American Journal of Botany 86(3): 367–371. 1999.

EFFECTS OF ENVIRONMENTAL FACTORS ON

DEVELOPMENT OF WOOD1

DAVID H. ARNOLD2 AND JAMES D. MAUSETH

Department of Botany, University of Texas, Austin, Texas 78713

This research tested the hypothesis that environmental factors (light, water, and nutrient levels) affect wood development.
Specimens were placed in treatments of low, medium, or high levels of light, water, nitrogen, or phosphorus for one year.
Control plants received medium levels of all factors, while experimental plants received medium levels of all factors except
the experimental factor; for example, ‘‘high light’’ treatment consisted of high light but medium levels of water, nitrogen,
and phosphorus. Some character changes seen in Cereus peruvianus were a reduction in mean vessel diameter and shoot
elongation as a result of low nitrogen and low phosphorus treatments and a reduction in mean vessel density due to low
light; high water induced broader vessels and greater shoot elongation. In Cereus tetragonus, low water treatment caused a
reduction in mean vessel diameter, and high nitrogen decreased the amount of wood produced. Whereas all characters
studied showed a significant correlation with at least one treatment in one species, few characters responded similarly
between species. Estimated specific conductivity of wood could be altered by treatments affecting either vessel density or
vessel diameter strongly or by treatments affecting both diameter and density weakly. Under the conditions tested, wood
structure was stable but estimated conducting capacity was more flexible.

Key words: anatomy; Cactaceae; Cereus; development; environment; wood.

As the concentration of greenhouse gasses increases in
the atmosphere, global temperatures rise, rainfall patterns
are altered, cloud cover changes the amount of light
reaching plants, and minerals are leached from the soil
in some areas, deposited in others (Larcher, 1995). En-
vironmental factors affect plant vigor and health, but little
is known about how such factors might influence a plant’s
internal structure. Many agricultural studies have exam-
ined responses such as crop yield, but there have been
few studies dealing with morphogenic effects (Nobel,
1988, 1996; Larcher, 1995). Penfound (1931) reported
that Helianthus annuus and Polygonum hydropiper pro-
duced more and larger vessels if given increased water.
Doley and Leyton (1968, 1970) found that moisture level
affected vessel diameter in Fraxinus excelsior, and Biss-
ing (1982) had similar results in a study of several dicots.
Formation of aerenchyma and root hairs was also affected
by environmental factors (Drew, Jackson, and Gifford,
1979; Kawase, 1979; Etherington, 1984; Heathcote, Da-
vies, and Etherington, 1987; Jackson, Manwaring, and
Caldwell, 1990), and increased light induced the forma-
tion of leaves with a higher density of stomata, veins,
and chlorophyll (Tselniker, 1978; Goryshina, 1980,
1989). Low nitrogen supply induced altered growth rate
and lignification of leaf epidermis and nonveinal scleren-
chyma in Poa, but few anatomical characters were ex-
amined (Van Arendonk et al., 1997). Even experiments
examining effects of elevated CO2 seem to study few
anatomical responses other than changes in stomatal den-
sity (Raschi et al., 1997). Surprisingly, this short list ap-
pears to be the bulk of the experiments related to the
effect of environmental factors on plant structure. Yet a
plant’s developmental response to environmental condi-
1 Manuscript received 21 October 1997; revision accepted 23 June
1998.
This research was supported by a grant from the Research Committee
of the Cactus and Succulent Society of America.
2 Author for correspondence.
tions is a factor in determining whether that species sur-
vives environmental change or is affected deleteriously
by it.
Due to the scarcity of experimental data we decided to
examine anatomical responsiveness to environment. Cacti
were chosen for the study because our experience has
shown them to be morphologically responsive to varia-
tions in environmental factors, and many have complex
polymorphic anatomy (Mauseth, 1993; Mauseth and Ple-
mons-Rodriguez, 1997, 1998) so anatomical responsive-
ness might also be expected. Because many cacti are
adapted to habitats of extreme aridity, intense insolation,
and mineral-rich soils, even small increases in global
rainfall and cloudiness would be significant changes for
them.
