73,HS3SQ9 OSeO-9

W. G. FARLOW

HARVARD BOTANICAL MEMOIRS

os

At a meeting of the Botanical Department held
Oct. 20, 1903, the following vote was passed:

‘“ Under the head of Harvard Botanical Memoirs it is
proposed to include all quarto publications issuing from
the Gray Herbarium, the Cryptogamic Herbarium, and
the Botanical Laboratories of Harvard University, including
theses presented for the degrees of Ph.D. and S.D. in Botany.
Inasmuch as some of the future publications are likely to be
continuations of subjects treated in quarto papers already
published, it seemed desirable to begin the numbering of the
Memoirs with the year 1880, the date of the first quarto
publication of any member of the botanical staff at present
connected with Harvard University.”

At a meeting on Nov. 25, 1916, it was voted to dis-
continue the series of Botanical Memoirs. In all, nine
numbers have been issued, the titles of which are given
below.

I. The Gymnosporangia or Cedar-Apples of the United States. _ a
By W. G. Farlow. Anniversary Memoirs, Boston Soc.
Nat. Hist. 1880. Pp. 38. Pls. 1 and 2.

II. The Entomopthoreae of the United States. By Roland /
Thaxter. Mem. Boston Soc. Nat. Hist., IV, No. 6. Pp.
133-201. Pls. 14-21. April, 1888.

III. The Flora of the Kurile Islands. By K. Miyabe. Mem.
Boston Soc. Nat. Hist., 1V, No. 7. Pp. 203-275. Pl. 22.
Feb. 1890. .

IV. A North American Anthurus: its Structure and Development. noe
By Edward A. Burt. Mem. Boston Soc. Nat. Hist., 111,

No. 14. Pp. 487-505. Pls. 49 and 50. Oct. 1894.

V. Contribution towards a Monograph of the Laboulbeniaceae. _/
By Roland Thaxter. Mem. American Acad. of Arts and
Sei. Boston. XII, No. 3. Pp. 189-429. Pls. 1-26.
Presented May 8, 1895. Issued Oct. 14, 1896.

VI. The Development, Structure, and Affinities of the Genus
Equisetum. By Edward C. Jeffrey. Mem. Boston Soc.
Nat. Hist., V. No. 5. Pp. 155-190. Pls. 26-30. April,
1899.

ae

VII. The Comparative Anatomy and Phyllogeny of the Coni-
ferales, Part I. The Genus Sequoia. By Edward C.
Jeffrey. Mem. Boston Soc. Nat. Hist., V, No. 10. Pp.
441-459. Pls. 68-71. Nov. 1903.

VIII. The Comparative Anatomy and Phyllogeny of the Coni-
ferales, Part II. The Abietineae. By Edward C. Jeffrey.
Mem. Boston Soc. Nat. Hist., VI, No. 1. Pp. 1-87. Pls.
1-7. Jan. 1905.

IX. Contributions towards a Monograph of the Laboulbeniaceae,
Part Il. By Roland Thaxter. Mem. American Acad. of
Arts and Seci., XIII, No. 6. Pp. 219-469. Pls. 28-71.
June, 1908.

“or

at ee ene

MEMOIRS.

OF THE

BOSTON SOCIETY OF NATURAL HISTORY;

VOLUME 5, NUMBER 10.

THE COMPARATIVE ANATOMY AND PHYLOGENY OF THE CONIFERALES.
PART 1.—THE GENUS SEQUOIA.

By EDWARD C. JEFFREY.

BOSTON :
PUBLISHED BY THE SOCIETY.
NovEMBER, 1903.

10. Tue ComparativE ANATOMY AND PHYLOGENY OF THE CONIFERALES.

Part 1.—Tue Genus Sequora.!

By Epwarp C. JEFFREY.

(Read April 15, 1903.)

INTRODUCTION.

In the present and following memoirs, it is my intention to describe certain features
of the anatomy of the Coniferales which appear to be of interest. The actual state of our
knowledge of the morphology and history of the group is not sufficient to justify much
certainty as to its classification, and consequently additions from any standpoint are likely
to be of value. The prevailing views as to the relationship of the various orders of the
Coniferales are based almost entirely on reproductive characters, and valuable as these
must always be, they constitute, nevertheless, but a single Ime of evidence. The study
of the anatomy of the older groups of Gymnosperms has done so much to clear up the
question of their affinities that it does not seem unreasonable to expect that a good deal
should be learned from investigations carried out on the same lines in the case of the
Coniferales, the prevailing Gymnosperms of the present day. It is to be anticipated that

the results of such investigations will serve to correct and supplement the conclusions

drawn from the study of the reproductive organs alone.

The existing Sequoias are chosen as the subject of the first memoir, both because
of their interest as the sole survivors of a genus which once flourished in many species
throughout the entire northern hemisphere, and because investigations, already some-
what extensive although as yet incomplete, make it apparent that the genus Sequoia
presents a striking example of those rare and important forms which so infrequently per-
sist as links of transition between distinct and different natural orders of the present day.

Tur Stem or Sequoia gigantea.

Figure 1, plate 68, illustrates the structure of the wood in Sequoia gigantea as it
appears in a transverse section of the heart of an old stem. There are three annual rings
represented in the figure. In contrast to S. sempervirens, to be described later, the

1 Harvard Botanical Memoirs.— No. 7.

442 EDWARD C. JEFFREY ON

autumnal tracheids form a very narrow zone in each annual ring of growth, and the
wood consequently lacks the strength which is characteristic of the latter species. The
radial rows of tracheids are interrupted at intervals, especially in the autumnal region,
by the resin cells which are so constant a feature of the wood of the Taxodineae and
Cupressineae. As has been pointed out by Penhallow (Generic characters of North
American Taxaceae and Coniferae, Proc. and trans. roy. soc. Canada, 1896, series 2, vol.
2, p. 53), the resin ducts, found so frequently in the wood of the Abietineae, are absent in
the mature ligneous cylinder of S. gigantea. Figure 2, plate 68, gives a general view of
the central cylinder of a branch of a young tree from the arboretum of Leland Stanford
university, California. The stem is a year old and shows as a consequence a single ring
of wood. The resin cells of the wood are poorly developed as yet, although those of the
phloem are very apparent in spite of the comparatively low magnification. The minute
structure of the fibrovascular tissues cannot be made out.

