Br srirsrry ise 4 babet sheet
Ui ERS SO ; rs

ii

om
Thon ban eee
Rar isventl
“ 4
$

HHith
th
: f sae

tay

2
i}
i M * f, Hi i
; peer siaiasten + H
nat rt
Tpipetsha petay
aaab isin eh:

niet
Hie Re eereeeriri
Hitiaaty sah aah
settee r eee
BiH
Tb ae ae ines
nent +

Srnrin

mornieeoeat are

jad

Aah 98

pehetire
sheastenehi bad

tibetan tha
pertst te serene ctitital ; i
tie

Tish eharateleitrrane
at PAE
i! H
Neer ee
Donan 4 Hite te
Sasbeeaaren Tabelenatatuy
esenaea tthe H
He
i Mt
sbasesasuataeceets H
rine
cbse st
nit

titiehareiae

sTihasiiely
ma } tiie

Tees distant
searteieb sie tt

Te et

teiaet
Hibs tetans

Wi i
THERE EERE

AT ET
HHH beth
cau

Edi hees saecitacdskrrsgey
ied vada

a Sty
tits
ade :

Oren te)
rene
en

eer)
RSH K
ime etna
. +e we titetestates thst etes eat

CE
ease Ctl cates and

EEN
Proreineien
bs

us od
WA racnee
: i

‘ se enasd alana
MEIER Pirie erie ity
CC nr hte
* Hate

eteay
eee

states tases wea
ass ee

est
hi
Nt a
Ne

1h

5 ane

< Rota scat ef BS i ry POS Se

es
ees YG ee Mee

Zee ek. ZOeaTy Ce PS ae
Ee A

; “MEMOIRS
GEOLOGICAL SURVEY

GREAT BRITAIN, &c.
VOL. IL—PART II.

On the Vegetation of the Carboniferous Period, as compared with that
of the present day. By Dr. Hooxrr, Botanist to the Geological
Survey of the United Kingdom.

THERE are few persons who have devoted any time to the study of
fossil plants, especially those of the coal formation, and have not been
particularly interrogated on the value of their results, compared with
those derivable from the investigation of animal remains. What that
value may be, is daily asked of the naturalist, while in the field, by the
uninitiated yet curious looker-on, who eagerly offers his aid as a col-
lector in exchange for information upon the materials he gives or
offers to procure. It is no less frequently proposed by the young
geologist, who, though skilled in seizing the characters presented by
a comparatively indestructible shell or bone, is at a loss to appreciate
those afforded by the always compressed and more or less mutilated
fragments of what were originally perishable plants.

It is with a view of instructing such enquirers that the following
introductory observations are thrown together. Relating exclusively
to the more obvious features of that formation which conspicuously
abounds in fossil vegetables, to their most prominent characters, and to
the botanical value of those features only, they can have no claim upon
the attention of the experienced paleontologist. They are little more
than the first impressions received by a naturalist, who, having been
almost exclusively occupied with an existing Flora, is called upon to
contrast with it the fragmentary remains of another Flora, whose
species are, without an exception, different from those now living, which
represent in part the vegetation of a period indefinitely antecedent to the
present, and have been succeeded by still other plants, equally diverse
from both, and which have likewise perished.

We ‘] 25

“

388 VEGETATION OF THE CARBONIFEROUS PERIOD

From the very outset it must be borne in mind, that whatever light
future investigations of hitherto unexplored coal-fields may throw upon ’
this most difficult subject, we can never hope thereby to arrive at any
great amount of precision in determining the species of vegetable re-
mains, nor to ascertain the degree of value due to the presence or
absence of certain forms, such as the animal kingdom so conspicuously
affords. Still less can we expect that they will prove equally appre-
ciable indices of the climate and other physical features of that portion
of the surface of the globe upon which they once flourished.

The great extent of the vegetable kingdom is hardly to be appre-
ciated except by the professed botanist ; and he must be an advanced
student who knows as much of its main features as he may acquire of
the animal creation during the course of an ordinary education. Every
one, for example, is familiar with the divisions of the class Animalia
into beasts, birds, fish, reptiles, shells, &c.; but much study is required
to attain an equal amount of acquaintance with the parallel divisions of
plants into exogenous, endogenous, &c. The technical terms, too,
_ employed in the one case are, very many of them, universally intelli-
gible; whilst the majority of those applied to the more conspicuous
organs of plants must be acquired by a special study. Lastly, the ex-
ternal organs of vegetables, and especially such as are generally avail-
able in the fossil state, are not the same guides to the affinity of the
objects themselves, to their habits, or to the nature of the area they
occupied, which the similarly conspicuous organs of animals are. Thus,
were fossil vegetables much more perfect than they are, the informa-
tion to be derived from their study will never hold a rank of equal
importance to the geologist with that afforded by animal remains.

It is partly owing to these circumstances that the study has been
comparatively neglected ; partly also because a far more comprehensive
knowledge of the existing forms of plants is required to make any
progress in fossil botany, than of recent zoology to advance equally in
palzontology ; for, whilst an acquaintance with a single class of ani-
mals (the shells for instance) enables the-student to understand and
distinguish whole formations, he cannot, without being somewhat con-
versant with all classes of living plants, appreciate the value of the most
perfect series of them in a fossil state.

Turning from this discouraging view of fossil botany in general to
that of the particular formation to whose consideration the remainder
of these pages will be devoted, it is satisfactory to find that it presents
facilities for the investigation of its vegetable remains such as is afforded
by no other. This is mainly due to the vast accumulation of specimens,
and to many of them being presented under very different conditions in
the under-clay, in the shales, in nodules of ironstone, and in the sand-

AS COMPARED WITH THAT OF THE PRESENT DAY. 389

stone. Had it not been for'these favourable circumstances, the study of
coal fossils would have been apparently hopeless ; for, whilst the clays and
ironstone and sandstone scarcely ever contain more than one large class
(ferns) in a fit state for determination, the shales preserve only the out-
lines of another (Svgillarie and Lepidodendrons), whose affinities could
hardly be guessed without a microscopical examination of their internal
tissues, as these are preserved in the ironstones and sandstones. Con-
sidering of how exceedingly lax and compressible a tissue the coal-
plants were composed, it is not wonderful that instructive specimens
are rare; but to appreciate to its full extent how universal is the com-
pression and how complete the mutilation of almost every individual, it
is necessary to study the whole bed or deposit, in situ. Thus will be
seen a layer of mineralized organic matter, exceeding in bulk and in
area whatever any other formation may present in equal purity; for
throughout the whole mass of the coal there will not be found one
pebble or even grain of sand: it is a deposit of vegetable matter, so
homogeneous that not a trace remains of the outward form of that
incalculable number of species and specimens of plants to whose decay
it owes its existence.

Plants, whose tissues are so lax as to be convertible after death
into a mass of such uniform structure as coal, evidently would not
retain their characters well during fossilization, under whatever favour-
able circumstances that operation may be conducted. We conse-
quently find that few specimens are available for scientific purposes.
Of the ferns, whose remains preponderate in the carboniferous Flora,
only one surface of the leaves or fronds, and that invariably the least
important (botanically speaking), is exposed to view ; and their mutila-
tion is so great that the identification of contiguous specimens is fre-
quently impossible, much more so of those from different parts of the same
or from other coal-fields. Were the species and genera identical with
those now in existence, this difficulty would be lessened, for we should
then know the variations in form which the individuals might be likely
to assume, or, at any rate, what dissimilarity between the isolated frag-
ments was due to their belonging to different parts of the same plant.
The naturalist is thus hampered in the outset by his inability to answer
questions relating to systematic and specific botany. And when he
turns to a general review of the whole, and seeks to reclothe the globe
with the vegetation to whose decomposition we are indebted for coal,
he labours under no lighter difficulties: the most casual inspection of
such a wreck suffices to show that the number of species, genera, and
even orders, of which scarce a trace remains, must far outnumber
those which are recognizable. Of the latter, again, but a small per-

centage is known in a tolerably complete state, only the larger and
252

390 VEGETATION OF THE CARBONIFEROUS PERIOD

better preserved specimens retaining those organs and appendages which
the most skilful botanist requires to examine in the living vegetable
before he can pronounce decidedly on its affinities. The female flowers
or fruit are distinguishable in very few cases, and they are so rare
that but one genus of coal-plants has thereby been referred with any
certainty to its proper position in the natural system. They occur in
the form of cones (aggregations of seed-vessels) or of isolated seed-
vessels. Their form alone is generally preserved, their interior having
been wholly destroyed, or presenting a crushed and shapeless mass of
disorganized tissue. The solitary nuts, again, may have grown in
cones or separately: they have never been found attached nor in a po-
sition relatively to any leaf, branch, or cone, that would justify their
having belonged with certainty to either. Of male flowers no traces
remain. Leaves and scales occur abundantly, but almost invariably
detached, as is generally the bark of the stems or trunks, so that the
very outline of the vegetable is frequently lost. Hence arises the ne-
cessity, in the infancy of this science, of describing the different portions,
perhaps of one plant, as species, and of arranging them provisionally
into genera ; the word genus signifying, not a natural group of species,
but a set of organs; and being synonymous in many cases with a
shorter and more expressive Latin word, long in use and better under-
stood. As an example, the genera Strobilites and Lepidostrobus may
be cited, whose species are various cones (s¢rodili), in some cases of
Lepidodendron, im others possibly of other plants, widely different ;
even the order to which they belong being distinct from that including
Lepidodendron.

This arrangement of portions of specimens under various genera is
highly detrimental to the progress of systematic botany, but is not
equally disadvantageous to the geologist, whose object it is to deter-
mine the relationship of strata, by means of a comparison of their
contained species, without so particular a reference to their affinities.
The identification of these is always open to question, from the errors
into which the imperfection of the specimens necessarily leads. Two
specimens of one plant, the one more perfect than the other, are
frequently described as different: this is eminently the case in the
Sigillarie, the markings upon the surface of whose bark differ from
those on the similar surface exposed by the removal of that bark, while
in many specimens it is exceedingly difficult to determine whether the
latter be present or not. The markings also vary extremely in
different parts of the same trunk; insomuch that fragments which had
been regarded as characteristic of six or eight separate species, have
been more recently found to belong to one, that one presenting a sur-
face equal to those six or eight fragments collectively, whereon the

AS COMPARED WITH THAT OF THE PRESENT DAY. 391

supposed species were founded. Again, as the specific characters used
in dividing this genus are drawn from what are considered very unim-
portant features in recent plants, namely the scars left by the fallen
leaves, it is evident that several distinct species may be merged into one,
in the absence of other distinctions beyond that solitary character which
does not suffice to recognize analogously marked living vegetables.

The last obstacle which demands a passing allusion, because tending
to retard our knowledge of coal fossils, is, that they cannot be investi-
gated independently. Representing the earliest known Flora, the
individuals composing it are, as might be expected, more unlike those
now living, than what any subsequent formation contains. The suc-
ceeding beds present us with plants, which occupy, in point of organi-
zation, as in date of creation, a middle position; and it is in many
cases through the investigation of these alone that a clue can be gained
to the relationship existing between the earliest known and the now
living vegetable forms. It is not so to an equal extent in the animal
kingdom. A knowledge of recent shells, for instance, can be brought
to bear upon those of the Silurian formation (independently of any study
of their allies in the more modern strata) far more effectually than an
equal acquaintance with living plants can, upon those preserved in our
coal-fields. Many and sufficiently obvious are the reasons for this:
the Silurian rocks contain but one or few orders of animals, the car-
boniferous many of plants. There is a greater external similarity
between the shells of all periods—they are better preserved, and their
external characters afford surer indications of their affinities, habits,
and localities. -

An examination of the coal vegetation being merely a comparison of
its tribes of plants with those we are more familiar with, the first object
of the vaturalist is, to reduce all the strange individual forms he here
for the first time sees, to the same classes and orders with existing
ones. When their affinities cannot be traced, he seeks to ally them to
living analogues, and thus, reproducing the whole Flora, he regards it
as probably characteristic of such physical features of soil, surface, and
climate, as accompany what he has determined to be the existing
types of the by-gone Flora. The general laws now affecting vege-
table life, are the only ones available in this comparison, and therefore
are adopted as correct; but to appreciate the extent of their application,
a very comprehensive knowledge of the distribution of plants is
necessary. Slight local causes may very materially modify the opera-
tion of these laws; and so plastic is vegetation under their influence,
that we find what appear to be entirely analogous positions with re-
gard to heat, light, soil, and moisture, tenanted. by whole genera, and
even orders of plants, of very opposite botanical characters, and that

392 VEGETATION OF THE CARBONIFEROUS PERIOD

such localities present a greater disparity of vegetation than do other
countries more remote in geographical position, and with less simi-
larity in their conditions.

It is the case with very many species of existing plants, that they
vary so considerably at various parts of the area over which they are
dispersed, as to draw all but those who know the intermediate links
(which may be comparatively scarce) into a belief, that the extreme
varieties are specifically distinct. This is eminently true of ferns, which
have very wide ranges, and are exceedingly sportive. If the difficulty
be great with living plants, of which complete specimens or whole indi-
viduals are procurable, it must be far more so with fragmentary fossils ;
and the coal formation being characterized by ferns to a very remark-
able degree, it follows, that with only imperfect specimens, all attempts
towards determining the species and limiting them must be very vague.
The amount of variation also is fluctuating, and it bears no necessary
reference to botanical affinity ; for whilst nine species of a genus may be
constant to their characters wherever they occur, a tenth may vary so

-widely that its extremes will appear far more dissimilar than are any
two of the other nine.

The knowledge of recent botany, which is needful to throw light

upon the study of fossil plants and the origin of coal, must be both.

varied and extended; though a profound acquaintance with any par-
ticular branch is not required to make a very considerable progress.
Those points with which the student should be most familiar are some
of them purely botanical ; whilst others are more general, and refer to
the dependence of vegetation upon the condition of the area it covers.

Some acquaintance with systematic botany is the first requisite :
through this alone can any approximation to the living affinities of the
fossil be obtained. It should embrace not only a knowledge of the
principal groups or natural order under which all plants are arranged,
but a familiarity with vegetable anatomy; for when the stem or trunk
alone is preserved, which is often the case, a minute examination of
its tissues is the sole method of determining its position in the natural
series.

A solution of the difficulties which this special knowledge will tend
to remove is of the highest interest to botanists, though comparatively
preliminary to the object of the geologist, whose inquiry is, what were the
general features of such a vegetation as has effected the formation of a
seam of coal, both as regards quantity and kind. As regards quantity,
inasmuch as the growth was either wonderfully rapid, or more tardy,
but prolonged under uniform conditions ; and as regards kind, from cer-
tain species, genera, or orders, iene particularly adapted by their
quick growth, their gregarious habits, and their continued appropri-

a,

AS COMPARED WITH THAT OF THE PRESENT DAY. 393

ation of certain areas to produce those vast accumulations, the expla-
nation of whose origin is still an unsolved problem. Other questions,
which a study of living plants alone can answer, refer to the sorts of
plants best calculated to thrive in such a uniform soil as the under-
clay, upon which each bed of coal rests, and into which some of the
vegetables have certainly been rooted. What form of surface is best
fitted to retain so mobile a débris as the coal was previous to its
compression and hardening; what degree of dryness would be most
favourable to such an accumulation, consistently with an energetic
growth of vegetation.

The above considerations pre-suppose some general ideas of the
vegetations both of the tropics and cooler latitudes, of mountain-chains,
table-lands, valleys, and estuaries, more especially of countries cha-
racterized by equable or by excessive or extreme climates, as compared
with continents, and of humid and desert districts ; in short, of all the
complex associations with, or dependence of botanical characters upon,
surface, soil, and climate, which the globe presents.

The want of this kind of information amongst many naturalists, and
the neglect of its application by others, have caused those utterly con-
tradictory opinions which have been expressed regarding the origin of
coal,* and unnecessarily complicated the subject. The botanist must
not seek to force a plant into a natural order, the habits of whose ex-
isting species are incompatible with those conditions under which a more
comprehensive view of the coal formation may assure him it must have
vegetated ; nor can the geologist put forward any theory which will
explain the features of that formation, if it be grounded on views op-
posed to those few certain data, which a study of the botany of the
period in question has afforded.

There is another branch of this investigation of equal importance to
the geologist and botanist, namely, the identification and comparison of
the species from different and sometimes remote coal-fields, or from the
various strata of the same field. This is as difficult as any of the
points which occupy the botanist ; and all questions connected with the
geographical distribution of the plants of that period being dependent
on the results thus obtained, it is one which requires extreme caution
in the working. The obvious tendency in the student is to regard as

* The looseness of the speculations hitherto advanced on the relationship of the coal flora
to such physical conditions as climate, cannot be better illustrated than by the fact that the
Sigillarie (which have undoubtedly contributed largely to the formation of coal) are
considered by some naturalists to be allied to the order Euphorbiacee, by others to Cacti,
and by the majority to ferns. The necessary conclusion to which those who place them
in the first two orders would lead us, is, that they were inhabitants of singularly arid and
desert countries ; whilst, if ferns, they are characteristic of diametrically opposite condi-
tions, a moist soil and a humid atmosphere.

394 VEGETATION OF THE CARBONIFEROUS PERIOD

identical the similar fossils in the various strata exposed in one mine ;
and as different the plants from remote coal-fields. From recent ob-
servations, it appears that subsequent movements may have isolated
portions of what once formed a continuous bed of coal, characterized by
a uniform vegetation throughout, and that hence a slight dissimilarity
between the plants of each portion may be attributable to a difference
in the conditions to which it was exposed in each. On the other
hand, it must be borne in mind, that at the present day a change in
position is almost surely accompanied by a very considerable change
. in vegetation. The labour of identification too is not confined to the
comparison of specimens, but includes the determination of their names
when previously described. This is often all but impossible, from the
nature of the specimens, and the difficulty of presenting them, in an
available form, without plates. Hence it happens, that the labour of
individual observers is overlooked. It is, perhaps, impossible to employ
similar materials to better purpose than has the author of the “ Flore
Fossile” those upon which he laboured; and yet the difficulty of
naming the species by that work is very great, and must be so; for the
specimens to be compared are, like the originals, mere fragments, and
the genera of Ferns adopted are far from being properly defined, though
as judiciously as the materials would permit. On the contrary, many
of these are not supported by the examination of living analogues ;
whilst others are unavoidably founded on isolated portions of plants,
whose appearances whilst living and affinities are alike unknown, whether
amongst their contemporaries by which the world was then inhabited,
or those hitherto unrecognised allies that may now surround us.

