290 DEVELOPMENT OF COMPOUND LEAVES. [BOOK i. 



remains after the birth of the leaflets, and the latter are then 

 originally inserted immediately on the axis. 



It seems from what precedes that the plan of development 

 of a compound leaf is generally the same as that of a simple 

 one, that is to say, it proceeds from the apex to the base. 

 But the longer the common petiole, the greater the diffe- 

 rences between the upper and lower parts ; and the latter 

 although last formed, are often much more considerably 

 developed than the former. Yet some observers, and espe- 

 cially Steinheil, have drawn from this fact a conclusion as 

 to the origin of these parts that is entirely unwarrantable. 



Each leaflet considered by itself is also developed from 

 above downwards ; but its stipels do not proceed as the 

 stipules of a simple leaf. In the small number of cases in 

 which I have observed these stipels, they have never been deve- 

 loped before the leaflets accompanying them ; moreover they 

 are not produced, like stipules, as parts of the lamina of the 

 leaflet, but as small tumours at the base of the partial petiole. 



A very young leaf (about 1^ millimetre long) of Melian- 

 thus major is composed of a palmate blade with five, seven, 

 or nine lobes of various lengths. The lamina has as yet no 

 petiole, and is applied at its base against the axis, from which, 

 however, it is separated by the interpetiolar stipule which 

 already exists as a rim. The three terminal lobes of the 

 leaf are considerably larger than the lower lobes. All these 

 lobes cross at their base, and they consequently penetrate 

 further and further into the lamina, whence results a five- 

 parted leaf, three or four millimetres long. Later, an opaque 

 line is found on the anterior surface of the terminal lobe, 

 which is gradually extended to its base ; a little while after, 

 the other lobes offer the same formation, which is, as we 

 know from what has already been stated, due to the develop- 

 ment of the lamina of the leaf. At the same time, the bases 

 of the leaflets, which were originally arranged almost in a 

 semicircle, separate more and more, so as to form a lower 

 and an upper pair, and a terminal leaflet; they have all a 

 short broad stalk, continuous with the medial line of the 

 terminal leaflet; the total length of the leaf is at about this 

 time from six to seven millimetres ; it already represents, on 



