764 DEFINITION OF THE ROOT. 



rounded, loosely-fitting cells, and on this outer side of the root-cap the cells are 

 also seen to be partially separated and torn off. As the outer cell-layers are rubbed 

 away by the advance of the root, and by unavoidable contact with the sur- 

 rounding soil, new cells are always being pushed forward from within, and in this 

 way the loss is made good, and the shield continually repaired. 



Obviously, aquatic roots do not require a shield of this kind at their apex, and 

 in aerial roots, at least in the form described, it would likewise be superfluous. 

 Even roots which penetrate into mud do not require it. Accordingly many water- 

 plants and the marsh-inhabiting mangroves do not develop a cap at their root-tip. 

 The root-cap is also entirely absent in parasitic plants which it would only hinder 

 from penetrating into the tissue of the host-plants. 



DEFINITION OF THE ROOT. 



In the preceding pages we have continually spoken of roots, although we have 

 not yet defined technically what a root is, and now, contrary to the usual custom 

 in scientific works, the definition of this organ has to come not at the beginning 

 but in the middle of the chapter. This alteration in position has been caused by 

 the necessity of establishing the definition on some peculiarities in the external 

 and internal structure of roots, with which we could not suppose all readers to be 

 familiar, and which therefore had to be described beforehand as far as required. 



But many readers will ask if any definition is required, if everyone does not 

 know without it what the root of a plant is, and how it can be distinguished from 

 a stem and leaves ? The case is exactly parallel with that of the leaf. Every one 

 who is not a botanist thinks he knows what is meant when he hears the word 

 "leaf", and cannot conceal his astonishment or possibly his smile when he is 

 informed that scientific men are not agreed about such a simple question, 

 and that they write violent polemics upon this question. To the impartial reader 

 debates as to whether a certain part of a plant is to be regarded as a root or not 

 doubtless appear hypercritical and a pedantic splitting of hairs, and with regard to 

 many of the discussions I would hardly venture to deny the justice of his position. 

 The savant who constructs for himself the picture of an ideal or primitive plant 

 from a sometimes larger, sometimes smaller number of single observations, who 

 finds out how the individual parts were situated in their succession as to time, 

 and in their mutual relations in space, and who distinguishes and defines the 

 various parts accordingly, is indeed very easily tempted to take the abstract ideal 

 he has created as a standard for the whole vegetable kingdom. From his point 

 of view, obtained by the consideration and comparison of so many individual cases, 

 all forms are arranged and explained, everything must fit into the now firmly 

 established groundwork, and where it will not coincide, he talks of exceptions, 

 forgetting that in such a case exceptions are not permissible, but are rather a proof 

 of inadequate generalization from the single cases observed. 



In the comprehension of the results of general comparative studies of this kind 