Figure 3, plate 68, is of the central cylinder of a five year old branch of S. gigantea
taken from a tree which had already produced cones and seeds. The autumnal wood is
relatively much thicker than in the old stem shown in figure 1. The feature of special
interest in this section is the presence of resin ducts in the inner region of the first
annual ring. Figure 4, plate 68, shows the first annual ring and the resin ducts more
highly magnified from another section, cut from a different twig of the same main
branch. As the material came from the Gray herbarium of Harvard university and
was conseqtiently in a dried condition, the cells surrounding the resin ducts show no
evidence of the presence of protoplasm and a nucleus; but the obviously resinous con-
tents of the ducts in the dry condition, as well as the parenchymatous nature of the cells
surrounding them, leave no doubt as to their identity. In the last mentioned figure,
the cells enclosing the lumen of the resin canals appear darker as to their walls than the
adjacent tracheids. This is due to the fact that they become stained very intensely with
haematoxylin, and thus are more non-actinic than the surrounding tracheids, counter-
stained less strongly with aqueous saffranin. The material presenting these peculiarities

in the first annual ring was taken from specimens accompanied by unusually large cones,

and labeled “J. G. Lemmon, California.” In order to test whether the structures in
question were not abnormalities, sections were made from two other lots of material
from the Gray herbarium, marked “Bridges 331” and “Henry 1086,” respectively.
Figure 5, plate 68, shows the central portion of a transverse section from the material
marked “ Bridges 331.” The resin ducts are obviously even better developed than they
are in figure 3. Sections from the other lot of material also showed the presence of
large and typical resin ducts in the first annual ring of the wood. The resin passages
just described disappear in the upper part of the annual ring as it reaches what was

~,

% Fl a

ee ee Pot ee ee ee

eee? Pet oie "os

|

ANATOMY AND PHYLOGENY OF THE CONIFERALES. 445

originally the end of the yearly growth in length of the branch. In no case which has
come under my observation did the resin ducts of the first year’s growth in one annual
segment of a branch become continuous with those in the next upper segment repre-
senting a subsequent year’s growth in length. As a consequence the resin ducts of each
first annual ring form a closed system and are confined entirely to their own segment of
the branch. There are, further, no lateral anastomoses of the ducts in the same annual
ring. In the weaker lateral twigs of the branches from the three lots of material
mentioned above, resin ducts were generally found to be entirely absent. Through the
kindness of Prof. C. V. Piper, I received formalose material of the branches of a young
tree twenty years old, grown at Pullman, in the state of Washington. Not even the
largest and most vigorous branches in this material showed any indication of the presence
of resin ducts. Similar observations were made in the case of alcoholic specimens
supplied to me by Dr. A. A. Lawson, of the Botanical department of Leland Stanford
university. It may consequently be stated that so far as the present observations go,
the presence of resin ducts in the first annual woody ring of S. gigantea is confined to
the more vigorous branches of older trees which have already produced female cones, and
that the phenomenon is normally absent in the smaller twigs of large trees as well as in
the entire branch system of young trees.

THe RepropuctivE Axis or S. gigantea.

In figure 6, plate 68, is seen the woody portion of the base of the female reproduc-
tive axis of S. gigantea in transverse section. In the first annual ring are numerous
large resin ducts similar to those described above in the case of the more vigorous vegeta-
tive shoots of adult trees. The second annual ring is very broad and is quite free from
resin passages. It is followed by a number of much narrower rings equally free from
resin canals. The peduncular portion of the reproductive axis of S. gigantea continues to
grow in thickness for some time after the seeds have been shed, and in this respect it
resembles certain species of Pinus with persistent cones, and at the same time presents a
marked contrast to S. sempervirens. Figure 7, plate 68, shows the center of the same
section more highly magnified. The size and distribution of the resin ducts and of the
resin cells also can be very well made out. Figure 8, plate 68, reproduces a cross section
of a higher region of the same peduncle. The narrow annual ring containing the resin
ducts seen in the last two figures, has disappeared, and the broad second ring of annual
growth now abuts directly on the pith. In figure 9, plate 69, there is represented an
intermediate region between those shown in figures 6, 7, and 8, of plate 68. Some of
the resin canals in this section are very much enlarged. This appearance is due to the

444 EDWARD C. JEFFREY ON

fact that the resin ducts end upwardly in resin pockets of considerable size. Figure 10,
plate 69, illustrates the structure of the reproductive axis proper as seen in cross section
at a height much above any of the foregoing. The resin ducts have reappeared in the
wood close to the medulla, which in the present figure presents an elliptical outline due
to the fact that two fructiferous scales of the cone derive their fibrovascular supply from
the central cylinder of the axis in this region. The resin canals are particularly well
developed around the bays corresponding to the departing traces of the fructiferous
scales. The figure includes only part of the first annual rig of wood. In the repro-
ductive axis of S. gigantea there are two annual rings of wood only, in the portion
corresponding to the cone proper, and as a consequence growth does not continue
after the second year when the seeds are ripe. The additional annual rings which are
characteristic of the peduncle of the cone gradually die away in the transitional region
between the peduncle and the axis proper. It is to be noted that there is no connection
between the resin ducts of the peduncle and those of the proper axis of the cone, a condi-
tion quite similar to that described above for the successive longitudinal annual segments
of the more vigorous vegetative branches of adult trees. I have not yet found any cone
of S. gigantea in which the resin canals were absent from the first year’s growth of the
wood.

