May, 1922.] N THE GERMINATION AND THE POLYEMBRYONY 51 



Similar results have also been reported by Guignard 12 (1881), 

 Jeffrey 17 (1895), Ernst 9 (1901-2), Schaffner 28 (1901), Hall 14 (1902). 



While I have not as yet followed through the stages in the 

 development of these buds my preparations indicate that the develop- 

 ment of the massive suspensor may account for the. peculiar types 

 of seedling shown in Fig. 3. In this case it would appear as though 

 by means of this budding two plumules and two roots have been 

 associated with a single coleoptile and coleorhiza. In another case 

 observed, but not illustrated, but a single root was developed with 

 the twin plumules. 



In addition to the above cases of polyembryony, numerous 

 instances were observed where the nucellar cells in the vicinity of 

 the micro pyle gave rise to two or three bud-like outgrowths similar 

 in character to those reported by Strasburger 31 (1878) in Citrus, 

 and by Ganong 11 (1898) in Opuntia. In contrast to these cases 

 quite frequent examples of bud-like outgrowths were also observed 

 at the antipodal region of the embryo sac, and these were often 

 characterized by a massive growth suggestive of the suspensory 

 development referred to above. Frequently, the nucellar cells on the 

 sides of the embryo sac also gave rise to embryonic buds, but these 

 outgrowths were always of a simple character as contrasted with 

 the more vigorous growths at the antipodal and micropylar regions. 



In all these various types of polyembryony it was noticed that 

 one of the embryos might be developed from the fertilized egg (Fig. 

 5), but on the other hand it not infrequently happened that the 

 normal embryo was sooner or later inhibited in its growth by the 

 development of the other embryos. 



In some instances the egg apparently did not function at all. 

 In others triple fusion was much delayed so that as a result the 

 embryos were in a multicellular stage and rather more advanced 

 than those reported by Murbeck 23 (1902). 



The antipodals varied as a rule from three to six ; rarely were 

 larger numbers observed. Not infrequently these cells became much 

 enlarged and assumed a spherical form. In all such cases these cells 

 showed indications of early disintegration, the cell contents being 

 strongly vacuolated and sometimes multinucleated. More frequently 

 the chromatic material was massed into irregular lumps or scattered 

 into irregular fragments. 



Apparently polyembryony and other irregularities are associated 

 with abnormal ovaries. The abnormality arises at different periods 