MATERIALS AND METHODS
In April 1994, 6-mo-old plants of two species of cacti (62 plants of
Cereus peruvianus (L.) Miller and 61 of Cereus tetragonus Miller) were
purchased from a commercial nursery. All plants were cuttings obtained
from single clones of each species. Plants were acclimated in a green-
house in Austin, Texas, for 60 d under conditions similar to those of
the nursery. All plants were in 7.6 cm (3-inch) plastic pots with a 1:1
mixture of potting soil and vermiculite. The soil contained 13-13-13
Osmocote time-release fertilizer (3.0 g per plant; NPK concentration of
1000, 1000, and 1000 parts per million [ppm], respectively); all plants
received an additional application of 18-18-18 instant (water soluble)
fertilizer (2.4 g/L NPK concentration of 800, 800, and 800 ppm of
solution) at 14-d intervals. Soil was thoroughly soaked once per week
with either fertilizer solution or water.
After acclimation, plants of each species were randomly assigned to
treatments. Plants were depotted and the roots gently washed to remove
all fertilizer, then the plants were repotted using a 1:1 mixture of potting
soil and vermiculite. Plants were measured for height and for diameter
at base, mid-height, and 5 cm below the apex. Plants were allowed to
recover from repotting by being irrigated with water only, once each
week, until treatments began in August 1994.
Experimental treatments were high, medium, and low levels of light,
water, nitrogen, and phosphorus. High light was full ambient light inside

367
368 AMERICAN JOURNAL
the greenhouse augmented with 12 h of fluorescent lighting per day.
This provided an average photosynthetically active radiation (PAR) of
1720 mmol photons·m22·s21. Medium and low treatments were the same
as high light except that plants were placed under shade cloth that
reduced light by 50 % (PAR of 860 mmol photons·m22·s21) or 90 %
(PAR of 170 mmol photons·m22·s21). PAR was measured ten times over
the year using a LI-COR, Inc. model LI-1600 (LI-COR, Lincoln, Ne-
braska) steady-state poromoter with light meter. Plants receiving high
or low light received medium water and medium nutrient levels.
Plants receiving high water treatment were watered twice each week,
once with water only and once with fertilizer, each time receiving
enough to exceed field capacity. The soil was never allowed to dry out.
Plants in medium water treatments received enough fertilizer solution
to exceed field capacity once each week. Soil in these pots was normally
dry 1 or 2 d each week. Plants receiving low water treatment received
60 mL of fertilizer solution each week, and the soil was normally dry
4 or 5 d between watering. Plants receiving experimental treatments of
water or one of the other factors were given medium light.
Fertilizer treatments were as follows. High nitrogen (53-20-20) was
prepared by mixing 33-0-0 and 20-20-20 (total NPK concentration of
3200, 800, and 800 ppm). High phosphorus (20-66-20) was prepared
with 20-20-20 plus 0-46-0 (total NPK concentration of 800, 3800, and
800 ppm). Medium fertilizer treatments were given 20-20-20 instant
mix alone (NPK concentration of 800, 800, and 800 ppm). Low nitrogen
treatment was a mix of 0-20-0 and 0-0-20 (NPK concentration of 0,
800, and 800 ppm). Low phosphorus treatment was prepared from 20-
0-0 and 0-0-20 (NPK concentration of 800, 0, and 800 ppm). The water-
soluble fertilizer used was Greenlight brand 20-20-20, produced by
Greenlight, Inc., San Antonio, Texas. The N (as NH4NO3), P (as P2O5
either 800 or 3000 ppm), and K (as K2O-soluble potash) used to modify
the solutions were granular fertilizers manufactured by Green Diamond,
Inc., Bay City, Texas. The granules were ground to the consistency of
fine powder to make them water soluble also. High treatment solutions
were prepared by mixing 2.4 g of commercial instant mix per litre of
water and the addition of either 0.8 g of NH4NO3 or 1.0 g of 3000 ppm
P2O5. Low nitrogen treatment was prepared by mixing 0.9 g of 800 ppm
P2O5 and 0.7 g of K. Low phosphorus treatment was a combination of
0.8 g of N (as NH3, NO3 and urea) and 0.8 g K (as K2O-soluble potash).
The treatments were as follows: ‘‘control’’ plants received medium
light, water, nitrogen, and phosphorus. Experimental plants (everything
except the control plants) received medium levels except for the exper-
imental factor; for example, plants in the ‘‘high light’’ treatment re-
ceived high light but medium water, nitrogen, and phosphorus.
