494 Berry: Origin of Stipules in Liriodendron 





modern tree represent the final stage in development from the 

 basilar leaf-lobes of the Cretaceous species, is amply proven, although 

 questioned in some quarters.* Still another instance where stipules 

 are mentioned as developing from leaf-lobes is implied in the fol- 

 lowing footnote on page 420 of The Smilaceae of North and Cen- 

 tral America. In speaking of those species which climb by means 

 of tendrils growing from the stipular wings on each side of the 

 petiole, Dr. Morong says : " De Candolle regards this appendage 

 as more in the nature of a modified leaf segment or leaflet than a 

 stipule, but it seems to me that a stipule is nothing else' than a leaf- 

 let at the base of the petiole." f The only other mention of winged 

 petioles in Liriodendron so far as I am aware, is a short communi- 

 cation to Meehan's Monthly for August, 1896 (p. 145), from Mrs. 



W. A. Kellerman recording her observation of several forms with 

 adnate stipules. 



The writer has been fortunate in securing- a large series of 



t> "■ ""b 



specimens of Liriodendron Tidipifera Linn, with winged petioles 



and adnate stipules in all stages of development, all of which 



strikingly confirm Dr. Hollick's theory of their origin. We 



have noted them briefly in the following pages in what seem to 



be the natural stages of their development as exhibited by our 

 material. 



We are greatly indebted to Mrs. W. A. Kellerman," of Co- 

 lumbus, Ohio, for many fine specimens. The first suggestion of 

 basilar lobes is found in an obtusely lobed leaf of the normal size 

 from a sapling. The leaf-blade is extended on each side at the 

 base to form more or less well-defined lobes nearly the- size of an 

 ordinary stipule, and strikingly like them in venation, attract- 

 ing attention at a glance ; this is particularly true of the right- 

 hand lobe which is entirely stipular in venation, quite different 

 from the usual venation of the basilar portion of ordinary leaves. 

 These lobes are well shown in Plate 41, Fig. 1. This form un- 

 doubtedly represents the first stage the ancestral leaf, or at least 

 its basal portion, assumed in the course of stipular development. 



% \ / i 1 A ft a _ 



We 



s 



v.- 



Ward's conclusions are rejected by A. G. Nathorst, Rev. of Types of Laramei 



Flora by L. F. Ward, in Neues Jahrbuch fiir Mineralogie, 2: 219-222. 1893. 

 t Morong, Bull. Torr. Club, 21: 420, footnote. 1894. 