The foregoing remarks admit of illustration, to a certain extent,
by particular instances. This may be useful, because indicating to the
student those errors into which he is most liable to fall. Since, however,
he may not be aware how closely the course of investigation pursued
in the examination of a living Flora ought to be followed in studying a
fossil one, it is, perhaps, well to enumerate those steps by which a know-
ledge of both can be obtained.

As a field for botanical research there is none so novel as the eoal
formation, the few yards of shaft being more than equivalent to the longest
voyage, in respect of the amount and kind of difference between the ve-
getation the naturalist is acquainted with and that he seeks to under-
stand. Whatever be the nature of the vegetation to which the botanist
is transported, he commences by observing :—1. What are the orders,
genera, and species of plants characterising the Flora; their mutual
affinities, and their relations to those of other countries ; their numbers,
and the relative proportions which the natural groups under which they
arrange themselves, bear to each other. 2. The geographical distribu-

AS COMPARED WITH THAT OF THE PRESENT DAY. 395

tion of the species, &c. ; their extension over the surface of the country,
and the replacement of one kind by another. 3. The relations which
may be traced between the species and the soil to which some are
peculiar, whilst others are common to it, and to very different soils ; the
quantity of moisture, heat, and light they require, and the effects of a
diminution or increase of any of these elements. 4. The reciprocal
influence of the whole mass of the vegetation upon the surface it
covers; the new soil, or alteration of the old, produced by its decay;
the extent and composition of accumulations of dead matter; the par-
ticular kinds of plants contributing most largely to, and the consequent
nature of, such deposits; the conservative influence of the vegetation upon
this deposit, which may be retained by roots, and sheltered by foliage
from the action of elements, which, in the absence of these protections,
would rapidly sweep it away.

An enumeration of these points, viewed in their bearing on the
subject of the Coal-flora, will show how limited is our knowledge of
any one of them, compared with what might be acquired from a very
superficial examination of any recent flora, or with what the geologist
may obtain from an inspection of the animal remains in many strata.

1. Of the mutual affinities of the groups under which the majority of
the genera of coal-plants arrange themselves, little more can be said,
than that the Ferns occupy the lower end of the series, and the Conifere
possibly the highest ; but this depends upon the view taken of the
‘affinities of Sigillarie, the most important group. These are classed
by some observers amongst Ferns; by others, with Conifere ; another
considers them as linking these two widely-different families ; whilst a
fourth ranks them much higher than either. The affinities of another
group, Calamites, are entirely unascertained. Of the whole amount of
species in each, no conjecture can be formed ; or any but a very rough
one of the number into which those with which we are familiar, as of
common occurrence, should be divided. The ferns far out-number,
probably, all the others ; but this again materially depends on the value
accorded to the markings of Sigillarie, as means of dividing that genus ;
for if the slight differences hitherto employed be insisted upon, the
number of the so-called species may be unlimitedly increased.

2. With regard to the geographical distribution of the species, &c.,
it appears that an uniformity once existed in the vegetation throughout
the extra-tropical countries of the northern hemisphere, to which there
is now no parallel ; and this was so, whether we consider the coal plants
as representing all the flora of the period, or a part only, consisting of
some widely-distributed forms that characterized certain local con-
ditions. Nor is this uniformity less conspicuous in what may be called
the vertical distribution, the fossils in the lowest coal-beds of one field,

396 ’ VEGETATION OF THE CARBONIFEROUS PERIOD

very frequently pervading all the succeeding beds, though so many as
thirty may be interposed between the highest and the lowest.

3. Of the relations between the soil and the plants nourished by it,
little more is recognizable, than that the Sigillarie have been particu-
larly abundant on the under-clay, which, judging from the absence of
any other fossils but Srgillarie roots (Stigmaria), seems to have been
either, in itself, unfriendly to vegetation, or so placed (perhaps from
being submerged) as to be incapable of supporting any other. The latter
is-the most probable, because both Sigzllarie and their Stigmaria roots
occur in other soils besides under-clay, and are there accompanied by
Calamites, Ferns, §c. 'The Conifere, again, are chiefly found in the
sandstones; and their remains being exceedingly rare in the clays,
shales, or ironstones, it may be concluded that they never were asso-
ciated with the Sigil/arie and other plants which abound in the coal
seams ; but that they flourished in the neighbourhood, and were at times
transported to these localities. The quantity of moisture to which
these plants were subjected must remain a question, so long as some
authors insist upon the Sigi/larie being allied to plants now character-
istic of deserts, and others, to such as are the inhabitants of moist and
insular climates. The singularly succulent texture and extraordinary
size: of both the vascular and cellular tissue of many, possibly indicate a
great amount of humidity. The question of light and heat involves a
yet more important consideration, some of the coal-plants of the Arctic
regions being considered identical with those of Britam. How these
can have existed in that latitude, under the now-prevailing distribution
of light and heat, has not been hitherto explained: they are too bulky
for comparison with any vegetables inhabiting those regions at the pre-
sent time, and of too lax a tissue to admit of a prolonged withdrawal
of the stimulus of light, or of their being subjected to continued frosts.

4. The consequence of the existence of the coal-plants has been the
formation of coal; but how this operation was conducted is a question
still unsolved. The under-clay, or soil upon which the coal rests, and
upon which some of the plants grew, seems in general to have suffered
little change thereby, further than what was effected by the intrusion
of a vast number of roots throughout its mass. The shales, on the other
hand, are composed of inorganic matter, materially altered by the
presence of the vegetable matter which they contain. The iron-clays
again present a third modification of this mixture of organic and in-
organic matter, often occurring in the form of nodules. These nodules
seem to be the result of a peculiar action of vegetable matter upon
water charged with soil and asalt of iron, the iron-stone nodules of *
existing peat-bogs appearing altogether analogous to those of the
carboniferous period, whether in form or in chemical constituents.

AS COMPARED WITH THAT OF THE PRESENT DAY. 397

Here, then, the botanist recognizes in one coal-seam a vegetable
‘ detritus under three distinct phases, and which has been acted upon in
each by very different causes. In the under-clay there are roots only :*
these permeate its mass, as those of the water-lily and other aquatic
plants do the silt at the bottom of still waters.

The coal is the detritus either of those plants whose roots are pre-
served in the under-clay, or of those, together with others which may
have grown amongst them, or at a distance, and have been afterwards
drifted to the same position.

Above the coal is the third soil, bearing evidence of the action of a
vigorous vegetation ; this is the shale, which has all the appearance of a
quiet deposit from water charged with mineral matter, and into which
broken pieces of plants have fallen. Here there is so clear a divisional
line between the coal and shale, that it is still a disputed point whether
the plants, contained in the latter, actually grew upon the former, or
were drifted to that position in the fluid which deposited the mineral
matter. Amongst the shales are also interspersed, in many cases, in-
numerable stumps of Sigillaria, similar to those whose roots occur in
the under-clay, and which are themselves found attached to those roots
in soils similar to the under-clays, but unconnected with any seam of
coal. These stumps are almost universally erect, are uniformly scat-
tered over the seams, and otherwise appear to have decidedly grown on
the surface of the coal: the shales likewise seem deposited between these
stumps. The rarity of Sigillarie roots (Stigmari) in this position, is
probably due to their being incorporated with the coal itself, though
they sometimes occur above that mineral and between the layers of
shale. The seams of iron-stone (or black-band) are the last modifica-
tions of soil by vegetable matter, to which allusion has been made: when
these are uniform beds or layers, they may be supposed to be the
deposit from water charged with iron and soil which has percolated
through the peat, and in so doing absorbed a great deal of vegetable
matter. “The layers of nodular iron-stone are simple modifications of
these, and may be caused by the sedimentary particles contained in the
fluid, which, instead of being deposited in a uniform stratum, are aggre-

* The absence of other parts of plants, and indeed of any plant but the roots of Sigil-
larie, in the under-clay, appears a fact of considerable importance, In the first place, it
indicates that that soil was in a condition unsuited to the growth of other vegetables (as
mentioned above), whose seeds might have accompanied those of that genus on its pre-
viously-naked surface. In the second place, this absence of other fossils in the under-
clay is opposed to the theory of the drift-origin of the vegétable matter comprising
coal; for there is nc interstratification of coal with this subjacent deposit, which might
haye been expected to occur over some portion of an extensive eoal-field ; whereas, the
gradual decay of these plants, whose roots struck into the under-clay, would produce a

uniform bed of peat, perhaps adapted to the growth of those ferns and other plants which
are fossilized in the superincumbent shales.

2

398 VEGETATION OF THE CARBONIFEROUS PERIOD

gated round broken bits of vegetable matter (as fern leaves, stems or
cones) which served as nuclei.

Now, though each of these points admits of some explanation when
taken independently, and some illustration from the action of an existing
vegetation upon soil, &c., it is very difficult to understand their com-
bined operation over so enormous a surface, for instance, as one of the
American coal-fields, and even more to account for their regular recur-
rence, according to some fixed law, in every successive coal-seam
throughout the whole carboniferous formation. These are problems of
the highest order and unsuited to this sketch, the remainder of which
shall refer to the plants themselves, and especially to those botanical
characters according to which the Coal Flora-has been grouped and
named, and to an illustration of these several points by a comparison of
them with what are afforded by recent plants.

ON THE PROBABLE EXTENT OF THE FLORA OF THE COAL
FORMATION IN BRITAIN.

No fewer than 300 species of plants have been enumerated as belong-
ing to the Coal Flora of Great Britain; but whether this gives any
approximation either to what was the amount of their species at one
period, or even to all those which contribute to form the cval, it is im-
possible to say. It need hardly be observed, that a collection of the
fragments imbedded in our most recent deposits is no index to the
general mass of the vegetation, nor are the remains necessarily those
of the commonest plants, or even of such as would a priori be judged
the best suited for becoming fossilized. That hitherto unknown species
do exist in an available state for the botanist cannot be doubted: they
are of frequent occurrence ; but that these are not so numerous as might
be expected from the enormous magnitude of a coal-field, is evident,
from the great uniformity that prevails throughout the formation. It
may indeed be a query, whether the number of species still to be
discovered will equal in amount that of the so-called species, which
being founded on imperfect specimens, will ultimately prove to belong
to previously described forms.

That the vegetation of the carboniferous period, whether confined to
the coal veins or not, was highly luxuriant, cannot be disputed. The
enormous bulk of carbon accumulated, and the prevalence of ferns
in all the fields, and the great size to which so many soft-tissued plants
attained, all prove this fact. A luxuriant vegetation is, however, no
index to a varied one; and, as many of our modern woods and even
great arez of tropical forests consist of but a few species multiplied ad
infinitum, so may the forests of the carboniferous period have been com-
posed of but a few Sigillarie and Lepidodendrons, sheltering an under-

———

AS COMPARED WITH THAT OF THE PRESENT DAY. 399

growth of a limited number of kinds of fern,* for a very limited number
of them (comparatively speaking) if as protean as some of their allies
are in our day, would embrace all the known species of the Fossil Flora.

In the temperate latitudes particularly, a recent Flora, marked by a
preponderance of ferns, is almost universally deficient in species of
other orders; as is thus shown. 1. Where one species prevails over a
considerable area, as the bracken (Pteris aquilina) does in parts of
Britain, and the P. esculenta in Van Diemen’s Land and New Zealand,
it generally monopolizes the soil, choking plants of a larger growth on
the one hand, and admitting no under-growth of smaller species on
the other. 2. A luxuriant vegetation of many species of ferns, con-
tinued through a great many degrees of latitude or longitude, especially
in the temperate regions of the globe, generally indicates a uniformity
of temperature throughout that area, and a paucity of species of flower-
ing plants. A comparison of the vegetations of Tasmania and New
Zealand illustrates this. The former of these islands, barely 200 miles
long, contains four times as many species of flowering-plants as New
Zealand, whose total length is 900 miles. On the other hand, this
latter country possesses more than four times as many kinds of fern as
Tasmania, and they are so uniformly distributed over its area, that
almost all those which are found at the southern extremity of the
island prevail also at the northern. The West Indian and Pacific
Islands again present a Flora remarkably rich in ferns, and in both
these instances we have very many of the species uniformly spread over
an enormous surface, in the one instance, from the Windward Islands
to Mexico, and in the other from New Zealand to the Society and Sand-
wich Islands. Take on the other hand the campos of Brazil, the sandy
flats of Southern Africa, and the somewhat similar plains of Australia,
and sterile though they appear at first sight, they will be found to
abound in many kinds of flowering plants, but unaccompanied with ferns.

This prevalence of ferns has been long adduced in proof of the
climate of the carboniferous period being temperate, equable and
humid ; and so it no doubt was; but I am not aware that it has been
hitherto regarded as probable evidence of the paucity of other plants,
and the general poverty of the whole Flora which characterized that
formation. If, however, the laws of existing vegetation are to be con-
sidered as having had equal force at that time when the fossil one

* This preponderance of ferns over flowering plants is common to many tropical islands,
and not confined to the smaller of them, as St. Helena and the Society group. In extra-
tropical islands, too, as New Zealand, I have collected as many as 36 kinds of fern in an
area not exceeding a few acres : they gave a most luxuriant aspect to the vegetation, which
presented scarcely a dozen flowering plants and trees besides. An equal area in the
neighbourhood of Sydney (in about the same latitude) would have yielded upwards of 100
flowering plants and but two or three ferns.

400 VEGETATION OF THE CARBONIFEROUS PERIOD

flourished, we must conclude that the predominance of ferns in general,
and of certain species of Pecopteris (a fern apparently allied to our
Pteris) over a great area, together with the remarkable similarity of the
English fossils with those of North America, are all indications that
the Flora of that period was poor in number of species.

Let it not be supposed that this prevalence of an order, which
in point of complexity of structure is low in the system of plants,
is a fact favourable to the hypothesis that the vegetation of which
it appears to form a large part, was less highly developed than
what succeeded it. We know too little of the structure of the ferns
of that day, to pronounce them either more or less complete than their
allies of the present time; while of the Liycopodiacee* it may be safely
asserted, that they were of a form and stature far more noble, and in
structure more complicated than any plants of that order now existing.

ON THE MOST PREVALENT GENERA OF PLANTS BELONGING TO THE
COAL FORMATION, AND THE CHARACTERS EMPLOYED IN THEIR
CLASSIFICATION, AND THAT OF THE SPECIES THEY CONTAIN.

§ Ferns, General Remarks.

Numerous as are the dissimilar groups of plants (whether genera
or orders) scattered through the various strata of the coal formation,

Fig. 1.

ORG

Re ;

SJ

f Ww
Pry oT

——

Pecopteris heterophylla (Coal-fields, Newcastle).

* See ‘‘ Essay on the Structure and Affinities of Lepidostrobus,’ in the present volume
of Survey Reports.

AS COMPARED WITH THAT OF THE PRESENT DAY. 401

there is but one which presents any obvious or recognizable close
relationship with an order now in existence. Such is that of the ferns.
These were not only undoubted ferns, such as we have now living, but
one of the many fossil genera included under this order is probably
identical with a living one, though none of the species comprised in it
are alive now. The genus here alluded to is Pecopteris (Fig. 1), the

fossil representative, if not congener, of the modern Pteris (Fig. 2).

Fig. 2.

\\
uN ENG
AAR

Pteris esculenta (New Zealand).
It is not improbable that there are other genera of living ferns fossilized
in the shales of the coal formation, but if so they are not so well pre-
served, probably from not offering the facilities for petrifaction (im a
determinable condition) that Pecopteris does. The perfect state in

402 VEGETATION OF THE CARBONIFEROUS PERIOD

which this genus is often found is attributable partly to its charac-
teristic fructification being copiously produced, to its fronds being of a
hard texture and coriaceous consistence, and to the absence of hairs
or glands upon the under surface where the spores are produced, and to
the large size, smoothness, and evenness of the involucre, or organ pro-
tecting these spores. A few other ferns of this formation present the
fructification well preserved, and especially the genus Senftenbergia of
Corda* (Fig. 3, 1 & 2), which is allied to the recent Anetmidictyon
(Fig. 3, 3) ; but in no case except Pecopteris can the genus be thereby,
even provisionally, identified with any living one.

Fig.3.
1

The ferns are a group of plants, whose genera and species vary
extremely in the form of their fronds, and are sometimes so bizarre that
without the organs of fructification it is impossible to ascertain even on
the order they sometimes belong to. Now the fruit is so universally
present and abundant in recent ferns, that there is no difficulty in
defining the limits of the order as it now exists ; but in the case of fossil
ferns, in most species of which the fruit is of rare occurrence, and par-
tially obliterated by fossilization, it is impossible to decide whether

* Corda “ Flora der Vorwelt,” Tab. 57, fig. 2. I have introduced a copy of Corda’s
figure of the remarkable fructification of this plant ; as from the scarceness and high price
of that work it must be inaccessible to most of our readers,

ee a a

AS COMPARED WITH THAT OF THE PRESENT DAY. 403

many of the genera put into the order (as Asterophyllites,* Spheno-
phyllum, &c.) really belong to it or not, until some evidences be gained
from their fructification.

This absence of fructification not only prevents our defining the
limits of the fossil ferns as an order, but deprives the botanist of one of
the most important generic characters that the plants comprised in
it afford. As will be hereafter shown, it is generally impossible to trace
the affinities of a fern without the fruit, and often even to refer it to the
section or subdivision in which it should be placed.

There are two other points which cannot fail to strike the observer of
fossil ferns: they are, the absence of a singular aspect of fronds in a
state of incipient vernation, that is curled up like a crozier, as the
fronds of all are in a young state. In the perennial existing species,
fronds in this state may be seen on the plants at all seasons, and even in
the annuals they are visible for many months; so that it were diffi-
cult to select a season of the year during which a modern fern Flora
could be fossilized without very many such fronds being preserved, as
well as fully expanded ones ; but in the coal formation these are exces-
sively rare. That they do occur is evidence that the evolution of the
leaves followed the same law then as now ; while this extreme infrequency
would seem to indicate, either that the majority of the species were
annual, and fossilized at an advanced season of the year, possibly coin-
ciding with a sudden depression of the surface; or that they were
perennial, and the old fronds dropping from time to time off the parent
trunk, were subsequently fossilized.

Again, the almost universal absence of any defined termination
to the stalks (stipites) of the fronds, or of any such masses of woody
roots or rhizomata as those of modern ferns, is highly remarkable.
In the case of recent species we find long and strong under-ground
woody roots, or more frequently great knotty masses above the surface
of the earth, from out of which the fronds spring. But in the case of
fossil ferns, there is hardly such a thing known as a specimen presenting
two fronds attached to one stem. This may readily be explained on
the supposition that the ferns.are all fragments, transported to the
position they now occupy, an hypothesis gravely objected to by many
close observers on geological grounds. Another suggestion which
offers itself is that the fronds are the deciduous ones of tree-ferns ; this
is much more plausible, though almost necessarily involving the con-

* My friend, Mr. Bunbury, informs me that he is possessed of American specimens of
this genus with the organs of fructification preserved.