Temperature was maintained between 248 and 278C.
All plants were maintained in treatment from 1 August 1994 until
they were harvested beginning 1 August 1995. At harvest, each plant
was again measured for height and diameter. For Cereus tetragonus,
whose plants are highly branched, all branches were removed and mea-
surements taken from the main stem.
Plants were harvested by severing the main stem at ground level. To
separate the original growth from growth that had occurred during treat-
ment each plant was cut in two transversely at the height the plant had
been when treatment was initiated. Histological samples were obtained
from basal, middle, and apical regions of both the lower portion and
the upper portion. All tissues in the upper portion had been initiated
and had developed during treatment; in the lower portion, all primary
tissues had been initiated and had matured prior to treatment. In the
lower portion of the stem, the innermost, oldest wood had been pro-
duced prior to treatment and the outermost wood, adjacent to the vas-
cular cambium, had been produced during treatment. Samples were
fixed and processed as described by Mauseth, Montenegro, and Wal-
ckowiak (1984). Tissues were analyzed using brightfield microscopy.
When collecting wood data from the base of the plants (in the region
of the stem that had existed before the experiment started), care was
taken to ensure that only cells produced during the experimental period
were used: measurements and counts were restricted to within 200 mm
OF BOTANY [Vol. 86
of the vascular cambium. For each species in each treatment, at least
250 vessels were measured for lumen diameter. They were also scored
as to occurrence in clusters or as solitary vessels. Vessel density was
measured as the number of vessels per square millimetre of axial wood
only; rays were not included in the measurement. Vessel density has
been calculated this way in other studies of cacti (Mauseth, 1996; Mau-
seth and Landrum, 1997; Mauseth and Plemens-Rodriguez, 1998) to
eliminate variability that would result from differences in relative vol-
ume of rays in wood. For each species in each treatment, 55 fields were
sampled at 403 (field size 5 21300 mm2) and all vessels in each field
were counted.
Within each species, treatment means for each character were ana-
lyzed using the ANOVA function of PROC GLM in PC-SAS v6.11. If
main effects were significant, then pairwise comparisons were made
between the control and other treatment means using PROC TTEST
with Cochran option (SAS, 1996).
RESULTS
Shoot elongation—Plants of Cereus peruvianus were
unbranched columns, and all elongation growth occurred
in the solitary shoot. In control conditions, the plants only
had a mean increase in height of 17.3 cm (Table 1, col-
umn 2) but were somewhat variable, with one plant grow-
ing 33.5 cm during treatment, another growing only 6.0
cm, and the other three plants growing a more moderate
17.5, 15.5, and 14.0 cm (SD 5 4.5 cm). Only low nitro-
gen treatment produced a statistically significant effect on
growth (reduced shoot elongation: mean elongation only
3.5 cm, SD 5 1.4 cm; P , 0.05). Low phosphorus treat-
ment had no significant effect on shoot elongation (P ,
0.06: one plant had unusually great elongation, all the
rest had very little). High water treatment produced al-
most double the mean amount of elongation growth of
the control but was not statistically significant (P , 0.1)
due to high variability, because a single sample (plant no.
5) was aberrant and elongated only 13.0 cm, whereas the
other four elongated 43.0, 32.0, 38.5, and 27.9 cm. The
low amount of growth in plant no. 5 might have been
due either to natural genetic variability or to insect or
fungal damage or some other adverse factor that was not
intended to be part of the experiment. If the aberrant plant
no. 5 was diseased, it should be removed from consid-
eration, and then mean elongation in high water treatment
would be 35.1 cm (SD 5 7.0), which would be significant
statistically (P , 0.05). But this plant appeared healthy
and other aspects of its anatomy were within normal pa-
rameters, so its small amount of elongation is probably a
valid datum. Low water treatment did not affect shoot
elongation, and here variability was due to several sam-
ples, not just a single aberrant sample. Neither high ni-
trogen nor high phosphorus had significant effects, and
neither did high light or low light. Low light was not dim
enough to induce any obvious sign of etiolation.
