MAIN RESEARCHES OF PLATEAU. 165 



It is possible, it is even probable that several of my experiments have been defective ; 

 I believe, however, that it is impossible that all of these observations and experiments, 

 made during five successive summers, are absolutely bad. A single consideration 

 diminishes my regret at not being able to carry them further, namely, that the present 

 work contains a number of documents which it will be neceessary to take into account in 

 writing the history of Macroglossa stellatarum." 



Entomophilous flowers little visited by insects. — The final memoir 

 by Plateau (1910:3) was devoted to experiments with entomophilous flow- 

 ers with bright colors but without nectar. The investigations were made 

 in the spring and summer of 1907, 1908, and 1909, and comprised 55 ex- 

 periments upon a dozen species. The substances employed to render the 

 flowers attractive were anisette, composed of very dilute alcohol, sugar 

 sirup, and a small amount of essence of anise, brown-sugar sirup with rum, 

 sugary juice of cooked cherries, and sugar sirup with a decoction of An- 

 gelica. The first two species employed were Fumaria officinalis and Poly- 

 gonum convolvulus, both with very small inconspicuous flowers. When 

 anisette was placed upon these by means of a brush, insects were immedi- 

 ately attracted to them, the visitors being chiefly flies. The experiments 

 dealt chiefly with large flowers of a vivid color, which are normally little 

 or not at all visited by insects. When Lilium candidum was supplied with 

 anisette or brown-sugar sirup, visitors appeared almost at once, the total 

 number reaching 28 in an hour, of which 25 were honey-bees. As a check, 

 flowers were supplied with sirup alone, but these were visited by only a 

 small number of flies, no bees noticing them. A single flower of Passifiora 

 with the corolla moistened with anisette received 34 visits by flies, while 

 another was supplied with brown-sugar sirup with rum a few days later 

 and yielded 40 visits in an hour, of which 1 was by Apis, 2 by Bombus, 

 and several by Vespa. Five experiments were made with Oenothera speci- 

 osa, the first one with anisette showing visits by Halictus and Prosopis 

 chiefly, while the second, on the afternoon of the same day, yielded 20 

 visits, 19 by Apis and 1 by Prosopis. Similarly treated, Pisum sativum 

 received 30 visits, chiefly of Bombus and Megachile, and Linum perenne, 

 15 visits, 11 of which were by flies, 3 by bees, and 1 by a hawk-moth. 

 The juice of cooked cherries was put in the flowers of 3 umbels of Pelar- 

 gonium zonale and anisette in those of another two. While the natural clusters 

 were not visited at all, those treated received a number of visitors, practi- 

 cally all flies. Clematis jackmanni perfumed with anisette gave 20 visits 

 in an hour, of which 14 were made by Bombus, and Convolvulus sepium 

 received numerous visits from 4 genera of Hymenoptera, 5 of Diptera, and 

 1 of Lepidoptera. In the case of Petunia hybrida provided with cherry 

 juice, 40 honey-bees came during a period of an hour, and visits continued 

 actively for a day or more, even to flowers completely faded, as long as 

 the juice was available. In the second experiment, cherry juice was placed 

 in the flowers of a group situated at a distance of 5 meters from the first 

 one, but no visits occurred for more than 4 hours, and the number then was 

 much smaller than in the previous case. 



