Structure of Plants. 197 



has influenced their shape and even note a sort of loseness and 

 wrinkling at the back which is due to their opening out. In P. 

 obconica the broad green funnel of the calyx is utterly unlike 

 that of most primulas, but again, looking at a young bud, one sees 

 that the sepals when young are small triangular flaps which fold 

 in and meet. This species, as every one knows, possesses an 

 irritating fluid distributed in small glands, which may cause a 

 dangerous eruption of the skin. If any minute insect dared to 

 enter the somewhat loosely fitting sepals it would surely be 

 shrivelled up. In the primrose and polyanthus the sepals are long 

 and ha\'e very hairy projecting mid-ribs, whilst the part between 

 the ribs is scarcely hairy. In the bud one sees that these hairy 

 ribs are prominent and their hairs touch or are entangled so as to 

 cover the intermediate tissue. There is a complex little system of 

 packing in the tips of the sepals which will at once be realised if 

 you examine them. The petals of primulas show a remarkable 

 adherence to one uniform plan, but there is a certain amount of 

 variety in the tips of the lobes. They are deeply cut in P. 

 obconica, distinctly lobed in the primrose, and show considerable 

 variation in both polyanthus and the auricula. Sometimes they 

 are scarcely cut at all, but in other cases there is a distinct lobe. 

 Both polyanthus and auricula are florist's flowers and show great 

 variation in the petals. When one examines the bud of P. 

 obconica, one finds that the two lobes of each petal are folded 

 inwards. All these intumed petal flaps can be seen if one looks 

 at a young bud. It is obviously careless about the opening up of 

 its flower, for it relies on its formic acid. In the primrose the 

 petal lobes are rolled round one another round a vertical axis, and 

 the little nick of the petal edge is clearly a necessity. In the 

 auricula and polyanthus they are both rolled in towards the centre 

 and rolled round a central axis in a complicated way. The petals 

 in each case seems to start as a concave half-dome and the tube is 

 scarcely visible in the youngest flowers. As they develop these 

 various modifications appear. It would not be ad\isable for me 

 to go further into the details, for I would require to take expert 

 advice from a master tailor or a dressmaker as to the effect of 

 scallops in packing or curving soft tissues. Yet it is clear that 

 these differences in the mature petal do depend upon the bud con- 

 dition. But I will take another case. The seeds of snowdrops 



