The Distribution and Possible Importance of a Woody
Tumor on Trees of the White Spruce, Picea glauca*
PHILIPR. WHITEANDWILLIAMF.
(Potcoe B. Jackson
Memorial Laboratory, Bar Harbor, Maine)
INTRODUCTION
Massive excresences or "burls" are not uncom
mon on individual specimens of nearly all species
of trees. Some, the Circassian walnut burl for ex
ample, are well recognized sources of fine cabinet
woods. Their causes are multiple and may or may
not be apparent. They are curiosities of no great
scientific importance.
When, however, a malady, and especially one of
tumorous character, occurs regularly within nar
row environmental limits and under definable con
ditions, it becomes of more than passing interest.
Clear limits imply a limited causation, capable of
analysis. Such was, for example, the well defined
syndrome of cancer of the scrotum prevalent
among the chimney sweeps of Britain a century
and a half ago (15). This disease was ultimately
shown to result from prolonged contact with chim
ney soot. Out of a careful study of the hydrocar
bons found in such soots Kennaway and his col
leagues laid the foundation for the very important
field of chemical carcinogenesis (6). It is with such
a tumor, occurring under sharply limited condi
tions, that this paper will deal. This is a tumor on
trees of the White Spruce, Picea glauca. Its inter
est lies in the fact that it occurs under conditions
which are clearly enough defined to suggest that a
single explanation can be given for the cause of all
examples and in great enough numbers to furnish
material for extensive and intensive study. In re
cent years, the basic similarity between some tu
mors of plants and malignant growths in animals
has been well established, leading us to believe
that a study of an epiphytotic tumor may furnish
valuable information in the elucidation of the gen
eral problem of tumorigenesis and of cancer.
A tumor affecting a perennial woody plant, a
tree, may be of great interest for another reason.
One of the great difficulties in tracing the origins
and identifying the early stages of animal tumors
lies in the fact that the animal body is in a con
stant state of flux. With the exception of the
nerves and muscle cells, few individual cells remain
long unaltered. Either they move from place to
place, as do the connective tissue cells, or they are
sloughed off and replaced, as is the case with many
epitheliums. Even the trabecular patterns in bone
are constantly shifting. It is therefore for the most
part impossible to trace from its histology the de
tailed history of an animal tissue. Only the hair,
teeth, and horns of mammals, the scales of fish, the
carapaces of tortoises and the shells of mollusks
preserve the details of their history, and none of
these is subject to neoplastic disease. When a
"tumor" appears spontaneously in the soft parts of
an animal we can only say that its inception oc
curred in an approximate location, some time with
in a matter of months or even years. Even in ex
perimentally induced tumors we cannot define
with cellular precision the location of inception,
nor can we state except in general terms its date.
The wood of a tree, on the other hand, preserves in
its anatomy a detailed and legible record of events,
a record so clear that tree ring studies, in the hands
of Douglass (7,9) and others, have been used in de
scribing weather conditions and dating historical
events over a period of nearly 3,000 years. While
this "tree ring" record is relatively macroscopic,
the same precise record exists at a cellular level so
that it is possible in a tree to trace a pathological
event back to its inception 20 or 50 or 100 years.
When that event has resulted in a tumor, it is pos
sible to trace exactly when and where the tumor
was initiated, to locate the single cell which has
first shown evidence of neoplastic change (see 21),
to draw relatively precise inferences as to how such
initiation may have taken place, and to formulate
* Work supported by grants from the American Cancer
methods of possible experimental testing of the
Society, on recommendation by the Committee on Growth of correctness of our inferences. These are great ad
the National Research Council.
vantages. They are advantages which certainly
t Present address: University of Wisconsin, Madison, Wis.
justify the study of a tumor of the type to be con
Received for publication October 8, 1958.
sidered here.
128
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1954 American Association for Cancer Research.