Plants of Cereus tetragonus were highly branched, and
increased vigor could be expressed as increased elonga-
tion or increased branching. In C. tetragonus, elongation
of the main stem was not significantly affected by any
treatment, but the number of branches was: mean number
of branches per plant ranged from a low of 24 to a high
of 67, and low nitrogen produced a significantly (P ,
0.04) low number of branches, and high nitrogen pro-
duced a significantly (P , 0.01) higher number of
branches (data not presented in Table 1). Light, water,
March 1999] ARNOLD AND MAUSETH—EFFECTS OF ENVIRONMENT ON WOOD ANATOMY 369

TABLE 1. Mean values and standard errors for shoot growth and anatomcal characters in Cereus.

Mean
Mean Mean Mean % Mean
Mean overall solitary clustered soli- vessel Mean
shoot vessel vessel vessel tary density Mean amount of estimated
elongation diameter diameter diameter ves- (no. wood/bundle conductance
Treatment N (cm) (mm) (mm) (mm) sels vessels/mm2) (mm2) (mm2)

C. peruvianus
Control 5 17.3 6 4.5 32.4 6 2.6 33.7 6 2.7 31.3 6 2.5 41 405 6 64 0.19 6 0.06 40.5 6 6.4
High light 5 14.5 6 2.6 32.3 6 3.0 35.6 6 3.2 31.4 6 2.8 22 580 6 145 0.17 6 0.04 60.6 6 15.2
Low light 4 15.7 6 1.5 28.9 6 4.2 31.1 6 4.5 26.1 6 3.2 51 173 6 40* 0.12 6 0.06 15.8 6 3.7*
High nitrogen 5 16.8 6 2.8 34.1 6 3.5 36.1 6 3.7 32.7 6 3.1 39 440 6 68 0.25 6 0.06 52.1 6 8.1
Low nitrogen 5 3.5 6 1.4* 23.3 6 1.9* 25.0 6 1.7* 21.5 6 1.9* 51 368 6 32 0.16 6 0.02 10.8 6 1.0*
High phosphorus 5 20.4 6 5.5 32.4 6 3.7 34.0 6 3.6 31.8 6 3.8 32 417 6 78 0.23 6 0.1 48.6 6 9.1
Low phosphorus 5 6.1 6 2.6m 24.4 6 1.1* 26.0 6 1.5m 23.5 6 1.0* 46 382 6 34 0.18 6 0.03 11.9 6 1.1
High water 5 30.7 6 5.2 35.5 6 1.4 38.1 6 1.7 33.3 6 1.1 41 291 6 27 0.50 6 0.15 40.2 6 3.9
Low water 4 6.4 6 2.1 35.2 6 2.9 36.9 6 2.9 33.2 6 2.9 48 374 6 117 0.18 6 0.03 46.5 6 14.5
C. tetragonus
Control 6 7.0 6 2.9 27.9 6 0.5 30.6 6 0.5 26.1 6 0.8 42 271 6 23 0.09 6 0.02 14.3 6 1.3
High lighta 3 12.2 6 2.1 25.2 6 3.6 26.1 6 3.8 24.5 6 3.2 35 233 6 78 0.18 6 0.08 8.6 6 2.9
High nitrogen 8 5.4 6 0.8 23.4 6 1.3 24.6 6 1.2* 22.3 6 1.5m 40 257 6 22 0.03 6 0.01* 7.1 6 0.6*
Low nitrogen 7 6.9 6 1.3 24.8 6 2.1 25.4 6 2.6 24.1 6 1.8 39 258 6 31 0.11 6 0.04 9.4 6 1.1*
High phosphorus 7 8.3 6 1.8 29.0 6 3.1 30.5 6 3.3 26.7 6 2.7 52 260 6 20 0.13 6 0.04 19.5 6 1.5*
Low phosphorus 7 12.6 6 1.8 28.0 6 1.7 29.2 6 1.7 26.6 6 1.4 51 224 6 20 0.13 6 0.03 12.6 6 1.1
High water 6 5.9 6 2.1 26.6 6 1.3 28.0 6 1.1 24.0 6 1.9 57 244 6 39 0.22 6 0.15 10.3 6 1.7
Low water 6 5.8 6 1.2 22.7 6 0.9* 23.8 6 1.1* 22.2 6 0.9* 34 297 6 35 0.03 6 0.01* 6.6 6 0.8*
* 5 significance P , 0.05; m 5 marginal significance P , 0.06–0.09.
a The low light samples for this species were lost during processing for microscopy.

