1382 THE NATURE-STUDY OF PLANTS 
Robert grows in profusion, but as a variety * which is 
in many respects unlike the type. 
I should like to draw attention to a few very 
easily observed points of difference, which may, I 
think, be attributed to the different surroundings. 
The plant is very much less luxuriant and prolific, 
it is smaller both as a whole and in its leaves and 
flowers ; moreover, it is more fleshy, considerably less 
hairy and brown-red instead of green in most of its 
parts. 
If we examine its various organs we shall find that 
the root system is quite different, but in appearance 
rather than in plan. Instead of being slender and 
much branched there is a thick fleshy main tap-root 
which grows downwards and does not branch until 
it reaches the moist layers some inches below the dry 
surface shingle, and then it does not branch much, 
The leaves have shorter stalks; in addition to being 
more fleshy, smaller, and less hairy, they are much 
more finely divided and with hardly any suggestion 
of green in those of the rosette The smaller flowers 
are frequently of a deeper colour, the petals are 
narrower, and their white lines less pronounced, while 
the seeds are without the long threads which are so use- 
ful to their brothers in the moist hedgerow ; they are, 
however, dispersed in the first instance by the same 
mechanism. It would be interesting to know exactly 
the cause and meaning of all these differences, but 
beyond suggesting that it is probable that the dry 
maritime habitat is at the bottom of them, I have not 
much to offer by way of explanation, though I may 
remark that when, in 1916, I planted some young 
specimens from the shingle in moist soil covered with 
* Var. purpureum. 