In figure 11, plate 69, is seen a transverse section through the woody portion of the —
base of a cone scale, including all of the first annual ring of growth and a narrow zone
of the second. On the lower side of the first annual ring, resin ducts appear near the
medulla. These form the direct continuation of the resin ducts of the axis shown in —
figure 10, plate 69. In figure 12, plate 69, is represented a section through a somewhat
higher region of another fructiferous scale. In the center appears the medulla, above and

below it is the wood of the fibrovascular system of the scale. The most interesting ,
feature of this figure is the large number of resin canals present in the wood. Of these
there are two series: the one composed of larger and less numerous ducts nearer the
pith, and the other made up of smaller and more numerous ducts, farther out in the wood
of the scale. Both series of ducts tend to become divided by radial partitions, and the
two systems are not infrequently united by radial anastomoses as well, comparable to the
radial canals running in the medullary rays of most of the Abietineae. The last two.
figures present extreme conditions in the modes of occurrence of resin canals in the wood |
of the fructiferous scales. Where the ducts are least abundant they are confined to the
lower side of the scale as in figure 11, plate 69, and form a single series. They may also
extend as a single series to the upper side of the scale. When the system of ducts is
double, it may be double on the lower side of the scale only, or on one or other wing: of
the scale as well, or finally as in figure 12, plate 69, the double series may extend all

ANATOMY AND PHYLOGENY OF THE CONIFERALES. 445

around the medulla. Whenever a double system of ducts extends more or less com-
pletely around the wood of the scale, the series nearer the medulla is always complete,
whatever may be the case with the more external series. The more complex arrange-
ments of the resin canals just described are more often found in the upper scales of the
cone. The outer system of ducts, unlike the inner one, never extends down into the
axis of the cone, although it may pass far up in the bundles into which the fibrovascular
tube of the base of the scale breaks above.

In figure 13, plate 69, is seen a transverse section of the upper broad portion of the
fructiferous scale. The scale is covered on both the upper and the lower surfaces with a
layer of-periderm. In the fundamental tissue of the scale are numerous large resin
passages as well as numerous sclerenchymatous cells. There are present two systems of
fibrovascular bundles oriented in opposite directions, of which the upper series is less well
devéloped and consists of somewhat smaller bundles. Along the upper margin of the
lower set of bundles, which are the better developed, can be seen small, light dots which
mark the position of intrafascicular resin ducts. These are continuous with those
described above as occurring in the lower portion of the woody skeleton of the scale.
Resin canals are much more commonly present in the lower series of bundles of the
flattened upper part of the fructiferous scale than they are in the upper series, possibly
on account of the greater robustness of the former. Figure 14, plate 69, shows a
portion of the same section more highly magnified. The structure of the fundamental
tissue can be more clearly seen. It consists of parenchymatous elements, of long scler-
enchymatous elements present in cross section, and of large resin passages. Within the
inner boundary of the woody tissue of most of the lower bundles can be seen one or more
resin ducts. These are absent from the upper bundles. Figure 15, plate 69, represents
another part of the same section. The fundamental tissue presents no difference from
that shown in the preceding figure; but the bundles which belong to the lower series,
towards the margin of the scale, are obviously united by strands of tissue running
from their inner borders. The tissue in question is made up of transfusion cells. The
smaller terminal bundles of the extreme upper portion of the fructiferous scale of S.
gigantea are all more or less completely united at their inner borders by transfusion
cells. The bundles likewise frequently terminate in transfusion tissue as well. Occasion-
ally towards the lower part of the course of the bundles, as they become united into the
fibrovascular tube of the base of the scale, the enormously developed transfusional borders
of the bundles bend inwards and fuse more or less completely together, giving rise to a
peculiar pseudomesarch type of bundle such as is shown in figure 16, plate 69. In this
figure, two resin canals are to be seen in the secondary wood. Subtending the part of
the secondary wood containing the two resin ducts, is a mass of parenchyma which is

446 EDWARD C. JEFFREY ON

enclosed incompletely on the opposite side by wing-like extensions of the secondary
xylem. These are in turn completely covered by a thick zone of very dense transfusion
tissue. The phenomenon just described is not of very common occurrence and in all

probability no very great morphological importance is to be attached to it.

THE SEEDLING oF S. gigantea.

Figure 17, plate 70, reproduces a transverse section of the woody portion of a five
8 2 >, FE} yP

year old seedling of S. gigantea, collected by Prof. Asa Gray in the famous Calaveras
grove of Big Trees, during his visit to California in 1872. I owe this duplicate, specimen
to the kindness of my colleague, Prof. B. L. Robinson, curator of the Gray herbarium.

y Sue, ? ,
Four annual rings are shown in the figure. The resin cells which are so conspicuous in
the wood of the adult stem are much less strikingly present in the seedling, and resin
ae ?

ducts are entirely absent even from the first annual ring. Through the kindness of Dr.
Joseph L. Goodale, I have had the opportunity of examining several other seedlings
of S. gigantea, grown on his estate at Ipswich. In none of these did I find any indi-

cation of the presence of resin canals in even the first annual ring of woody growth.

Errects oF WouNnpDs ON THE Woopy TissuEs oF S. gigantea.

Figure 18, plate 70, represents part of a transverse section of a wounded root of S.
gigantea, in which the wounded area has become completely callused over. Towards the
right in the figure, may be seen a mass of callus tissue. In the ring of growth just below
the callus, is to be made out a long row of traumatic resin ducts stretching completely
across the figure. The annual ring containing the resin canals, although axial to the
callus, was formed subsequently to the wounding of the tissues, as was made out by
examining the whole line of injury. The resin ducts are quite similar to those occurring
normally in many of the Abietineae, and are generally surrounded by but a single layer
of cells. Of these cells some are normal glandular cells, such as usually surround the
lumen of resin canals where they occur in the wood of the Abietineae, while others are

‘““resin cells,”

and are distinguished by their thick and strikingly pitted walls. In figure
19, plate 70, is represented part of the outer margin of the wood in a transverse section
of the peduncular portion of the cone of S. gigantea. Very large resin canals appear
along the inner border of the vernal wood in the last annual rmg. These are surrounded
by secretory cells; but as the figure was made from dry material the latter have almost }
completely collapsed. The formation of true resin canals in this case was due to the

same cause as in the root just described above. I have seen a large number of instances

Pe

ANATOMY AND PHYLOGENY OF THE CONIFERALES. 447

of the formation of traumatic resin canals in the wood of both the root and the stem of
S. gigantea and consequently am of the opinion that it is quite a usual result of injury.
It should be added, however, that the formation of parenchymatous wood containing a
large number of resin cells and ordinary parenchyma cells is even more commonly
present in the case of wounds. This is especially the case with large exposed wounds on
the aerial portions of the stem, where there is much necrosis and drying out of the woody
tissue. In smaller wounds, however, where there is not too great an exposure of the
ligneous cylinder to desiccation, the formation of the traumatic resin passages described
above is very common. It should be noted finally that the traumatic resin canals of S.
gigantea are formed in the vernal wood and in this respect are in contrast to those occur-
ring in S. sempervirens under similar conditions, which, as will be shown later, make their
appearance in the autumnal wood. I have found examples of the formation of resin
canals in the phloem of S. gigantea, as well as in the wood, and as the result of a similar
cause. I have discovered no instance of the occurrence of resin ducts in the fibrovascular
cylinder of this species, except in the first year’s growth of vigorous young branches of
adult trees and in the axis of female cones (as described in earlier paragraphs), which
could not be clearly traced to a previous injury. It seems extremely probable that the
formation of traumatic resin canals in S. gigantea is to be regarded as a case of reversion
to an ancestral condition, especially in view of the mode of occurrence of normal resin
canals in certain instances as described above. The arguments in favor of this view, how-
ever, are best deferred to a later stage.