and phosphorus treatments produced no significant dif-
ferences in branching.
Mean diameter of all vessels—In both species, vessels
occurred either as solitary vessels or as small clusters of
2–5 vessels. In all treatments, vessels were narrow, with
a grand mean diameter (mean 6 1 SE of all vessels in
all treatments) of 31.0 6 1.5 mm in C. peruvianus and
26.0 6 0.8 mm in C. tetragonus. Mean vessel diameters
in control plants (those receiving medium levels of light,
water, and nutrients) were (32.4 6 2.6 mm in C. peruvi-
anus and 27.9 6 0.5 mm in C. tetragonus) (Table 1, col-
umn 3). The only treatments that significantly altered
mean vessel diameter were those of deprivation: low wa-
ter, low nitrogen, and low phosphorus. In C. tetragonus,
low water was correlated with narrower vessels (P ,
0.05), but in plants receiving high water, vessels were not
significantly different from those of either the control or
the grand mean of all treatments. In C. peruvianus, nei-
ther water treatment produced a significant effect, even
though high water had caused most plants to elongate
greatly. Even if the aberrantly short plant no. 5 should
be removed, mean vessel diameter would not change and
would not become significantly different from the con-
trol. Low nitrogen and low phosphorus in C. peruvianus
(which caused reduced shoot elongation) induced a sig-
nificant reduction in vessel diameter relative to vessels in
the control plants and relative to the grand mean of all
treatments, but low light exerted no effect. In C. tetra-
gonus, low nitrogen and low phosphorus did not produce
any significant alteration in vessel diameter.
High levels of light, water, nitrogen, and phosphorus
were not correlated with any significant change in vessel
diameter. No treatment resulted in vessels significantly
wider than those of control plants.
Diameter of solitary vessels vs. clustered vessels—Sol-
itary vessels were significantly (P , 0.05) wider than
clustered ones in both species: in every treatment, mean
diameter of solitary vessels was wider than that of clus-
tered vessels (Table 1, columns 4 and 5). Consequently,
the effects of low nutrient treatments discussed above
could have been due to the effects on vessels in general
or due to the effects on just solitary vessels or just clus-
tered vessels or on the ratio of solitary to clustered ves-
sels. In C. tetragonus, low water treatment resulted in
narrower mean vessel diameter due to significant decreas-
es in the diameters of both solitary (P , 0.002) and clus-
tered (P , 0.02) vessels. In C. peruvianus, low nitrogen
also acted by significantly (P , 0.05) reducing mean di-
ameters of both solitary and clustered vessels. However,
the low phosphorus effect was due to a significant nar-
rowing only of clustered vessels; solitary vessels were
not significantly affected (P , 0.06). When both vessel
types were considered together (above), variability was
so great that high nitrogen treatment seemed to have no
effect, but by obtaining mean values for solitary and clus-
tered vessels separately, variability was reduced and re-
vealed that in C. tetragonus high nitrogen treatment was
correlated with a significant (P , 0.01) narrowing of sol-
itary vessels but only a moderate (P , 0.06) narrowing
of clustered ones.
Ratio of solitary to clustered vessels—The percentage
of vessels that were solitary was extremely variable with-
in each treatment in both species. In C. tetragonus, be-
tween 34 and 57% of all vessels occurred as solitary ves-
sels (Table 1, column 6), but neither was statistically sig-
nificant because of the high variability within each treat-
ment. In C. peruvianus, the high light treatment had a
significantly (P , 0.03) low percentage of solitary ves-
370 AMERICAN JOURNAL
sels (22%), but no other treatment was significantly dif-
ferent from the controls.
Vessel density—Vessel density was high in both species,
with a grand mean of all treatments of 381 vessels/mm2 in
C. peruvianus and 256 vessels/mm2 in C. tetragonus; in
most treatments, density was quite variable with standard
deviations being 10–30% of their respective means. In C.
tetragonus, mean density varied from 224 to 297 vessels/
mm2 (Table 1, column 7) but no treatment caused a signif-
icantly high or low density. In C. peruvianus, the range was
greater, from 173 to 580 vessels/mm2, the low value dif-
fered significantly (P , 0.03) from the control, and both of
these extremes occurred in light treatments: low vessel den-
sity in low light and high density in high light. High water
treatment, which had induced greatly increased shoot elon-
gation in four of five samples of C. peruvianus, did not
cause increased vessel density.