+ Several authors have figured small ferns, in which both the fronds, their axis of sup-
port, and root, are well preserved; but the species are very scarce, much smaller, and
of a different character to the prevailing forms of the coal formation,

II. a 3

404 VEGETATION OF THE CARBONIFEROUS PERIOD

clusion that Sigillarie are the trunks of those tree-ferns, a subject
to which I shall revert at a future part of this essay. A third and the
only other explanation that I can propose is, that the roots were planted
in the peat, and with that are now turned to coal, thus being as effectu-
ally obliterated as are the roots of those numerous upright stumps
of Sigillarie which I consider to have undoubtedly grown on the surface
of the coal, and spread their roots into, or along the surface of its mass.

No fewer than one hundred and forty species of ferns are enumerated
as having inhabited those few isolated aree in England over which the
coal has been worked, at the time when the latter was formed. This
is a strange contrast to our existing Flora, which boasts but 50 species
of that order, upon a surface of incomparably greater extent than what
we have examined of the carboniferous period. It is, indeed, doubtful
whether all the fronds now in Great Britain would equal in number
those contained in the largest seams; so that under whatever light the
predominance of the ferns be regarded, whether in amount of species or
specimens, they indicate a climate far different from the present. I
have before said, that it is only in the tropics, and in the equable, moist,
higher latitudes of the southern hemisphere, that any remarkable
luxuriance of ferns is met with. A climate warmer than ours now is
would probably be indicated by the presence of an increased number
of flowering plants, which would doubtless have been fossilized with
the ferns ; whilst a lower temperature, equal to the mean of the seasons
now prevailing, would assimilate our climate to that of such cooler
countries as are characterized by a disproportionate amount of ferns.
This then is an argument unfavourable to the theory of central heat
having warmed the surface, or of the direction of the poles being so
altered as to have exposed Great Britain to a tropical climate, and
demanding only a different disposition, and perhaps proportion of land
and water to that now existing ; judging from the southern hemisphere,
where it is seen that the relative proportions of land and water modify
the Flora most materially.

With regard to the distribution of these fossil ferns out of Britain, it
appears that their ranges were as wide as are those of the present day ;
perhaps more so, when we consider how extremely difficult is the deter-
mination, and no less the identification of the species and specimens,
and that the general tendency is probably to the multiplication of
species. Of the British species, about 50 are known to occur in the
coal-beds of North America, some of them ranging there from the
latitude of Nova Scotia to 35° S., and abounding in these and various
other intermediate coal-fields. On the continent of Europe again there
are about as many of the British species.

Again, turning to the living ferns, we find that of the 50 inhabiting

AS’ COMPARED WITH THAT OF THE PRESENT DAY. 405

Great Britain, little more than a half are found in North America,
A greater similarity between the climates of these countries is required
to account for the preponderance of the species common to both during
the carboniferous epoch, and possibly the presence of land between the
two to have favoured the transmission of seeds from one to another.
That such an aid as an intermediate land was necessary, however, may
be'disputed, on the ground that the Lycopodia of Europe are all found
in North America, their spores being apparently no smaller nor easier
wafted than are those of the ferns.

Nothing satisfactory has been elicited regarding the vertical dis-
tribution of the species, beyond the fact that no decided difference
between those of the uppermost and lowermost beds has been hitherto
ascertained. A few of the species are found in the Old Red Sandstone,
but none enter the superior beds, as the Oolite, &c.

Ferns—Botanical Remarks, Se.

Habit.—Upon this point we absolutely know nothing. Although the
fronds occur in countless myriads throughout all the beds, they offer no
characters, either relative. or individual, by which we can pronounce
whether they were terrestrial or epiphytal (growing on trunks of trees),
if the stems were erect, inclined, or creeping; nor, what is most
remarkable of all, do the fronds ever occur attached ; so that we are
ignorant whether any or all of these kinds belonged to tree-ferns,
or were humble individuals, with stems scarcely rising above the
ground.

This ignorance of the habit of the plants is a most serious drawback,
and till it is removed we have little hope of gaining a clear idea of the
features of that vegetation to which they so largely contributed. It
may be urged in favour of a great proportion being arborescent, that
none are found attached, as stated above, to roots or slender stems; and
that some.may have belonged to Sigillarie, with whose foliage we are
otherwise unacquainted. Against this supposition, again, stands the
extreme rarity of any acknowledged tree-fern stems in the coal-beds,
and that whether drifted or deposited on the spot, the fronds must, if
arborescent, have been accompanied by such stems.

The infrequency of fructification upon the fronds of the fossil ferns
belonging to this formation appears as possibly another argument in
favour of many of those appertaining to tree-ferns; for, while the
herbaceous and caulescent ferns of New Zealand are scarcely ever
barren, the arborescent species are almost invariably so. I think I am
safe in saying that of two or three kinds of New Zealand tree-fern, not
one specimen in a thousand bears a single fertile frond, though all

abound in barren ones.
24 2

406 VEGETATION OF THE CARBONIFEROUS PERIOD

Fructification.—The sporules (fruit or seeds) of the ferns are invari-
ably produced on the under-side of the frond, and as that is all but
universally the concave surface, and generally presents other inequalities
beyond those caused by the organs of reproduction, it becomes firmly
attached to the rock during petrifaction, and very rarely exposed to
view. Hence, when the frond is coriaceous, it cannot even be deter-
mined whether the plant was in fruit or not, except in such rare case
as that of Pecopteris lonchitica, to which I have already alluded. When,
however, the texture of the frond is membranous, the position of the
sori (or clusters of spores) is indicated by a prominence on the upper
and exposed surface. This stamping through is usually the only appa-
rent sign of fertility in fossil ferns, and is conspicuous in several British
species (as Pecopteris obtusifolia) ; also in three American species, indi-
cated by Mr. Bunbury,* who has lately drawn attention to the circum-
stance. Lindley and Hutton were, I believe, the first to point this out
as the probable cause why the fructification of fossil ferns should be very
seldom apparent, even supposing fruiting specimens more frequent than
they really are.t| Goeppert { again finds that in producing artificially
fossilized ferns, the under surface invariably remains attached to the
matrix, which separates from the upper.

It is, unfortunately, impracticable to class the most perfect specimens
of living ferns without the aid of characters derived from their sori.
Much more is this the case with the imperfect ones of the fossil. The
former indeed are almost universally rejected from the Herbarium,
many species being in every other respect identical.

In illustration of this fact two wood-cuts have been prepared, each
displaying how great a difference may occur in fructification unaccom-
panied by any disparity in the outline, form, surface, texture, or vena-
tion of the frond.

The first (Fig. 4) represents a portion of a frond whose outline and
venation answer either for a species of Leptogramma (A) or of Asple-
nium (B), genera which are precisely alike in those points ; but if the
fructification be regarded, it will be seen, that the sori of the portion,
marked (B) and enclosed in the black line, are covered with a protecting
scale (involucre or indusium) as is more distinctly shown in the magni-
fied portion, whilst those of the rest of the figure are naked. Now this
absence of an involucre in Leptogramma is not accidental in that fern,
but a constant character, separating both it and many allied species
(all equally wanting the involucre) from Asplenium, and a similar group
which possess that organ.

* Fossil Flora sub Meuropteris undulata, t. 83.
t “ Quarterly Journal of London Geolog. Soe.,” yol. ii. p. 85.
t “Systema Filicum Fossilium,” p. 293.

4

AS COMPARED WITH THAT OF THE PRESENT DAY. 407

At Fig. 5 again a portion of a fern frond is figured, which, when out
of fruit, would be referred indiscriminately to one of four species, differ-
ing only in the form, protection, or position of their sori. Thus, at

Fig. A, the sori are placed on the middle of a venule, and are unpro-
vided with an involucre: were the whole frond thus characterized it
would be that of Goniopteris crenata. At B the sori are elongated,
covering the whole venule; it is Stenogramma aspidioides. Fig. C
shows the sori to be situated at the middle of a vein (as in A), but
they are here covered with an involucre, and not exposed ; such is the
case with Nephrodium glandulosum. The sori at D again are naked
and round (as at A), but instead of being placed in the middle of a

408 VEGETATION OF THE CARBONIFEROUS PERIOD

venule, they occur only where two venules meet : were the whole frond
so furnished it would represent Meniscium cuspidatum.*
Thus we have four plants, wholly indistinguishable except by their

Fig. 5.

* To illustrate still further the confusion which the absence of fructification might
entail, it may be mentioned that three of these ferns are natives of Jaya, and would, if
petrified without the fructification being displayed, be unsuspectingly referred to one
species ; the fourth, a native of Jamaica, would, if similarly fossilized, be adduced as an
instance of the similarity in the vegetable productions of the Old and New World at the
epoch when they both flourished! Yet four more really and truly distinct plants cannot
well be. -

es

Se Ss

AS COMPARED WITH THAT OF THE PRESENT DAY. 409

*

fructification, which proves them to belong to different groups of the
order. It has been already remarked that the fruit is rarely present in
fossil ferns; whence it may be presumed, that did the four ferns just
described occur fossil, the probability is great that whatever formation
they inhabited, or however widely separated they were in geographical
position, they still would all be included under one species.

Fig. 6.

Again, very many existing species of ferns have the fertile frond
quite distinct from the barren ones in size and shape. The Niphobolus
rupestris of New Zealand is an instance: a wood-cut of it (Fig. 6) is
added. In such a case the absence of any fructification on the one
hand, and the severance of both the fertile and barren fronds from the

410 VEGETATION OF THE CARBONIFEROUS PERIOD

caudex previous to fossilization, would effectually prevent the two parts
being recognised as portions of one plant.

The tendency in this case is towards the multiplication of species,
while in the former we should unite what are nearly totally distinct.

Venation. In the absence of those characters afforded by fructification
which are acknowledged to be of the very highest value in this order,
if indeed they are not the only ones by a study of which the primary
groups of ferns are to be united, the venation, or arrangement of the
nerves, is adopted. This character is considered by many of secondary
importance, and one by which the primary groups may be divided
into minor groups without violence to nature, especially according to
whether the veins after branching join again, or continue free to the
margin of the frond. Now, there are doubtless many sub-orders both
of fossils and recent ferns distinguished by the organs of fructifica-
tion, each of which again is sub-divided by characters in the venation ;
but instead of adopting the primary characters (of fructification) in
the formation of these sub-orders, the secondary ones (of venation)
are employed: it follows, then, that every such group (and they are
genera) of fossil ferns may contain species generically widely apart.
The genera must, of course, be wholly artificial, and include plants
belonging to separate sub-orders, as would be proved did their fructifi-
cation occur.

Even venation is a character to be used with caution ; for though the
presence or absence of an involucre to the sorus is constant in one
species, and so are the form and position of the sori, the venation is found
occasionally to vary materially in different parts of the same plant, or
even the same frond. Hence, while some sub-orders of ferns may be
trenchantly divided according to the branching of the nerves, others
cannot, because the individual species at times assume both forms in
one frond.

This is remarkably the case with the Diplazium Malabaricum,* of
whose fronds two states are figured (Fig. 7). In the upper and older
portions it will be seen that the veins, after branching, meet; while in
the lower the branches run free to the margin of the frond. Fossil frag-
ments of this fern would be included under two genera, those as resem-
bling Fig. A to Callipteris, and those veined in B to Digrammaria. It
would be out of place to exhibit through what transition stages the pas-
sage between A and B iseffected: any one frond, however, of this par-

* This plant has caused no little perplexity to the students of existing ferns. So pro-
tean are its fronds, that different states of this one fern haye been made into two separate
genera, Anisogramma, Pres., and Digrammaria, Presl. Mr. John Smith, of Kew, to
whom I owe many illustrative specimens of these points, has separated this plant from
Diplazium, and given it the name of Callipteris,

AS COMPARED WITH THAT OF THE PRESENT DAY. 411

ticular species shows it, and would afford material, if fossilized, for
many. bad species.
Generally speaking, however, venation is a character of the greatest

Coen a

ayn. wo OP GL

|

importance ; and to show how closely it should be scrutinized, an illus-
tration is added (Fig. 8) of a form of frond and fructification common
to two very different ferns, which are permanently and effectually dis-
tinguishable by the length of one branch of a repeatedly branched vein.

Fig. 8.

The general character of the frond of both A and B is to have the
veins alternately simple and branched, the simple one is in both pro-
duced only half-way towards the margin, and bears a sorus on its point,

412 VEGETATION OF THE CARBONIFEROUS PERIOD

the longer is in both three times branched, and bears a sorus on the first
or right-hand branch. But whereas in A (Cyclopeltis Presliana, Sm.)
the said first right-hand branch or venule is not produced to the margin
of the frond, and bears its sorus on the point, in B (Cyel. semicordata,
Sm.) the same venule is produced to the margin, and has the sorus on
its middle. Now this arises from no difference of specimens, but of
species, and it is the only character by which a plant exclusively in-
habiting the Old World can be distinguished from one confined to the
New.

Though such an anomaly as that of a growing fern bearing fronds
with two different types of venation is readily recognized amongst
recent plants whose whole frond is presented to the eye, it is not so
in the petrified specimen, of which isolated fragments must be classed
under different genera according to the arrangement of the veins.
This is probably a fertile agent in dismembering genera, an error no
less grave than that to which so slight a character as that distinguishing
Cyclopeltis Presliana from C. semicordata, would, if overlooked, lead ;
for hence we might specifically unite the widely different ferns of
two more differently circumstanced countries than Europe and North
America.

Lastly, many recent ferns are of so dense and leathery a texture
that their venation cannot be ascertained without much trouble, arising
from the necessity of macerating the frond and dissecting out the veins.
When such are petrified, they cannot be referred with propriety to any
previously established genus, the characters afforded by texture, and
comparative denseness of substance, being of specific importance only.

The conclusion to be drawn from the above resumé of the value of
the characters drawn from the venation of fossil ferns is that the pre-
vailing tendency of dwelling too much upon it leads to the dismem-
berment of species, and placing the individual parts under different
genera.

Outline and Division of Fronds.—A character considered next in im-
portance to venation, both amongst recent and fossil ferns.. It is seldom
of more than sectional or specific value, and frequently not that, the
species of a genus being frequently grouped into those with simple
fronds, those with the fronds lobed, deeply divided, once, twice, and so
forth.

Tn some cases of living ferns even the amount of variation is very
startling, for instance the Polypodium pustulatum (Fig. 9), where the
fronds are sometimes entire, at others deeply lobed; or still more
remarkably in the case of Lindsea cordata (Fig. 10, p. 414), whose fronds
vary so extraordinarily, that did the various forms occur petrified, and
detached from the parent plant, it would be impossible to deny specific

AS COMPARED WITH THAT OF THE PRESENT DAY. 413

value to some dozen, or more, of such isolated fragments as would
alone be found. Further instances of variation of form in the same
plant are given at Figs. 1 and 6 of the illustrative wood-cuts.

Fig. 9.

On the whole, it is probable that the irregularity of outline and
division, prevalent in recent ferns, is the most fertile source of error
in our investigations amongst the fossil, because the individual fronds
are universally detached, and are seldom seen entire, so that we are
ignorant to what portion of the plant the fragments belong. The
result is of course the multiplication of species to a degree only com-
mensurate with the protean nature of the species.

414 VEGETATION OF THE CARBONIFEROUS PERIOD

The zoologist is not equally hampered with the three classes of diffi-
culties described above. Take the mollusca for instance; there is
hardly an instance of two shells, identical in appearance, yet belonging
to different genera of that order, and only distinguishable by the animal
they contain, or by one organ of that animal ; far less of four genera of
shells so circumstanced, and each the type, not of a genus, but of
groups—groups, too, differing in geographical distribution as well as in
their most important characters.

Fig. 10.

Outline again, a fallacious guide to the determination especially of
generic affinity amongst plants, is of eminent importance in the animal
kingdom, where the habits of the different groups, all locomotive, de-
mand peculiar modifications of the hard parts, whether of the external
or internal skeleton.

Surxface.—The presence or absence of hairs or scales on the surface
of the fronds, and sometimes of glands also, affords characters of the
greatest constancy for distinguishing species, being often available when
the outlive of the fern is so variable, that by it alone the species cannot
be traced. The larger and most characteristic hairs or scales occur

AS COMPARED WITH THAT OF THE PRESENT DAY. Al5

chiefly at the base of the frond-stem or stipes, a part of the plant inva-
riably wanting in the fossil. Smaller scales, hairs, and glands are most
frequently confined to the under surface of the frond, which, from
various causes, and this amongst others, is almost invariably the one
that is united to the rock in petrified ferns, and therefore unavailable
for botanical examination.

Sigillaria.

Perhaps the most important plant in the coal formation, forming a
conspicuous feature in almost every field, appearing in all the strata, and
distributed from Spain to Scotland and from Virginia to Newfoundland
in America. Upwards of 60 species have been described ; with how little
precision may be inferred from the fact mentioned above, that many of
them have been proved mere varieties, in a solitary case even four
being reduceable to one.* Under the names of “ bottoms” and “ bell-
moulds” the stumps of Sigil/aria are well known to the colliers, as dan-
gerous associates of the shale roofs in the workings, immediately above
the coal. They are generally but a few feet high, though sometimes two
yards broad at their expanded bases, they are truncated at the top, and
retain their position in the shales after the removal of the coal upon
which they rested, being supported by the pressure of the atmosphere
and the cohesion of their smooth sides. Unless propped, they, after
a time, lose their hold and fall in, sometimes severely injuring the
workmen. So common are they, that I have in many South Wales and
other collieries counted five or six in the space of a few fathoms, always
suggesting the idea of the erect stumps of trees in a forest.

In the shales surrounding the bases of these stumps are found pros-
trate stems, bearing markings similar to what sometimes appear on the
stumps. It is, however, generally the case that the markings of the
stumps are very obscure or wholly obliterated, as usually occurs on the
lower parts of the stems of those plants which at an earlier stage bear
deciduous organs. This absence of marking has been urged in proof
of these stumps not being Sigil/arie at all; but we know no other
genus to which they can be referred; and in the noble erect specimens
discovered and examined by Mr. Binney, the gradual evanescence of the
scars or even flutings of the trunk at a few feet above the root, is very
decidedly shown in most instances, though in some they may be traced
almost to the base.

