April 1892.] THE BOTANICAL SOCIETY OF EDINBURGH. 373 



4, Oue nucleus gives rise to the antipodal primordial cell, 



and the third antipode which 

 is placed on a slightly higher 

 level than the two other anti- 

 podes. 



Do these four groups, of two cells each, correspond to four 

 two-celled prothalli ? Are we dealing with eight spores ? Or 

 what interpretation has to be given to the eight cells ? 



Much light has been thrown on this difficult question by 

 Dodel * in an able paper, which the author was so kind as 

 to send me. While investigating the phenomena of fertilisa- 

 tion in Iris sihirica the author incidentally discovered the 

 important fact that besides the egg-cell, one or both synergidte 

 may occasionally be fertilised, and even give rise to a few- 

 celled embryo or embryos {I.e., figs. 14, 16, 17). This dis- 

 covery is summed up in the words : " There can no longer 

 be any doubt that the synergidse of Iris sihirica not unfre- 

 quently possess the characters of an egg-cell, as they have 

 the power of receiving a sperm-nucleus, and of undergoing a 

 true fertilisation in consequence of which even a several- 

 celled embryo may develop." 



Quite independently of Dodel's researches Overton t has 

 discovered the same phenomenon in Lilium Martacjon, and 

 figured it. If then both synergidae and the ovum may be 

 fertilised by sperm-nuclei they must all three be sexual cells ; 

 but also the fourth apical cell (the micropylar primordial cell) 

 must be a sexual cell as it conjugates with a cell from the 

 basal or antipodal end of the embryo-sac. We find thus 

 the micropylar half of the embryo-sac to contain four sexual 

 cells, while the antipodal half contains normally only one. 

 If we interpret the embryo-sac as the equivalent of two pro- 

 thalli, we would have to consider the micropylar prothallus 

 as consisting of four sexual and no vegetative cells, while the 

 antipodal prothallus would be a prothallus giving rise to 

 three vegetative cells and one sexual cell, which latter fuses 

 with one sexual cell from the micropylar region. 



The objection I have to the two-spore-hypothesis is 

 shortly this : — Sporocytes normally divide into four spores 



* Dodel, Befruchtungs-Erscheinungen b. Iris sibirica, Zurich, 1891. 

 t E. Overton, Entwickelung und Vereinigung d. Geschlechts-producte b. 

 Lilium Martagon, Ziiiich, 1891, fig. 16. 