Mean amount of wood per axial bundle—Interfascic-
ular cambium develops only slowly in these species; con-
sequently during the course of the experiment all wood
was produced by fascicular cambia. Each sympodium re-
mained a discrete bundle separated from adjacent bundles
by primary rays. Mean amount of wood per bundle was
calculated in transverse section from ten bundles per
plant. Radial length of each bundle was measured from
metaxylem to vascular cambium, and width was mea-
sured at midlength. In C. tetragonus, high nitrogen (P ,
0.02) and low water (P , 0.04) were correlated with
significant decreases in the mean amount of wood pro-
duced per axial bundle (Table 1, column 8). In C. peru-
vianus the mean amount of wood per bundle was not
significantly affected by any treatment, even high water,
and even the one plant in this set that was aberrantly short
(see above) had an average amount of wood per bundle.
Estimated specific conductance—An approximate es-
timate of the potential conducting capacity per square
millimetre of wood can be obtained by multiplying mean
vessel density by mean vessel radius taken to the fourth
power (conductance is proportional to r4; Zimmerman,
1983); this is the estimated specific conductance. In C.
peruvianus, low nitrogen (P , 0.01) and low phosphorus
(P , 0.01) caused a reduction in estimated specific con-
ductance due to a significant reduction in mean vessel
diameter (see Table 1, columns 3, 7, and 9). Low light
caused a significant (P , 0.03) reduction of estimated
specific conductance due to a significant decrease in
mean vessel density (Table 1, column 7) rather than di-
ameter. A significant decrease in the estimated specific
conductance values for C. tetragonus occurred with high
nitrogen (P , 0.003), low nitrogen (P , 0.03), and low
water (P , 0.03) treatments. High phosphorus (P , 0.04)
produced a significant increase in estimated specific con-
ductance. In all cases except low water, the significant
effect on estimated specific conductance resulted from
nonsignificant changes in vessel diameter and density.
DISCUSSION
Most aspects of wood development are stable relative
to the environmental factors tested here. Although several
OF BOTANY [Vol. 86
treatments could cause increased or decreased shoot elon-
gation, only a few caused significant character changes
in wood. Even vessel diameter, the character most often
affected by these experimental treatments, was signifi-
cantly altered in only two of eight treatments in Cereus
peruvianus and one of seven in Cereus tetragonus, and
even these results must be viewed with caution: the two
species, although closely related, do not respond similar-
ly. Either the factors tested have little involvement in con-
trol of wood morphogenesis or the plants have resilient
mechanisms of homeostasis. This result was particularly
surprising because all woody plants produce vessels with
a diversity of sizes (Carlquist, 1988; Mauseth, 1988),
even cacti (Mauseth and Plemons-Rodriguez, 1997,
1998). It seems logical that environmental factors that
increase growth rate, especially an abundance of water,
should induce plants to produce broader vessels, espe-
cially because this would not involve production of a new
type of cell but rather producing more of a cell type that
it already produces (broad vessels) and producing fewer
of another (narrow vessels). This was reported for He-
lianthus and Polygonum (Penfound, 1931) and Fraxinus
(Doley and Leyton, 1968, 1970) but not in cacti. For
desert-adapted plants, water is available episodically as
brief rainy periods that alternate with long droughts, so
it may be more adaptive for their wood morphogenic
mechanisms to respond to long-term average conditions
rather than to transitory ones. If so, morphogenic mech-
anisms should be relatively self-controlling rather than be
responsive to environmental conditions.
Significant effects were seen primarily in treatments of
deprivation. The low light treatment, 10% of ambient,
almost certainly must have been a severe deprivation for
these species whose natural habitat is full-light areas.