WHITE ANDMILLINGTON—A
Woody Tumor in While Spruce
129
In 1929, while working at the Mount Desert
Island Biological Laboratory, Salisbury Cove,
Maine, the senior author (P. R. W.) noted on
Otter Cliffs, on the eastern side of Mt. Desert Is
land, a great many spruce trees affected with
burls. Such trees were observed near the shore, but
burls were not seen inland, although the same spe
cies of spruce occurred throughout the Island. The
growths were certainly neither "occasional" nor
tion. Although sometimes associated with broken
branches, these associations are too sporadic to
appear to have any significance. The character of
the wood itself is not greatly altered, the annual
rings being quite regular, only wider than in the
normal wood. (For details see [21j.) Only two gen
eral characteristics appear important: the unusual
ly rapid growth of tumor wood and the fact that
the tumors usually and in all probability always
randomly distributed. Closer study was not under
originate in very young wood close enough to the
taken at the time, and it was not until 1947 that it pith to indicate that the tumefacient change must
could be resumed. In that year, with the support of have occurred very early in the life of any affected
a grant from the American Cancer Society made leader, twig, or root, when it was relatively soft
on recommendation of the Committee on Growth and tender.
of the National Research Council, a program of de
DISTRIBUTION
tailed investigation was undertaken. It was
Our information on the distribution of these
planned that this should encompass four phases:
(a) a thorough survey of the geographical, spatial, growths was gathered in three ways. First, we have
and ecological distribution of the tumors, with the personally examined in considerable detail the
hope of obtaining some suggestion of possible forests of Mount Desert Island, Schoodic Peninsu
la, and neighboring smaller islands and promi
causes for these growths; (6) a study of the anato
my and histology of the tumors in order that the nences in and on Frenchmans Bay and Blue Hill
This has been carried on by the senior author
nature of the changes involved might be more Bay.
(P. R. W.) and helpers during seven summers' resi
clearly defined and possible causative or contribu
tory factors recognized; (c) studies aimed at the dence on and near Mount Desert Island and 2
cultivation in vitro of the tissues of tumors and of years, winter and summer at Bar Harbor; and by
unaffected plant parts as a means of examining the the junior author (W. F. M.) during 2 years full
physiological and pathological characteristics of time study of this tumor problem. Although there
are regions, particularly along the low western
the affected tissues and the degree of their malig
nancy; and (d) experiments in the laboratory and in part of Mount Desert Island which we have not
intensively on foot, we have driven over
the field planned to elucidate further the nature of covered
all the Island's roads and have examined on foot
the malady in its various phases. This paper covers
our progress to date on the first of these four most localities which we believe are likely sites for
phases: the distribution and occurrence of the such growths.
Second, we have contacted, personally and by
tumors in question. Since 1951 the work has been
letter,
foresters, naturalists, and tree surgeons in
done in large part by the junior author (W. F. M.).
many sections included in the range of the white
OBSERVATIONS
spruce. We have had the close cooperation of the
Forestry and National Park services of the United
CHARACTERIZATION
The tumors in question are mostly regular, States, the Department of Resources and Develop
smooth, woody growths, not fungating or fissured, ment of Canada, and the forestry departments of
with continuous bark. They may be of all sizes. the University of Maine, and of the University of
Montana. We have specifically asked these coop
The smallest which we have been able to recog
erating
agencies for information on the occurrence,
nize, so small that they are evident only when the
bark of a twig is stripped off, are in the form of range, and ecological peculiarities of any similar
ridges about a millimeter wide and 3-4 mm. long growths which might be observed.
Third, we caused to be published in the news
(Fig. 1). From these tiny excrescences they range
up to burls a meter or more in diameter (Fig. 2). papers of Bangor (inland) and Portland (seaThese may be solitary, only one to a tree (Fig. 8), coast) and to be released to other papers through
or there may be hundreds of them scattered over one of the national press agencies an article, in
the twigs, branches, trunk and even exposed roots cluding photographs, requesting from the general
of an affected individual (Fig. 4). They appear not public any information about such tree tumors.
to be distributed in any special manner on a tree, Since we received replies and clippings from San
on one side to the exclusion of another, nor on the José,California; Billings, Montana; Vicksburg,
Mississippi; Galveston, Texas; West Paducah,
top or bottom of branches. Thus neither gravity
Kentucky, and other distant points, it is clear that
nor sunlight appears to influence their distribu
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1954 American Association for Cancer Research.
