56 THE ESSEX NATURALIST.
But to return. Paeonia is an example of a genus with a
single species in the Catalogue ; there are also some families
represented only by one or very few, and in all such cases the
rule of not separating close allies must be respected. Thus,
whilst Actaea and Paeonia may be put together, Frankenia, for
example, should go with Dianthus rather than with Polygala
because of its affinities with the Caryophyllacea. There is no
need to give a whole box to a family or a genus merely because
there is only one species ; nothing is lost by putting into the
same box widely separated species that follow one another in
the Catalogue, my point being that something is gained by not
putting those that are closely allied into different boxes if it
can be helped.
Paeonia brings me quite naturally to Section IV., because its
ripe fruit is too large to be put into a green store box. It is
one of the delightful things about natural history that whether
the system be confessedly artificial or one that is fondly imagined
to be natural, there is always something left over which will not
fit in ; hence the necessity for this section. In addition to the
things that are too big for the other boxes, there are seeds that
are far too small to be seen individually with the naked eye,
as well as happenings connected with dispersal which cannot be
dealt with in any of the other three sections. So recourse
must be had not only to the microscope slide but also to larger
boxes or a cabinet of drawers which will hold a thing as long,
for instance, as the spent fruiting axis of the foxglove, or as
tall as the ripe capsule of Datura, a full inch and a half or more.
There is no real need to preserve a spent infructescence of
all our plants; a few that are typical of the various modes which
obtain in the family will, with the help of the herbarium, be
quite enough for all practical purposes, but there is another
class which is neither seeds nor capsules, but which is yet con-
cerned with the dispersal, for example, the hygroscopic capitulum
of Carlina, which closes when wet, or that of Arnoseris, which, like
the whole infructescence of Prunella, opens in response to moisture.
Then there are the cupule of the Beech and the involucre
of the Burdock, and it is no use mounting any of the hygroscopic
subjects on cards, because that power cannot be shown unless
they be immersed or soaked first. Again, there is one special
group of seeds which can be understood much better when