70 MR. G. C. DRUCE ON THE 
are found not only on estuarine mud-flats at the coast-level, but 
also on mountain tops, the plant with bare intercostal spaces 
occurring on Ben Lawers, Snowdon, Little Culrannoch, and 
Carran Tual, while others from Ben Heasgarnich in Glen Lyon 
belong to the form with hairy intercostal spaces. 
The geographical distribution of the two plants, so far as the 
British Isles are concerned, is also very similar. But although 
the character alluded to is trifling it is, as I have said, constant, 
and hitherto I have been unable to find two fruits from the same 
plant that differ in this particular: not only so, but all the 
individual specimens (numbering a hundred) which I examined 
on the Snowdonian cliffs belonged to the variety with the inter- 
costal spaces bare; while specimens gathered by Dillenius 174 
years previously, and others in the British Museum Herbarium 
obtained nearly half a century ago from Snowdon Peak, and 
those collected by Babington from Crib Goch in 1847 and from 
Twll Dhu in 1832, belong to the same form. From this fact we 
may presume that the character bas some degree of constancy 
and can be perpetuated by seed, although at present we lack 
direct evidence on this latter point. It is not improbable that 
the numerous plants of the Thrift on the Snowdonian cliffs may 
have originated from a comparatively small number of wind- 
blown seeds from the neighbouring coast. Various theories 
have at one time or another been suggested to account for the 
occurrence of the Maritime Thrift in alpine places. I am not 
one of those who believe that it is caused by any advancing or 
receding thermal changes, but think that its presence on the 
mountains can be accounted for by the wind carrying the seeds 
from the coast to a place which, being bare of other vegetation, 
is suitable for the Thrift ; and its comparative rarity on the lower 
levels is probably due to the fact that in these situations there 
is a much greater amount of competing vegetation against which 
the Thrift cannot successfully colonise. Mountains offer the 
not less important factor of a moisture-laden air, which is not so 
uniformly present in the lower areas away from the coast. An 
examination of the structure of the fruit with its enveloping 
calyx will show how highly specialized it has become, and chiefly 
perhaps in those structural characters which enable it to be 
more readily carried by the wind to a suitable place of growth. 
I hope to have the assistance of Mr. Poulton, Jun., of Oxtord, 
in making some experiments on the comparative culture of the 
