Feb. 8, 1877] 



NATURE 



Z^l 



FERTILISATION OF FLOWERS BY INSECTS^ 



XV. 



Alpine Species of Genliana. 



T N previous artxles I have attempted to show that in 

 -•■ the Alpine region Lepidoptera are relatively much 

 more frequent visitors and fertilisers of flowers than in 

 the plain and lower mountain region ; and that, in 

 connection with this fact, in the Alpine region certain 

 flowers are found adapted to cross-fertilisation by butter- 



FiG. 93. 



Fig. 94. 

 Figs. 94, gs — Gentiana lutea, L.— Fig. 94. — Whole flower, a little mag- 

 nified, seen obliquely from above. Fig 95. —Undermost portion of the 

 ovary, showing the nectary and two filaments.^ 



flies and moths, the nearest allied of which, inhabiting 

 the plain or lower mountain region, are adapted to cross- 

 fertilisation by bees. As a further confirmation of this 

 statement, we may consider the genus Gentiana, which, 

 besides some species inhabiting the plain and lower 

 mountain region, includes various beautiful Alpine forms. 

 The former, G. cruciata, G. p7ieumonanlhe, and G. ciliataj^ 

 are all adapted to cross-fertilisation by larger Apidas, 

 chiefly by humble-bees, whereas in the Alpine region, 

 besides many species adapted to humble-bees and one 

 accessible to insects of all orders, there are also nume- 

 rous species adapted to Lepidoptera. 



Fig. 96 

 Figs. 96, gj. — Gentiana punctata, L. — FiG. 96. — Flower in its natural posi- 

 tion (nearly 1^:1), the anterior part of the corolla having been removed, 

 as far as the filaments, which are not united with it. Fig. 97. — Pistil of 

 the same flower. 



It may be worth while considering by what modifica- 

 tions of structure the adaptation of one and the same 

 genus to such different visitors has been e fected. 



I . A Ipine Species of Gentiana accessible lo Insects of all 



' Continued from vol xiv., p. 175. 



' The'foUowing explanation of the lettering applies to all the figures : - 

 a =anth-rs, ca = calyx, c/« = channels conducting to the honey, co = corolla, 

 '/ = filaments, n = nectary, o =■ openings conducting to the honey, ov = ovary, 

 / = petals, /r = pro;ecting hairs (Sprengcl's iiaftdecke), ;f = sepals, st — 

 stigma 



i G carnpcstris, gemtanica, and amarella, inhabit the Alpine region, 

 the mountain region, and the plain. 



Orders. — By far the most simple structure of flowers 

 among all the Gentianae is to be found in G. lutea (Figs. 

 94, 95), which may therefore perhaps be considered as the 

 nearest allied to the common ancestor of the whole genus. 

 Its flowers are perfectly open ; the anthers and stigma 

 are developed simultaneously, and in some flowers one 

 of the anthers is found in contact with the stigma, so that 

 self-fertilisation is by no means excluded. The honey — 

 being secreted by an annular swelling of the base of the 

 pistil («, Fig. 95) so copiously that a large drop of it 

 completely covers the excavated base at each of the five 



Fig. 98. 

 Figs. 98-102. — Gentmna tenella, Rottb. iglaciaiis. Thorn.). — Fig. 98. — 

 Flower seen from above (3J : i). Fig. 99. —The middle part of the same 

 flower (7:1). Fig. 100. — Lateral view of the same flower (3^:1). 

 Fig. ioi. — A piece of the corolla with the adherent filaments and nec- 

 taries. Fig. 102. — Flower bisected longitudinally (32 : i). 



petals and touches the two neighbouring filaments — is 

 visible and accessible to flying insects of all orders, whilst 

 ants and other insects creeping to the flowers are fre- 

 quently prevented from gaining the honey by the basal 

 lobes of the opposite leaves uniting round the stem, so as 

 to form a kind of basin in which rain-water is collected.^ 



The splendid yellow colour of the large flowers, which 

 are grouped in numerous whorls round stems of more than 

 a man's height, makes them more conspicuous than the 

 flowers of any other species, and attracts plenty of various 

 insects, which alight on these flowers for honey and for 

 pollen.^ Some of them alighting in the middle of the 

 flower will first touch the stigma and dust it with pollen 

 from previously-visited flowers, and thus effect cross-ferti- 

 lisation. This, however, is by no means secured, and 

 many flowers, in spite of numerous visits of insects, may 

 remain quite unfertilised by them, so that the possibiUty 





Fig. 99. 



of self-fertilisation above alluded to is probably not use- 

 less to the plant. 



2. Alpine Species of Gentiana adapted to Humble-bees. — 



I See Kerner, Die Schutzmittel der BlUthen gegen unbtru/ene Caste. 

 Wien, 1876, p. 207. 



* Only once have I had the opportunity of watching G. lutea, in the Roseg 

 Valley, near Pontresina, July 29, 1876. Here I found its flowers visited by 

 CoLEOPTKRA : Maltkodes jlavogiitiatus, some specimens ; Anthiophagus 

 alpinus, numerous specimen.s ; Epuraea aestiva, in the largest number ; 

 DiPTERA : some species not yet known to me ; Lepidoptera : Agrotis 

 ocellina, pretty frequently, sucking ; Hymenottera : Tentkredo species 

 (similar to T. notha), some specimens ; Anthiophora spec, not yet known 

 to nie, some specimens ; and once, Bombus pratorum, 5i the last two both 

 sucking and collecting pollen. 



Q 2 



