474 



NATURE 



{March 29, 1877 



manner, it passed over to Trifolium nivale, which it then 

 sucked perseveringly. If from these observations we may 

 infer that the forcible enlargement of the narrow corolla- 

 tube of the species of Gentiana in question is too in- 

 convenient for the humble bees, then the only insects 

 capable of gaining the honey in all the species of the pre- 

 sent group and capable of regularly visiting and cross- 

 fertilising them, are Macroglossa stellatarum, and perhaps 

 some other Sphingidae not yet observed by myself in the 



Fig. ioq. 

 Figs. 109-111. — Gentiana ver7ia (3J : i). Fig. log. — Flower, completely 

 opened, seen from above. Jig. iic— Upper portion of the same 

 flower, bisected longitudinally, showing the pistil and the anthers. Fjg. 

 III. — Lowermost portion of the same flower, showing the nectary. 



Alps. The smallest of the species in question, G. nivalis 

 (13-16 mm.), G. iinbricata (15 mm.), and G. pumila (16-18 

 mm.), may also be fertilised by some moths^ and by many 

 butterflies^ which will all easily insert their slender pro- 

 boscis into one of the small openings at the base of the 

 corolla-tube, then, withdrawing it smeared with pollen, will 

 leave some pollen-grains on the margin of the stigmatic 

 disc, and, when inserting the proboscis into another flower, 

 will efifect its cross-fertilisation by stripping ofif some of the 

 pollen-grains from the hair-like papillae at the margin of 



- St 



Fig. 1 10. 



the stigmatic discs. Humble-bees, on the contrary, at all 

 events, seem to be of very small or no importance as 

 fertilisers of this group. If thus, by indirect inference, 

 we are led to the conclusion that the present group 



' Plusta gamma, 15-16 mm. ; P. interrogationis, 15 ; P. Hochenwarti, 



.^ l^^^y."*^^^ "'^'i'P^' 13-14; A.aglaia, i^ik; A. niobe.yxt. eris,\y\6; 

 A. pafhia,\i--i^; Colias phicomone, 13-14; Erebia goante, ii-n; Hesperia 

 syivanus, 16 ; PaJ>ilio machaon, 18-20 ; Pararge maera, 13-14 ; Pamassius 

 apoUo, 12-13 : P- delius, 11-16; Pieris brassicce, 15-16: P. cratcegi, 15 ; P. 

 rapce, 14-18 ; Vanessa cardui, 15 ; and V. -urtica, 14-15 mm. 



is exclusively adapted to Lepidoptera, we ought to 

 embrace this opinion with due precaution ; for all 

 visits of Lepidoptera, hitherto directly observed by the 

 comparison of the length of their proboscis with the 

 length of the corolla-tube, prove to have been only 

 fruitless attempts. On G. nivalis (corolla- tube 13-16 

 mm.), near Pontresina, August 4, 1876, I observed Ca^fio- 

 ny7npha salyrioft, Esp. (proboscis, 7 mm.), and Hesperia 

 serratulce, Rbr. (10-11 mm.) ; on G. verna (corolla-tube, 

 23 mm.) in the Aibula Pass, July 28, 1876, I found Melitaa 

 aurinia, var. Mero_pe,Frunn. (7 mm.), Argymtis pales, S. V. 

 (9-10 mm.), and Erebia lappona, Esp. (8-9 mm.) ; on G. 

 bavarica (20-22 mm.), Kerner' observed Agrotis cuprea 

 (12 mm.), I myself, upon the Piz Umbrail, July 15, 1875, 

 saw Erebia lappona, Esp. (8-9 mm.) ; in the AlbuIa Pass, 

 July 27, 1876, E. lappona and Melitcea asteria (5-6 mm.) ; 

 in the Val del Fain, August 5, 1876, Melitcea aurinia, 

 var. Merope, Prunn. (7 mm.), all easily inserting their 

 proboscis into the corolla- tubes, but all apparently 

 without any advantage to themselves, though, by their 

 repeated fruitless attempts, some cross-fertilisation of the 

 flowers may have been effected. No direct observation 



Fig. 112. Fig. 113. 



Figs. 112-114 — Gentiana nivalis [■i\\-i). FiG 112.— Flower seen from above. 

 Fig. 113. — Pistil. Fig. 114 —Upper portion of the flower bisected longi- 

 tudinally, showing one of the anthers (a') in contact with the margins of 

 the stigma. 



of the true natural fertilisers of the species of Gentiana 

 belonging to our fourth group has yet been made. 



In former articles I have shown that frequently, of 

 different species of the same genus, those 

 possessing the most conspicuous flowers 

 are adapted to cross-fertilisation by in- 

 sects, whilst other species of the same 

 genus, possessing less conspicuous flowers, 

 have recourse to self- fertilisation, in case 

 the visits of insects are wanting. This 

 statement is also confirmed by the species 

 of Gentiana in question. For in G. verna 

 and bavarica, differing from nivalis not eo~% 

 only by the considerably larger size of 

 the separate flowers, but also by a num- 

 ber of flowers which stand close together, 

 thus being easily seen from a great dis- 

 tance, the possibility of self-fertilisation is excluded by 

 the position of the stigma and the anthers (as shown by 

 Fig. 106 and no) ; whereas in Gentiana nivalis, whose 

 flowers are much smaller and more distant one from 

 another, one or some of the anthers commonly come into 

 contact with the margin of the stigma (as shown by Fig. 

 1 14), and efifect self-fertilisation in case cross-fertilisation \ 

 by insects is wanting. 



' Kerner, " Schutzmittel der BlUthen gegen unberufene Gaste." (Wien 

 i876)j 



Fio. 114. 



