Moore : The study of winter buds 123 



' number of leaves is packed away in the bud before the approach 

 of winter. 



m 



Miiller (1866-2) made an extensive study of Fraximts from 

 early spring to midsummer and found that the bud scales of the 

 next season's terminal bud began a rapid development in April. 

 The present study of Fraximts americana confirms MiJller's 

 observations on this species, for,^as will be seen by reference to 

 FIGURES 34 and 35, no addition of leaves occurs during the winter 

 months. The new additions, which take place after April 16, are 

 the scales whose development has been traced by Miiller, 



Philadelphus enters upon the winter rest with the buds buried 

 beneath the leaf scars. These buds were examined with difficulty, 

 not only on account of their small size but because o{ the simi- 

 larity between the outer coverings and the inner leaves. In the 

 stained sections of the buds, these structures are differentiated 

 and the leaves easily enumerated. Eight to ten whorls of true 

 leaves are generally present and this is the usual number that 

 unfolds. Figures 44 and 45 show median sections through a 

 bud. The young whorls of leaves are closely appressed until the 

 blades are quite well expanded. When the leaves are unfolded 

 in May, there appear between the last two leaves two small, yel- 

 lowish scale-like leaves which soon die away, evidently because 

 the meristematic point itself ceases to function and thus terminates 

 further growth at this point of the shoot. 



There is little question that in the one conifer studied, Picea 

 excelsa^ the buds contain through the winter the complete number 

 of rudimentary leaves. Kiister (1S99) ^^X^ ^^^ '^^^ Abies and in 

 Pinus the number o^ leaves in the bud is probably the same in the 

 fall as during the time of unfolding. The present study would 

 tend to confirm this view for Picca, since the earliest growth in the 

 spring indicates a production of tiny scales at the growing point 

 of the bud rather than the organization of more needle-like leaves. 



The yo\xx\% leaves in the Rhododendron buds correspond fre- 

 quently to the number of leaves and nodes present on shoots of 

 the preceding season. The irregularity in branching renders the 

 count difficult in growths of two successive seasons. The largest 

 number of leaves counted in any bud is twelve; the largest num- 

 ber o{ nodes for previous years, fourteen. Because of the more or 



