SEED-STRUCTURE ETC. OF PACHIRA AQUAT1CA. 147 



the small number of pollen-grains produced. I believe I am cor- 

 rect in stating that no seeds have ever been produced by any 

 means in a state of culture. 



Meyenia erecta is an Acanthaceous shrub of western tropical 

 Africa, and it has purple or white flowers. 



On the Seed-structure and Germination of Pachira aquatica. 

 By E. Irwin Lyis-feu, of Kew Gardens. (Communicated 

 by Dr. J. Murie, F.L.S.). 



[Bead May 2, 1878.] 

 (Plate VIII.) 



Comparatively few observations have been made on peculiar 

 seed-structure and germination, and therefore I bring before the 

 Society an example possessed of much interest. The structure of 

 the seed is of the highest importance, seeing that all that follows 

 in germination may be looked on as an evident consequence. 



The seeds now in question were received at Kew from Hon- 

 duras in July of last year as those of the " Provision Tree "; but 

 to what tree this appellation is properly applied I was at first 

 unable to discover, though being, without doubt, from the aspect 

 of the young plants, a species of Pachira. Specimens in flower, 

 sent by request for the purpose of identification, show it to be 



P. aquatica. 



The seeds vary in size and slightly in form, apparently from 

 position and compression in the fruit. They are without albu- 

 men, and consist in bulk of but one cotyledon, which is very 

 fleshy and lobed, in a manner well shown by the sketches in 

 Plate VIII. This cotyledon is evidently a store of much nutri- 

 ment, and, after germination, persists a long time. The other, on 

 the contrary, soon falls and appears to have little or no function ; 

 it is always diminutive in size, and is not fleshy. The cotyledons 

 are not quite opposite ; the smaller is always the highest, and is also 

 attached a little on one side towards one or other of the angles of 

 the larger, which it embraces by its under face. The larger coty- 

 ledon determines the form of the seed, and requires particular de- 

 scription. Typically it has one plano-convex fold at right angles 

 with the attachment, forming half of the spheroid of which the 

 whole may be said to consist. There are, then, two lateral lobe* 

 or folds which together form the other half of the spheroid, each 



