138 



baccus, and Rhus radicans, the last with four or five partial or 

 complete leaflets) or again on the older parts of the Japanese ivy 

 {Parthenocissus tricuspidatci). 



Setting aside these cases of the ash and other compound-leaved 

 plants, the explanation of the above-described phenomenon is to 

 be found in the behavior of the leaf fundaments in the buds. In 

 the case of the lilac the pair of youngest leaves is so disposed in 

 the bud that the ventral surfaces of the two are faced and that their 

 margins lie each against the other, and match exactly. As the 

 leaves enlarge their blades become thinner and so curved that 

 one leaf comes to be infolded by the other. It sometimes 

 happens however that during the time that blades begin to over- 

 lap, the margins, as the result of unequal growth or pressure 

 cross one another at one or more points, and further development 

 is retarded at those places. That this crossing is the cause of 

 the notching or lobing there can be no doubt, since I have found 

 a pair of leaves in which the notches were so deep that the 

 blades became too closely interlocked to be able, during their ex- 

 pansion from the bud, to separate. 



That other irregularities of form, also, such as a sudden nar- 

 rowing of the blade toward the apex, are due to similar causes, is 

 apparent from the circumstance that these, too, are often asym- 

 metrically mutual in opposite leaves, and that upon careful 

 examination, other evidences of compression are to be seen. 



Examination of the leaf fundaments in Lonicera and Forsythia 

 shows that the above explanation is undoubtedly correct for 

 them also. Whether a similar mechanical explanation is true 

 also for the ash I have some doubt, and the cases of analogous 

 appearances in alternate leaves cited are still more puzzling. 



That the variations in the ash occur on the same side of the 

 leaflets speaks against the application of the same explanation 

 must be admitted, although it is to be noted that in the lilac the 

 variations are sometimes mutual in the same way, namely on the 

 same side of the midrib. This is to be explained by the fact that 

 the crossing of the leaf margins does not always set up a disturb- 

 ance of growth in both leaves involved at once. 



This apparently unimportant phase of the study of leaf varia- 



