NATURE 



\Dlx. io, 1874 



PER Tins A TION OF FLO WERS B Y INSECTS * 



VIII. 

 Alpine Species adapted to Cross-fertilisation by Butter- 

 Jlics, while the most nearly allied species which inhabit 

 the plain or lower mountain region are adapted to 

 Cross-fertilisation by Bees. 



IN the last article I attempted to show that in the 

 Alpine region Lepidoptera are far more frequent 

 visitors of flowers than in the plain and lower mountain 

 region, while the frequency of Apida^, not only absolutely 

 but to a still greater extent relatively, is greatly diminished 

 towards the snow-line. If this be so, whatever may be 

 the cause of the fact, it is hardly to be supposed that the 



Fig. 41. — Daphne Mezereitjn, L , 

 dissected longitudinally. 



(Both figures 



42. -The same f 

 from abo 

 atural sue ) 



different proportion of visitors of such different struc- 

 ture as butterflies and bees should not have in any way 

 influenced the adaptations of the flowers ; and indeed, 

 even during my short stay in the Alps, I succeeded in 

 finding some species of flowers adapted to cross fertilisa- 

 tion by butterflies, their most nearly allied species which 

 inhabit the plain or lower mountain region being adapted 

 to c OSS-fertilisation by bees. 



I. Daphne Mecercum and striata.— \n both species 

 (Figs. 41-44) the nectar is secreted in an annular swelling 

 {h) at the base of the ovary {01'), and is contained in the 

 lowest part of the tubular corolla, which includes (1) the 

 ovary {ov), terminated by a short-styled, knobbed stigma 



. ^-^.—Daplmi striata, Trat , 

 dissected longitudinally. 



(Doth figures 3! tin 



{st) ; (2) four lower anthers inserted above the centre of 

 the corolla-tube ((?') ; and (3) four higher anthers inserted 

 near its mouth (d'-). In both species, therefore, the pro- 

 boscis of a visiting insect, when in search of the honey, 

 grazes at first the higher, then the lower anthers, and at 

 last the stigma ; but the pollen-grains, being only slightly 

 sticky, scarcely adhere to the proboscis, and, at the most, 

 some few grains will be brought by it to the stigma of the 

 same flower. Only when retreating out of the flo.ver will 

 the probojc's, wetted with honey, be dusted by any con- 



* Ojutiniied Tr in p. ^53. 



siderable number of pollen-grains, which will partly be 

 deposited on the stigma of the next visited flower. Thus 

 cross-fertilisation is secured in case suitable insects visit 

 the flowers, whereas when visits of suitable insects are 

 wanting, pollen may easily fall down in both species from 

 the anthers upon the stigma of the same flower, and effect 

 self-fertilisation. 



Agreeing thus far, the flowers of the two species differ 

 remarkably in the length and width of the corolla and in 

 the insects which they attract. The corolla-tube oi Daphne 

 Me::ereiiin being 6 mm. long and 2 mm. wide, its honey is 

 accessible to a great number of bees, among them to all 

 humble-bees, and to some flies (Eristalis, Rhingia), which 

 will be attracted by the bright red colour, and when seek- 

 ing for honey and flying from flower to flower will regu- 

 larly effect cross-fertilisation. The honey is also acces- 

 sible to butterflies, but in consequence of the width of 



Fig 45. Fin, af, Fig. .17. Fig 48. 



Fig. 4 s —Primula villosa, Facq. Long-styled flower, natural size. Fig. 

 46. -Lower part of the same flower, longitudinally dissected : 3J times 

 natural size. FlG. 47.— Short-styled flower, natural size. Fig. 48.— 

 Lo.vcr part of same flower, longitudinally dissected : 3^ times natural size 

 «. nectary ; oz\ ovarj- ; st, stigma ; a. anthers. 



the corolla-tube the s!ender proboscis of these insects 

 will often be entered and retracted without touching 

 anthers and stigma. Daphne striata, on the contrary, 

 with corolla-tubes of lo-ii mm. long, the entrance of 

 which is only i mm. wide, is hardly accessible to any 

 insects except Lepidoptera ; and the pale rose or whitish 

 colour of its flowers, crowded together in tens or twenties 

 into umbels, and the entire absence (or nearly so) of scent 

 in the day-time, while they emit a remarkably sweet scent 

 during the evening twilight, prove them to be adapted to 

 Sphingidic and moths,* which, when visiting the flowers, 

 in consequence of the narrowness of the corolla-tube, 

 cannot avoid grazing the anthers and stigma and regu- 

 larly effecting cross-fertilisation. 



Fig 49. Fig. 50. 



FiG. ^q.— Primula officinalis, Facq. F1G.50.— The same : Short-styled 



Long-styled. Natural size. flower, longitudinally dissected. 



(Copied from Hildebrand, ** Geschlechtervertheilung," p. 31.) 



2. Primula officinalis and villosa (Figs. 45-50) are con- 

 nected with one another by a relation analogous to that 

 between Daphne Me:erei/in and striata. Both offer the 

 remarkable contrivances for cross-fertihsation which Mr. 

 Darwin has discussed in so masterly a manner in his 

 paper on Primula,f that is to say, both possess two forms 

 of flowers, a long-styled (Figs. 46,49) and a short-styled 

 (Figs. 48, 50), growing on different stems and existing in 

 nature in about equal number. As is evident from the 



* I have not yet succeeded in actually obsirving the fertilisation of either 

 of these two species ol Daphne. 



t ■' On the Two I'orms or Dimorphic Condition in the Species of Primula 

 and their remarkable Se.-suil Rekafons." Proc. Linn. Soc. vi. (1862'. Bot. 

 pp. 77-99. 



