yune 18, 1874J 



NA TURE 



129 



FERTILISA TION OF FLO WERS B V INSECTS * 

 VI. 



Difftrt-)it Modes of Self-ftitiliSiXtioii lulurc Visils 0/ 



Insects are ■wantiu" 

 'T'HE two functions of cross-fertilisation and self-fertili- 

 ■*■ sation, which in the previous articles we have seen 

 to occur in different forms of the same species or 

 genus, are in most cases successively presented by the 

 same form of flower ; and the modifications by which 

 self-fertilisation is attained by different plants, where 

 visits of insects are wanting, are almost as various as the 



When the flower opens, it stretches forth its small 

 petals (/), which serve as nectaries (//"), and offer a small 

 drop of honey IJi), by which very minute insects are at- 

 tracted in sunny weather. These visitors arc for the most 

 part Diptera not exceeding i-iimm. in length, belonging 

 to the genera Sciara, Chironomus,* Hydrellia,* Scatopse,t 

 Phora, Cecidomyia, Oscini;;, and Microphorus. I ob- 

 served also a single specimen of Mclanostoma mcllina 

 (Syrphidas), some Anthomyia (Muscida;), a small Haltica, 

 some Pteromalidas, and small Ichneumonids. These 

 minute visitors licking up the drops of honey, and 



Fig. 32. — Side view of flower of Myosunts luiniiuHS before opening. Kio. 

 33. — Side view of a flower just open. 



contrivances by which cross-fertilisation by insects is 

 secured. Of these various modes 1 shall here speak only 

 of some not yet referred to in my book on fertilisation."!" 



Myosunts niiniinus is as remarkable for the great varia- 

 bility in the size of its flower (compare Figs. 35 and 38) 

 and in the number of its parts,! as for the enormous 

 growth of the cone of pistils, which affords no other 

 benefit to the plant but the self-fertilisation ot the greater 



Fig. 37. 



Fig. 36. — Stigma of the 



-r, Fig. 33, provided wilIi pollen-grains 

 have already emitted their pollen-tubes. 

 Fig. 37 —Stigma of the ovary y. Fig. 35, more developed than the 

 slignia, Fig. -,6, provided with two pollen-grains fallen doivn from the 

 aniher a. Fig. 38.— One of the smallest flowers : s. sepal, /), petal, 

 tt. nectary. It, honey, k, cone of ovaries. Figs. 3;-35 and 3S, seven 

 limes natural size. ^^ 



walking round the cones of ovaries, stop many seconds 

 in a single flower before visiting another. The an- 

 thers, lying close round the cone of ovaries, open by 

 two lateral slits, and are soon afterwards covered with 

 pollen on their whole outside ; consequently, insects walk- 

 ing round the ovaries may easily be charged with pollen, 

 and flying to another flower effect cross-fertilisation. But, 

 upon the whole, the flowers, because of their being scent- 

 less and very inconspicuous, are so scantily visited 

 by insects, that, after repeated careful e.xaminations, I 

 believe that even in sunny weather more than 90 per cent, 

 of the flowers remain without any visit. This deficiency 

 of secured cross-fertilisation is supplied by regular self- 

 fertilisation in the following manner. The axis of the 

 flower, extending gradually during the blooming-time into 



Fig. 34. 

 Fig. 3.1.— The sanit 



part of the numerous stigmas by the small number of an- 

 thers, in case it is not visited by insects. 



* Continued from vol. ix. p. 166. 



t " Die Befruchtung der Blumen durch Insecten," Leipsig, 1873. 



1 I examined a hundred flowers as to the number of sepals, petals, 

 and anthers, and found the number of sepals in 3 flowers = 4, in 94 flowers 

 = 5, in 3 flowers = 6 ; pet ils in 2 flowers = 2, in 20 flowers = 3, in 3 flowers 

 = 4, in 35 flowers = 5 : anthers in 2 flowers = 3, in 2 flowers = 4, in 11 

 flowers = 5, in 22 flowers = 6, in 31 flowers = 7, in 46 flowers = 8, in 5 

 flowers = 9, in I flower — 11. 



Fig. 39. — Flower of Myosotis ■z'ankolor v/hen just opened, dissected lon- 

 giludinaily. Fig. 40.— .^n older flower, with full-grown corolla (seven 

 times natural size) ; ca caly.v, Co corolla, n anthers, si stigma, « nec- 

 tary, ov ovary. 



a long cone, brings a great part of tlie stigmas into con- 

 tact with the lateral pollen-grains of the anthers ; those 

 ovaries which now are in contact with the anthers soon 

 afterwards overgrowing them, and others now bjlow the 

 anthers reaching them.J Thus a number of styles 

 grazing the anthers during the growth of the long cone 

 are self-fertilised by about five or more pollen-grains 

 (Fig. 36) ; besides, also, the lowest stigmas of the flowe 

 are fertilised by their own pollen, many pollen-grains fall- 

 ing down from the anthers (Fig. 35 y, Fig. 37). Conse- 

 quently, only those ovaries are never self fertilised which 



• For instance, Cliirofiomns hyssinus Schrk., Uy.lrcllU ehrysoslotna 

 Meig., zn&griseola Fall., after Prof. Milt's (of Vienna) examination, 

 t Scatopse brevkornis LOw. 

 X This is easy to be seen by marking some of the ovaries with a spot of ink. 