That the Sigillarie were of a very brittle and probably lax tissue, is
I think evident by the constantly truncated upper end of the stumps
in the shales, their never being prostrated entire, and the singular com-

* Mr. Binney, of Manchester, showed me a Sigillaria, in which the characters of

S. catenulata, renifornus, organum, and alternans were all displayed, a fact he has alluded
to in the Manchester Philosophical Magazine for October, 1844.

416 VEGETATION OF THE CARBONIFEROUS PERIOD

pression of the prostrate portions. I am not aware that they have ever
been found prolonged upwards from the roof of one coal-seam through
another one, as the coniferous * fossil discovered by Mr. Binney was.
This may be owing either to their generally lower stature, or more pro-
bably to their not being capable of surviving the geological changes
which that harder-wooded plant did.

External Form of Sigillarie.—It is highly singular that this fossil
should be of such universal occurrence and yet be unaccompanied by
any evident traces of branches, leaves, flowers, or fruit, and that till very
recently the probability of its roots beg Stigmarie was called in ques-
tion by most. Except, however, we assume that some Lepidodendrons,
which almost universally accompany Sigillarie, were the branches of
that genus, or that the fern fronds deposited round their bases were
their foliage, we cannot hazard a conjecture of the appearance of the
growing plant.

In favour of certain so-called species of Lepidodendron having been
the young state, or the branches, of the older Sigillarie, it may be
urged, that though abundant along with the latter genus, they are
very seldom found erect, or as it were growing, that there is no real
line of distinction between the two genera, and that some plants may be
indiscriminately referred to either. Opposed to this theory, however, is
the fact, that the Sigil/arie seldom or never show any tendency to ramify,
or present scars of fallen branches.

Above the sudden enlargement of the base, this genus presents a
columnar trunk of nearly equal diameter, very different from the gra-
dually tapering Lepidodendron, and more like the caudex of a tree-
fern. ‘The great size and peculiar form of the scars also rather resemble
those left by the fall of a frond than by a leaf such as the Lepidodendron
bore. Another fact, favouring this view of their having simulated mo-
dern tree-ferns in mode of growth, is the absence of any hitherto recog-
nized specimens of very small Sigillarie which, from being of consider-
ably less diameter than the old, would answer to the young+ of that
genus, and also the absence of slender Stigmarie roots. In short, it ap-

* This remarkable case of an erect fossil many feet long having deposited around it
as many feet of sandstone, followed by underclay, a bed of coal, shale, and other successive
deposits, is a startling proof of the rapidity with which the coal-beds were formed, of
the rapid decomposition of those which constituted the coal in comparison with the coni-
ferous wood, and of the probable composition of that deposit of very soft tissued plants.

+ This absence of young specimens may be otherwise accounted for by supposing on
the one hand, that the shale deposit was drifted, and that the old specimens alone with-
stood the effects of the transport; and on the other, to me more probable, hypothesis,
that they grew on the spot, and the older specimens alone fell and were fossilized.
I was once inclined to consider some of the Lepidophylla as the foliage of Sigillarie,
but the specimens I have seen of certain fossils with those attached, and the figure given
by Brongniart, “‘ Hist. Veg. Foss.,” vol. ii, t. 23, fig. 6, are against this supposition.

AS COMPARED WITH THAT OF THE PRESENT DAY. 4L7

pears far from impossible, that the Sigillarie was strictly Acrogenous
(increasing not laterally but in length) in its growth, that it at first
assumed nearly its full diameter, as a tree-fern or palm does, throw-
ing out at the same time its stout Stigmarie roots, which we know
to be of a very uniform diameter in all stages, and throughout their
whole length (pomts which will be dwelt upon in a separate essay).
It may further be conjectured, that the Sigillaria, as it grew upwards
with some rapidity, threw off the lowest fronds successively, leaving
the broad scars on the stems.

I do not attempt to enter into any further or more rigid comparison
between tree-ferns and Sigillarie, confining the above only to what may
subsist between the mode of growth of these two very distinct tribes of
plants, which does not necessarily indicate a close botanical affinity
between them.

The succulent nature of the Stgillariz, which I have elsewhere dwelt
upon, can hardly be considered an objection to these trunks having
been the supports to the fern fronds* so abundantly scattered about their
roots; for it may be remarked, that the stipites (stems of the fronds
themselves) appear to have been of a succulent texture likewise, if we
may judge of their invariable compression in the shales, and the rarity
of their occurrence in the nodules of iron-stone, &c., in a state that
shows structure. Succulent caudices (subarborescent) and stipites, too,
are further characteristic of some of our largest and most coriaceous-
fronded recent ferns, as the Marattiacee and Daneacee. In these tribes,
the succulence of their organs is remarkably contrasted with the woodi-
ness of those of most other ferns, the fronds of which are already assumed
(on characters drawn from the fructification only) to occur in the coal
formation.

A yet more remarkable and anomalous structure in Sigillarie than
either that of their stigmaroid roots or fluted stems was pointed out to
me by Mr. Binney, of Manchester, to whom I am mainly indebted for
all I know of the most important features of this genus, and whose in-
vestigations of their habit, mode of growth, and of their connexion with
the Stigmarie@ are beyond praise.t ‘This is the curious crucial mark

* This is, I believe, the opinion of Mr. Binney, who mentioned it to me with some
hesitation arising from his never having found the fronds attached, and from his having
had no opportunity of comparing the Sigillarie with existing tree-ferns.

7 In the Manchester Museum there is a room almost entirely devoted to illustrating
the botany of the coal formation. The original specimens of some Sigillarig and models
of others of the natural size, collected and transported with great labour, and arranged
under Mr. Binney’s direction, present to the eye the grandest features of the coal flora.
Accustomed as I had been to see these fossils in situ, both in the pits and in quarries, I
had previously no adequate conception of their gigantic size, nor of the rapidity with

which coal may have been formed, if the tissues of these yegetables were as lax as I sup-
pose them to have been.

418 VEGETATION OF THE CARBONIFEROUS PERIOD

which quarters the base ofthe trunk. The Svgillaria generally divides
into four main roots at the base, which unite to form that crown of the
dome described by Lindley and Hutton; and it is along the line of
union of these four roots that these strongly marked lines run, all
meeting at the centre of the dome. I know of nothing analogous to
this either in recent or fossil botany.

Of the foliage of Sigillaria little or nothing is known; the S. lepido-
dendrifolia* not appearing to be a typical, if at all a species, of the
genus. The scars are, especially in the larger species, much too broad
to be regarded as the point of attachment of leaves like those of Lepi-
dodendron.

Internal Structure of Sigillarie.—The beautiful plate and excellent
description of the anatomy of Stgillaria elegans given by M. Brongniart+
are, though as regards the specimen most satisfactory, far from com-
pleting our knowledge of the genus. The drawing of the entire plant
is evidently not that of a fluted Sigillaria of the common form. It may
belong to a young plant of this genus upon which the organs indicated
by the scars were very closely packed; but this is a supposition only ;
and until the figured fossil is better known, orsome of the fluted species
are found to contain structure, we must hold that the tissues of Sigz-
laria proper are to be described. M. Brongniart’s plant is no doubt
allied to this genus, and very distinct from Lepidodendron: there is a
probability, too, of its belonging to the same plant as Stigmaria, which
is evidenced by the similarity of the arrangement of the vascular
tissues in both. This tissue in St¢gmarie is broken into wedges, indi-
cating a higher organization than is found in Lepidodendron, and a
still more complex arrangement in the trunks of which Stigmarie@ are
the root. Such we find in M. Brongniart’s S. elegans, wherein are
two distinct series of bundles of vascular tissue; one disposed in
wedges separated by medullary? rays,} the other in as many bundles,
each placed at the back or small end of one of these wedges.

The great bulk of Sigillarie seems to have been inimical to the pre-
servation of their tissues, the process of decay being generally effected
on a grand scale in the substance of a plant evidently almost entirely
composed of a lax tissue. The remains of a central column are,
however, sufficiently obvious in the upright stems of many Sigillarie,

* “Brongn. Hist. Veg. Foss.,”’ p. 426, t. 161.

+ ‘“ Archives du Musée d’Hist. Nat.’’ vol. i. p. 404.

+ The tissue of the axis of this plant and of its rays is wanting. This is generally
the case in the spaces corresponding to what are filled with cellular tissue in analogously
formed vegetables. In all carboniferous fossils that I have examined the cellular tissue
is best preserved at the coaly circumference, I have, however, observed it elsewhere; and
in one Stigmaria it occupies nearly all the broad space intervening between the circum-
ference and vascular column.

AS COMPARED WITH THAT OF THE PRESENT DAY. 419

these have been called “ Endogenites,”’ are scarcely two inches in
diameter, and are generally obliquely placed in the substance of
specimens five feet and upwards in girth. That this slender column
represented all the vascular tissue of this plant, I cannot doubt, from
examination of Stigmari@, whose vascular column often assumes the
same appearance.

Affinities—These have been very much discussed by various natu-
ralists, who have eagerly seized upon the few botanical characters the
very imperfect fossils present.

In noticing its outward appearance, little allusion was made to the
S. lepidodendrifolia, the only species of the genus whose foliage has
been described, and which, if really a congener of S. Organum, &c.,
would, doubtless, ally all these plants closely to Lepidodendron itself.
But Iam inclined to pronounce M. Brongniart’s figure of this plant* to
be a true Lepidodendron, though possessed of scars more resembling
those of many Stgillarie, for it wants the fluted stem. The leaves of
this S. lepidodendrifolia seem to be what was considered a species of
grass or sedge, and placed in the genus Cyperites.

Did the great Sigillarie bear leaves like those of Lepidodendron,
they could hardly escape petrifaction in the shales, especially where
the Cyperites so abound ; but there is nothing whatever in the shales
which we can recognise as having been in any probability an organ of
or appendage to Srgillarie.

One point which must not be overlooked, Fig. 11.
namely, the possibility of the scars upon young
Sigillarie being quincuncially or spirally
arranged, as in Lepidodendron, and after-
wards owing to the dilatation of the trunk
being disposed in lines. The reverse of this
is sufficiently obvious in some recent Conifere,
as the Pinus Webbiana, where the young leaf-
sears are disposed in parallel lines, which be-
come disturbed through age, and appear ulti-
mately arranged in the old branches, as shown
in the accompanying wood-cut (Fig. 11).

That there are Lepidodendrons generically
distinct from Sigillarie cannot be doubted ;
for specimens of the former genus, many feet
long, are known, having their scars spirally
arranged throughout; but this is not incon-
sistent with the hypothesis that other plants
in a certain state, apparently belonging tu the

* Sigillaria lepidodendrifolia, “ Brong. Hist. Veg. Foss.,’’ vol. ii. t. 161.
II, 26

420 VEGETATION OF THE CARBONIFEROUS PERIOD

one genus, ought really to be referred to the other, and that the
Sigillaria elegans, whose structure M. Brongniart has ably illustrated,
may be amongst them.

Although the S. elegans here referred to displays no fluting, it is
possible that in an older state it might, and that the arrangements of
the vascular tissue into separate wedge-shaped bundles, may indicate
such a fluting, each ridge answering to one of the wedges ; a conjecture
the more probable, from the mass of vessels which supply the leaves,
and which lead to scars on the ridges being placed exactly opposite to
the wedge-shaped bundles,

The opinions previously held of the affinities of Sigillarie are very
various, and no less vague, being grounded upon supposed resemblances
in their external characters, to certain natural orders of living plants.
Thus Artis*, Lindley, and Hutton,t and more recently Corda,} have
referred them, with more or less confidence, to Huphorbiacee.

Schlotheim § to Palme.

Von Martius to Cact?.

Brongniart, originally, || and

Sternberg to Ferns.

In the earlier years of the study of fossil plants, when their affinities
were scarcely known, it was no wonder that observers sought the affini-
ties of Sigillarie amongst Dicotyledones ; but now that each successive
observation tends to prove that every order of that class,** except
Cycadee and Conifere, were absent from the coal-fields, it is not a little
bold to persist, as M. Corda has done, and on very insufficient grounds,
in ranking them among ELuphorbiacee. The absence of true woody
fibre in any part of these plants, pointed out by Mr. Bunbury} j is con-
clusive against Sigillarie being allied to Euphorbiacee: I may add,
that the same objection affects M. Corda’s reference of Lepidodendron
to Sempervivee.

Their reference to Palme by Schlotheim, is yet more unsatisfac-
tory. -The central column is of itself a fatal objection, and it is further
exceedingly improbable that such numerous Palme would be petrified
without the occurrence of structural specimens.

The arguments in defence of their being Cacti are nearly what other
writers have thought conclusive as to their being Huphorbiacee.

* Artis, Antediluy. Phytol., t. 3, 10, 18.

+ Lindley and Hutton, Fossil Flora, I. pp. 94—110; t. 31—36; II. p. 13; III. p. 47,
t. 166.

{ Corda, Flora der Vorwelt, p. 34. § Schlotheim, Essai, p. 23, t. 12.

|| Brongniart, Mem. Mus. d’Hist. Nat., v. 8, t.1, f. 7.

{1 Sternberg, Vers. I. p. 4, t. 38. Vers. II. p. 209, t. 15.

** See M. Brongniart’s excellent remarks on this subject in the “ Annales des Sciences
Naturelles,’’ 3rd Series, vol. vy. p. 52.

++ Quarterly Journal of Geological Society, vol. ii. p. 120.

AS COMPARED WITH THAT OF THE PRESENT DAY, 421

Judging from the great preponderance of Flices in the coal measures,
the probabilities are somewhat in favour of Sigillarie belonging to that
class ; and this opinion has accordingly the most advocates. Lindley
and Hutton* indeed adopted it at first, but rejected it afterwards on
the examination of Caulopteris ; + a most decided tree-fern.

The latest observations, however, on this subject, are those of
M. Brongniart, who after an examination of the S. elegans} slightly modi-
fies his opinion, placing the genus between Lycopodiacee and Cycadee,
but nearest to the latter.

Assuming the S. elegans to be a true Sigillaria, it appears to afford
slender grounds for the adoption of the above view, as regards its
uniting such diverse and distinct orders as Cycadee and Lycopodiacee.
It is true that it departs signally from the ordinary structure of the
latter order, which however itself exhibits various modifications of the
arrangements of scalariform tissue; but it requires stronger evidence
than the more perfect structure and regular arrangement of the bundles
of vascular tissue to ally it to Cycadee, with which, in general appear-
ance, habit, fluting, markings, stigmaroid roots, absence of accompany-
ing foliage, and many other points, it has nothing in common.

The Cycadee, on the other hand, and especially the Zamia integrifolia,
which M. Brongniart instances as peculiarly favourable to his views,
possess broad vessels perforated with very large circular apertures §
(not seen in Sigillarie), and they want the double zone of bundles of
vascular tissue and diverging fascicles of Sigillaria.

That the Sigillarie were allied to Lycopodiacee is evident, their
tissues and scarring being very like those of Lepidodendron, in which,
however, there is but one series or zone of vascular tissue. In both, the
great mass of the woody axis is formed of large tubular vessels iden-
tical in structure, and in both, fascicles are given off from the central
mass to the scars on the circumference, which fascicles consist of slen-
derer tubes than the axis does. In Lepidodendron there are no
medullary rays, the vascular zone being continuous and surrounded by

* Fossil Flora, Introduction to, vol.i.

+ Fossil Flora, tab. 42, sub Caulopteris primeva, et tab. 54, sub Sigillaria pachyderma.

t~ Archives du Muséum, t. i. p. 426.

§ A very similar tissue is exceedingly common in the coal-beds: all the mineral char-
coals (commonly called mother-coals in 8. Wales) present it abundantly, and appear
wholly formed of such vessels held together by fragments of medullary rays. In spe-
cimens of charcoals from the Virginian coal-mines given me by Mr. Lyell, the same tissue
occurs; but it was only during the passage of this sheet through the press that I had the
good fortune to meet with a fragment of the fossil wood to which these charcoals belong.
In examining the collection of Mr. Darker, an acute obseryer and excellent lapidary, he
showed me relics of a small rolled nodule of fossil wood from the coal-seams of the Yard
‘Mine, Durham, in which I at once recognized this tissue, of which I had been long in search.
The nodule was little larger than a hasel-nut, and originally covered with quartz crystals
Mr, Darker had himself conjectured its probable alliance to Cycadee.

2.6.2

422 VEGETATION OF THE CARBONIFEROUS PERIOD

those slender vessels from which the fascicles diverge and run to the
scars. In Stgillaria elegans, again, there are medullary rays; and
M. Brongniart states that the fascicles which run to the scars in that
plant originate in the inner zone of slender vascular tissue. I am how-
ever more inclined to suppose that they belong to a system of vessels
exterior to the main body of the woody tissue, for if they passed out by
the medullary rays they would be alternate with the latter, instead of
being placed immediately opposite to them.

In Stigmarie the bundles of slender vessels are distinctly seen arising
from behind the woody zone and passing out by the medullary rays
to which they are opposite, not alternate, as in Sigillaria elegans.

It is not by solitary characters, and least of all by such as the
arrangement of the tissues in the axis affords, that genera of plants are
referred to their natural orders. Amongst recent plants we see many
instances of plants indisputably belonging to one natural family having
the peculiar woody tissues of another and far distant group in the
system ; but these are mere analogical resemblances and by no means
indications of affinity.

We may conclude then that the Sigillaria elegans was not far from
Lycopodiacee, and especially from Lepidodendron : — that it was of
much completer structure and higher organization than either is incon-
testable ; but the indications of a relationship with any individual group
higher in the series, or with Cycadee in particular, appear to me far too
feeble to justify our considering it as tending to unite these two natural
orders. It is a plant which (until we know its foliage or fructification)
must be considered as belonging to the great family of ferns (including
Lycopodiacee), displaying a relationship, though only of analogy, to
Cycadee in one point, and to Huphorbiacee and Cacti in others.

Lepidodendron. —'The anatomy of this genus has been well illus-
trated by Lindley and Hutton,* by Witham,y and more recently by
M. Brongniart, { whose analysis has left little to be desired regarding
the structure of its stem. Under the remarks upon Lepidostrobus which
follow this Essay will be found notices of a few remarkable points con-
nected with the occurrences of cylindrical specimens of the trunk of this
genus containing cones in their hollow axes; and of the sculpture of
its surface as it appears before that compression to which the spe-
cimens are almost apyaniably subjected previous to or during their
petrifaction.

Of the stems, pentane? leaves, and fructification, we have thus a
very satisfactory knowledge ; but the nature of their roots is not ascer-
tained. Mr. Dawes, of West Bromwich, to whom I am indebted for
much information regarding the structural characters of coal-fossils, is

* Lindley and. Hutton. t Witham. ¢ Brongniart.

