1900] DOUBLE FERTILIZATION IN COMPOSITAE 257 
clearly. The cause of the peculiar appearance of the endo- 
sperm nucleolus shown in fig. 7 is not known, but it is not 
believed to be due to reagents, as the other structures appear to 
benormal. fig. 8 also shows the coiled male cells lying against 
the egg and the endosperm nucleus. /ig. 9 shows a more 
advanced stage of fertilization, the male cell fusing with the 
egg (2), having lost its coiled form. In the other (d), a trace 
of the coil may still be seen faintly. One is closely applied to 
the egg nucleus, the other to the endosperm nucleus. The 
regular shape of the endosperm nucleus shown in fig. 7 is also 
sac here. Nawaschin, in a paper received since the above 
mentioned figures were drawn, describes coiled male cells in the 
embryo sac of Helianthus annuus and of Rudbeckia speciosa. His 
description of the reticulated porous structure shows them to be 
Similar to those found in Silphium. 
The large amount of food present in the egg is shown in fig. 
, the deeply stained bodies being starch grains. 
ae fusion with the male cell the endosperm nucle 
tapidly (ee, fig. ro), and soon the sac is filled with a 
a : nuclei. The fertilized egg (fo, fig. ro) does not divide 
: lately after fertilization, but rests for some time. It may 
“hee to note that the nucleus of the upper antipodal 
antipo =e large at this stage. The appearance of the 
Suggests that they, for a time at least, may possi 
ie to transmit food to the embryo, and also to the 
sperm, ! 
— shown that eight is the characteristic cexte 
is the a: of chromosomes, and he scheastters that oe 
ec. S In the tapetal cells. The writer was unab ae 
. ay count, but sixteen appears to be the ae 
eiddosperin.. ae oe than sixteen were oe — 
that the aa : is believed, as the result of many te é 
ae “ in the endosperm is twenty-four. Lee 
Sperm ac. € should expect here, for the primary ee 
‘aining sizhi Nada from the fusion of three nuclei, eacd ©©" 
romosomes. 
in 