TRAUMATIC RESIN CANALS OF THE Abietineae.

It seems necessary in the meantime to refer preliminarily to certain other facts of a
similar nature which have been made out in the case of the Abietineae, in a series of
researches of which the present investigation forms a part. It has long been known that
resin ducts are generally absent in the wood of the Abietineous genera, Abies, Tsuga, and
Cedrus, and that they are replaced in many instances by resin cells. For the present the
first-named genus will alone be considered; for as I hope to show subsequently, the other
two genera resemble it very closely in the features which are of interest in this con-
nection. Abies firma is described by Prantl (Engler u. Prantl, Nat. pflanzenfamilien,
Coniferae, 1889, p. 37) as differing from the other species of Abies in normally possessing
resin ducts in its wood. Penhallow (op. cit.) has more recently described the presence
of resin canals, occurring in isolated annual rings of Abies nobilis and Abies bracteata.
I have been able to confirm these observations in the beautiful series of sections of North
American woods published a few years ago by Professor Penhallow. The resin ducts

448 EDWARD C. JEFFREY ON

occur in long tangential rows in both the last-named species. I have myself examined
a number of species of the genus Abies, and have found resin canals of very frequent
occurrence under conditions to be more fully described on another occasion. For the
present, however, it may be stated that resin ducts are extremely apt to be present in
the female reproductive axis of various species of Abies, even when they are quite absent
from the woody tissues of the ordinary vegetative stem. For example, in Abies grandis,
according to Penhallow, there are no resin ducts present in the wood. Figure 20, plate
70, shows the general structure of the woody axis of the upper portion of the female cone
of A. grandis. The specimen from which the figure was made, was secured at the Gray
herbarium, and as a consequence the tissues have suffered a good deal from long desicca-
tion. Nevertheless, indubitable resin ducts can be clearly made out in large numbers in
practically the whole circumference of the woody cylinder of the cone. They are of quite
the normal type and are surrounded by glandular cells which still retain the remains of
their protoplasm. The section through the small upper end of the cone axis was chosen
because it permitted of showing the structure of the whole of the ligneous portion of the
tissues of the cone on a sufficient scale of magnification to make clear the presence and
mode of distribution of the resin canals. In figure 21, plate 70, is seen a similar view of
the woody tissues in the cone of Abies balsamea. This species is devoid of resin ducts
not only in the wood of the vegetative axis but also in that of the cone. The figure in
this case was made from material which had been properly preserved. Figure 22, plate
70, is made from a section through an injured root of a small tree of Abies balsamea
growing in the Botanic garden of Harvard university. The figure includes all the xylem

as well as a portion of the phloem of the injured root. There is an interruption of con-
tinuity in the latter tissue on the lower side, due to an incised wound which was made in
all probability by a spade used in digging the soil about the tree. The woody tissue ia%
likewise interrupted in the same region as the phloem, and on one side the last annual
ring is separated for some distance from the tissues below. There is a considerable forma-
tion of resinous tissue along the inner margin of the last annual ring on both sides of the —
wound, which takes the form of resin cells. Farther away from the wounded portion, on —
the left, the resinous tissue gives place to a tangential row of resin ducts continuous —
through almost half the circumference of the root, and passing on the right into a narrow
zone of resin cells, which are in turn continuous with the thick mass of resinous tissue
on the immediate border of the wound. The center of the root is occupied by a resin
duct surrounded by resinous cells, such as is commonly found in the axial wood of the
root in the genus Abies (de Bary, Comp. anat., 1884). My investigations on this genus,
although not yet complete, show that resin canals occur very frequently in the female
reproductive axis, even when they are quite absent from the wood of the vegetative stem.

ARTE AE RE yg! a SR Ty <t

TR eat = ™

ANATOMY AND PHYLOGENY OF THE CONIFERALES. 449

and from the root as well, except in the axial primary wood as just mentioned above.
In all cases where I have been able to make the experiment, however, I have succeeded
in bringing about the formation of resin canals as a result of injury, in the wood of
species of Abies which are normally without them, even in their reproductive axis. It
is interesting in this connection to note that certain fungous diseases causing injury to
the wood of species of Abies produce the same result. It seems necessary to record the
traumatic reactions of the woody tissues of the genus Abies because, as I shall attempt to
show at the end of this essay, they supply an interpretation of the similar phenomena
in the case of Sequoia.

Tne Lear or S. gigantea.

In figure 23, plate 70, is seen part of a section through the base of the leaf of S.
gigantea. In the center of the figure lies the leaf trace flanked by the two lateral wings
of transfusion tissue, which are so characteristic of the lower part of the foliar bundle
in Sequoia and its allies. Above and below the leaf trace are masses of collenchyma-
toid tissue. The feature of greatest interest in the figure is, that contrary to other
described coniferous leaf traces, the foliar bundles of S. gigantea contain a resin duct.
The duct is quite of the normal type and is surrounded by a single, almost complete row
of resiniparous cells. In some cases, however, as is often found in the resin canals of the
Sequoias, the parenchymatous lining of the canal is far from being continuous, and the
tracheids as a result often abut directly on its lumen. It must not be supposed that resin
ducts occur in all the leaf traces of S. gigantea, for this is not the case. They appear
only to be present in the bundles of the very large leaves of exceptionally vigorous
branches of mature trees (7. e., trees which have already ripened seed). All the material
which I have, showing this feature, came from the Gray herbarium of Harvard univer-
sity; but in spite of the necessarily bad condition of preservation consequent on its
origin, there seems to be no reasonable doubt as to the nature of the canals, which I have
described above as resin ducts. The longitudinal range of the resin ducts of the foliar
bundles of the largest leaves in mature trees of Sequoia gigantea is quite limited; for
they appear only after the leaf trace has passed quite out into the cortex of the branch,
and indeed after it has traversed a considerable part of its upward and outward course.
There is as a consequence no communication between the resin ducts of the first annual
ring of wood in the branches and those appearing in the leaf traces. Like the other
modes of occurrence of resin canals described above, there is an entire absence of correla-
tion with other similar tissues. The resin ducts of the leaves disappear again very shortly

450 EDWARD C. JEFFREY ON

after the leaves have emerged from the surface of the branches, so, as has just been stated
above, the longitudinal course of the resin canals of the leaf traces, where they occur, is

very short indeed.

