I 



24 BOTANICAL GAZETTE [jxjly 



■plant persists. In fig. j a root-shoot is represented which at first 

 glance may suggest a *^ slender tuberiferous rootstock" with a basal 

 stem-portion, subterranean and covered with scale-like leaves. In 

 fig. 4, however, a larger tuberous body is seen, from which several 

 stems have developed, whose irregular position does not indicate 

 that they have arisen from the axils of opposite leaves. As a matter 

 of fact, there are neither leaves nor leaf scars to be observed upon 

 this tuberous organ. The ^^ rhizome'' represented in fig, 4 is the one 

 most frequently found in flowering or fruiting specimens. A closer 

 examination shows that the tuberous organs are not tubers, but tuber- 

 ous roots. Moreover, in fig. j the small tuberous body is borne upon 

 a slender branch, which is seen to be a secondary root and not a stolon, 

 since it bears no leaves. If these organs were stems, they would have 

 been provided with opposite leaves like those of the aerial shoots, 

 and the shoots would necessarily have occupied a much more definite 

 position than those shown in fig. 4. Moreover, the internal structure 

 of these organs proved to be that of a true root. This is the only type 

 of *^ rhizome'' that I have found in the very many flowering and fruit- 

 ing specimens examined. 



The development of root-shoots is a phenomenon very frequently 

 observ^ed in our herbaceous and woody plants, but thus far we know 

 a very limited number of plants that persist only as root-shoots. It 

 was very desirable, therefore, to study the younger stages and especially 

 the seedlings of Rhexia. On account of their diminutive size they 

 are difficult to find in nature, since they are always more or less 

 hidden by the rank vegetation that covers the swamps. However, 

 I succeeded in detecting some small patches of young plants among 

 which there were many seedlings. One of these is shown in fig. i, 

 which represents the complete root system, the very small epigeic 

 cotyledons {Cot)^ and the basal internodes of the aerial shoot. At 

 this stage the primary root {R) is the main root of the young shoot; 

 it bears several lateral and very thin branches, two of which are oppo- 

 site and developed beneath the cotyledons. These two roots com- 

 mence to swell, and during the summer they show a local thickening 

 in the shape of a fusiform tuber (r). A few secondary roots sometimes 

 develop from the first intemode a short distance above the cotyledons. 

 At the end of the first season the shoot dies down to the ground, and 



1 