AS COMPARED WITH THAT OF THE PRESENT DAY. 423

inclined to regard the species of Halonia as roots of Lepidodendron, on
which opinion I have no remarks to offer.

Nearly forty species of Lepidodendron are described; and the cha-
racters of the majority of them are drawn from the arrangement and
form of the scars on the stem. There are, besides, seven or eight
species of Lepidostrobus, some of which (perhaps all) are parts of one
or other of the Lepidodendrons. Halonia, another genus of the Lepz-
dodendree, is composed of three species, possibly the roots of these or
others of the same genus. Some Lepidophylla are certainly its leaves:
a few of the carboniferous Lycopodites are founded on its slender ter-
minal branches ; and one or more Trigonocarpi (as I shall show in some
remarks upon Lepidostrobus) are probably the seed-vessels (sporangia)
of the same.

This reduction of genera to species, and in some cases of genera
contained in one order (as Trigonocarpum, which is placed in Scitami-
nec by Unger) to the species of another, materially diminishes our ideas
of the great amount of species preserved in the coal. Such a reunion
is not, however, available in practice, nor is it here for the first time
forced on our attention; though it is not sufficiently considered, in the
hurry of describing new forms, that the result of further discoveries
amongst those already described, is to reduce the many supposed
species to fragments of a few certain ones. Hence, the object of the
fossil botanist should be now, in preference to describing what appear
new species, and thus increasing the synonomy, to throw more light
on the old, about all of which we know, botanically speaking, very
little.

But it is not merely by the reduction of genera, founded on the parts
of plants, that this is to be carried into effect. No one accustomed to —
study existing individuals, can regard the characters adopted for dis-
tinguishing the Lepidodendrons, without comparing their validity with
those drawn from the same parts of their living analogues. Applying
this rule to Lepidodendron, it will be seen that if the species of that
genus were as prone to vary in the foliage as are those of Lycopodium,
our available means for distinguishing them are wholly insufficient.
Take, for example, the noblest of the genus Lycopodium, the L. densum
of New Zealand, which in stature, and probably in general habit, &c.,
approaches nearer to Lepidodendron than any of its congeners. In this
species, not only do the shape and arrangement of the leaves vary in
different specimens, but leaves from different parts of the same indi-
vidual are very unlike. The three accompanying cuts represent as
many states or varieties of this plant, gathered by myself in New Zea-
land, where multitudes of specimens of all were growing intermixed.
Of these, the dense fastigiate one (Fig. 12) is the most common; and

424 VEGETATION OF THE CARBONIFEROUS PERIOD

this is represented with the cones at the ends of the branches, borne as
the Lepidodendrons did theirs, and all of a uniform size.* The same
variety, however, repeatedly occurred with some of its branches, as in
(Fig. 13), clothed with slender patent squarrose leaves ; whilst others had

Fig. 12.

the branches slender, as in (Fig. 14), and covered with small and closely — 7

appressed leaves. In figure (13) the whole habit is widely different ;
and the leaves from three parts of the plant are so dissimilar, that had
* Though this species bears cones whose size, shape, &c., is constant, the majority of

the species are extremely yariable in this respect, as were the Lepidostrobi : this is illus-
trated in my remarks on the latter genus.

ey . za ~ 7. ,

ee

eee

AS COMPARED WITH THAT OF THE PRESENT DAY. ~ 495

we only portions of branches, &c., such as we possess of Lepidodendron,
it were impossible to recognize the plant. At letter a is shown the
base of the stem, covered with closely imbricating leaves; at letter b, a
portion from the middle, with very long tapering appressed leaves ; at
letter c, the branch, with its elongated spreading and squarrose foliage.

Fig. 13.

The third modification of this plant is represented at (Fig. 14), in which
the branches are uniformly slender, and clothed with similar appressed
leaves throughout; but the arrangement of these leaves is exceedingly
variable ; for at letter a they are closely imbricated and cover all the

426 VEGETATION OF THE CARBONIFEROUS PERIOD

stem, whilst at 6 they are disposed in whorls, separated from one
another by more than the length of the leaf itself.

It may well be asked how the botanist can pronounce these dissimilar
forms to belong to one plant. This can only be done by comparing a
large series of perfect specimens, or by growing the different varieties
from the seeds of one. The plant has, however, some constant cha-
racters and a certain habit, the value of which is at once perceived and
admitted by the skilful botanist, though the uninitiated eye cannot
seize it.

¢

AS COMPARED WITH THAT OF THE PRESENT DAY. 497

Ulodendron.—This very remarkable genus scarcely differs from Lep?-
dodendron in internal structure: its external aspect widely departs from
that of any plant, recent or fossil, with which I am
acquainted. I have seen in collections, speci-
mens which have been fossilized, apparently erect,
or at any rate, under very different circumstances
from those preserved in the shales over the coal.
They present the appearance of a large un-
branched zigzag trunk, with two rows (opposite
one another) of alternating cup-shaped deep
depressions, one at every projecting angle of the
trunk. Some idea of the appearance of these
specimens might be gathered from the accom-
panying wood-cut. Mr. Dawes showed me a
specimen preserved in sandstone, with a large
organ which he considers to be a cone inserted
into one of the cup-shaped depressions. I could
not, however, form any conclusion concerning
the real nature of this highly interesting ex-
ample.

Knorria. — A genus of which very little is
known. Dr. Lindley* pronounces the K. taxina
to be certainly coniferous, comparing it with a
leafless branch of the common Yew; to me it
appears to have quite as much resemblance to
the stem of a Lycopodium, especially of L. phlegmaria.t

The K. Sellonii again, is remarkably similar to what I have con-
sidered as much compressed specimens of Stigmaria ficoides, which are
very much compressed but not entirely flattened, and in which the
cellular tissue has collapsed, leaving the vascular bundles which meet
the circumference at the areole projecting beyond the surface :—this,
however, is only a suggestion.

Fig. 15.

Calamitee.

T have in vain sought for any traces of structure in carefully prepared
species of this genus; or for evidence of their being Equisetacee in the
presence of those siliceous stomata with which that order abounds, and
which would surely have been preserved in the fossil state.

Very fine specimens of this genus were pointed out to me on a cut-
ting of the Manchester and Bolton Railway by Mr. Binney : they were

* Lindley and Hutton, Fossil Flora, sub. K. taxina, tab. 95.
+ Compare the figure of the stem of this plant given in Brongniart’s Hist. Veg. Foss.,
tab. 7, fig. 11, with Lindley and Hutton’s figure v. 2, t. 95.

428 VEGETATION OF THE CARBONIFEROUS PERIOD

standing erect as they grew in a forest of Sigillaria, distinctly proving
that what have been taken for leaves by the authors of the Fossil Flora,
are in reality roots.

M. Brongniart* ranks them, I am not aware upon what ground, with
dicotyledonous plants, allied to Lepidofloyos, whose tissue has been
illustrated by Corda.

Cycadee.

The only evidence I have seen that this order existed during the car-
boniferous period, rests on that afforded by the charcoal (or mother-
coal), and rolled nodule of fossil wood, both alluded to at p. 421; and
it is confirmed by the curious observations of Brongniart | upon Wog-
gerathia.

On the Order Conifere, and some obscure ones supposed to be allied
to Filices, I have no remarks to make: future collectors will doubtless
throw some light upon their true nature, especially when it is considered
how much these last few months have done for the Lepidodendrons,
Lepidostrobi, and Stigmaria. Our provincial collections, and even the
still rudimentary one of the Geological Survey, abound in specimens
suggestive of most interesting points, yet to be worked out. ‘These, it
is trusted, will form the materials for a succession of essays in the
Memoirs of the Geological Survey of the United Kingdom.

I cannot conclude these desultory, and I fear unsatisfactory remarks,
the fruits of one short year’s study in the vast field of inquiry to which
they relate, without expressing a hope, that my observations on the dis-
couraging aspect of the science will not deter the beginner from pur-
suing his investigations: still less that they will lead the geologist to
reject such information as the botanist can supply, because it has
hitherto been encumbered with loose speculations on the affinities of the
genera, distribution of the species, and value of the characters which the
latter display. ‘Too much has been expected from the botanist, who
wants materials for those bold generalizations which the fossils of the
animal kingdom so abundantly supply. Except to individuals who have
great facilities for this study, the collection and examination of the
waifs and strays of a by-gone Flora is a forbidding pursuit. It can be
undertaken to advantage only by him to whom the existing Flora is in
some measure familiar, and such an one cannot see the rapid advances
in paleontology which are due to the exertions of the zoologist, without
feeling a conviction, that some undistinguishing geologist will expect
more definite and immediate results from his labours than the speci-
mens at his command may ever afford.

* Brongniart, Annales des Sciences Naturelles, Ser. 3rd, vol. 5, p. 52.
+ Brongniart, Annales des Sciences Naturelles, 1. ¢.

AS COMPARED WITH THAT OF THE PRESENT DAY. 429

The very desire to give a definite answer to the many puzzling ques-
tions proposed to the observer, has mainly increased the prevailing
readiness to draw arbitrary conclusions from the characters which the
outward appearance of a fossil affords: against this the student cannot
be too frequently warned. ‘The earliest inquirers saw (and some see
still) the acorn, bean, hasel-nut, &c., in the strata of the coal formation,
parts of plants belonging to orders which are not acknowledged by the
sober geologist to have existed at that period. So the rude collier,
denying a vegetable origin to the ferns of the shale, beholds in them
merely the effort of a creative power which has fashioned stones in the
likeness of plants; whilst his more reflective master recognizes fossil
snakes in Lepidodendron and mail-clad crocodiles in Sigillarie. He who
acknowledges all to be vegetables has advanced a step further, but
there he too may stop, if he does not pursue his investigations with due
caution ; for the measure of success which will reward his study will be
proportionate to the comprehensiveness of the view he takes of the whole
vegetation of the period in question, whether such view be the result of
his own experience, or adopted from the conclusions of others. Let
him remember that as remote districts on the globe are peopled, not
only by different species of plants but by different prevailing natural
orders, so the outward forms of one locality are imitated in the other by
objects which have no further relationship, and nothing else in common.
In this respect, intervals in time are marked by the same changes as
intervals in space. Let him likewise bear in memory, that the two most
experienced and distinguished observers have attained the same conclu-
sion regarding the general features of the coal Flora, which I shall give
in their own words. Goeppert.j in his resumé of the characters and
affinities of Stigmaria, says :— :

«_________ q’ou _s’ensuit une nouvelle preuve pour l’opinion deja
emanée tant de fois, que la vegetation actuelle et la primitive ne for-
ment q’une seule Flore, dans laquelle les familles separées forment
actuellement un ensemble harmonieux, au moyen de formes interme-
diaires multipliées qui se trouvent tantot dans le monde actuel, tantot
dans le monde primitif.”

'M. Brongniart, the most successful cultivator of the science, after
a careful review of the whole vegetation of the coal, excludes all existing
orders of flowering plants except Conifere and Cycadee, and thus con-
cludes his essay on Noggerathia :— {

* The tail-piece represents a very singular fossil fruit of the coal formation, possibly
belonging to the highest order of plants of any known during that epoch. It was dis-
covered by Mr. Wilson, of Barnsley, in the Oak’s Quarry sandstone, and is apparently the
Trigonocarpum ovatum, Lindley and Hutton (Fossil Flora, t. 142 A.)

+ Goeppert, Genera des Plantes Fossiles, under Stigmaria, p. 29.
{ Annales des Sciences Naturelles, Ser. 3, vol. 5, p. 61.

430 VEGETATION OF THE CARBONIFEROUS PERIOD, ETC. ;

“« Ainsi, tout semble nous porter & conclure, des recherches faites
jusqu’a ce jour, que la végétation terrestre de l’epoque houillére était
limitée a deux des grandes divisions du regne végétal; les Cryptogames :
acrogenes ou vasculaires, et les Phanérogames dicotylédones gymnos-
permes.” |

Fig. 16.

fe 4a1?

On some Peculiarities inthe Structure of Stiamaria. By Dr. Hooker,
F.R.S., &c., Botanist to the Geological Survey of the United
Kingdom.

Mucu as has been said and written upon the structure and appear-
ance of Stigmaria, and long as this genus has been recognized as the
prominent feature in all discussions regarding the origin of coal, there
are still some very important points in its history which have hitherto
escaped observation. Of these, the most remarkable is the nature of
the surface previous to compression, and of the bases and attachment of
the rootlets.

The surface of the Stigmaria has hitherto been supposed to be nearly
even and uniform, interrupted only by shallow scars, variously described
as circular by some, oblong by others, and lanceolate by still a third class
of observers, according to the state of the specimens they have ex-
amined. ‘The depth of these scars seldom exceeds 7'sth to 3th of an
inch, in which respect the appearance of the surface of this fossil is so
constantly the same, that no geologist has at first sight recognized the
specimens figured at Plate 1, as belonging even to the genus Stigmaria.

The above-mentioned two specimens are fragments of plants which
have evidently been preserved under little or no compression: the per-
fectly circular outline and uniform depth and figure of the cavities,
together with the cylindrical form of the organ they contain, all attest
this positively ; whilst the depression of the surface between these cavi-
ties, and the wrinkled substance, seem to indicate that a slight collapse
of the tissues composing the plant has taken place.

After having convinced myself that the specimens of Lepidodendron,
ficured at Plate 8, fig. 12, and that the portion of bark enclosed in the
fossil,* Plate 10, fig. 7, and indicated by the daggers, were the only
specimens that had come under my observation displaying any approach
to the original and unmutilated state of that genus, I was more than
ever assured of the extreme laxity of the tissues of the coal-plants, and
persuaded in my own mind that the ordinary run of specimens of all were
calculated to lead to erroneous views of their original appearance. Being
thus prepared to find familiar forms under a new aspect, I was not a
little gratified, when inspecting the cabinet of Miss Jukes, of Birming-
ham, to recognize the specimen of Stigmaria ficoides, figured at Plate
1, fig. 3, and which was obligingly lent for the use of the Survey. On
my return to town I found that a similar, but more perfect and much

* See Memoirs of Geolog. Survey, in Essay on Lepidostrobus.

432 STRUCTURE OF STIGMARIA.

larger, specimen had been sent to the Museum, by Mr. Ormerod, and
ficured for me by Mr. Baily. In this latter the cavities are of somewhat
less diameter, and contain the obconical or flaggon-shaped bases of the
rootlets, whose summits are level with the mouth of the cavity, and are
depressed at the apex.

In neither of these cases could I gain any information upon the par-
ticular circumstances under which these valuable specimens were pre-

served, to which however they must owe their singularly perfect —

condition. ‘The precise locality of the specimen, Fig. 3, is unknown:
Mr. Ormerod’s was found overlying the white coal in the Peel Quarries,
Lancashire.

Comparing these with the usual state of the plant, figured at Plate 2,
fig. 4, the amount of collapse of the tissue which shall bring the base of
a cavity a full half-inch deep, to the surface, must appear startling ; but
it is not really so remarkable as that the whole cylinder, including the
vascular axis, should be reduced by pressure from the thickness of three
or four inches to that of an inch, or even a few lines, or to a mere flake,
as is generally the case with this plant when preserved (if this deserves
the name of preservation) in the shales above the coals.*

While upon this branch of the subject I may allude to the singular
difference in the appearance of the surface in different parts of the same
Stigmaria. In the Bolton railway trees, to which Mr. Binney conducted
me, the Sigmaria roots extended uninterruptedly for upwards of 20 feet,
and at different points along that distance from the trunk of one indi-
vidual we had many opportunities of marking the appearance of the
surface: this, which was characteristic of S. ficoides throughout the
greater part of their length, when within a few feet of the tree was seen
to be materially altered. The roots themselves here became broader,
more vertically compressed, the scars densely packed, and forming
transversely elongated areole, the fossil thus possessing all the charac-
ters of S. conferta of Corda.t

Though so much swollen below, the surface of attachment between
the base of the rootlet and of the cavity into which it is inserted must
have been but limited and the union slight; for in several instances
these organs are loose in the cavity, and resemble so closely a nut in its
shell, that this fossil has been taken for a fruit with many nuts borne
on a pitted receptacle. In the admirable and very complete account
of the genus, published by M. Goeppert,{ there is a figure (Table
XIII.), representing the outline of the rootlet as produced within the

* The Stigmaria is rare in the shales above the coal; but I have observed it to occupy

that position in the Oakwood level of Cwm Afon (inS, Wales). I have elsewhere (p. 396)

given my views on the subject of the prodominance of this yegetable in the underclay.
} Corda, Flora der Vorwelt, Tab. 13.
{ Goeppert.

———=- >

STRUCTURE OF STIGMARIA. 433

.

circumference of the stem, and circumscribed by a dark line: this line
probably indicates the surface of the deep cavity, to the base of which
the rootlet is attached.

The apices of those bases of the rootlets are marked with a circular
depression, and so is the lower extremity of the rootlet figured at Plate
2, fig. 1, suggesting the possibility of the point of the latter (rootlet)
having been jointed unto the basal portion, at the level of the aperture
through which it emerges from the root. These corresponding depres-
sions may, however, have been caused by the collapse of the cellular
tissue at the fractured end, and thus indicate no articulation at all. In
none of M. Goeppert’s figures is any such articulation represented, but

there is something of the sort in Corda’s Tab. XIL., Fig. 4.

For the highly interesting specimen of a rootlet in which the cellular
tissue is preserved, the Museum is indebted to Mr. Warington Smyth
(Mining Geologist to the Geological Survey). It is figured at Plate 2,
Fig. 1, and consists of the silicified lower portion of a rootlet, whose
sides have collapsed, so as to reduce its originally cylindrical form to
that of an irregularly four-sided prism. ‘The substance is formed of a
network of exceedingly delicate cellular tissue, (Plate 2, Figs. 2 and 3,)
composed of hexagonal cells ; and it is traversed throughout its length
by a dark line, (Fig. 2 a,) no doubt indicating the prolongation of one
of those slender bundles of vascular tissue which issue from the medullary
rays of the axis, (a of Figs. 6 and 7,) and thence proceed to the mamille
on the surface of the specimen, figured at Fig. 4, or the bases of the
cavities in the figures in Plate 1.

This very simple structure of rootlet is similar to that of several
Lycopodia, and indeed of many other plants, both Monocotyledonus and
Dicotyledonus, so far as can be ascertained without a further knowledge
of the axis of the organ; of whose structure, however, there can exist
no reasonable doubt, since it may be confidently assumed to consist of
a bundle of vessels, similar to those represented at Fig. 7 a.

