THE PRE-GLACIAL FLORA OF BRITAIN. 201 
pretend to be able to distinguish the different fruticose Rubi by their stones, 
nor can we find satisfactory specifie characters in the carpels of Rosa, or of 
the Batrachian Aanuneuli. The closely allied Stellaria media, S. Boraeana, 
and S. neglecta, however, have seeds noticeably different ; Arum maculatum 
and A. italieum have seeds remarkably distinct. 
Seeds (or the hard parts of fruits) being thus critically distinguishable, 
it seems advisable to place on record the evidence by which we have been 
guided in determining the fossil plants. This is the more necessary as many 
of the seeds are impregnated with pyrites, which tends to decompose and to 
destroy the specimen. For this reason it is important to preserve exact 
photographie representations, so that if the fossil itself disappears there may 
still be a satisfactory record of its discovery *. 
Fruits and seeds of. our recent. plants are seldom figured and deseribed in 
such a way as to enable the botanist to recognise them specifically, especially 
if detached from the plant. Seeds are constantly described as smooth or 
shining, when in reality they have a highly characteristic sculpture, only 
masked by a coat of varnish. Our fossil fruits and seeds also are preserved 
in an altered state, the ‘alteration in some cases being so great that their 
identity with the living plants is not suspected until recent seeds have been 
reduced by maceration or carbonising to a similar condition. It is not that 
the fossils are less determinable than the recent, often the intricate patterns 
on the outside are far clearer. But any botanist finding, for instance, fossil 
fruits of Conium in the state of preservation shown on PI. 12. fig. 60 would 
not recognise them, unless he had treated the recent fruits in such a way as 
to remove their ridges. In this case slow carbonising shows the identity of 
the recent with the fossil fruit, and this carbonising also causes the recent 
fruit to shrink till it is as small as the fossil specimens. This difficulty of 
comparison has given much trouble, and in many cases we have had the 
fossil for years before we recognised it as being the seed or fruit of a common 
British plant. + 
Tt would need a large number of magnified photographs to illustrate all 
the details of cell-structure, ete, on which our determinations have been 
based ; but the figures and descriptions now given should be sufficient at 
any rate to help the botanist, and to prevent the geologist and archæologist 
from throwing away what looks like most unpromising material for the 
study of an ancient flora. 
The mode of collecting and washing the material need not again be 
described ; but a few words are required as to the best methods of preserving 
the seeds in such a state as to be fit for study. One remark, however, 
must be made as to the preliminary work. Fossil seeds if possible should 
not be allowed to dry, at any rate not after they have been taken out of the 
* We have not thought it necessary to reproduce our photographs of the leaves, as the 
forms and venation of leaves are well known, and the specimens themselves are generally 
preserved in slabs of ironstone. 