THE STEM OF Sequoia sempervirens.

Penhallow (op. cit.) has noticed the occurrence of continuous tangential rows of
what he terms “imperfect resin ducts” in the wood of the old stem of S. sempervirens.
The phenomenon seems to be extremely rare, for | have found very few examples
although a large number of different specimens of the wood of this species has been
examined in this connection. Professor Penhallow has been so kind as to send me a
very good example illustrating this peculiarity, and as it is much more striking than any
other I have seen, all the accompanying figures of the mature wood of S. sempervirens
are made from this material. In figure 24, plate 70, a part of a transverse section of the
wood of S. sempervirens is represented under a low magnification. There are four
annual rings present in the figure. Mingled with the tracheids in all the four woody
rings are numerous resin cells, such as have been described above in S. gigantea, and
such as occur commonly in the other Taxodineae. The autumnal wood in S. semper-
virens forms a much thicker zone than in S. gigantea, which is the cause of the greater
weight and strength of the wood in the former species. In the second annual ring from
the top of the figure are to be seen the resin ducts first described for S. sempervirens by
Penhallow (op. cit., p. 39, pl. 6, fig. 2). They are obviously situated in the autumnal
wood and in this respect present a marked contrast to those of S. gigantea described
in the foregoing paragraphs, for in the latter species they occur in the vernal portion of
the annual rmg. In their distribution the resin canals in the present instance resemble
those formed as the result of injury in S. gigantea, species of Abies, ete., and form a
continuous tangential zone, such as is characteristic of traumatic resin canals. Figure
25, plate 71, shows a portion of the annual ring containing resin ducts more highly
magnified. There is obviously a sharp transition from the autumn wood to that of the
following spring in the size of the tracheids and the thickness of their walls. There are
numerous resin cells on the outer margin of the autumnal wood. Farther inwards, the
autumnal tracheids give place to parenchyma cells, which, on account of the fact that
the material is from the heart wood of an old tree, are quite devoid of protoplasmic
contents. They are unmistakable, however, since their walls are composed of cellulose
and stain a deep blue with haematoxylin. About the lumina of the resin ducts, the
parenchymatous cells are as a rule strongly pitted and closely resemble the resin cells
of the medullary rays in certain Conifers.

Er ye os a

Oe! «

Qtech eT ee a

ANATOMY AND PHYLOGENY OF THE CONIFERALES. 45]

In figure 26, plate 71, is shown a transverse section of a young branch of S. semper-
virens from material sent to me by Dr. A. A. Lawson of the Botanical department of
Leland Stanford university. The bases of four leaves clearly appear on the margin of the
figure. These originate from the branch by a flat base in contrast to the rounded mode
of origin found in S. gigantea. In two of the leaves there are obviously three resin
canals in the mesophyll, another point of contrast to S. gigantea and most of the other
Taxodineae and Cupressineae, where there is as a rule but a single resin duct in the
parenchyma of the leaf, and immediately under the fibrovascular bundle. The material
from which the illustration has been made was cut and preserved in California on the
2d of May. The growth of the present year’s ring of wood has begun and has already
reached a considerable thickness. Branches of S. gigantea cut from trees at the same
spot and at the same time show scarcely any signs of the commencement of cambial
activity. The slowness of initiation of cambial activity in the Big Tree is no doubt one,
at least, of the causes of its greater hardiness. There are no resin duets to be seen in
the wood of the young branch, appearing in transverse section in the figure, and in this
respect the illustration might stand for any branch of S. sempervirens; for I have not
found even in the most robust ramifications of the mature trees of this species any indica-
tion of the resin canals, which are so often present in the first annual woody ring of S.
gigantea.

Traumatic Resin Ducts 1n S. sempervirens.

In some of the specimens sent me by Dr. Lawson from the arboretum of Leland
Stanford university the phenomenon of fasciation was present. As this feature of S.
sempervirens has recently been fully described by Pierce (Studies on the Coast Redwood,
Cal. acad. sci., 1901), it will not be necessary to refer to it here. The wood in most of
the fasciated specimens was quite normal as regards the absence of resin canals; but
in the subterranean portion of a dead branch which had taken on the flattened mode
of growth so characteristic of fasciation, numerous resin ducts were found in tangential
rows. The shoot in question had obviously been injured and the appearance of resin
canals is doubtless to be correlated with the diseased condition of the tissues. Figure
27, plate 71, is made from a section through this branch. There are two annual rings
present and it is in the first of these that the resin ducts are to be seen. The ducts are
in tangential rows and are situated in the autumnal wood just as in figure 25, plate 71.
In figure 28, plate 71, are represented some resin canals from another section of the same
material, highly magnified. The resin ducts are obviously surrounded for the most part
by cells still containing protoplasm and a nucleus, although in some cases the tracheids

452 EDWARD C. JEFFREY ON

abut directly on the lumen of the resin canal. The structure of the ducts shown in the
figure is quite typical of the resin passages I have found in the Sequoias in general.
It is a fact of considerable interest in the present connection, that the resin ducts first
described by Penhallow for the wood of the old stem of S. sempervirens occur in the
same tangential rows which are found in the obviously injured stem of the species under
discussion, and under similar circumstances in the Abietineae as well. It is often the case
that traumatic resin canals extend far above and below the actual spot of injury and it
accordingly seems extremely probable that the rows of imperfect resin canals found in
the otherwise quite normal fragments of wood of S. sempervirens are really due to an

injury above or below the region from which the pieces of wood have been taken.

THE RepropuctTivE Axis or S. sempervirens.

