1902] 



NOTES ON THE PHYLOGENY OF LIRIODENDRON 



55 



of Liriodendropsis angustifolia to secure a form similar to Lirioden- 

 droit alatuniy and by a slight shortening and broadening of the 

 blade in the latter we obtain a form from which it is an easy 

 transition to Liriophyllum popiiloides, through forms with slightly 

 more ascending apical lobes, and with a greater development of 

 the basal portion of the blade 



in a lateral direction. 



I 



would consider LiriopJiylhim 

 ob cor datum as c o - 1 a t e r a 1 1 v 

 descended from some simple 

 ovate form such as some of 

 the forms referred to Liriode?t~ 

 dron primaevum. In either 

 case, a Icngtheninjr of the 



E.WB.cJUX. 



next to the last pair of pri- 

 maries, and the resulting 

 lobes, would give a form 

 identical with Liriophyllum 

 ob cor datum. However, we do 

 not consider this a true 

 species of Liriophyllum ; it 

 IS either a young and ano 

 lous Liriodendron leaf, or a '-">"■/-"•, 

 leaflet of some species of Leguminosae, preferably the former. 

 Linophyllum populoides, the type of the genus, as the winged 

 argins of the petiole show, had advanced several stages in the 

 development of stipules. It is undoubtedly related to the ovate 

 simple forms of the ancestral leaf through a form similar to 

 Liriodc?idron alatum, as was just pointed out, possessing, as it 



3 



ma- 



Certain forms of leaves of Liriodendron 



Tulipifera. 



m 



does, the winged petioles of the latter. Lesquereux's two figures 

 of Liriophyllum populoides have the petioles broken off, but both 

 show the petiole widening as it descends. In the figure of a 

 young leaf, the wings are in juxtaposition with the leaf blade, 

 while in the older leaf there is an interval of petiole, apparently 

 shownig that the wings grew down the petiole as it lengthened. 



Th 



e immature leaf furnishes additional proof of its origin from 



J 



