144 THE NATURE-STUDY OF PLANTS 
microscope must be used in order to see the pattern, 
which recalls a honeycomb. 
This is really a most interesting point of difference ; 
an insect-like knob would be of no dispersal value to 
a plant that throws the seed free of it, and a sculptured 
coat would, so far as one knows, be superfluous to a 
seed that is ejected while still enclosed in the knob: 
but whatever the exact use of either,may be it is 
delightful to find that neither of them is possessed by 
those species to which apparently it would be of no 
service, while on the other hand one can readily under- 
stand the possibility of their very considerable value 
to their actual possessors. 
I have already suggested a way in which a sculp- 
tured seed-coat may be of use, and a wrinkled knob, 
apart from the resemblance to an insect, may likewise 
subserve burial and anchorage, while the retention 
of moisture in its depressions would be favourable for 
germination at the close of the dispersal period. 
With regard to the interval that elapses between 
sowing the seeds and the appearance above ground of 
the cotyledons, and confining myself to the few species 
to which I have given most attention, I find that, 
broadly speaking, the small-flowered ones, the Doves- 
foot, the Shining and the Round-leaved Cranesbills, 
agree with the Herb Robert in requiring a minimum 
of about three weeks or less, but the large-flowered 
Meadow Cranesbill took thirty-nine days, not far short 
of twice as long. In my trials, too, I noticed that its 
seedlings were eaten much more frequently than those 
of any of the other four. I did not succeed in catching 
the culprit in the act, but I have no doubt that it was 
my inveterate enemy the Garden Snail: he takes his 
daily meal at night, which perhaps sounds a little 
