Notes . 
7 1 5 
The first two leaves after the cotyledons, when roughly examined with a lens, 
may appear to bear nothing but malpighian hairs, like the leaves of the mature plant, 
but more careful examination under a low power of the microscope generally reveals 
the presence of a small number of three-armed hairs. Some facts connected with the 
distribution of these hairs are worth recording. 
Usually one or more three-armed hairs are to be found on one or both of the 
first two foliage-leaves, but on the third, fourth, and fifth leaves they are rare, and 
none were found on later leaves borne by the main axis. A noticeable feature in 
the distribution of the three-armed hairs is that they are nearly always restricted to 
the distal half of the leaf, and generally (especially when there are only two or three 
of them) they lie quite near the apex of the leaf. 
Another point of interest lies in the fact that the nature of the soil apparently 
influences the number and distribution of the 
three-armed hairs. Twenty seedlings grown in 
loam were compared with twenty seedlings grown 
in sand, and the following results were obtained. 
On the first two leaves, three-armed hairs were 
decidedly more numerous on the plants grown 
on sand, the total numbers being 218 for the 
twenty plants on sand, and 97 for the twenty 
plants on loam. 1 These figures give an average 
of about 5*4 three-armed hairs per leaf in the 
first case, and 2*4 in the second; but the actual 
numbers vary greatly for different seedlings, thus, 
while on more that 50 per cent, of these leaves 
the number of three-armed hairs was o, 1, or 2, 
the largest number observed was 34 on sand, and 
1 8 on loam. c r . , , f ,. 
Fig. 1. Some of the types of hair 
On the two kinds of soil an equally marked found on the cotyledons of the wall- 
difference in the distribution of the three-armed fl . ower - ? e " eral . ou ‘ lin ' s are 
shown without indicating the lrregu- 
hairs was shown. On loam they were more larities caused by the knobs, x 60. 
numerous on the upper side of the leaf (66 upper 
side, 31 lower), while on sand the numbers on the two sides were nearly equal 
(no upper side, 108 lower). 
On the third and fourth leaves of the forty seedlings, eleven three-armed hairs 
were found, three on loam, and eight on sand, while the fifth leaf of twenty plants 
yielded four of these hairs. Out of the forty seedlings only four (all from loam) 
showed no three-armed hairs on leaves after the cotyledons. 
The conclusion, perhaps, most naturally suggested by the facts described above 
is that the presence of hairs with three or more arms is an ancestral character 
preserved in the early stages of the ontogeny of the wallflower, and formerly common 
to all the foliage-leaves of the plant. This view gains some support from the fact that 
1 On sand the leaves were considerably smaller than on loam, but from some rough estimates 
it appeared that the total number of hairs (malpighian and three-armed) per leaf was about the 
same in the two cases, the hairs being more crowded on the smaller leaves. 
3 C 2 