In M. Goeppert’s* figure (Table XIII. Fig. 8, E d) this axis is
represented as formed of such tissue, surrounded by cellular tissues dis-
posed in two concentric rings, between which a black line is interposed.
There is no appearance of any similar arrangement in the specimen I
have examined, the net-work being continuous from the vascular axis to
the circumference. As there appears to be a rupture of the tissue to a
considerable extent in M. Goeppert’s figure near this line, and as where
there is no such disunion the cellular tissue is continuous across the line,
I am inclined to regard the latter as a coaly deposit, consequent possibly
upon the process of petrifaction having been arrested at that point.

Mr. Henry Beckett of Wolverhampton having forwarded to the Geo-

* Goeppert, l. c. t. xvi. fig. 48.

434 STRUCTURE OF STIGMARIA.

logical Survey most beautiful specimens of the vascular axis of Stig-
maria, they were sliced for examination by Mr. Cuttle, and some of the
points they perfectly illustrate are drawn by Mr. Bone.

Plate 2, fig. 4, represents the surface of the specimen in which the
axis Fig. 14is contained. There was not the slightest appreciable dis-
tinguishing characters between the markings of this and that specimen
containing the very different looking axis Fig. 5. I shall therefore de-
scribe them as specifically the same, for in microscopical characters also
they are identical.

The axis consists of a broad cylinder, composed of parallel elon-
gated tetragonal or hexagonal tubes, of equal diameter throughout, for
the greater part of their length obtuse and rounded at either extremity,
and everywhere marked with crowded parallel lines, which are free or
anastomising all over the surface. The tubes towards the axis are of
the smallest diameter: they gradually enlarge towards the circum-
ference, where the largest are situated, though bundles of smaller tubes
occasionally occur among the larger (ig. 7 d). The cylindical axis
(which may for convenience be called the woody system of the plant) is

divided into elongated wedge-shaped masses, rounded at their posterior ©

or inner extremity by numerous medullary rays of various breadths, some
much narrower than the diameter of the tubes, others considerably
broader, but none are conspicuous to the naked eye, except towards the
outer circumference.

The medullary rays, even the narrowest, are traversed by bundles of
tubes half the diameter of the largest vessels of the axis (or wood) or
even less (Plate 2, figs. 6, 7 a, and 8.) The transverse lines on their
surface are generally finer and less crowded. These bundles evidently
originate in the cellular axis of the stem, and do not belong to the
wedges of vascular tissue (or wood) between which they run, as they ap-
pear to have done in M. Brongniart’s* specimen of the plant, both from
his figure and description. I cannot, however, but conclude the latter
to be erroneous, because M. Goeppert, whose specimens appear to have
been in this respect more perfect than any hitherto illustrated, repre-
sents the bundle of vessels which proceed from the axis, run between
the wedges of wood, and communicate with the rootlets (leaves,
Goepp.) as originating in isolated bundles, irregularly scattered in the
medullary axis of the stem. Of the existence of these bundles there
are some indications in my own specimens, though for the most part
they have been destroyed with the cellular tissue of the plant, which
indeed often takes place with the system of vessels from which the leaves,
rootlets, or scales of the cones in these fossils are supplied. It is so in
the stems of Lepidodendron, in the axis of the Lepidostrobi, in the

* Archives du Museum d’Hist. Nat. v, i, p. 406, t. 29, f.3 d.

STRUCTURE OF STIGMARIA. 435

portion of the Sigillarie figured by M. Brongniart, and in other fossils
contained in the Museum of the Survey, and is probably owing to their
great delicacy, for they are much more membranous in appearance than
the similarly marked vessels of the wood.

The most important circumstance thus developed is the existence
of a double system of vessels in Stigmaria, first shown by Goeppert,
and the consequent approach in this respect to Diploxylon,* Corda.
In Diploxylon, however, the inner system forms a continuous cylinder,
concentric with and in juxta-position to the wedges of wood forming the
outer ; whilst in St:gmaria the same inner system is broken up into
scattered bundles apparently unsymmetrically arranged in the medullary
axis or pith of the plant.

In Lepidodendron again there is the same double vascular system ;
but that, from which the bundles arise which proceed to the leaves, is
placed externally to the wood, where it forms a continuous zone with a
well-defined inner edge (in juxta-position with the outer circumference
of the inner zone) and a sinuous outer edge, from which the diverging
bundles are given off. Such is also the arrangement in the axis of
Lepidostrobus.

Sigillaria elegans differs from all the above. The woody system is
(as in Stigmaria) broken up into wedge-shaped plates, separated by
medullary rays; and another vascular system (from which, possibly, the
leaves are supplied) forms f a series of bundles in the medullary axis of
the stem, each placed opposite the wedges of wood. The bundles
which immediately supply the leaves again are also placed opposite the
' wedges of wood, but externally to them, in this respect differing from
Stigmaria, where they are opposite the medullary rays, and from Lepi-
dodendron, where they form acontinuous zone. It is, however, doubtful
to M. Brongniart whether in his Scgillarie the system of vessels which
supplies the leaves really communicates with that immediately sur-
rounding the pith. To me that structure appears improbable, from the
position of both being alternate with the medullary rays through which
the vessels should pass. This then is a very important point to clear
up; for if the inner and outer bundles of tissue in Sigillarie communi-
cate through the medullary rays, then will the arrangement be nearly
identical with that of Stigmaria. If, according to Brongniart’s views,
on the other hand, the outer bundles from which the leaves are supplied
haye no communication with the innermost, and form an independent

* Corda, Flora der Vorwelt, p. 34, tab. xi.

+ On this point M. Brongniart is not satisfied, not haying seen the bundles which
traverse the cutical portion in the medullary rays, which might be expected if they com-
municated with the system of vessels within the wood and immediately surrounding the
pith,

II. 24H

436 - STRUCTURE OF STIGMARIA.

series concentric with but exterior to the wood, then will the arrange-
ment coincide with that of Lepidodendron in these important points, and
the Sigillaria in question may be pronounced a Lepidodendron with the
vascular system broken up into wedges divided by medullary rays, and
further possessed of a third system of vessels placed in juxta-position to
and within that of the wood.

The accompanying wood-cuts explain these arrangements of the
tissues in Stigmaria (A.), Lepidodendron (B.), and Sigillaria (c.), the
latter both according to M. Brongniart’s view, and in the light wherein
I regard it. In all, the letter a indicates the vascular tissue of the
wood; b, that from which the leaves are supplied; and ¢ indicates the
third series in Sigillaria.

NS
Ie
Bs Wi

Ws

Under any circumstances, the most complex of these plants is the
Sigillaria, mm which the vascular tissue surrounding the axis is as
independent of the foliage and as free, as are the similar vessels in
Lycopodium, which also pass down the pith within the wood, and send
off no bundles to the leaves.

STRUCTURE OF STIGMARIA. 437

Should the conclusion of M. Brongniart hold good, and the vascular
system of the axis ¢ give origin to the bundles 4, then will there be
some botanical evidence of Stigmaria constituting a part of Sigillaria.
If, on the other hand, it does not prove correct, still its failure would
afford neither proof nor even presumptive evidence against these genera
holding the respective place of root and stem.

As, however, I have elsewhere stated, I consider Mr. Binney’s dis-
coveries to be conclusive upon the origin of Stigmaria. That gentleman
had the kindness himself to conduct me to the spot where the great fossil
trees were, and we cleared the soil from the roots of several well-marked
Sigillarie, which roots were indisputably the plant called Stigmaria.

M. Goeppert, who holds this genus to be an individual plant, generic-
ally distinct from any other, refers it to the Lycopodiacee as its nearest
allies, but suggests the probability of its being intermediate between
that order and the Cycadee. This exact coincidence with the opinion
of M. Brongniart respecting the affinities of Sigillarie is sufficiently
striking, especially since these authors entertain different views of the
nature of Stigmaria. I have, under my general remarks on Sigillaria,
expressed doubts of the correctness of this opinion, in reference to the
connexion of two such widely different families by so slender and falla-
cious a character as the arrangement of the tissues in the stem. The
points by which Sigillaria (and Stigmaria) is allied to Lycopodiacee,
especially through the Lepidodendra, are probably quite sufficient ; but
when we consider that the Cycadee have sexes, a complicated repro-
ductive system, and a perfect seed with embryo, radicles, albumen, &c.,
whilst Lycopodiacee are asexual, have no reproductive system of male
and female flowers, and increase by spores, it will be evident that it is
by modifications of these important organs that the intermediate family
would be indicated. The noble Tree-fern departs from its humble
congeners in assuming the aspect of a Palm, and the Cephalotus (or
Pitcher-plant of New Holland) has the leaf of Nepenthes; but the
Tree-fern does not unite the Ferns with the Palms, and the Cephalotus
is not the less a rosaceous plant because it has the pitcher of Mepenthes ;
and so we may find that the Stigmaria is no less a fern from its tissues
being modified in arrangement. A still better illustration is furnished
by some of the natural order Magnoliacee, that have the woody tissue
closely resembling Conifere, and the parasitical Myzodendrons of Cape
Horn, whose wood is like that of the ferns in many respects. These
are genera certainly departing from their immediate allies in these cha-
racters, but not therefore allied to the plants which in those particulars
they resemble.

The symmetrical quincuncial arrangement of the scars has been
considered by very many observers to be all but segs aan the

H

438 STRUCTURE OF STIGMARIA.

theory of Stigmaria being a root. We know, however, very little of
the roots of living plants, their importance in systematic botany being
trifling. There is, however, no good cause for supposing that symmetry
should not pervade the subterraneous as well as the superior organs of
plants; and in the case of the roots of some Palme in common use,
but of unknown origin, we have an arrangement of rootlets round a
prostrate root (analogous to the root of Stigmaria), as symmetrical as
in Stigmaria itself.

M. Goeppert has recognised, in the majority of those species of Stig-
maria which are founded on variations in the form and disposition of
the sexes, only one. That such a view is concurrent with the fact of
the genus representing merely a root, is obvious; for such an organ
cannot be expected to afford specific characters. M. Corda, on the
contrary, has given the name of Stigmaria ficoides to the dissection
of a plant which evidently bears no very close relationship to the cha-
racteristic fossil of our British under-clay, and which is, doubtless, the
true S. ficoides of authors. M. Corda’s S. jficoides is, on the other
hand, possibly a stem, and not a root.

There are in the Museum of Practical Geology so many Stzgmaria
specimens, the majority having only the external characters preserved
(though in not a few the internal tissues, and in some both, are admira-
bly displayed), that I can confidently affirm, that specimens figured by
Corda (Tab. XII.), do belong to the same genus, and probably species,
as M. Brongniart’s dissections in the Archives. In proof of this asser-
tion I would adduce the figures and dissections of Plate 2, Bot. figs.
5—13, all having been made from the one specimen, fig. 4.

Explanation of Plates.

Plate 1, figs. 1 and 2. Two views of a specimen of Stigmaria,
fossilized without compression, and showing the pits or cavities from
which the rootlets emerge in an unmutilated condition. The bases
of these rootlets are generally retained within the cavity. Specimen
from Peel Quarry, Lancashire, above the white-coal seam,—Mr.
Ormerod’s cabinet. Fig. 3, specimen of a similarly preserved fossil,
from the cabinet of Miss Jukes, of Birmingham.

Plate 2, fig. 1, rootlet of Stigmaria ficoides. Vig. 2, horizontal
section of ditto, very highly magnified, showing at a the position of
the vascular axis. Fig. 3, cellular tissue from ditto, more highly
magnified. Fig. 4, portion of Stigmaria ficoides, showing the scars as”
they usually appear, and enclosing an axis from which the dissections
5—13 were taken. Fig. 5, horizontal section of axis in fig. 4, of
natural size. Fig. 6, posterior ends (those towards axis) of the vas-
cular tissue (6) forming the wood, divided by broad and narrow medul-

STRUCTURE OF STIGMARIA. 439

lary rays (a). Fig. 7, anterior (or circumferential) end of vascular
axis; a, vessels passing through the medullary rays to the rootlets ;
6, vascular tissue of wood; d, bundles of small vessels towards circum-
ference. Fig. 8, very highly magnified view of one of the narrow
medullary rays (fig. 6 ¢), forming a bundle of tubes of less diameter
than those forming the wood. Fig. 9, vertical section through the axis
of fig. 5. Fig. 10, magnified portion of fig. 9, showing at a the longi-
tudinal vessels of the wood ; at 6, the horizontal ones passing through
the medullary rays. Fig. 11, another magnified portion from the cir-
cumference of the same; a, the longitudinal vessels displaced by the
horizontal ones, 0, which are fractured in the fossil, and pass upwards
towards the rootlets at 6’. Fig. 12, section of the axis (fig. 5) taken at
right angles to the medullary rays. Fig. 13, very highly magnified
portion of ditto, showing the longitudinal vessels of the wood, a, dis-
placed by the horizontal ones, 6, whose mouths are cut across, issuing
through the medullary rays. Fig. 14, horizontal section of another
axis, from a specimen marked externally asin fig. 4, and whose structure
accords with that of figs. 5—13.

[ 440 ]

Remarks on the Structure and Affinities of some Lepipostrosi. By
Dr. Hooxrer, F.R.S., &c., Botanist to the Geological Survey of the
United Kingdom.

The numerous detached petrified cones which are scattered through
the various strata of the coal formation, being obviously organs of fructi-
fication, and having therefore belonged to some of the arborescent
plants whose remains they accompany, are objects of peculiar inte-
rest to the fossil-botanist and geologist. Such of them as are preserved
in the nodules of iron-stone, or are otherwise mineralized without pres-
sure, alone offer the means of ascertaining to what existing families of
plants they are most nearly allied ; for in those that are crushed flat
in the shales the internal structure is wholly destroyed. Many of the
better preserved specimens have been sliced, polished and examined
with the greatest care ; but this expensive operation has hitherto thrown
little light upon the true nature of the objects thus investigated.

This is because the three following conditions for their complete
illustration has never been displayed by one specimen, and the most
important point—the nature of the organs of fructification—has
hitherto wholly escaped observation in all. Every cone being an
ageregation of organs of some kind, it becomes necessary to ascertain
not only the arrangement of these organs, but the nature of the tissues
composing them, and (even more essential) their contents, before satis-
factory conclusions can be drawn as to their relationship to any of the
vegetable remains they accompany, or to whatever existing order of
plants they are allied :—

1. The arrangement of the individual organs of fructification, of
which the cone is an aggregation, and the nature of the scales support-
ing them. These are characters sometimes displayed on the fracture
of the specimen by ordinary means, though rarely, from the parts ap-
pearing to have suffered partial decay previous to or during petrifac-
tion. The imbricating apices of the scales, which lie over one another
like those of a pine-cone, are generally removed with the matrix
wherein the fossil is imbedded, as is seen by contrasting plate 8, fig. 1,
with fig. 38, in the former of which the apices of the scales have per-
sisted, whilst in the latter they are uniformly broken away. The muti-
lated is the more usual state of the surface, for reasons which shall be
afterwards detailed.

2. The tissues, or anatomical structure of the various organs com-
posing the cone ; viz., of the central axis, which is a continuation of the

STRUCTURE AND AFFINITIES OF LEPIDOSTROBI. 44]

stem of the plant ;—of the scales, which, being inserted into the axis,
support the individual male or female organs ;—and of the latter them-
selves. These tissues can only be displayed by slicing fossils in the
very best state of preservation, and in such as are changed into a more
or less transparent mineral. Specimens of this description are ex-
ceedingly rare, and have not hitherto been described.

3. The two above considerations are secondary to the remaining one,
the nature of the contents of the cones. ‘There may be stamens or
male organs,—ovaria, or female ones ;—-or lastly, capsules containing
reproductive spores (which are peculiar to plants having no sexual
system) ; for these three kinds of organs all occur arranged in the
form of cones, undistinguishable from one another by any external
marks. Up to the present time no carboniferous fossil cone has ever
been known to supply this great desideratum, without which we can
arrive at no exact conclusion as to whether these curious objects are
clusters of flowers, or fruits, or are the spore-bearing organs of flower-
less vegetables as mentioned above.

Sections of numerous fossil cones have been prepared for the Geolo-
gical Survey, with the view of illustrating these several points. Many
have hardly repaid the trouble and expense of slicing, whilst others,
which from external characters had been considered the least pro-
fitable, have, on the contrary, turned out the most instructive.- It is a
fact, that neither the most experienced lapidaries, or collectors, nor
the geologists, mineralogists or botanists, have been able to decide in
the majority of cases by the external appearance of a fossil cone
how its internal and microscopical tissues shall be preserved. As an
instance, it may be mentioned, that the most beautiful of all the Lepz-
dostrobi which I have seen (that figured at plate 8, fig. 1) is utterly
worthless as a structural specimen. This circumstance, combined with
the rarity and singularly elegant forms of these plants, must ever render
their collection and investigation among the most attractive and curious
branches of Fossil Botany.

The circumstances under which these structural specimens occur are
not the least interesting points in their history. All are found in the
seams or nodules of clay iron-stone, and are very highly mineralized,
sometimes containing crystals of iron, and the cavities in their substance
being filled with white carbonate of lime and magnesia. Those which
are most complete always form the nuclei to nodules of clay iron-
stone—such are those figured at plate 7 and plate 8, fig. 1; but in these
the internal tissues are nearly obliterated. Others again, including all
in which the spores are preserved, have occurred as broken frustules,
within stems of Lepidodendron elegans and other species of that genus.

I have examined as many as thirty specimens of cylindrical truncheons

442 STRUCTURE AND AFFINITIES OF LEPIDOSTROBI.

of Lepidodendron, each from two or three inches to a foot long and
from two to five inches in diameter, from the Staffordshire coal-fields,
all of them containing one or more fragments of Lepidostrobi.* ‘These
Lepidostrobi ave of various lengths, they run parallel to the stem in which
they are enclosed, are of all ages, and in various stages of decomposition.

In one large stem, figured at plate 9, fig. 2, and plate 10, fig. 1,
the remains of upwards of thirty Lepidostrobi are crowded together
with broken-up portions of the bark and of other parts of the Lepz-
dodendron itself. When they are solitary, which is very frequently the
case, they resemble the vascular axis of the Lepidodendron, and are
usually taken for such; but a very slight examination suffices to shew
their true nature.

Usually the fragments of Lepidostrobi are not more than half an inch
long, and very frequently are mere discs ; so that though there is often
the appearance of one several inches long, and traversing the whole
length of the fragment of Lepidodendron, it will generally be found
that this is owing to two being placed each at an extremity of the
truncheon, and opposite to one another. That all were exceedingly
brittle cannot be doubted, for no modern cone of any natural order
could be broken up into the shallow discs which many of these fossils
present. This brittleness of Lepidostrobi, combined with that of Lepi-
dodendron, is in consonance with the remark I have elsewhere made,
that the tribes of plants that were converted into coal, were all of a
singularly compressible and succulent texture.

