AND CROSS-BREEDING. 143 



connected with these arrangements is, that closely-allied species 

 of the same genus exhibit sometimes exactly opposite pecu- 

 liarities in this respect ; and it is even uncertain whether the 

 same species does not vary under different conditions. A 

 very interesting account of the phenomena presented by a 

 number of plants of the Pea tribe belonging to the natural 

 order Leguminosae, by Mr T. H. Fairer, will be found in 

 ' Nature,' vol. vi. We may give a single very good example of 

 this in the two common Mallows. In the large Mallow (Malva 

 sylvestris) the stamens are collected together into a bundle 

 completely surrounding and overtopping the pistil. At a later 

 stage the empty anthers are bent down out of the way of the 

 stigmas, which are even yet not in a receptive condition. 

 Spontaneous self-fertilisation is in this case scarcely possible. 

 In the smaller species (Malva rotundifolia) the structure is the 

 same up to a certain point, but the stigmas mature earlier, and 

 when in this condition coil themselves among the anthers, 

 there being still sufficient pollen left in the anthers to insure 

 the self-fertilisation of the flower. The two species often grow 

 intermixed ; both are scentless ; insects are, however, abun- 

 dantly attracted by the large showy flowers of M. sylvestris^ 

 which are also beautifully streaked, the streaks all pointing 

 towards the nectar-glands, at the base of the tube formed by 

 the filaments. The flowers of M. rotimdifolia are much 

 smaller and of paler colour and are not streaked, and hence 

 not so attractive to insects. Dr Miiller records thirty-one 

 species of insects, chiefly Hymenoptera, which he detected 

 visiting the former, whilst only four were observed to frequent 

 the latter species. Dianthus deltoides, the " Maiden Pink," 

 is scentless ; but each of the five petals is provided with a 

 number of purple spots, which seem to indicate to the butter- 

 flies, by which they are chiefly visited, the exact place wherein 

 to insert their proboscis in order to reach the honey glands. 

 The anther, at this time discharging pollen, is placed imme- 

 diately over each petal, and the butterfly cannot fail to carry 

 off some of the dust on its head. A second inner row of five 

 stamens, at this period completely concealed within the tube of 

 the corolla, do not mature till later \ and it is only after all the 

 anthers have dropped off that the two stigmas, previously coiled 

 round one another, separate and develop the hairs which 

 serve for the detention of the pollen. While the various con- 

 trivances connected with the arrangements of the male and 

 female organs have been more or less known to botanists for 

 three-quarters of a century, very little attention has been paid, 



