204 NATURAL SCIENCE. March. 



Aroideae, in which the flowers are frequently proterogynous and as 

 frequently proterandrous ; in Portulaca pilosa, the flowers of which open 

 only in the sunshine, and yet seed abundantly when grown in the 

 shade ; in Cuphea Zimpani (Lythraceae), Lopezia coronata (Onagraceae), in 

 several species ot Lonicera, Phytolacca decandva, Lycopevsicon escnlentinn, 

 Hamamelis virginica, &c. A large number of the American Compositae 

 he asserts to be self-pollinated. He has observed pollen-tubes 

 entering the clefts of the bilobed stigma before it opens; and the 

 pollen-grains, even when large and brightly coloured, frequently 

 fall on the stigma before any insect can possibly enter the flower. 

 Mr. Meehan, indeed, goes so far as to say that whenever a species is 

 unusually productive, he finds, as a rule, arrangements for self- 

 pollination. An Italian botanist, Dr. Terraciano, states that in 

 many species of Nigella, notwithstanding the conspicuousness of the 

 flower and the presence of nectaries, the structure is adapted for self- 

 pollination. According to Warming, all the Greenland and Iceland 

 species of Euphrasia are self-pollinated. Even so decided a Dar- 

 winian as Professor Caruel, of Florence, is inclined to doubt whether 

 too much has not been taken for granted in the prevalent theory of 

 the part played by the bright colour and sweet scent of flowers in 

 attracting insects, seeing that, as far as we know, the visual and 

 olfactory organs and perceptions of many animals are very different 

 from our own. In particular, he calls attention to Sir John Lubbock's 

 observations on the different effect of colours on many animals from 

 that which they produce on us ; to the eyes formed of facets and to 

 the ocelli of insects, and to their sensitiveness to the ultra-violet rays 

 of the solar spectrum. Rosen asserts that even with many wind- 

 pollinated plants, such as Carex and Fesiuca, self-pollination is the rule. 

 On the other hand, the orthodox view receives support from a 

 very large number of fresh observations which reveal adaptations 

 clearly intended to promote cross-pollination by the agency of insects. 

 In a series of papers in the Botanical Gazette and elsewhere, Mr. C. 

 Robertson describes a large number of observations to this effect on 

 American plants. Professor Halsted states that the flowers of the 

 barberry are very rarely, if ever, self-pollinated, and calls attention 

 to the remarkable irritability of the stamens of species of Portulaca, 

 which promotes the scattering of the pollen over the bodies of 

 insects visiting the flowers. The structure of the flowers of ^r«»«y 

 Dvacunculus, Helicodiceros, Arisama, Ainorphophallus, and other species 

 of Aroideae, has been made a special study by several Italian 

 botanists. The extraordinary simulation by the open flower of the 

 appearance and odour of decomposing flesh, appears specially 

 designed to deceive and attract necrophagous coleoptera and diptera, 

 which aid in the carriage of pollen. Schulz asserts that in the 

 family Sileneae of Caryophyllaceae the proterandry is in many cases 

 so marked as to render self-pollination impossible. Mr. G. F. Scott 

 Elliot has, in the Annals of Botany, a very interesting paper on the 