It is difficult to account for the presence of these fragments of Lepi-
dostrobi, and especially of so many in one Lepidodendron stem as are
shewn in plate 10, fig. 1. We can but conjecture that the trunks of
the latter were erect stumps whose interior was hollowed out by decay,
that these stumps were covered with water, charged, so to speak, with
myriads of broken fragments of Lepidostrobi and other vegetable matter,
which were washed by the fluid into the stumps. These points, how-
ever, demand a separate consideration, and will be better understood
by a reference to the figures in plate 9, and those of transverse sec-
tions of the same fossils in plate 10, bearing in mind that the Lepido-
dendron stems or stumps are stuffed throughout with fragments similar
to those figured in plate 10>

1. The stumps appear to have been rooted and erect, and to have
received the cone fragments into their cavity, as fern fronds find their
way into the axis of Sigillarie. Were the stumps mere prostrate por-

* There is in Lord Stamford’s Museum, at Enyille Hall, a very valuable collection of
coal-fossils, made by Mr. Beckett, of Wolverhampton: amongst them is a cylindrical
specimen of Lepidodendron, containing three species of shells labelled Unio aquilinus
Modiola carinata, and Mytilus triangularis. ‘

STRUCTURE AND AFFINITIES OF LEPIDOSTROBI. 443

tions of stems, it is evident that cones would have lain horizontally in
them, and that no washing or drifting could have induced the fragments
of these cones to lie with their axis parallel to them, or could have
introduced so many into one trunk ; and the latter would certainly have
been materially compressed had they received on one side the pressure
of the superincumbent shales.

2. The stumps must have been submerged, and the fragments quietly
deposited from the water. Had the cones fallen into the stumps from
an overhanging forest, they would have alighted in all manner of irre-
gular positions, and in some cases overlain one another, which I have
never seen to be the case.

3. The deposit appears to have been effected by the gradual subsi-
dence of the water, and not by a sudden rush or current. This again
is proved by the non-interference of the cones, and their uniformly
vertical position with respect to the Lepidodendron, a fact to which I
cannot too strongly call attention. It might be expected, that how-
ever tranquilly they were deposited, some irregularity should occur in
their arrangement, and such may yet be observed, though in the various
sections I have had the advantage of seeing made under my own direc-
tion none has been noticed.

It is hard to account for the accession of so large a volume of
water as should submerge these stumps and deposit the fragments,
and yet exhibit no evidence of drifting in its course. The sudden fall
of a tropical torrent of rain, on a Lepidodendron forest, in which
were hollow stumps of these trees, must at once suggest itself. This
would both carry down the Lepidostrobi from the trees and float up
the fragments on the ground, depositing them together in the stumps.
Another effect of such a fall would be to break down some of the
older trees, whose decaying stumps would be prepared to enclose other
Lepidostrobi on the precipitation of the next similar torrent.

It may be asked, why the same agent that produced all the deposits
over the coal, should not both have filled the trees and enveloped the
whole in silt. ‘To this I would answer, that the action does not appear
to have been sufficiently violent, that the beds of iron-stone nodules, or
continuous seams of clay iron-stone in which this kind of fossil occurs,
bear no evidence of having contained the remains of a luxuriant forest
vegetation, similar to what the coal-shales present; and that the repe-
tition of such phenomena as I have assumed to have filled the stumps
with fragments, is very likely to have impregnated the iron-charged
mud with carbonized matter in a state of the most subtle comminution,
which is the character of the carboniferous iron-stone wherein these
specimens are imbedded.

The extreme fragility of the Lepidostrobi, displayed by these speci-

444 STRUCTURE AND AFFINITIES OF LEPIDOSTROBI.

mens, is to me very satisfactory, as the Lepidodendrons, of which they
are the fruit, no doubt partook of this character, which is eminently
favourable to a rapid decomposition and intimate union with the silt or
mud which is the basis of the clay iron-stone in the one case, and the
formation of a homogeneous bed of vegetable matter, such as the coal
presents, in another. The extraordinary abundance of the fragments,
too, indicates a most vigorous vegetation, for they must indeed have
been profusely scattered to be deposited in such numbers within narrow
cylinders, into which no current appears to have been directed.

It is worthy of remark, that no fern leaves are contained in any of
these Lepidodendron stems; and their absence is the more singular,
from their being commonly deposited, along with branches of Calamites,
&c., in the erect stumps of Svgillaria resting on the coal-shales. This
is no doubt connected with the well-known fact of the Sigillaria stumps
being filled with sandstone, or the same materials as those composing
the stratum above the shales they root into; whilst the fossil Lepidoden-
dron of the clay iron-stone seams is of the same mineral as that wherein
it is imbedded. Were the fragments of Lepidostrobi washed into their
enclosing stumps by any current, that agent would in all probability
have transported the remains of other plants to the same spot.

The perfect preservation in which these fragments occur must be
attributed to the protection afforded them by the surrounding Lepido-
dendron bark. That the circumference of the latter has been subjected
to pressure may be inferred from the flattening of the prominences to
which the leaves were attached. This pressure was moreover very con-
siderable, as may be proved by comparing the evenness of their surface
with that of a piece of Lepidodendron bark fossilized without pressure
and embedded within the stem along with the Lepidostrobi (plate 10,
fig. 1). From this it may be seen that the scars of the surface occupied
the faces of diamond-shaped projections, elevated a full } of an inch and
more above the surface of the stem, and that they were separated from
one another by deep grooves which dilate inwards: in other words, the
prominences are broader at their bases than on their surface.

To illustrate this beautiful and hitherto unobserved character of
Lepidodendron bark; I have figured (plate 8, fig. 12), a unique speci-
men preserved in a very hard and tough iron-stone, for the loan of
which the Geological Survey is indebted to Mr. Cooper, of Bilston. It
is not known to what favourable combination of circumstances this fossil
owes its rare and perfect state of preservation: it is the only specimen
of the kind that I have seen, and, like the Stigmaria fragments figured
at plate 1, it throws a new light upon the subject it illustrates. In this
the surface that bore each leaf projects a full } of an inch beyond -
that of the stem itself, and is perforated by a tubular cavity through

STRUCTURE AND AFFINITIES OF LEPIDOSTROBI. 445

which the bundle of vessels that diverged from the vascular axis of the
stem to the leaf passed out.

The uniformity of the surface of the Lepidodendrons figured at plate
8, is an argument for their having been erect when fossilized, and
proves that the deposition of the Lepidostrobi, &c., into their interior
was effected by, comparatively speaking, tranquil waters. Had the
specimens been drifted, or exposed to the action of a rushing current,
on one side or the other, they must have shown it by the obliteration of
some of the scars, or probably the adherence of other fossils to their
surface. Again, had the stems been prostrate, the same pressure that
flattened the once prominent scars, would have compressed the cylinder
also.

The above are the only observations I have to offer upon these spe-
cimens relatively to the formation wherein they were found, and the
circumstances under which they appear to have been fossilized: the re-
mainder refer to the botanical characters of the Lepidostrobi them-
selves, to the strong evidence these specimens contain of their belonging
to the Lepidodendrons that enclose them, and to the proofs which the
discovery of their spores and tissues yield of both being allied to the
existing genus Lycopodium.

Description of the Specimens.*

Plate 3. This and the following plate represent the fragment of a
Lepidodendron, enclosing in its once hollowed axis several Lepidostrobi.
The specimen was received through Mr. Jukes from Mr. Lowe, of
Wolverhampton, who obtained it from the iron-stone seams in the
neighbourhood of that town.

Fig. 1. External surface of the fossil, displaying the markings of the
Lepidodendron Harcourtii of the natural size.

Fig. 2. Lower surface of the same, showing the fractured surfaces of
a mature and very young Lepzdostrobus.

* It is with regret that I find myself obliged to introduce here a detailed description
of the specimens, of which the plates are faithful and beautiful copies; but to under-
stand all the points of structure in these interesting fossils, it is essential that the
reader be familiar with these. Perhaps there is no more tedious and perplexing task
than that of restoring a fossil plant of the coal formation. WNo solitary specimen among
the many prepared at great cost and labour by the Geological Survey has fully illustrated
any of the various points now, I hope satisfactorily, cleared up. The restoration of a
Lepidostrobus is effected by taking a cell here and a cell there out of many individuals,
and the tubes of vascular tissue, spores, &c., from various other cones. The adaptation
of these to one another, and arrangement of the whole, are, after all, a mental operation.
It is, however, imperative on the restorer to give every feature he has availed himself of
in detail ; that the experienced observer may pronounce on the justice of his views, or
point out the error of his judgment; and that the tyro may know how to proceed in follow-
ing similar investigations.

446 STRUCTURE AND AFFINITIES OF LEPIDOSTROBI.

Fig. 3. Magnified view of the four somewhat distorted scars on the
surface of fig. 1.

Plate 4, Fig. 1. Opposite surface of the fossil from that figured in
Plate 3, showing a large Lepidostrobus, whose base rests on the curved
crest of what was perhaps a branch supporting the cone.

Fig. 2. Upper fractured surface of the fossil, showing the large
mature cone a, and a somewhat smaller one lying parallel to it. The
point, where lines, drawn from the daggers a@ and 6b would intersect,
indicates the position of the very perfect apex of one of the sporangia
which are arranged round the axis.

Fig. 3 represents four scales (two restored) from the surface of the
cone a.

Fig. 4. Sporangia chipped off the cone a.

Fig. 5. Magnified view of a sporangium, restored according to the
appearances presented by the apex of the sporangium of the smaller
cone seen in fig. 2. -

Plate 5. Vertical sections from the Lepidostrobus figured in Plates 3
and 4, taken through the axes of the cones.

Fig. 1. A vertical section taken from the cone, Plate 4, figs. 1 and
2a, in the direction of the dagger a, cutting through the axis and por-
tions of several sporangia on either side. The axis retains hardly a
trace of organization beyond a little cellular tissue on its outer margin.
There are no remains of the bases of the scales supporting the spo-
rangia, which were once inserted into the axis. The sporangia are
large and long, occupying nearly the whole semidiameter of the cone,
and slightly curved, with the convexity upwards, as shown at Plate 4,
fig. 4. Their walls are of extreme tenuity, formed of a single row of
parallel cells (seen at Plate 6, fig. 4c and 10), and they contain the
spores.

Fig. 2 is a magnified view of some of the tissues of the walls of the
sporangia 6, and the vascular tissue of the scales a, the latter composed
of tubular vessels which communicate from the apices of the scales and
the axis of the cone.

Fig. 3 shows the dorsal portion (that nearest the axis) of a sporan-
gium, whose position is indicated by the intersection of lines drawn
from the daggers a, of fig. 1. Towards the back of this are seen
numerous spores from the opaque mass which fills the greater part of the
sporangium.

Fig. 4 is a portion of the same, so magnified as to define the
spores.

Fig. 5. Dorsal portions of two other sporangia, indicated in fig. 1 by
the daggers bc, between the bases of which some escaped spores are
enclosed in a transparent mineral -.

|
,
.

STRUCTURE AND AFFINITIES OF LEPIDOSTRODI. 441]

Fig. 6. The portion marked z in fig. 5, magnified equally with fig. 4.

Fig. 7. A group of spores from fig. 5, very highly magnified, exhi-
biting their characteristic trisection by opaque lines.

Fig. 8. A smaller group of more advanced spores from fig. 8, still
more highly magnified.

Fig. 9. Individual spores selected from figs. 4 and 6. They are ge-
nerally tetrahedral bodies, divided into three by broad transparent
spaces. ‘The youngest, a, b,c, are immature, and present spurious pro-
jections at their three angles, which spines are continuous with the
transparent spaces and acute, deformed by prolongations of the con-
tiguous apices of the three sporules comprising the spore. Letter d
indicates a more mature spore, with the spines very nearly obliterated.
At letter e is a still more matured spore, whose angles are rounded, and
in which the transparent spaces are broader, indicating that the spore
has all but broken up into three sporules.

Fig. 10. A vertical section, of the natural size, of a portion of the
cone, Plate 4, fig. 2 a. In this the tissues of the axis are preserved.
This axis is hollow in the centre from the obliteration of the once cel-
lular (?) tissue filling it.

Fig. 11. A semi-diameter of the axis of fig. 10, showing at a the
tubes of vascular tissue, which once formed a continuous zone round
the pith; 6, the system of vascular tissue exterior to the former, which
gives off the bundles c to the scales of the cone. The arrangement of
these tissues will be best observed by referring to the wood-cut of these
parts in Lepidodendron, at p. 436 of this volume,

Fig. 12. Tubes of the perpendicular vascular tissues a, of fig. 11.
These are marked with horizontal rings, sometimes free, at others anas-
tomosing at various points.

Fig. 13. Slender tubes 6, of the vascular system, ¢ of fig. 11, sup-
plying the scales. They are surrounded by cellular tissue d, formed of
elongated utriculi. The tubes are much more delicate and slender than
those of the other system of vessels.

Plate 6. Horizontal sections of the cones, Plates 3 and 4.

Fig. 1. Slice of the mature, a, and immature, 0), cones, seen in
Plate 3, fig. 2.

Fig. 2. A portion of the coaly bark close to the surface of the Lepi-
dodendron, showing its cellular tissue.

Fig. 3. A highly magnified portion of the cone, fig. 1a, showing a
section of the apex of one scale, with several others imbricating over it.

Fig. 4. Still more highly magnified view of a portion of a similar
scale, edged by the walls of a sporangium c. The scale is chiefly formed
of cellular tissue, enclosing in its centre a mass of vascular tissue a,
which turns up to the ascending apex of the scale at 6.

448 STRUCTURE AND AFFINITIES OF LEPIDOSTROBI.

Fig. 5. Vessels seen at fig. 4 4, as they ascend towards the apices of
the scale, fig. 4, and are consequently cut across by the lapidary.
These vessels are figured at Plate 5, fig. 2 a, and Plate 8, figs. 6, 7,
and 11a.

Figs. 6 and 7. Portions of the crystallized cellular tissue of the axis
of the cone, fig. 1 a, in which the circular holes, a, indicate the position
of the vascular bundles supplying the scales, which are vertically sliced
in Plate 5, fig. 11 e¢.

Figs. 8 and 9. Magnified portions of the small mutilated cone,
fig. 1 b, showing the positions of the spores which are retained in the
young cone, but of which there are no traces in the large cone, fig. 1a.

Fig. 10. A portion of the walls of a sporangium, cut obliquely by a
horizontal line, so that at one portion, a, the lateral cells are divided
according to their long axis, whilst at the other, }, the cells forming the
floor, are cut across their short axis.

Fig. 11. Ripe spores from the small cone, fig. 1 5, one of them, a,
divided into four cohering sporules.

Fig. 12. Other spores more advanced, a, ready to separate; 6, the.
sporules detached from one another ; 3, a sporule detached wholly, pro-
bably in a state once ready for germination.

Fig. 13. Horizontal section of another part of the cone figured at
Plate 3, fig. 2. This is crushed on one side, but tolerably perfect on the
other, and there displaying some ripe sporangia, with a few spores in
each.

Fig. 14. Magnified view of the axis.

Fig. 15, Portions of three sporangia, with a few spores scattered at
their bases.

Fig. 16. More highly magnified view of portions of these sporangia
and their contained spores.

Fig. 17. Spores from the same, in various stages of development.

Fig. 18. Some spores very highly magnified, all young, and the com-
ponent sporules not only separated, but bounded by a transparent space,
and the whole surrounded by a jagged border, probably the effect of
crystallization.

Plate 7, illustrates a beautiful specimen of Lepidostrobus ornatus,
preserved in a nodule of iron-stone. The specimen is from Glamor-
ganshire, is preserved in the Bristol Museum, and liberally placed
at the service of the Survey, not only for illustration, but for slicing,
by Mr. Stutchbury, with the approbation of the directors of that
institution. : :

Fig. 1. Lepidostrobus ornatus, Lindl. and Hutt., of the natural size.

Fig. 2. Vertical slice, taken through the axis, showing that the apex
is undeveloped, and the cones hence hardly mature.

STRUCTURE AND AFFINITIES OF LEPIDOSTROBI. 449

Fig. 3. Vertical slice of lowest scales and sporangia.

Fig. 4. Horizontal slice of semi-diameter of cone through the middle.

Fig. 5. Vascular tissue of the scales supporting the sporangia.

Fig. 6. Cellular tissue of ditto, restored from cells in various scales.

Fig. 7. Cellular tissue comprising the walls of one of the sporangia.

Fig. 8. Ideal representation of a horizontal section of a restored cone,
showing eight scales, radiating round, and inserted into the axis, each
bearing a sporangium.

Fig. 9. Ideal representation of a vertical section of two sporangia and
their supporting scales, the latter traversed by vascular tissue.

Plate 8, Fig. 1. Fragment of a cone of Lepidostrobus ornatus, in
which the apices of the scales are admirably preserved. Specimen from
Mr. Beckett of Wolverhampton.

Fig. 2. Apices of scales of ditto magnified.

Fig. 3. Upper extremity of a Lepidostrobus from which the apices of
the scales are removed. In the collection of Miss Jukes, of Birmingham.

Fig. 4. Vertical section of the same, showing that the axis has fallen
away, as also the apices of the scales.

Fig. 5. Magnified view of the fractured terminations of the scales,
each with a prominence in its hollow, which is the projecting bundle of
vascular tissue, shown at fig. 11 a.

Fig. 6. Portion from another specimen of L. ornatus, showing the
union of the walls of the sporangium ec, with the scale d, at the point 4;
ce is the vascular axis of the scale.

Fig. 7. More highly magnified portions of fig. 6, showing the dispo-
sition of the cells at the junction of the walls of the sporangium, ¢, with
the scale, d. At letter a is seen the vascular tissue of the latter, and at
e an escaped spore.

Fig. 8. Vertical slice of another Lepidostrobus ornatus.

- Fig. 9. Mutilated apex of ditto, with spores escaped and mineralized
in a transparent medium.

Fig. 10. Spores from the same, some of them broken up into their
component sporules.

Fig. 11. Restored view of axis of cone, x, with two supporting scales,
6, transversed by vascular tissue, a, the lower one bearing a sporangium,
s, with its contained spores. The apices of the scales are represented
broken off, as at figs. 83, 4, and 5, but restored by dotted lines, as in
figs. 1 and 2.

Fig. 12. Fragment of Lepidodendron elegans, preserved in iron-stone,
without any material compression, and hence showing the surface scars
as prominent bodies. It is in the collection of Mr. Cooper, of Bilston,
who procured it near that town.

