INDEPENDENT ORIGIN OF PARTS 183 
than has often been assumed, and comparative arguments based on 
embryological facts must be used with the greatest caution. 
The independence of origin of the separate parts thus seen in some 
degree in the embryo calls for further consideration, since it is shown also 
elsewhere than in the normal embryo, and it will affect in some degree 
the conception of the nature of the parts of the plant. It is a common 
experience in the plant at large that roots may arise independently of 
other parts: frequently their occurrence is irregular both in number and 
position, and this finds its illustration in almost all the large groups of 
plants. Goebel! quotes examples of “free-living” roots, which do not 
spring from a shoot at all, in Pyrola and Monotropa: he regards these 
as derived from the normal in accordance with the saprophytic mode of 
life of these plants. A very peculiar illustration of the detachment of origin 
of roots is shown in the 
abnormal cases of apogamy 
described by Lang (Fig. 
93); for here numerous 
roots were formed inde- 
pendently of any other 
parts of the sporophyte ; 
thus the idea of detach- 
ment of the root is already 
a familiar one. On_ the 
other hand, the current 
conception of the leaf is 
of a part in close genetic 
connection with the axis: 
but this also has been 
shown by Goebel to be Fic. 93. 
open to exceptions. He Scolopendriuim vulgare. Prothallus from the branched cylindrical 
. process of which ten roots arose: eight of these are visible in the 
describes cases of free- drawing. x about 6. (After Lang.) 
living leaves? The old 
morphological dogma asserted that a leaf could only arise out of the 
vegetative point of a shoot; but Goebel accepts the facts disclosed in 
Lemna and Utricularia, as well as the condition of the embryo in many 
Monocotyledons, as overthrowing this dogma. In the latter case the 
cotyledon arises without any vegetative point of an axis being visible. He 
also quotes the case of Adiantum Edgeworth?, a Fern which produces buds 
at each leaf-tip.2 This case I regard as being important for comparison 
with the condition seen in embryos; for according to Goebel’s description 
and drawings (Fig. 94), the first leaf of the new bud arises not from the 
1 Organography, p. 234. AL iy De 235% 
* Lic, p. 241. See also Kupper, Flora, 1906, p. 337, who found that in Adzantum- 
species three, and in Avezmia rotundifolia even six leaves originated before the stem-apex 
was defined. 