In figure 29, plate am is represented the fibrovascular cylinder of the lower third of
the reproductive axis of S. sempervirens. Although the cone from which the section was
made had long shed its seeds, and had assumed an almost black hue from long exposure
to the weather, there are obviously no indications of annual rings in the photograph.
While the absence of annual rings does not absolutely prove that there are no additions
to the secondary wood of the reproductive axis after the first year, still it appears very
probable from the state of affairs in the cones of the other living species of Sequoia that
such is the case. This view of the matter is further strengthened by the occurrence of
perfectly normal annual rings in the vegetative organs of S. sempervirens. The question
can, however, only be finally settled by the study of the growth of cones in the field.
There are no annual rings in the peduncular portion of the cone of the species under —

consideration and in this respect it also presents a contrast to S. gigantea, where, as has :
been described above, there are numerous annual rings in that region of the reproductive
axis. Further, resin ducts are entirely absent from all parts of the wood of the cone of S.
sempervirens. Figure 30, plate 71, shows the structure of a cross section through the
base of a cone scale of this species. The dense wood of the tubular central cylinder of —
this portion of the fructiferous scale shows no evidence of the presence either of the
‘ annual rings or of resin ducts, which are such notable features of the cone scales of S.
gigantea. ‘The specimen shown in figure 30 is quite typical in these respects; for I have
examined a number of cones from different sources, without findmg any examples of the
occurrence of either of these structures. In figure 31, plate 71, appears a portion of a
transverse section of the upper third of the cone scale of S. sempervirens. The large
bundles along the lower side of the figure belong to the lower series. They are very
much larger than those near the upper surface of the scale and are inversely oriented.

ANATOMY AND PHYLOGENY OF THE CONIFERALES. 4538

In the species under consideration there is a very much greater relative reduction of the
upper series of bundles than is the case in S. gigantea. The commissures of transfusion
tissue, which are so marked, especially in the lower bundles of the cone scales of the
latter species, are much less well developed in the Redwood. The fibrovascular bundles
of the fructiferous scales of S. sempervirens show no annual increments of growth and
never contain resin ducts. :

Tuk Roor or THE SEQUOIAS.

Figure 32, plate 71, illustrates the structure of a small root in S. gigantea. It is
obviously pentarchous in its organization and very much resembles a root of Pins
strobus or some other Abietineous species with large polyarch roots. The protoxylem
groups, however, which occupy the apices of the five rays of the primary wood, differ
from those of species of Pinus, efc., in not embracing resin ducts. Although the roots of
the species under consideration are generally pentarchous, examples are not infrequently
found in which there are only four rays of primary wood. ‘This state of affairs also occurs
in Pinus strobus. The primary root of all seedlings of S. gigantea which I have exam-
ined is of the tetrarchous type. I have never found resin canals present in the secondary
wood of this species except as the result of injury.

The root of S. sempervirens is so similar to that of S. gigantea as scarcely to merit a
separate description. The symmetry of the root is, however, almost always tetrarchous
and the various histological elements are somewhat larger than in the latter species.

CONCLUSIONS.

In the foregoing paragraphs, attention has been called to the occurrence of resin
duets, under certain definite conditions, in the wood of Sequoia. In S. gigantea, which in
all probability is the more primitive of the two surviving species, resin canals are con-
stantly present in the wood of the peduncle, of the axis, and of the scales of the female
cone. They also frequently occur in the first annual ring of the more vigorous branches
of adult trees. Resin ducts likewise are sometimes found in the leaf traces of the larger
leaves of well nourished, mature individuals. Finally, it is possible to bring about the
formation of resin canals in tangential rows as the result of injury to the woody cylinder
of the root stem and cone of S. gigantea. The resin ducts occurring in the various situ-
ations cited above, form entirely separate systems which do not communicate in any way
with each other. In S. sempervirens, in contrast to the species just referred to, there are

no resin canals in the woody tissues either of the reproductive organs, or of young

454 EDWARD C. JEFFREY ON

branches, or even of vigorous leaf traces. In S. sempervirens, resin ducts appear only as
tangential rows in isolated annual rings, probably in all cases in response to injury to the
tissues of the wood.

The occurrence of resin canals in the reproductive axis of 8. gigantea is in all proba-
bility to be regarded as an ancestral feature. It is significant in this connection that
certain species of Abies, which have no resin ducts in the wood of their vegetative organs,
retain them in the ligneous cylinder of their cones, e. g., A. grandis, A. nobilis, etc., ete.
In the case of these species the presence of resin ducts as a general feature of structure
of the wood in the order to which they belong, justifies the conclusion that the occurrence
of resin canals in the wood of the female reproductive axis is an ancestral character, which
has disappeared in the wood of the vegetative stem. A parallel case is presented by
the peculiar mesarch bundles found by Dr. Scott in the peduncle of the cones in certain
Cycads (The anatomical characters presented by the peduncle of Cycadaceae, Ann. bot.,
1897, vol. 11). Bundles of this nature were present in the medullary crown of the vege-
tative axis of certain of the primitive fossil Gymnosperms and of the Cycadofilices. In
the living Cycads, mesarch bundles have disappeared in the vegetative stem, but are some-
times retained in the reproductive axis and quite generally in the vegetative and repro-
ductive leaves. In S. gigantea, resin ducts retained in the wood of parts of the female
cones, likewise occasionally occur in the leaf traces as well, thus presenting a striking
parallel to the mesarch bundles of the Cycads, in the manner of their distribution. The
presence of resin canals under certain quite definite conditions in the first annual ring of
the branches of S. gigantea, supplies a further argument for regarding the occurrence of
resin ducts in the wood of this species as the retention of an ancestral feature, for it is
only natural to find such a character reappearing in the first zone of woody growth of
the young branch. This conclusion is also confirmed by the presence of a similar feature
in the young branches of certain Abietineous species to be subsequently described.

The phenomena of injury furnish additional evidence in favor of the view that the
ancestral stock from which the Sequoias have been derived, was characterized by the
presence of ligneous resin ducts, for it appears justifiable to regard the formation of
tangential rows of traumatic resin canals in the injured wood of both living species of
Sequoia as a reversion to an ancestral condition, especially in view of the phenomena
described in the last paragraph. This explanation of the phenomena resulting from —
injuries to the woody tissues in the species under discussion is confirmed by the fact —
that similar traumatic resin ducts are commonly formed as the result of injury to the
wood in the Abietineae, in contrast to the Cupressineae, ete.

