ON THE NATURE OF STIPULES 205 



well shown in species where the leaf-development is 

 basipetal that is to say, the upper leaflets are formed 

 first, and the lower ones in succession. Now, if stipules 

 were ordinary leaflets, they would in such cases be formed 

 last, while, as a matter of fact, they make their appear- 

 ance with, or even before, the first leaflets. 



A strong reason against regarding stipules as mere 

 subordinate structures dependent on the leaf-blade is 

 that they arise so early, often, indeed, before the leaf- 

 blade with which they are associated, as, for instance, in 

 the Pea (fig. 26, p. 23), Buchlandia (fig. 190, p. 122), &c. 

 A further reason against regarding stipules as mere de- 

 pendent parts of the leaf is, that while in many cases 

 leaves have no stipules, in others stipules take the place 

 of leaves using the word, that is to say, in the usual 

 sense. 



Thus, in Lathyrus Apliaca (fig. 304, p. 179) there are 

 no leaflets, the leaf-blade being replaced by a tendril, 

 and the ordinary function of the leaf-blade is performed 

 by two enlarged stipules. 



Again, in some species of Adesmia (fig. 311, p. 187), 

 a South American leguminous genus, leaves are present 

 only on the lower part of the long, straggling branches, 

 the upper part bearing stipules only, which gradually 

 pass into bracts. 



It seems, then, incorrect to regard stipules as mere 

 dependencies of the leaf-blade with which they are 

 associated. 



