OcTOBER 8, 1897. ] 
dicecious Spermatophyta, the inference has 
generally been drawn that nutrition bears 
an important relation to the development 
of the staminate and pistillate flowers ; that 
scanty nutrition produces a preponderance 
of staminate plants, while an abundance of 
nutrition produces a preponderance of pis- 
tillate plants. For a period covering three 
decades several investigators have dealt 
with this question experimentally, notably 
K. Miller (64), Haberlandt (’75, ’77) and 
Hoffmann (’85). These experiments in 
general give some support to the inferences 
from obseryations, yet the results indicate 
that other influences are also at work, for 
the ratios of preponderance either way are 
not large enough to argue for this influence 
alone. In a majority of cases thick sow- 
ipgs, which in reality correspond to scanty 
nutrition, tend to produce staminate plants; 
while thin sowings tend to produce pistil- 
late plants. In the case of the hemp 
(Cannabis sativa) Hoffmann found that 
these conditions had practically no influ- 
ence. He suggests that the character of 
each may have been fixed during the de- 
velopment of the seed, or even that it may 
be due to late or early fecundation (’71). 
In moneecious plants it has often been ob- 
served that pistillate flowers change to 
staminate ones and vice versa, and in dic- 
cious plants pistillate ones sometimes are 
observed to change to staminate ones (the 
hemp, for example; see Nagel, 1879). 
K. Muller (’64) states that by scanty nu- 
trition the pistillate flowers of Zea mays can 
be reduced to staminate ones. 
Among the pines what are called androg- 
ynous cones have in some instances been 
observed. In Pinus rigida and P. thunbergii, 
for example, they occur (Masters (’68). 
Matsuda (’92) has described, in the case of 
Pinus densiflora of Japan, pistillate and an- 
drogynous flowers, which developed in place 
of the staminate flowers, and conversely 
Staminate and androgynous flowers in place 
SCIENCE. 
545 
of pistillate ones. Fujii (’95) has observed 
that where the pistillate or androgynous 
flowers of Pinus densijlora occur in place of 
the staminate ones they are usually limited 
to the long shoots which are developed from 
the short ones of the previous year. The 
proximity of these transformed short shoots 
(Kurztriebe) to injuries of the long ones 
suggested that the cutting away of the long 
ones might induce the short ones to develop 
into long ones and the flowers which were 
in the position for staminate ones to become 
pistillate. 
Fujii says: ‘In fact, the injuries pro- 
ducing such effect are frequently given by 
Japanese gardeners to the shoots of the year 
of Pinus densiflora in their operations of an- 
nual pollarding. But the ‘ Langtrieb’ which 
is transformed from a ‘Kurztrieb’ of the 
last year does not necessarily bear female or 
hermaphrodite flowers in the positions of 
male flowers.” ‘To determine the influence 
of pollarding of the shoots he carried on ex- 
periments on this pine in the spring of 1895. 
He pollarded the shoots, so as, as he 
terms it, to induce the nourishment to be 
employed in the development of the flowers 
and short shoots near the seat of injury ; in 
other cases one or two shoots were pre- 
served, while all the adjacent shoots of last 
year’s growth at the top of the branch 
were removed, and, farther, both of these 
processes were combined. Out of the 45 
branches experimented on, and on which 
there were no signs of previous injury, 
there were nine pistillate or androgynous 
flowers in place of staminate ones; in 21 
branches with signs of previous injury five 
were transformed, while in 2,283 not experi- 
mented on, and with no signs of previous 
injury, only seven were transformed. Such 
abnormal flowers, then, are due largely to 
the injuries upon the adjacent shoots, and, 
Fujii thinks, largely to the increased amount 
of nourishment which is conveyed to them 
as a result of this. 
