GOETHE ON THE METAMORPHOSIS OF PLANTS. 363 
XI. Of the immediate Covering of the Seed. 
82. We find the seed, on the contrary, in the highest degree of con- 
traction, but internally perfect. It may be peiteived, in various seeds, 
that transformed leaves constitute their first covering, that they more 
or less adapt this covering to their shape, and in most instances that 
they have the power of closely attaching it and of entirely changing its 
orm. Having seen above, that many seeds are developed in and from 
a single leaf, we need feel no surprise that a single embryo should 
clothe itself with a leafy covering.* 
83. We see in many winged seed-vessels traces of such modified 
leaves imperfectly fitted to the seed,—in those, for instance, of the Maple, 
the Elm, the Ash, and the Birch. The Marigold affords us a very re- 
markable example, in its three circles of differently-shaped seeds (fruits), 
of the manner in which the embryo gradually contracts a covering of 
larger dimensions than itself, and closely adapts it to its own form. In 
the outer series the seed-vessel still retains a shape resembling that of 
the leaflets of the involucre, except that the rudimentary seed occasions 
a strain on the midrib, and curves the leaf, the inner curved surface 
being longitudinally divided by a membrane into two parts. 
In the next circle a still further change takes place; the little leaf is 
both narrower and shorter, the membrane has entirely disappeared, 
and the rudimentary seed is more plainly shown at the back, on whic 
moreover little excrescences are now perceptible ; these two circles ap- 
pear to be either not at all, or imperfectly fructified. In the third circle 
the curved shape of the seed is undisguised, the covering fits closely, 
and all its ridges and excrescences are complete.t Here we see a fresh 
* For instances of the reversion of seeds or ovules to leaves, see Lindley, * Ele- 
ments of Botany,' p. 88; Moquin-Tandon, * Tératologie Végétale,’ p- 205, ete. 
Tt can har ae ts of the ovule or seed is yet 
understood. While there is much evidence to show the s0 
much in favour of their intrinsic axial nature. 
ce A. Braun, Mém. sur les Transform. de POvule Végétale, ete. ; Aun. Se. Nat. 
éri ooker on ovule of Welwitschia, Trans. Linn. Soc. 
s especi J 
argins, and the third from the middle of the inner surface, 
whieh it “divida into two parts.” Tn Jepieres, * South African genus closely 
the fruits are even more decidedly three-sided and three-winged. 
np ane 
