32 MESSRS. NEWELL ARBEH AND J. PAHKIN ON 
The subsequent tendency of various lines of research, until quite recently, 
has been on the one hand to draw closer together the ties of relationship 
existing between Gymnosperms and Pteridophytes, and on the other hand to 
increase the isolation of the Angiosperms. For instance, one of the most 
important embryological facts, recently brought to light, linking together the 
Gymnosperms and the Pteridophytes, is the formation of antherozoids in the 
pollen-tubes of Cycas and Ginkgo. 
Much work has been done on the embryo-sac of Angiosperms, primarily 
with the hope of throwing light on the question of its homologies, and the 
line of descent of the group. Practically every Angiospermous family, 
e 
which is of interest phylogenetically, has now been examined, including quite 
recently the Magnoliaceze *. The outcome of the whole of this vast investi- 
gation has merely emphasized the great difference which exists between the 
Angiospermous and Gymnospermous embryo-saes, and in addition the great 
similarity between those of the Dicotyledons and the Monocotyledons. 
Variations do occur, but these appear to us to be points of detail rather than 
of fundamental importance. In fact they are of such a kind that it is 
uncertain whether they should be best regarded as primitive or as recently 
acquired. ‘This is particularly true of the antipodals, a group of cells more 
variable perhaps than any of the other constituents of the embryo-sac. 
Though the net result of these studies has so far not enabled us to bridge the 
gap between the Angiospermous and Gymnospermous embryo-sacs, yet 
additional discoveries of great interest have been made, e. g. double fertilization 
and chalazogamy. The former seems to increase rather than to diminish the 
difficulty of explaining the Angiospermous embryo-sac and especially its 
endosperm in terms of the fern-prothallus, or the female gametophyte of the 
Gymnosperm. 
Turning now to paleeobotanical work, the main result has been the same. 
Remarkable fossils have been found connecting the Gymnosperms more closely 
with the Ferns, but anything of a like nature bearing on the Angiosperms has 
remained hidden. The rocks have been singularly silent as regards the origin 
of the latter group, now predominant in the vegetation of the world. 
The existing Cycads, and less clearly the Conifers, have been linked up with 
the Ferns by means of the anatomical investigations of certain Palaeozoic 
petrified stems possessing fern-like characters, known as the Cycadofilices, and 
by the discovery of the seeds and male fronds of these plants. The old idea of 
connecting the Gymnosperms with the Lycopods is now no longer tenable, at 
any rate so far as the Cycads are concerned. The Angiosperms, on the 
other hand, have been considered to have sprung from the Ferns; yet 
no work on the existing Filices has shown any direct connection between the 
two groups. 16 15 true that Zsoetes f has been brought forward as revealing 
in the mode of origin of its stem, root, and first leaf, as well as in its adult 
* Strasburger (1905). t Campbell (1891) pp. 253-254. 
