238 PHYSIOLOGICAL BOTANY. 



the vessels were empty in summer ; towards the end of 

 October and at the beginning of November they only 

 contained a small number of cellules adhering to the walls 

 of the vessels ; but a month later he found them copiously 

 furnished with both large and small cells . In a fourth 

 year's branch of Robinia pseudacacia, the most external 

 annual zone was in a condition similar to that of a first 

 year's shoot of the same plant ; the three inner ones were 

 completely filled with cells. As regards their being 

 adherent, he makes the remarkable observation, that the 

 small cells are always attached to the side of the vessel 

 when it is surrounded by cells of woody tissue or the 

 parenchyma of the medullary rays, but never to a wall 

 which is bounded by an adjacent vessel. Moreover he 

 observed that one of these cells always lies in front of 

 the dot of a vessel, which corresponds with the dot of the 

 adjacent external cell. He also believes he has observed 

 that the membrane of the utricle has some relation with 

 the primary membrane (which belongs to the external 

 cell and the vessel, and which closes the two dotted canals), 

 and that in its earliest stage it is an expansion of this 

 primary membrane into the cavity of the vessel. Hence 

 the formation of the inner cell depends upon the activity 

 and development of an external adjacent cell. This is 

 most distinctly seen when preparations of these vessels in 

 Vitis vinifera and Sambucus nigra are treated with 

 caustic potash. To avoid unnecessary circumlocution, he 

 adds, he cannot avoid assigning names to the objects, 

 especially to distinguish the old cell from the utricular 

 sac, which he denominates the tliylle, these two composing 

 one compound organ. Considerations upon the forma- 

 tion and development of these and other cells follow. 

 The investigations of the author deserve the greatest 

 attention and careful repetition, for the purpose of either 

 confirming or correcting these observations upon a re- 

 markable phenomenon. 



We have received accurate and comparative experiments 



