1907] CHR YSLER—PO TA MOGETONA CEA E 181 



studies. Since the correctness of this view is to be premised in the 

 following discussion, it will be well to review the evidence for it. The 

 two characteristic monocotyledonous features — scattered arrange- 

 ment of the bundles and amphivasal form of bundle — are found not 

 to occur in the seedling in families which have been examined in this 



(J 



On the contrary, 



the seedling shows a tubular stele with collateral bundles, such as is 

 typical of a dicotyledon. Sooner or later some of the bundles turn 

 into the medulla and show the amphivasal feature at some part of 

 their course, eventually becoming leaf traces. These stages in onto- 

 geny are believed to repeat stages in the phylogeny of the group. 

 Occasional cases of apparent reversion to the primitive type occur in 

 adult rhizomes, such as that of Clintonia. The tubular stele is found 

 in the floral axis of some monocotyledons, and its occurrence here has 

 been interpreted as the persistence of an ancestral feature. In this 

 organ and in the leaves, moreover, only collateral bundles are found, 

 which is the prevailing type in dicotyledons. A cambium, the lack of 



F 



w^hich was supposed to be a distinguishing feature of monocotyledons, 

 has been demonstrated in Gloriosa by Queva (20), in the seedlings of 

 a number of monocotyledons by Andersson (i), in Cyperaceae by 

 Plowman (19), and most recently in the nodes of grasses by the 

 present writer (6). It has been argued that in the last-mentioned 

 case the cambium has persisted in regions where it is of use, and is 

 therefore a vestigial not a rudimentary structure. As to geological 

 evidence, many of the supposedly ancient monocotyledons have 

 turned out to be gymnosperms, and there is no reason for believing 

 that monocotyledons existed earlier than dicotyledons. Finally the 

 researches of Miss Sargant have shown how a monocotyledonous 

 embryo may have been derived from a dicotyledonous one. 



The present research has added to the evidence along at least one 

 of the lines just mentioned. It has been shown that in Triglochin 

 and various species of Potamogeton the floral axis contains a circle 

 of collateral bundles, in which respect it is sharply distinguished from 

 the leafy shoot, especially in Triglochin. It has also been pointed 

 out that in the creeping stem of Potamogeton, where no large leaves 

 send in traces, the central cylinder has a simple tubular form, though 

 this appearance is often masked by the large size of the dorsal and 