130
Cancer Research
this request was widely disseminated. We have no
information on how complete the coverage may
have been.
From these three sources of information emerge
certain facts. Picea glauca has a distribution which
includes all of Canada except the west coast, where
it is replaced by Picea sitchensis and Picea engelmanni, all of the states of Maine, New Hampshire,
and Vermont, much of northern New York and
Michigan, most of Wisconsin and northern Minne
sota, and scattered areas in Montana, Wyoming,
and South Dakota (Chart 1). Yet the reports we
have of tumors of this type are restricted, with three
photographs. They appear to represent similar
growths but must be accepted with some reserve in
the absence of personal verification. We say this
especially because information from the Botany
Department of the University of Montana, repre
senting observations from rangers in the Missoula,
Lolo, Bitterroot, Blackfeet, and Selway forests of
the northern Rocky Mountains in the United
States indicates that such growths have not been
reported from this region, only a few hundred
miles south of Banff. On a continental scale these
intumescences are certainly most common along
the seacoast, but may possibly occur in wet,
swampy, and exposed localities at some inland
points.
Within the eastern part of the range of Picea
glauca the distribution of tumors likewise does not
correspond to the general range but is closely lim
ited. We have no reports of tumors from the inte
rior of Maine, New Hampshire, Vermont,1 or New
York, yet we have dozens of reports from the
coastal regions from Kittery north to Nova Scotia.
These are concentrated around the Boothbay Har
bor region, the headlands and islands of Penobscot
Bay, Blue Hill Bay, and Frenchmans Bay. The
only reports from New Brunswick and Nova
Scotia have come from the Bay of Fundy National
Park and Cape Breton Highlands National Park,
which are wild and rugged terrain (Chart 2). We
have no assurance that our requests have com
pletely covered these states, nor that everyone
who has seen the tumors has reported to us. The
population of Maine is mostly concentrated along
the seacoast, so that observation is perhaps
weighted in favor of these regions. We might thus
expect a somewhat greater number of reports from
these regions, even if the tumors were evenly dis
tributed throughout the state. However, this
would not account for the complete absence of re
CHABT1.—Mapof North America showing the distribution
ports from inland. We believe that our reports do
of White Spruce (shaded area) and of reported tumors
represent the true population density of the tu
(blacked-in areas). (Compare Charts 3 and 3.)
mors themselves, not merely the population densi
exceptions, to a narrow band along the eastern sea ty of possible observers. It seems clear that the
coast. These exceptions are: a report of similar affected trees are concentrated along the coast.
tumors "along streams, in swampy areas, at the
This conclusion is borne out quite sharply by
heads of lakes and occasionally near timber line" the facts as disclosed by our personal and detailed
in the vicinity of Banff National Park (Canada) ; survey of Mount Desert Island and neighboring
regions. In this area we have been able to assure
a report of occasional tumors on spruce in the up
per Lake Superior region and in the Isle Royale ourselves of fairly complete coverage. There are
three species of spruce on the Island, Picea glauca,
National Park and similar south central Canadi
P. rubens and P. mariana. Tumors are found only
an locations; and a report in the lumbering litera
1Since completion of this paper we have received a report
ture (1) of similar growths on Sitka Spruce around
Victoria, Canada, in locations similar to those re from the Vermont State Forest Service of similar growth in
five localities in Vermont, but with no details. These will be
corded in Maine. The reports from Banff and Isle investigated.
One of the purposes in publishing this paper is
Royale are from the Canadian Department of Re
to arouse wider interest in the problem and to obtain detailed
sources and Development and are accompanied by reports from other observers.
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1954 American Association for Cancer Research.
CHABT2.—Map of the northeastern coastal region showing
the distribution of tumors reported from this region.