The hypothesis that the Sequoias have come from ancestors characterized by the
presence of resin ducts in their woody tissues, receives a further support from the con-_

r= ~~ ew

en

GE EDI A ee pcr RC, CODY te

ANATOMY AND PHYLOGENY OF THE CONIFERALES. 455

sideration of the anatomical structure of certain fossil species of Sequoia. Penhallow has
described the presence of resin canals in the wood of S. langsdorfii (Notes on the
Cretaceous and Tertiary plants of Canada, Proc. and trans. roy. soc. Canada, 1902, series
2, vol. 2, p. 44-45). The wood of this species, known heretofore only by its foliage and
its fruit, is described as having occasionally imperfect resin ducts in its autumnal wood.
Unfortunately the author does not figure nor further describe the distribution of the resin
ducts in question, so it is not possible to have an opinion as to whether they were of
traumatic or of normal origin. The resin canals in the case of S. langsdorfii had a longi-
tudinal course in the wood. Quite recently, Professor Penhallow has announced the
occurrence of radial resin canals in the wood of another species of Sequoia; but as yet
no published account has appeared.

It seems highly probable, if the various arguments advanced above are valid, that the
Sequoias have come from a stock characterized by wood penetrated by resin canals. If
this probability be granted, the question immediately arises, whether they have come off
from the same stock as the Abietineae, which have as a striking anatomical peculiarity the
presence of resin canals in their ligneous tissues. At first sight this would appear to be
an unlikely hypothesis; for the Abietineae are generally regarded (cf. Potonie, Pflanzen-
palaeontologie, 1899, p. 522; Coulter and Chamberlain, Morphology of Spermatophytes,
1901, p. 108; Strasburger, Coniferen und Gnetaceen, 1872, p. 265; etc.) as a very modern
order of the Coniferales. The reasons for this view are, however, not entirely apparent,
for the genera Pinus and Cedrus appear in the infra-Cretaceous, and are thus as far as
the palaeontological record at present goes, synchronous in their first appearance with
Sequoia (R. Zeiller, Palaeobotanique, 1900, pp. 271 and 277). Further, so competent
a judge as Dr. Scott states that “on the whole, it is impossible in the present state of
our knowledge, to say which tribe or family of the Coniferae is the most ancient”
(Studies in fossil botany, 1900, p. 483).

Much importance is justly attached to the structure of the female cone in the classi-
fication of the Coniferales, and any hypothesis of the phylogeny of the group must
accordingly harmonize with the facts gathered from this source, when properly inter-
preted. In the Abietineae, the female cone is made up of double superposed scales.
The most generally accepted explanation of the morphological nature of each pair of
superposed scales is that the upper ovule-bearing scale represents the fusion of a pair of
leaves, belonging to a reduced axis, axillary to the lower subtending bract (for an admira-
ble account of this difficult question, see Worsdell, Structure of the female “flower” in the
Coniferae, Ann. bot., 1900, vol. 14). This hypothesis finds a strong support from the
standpoint of comparative anatomy, and as has recently been aptly stated by Professor
Coulter, the clear and parallel case of reduction of an axillary (vegetative) shoot in

456 EDWARD C. JEFFREY ON

Sciadopitys is almost a demonstration of its accuracy. The hypothesis which explains the
ovuliferous scale of the Abietineae as a reduced and modified short shoot (brachyblast) , is
likewise in harmony with teratological evidence, for im the so called proliferous cones of
certain Abietineae, the ovuliterous scale becomes more or less completely transformed into
a leafy bud and the subtending bract into an ordinary vegetative leaf. It is a curious and
interesting fact, which does not seem to have been properly considered in regard to its
bearing on the phylogeny of the Coniferales, that cases of proliferous female cones are
confined, in the present state of our knowledge, to the Abietineae and the Taxodineae
(Penzig, Pflanzenteratologie, 1894, vol. 2, p. 485-514). To those who accept the brachy-
blastic theory of the nature of the ovuliferous apparatus in the Coniferales, this state of
affairs must appear weighty evidence in favor of the view that the Abietineae, and the
Taxodineae as well, are somewhat primitive orders of the group, for there would naturally
be the greatest tendency to reversion and the clearest anatomical evidence of the shoot
value of the ovuliferous scale in the orders which are nearest to the ancestral stock.

In the case of the Taxodineae, the scales of the female cone are not superposed in
pairs but consist of single, generally more or less thickened ovule-bearing organs, in
which there is present a double system of bundles consisting of an upper and a lower
series oriented in opposite directions, as are those of the separate superposed scales of the
Abietineae. The generally accepted description of the state of affairs in the cone scales
of the Vaxodineae, is that there is a fusion of the ovuliferous and sterile bracts. It is
scarcely logical to speak of organs as being “fused” unless they were originally sepa-
rate. The view that they were primitively separate, superposed: scales in the ancestral
stock of the Taxodineae is strengthened by a consideration of the anatomical facts in
the case of the genus Sequoia; for it is not easy to conceive that the series of massive
bundles near the upper surface of the cone scale in this genus is arising de novo for the
benefit of the reproductive organs, since the ovular bundles are extremely small and
quite out of all proportion to the fibrovascular strands of the upper system from which
they take their origin. The more reasonable explanation seems to be that the thick,
single scale which bears the ovules in the case of most of the Taxodineae, is made up of
a fusion of two separate scales which existed in a more primitive group, probably the
Abietineae or their parent stock. This view of the matter receives further support from
the peculiar manner of the occurrence of resin ducts in the woody tissues of living and

-

fossil Sequoias.

