356 BOTANICAL GAZETTE [November 



r 



cell (or cells) up to the telophase stage. There is no evidence of a wall 

 forming between the two nuclei either in the telophase stage or later; 

 and the two nuclei are thus left free in the cytoplasm. This stage 

 was seen repeatedly, and never was there a wall separating the 

 megaspores formed by the second division. Figs. 20, 21^ 2j, 24 

 represent conditions common in the material studied, which was 



July 



This is the con- 



dition in most flowers when one looks for fertilization stages; but 

 here the megaspores are just forming in many of the ovules, although 

 in an occasional one there was some evidence of fertilization. The 

 figures give a rather complete series of the stages that would be 

 expected to show indications of wall formation; but there is no 

 suggestion of an ephemeral wall even, as might be expected in case of 

 its relatively recent elimination. Fig. 22 shows tw^o daughter cells 

 in which the nucleus of the micropylar one is developing, but this 

 was the only case of the kind seen. 



EMBRYO SAC 



Immediately after the division of the inner daughter nucleus 



megaspore 



embr}^o sac {fig. 2j), which incr 

 erv vacuolate. The two mega 



form spirems {jig. 24) which thicken, and the usual mitotic division 

 follows {fig. 25). The spindles for this division, resulting in a four- 

 nucleate sac, are small and apparently at right angles to one another. 

 No evidence of another division was found, although at least three 



exami 



for this particular stage. When the sac is ready for fertilization, 

 four nuclei are present, so that if other nuclei are formed they are 

 very ephemeral. The micropylar daughter cell resulting from the 

 division of the megaspore mother cell may still be distinguished, 

 though it is e\idently fast disintegrating {figs. 2j, 24, 25). The four 

 nuclei of the sac are usually of the same size {fig. 26), but in one 

 instance the two micropylar ones were larger than the others {fig. 27). 

 Probably fig. 28 gives the explanation; in this sac the two megaspore 

 nuclei are not dividing simultaneously, the antipodal one being in 

 late prophase, while the micropylar one has reached the late telophase 



