698 SECONDARY GROWTH IN THICKNESS 



succeeded by a third cambial layer, which originates at the outer limit 

 of the phloem produced by its predecessor. This process ' may be 

 repeated several times. The result is the formation of concentric zones 

 composed alternately of xylem and phloem, the latter constituting the 

 bulk of the storage-tissue. According to Trecul, the roots of Myrrhis 

 odorata likewise develop several additional cambial layers, which, how- 

 ever, originate on the inner side of the primary cambium, and cut off 

 xylem and phloem in inverse order. 



The most complicated variety of this type of anomalous secondary 

 thickening is found in certain Convolvulaceae, and in the genus 

 Rumcx, where partial or complete additional cambial cylinders arise 

 successively, sometimes within and sometimes outside the primary 

 cambium, all forming xylem and phloem quite independently of one 

 another. It is quite unnecessary to discuss these highly complicated 

 cases in detail/" 



Attention has already been drawn to the fact that such peculiarities 

 of secondary growth do not merely serve to ensure the production of the 

 requisite amount of storage-tissue ; another object of these " anomalous " 

 processes is the thorough inter-penetration of storing and conducting 

 tissues, an arrangement which renders the deposition and renewal of 

 reserve-materials easier and more effective. The subsequent differentia- 

 tion of (tertiary) leptome-strands within the secondary xylem cannot 

 be explained from any other point of view. 



IV NON- ADAPTIVE ANOMALIES OF SECONDARY 

 GROWTH IN THICKNESS. 



In conclusion, a few words must be devoted to those anomalous 

 modes of secondary thickening which appear to be merely cases of 

 variation of design. The anomalies in question occur more parti- 

 cularly in certain herbs and small shrubs belonging to the Cheno- 

 podiaceae, Amarantaceae, Nyctaginaceae, and Tetragonieae, and 

 in species of Mesembryanthemum and Phytolacca?'* That we are really 

 dealing with non-adaptive features, in such cases, follows at once from 

 the circumstance that the abnormal features may vary considerably, not 

 only within the limits of a single family, but even in different species 

 of the same genus, quite independently of any differences in the 

 ecological relations of the plants concerned. 



The secondary growth in thickness of the Chenopodiaceae and 

 of the above-mentioned allied families has been investigated by Morot, 

 Herail, and Leisering. In these plants one or more cambial layers 

 arise in the pericycle that is, outside the circle of primary leaf-trace 

 bundles ; these extra-fascicular meristematic layers, which cut off 