CHART3.—Map of Mt. Desert Island and neighboring
region showing distribution of tumor-affected trees person
personally
examined by the authors.
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1954 American Association for Cancer Research.
132
Cancer Research
on the first species. Picea glauca occurs in all parts
of the Island, at all altitudes from sea level to
1,500 feet—but the tumors do not (Chart 3).
Probably 99 per cent of the growths observed occur
in a narrow strip, never more than a quarter mile
wide and generally less than 100 feet wide, border
ing the sea. This strip begins at Bar Harbor, rough
ly at the northeastern corner of the Island and ex
tends along the exposed eastern and southern
sides to Northeast Harbor. A few affected trees
occur at Ship Harbor, on the extreme southern
tip. These are the two sides of the Island which
face the open ocean. On the other two sides—west,
facing Blue Hill Bay and north, facing Frenchmans Bay—tumors occur very rarely. We have
noted them in the region of Latty Cove at the ex
posed southwest side of the Island and at the
head of Somes Sound (but not along its sides) and
have seen two affected trees directly on the shore
within 10 feet of high tide mark at Salisbury Cove
on the north side facing Frenchmans Bay which is
about 2 miles wide at this point. We have observed
no affected trees outside this strip, and only on the
low flat, back of Anemone Cave, are the affected
trees more than a hundred feet from high tide
mark.
The same characteristic limitations hold for
neighboring islands and headlands. We have seen
a single affected tree near the shore on Burnt Por
cupine Island, about a score of trees on the exposed
low western side of Iron Bound Island, a single tree
on the exposed tip of Schoodic Peninsula, and a
single tree on the exposed shore of Islesford. An
other tree on Islesford which was cut down in 1906,
was considered so unusual and unique that it was
photographed and kept for many years in a local
home. The photograph is still preserved in the
Islesford Historical Âluseum. We have reports
(not personally verified) of a single tree on Norris
Island off Winter Harbor and of several on the
exposed tip of the peninsula at Brooklyn.
The distribution on Schoodic Peninsula, Iron
Bound Island, Bald Porcupine Island, and Schoon
er Head is perhaps pertinent. An intensive sur
vey of Schoodic Peninsula in which a traverse was
made east to west across the summit, then north
to south from the summit down across the "Anvil"
and along the east, south, and west shores, revealed
no affected trees except a single one on the exposed
southern tip. On Iron Bound Island a similar eastwest traverse, a north-south traverse and a survey
of the shore located two groups of trees, both near
the shore on the exposed western side. No affected
trees were found on the high cliffs at the south
end, although this likewise is exposed. Similarly,
no affected trees were found in a careful examina
tion of Bald Porcupine Island, whose exposed
shores are all high, hence well away from the
water and so placed that spray would not normally
be driven up on to them except in extreme souther
ly storms. On Schooner Head, however, less than
a mile away and with a similar exposure but in
which the shore is made up of a series of rocky
steps over which the surf breaks and the spray is
borne inland in every storm, there are thousands of
affected trees. Sprengel (16) has reported that tu
mors occur in Germany on spruce trees (species
not designated) on swampy ground and at the
edges of clearings but more rarely in the midst of
dense stands.
It seems clear that, with rare and uncertain ex
ceptions, the affected trees are all located on the
wet and rocky exposed sea shore, within reach of
salt spray. Those reported from Banff, Isle Royale,
and similar scattered inland locations appear to be
exceptions to the rule of exposure to salt water,
but these likewise are from wet and exposed
points. Exposure, and especially exposure to salt
water, appears to be sufficiently common to con
stitute a significant factor in tumor initiation.
DISCUSSION
These tumors are, then, not a general charac
teristic of White Spruce, but are found only on
those White Spruce trees growing within a nar
row, sharply defined and closely localized segment
of the total range of the species. Anatomical
studies (21) have shown that they originate from
single cells or small groups of cells at closely local
ized points restricted to the cambium of young
twigs. Whatever agency or agencies have produced
these growths must be limited at some stage to
these restricted localities and must also be capable
of explaining the observed anatomical limitations.
