614 COTYLEDONS. 



The external form of the seed and the position which it consequently assumes 

 on falling to the ground is by no means an unimportant item in this connection. 

 If the seed comes to lie so that the axis of the hypocotyl is perpendicular to the 

 surface and the tip of the radicle is directed downwards, we seem at first sight to 

 have a very favourable position for germination; but it is not so in reality. In 

 this position the hypocotyl would have to perform the most complicated curves in 

 order to be able to withdraw the cotyledons from the seed. On the other hand, 

 the most favourable condition is obtained when the axis of the hypocotyl and 

 radicle lie parallel to the surface of the ground, as, for example, in the gourd 

 seeds illustrated in fig. 145 \ Here the radicle immediately after leaving the seed- 

 coat can bend down at a right angle and grow into the earth, and the hypocotyl 

 is able to withdraw the cotyledons very rapidly. When seeds of this sort are 

 sown, they usually assume the last -mentioned position. Flat or compressed 

 seeds lie with their broad surface on the ground; oval and elongated cylindrical 

 seeds fall so that their longer axis is parallel to the substratum; whilst in 

 spherical seeds the centre of gravity is so situated that the most favourable posi- 

 tion possible for germination is obtained. 



The importance of numerous developments on the exterior of the seed- 

 coat or pericarp will at once become evident to anyone who observes attentively 

 the process of withdrawal of cotyledons. It is manifest that the withdrawal 

 only occurs without delay when the seed is in some way or other firmly fixed 

 and when arrangements are present which prevent a favourable position being 

 lost when once assumed. This would not be so were the seed the plaything of 

 every gust of wind or current of water. Equipments for retaining fruits and 

 seeds in the position of germination occur in great number and variety. Even 

 the wing-like and hairy appendages, the curved, pointed, and barbed processes, 

 and the various adhesive arrangements of fruits and seeds, which in the first 

 instance have the function of agents for distributing the fruits, often afford 

 this advantage, viz. that by their means the seeds are fixed where germination 

 can successfully take place. If we look at the damp mud by a river's bank, 

 towards the end of May, when the fluffy seeds of willows and poplars are 

 escaping from the dehisced fruit-capsules and are carried along by the wind, 

 we there see countless numbers of these seeds sticking by their hairs to the 

 mud so tightly that they cannot readily be displaced. All such seeds (differing 

 from the generality of seeds) germinate in a few days, while seed lying on the 

 ground in loose flakes close by do not germinate. The hairy coat which first 

 served as an agent for distributing the seed, now functions as an agent for 

 fixing it in the germinating bed. This also applies to the tufts of silk 

 adorning the email seeds of tropical tillandsias (Tillandsia usneoides and 

 T. recurva) which grow as epiphytes on the bark of trees. These first serve 

 as wings, and the tiny seeds are carried by the wind far away from the 

 burst fruit-capsules. If these seeds are stranded on the bark of a tree- 

 trunk which is swept by the wind, the hairs cling firmly and bring the seeds 



