THE THREE FOHMS OF TA-THBUM SALICAEIA. 177 



stigma of each form pollen from the stamens of corresponding 

 length ; and we shall presently see the importance of this adapta- 

 tion. It must not, however, be supposed that the bees do not 

 get more or loss dusted aU over with the several kinds of poUen ; 

 they certainly do, as could be seen with the green pollen from 

 the longest stamens. Moreover, a case wiU presently be given of 

 a long-styled plant which grew absolutely by itself, and produced 

 an abundance of capsules, which must have been fertilized by its 

 own two kinds of pollen ; but these capsules contained a very 

 poor average of seed. Hence insects, and chieily bees, act both 

 as general carriers of pollen, and as special carriers of the right 

 kind *. 



Variability. — Before passing on to more important topics, I 

 must say a few words on this head. Wirtgcn remarks t on the 

 variability in the branching of the stem, .in the length of the 

 bractcaa, size of the petals, and in several other respects. The 

 plants now growing in my garden have their leaves arranged op- 

 positely, alternately, and in whorls of three, and differ greatly in 

 shape. The stems of the plants bearing leaves in whorls are 

 hexagonal ; those of the other plants are quadrangular. But we 

 are concerned only with the reproductive organs : the upward 

 bending of the pistil is variable, and in a remarkable degree jn 

 the short-styled form, in which it is sometimes straight, Tsome- 

 times slightly curved, but generally upturned at right angles. 

 The stigma of the long-styled pistil frequently has longer papillaa 

 or is rougher than that of the mid-styled, and this than that of 

 the short-styled form ; but this character, though fixed and uni- 

 form with the two forms of Primula, is here variable, and I have 



* In my paper ou the two forms of Frimula (Journal Proc. Linn. Soc. 1862, 

 p. 85) I stated that I had only occasionally seen Immble-bes sucking the flowers 

 of the Cowshp (P. veris). Since then I have had some beds in my garden 

 eontauiing nearly 700 plants, and these were incessantly visited by Bomlua 

 hortornm and S. muscormn. I caught some of these bees, and I found (as I 

 had anticipated in my paper, p. 86) that a vast majority of the poUen-grains 

 which adhered to the base of the proboscis were large-sized and had come from 

 the long stamens of the short-styled form, and were thus placed ready to ferti- 

 lize the stigma of the long-styled form. On the other hand, on the middle, and 

 near tlie tip of the proboscis, a very large proportion of the pollen -grams were 

 of the small size, and had come from the short stamens of the long-styled form. 

 My son caught, also, a moth (CuciilUa wriasci) hovering over the bed, and I 

 found on its proboscis a similar distribution of the two kinds of pollen-grains. 

 I give these facts as a further illustration of the importance of the relative 

 lengths of the stamens and pistil. 



t Vei-hand. des iiaturiiiet. Vereins, 5. Jahrgang, 1848, S. 11^ 13. 



p2 



