76 



rjJINOIPLES OF PLANT-T KRATOLOGrY. 



In some cases roots may be caused to form without 

 the aid of mutilation, as in the stem of MyriophylJum 

 figured by Goebel, a portion of which was bent under 

 water, when roots developed abundantly from the 

 convex, but hardly at all from the concave side. 



In some shrubs, e.g. Ribes, Jasminum, Viburnum 

 Ovulus var. nanum, which form very dense growth, 

 inducing a tendency to suffocation, curious gall-like 

 structures arise immediately below most of the nodes, 

 which are due to the formation of a mass of adven- 

 titious roots ; these do not grow out into the air, 

 hence the gall-like appearance. 



(3) On Leaves. 



Adventitious roots are not infrequently produced in 

 the axils of the leaves of the water-cress (Nasturtium 

 officinale) under , normal conditions, doubtless induced 

 by the aquatic habitat. But they are in most if 

 not all cases merely the forerunners of adventitious 

 shoots and perhaps were attached to rudiments of 

 these which are still within the tissues of the leaf. 

 Mr. Howard sent a carnation from the Agricultural 

 College at Wye which had evidently been grown under 

 very moist conditions, as adventitious roots had been 

 emitted, not only all the way up the stem, but also all 

 along the lower side of the midrib of the leaves. 

 Irmisch mentions roots arising from the stalk of the 

 cotyledon in G a/rum Bulboeastanum, and Masters saw 

 the same phenomenon in the mango. 



Roots are most usually emitted by leaves as a result 

 of mutilation. If leaves of Bryophylhim calyeinum are 

 detached from the plant, each marginal notch may 

 produce a single root (fig. 21) ; this must be regarded 

 as the first root of the adventitious bud whose rudi- 

 ment has not as yet appeared outside the tissues of 

 the leaf.* Goebel figures cases of root-formation on 



* Cf. the mode of appearance of the first leaf in the lateral buds of Lyco- 

 jpodium Selago. 