However, none of the plants manifested signs of etiola-
tion or even slightly abnormal growth. It is difficult to
be certain of the true level of deprivation of nitrogen,
phosphorus, and water: residual amounts would have re-
mained in the plant bodies, although nutrient reserves
should have dropped to extremely low levels after a full
year of deprivation. The objective of this study was not
to study wood morphogenesis in the absolute absence of
these minerals but rather at low levels that might be en-
countered in nature. These results indicate that plants in
nature or in cultivation are unlikely to experience such
low levels of these factors that wood morphogenesis
would be significantly affected.
In the presence of high amounts of light, nitrogen,
phosphorus, and water, wood undergoes normal morpho-
genesis and develops characters almost indistinguishable
from those of control plants, that is, plants receiving
moderate amounts of these factors. Apparently once low
or moderate amounts of these factors are present, wood
develops normally.
In most woody plants, annual wood production is
closely correlated with environmental conditions. Annual
rings of trees in temperate forests are broader if produced
during favorable years than if produced during times of
stress. Many cacti, these included, produce so little wood
each year that the production of even a few more cells
should have been readily detectable.
While this study was in progress, other investigations
revealed that whereas numerous types of highly modified
March 1999] ARNOLD AND MAUSETH—EFFECTS
wood have evolved in Cactaceae (Mauseth, 1993; Mau-
seth et al., 1995; Mauseth and Plemons, 1995; Mauseth
and Plemons-Rodriguez, 1997, 1998), fibrous woods such
as those in C. peruvianus and C. tetragonus have been
extremely stable and unchanging evolutionarily (Mauseth
and Landrum, 1997; Mauseth and Plemons-Rodriguez,
1998). The fibrous woods of cacti adapted to extreme
desert conditions, to mesic habitats, and even to rain for-
ests are similar in terms of vessel diameter and density,
whereas the percentage of vessels that are solitary varies
greatly in no discernible pattern. Apparently the morpho-
genic mechanisms that control maturation of fibrous
woods are extremely stable developmentally (as demon-
strated here) and evolutionarily.
Several cactus species produce dimorphic wood, that
is, they produce one type while young and another type
while older (Mauseth and Plemons, 1995; Mauseth and
Plemons-Rodriguez, 1997, 1998). The experimental treat-
ments here did not induce any type of wood dimorphism.
Even though wood anatomical characters appear stable
despite environmental treatments, the functional capacity
of the wood, as approximated by the estimated specific
conductance, appears to change significantly in response
to some treatments. In C. peruvianus, low light levels
caused such a major reduction in vessel density that the
estimated specific conductance decreased. Similarly, low
nitrogen and low phosphorus produced a lower estimated
specific conductance by inducing the formation of nar-
rower vessels. However, other treatments reduced the es-
timated specific conductivity by inducing the formation
of vessels that were only slightly narrower and slightly
less abundant than those in control plants. If estimated
specific conductance is a reasonably good estimator of a
plant’s capacity to conduct water, then a plant might be
able to alter its hydraulic conductance by making statis-
tically imperceptible changes to several characters si-
multaneously.
From these results, it seems that if global environmen-
tal change causes the habitat of these two cacti to become
cloudier and wetter, the two species will be adversely
affected. Most individuals do not undergo altered devel-
opment that might make them more adapted to the new
conditions. Assuming that their bodies are well-adapted
for current climatic conditions, as the environment chang-
es, they will continue to produce the same type of body,
even though it is not adapted to the new conditions. Some
individuals (such as sample no. 5) would possibly suc-
cumb quickly, altering the gene pool.
LITERATURE CITED
BISSING, D. R. 1982. Variation in qualitative anatomical features of the
xylem of selected dicotyledonous woods in relation to water avail-
ability. Bulletin of the Torrey Botanical Club 109: 371–384.
CARLQUIST, S. 1988. Comparative wood anatomy. Springer-Verlag,
Berlin.
DOLEY, D., AND L. LEYTON. 1968. Effects of growth regulating sub-
stances and water potential on the development of secondary xylem
in Fraxinus. New Phytologist 67: 579–594.
———, AND ———. 1970. Effects of growth regulating substances
OF ENVIRONMENT ON WOOD ANATOMY 371
and water potential on the development of wound callus in Frax-
inus. New Phytologist 69: 87–102.
DREW, M. C., M. B. JACKSON, AND S. GIFFORD. 1979. Ethylene-pro-
moted adventitious rooting and development of cortical air spaces
(aerenchyma) in roots may be adaptive responses to flooding in
Zea mays L. Planta 147: 83–88.