Plate 9, Fig. 1. Specimen of Lepidodendron elegans in the cabinet of

450 STRUCTURE AND AFFINITIES OF LEPIDOSTROBI.

Mr. Cooper, of Bilston, containing three tolerably perfect Lepidostrobe,
and the remains of very many others, all vertically disposed in its axis
—of the natural size.

Fig. 2. A much larger specimen of the same fossil, reduced one-third,
from Mr. Beckett, of Wolverhampton, containing more or less perfect
fragments of upwards of thirty cones of Lepidostrobus ornatus, besides
other vegetable matter.

Plate 10, Fig. 1. A horizontal section of the fossil represented in
Plate 9, fig. 1, showing the arrangement of the Lepidostrobi in its axis.

Fig. 2. Magnified view of semi-diameter of a very young cone,
whose position in Fig. 1, is indicated by the two arrows. It is also
probably the upper end of a cone, and contains sporangia full of spores.

Fig. 3. Spores and sporules from the same, of various ages and sizes,
all rather mutilated. .

Figs. 5 and 6. Fragments of vessels forming the vascular axis of the
scales, in the same specimen.

Fig. 7. Horizontal view of Plate 9, Fig. 2, of the natural size, show-
ing the numerous fragments of cones, and a curved portion of the bark of
Lepidodendron, the position of which is indicated by the four arrows.

If the several points thus obtained be, so far as they refer to the
structure of the cones, singled out and arranged, it will be seen that in
the aggregate they afford a very complete knowledge of every part of
the genus Lepidostrobus, which may be thus described :—

Cone variable in length, cylindrical, obtuse at both ends, gradually
tapering towards the apex, formed of a perpendicular axis, around
which are arranged horizontal scales, each bearing a sporangium, or
hollow vessel filled with spores. .

Axis cylindrical, consisting chiefly of cellular tissue, traversed by
tubular vessels, which compose the vascular tissue. Vascular tissue of
wood, Plate 5, fig. 11 a, forming a continuous ring enclosing the pith
of the axis, composed of long hexagonal tubes, whose sides are marked
with free or anastomosing transverse lines. Around this vascular tissue
of wood are arranged bundles of smaller and more delicate tubes,
Plate 5, fig. 11 6, which radiate outward to the basin of the scales.

Scales horizontal, 8-16 in a whorl, composed of two parts. Firstly, a
slender pedicel, inserted into the axis and supporting the sporangium,
Plate 8, fig. 11 5, formed of vascular tissue, through the axis of which
runs the bundle of vascular tissue, fig.11 a. Secondly, a broad dilated
apex, at right angles to the pedicel, fig. 11 c, produced upwards into a
triangular acute point, fig. 2, and downwards into a blunt lobe, fig. 11 e,
also traversed throughout its length by a continuation of the vascular
bundle of the pedicel. This dilated apex is formed of very loose
cellular tissue laxer, especially towards the centre, which is generally

E
f
p
7
‘
:
;
|

oy) ae a he A

STRUCTURE AND AFFINITIES OF LEPIDOSTROBI. 451

hollow ; the anterior portion of this body is hence frequently removed,
as seen at Plate 8, figs. 8 and 5, leaving a ‘pitted surface to the
cones.

Sporangia oblong, resting on the pedicel, and each attached by its
lower face towards the apex of the latter by a small surface. Some,
perhaps all, of these sporangia are furnished with ribs, Plate 4, fig. 5.
Walls formed of a single layer of transversely elongated cells, Plate 6,
figs. 4 and 10, and Plate 8, fig. 6 a. Mode of dehiscence or bursting
to allow the escape of the spores unknown.

Spores consisting of three, or rarely four sporules, which are after-
wards separated from one another, the immature produced at their
angles into acute spines, the older suborbicular.

The two cones, from which the above general views have been de-
duced, are apparently from different species, and I shall therefore cha-
racterize them as such. Further, as they seem to have belonged to the
fossil Lepidodendron enclosing them, I shall for the future notice them
as really the fruit of that genus.

1. LeprpopENDRON elegans. Cone slender, ? inch in diameter, 4-10
inches long, sporangia 8 in a whorl.

2. LErIDODENDRON Harcourtii ? Cone broad, 14 inch in diameter,
sporangia about 16 in a whorl.

If, now, these cones, be examined with reference to the known con-
temporaneous fossils which accompany them, it will appear impossible to
deny their having the reproductive organs of Lepidodendron, not only
from their association with the fragments of that genus, because the
arrangement of the tissue in the axis of the cone entirely accords with
that of the stem of Lepidodendron ; just as we find, in modern cones of
Lycopodiacee and Conifere, that the axis is a continuation of the
branch which bears leaves, modified into organs adapted to support and
protect the parts of fructification.

The most positive evidence that can be adduced of Lepidostrobi bemg
a genus allied to Lycopodium, is afforded by the spores, the presence of
which not only removes them from Cycadee, Conifere, or any other
order of flowering plants, but directly refers them to the family of
Lycopodiacee and Conifere. In both, the original spore divides into
three or nearly four sporules, which are angular, and form the repro-
ductive system of the plant. Not only do these groups coincide in the
essential character of their spores, but, in many minor points, the
strongest similarity exists between them. The arrangement of the
scales is the same in both, and so are the scales themselves in general
features, especially towards their dilated apices. The situation of the
sporangia, too, is alike, and their attachment by a very narrow surface
to the scale.

II. 21

452 STRUCTURE AND AFFINITIES OF LEPIDOSTROBI.

To illustrate this properly, the following illustrations (fig. 1), were
prepared. That at Plate 8, fig. 11, should be compared with this
wood-cut, where the only material
difference between them is in the
shape of the sporangium. The
spores ¢, from L. Selago, should be
compared with those of Plate 5,
figs. 7, 8, 9; Plate 6, figs. 11 and
12; Plate 8, fig. 10; and Plate
10, fig. 3.

Previous to the investigations
of M. Brongniart, the prevailing
opinion with regard to these cones
was, that they belonged to plants
of the Pine tribe. Their great
size and the presence of coniferous
wood in the coal formation both
favour this supposition.

It is, however, well known to the
botanist, not only that cones are
far from being peculiar to one
natural order of plants, but that
their extreme form is no indica-
tion, either of their contents, or of the affinities of the plants which
produced them. Accordingly, we find that Dr. Lindley, the first
English observer who published any extended views on the affinities of
these plants, suggests the probability of their being referable either to
Conifere, Lycopodiacee, or more probably still to Cycadew.*

The scales, to whose arrangement the production of a cone is due,
are in no essential respect different from those that occur on other parts
of the plant. The doctrine of morphology teaches us, that the cone
is nothing more than the leafy apex of a branch, whose leaves are
modified in form, generally to the end that they shall perform the
office of protecting organs to reproductive bodies ; this is the case with
the Pine cone, that of the Lycopodium or club-moss, and many other
plants.

Cases are however not unfrequent, of what may be termed false
cones, which have no relation to the reproductive organs of the plant.
In some species, they are common, and are naturally produced, as in
the well-known cone-bearing willow, and in the larch, which produces

barren cones, unconnected in every way with those containing the
seeds.

* Lindley and Hutton, Fossil Flora, t. 11, sub. Z. variabilis.

STRUCTURE AND AFFINITIES OF LEPIDOSTROBI. 458

The normal arrangement of the leaves of a plant is spiral around the
branch; when the latter is very short, and the leaves also short and
crowded, the whole branch becomes a cone ; this is frequently seen on
the apices of the branches of some plants, and is also artificially effected
in others, by an injury to the branch, which prevents its elongation.
In Tierra del Fuego there is a genus, several species of which have the
branches punctured by an insect; the consequence is, that the wounded
portion is arrested in its tendency to prolong, the sap is thrown into the
leaves, which become broader and otherwise of different appearance
from the usual form of leaf, and lap over one another, so as to produce
a cone.

Ihave thus long dwelt upon the formation of spurious cones, because
some of the so-called Lepidostrobi may be of this nature, —- witness
the Lepidodendron oocephalum, * of which it is impossible to say
whether it be a Lepidostrobus or the apex of a branch crowded with
short leaves.

The Fuegian plant mentioned above illustrates this subject, and a
wood-cut of two species of the genus alluded to (Pernettya, allied to
the “ Heaths”) is added. Both were collected at Cape Horn, where
the plants were invariably thus coniferous in the diseased branches,
while the healthy bore little bell-shaped flowers. Instances of this sort

are not very uncommon on individual species, but a general tendency
in the development in question is sufficiently curious. Were the above
figured plants to occur fossil, the probability is, that their cones would
be regarded as the undoubted reproductive organs, and the plants
themselves be provisionally referred to Conifere ; in further evidence of
which opinion the cold climate they inhabit might be adduced, and

* Lindley and Hutton, Fossil Flora, v. iii., t. 206.
212

454 STRUCTURE AND AFFINITIES OF LEPIDOSTROBI.

their general similarity to some really coniferous and totally different
plants of Tasmania.

Cones, again, instead of being fruit-bearing organs, are frequently
mere aggregations of male flowers. This is the case especially with
the Cycadee, the cones of one species of which (Dion edule of Mexico)
entirely resemble those of the truly coniferous genus Araucaria.

Proteacee are an order whose species abound in, and are almost con-
fined to, Australia and South Africa, and whose (hermaphrodite) flowers
are very frequently arranged in cones. Such cones would, if fossilized,

and in the absence of any other evi-

Soba dence, be considered to have belonged

to a coniferous plant. An illustration

is added of such a Proteaceous cone,

which, without examination, is calcu-

lated to deceive, even when recent, and

if petrified, would assuredly be con-
sidered as belonging to a pine.

Lycopodiacee are the last (though not
the only other) cone-bearing genera I
shall mention ; but as Lepidostrobi are
to be directly compared with these, I omit any further mention of them
here.

The size, form, and arrangement of their parts, together with the
general disposition of previous observers to class the Lepidostrobi among
Conifere, did not deceive M. Brongniart, who rightly conjectured them
to belong to Lycopodiacee, and that too without any of that direct
evidence, afforded by the specimen in the museum of the Geological
Survey. Dr. Lindley, with his usual sagacity, had previously referred
the cones to Lepidodendron, though ignorant of the nature of those
tissues which also it has been our good fortune to obtain. In this
opinion M. Brongniart concurs, and proceeds, in a review of the cha-
racters afforded by both Lepidodendron and Lepidostrobus, to cite his
reasons for allying them to Lycopodiacee, rather than to Conifere,
under which the authors of the Fossil Flora of Great Britain had
arranged Lepidodendron.*

From the curious dichotomous ramification of the stem, M. Brong-

* The extreme perplexity of this subject may be gathered from the hesitation with
which so excellent a botanist as Dr. Lindley always speaks of the affinities of Lepido-
dendron, In the introduction to the Fossil Flora, he regards it as Lycopodiaceous ; under
Halonia gracilis (Tab. 86), as probably coniferous; Lepid. Harcourtii (Tab. 89, 90) is
pronounced intermediate between Zycopod. and Conifere ; Lep. selaginoides (Tab. 113)
is, he continues, rather Coniferous than’ Lycopodiaceous ; and finally, under Lep. elegans
(Tab. 118) and Megaphyton approximatum (Tab. 116), the Lepidodendree are definitively
referred to Conifere.

ee hd ee

t

STRUCTURE AND AFFINITIES OF LEPIDOSTROBI. 455

niart inferred that Lepidodendron could not be a Dicotyledonous plant.
The arrangement of its tissues clearly forbade its belonging to Mono-
cotyledones. ‘The cylindrical form of many of their (erroneously so
called) cones was conclusive against their belonging to any coniferous
plant, where these organs are always truly conical; whilst the form of
these organs and the arrangement of their parts entirely tallied with
Lycopodiacee.

In the details of M. Brongniart’s description of the genus Lepi-
dostrobus, there are some errors which, with the aid of more copious
and perfect specimens, may now be corrected. It is satisfactory to
know, that these rectifications throw light upon certain important
points, in which it was supposed that this genus differed from any
existing vegetable, presenting a most anomalous feature.

The first error then relates to the arrangement of the sporangia
upon the supporting scales; they were imagined to be borne on the
under-surface of the scales,* a misconception probably arising from
mistaking the base for the apex of the cone, as shown at Fig. 2 a of
Tab. 23.—( Végétaux Fossile, vol. ii.)

M. Brongniart further supposed the horizontal portion or pedicel
of the scale, to be very broad, and to surround the base and two sides of
the sporangium, which was thus enclosed between its own scale and
axis, on all but the upper surface. In the specimens I have examined,
on the contrary, the pedicel of the scale is exceedingly slender in com-
parison to its dilated apex, and of far less breadth than the sporangium,
whose point of attachment to it is so small, that it can seldom be divided
during slicing by the lapidary.

Lastly, M. Brongniart hesitates on pronouncing the ZL. ornatus of
Lindley and Hutton to be Lycopodiaceous. If, however, as I suppose,
the specimen, figured: at Plate 7, is rightly referred to that plant, its
affinities are undoubtedly with the other Lepidostrobi.

With regard to the number of species of Lepidostrobus, comprised in
the Plates 3, 4, 7, 8, 9, 10, I do not recognize more than two, with any
degree of certainty. One of them is that contamed in the trunk of
Lepidodendron (figured at Plates 3 and 4), and which is probably iden-
tical with the fragments figured in Végétaux Fossile, (vol. i, t. 23,
fig. 5 a,b). The second is the very long cylindrical cone, included
within the trunk of Lepidodendron elegans (figured at Plate 8, fig. 1),
and to which, in all probability, should be referred the cones Plates 7,
8, figs. 3 and 8, and Plate 10. This is allowing for great variation
in the length of the cone, in one species, not more, however, than may
be seen in very many of their existing allies the Lycopodia as the

* Végétaux Fossile, vy. ii., p. 52.

456 STRUCTURE AND AFFINITIES OF LEPIDOSTROBI. <

accompanying wood-cut of the cones of the LZ. Magellanicum will show.
The internal structure, form of the sporan-
gium, the length of the apices of the scales,
and breadth of the cones, are very nearly
alike in all the specimens. To this L.
ornatus (Brongniart’s Figs. 1, 2, and 4
of vol. ii., Tab. 23), should, possibly, be
referred. .

The beautiful little nut, of which a —
wood-cut is added,* is most probably a
sporangium, belonging to this cone of Lepi-
dodendron elegans ; though differing in
shape, its sculpturing corresponds with that
of sporangium, removed and restored from
Lepidostrobus, Plate 4, fig. 5.

The young state of this species is the
very elegant specimen contained in a no-
dule of elay-iron-stone, figured at Plate 7.
Besides the more conical and shorter form 4
of the cones, they are distinctly attenuated 4
to the apex, as if hardly matured. The :
apices of the scales, too, do not appear to a
be produced downwards to the degree of 7
those of the former species. =

Fig. 5.

* From the collection of John Gray, Esq., of Dudley. It is fossilized in a nodule of
iron pyrites.

Geological Smbey othe United Kingdom,

ROOTLETS ef STIGMARIA.

Vol.2, P? 2

¢ if)
i}

4 H Hj | « } i \ (s2 ROSA Ae
Ycolog Lea | Survey ot the United bicaian oO. Plate IT

:
i

ee
ty easy Fae
NTR

1

) WiBaily del.

LEFIDODENDRON avd LEPIDOSTROBUS

Plate. Tid

boo sin as “y
aS ia eF
a

Vals eae

logical Surbep of the nited Kingdon. Plate 10

—_—— ———_—-- =

ERT DOS MRO BUS

Vol.2. P? 2

PGeologisal Surbev of the United Kington. Plate V

LEPIDOSTROBUS

JID Hooker IW.Lowre ft

(O5¢0 lagien l Surbe p of the Mnitten SSmgbou.

LEPIDOSTROBUS

J.D. Hooker

N. Baaly del.

Vol.2.L? 2

Plate V/

os

he

Nor er 5}

Vou.2. P? 2

maa | cle £4 a wes : 1 ¢ % oe
Geological Surbep of the Rnited Kingdom. Plate VIL

usc (UCU UICC

1 rm

accent (HU cuce TU UU ECAR UC UEEETUCEUC UHC

TUCHTATHTMMUAMLMMMMMNA Hi)
AT}

LEPIDOSTROBUS orRNAtTUS

ID. Hooker UW. Lowry

Vol. 2. £*

‘beological Survev of the United Kimgdam. are

Oe 7

= © sess

- >>

-_?

>
—, >

LEPIDOSTROBUS orwatus avd LEPIDODENDRON ELEGANS.

Seals Mewl Survey of the tH

LEPIDODENDRON ELEGANS

q CR.Bone del. J.D. Hooker

LEPIDODENDRON

TROBI zz

LEPIDOS

ee

AH a

ae
vie
‘

1

a

i i St
Ria i yy “
Be ie

pew

wy

i i HN Cy y a =)
4) na hiy ely * +t a i=

% Paty a
j ‘W iki,

ate

i 2 ;
a 1)
'
ae ill
hie {
1
Pa
7 t' ¢
1 i
a ,
Me
f \
,
i \
Mare |
]
' 1%,
rr , q :
H ‘ ‘ r
\ 1
5
| if
{
{ \
i 4 i) jel vol
ie in
hoy i , \y
‘ : \
Ver x ee
i i ie a te
5 i ‘| i!
mit Ageia i) dl
i, rata o
eet i Wh lad hu
' ' i We

ies

ms

4
vi

* Vrs es aa ,
ee

en

SE e rer ene

*
rere ete!, -

wee
Aen
Poi ciate gto

Per PEE
¥

Pusan tt
Dew hies
nee

oe
Siarery

Lely BP presee bes tras
i MH 1 *
tite

yt

patente
Peery
: .
F ’ sth ;
Htietate eats teers
wiht PUDMT Breas prata as teatatrrad ep nresateeetatetate®
i vr Sit Beatty : perverse

posi tetts
Thtube ese

LEED beet ares,
eee
hie ve
AB fac

a sresPat etait
wyeeraet
balk

HORT
Petey

itr roa
gpa

Vierasg
wastes

pees

Weta Wee
Usbseeatysisrtters cane
Ntenit

fines

rrr
nrrarhitn
eae

PU orrerr:

Hileaut
betas
Seu
rile
Heit

i : oe
esa ; Hu) ae
nie aie pt ye ;

" a3 ite

CUT rresesier trate r Hite be te
rire 7 2 A beta eat stare gure:
aa tree i ’ fe
+] he bike
Sheet &
an
Ene ie eo
rte is iY ve 6