152 Saunders . — The Leaf-skin Theory of the Stem: 
tute of hairs. Under conditions still more favourable to hair formation, on 
the other hand, the whole surface may be covered with hairs. It becomes 
evident, in fact, that the lines of fusion of the cotyledon extensions remain 
invisible, or are brought out, or obliterated, according to the extent to 
which the conditions happen to be favourable to hair development. In 
the four species of Veronica named above, the first leaves are always 
opposite. This arrangement may persist up the stem ( V. arvensis , V. 
Chamaedrys ), but in the flower spikes, if not before, the arrangement be- 
comes spiral with a f- divergence. In V. hederaefolia and V. polita , 
where the flowers are solitary and axillary, the spiral arrangement .is 
assumed at an earlier stage. Coincident with this change in the leaf- 
arrangement we find in individuals in which the hair lines are apparent 
a corresponding change of pattern. In place of two lines to the inter- 
node, both extending only from one node to the next, we now have a 
three-line pattern resulting from a ‘ pick-up 5 along the lines of fusion of 
the kind already described (see p. 140 and Fig. 11). Of the two lines arising 
at each insertion one now descends through two internodes and the other 
through three, the one from the right or from the left edge being the longer 
of the two according to whether the individual is one in which the spiral 
proceeds from right to left or from left to right. In V. hederaefolia a well- 
defined hair line is also traceable along the whole length of the flower stalk, 
marking the line of fusion of the extensions of the two postero-lateral 
sepals, and ending in the mid-axil of the subtending leaf where it meets the 
hair line on the main axis terminating at the same point. 
Lopezia coronata. Leaves opposite below, becoming spiral above, 
upper surface slightly hairy towards the base, margins finely ciliate. Here 
again under the appropriate conditions two lines of hairs are developed on 
the hypocotyl, descending from the points where the free margins of the 
cotyledons coalesce (see Fig. 4). A decussate arrangement of the leaves is 
continued for perhaps fourteen or fifteen nodes, a difference in the level 
of insertion of the two members of a pair, which is slight to begin with, 
becoming very considerable higher up. The internodes in this lower region 
show two contour lines flanked by lines of hairs which are absent from 
the surface elsewhere (see Fig. 22). Some interest attaches to a case in 
which an irregularity in the leaf arrangement occurred in view of the con- 
ception of the ‘potential’ edge which has been given above (p. 138). In 
the specimen in question two leaves were borne at the first and second 
nodes, a slight difference in the level of insertion being noticeable in the 
latter case ; at the third node only one leaf was formed, while three appeared 
at the node above. It was evident that one of the pair belonging properly 
to node 3 had become fused with the axis in its upward development, and 
hence only became free at the node above its true insertion. In consequence 
of this disarrangement, the leaf which was carried from node 3 and one 