Only two agencies are known to cause autono
mous tumorous growths on plants. The first is
heredity, exemplified in the genetically induced
tumors affecting the progeny of certain inter
specific hybrids within the genus Nicotiana (5, 12,
14) and paralleled in animals by the hereditary
melanomas of certain crosses among the "Swordtail" fishes (10, 11). These are true malignancies,
capable of artificial metastasis (17). We believe
that hereditary factors are unlikely to prove im
portant as primary or even major contributory
factors in the etiology of these spruce tumors. They
can hardly explain either the differences between
tumor-bearing and tumor-free trees within the spe
cies P. glauca, or the geographical distribution of
affected trees. Hereditary defects may arise either
by mutation or by hybridization. A single mutant
should give rise to a family of trees all of which
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1954 American Association for Cancer Research.
WHITE ANDMILLINGTON—A
Woody Tumor in White Spruce
should occur within the range to which seeds of the
mutated individual are distributed by normal
means. Certainly there is no reason to suppose
that such seeds would be carried from island to
island, even by sea birds, without some continuity
and without penetrating into the interior of the
affected locales. Repeated mutations should occur
at random over the entire range of the species. If
the defect were the result of hybridization with
Picea rubens or P. mañana,it should show some
correlation with the range of one of these species.
The facts do not agree with any of these possibili
ties. While it is possible that a hereditary defect
might sensitize certain individuals of Picea glauca
to some external agent which in turn induces tu
mors, it is still true that the inducing agent must
in some respect have the peculiar distribution
noted.
The second known cause of malignant tumors
in plants is the crown-gall bacterium, Agrobacterium tumefaciens. Past experience has taught us that
this agent often brings about the change from a
normal to a malignant state during a relatively
brief exposure period and then disappears, leaving
a sterile tumor (2, 3, 19, 20). We have found no
evidence of bacteria being present in older tumors,
either in sections or in the course of experiments
designed to cultivate the tumor tissues in vitro (to
be reported elsewhere), but this in no way rules
out the possibility that such bacteria might have
served as transient tumefacient agents. Among
known causes of plant malignancies, such bacteria
constitute an example which does not conflict with
our evidence. Nevertheless, if they have in fact
served as such agents, an explanation must be
found—first for their entry, or at least their effec
tiveness only at characteristically restricted sites
in the plant, and second for the observed geo
graphical limitations in their effectiveness. They
cannot represent the sole cause of Spruce tumors.
In searching for possible agents capable of pin
pointing the origin of these growths in single cells
or small groups of cells on young twigs of trees,
there appears to be only one well recognized type
of such agents. This is a sucking or ovipositing in
sect vector. While we have found no evidence of
lesions, we must recognize that such an insect
would make a wound so tiny that it might easily
be obliterated during the 3 or 4 years' lapse be
tween tumor initiation and the youngest stage
which we have been able to study.
At least one insect does attack spruce trees
which is capable of fulfilling the requirements
posed by this problem. The Spruce Gall Aphis,
Chermes abietis, lays its eggs at the bases of the
leaves in young buds of spruce trees. Trees affected
133
by this pest are to be seen throughout the regions
in which tumors occur. Often 90 per cent of the
terminal twigs will be damaged. The twigs develop
into galls a centimeter in diameter and 3-6 centi
meters long, the gall being formed by hypertrophy
of the cortical tissues and leaf bases. Usually these
galls affect the entire circumference of the twig.
The twig is then killed. However, the eggs are occa
sionally laid on only one side of the twig, and the
developing gall is lateral, leaving a sector of normal
cortex and cambium at one side. Since the gall per
sists only for a single season as a living mass of
hypertrophied tissue and then dies, in those cases
where living cambium persists the dead gall is cut
off by a cork cambium and is sloughed, leaving a
smooth scar devoid the leaf bases which charac
terize normal 2d-, 3d-, and 4th-year bark. The
twig then grows on into a branch or leader. It
seems conceivable that these occasional sloughed
off lateral galls might serve as points of entry for
a tumefacient agent such as crown-gall bacteria.