ETHERINGTON, J. R. 1984. Comparative studies of plant growth and
distribution in relation to water logging. Journal of Ecology 72:
389–404.
GORYSHINA, T. K. 1980. Structural and functional features of the leaf
assimilatory apparatus in plants of a forest-steppe oakwood. I. Leaf
plastid apparatus in plants of various forest strata. Acta Oecologica
1: 47–54.
———. 1989. Fotosinteticheskii apparat rasteni I usloviya sredy. Iz-
datel’stvo Leningrad University, Leningrad.
HEATHCOTE, C. A., M. S. DAVIES, AND J. R. ETHERINGTON. 1987. Phe-
notypic flexibility of Carex flacca Schreb. New Phytologist 105:
381–391.
JACKSON, R. B., J. H. MANWARING, AND M. M. CALDWELL. 1990. Rapid
physiological adjustment of roots to localized soil enrichment. Na-
ture 344: 58–60.
KAWASE, M. 1979. Role of cellulase in aerenchyma development in
sunflower. American Journal of Botany 66: 183–190.
LARCHER, W. 1995. Physiological plant ecology. Springer-Verlag, Ber-
lin.
MAUSETH, J. D. 1988. Plant anatomy. Benjamin/Cummings, Redwood
City, CA.
———. 1993. Water-storing and cavitation-preventing adaptations in
wood of cacti. Annals of Botany 72: 81–89.
———. 1996. Comparative anatomy of tribes Cereeae and Brownin-
gieae (Cactaceae). Bradleya 14: 66–81.
———, AND J. V. LANDRUM. 1997. Relictual vegetative anatomical
characters in Cactaceae: the genus Pereskia. Journal of Plant Re-
search 110: 55–64.
———, G. MONTENEGRO, AND A. M. WALCKOWIAK. 1984. Studies of
the holoparasite Tristerex aphyllus (Loranthaceae) infecting Tri-
chocereus chilensis (Cactaceae). Canadian Journal of Botany 62:
847–857.
———, AND B. J. PLEMONS. 1995. Developmentally variable, poly-
morphic woods in cacti. American Journal of Botany 82: 1199–
1205.
———, AND B. J. PLEMONS-RODRIGUEZ. 1997. Presence of paratracheal
water storage tissue does not alter vessel characters in cactus wood.
American Journal of Botany 84: 815–822.
———, AND ———. 1998. Evolution of extreme xeromorphic char-
acters in wood: a study of nine evolutionary lines in Cactaceae.
American Journal of Botany 85: 209–218.
———, Y. UOZUMI, B. J. PLEMONS, AND J. V. LANDRUM. 1995. Struc-
tural and systematic study of an unusual tracheid type in cacti.
Journal of Plant Research 108: 517–526.
NOBEL. P. S. 1988. Environmental biology of agaves and cacti. Cam-
bridge University Press, New York, NY.
———. 1996. Shading, osmoticum, and hormone effects on organ de-
velopment for detached cladodes of Opuntia ficus-indica. Interna-
tional Journal of Plant Science 157: 722–728.
PENFOUND, W. T. 1931. Plant anatomy as conditioned by light intensity
and soil moisture. American Journal of Botany 18: 558–572.
RASCHI, A., F. MIGLIETTA, R. TOGNETTI, AND P. R. VAN GARDINGEN [EDS.]
1997. Plant responses to Elevated CO2 evidence from natural
springs. Cambridge University Press, New York, NY
SAS. 1996. SAS for windows v6.11. SAS Institute, Cary, NC.
TSELNIKER, Y. 1978. Fiziologicheskie osnovy tenevynoslivosti dreves-
nykh rastenii. Nauka, Moskva.
VAN ARENDONK, J. J. C. M., G. J. NIEMANN, J. J. BOON, AND H. LAMB-
ERS. 1997. Effects of nitrogen supply on the anatomy and chemical
composition of leaves of four grass species belonging to the genus
Poa, as determined by image-processing analysis and pyrolysis-
mass spectrometry. Plant, Cell and Environment 20: 881–897.
ZIMMERMAN, M. H. 1983. Xylem structure and the ascent of sap.
Springer-Verlag, Berlin.