24 COLIN CLOUT'S CALENDAR. 



more perfect fertilisation, and to make the pollen adhere 

 more certainly to the visiting bees than in other flowers. 

 Outside all we get the six blue petals, three of which are 

 really calyx pieces, indistinguishable in colour and shape 

 from the true petals, but recognisable as to their real 

 nature by two signs first, that they slightly overlap the 

 others, and secondly that they have the long stamens 

 of the outer row opposite to them and combined with 

 them. In all the lilies the calyx pieces and petals are 

 very much alike and similarly coloured ; but in the wild 

 hyacinth the similarity is even closer than elsewhere. 

 This is doubtless due to the shape of the flower, which, 

 in order to accommodate its favourite bees, closely simu- 

 lates a true tubular blossom, like the Canterbury bell. 

 At first sight, indeed, one might almost take it for such 

 a perfect tube ; but when you pull it to pieces, you see 

 that the six apparent petals are really distinct, though 

 they converge so as practically to form a bell-flower, 

 with a tiny drop of honey glistening at its base. In the 

 true hyacinths of our gardens the six pieces have actually 

 coalesced into a solid and well-soldered tube, which marks 

 a still higher level of adaptation to insect visits : and 

 even our own wild species shows a slight tendency in 

 the same direction, for its pieces are often very shortly 

 united together at the bottom. It is from such small 

 beginnings as this that selective agency slowly produces 

 the greatest changes ; and perhaps after the lapse of 

 many ages our own wild hyacinths may become really 

 tubular too, under the modifying influence of insect 

 selection. But at present the frequent recurrence of 

 white varieties a probable reversion to some earlier 