It would be anticipating unduly the results to be recorded: in subsequent memoirs on

the Coniferales, to express more than a provisional opinion as to the position of the genus

*

Sequoia but this much may be safely stated. The anatomy of living and fossil species of

Sequoia goes to show that the genus has come from a stock possessing ligneous resin —

ANATOMY AND PHYLOGENY OF THE CONIFERALES. 457

ducts, such as prevail in the Abietineae of the present day. It may be urged that this in
itself is a very small point. Reply may be made that the mesarch structure of the pri-
mary woody bundles of the more primitive Gymnosperms and the adhesion of the stamens
to the calyx in the Rosaceae are likewise small points, yet no well informed botanist
doubts that they are important taxonomic criteria. It appears to be rather the signifi-
cance than the magnitude of anatomical or morphological features which makes them of
classificatory value. The peculiar and apparently most significant mode of occurrence of
resin canals in the wood of the Sequoias points to their derivation from an Abietineous
stock, which may of course have been ancestral not only to the Sequoias but also to the
living genera of the Abietineae as well. This conclusion appears to be supported by a
consideration of the structure of the female cone in the Abietineae, Taxodineae, and the
Coniferales in general. Finally the derivation of the Sequoias from an Abietineous source
is in no way contradictory to the palaeontological record.

SUMMARY.

1. Typical resin ducts occur in the wood of the peduncle, axis, and scales of the
female cone of Sequoia gigantea.
‘ 2. Resin ducts are likewise present in the first annual ring of vigorous branches of
adult trees of the same species. Resin ducts are normally absent in all the branches of
immature trees.

3. Resin ducts are also found in the leaf traces of very vigorous leaves of adult
trees of Sequoia gigantea.

4. Resin ducts are entirely absent from the wood of all parts of the cones of Sequoia
sempervirens.

5. The same statement holds true of the branches and leaves of this species.

6. Resin ducts appear in tangential rows in the wood of root and shoot in both
Sequoia gigantea and Sequoia sempervirens as a result of injury to the tissues.

7. A consideration of the mode of occurrence of resin ducts in the Sequoias leads to
- the conclusion that they are an ancestral feature of structure in the wood.
8. The anatomy of the vegetative organs and floral organs of the living Sequoias
_ points strongly to their derivation from an Abietineous stock. This conclusion receives
confirmation from what is known of the anatomy of the fossils, and finally is not in
contradiction to the palaeontological record.
; For help im securing the material for the present investigation, I wish to offer

my best thanks to Miss Alice’ Eastwood, Sir Wiliam Thiselton-Dyer, Dr. A. A. Lawson,

458

EDWARD C. JEFFREY ON

Dr. D. H. Scott, Prof. D. P. Penhallow, Prof. C. V. Piper, Professors Farlow, Goodale,
Robinson, and Thaxter, and last but not least, to Mr. George 8. Shaw, chairman of the

visiting committee on botany.

BoranicaL Museum or HArvArpD UNIVERSITY,

Fig.
Fig.
Fig.
Fig.
Fig.
Fig.
Fig.
Fig.

Fig.
Fig.
Fig.
Fig.

Fig.

Fig.

Fig.

Fig.

Fig.
Fig.

ducts in a tangential row. x 40.

Fig.

Fig.

presence of resin canals. XX 40.

iT,
2.
3.
4.
5
6
ip
8

Life
18.

19.
20.

Fig. 21.

x 36.

Cambridge, Mass.

EXPLANATION OF PLATES.

All the figures are from photomicrographs.
PLATE 68.

Transverse section of the wood of Sequoia gigantea. x 50.

Transverse section of a branch of a young tree of S. gigantea. x 60.

Transverse section of a vigorous branch of an adult tree of S. gigantea. x 40.

Transverse section of the medullary region of a similar branch. x 100.

Medullary region of a similar branch from another adult tree of S. gigantea. X 50.
Transverse section of the woody portion of the peduncle of an old cone of S. gigantea. X 30.
Transverse section of the central portion of the same. x 50.

Higher region of the peduncle of the same cone. X 30.

PLATE 69. \

Transverse section of the medullary region between those shown in figures 6 and 8. x 50.
Transverse section through the axis of the cone of S. gigantea. X 40.

Transverse section of the woody axis of the base of a cone scale in 8. gigantea. 50.
Transverse section of the woody axis of another cone scale of S. gigantea. X 50. ;
Transverse section through the upper portion of a cone scale of S. gigantea. X 3.
Transverse section of part of the same. > 450.

Transverse section of another part of the same. X 50.

Transverse section through an abnormal bundle of the cone scale of S. gigantea. 60.

PLATE 70.

Transverse section of the woody part of the axis in a five year old seedling of S. gigantea. X 40.
Transverse section through part of an injured root of S. gigantea, showing the formation of traumatic res’
Transverse section through part of the peduncle of a wounded cone of S. gigantea. x 50.

Transverse section of the upper portion of the woody axis of a female cone of Abies grandis, showing

Transverse section through the woody axis of the cone of Abies balsamea, showing the absence of resin duct

ANATOMY AND PHYLOGENY OF THE CONIFERALES. 459

Fig. 22. Transverse section through the central portion of a wounded root of Abies balsamea, showing the presence of
a tangential row of traumatic resin ducts. X 32.

Fig. 23, ‘Transverse section through a leat trace of a vigorous leat of a mature tree of S. gigantea. X 200.

Fig. 24. Transverse section of the wood of S. sempervirens, showing the presence of a tangential row of resin ducts.
x 35.

PLATE 71.

Fig. 25. Part of the last section more highly magnified, showing the structure of the wood and the resin ducts.
x 100.

Fig. 26. Transverse section of a young branch from an immature tree of S. sempervirens. X 30.

Fig. 27. Transverse section of part of an injured stem of S. sempervirens, showing the presence of tangential rows of
‘traumatic resin ducts in the autumnal wood. x 50.

Fig. 28. Transverse section of another similar specimen, showing details of the structure of the traumatic resin ducts.
x 200.

Fig. 29. Transverse section through the woody axis of the cone of S. sempervirens. X 35.

Fig. 30. Transverse section through the woody portion of the base of the cone scale of the same species. X 50.

Fig. 31. Transverse section through the upper portion of the cone scale of S. sempervirens. X 40.

Fig. 32. Transverse section through the root of S. gigantea. x 40.

Printed, November, 1903.

T.S0C.NAT.Hist. Vou.5

or
2)

Bos

MOIRS

=}
4

ME

ate ee

Page ae

LS oS tee, ae

fn
ry

MEM

}

-

“

MEMOIRS

Bost. S¢

7

JEFFREY ON SE

>
&

IU

“

a

LLL sso’
~~

eA COOHICD

NOOOITTIY YP)

ro)
‘7a

106°

i,

4
“a
%

.