We have, in fact, occasionally observed incipient
tumors under such scars.
However, even if we postulate an insect lesion
as a first step in tumor causation, and a subsequent
infection with crown-gall bacteria as a second step,
this still leaves unexplained the geographical dis
tribution of the affected trees. Braun has shown
that while Agrobacterium tumefaciens may become
established and multiply in certain plants under a
wide range of conditions, the infection will result
in a tumorous change in the host cells only under
much more restricted conditions (2—4).He postu
lates two steps in the tumefacient process: first, a
change in cellular potency, and, second, an evoca
tion of that potency into an actively growing neo
plasm. It appears in this case that some factor
closely associated with the proximity to the ocean
might serve as such an activating factor. One of
us (P. R. W.) has postulated that salt spray may
be this factor. According to this hypothesis, tumorization of these spruce trees involves three
steps: (a) a lesion, possibly caused by an insect,
which penetrates to the cambium and localizes the
attack, (6) an infection of this lesion by an agent
capable of bringing about the transformation of
cambial cells into potential tumor cells but not
capable of evoking overt tumors, and (c) an evok
ing agent which likewise gains access to the altered
cambial cells through the initial lesion but which
comes only from the sea, hence restricts the overt
tumors to a narrow strip having access to salt
spray. We have no direct evidence of any of these
three steps, but they serve as possible leads
capable of serving as bases for planned experi
ments.
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1954 American Association for Cancer Research.
134
Cancer Research
This hypothesis agrees with all our evidence,
with one exception. This is the reported occurrence
of tumors in the regions of Banff National Park
and of upper Lake Superior. Two explanations
seem possible. These exceptional tumors may, in
fact, represent a quite different sort of growth hav
ing a different set of causes, as do so many sporadic
but well known burls. In the absence of personal
verification we must consider this possibility. If,
however, these inland tumors do prove to be iden
tical with our coastal growths, then salt spray is
not an acceptable explanation of the third step in
our postulated sequence, and we must search for
some other factor associated with moist localities
but not with the ocean. In this case we would be
faced with, what appears to the authors, the much
more difficult task of explaining why tumors
should occur on inland lakes of the deep interior
(Lake Louise, Lake Superior, etc.) but not on the
inland lakes of Maine (The Rangeley Lakes, for
example) nor on the fresh water lakes of Mount
Desert Island.
REFERENCES
1. ANONYMOUS.
American Lumberman. (Cited by Graf von
Schwerin, "Merkwürdige Stammbildungen der Sitka
Fichte," Mitt. Deutsch Dendrol. Gesell., 26:227-28,1917.
sidering all the evidence, a hypothesis has been
formulated to explain these growths and on which
to base future investigations. This hypothesis
visualizes three consecutive causal factors: an in
sect lesion, a sensitizing agent such as the crowngall bacterium, and an evoking agent such as salt
spray. This hypothesis is consonant with all the
observed facts except for the reported occasional
occurrence of such tumors at two widely separated
locations in the interior of Canada. It is suggested
that these exceptions may represent a similar ap
pearing tumor but of different origin from the one
investigated here.
The citation is defective and we have been unable to find
the original.)
2. BRAUN,A. C. Thermal Studies on the Factors Responsible
for Tumor Initiation in Crown-Gall. Am. J. Bot., 34:23440, 1947.
3.
. Conditioning of the Host Cell as a Factor in the
Transformation Process in Crown-Gall. Growth, 16:6574, 1952.
4. BRAUN,A. C., and HANDLE,R. J. Studies on the Inactivation of the Tumor-inducing Principle in Crown-Gall.
Growth, 12:255-69, 1948.
5. BRIEOER,F. G., and FÖRSTER,
R. Tumores em certes
hÃ-bridosdo genero Nicotiana. Bragantia, 2:259-74, 1942.
6. Cook, J. W.; HASLEWOOD,G. A. D.; HEWETT, C. L.;
HIEGEH, I.; KENNAWAT,E. L.; and MATNEABD,W. V.
