750 RELATION OF STRUCTUKE TO FUNCTION IN ROOTS. 



fresh building-materials as those with which it was provided Viy the parent laecome 

 exhausted. The tissues of the young seedling always contain cells for the absorp- 

 tion of the dissolved food-salts and gases; these enter immediately into a close 

 union with the substratum, whether it consists of inorganic earth, of decaying 

 organic matter, or of a living host-plant. 



There are plants in whose seeds no differentiation into various parts, no 

 separation into embryo and food-stores, can be recognized, and in the seeds of 

 many thousand species we cannot even distinguish an embryo with cotyledons, in 

 which case the whole of the group of cells forming the seed must be regarded as 

 embryo. This group of cells first grows up, at the expense of its own materials, 

 into a structure having the form of a small tubercle, which on one side joins with 

 the substratum by absorbent cells, and on the other sends out a shoot, but develops 

 no system of tissues which could be called a root. This occurs, for example, in 

 Monotropa and in the Coral-root {Gorallorliiza), described on p. Ill, which are 

 usually termed rootless plants. In other examples of this group, in which the 

 undifferentiated embryo grows up directly into a small tubercle or stem, warts, 

 papillte, pegs and vermiform structures, equipped with absorbent cells, develop 

 on this tubercle, and join with the substratum; these are, therefore, of the 

 nature of roots. These structures always originate in gi-eat numbers from the 

 tubercle, i.e. from the enlarged developing embryo ; in many orchids living 

 epiphytically on the bark of trees they are formed on the side turned towai'ds 

 the tree; in parasitic Orobanchacese around the thickened lower end of the 

 tissue-body (c/. figs. 34" and 34^^ on p. 173), and in species of Cuscuta and 

 Cassytha laterally on the thread-like embryo where it has attached itself to a 

 host-plant. 



In plants whose seed contains an embryo differentiated into stem and leaf, 

 only a single, wart-like or conical body arises at one end of the hypocotyl, opposite 

 the bud of the epicotyl; it grows at germination into a cylindrical root provided 

 with absorbent cells, and later appears as a straight, downwardly-directed con- 

 tinuation of the hypocotyl. 



Neither the abundant roots proceeding from the undifferentiated embryo, nor 

 still less the single root springing from the membered embryo, suffice for the 

 requirements of the shoot arising from it. In proportion as this increases in size, 

 forming one internode above another, developing leaves with buds in their axils 

 which grow out into lateral shoots, the need of water and food-salts becomes 

 greater and greater. Fresh sources must be obtained for these materials, and 

 new conducting mechanisms must be established — in a word, new roots must be 

 formed. When only a single primary root is present in the embryo, the new 

 roots frequently spring from this as lateral branches, and it is customary to say 

 that the primary or main root has become branched, that it has formed lateral 

 roots. Of course each branch can again ramify, and indeed the branching is often 

 repeated beyond measure. The branched root (radix ratnosa) is to be seen espe- 

 cially in annual land-plants with erect leafy stems. Almost as often it happens 



