92 MR. R. H. COMPTON: AN INVESTIGATION OF THE 
small cotyledons are hypogeal, while many of the Amherstiee, &c., with 
large cotyledons are epigeal ; moreover, within the Viciez, which are almost 
exclusively hypogeal, a great variation in size is found. Probably the 
difference depends on the magnitude of some such ratio as 
bulk of cotvledons 
diameter of hypocotyl ' 
when this ratio is above a certain limit the germination is hypogeal; when 
below this limit, epigeal. 
The question as to which type is to be regarded as the more primitive 
probably cannot receive the same answer in all cases ; the frequency of the 
change from one to the other renders this unlikely à priori. Moreover, 
everything depends on what group is selected as a starting-point. If we go 
back to the Pteridophyta we find no signs of hypogeal germination, for the 
so-called cotyledons are neither absorbing nor storage organs. Among the 
Gymnosperms specialisation of function has occurred in the cotyledons, and 
both types occur in plenty. 
Almost all the various families of Gymnosperms have been regarded as 
primitive by different writers. There seems, however, to be a wide agree- 
ment that the Gnetales, Abietinez, Taxodiinze, and Cupressine are relatively 
advanced, and it is noteworthy that it is in these groups that epigeal ger- 
mination preponderates * The Тахасеге, whose phylogenetic position is very 
doubtful, are imperfectly studied, but both types of germination occur t. 
The Araucarine ў, Cycadales $, and Ginkgoales|| show many undoubtedly 
primitive characters, and here we find hypogeal germination occurring 
regularly. Araucaria § Lutacta is the only exception, so far as known. 
While the distribution of types among the Gymnosperms would indicate 
that hypogeal germination is primitive, it must be noticed that there is direct 
evidence in the Cycads and Ginkgo that it is a condition derived from the 
epigeal type €. There is no reason to assume that the epigeal type of the 
Abietinem, &e., was derived from hypogeal ancestry : it seems preferable to 
regard the hypogeal germination in Cycads, Ginkgo, Araucaria $ Colymbea, 
and Torreya as specialised in relation to food-storuge. 
It seems probable that in some cases epigeal germination has been derived 
from hypogeal. For instance, Phaseolus multiflorus is hypogeal; P. vulgaris, 
* Hill and de Fraine (1908-1910), [Also the next four footnotes.] 
T Chick (1903, p. 83). 
| Seward & Ford (1906, p. 521) ; Shaw (1909). 
$ Matte (1904, p. 350) ; Dorety (1908, p. 206) ; Thomas (1907, p. 81) ; Thiessen (1908, 
p. 368), 
| Thomas (1907, p. 83); Sprecher (1907, p. 16) ; Seward & Gowan (1900, p. 115). 
4] Wigglesworth (1903, p. 789). 
