114 BOTANICAL GAZETTE [ FEBRUARY 
cytoplasm is not always visible. Fig. 61, f shows as nearly as 
possible the appearance of the fully developed male cell, and 
brings out also the fact that the sinuous or spiral form is finally 
lost, and the nucleus is merely slightly bent. 
Strasburger and others have figured somewhat elongated 
male cells in various plants, but no more than could be explained 
by the need of such a form in order to pass down a slender pol- 
len tube. In one account, that of Golinski (21), the male cells 
of Triticum and other Gramineze were described as being ‘‘ not 
at all unlike an antherozoid of the Characee or ferns.’ He 
even records the presence of a vesicle, persisting for some time 
in a loop of the sex cell. Nothing of this kind occurs in Sil- 
phium, but, judging by the figures given, the spermatozoid form 
is here far more pronounced. 
Until recently, Golinski’s account was the only one which 
offered any parallel to what has been found in Silphium, Such 
an association of a composite with the grasses would seem at 
first to be an absurdity. But it is not a new one, for the Gram- 
inee stand preeminent among the monocotyledons as the 
formers of tissue by the antipodal cells, as do the Compositae 
among the dicotyledons; and if the antipodal cells represent a 
prothallial region, there is thus a primitive character in the 
female gametophyte of both groups, matched now by the pecu- 
liar form of the male cells. 
Nawaschin’s recent announcement (28) of long, vermi- 
form male nuclei in Lilium Martagon recalled a figure published 
last year by Mottier (25), showing a coiled male nucleus 
lying against the female pronucleus. Fuller confirmation of 
Nawaschin’s preliminary statement has already come from 
Guignard (29), who not only has given us excellent figures, 
but has extended his observations to several other species of the 
Liliaceez. The male nuclei, as he figures them, resemble closely 
those of Silphium. Their fusion with the polar nuclei, reported 
by both Nawaschin and Guignard, has not been found in Silphium, 
which, by virtue of the earlier formation of the definitive nucleus, 
is not a favorable object for the study of this point. 
