156 PROCEEDINGS OF THE ACADEMY OP 



donous mass had been shaped as we find it finall}^ developed in 

 the seed. 



In the case of Quercus rohur a plurality of plantlets from one 

 sprouting seed was not uncommon. He had found dozens in 

 a peck of seed. These were usually in tAvos, but occasionally 

 in threes. Of the last he exhibited only one specimen. In all 

 these cases there Avas no more matter emploj'ed than if only one 

 dicotj-ledonous seed had been perfected in the usual way. The 

 division of the mass into four or six cotyledons was ver}' arbi- 

 trary. In some cases one cotjdedon Avould have one-third the 

 whole to itself, the remaining being divided between the rest. 

 There was also the same irregularity^ in direction. In some cases 

 the division, after going longitudinally perhaps one-third the dis- 

 tance, would take a lateral direction, giving a broken-off appear- 

 ance to the one cotyledon, similar to the leaf of a Liriodendron^ 

 while the opposite division would thus be left with a thick gib- 

 bous tip. The divisions in other cases would take a somewhat 

 spiral course. It was evident that there was little more rule in 

 these divisions than there would be in the breaking of a piece of 

 glass. 



He had examined a half peck of sprouting acorns of Qnercus 

 jjalustris and another of Quercus macrocarpa^ but in these he 

 could detect no sign of variation — each seed seemed cleft smoothly 

 and directly through the centre into two regularly equal halves. 

 But in Quercus rubra there were some remarkable phenomena. 

 Of these he exhibited numerous specimens. In none of them did 

 he find a plurality of embryos, the interest was in the division of 

 the pair of cotjdedons. All the specimens of Quercus rubra 

 examined had fissures extending from the outside towards the 

 centre. These were two, three, or four. In some cases the fissures 

 extended but a mere line in depth ; at other times they would 

 penetrate to near the centre of the mass. In all these instances 

 the surfaces in the fissures were clothed with membrane, just as 

 we find to be the case in the walnut. The divisions of the coty- 

 ledons ivere_ always in the direction of these Jissm^es, giving the 

 lobes a very irregular appearance. In some cases a triangular 

 section served for one lobe, while the sum of two united tri- 

 angular sections served for the second one. It was so evident 

 here that the divisions of the seed into cotyledons were in the 

 weakest lines of union, and that these weak lines existed prior to 

 the division, that he thought no one of the gentlemen present who 

 were examining them could possibly dispute the conclusion. 



He further suggested that the absence of au}^ secondai'y mem- 

 brane from the face of the cotyledon was an argument in favor of a 

 late and smaller division. In ruminated seeds, such as the walnut 

 or papaw, this membrane followed tlie innei'most outline of the 

 surface, and was no doubt a formation directly from the cells of 

 the cotj'ledouous mass ; but the suddenness of cotyledon forma- 



[ August 15 



