McAllister—Cytology and Embryology. 617 
degenerates, after which the remaining single nucleus divides 
twice to form the four-celled embryo-sac, which consists of two 
synergids, the egg and the upper polar nucleus. 
Went has figured and described synapsis as occurring in the 
mother cell of Oenone Imthurma previous to the first division 
of the embryo-sac mother cell, which seems to point clearly to 
the fact that'the reduction divisions coincide with the first two 
divisions of the mother cell. 
He also finds consistently in all forms studied the large 
pseudo-embryo-sac which is formed at the base of the small 
true embryo-sac and into, which the embryo grows after fer¬ 
tilization, the true embryo-sac degenerating. 
The more recent literature of polyembryony has been admir¬ 
ably summarized and the different methods of the formation of 
embryos classified by Ernst (21). Since Ernst’s review Mur- 
beck (89) has reported embryos arising from nucellar tissue in 
Alchemilla pastoralis and embryos arising from the synergids 
in Alchemilla sericata. Guiguard (35) has also reported em¬ 
bryos arising from the synergids of JSTaias major. Treub (103) 
has reported the occurrence of the peculiar endosperm embryos 
in Balanophora elongata. Cook (17) has recently reported 
polyembryony in Mangifera indica and in Eugenia Jambos, the 
embryos arising from the nucellus. 
The present study of Smilacina racemosa was originally un¬ 
dertaken with the aim of comparing the macrospore formation 
and embryo-sac development in this form with that in Smilaci¬ 
na stellata (50). The problem has since been extended to in¬ 
clude the embryo development as well as the reduction phe¬ 
nomena in the pollen mother cells. 
The materials for this study were collected in the vicinity of 
Beloit, Wis. during Miay and June of 1906, 1907, 1908 and 
1909. Several fixatives were used but the material killed in 
Flemming’s strong solution was finally used exclusively for the 
study of the reduction divisions while the weaker chrom-acetie 
fixatives gave better results with large embryo-sacs. Whole 
racemes were fixed and sectioned for the study of the pollen 
mother cells. For late stages of embryo-sac and embryo de- 
