188 PART II. THE INTIMATE STRUCTURE OF PLANTS. [ 34. 



It will be understood that, in order to reach the surface, the lateral secondarj* 

 roots must penetrate the external tissues of the parent root. This is not 

 effected by purely mechanical means, but by chemical action, leading to solution 

 and absorption, exerted on the tissues, either by the rootlet itself, or, more 

 commonly, by a digestive sac which invests the rootlet, and is formed in Phanero- 

 gams by the growth and division of the cells of the endodermis (and sometimes 

 one or two layers of cortical cells), in Vascular Cryptogams, by the growth and 

 division of one or more of inner layers of cortical cells just external to the endo- 

 dermis, of the parent root. 



D. Development of Hairs. These structures are in all cases 

 developed from the superficial cells of the parent member, that is, 

 from dermatogen-cells in those parts in which this layer is differ- 

 entiated ; in the great majority of cases each hair arises from a 

 single superficial cell. Hairs are generally developed in acropetai 

 succession, but considerable irregularity is not uncommon, and 

 they are frequently developed on members in which the tissues 

 have already acquired their permanent characters (see p. 64). 



E. Development of Emergences. When exogenous they are 

 developed from the superficial and from one or more of the sub- 

 jacent layers of tissue of the parent member, that is, from the 

 dermatogen and periblem of those members in which this differen- 

 tiation of the primary meristem obtains. When they are endo- 

 genous {e.g. haustoria of Cuscuta, see p. 67), they are developed 

 exclusively from the periblem. 



The haustoria of these parasitic plants are frequently regarded as roots, either 

 normal or adventitious, especially in the exceptional case of endogenous origin ; 

 but this view is inadmissible, in the first place, because they are developed, not 

 from the pericycle, but from the cortex of the parent-member ; and in the 

 second, because, with the mode of origin, they are developed relatively late, 

 whereas, in accordance with the general rule (see p. 190), they ought, were 

 they roots, to take origin from more deeply seated tissues of the parent 

 member. 



E. Development of Reproductive Organs. The question as io the 

 relation of these members to the primary meristem, only arises 

 with reference to those plants, the bodies of which consist of 

 many layers of tissue ; their origin in plants, the bodies of which 

 consist either of filaments, or of flattened expansions of a single 

 layer of cells, need not be considered here. 



In the cases under consideration, the reproductive organs may 

 be developed either from the superficial layer alone, or from that 

 and one or more of the subjacent layers. 



