28 M. Trecul on the Structure of the Leaves of Palms. 



uules of tlie pinnate leaves formed like those of dicotyledonous 

 plants ? I do not hesitate to say at once, that the phaeuomenou 

 is very different, and that the origin of the ligula is by no means 

 that which has been supposed by very celebrated anatomists. 



In examining palms with pinnate leaves, one of the leaves of 

 which is just expanding, it will be seen that the old leaves have 

 the pinnules distant from each other on the rachis and free at 

 their extremities, whilst that which is just emerging from its 

 enclosing sheath presents a very singular appearance. The elon- 

 gation of the rachis removes the pinnules from one another, but 

 tliey are all joined together at the apex ; sometimes they are 

 united in this manner by a cellulo-fibrous thread which even 

 contains vessels (I have seen spiral and streaked vessels in the 

 threads of Phoenix sylvestris), and which extends from the base 

 of the limb to its apex. The sheath of this leaf encloses another, 

 all the leaflets of which are frequently so compressed together, 

 that they ap])ear to form a single piece without any parts distin- 

 guishable by the naked eye ; in other cases in which the com- 

 pression is less, all the partes of the leaf are apparent. 



What is the mode of formation of these leaflets, so singularly 

 attached to one another ? 



In examining a Chamcedorea Martiuna we find that in this leaf 

 in which all the parts are pressed together, the upper leaflets are 

 much longer than the lower ones. In a leaf of 16 centimetres 

 (about 6i inches) in length, the upper pinnules were 13 centi- 

 metres (about 5^ inches) long, whilst the lower were only 3 mil- 

 limetres (about ^ inch), and this disproportion may increase 

 when the apex of the leaf becomes extended. In this leaf there 

 were thirteen pinnules on each side of a rachis of 2 centimetres 

 (about ^ inch) in length, and the two rows being placed on the 

 inner surface, the back of the rachis only could be seen. 



A iew millimetres below the lower leaflets is the opening of 

 the sheath. If this be removed, a \ery remarkable conical body 

 is exposed. Extracted from a leaf of 16 centimetres in length, it 

 measured 3^ millimetres. This is also a leaf, which when viewed 

 from behind presents an entire surface, but in front is divided 

 into two portions ; the lower poi'tion is cylindrical and notched 

 at the apex at the opening of the sheath ; the upper portion, 

 which is conical, is divided longitudinally on the inner surface 

 into two rolls, which diverge towards the base and become atte- 

 nuated towards the apex. These are the two rows of leaflets in 

 course of formation. Each roll is transversely striated on the 

 sides, and the strise or furrows of one side of the roll alternating 

 with those of the other side of the same render the longitudinal 

 ridges sinuous. 



By opening the sheath of this leaf I obtained another which 