Chemical Compounds as Carcinogenic Agents. Am. J.
Cancer, 29:219-59, 1937.
7. DOUGLASS,A. E. Evidences of Cycles in Tree Ring
Records. Proc. Nat. Acad. Sc., 19:350-60, 1933.
8.
. Tree-Ring Work. Tree-Ring Bull., 4:3-6, 19S7.
9.
. Crossdating in Dendrochronology. J. Forestry,
39:825-31, 1941.
10. GORDON,M. Genetic and Correlated Studies of Normal
and Atypical Pigment Cell Growth. Growth, 16 (suppl.):
153-218, 1951.
11. GORDON,M., and SMITH,G. M. The Production of a
Melanotic Neoplastic Disease in Fishes by Selective
Matings. IV. Genetics of Geographical Species Hybrids.
Am. J. Cancer, 34:543-65, 1938.
12. KEHR, A. E. Genetic Tumors in Nicotiana. Am. Nat.,
86:51-64, 1951.
13. KENNAWAT,E. L. The Formation of a Cancer-producing
Substance from Isoprene (2-Methylbutadiene). J. Path. &
Bact., 27:233-38, 1924.
14. KOSTOFF,D. Tumors and Other Malformations on Certain
Nicotiana Hybrids. Zbl. Bakt. Parasit. Infect., 81:244-60,
1930.
15. POTT, P. Chirurgical Observations Relative to the Cata
ract, the Polypus of the Nose, the Cancer of the Scrotum,
the Different Kinds of Ruptures and the Mortification of
the Toes and Feet. London, 1775.
16. SPRENGEL,F. Ueber die Kropfkrankheit an Eiche, Kiefer
und Fichte. Phytopath. Ztschr., 9:583-635, 1936.
17. WHITE,P. R. Transplantation of Plant Tumors of Genetic
Origin. Cancer Research, 4:791-94, 1941.
18.
. Metastatic (Graft) Tumors of Bacteria-free
Crown-Galls on Vinca rosea. Am. J. Bot., 32:237-41,
1945.
19.
. Neoplastic Growth in Plants. Quart. Rev. Biol.,
26:1-16, 1951.
20. WHITE,P. R., and BRAUN,A. C. A Cancerous Neoplasm
of Plants. Autonomous Bacteria-free Crown-Gall Tissue.
Cancer Research, 2:597-617,1942.
21. WHITE,P. R., and MILLINGTON,
W. F. The Structure and
Development of a Woody Tumor Affecting Picea glauca.
Am. J. Bot. (in press).
FIG. 1.—Asmall tumor on a young twig. Note that only an
indistinct swelling is evident on the bark, the tumor being
clearly visible only when the bark is removed.
FIG. 2.—Alarge tumor near the base of a tree on Schooner
Head, Mt. Desert Island.
FIG. S.—Anexample of a tree, near Anemone Cave, Mt.
Desert Island, which bore only a single large tumor.
FIG. 4.—Anexample of a heavily affected tree. Every twig
of this tree bore minute tumors and the tree itself was dead
when photographed. This tree was on a small, low, semide
tached island near Seal Harbor, which was periodically inun
dated during heavy storms.
SUMMARY
A tumor affecting trunks, branches, and roots of
trees of the White Spruce, Picea glauca, has been
studied. A survey of distribution has shown this
tumor to be restricted, with a few possible and un
verified exceptions, to trees growing within a few
feet of the ocean, on exposed shores and headlands.
The exceptions are trees reported from the Canadi
an Rockies and from the upper Lake Superior re
gion. The tumors themselves are initiated within
the first year's growth of any affected branch. Con
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1954 American Association for Cancer Research.
•'\
*-~J
-/i
»o*.
"--Al
TSf>
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1954 American Association for Cancer Research.
The Distribution and Possible Importance of a Woody Tumor on
Trees of the White Spruce, Picea glauca
Philip R. White and William F. Millington
Cancer Res 1954;14:128-134.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/14/2/128
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1954 American Association for Cancer Research.