JulyZ, 1875J 



NATURE 



191 



flower has just opened, only a single very small open- 

 ing is commonly left free (<?, Fig. 68) ; somewhat later, 

 when the longer anthers have advanced a Httle further, 

 two small openings are frequently obvious {p 0, Fig. 69), 

 by which Lepidoptera can insert their proboscis. The 

 exclusion, however, of all other insects from the honey 

 would be useless or even fatal to this, as well as to the 

 above mentioned flowers, unless by particular contri- 

 vances, ( I ) increased frequency of the visits of Lepidoptera, 



iW \ 



Fig. 69.— The same, .at a somewhat later period. 



and (2) certain cross-fertilisation by them were effected. 

 Hesperis trisiis, by the very inconspicuous colour of its 

 flowers, which are yellow reticulated with purplish streaks, 

 by opening them in the afternoon, and by having no smell 

 in the daytime whilst very fragrant towards the evening, 

 proves to be adapted exclusively to crepuscular and noc- 

 turnal Lepidoptera, which, attracted from afar by the 

 sweet odour, are induced to pay frequent visits. The 

 base of each of the two shorter filaments is surrounded by 

 a greenish swelling (//, Figs. 66, 67), which secretes on its 

 inside honey so copiously that it rises in the interstice 

 between the shorter and the two adjacent longer filaments. 

 Cross-fertilisation by the visits of moths is secured in the 

 following manner. From the one or two small openings (<?, 

 Figs. 68, 69) the proboscis of the moth is guided down- 

 wards by the longer filaments as in a channel, first along 

 one side of the stigma {st, Fig. 66), which has bent down- 

 wards on both sides just into the way of the proboscis, then 



Fig. 70.— The same in its last state. 

 (Figs. 66-70 are seven times magnified.) 



along the shorter anther («', Fig. 66), which from the 

 other side has turned its pollen-covered front likewise 

 exactly into the way of the proboscis, until at last it 

 reaches the honey {h, Fig. 66) ; the proboscis afterwards 

 wetted with honey at its tip, when retracted, first touches 

 again the anther a' with one sid«, which is thus charged 

 with pollen, then with the other side the stigma, which thus 

 escapes fertilisation with its own pollen, and when in the 



next visited flower the tip of the proboscis with its pollen- 

 charged side touches i^the stigma, cross-fertilisation is 

 effected. 



My daughter Agnes, perseveringly watching Hesperis 

 tristis during several mild evenings in the month of May, 

 has succeeded in observing and catching the following 

 fertihsers of it : — (i) Plusia gamma, frequently (length of 

 the proboscis 16-18 mm.); (2) Hadena sp. (11 mm.); 

 (3) Dtatttha'cia conspersa, W.V., twice (13 mm.) ; (4) 

 lodis lactearia, L. ; (5) Botys forhcalis, L., three times. 



But although in calm and warm evenings, as is proved 

 by these observations, cross-fertilisation may be suffi- 

 ciently effected ; yet in unfavourable weather all flowers 

 of many individuals develop and fade without experiencing 

 any visit of fertilisers. In this case, nevertheless, almost 

 every ovary develops and brings to maturity its seed- 

 vessels, self-fertilisation being regularly effected by the 

 pistil growing and the stigma coming into contact with 

 pollen-grains of the four longer anthers. 



Thus, in these flowers the four longer anthers have 

 apparently no other function in the first period of flowering 

 but to exclude incompetent visitors from the honey, by 

 stopping the entrance of the flower, and, by the direction 

 of their filaments, to keep the proboscis of the fertilisers in 

 the right direction, whilst in a later period, in case visits 

 of moths have been wanting, they regularly effect self- 

 fertilisation. The two shorter anthers, on the contrary, are 

 exclusively adapted to cross-fertilisation by visiting moths. 



Lippstadt Hermann Muller 



JOSEPH WINLOCK 



THE following details concerning the late Prof. 

 Winlock, whose death we announced last week, we 

 take from the New \ork Nation : — 



Prof. Joseph Winlock, Director of the Observatory of 

 Harvard College, died suddenly after a brief illness last 

 Friday morning, June 11, at the age of forty-nine. One 

 of the foremost of American astronomers, whose honour- 

 able career in science began thirty years ago, who has 

 filled with great credit several important positions of 

 scientific labour and trust, is thus cut off in the midst of a 

 life whose usefulness cannot be estimated by ordinary 

 standards. Well known and highly estimated by all active 

 collaborators in astronomy both at home and abroad, he 

 was never so well known to others or to the public as his 

 important services deserved. This was chiefly on account 

 of a modest shrinking from any candidacy for honours, 

 amounting almost to an aversion from them, and an in- 

 difference to an uncritical or merely popular reputation. 

 Immediately upon graduating from Shelby College, 

 Kentucky, in 1845, he was appointed Professor of Mathe- 

 matics and Astronomy in that College, where he re- 

 mained until 1852, when he removed to Cambridge, Mass., 

 and took part in the computations of the Americdn 

 Ephemcris atid Nautical Almanac, then under the 

 superintendence of Admiral C. H. Davis. In 1857 he 

 was appointed Professor of Mathematics of the United 

 States Navy, and in that capacity served in succession as 

 Assistant at the Naval Observatory at Washington, as 

 Superintendent of the Nautical Almanac, and as Director 

 of the Mathematical Department of the Naval Academy 

 at Annapolis, Md. On the breaking out of the war, in 

 1 86 r, he was a second time made Superintendent of the 

 Nautical Almanac. His next service to astronomy was 

 in the position of Director of the Observatory at Harvard 

 College, and Phillips Professor of Astronomy, to which 

 he was appointed in 1865— a position already made 

 highly honourable by the labours of his predecessors, the 

 distinguished astronomers. Professors W. C. Bond and 

 G. P. Bond. He has also served at the same time as 

 Professor of Geodesy in the Mining School of Harvard 

 College. Only a few months ago, Mr. Bristow appointed 

 him the chairman of the Congressional Commission for 



