways. I have described (' Bl. u. Ins. a. d. 

 Nordfries. Ins.,' p. 52) and figured (see Fig. 15) such variable nectar-guides in 

 Erodium Cicutarium. 



The correctness of Sprengel's theory of nectar-guides has often been doubted ; 

 but it may be accepted as valid till some other and better explanation can be given of 

 the dots, lines, streaks, and markings that occur on the petals of flowers. 



In support of Sprengel's theory of nectar-guides Hildebrand mentions (' Die 

 Farben der Bliiten,' p. 72) a phenomenon frequently exemplified by double flowers. 

 It appears that when nectar-guides are strikingly developed in the simple form of 



