— 152 — 



The investigations of Juel (5) and Murbeck(13) have shown, 

 that some species of Compositae are apogamic. Antennaria alpina 

 certainly shows the most peculiar development, with which the 

 EMC directly becomes the embryo sac but at the same time with 

 the unreduced number of chromosomes. In Taraxacum there is 

 still a remnant of the tetrad division as one division of the EMC 

 takes place, but in this case as in the former with an unreduced 

 number of chromosomes. 



As regards Hieracium, Mur beck (13) has shown that the 

 embryo develops „parthenogenetically", that is so say, the egg cell 

 develops without fertilization. My investigations on Hieracium have 

 also clearly shown that several different forms of embryo formation 

 are to be found here. 



According to Ostenfeld's experiments (IG) H. excellens can 

 produce ripe seeds without fertilization, but can also produce 

 hybrids after pollination with, for instance, H. aurantiacum. A 

 similar case has been found and carefully examined by Overton 

 (18) in Thalictrum purpurascens. Here it was seen, that in some 

 ovules a typical reduction division is carried on, and thus normal 

 ES is formed, which could possibly be fertilized, whilst in other 

 ovules „keine Reduktion der Chromosomenzahl stattfindet". In 

 these latter ovules ES is formed with an unreduced number of 

 chromosomes and it is probable in this ES the egg cell, which can 

 be developed without fertilization. 



The material which was at my disposal was very complete 

 and its origin well known. If not specially mentioned all my 

 investigations are carried out on material from descendants of 

 castrated or isolated flowers, in which case fertilization has cer- 

 tainly not been the cause of the development of the egg cell. 



The Pollen Development. 

 Hieracium excellens. 



In H. excellens in full bloom the anthers are quite devoid of 

 pollen grains, but the story of development shows that here, as in 

 Taraxacum, pollen tetrads and pollen cells are to be found which, 

 however, are at a later period desorganised. 



The PMC shows in the beginning about the same condition 

 as in H. auricula. There is, at least there seems to be, an out- 



