1916] CURRENT LITERATURE 163 
distribution of various types of leaves in the principal phytogeographic regions, 
aving in view ‘‘the question of the conservatism of foliar characters and their 
i i The present paper summarizes the 
entire leaves and leaflets of dicotyledons. The authors conclude that there is 
a marked correlation between leaf margin and environment in the distribution 
of dicotyledons. Entire margins predominate in lowland tropical regions; 
while non-entire margins predominate in mesophytic cold-temperate regions. 
In the tropics non-entire margins are favored by ‘moist uplands, equitable 
environments, and protected, comparatively cool habitats.’’ In the cold 
temperates, entire margins are favored by “arid environments and other 
physiologically dry habitats.” These correlations ibite trikingly 
by trees and large shrubs than by herbs. If these conclusions are tenable, “the 
determination of the percentages of the entire and non-entire leaves in Creta- 
ceous and Tertiary dicotyledonous floras affords a simple and rapid means of 
gauging the general climatic conditions.’’ Attention is called to the danger of 
inferring that leaves are conservative or inconstant because in some cases they 
of climate and the evolution of plants is becoming increasingly evident.— 
MC 
Embryo sac of Onagraceae.—The embryo sac of this family, the subject 
of a good many investigations, has been reported as persistently 4-nucleate, 
with an occasional more or less doubtful exception. TAcKHOLM” has investi- 
gated the situation and confirms the constancy of this feature, stating that he 
never observed an embryo sac that was not 4-nucleate, nor did he find any 
. transition stages to the ordinary 8-nucleate type. Previous statements in 
reference to vanishing antipodals, thus reducing the numberof nuclei to 4, are 
thought to be accounted for by confusing them with the remnants of the 3 
lower megaspores, the uppermost megaspore in this family developing the 
embryo sac. In fact, it seems that one or more of the 3 lower megaspores 
often persist attached to the embryo sac, and this sometimes results in a strik- 
ing resemblance to 3 antipodals in a linear series. This union of persistent 
megaspore nuclei left free in the antipodal region of the sac. As to the possible 
nutritive function of these persistent and occasionally proliferating megaspore 
ear there has been much discussion, but no certain evidence. In any event, 
is behavior of the Onagraceae seems to be unique, and its possible eae 
ae significance must be left at present to the imagination.—J. M. C. 
e Abaeags HOLM, GuUNNAR, Beobachtungen iiber die Samenentwicklung einiger 
Onagraceen. Svensk. Bot. Tidsk. 9:294-361. figs. 16. 1915. 
