water used from storage was much greater and increased as the
season proceeded (from about 35 to 55 dm
3
). Most of the
stored water in the study tree was located in the relatively rigid
stem sapwood below 46 m. Water released by cavitation of
vascular elements may prevent desiccation of leaves and other
living tissues (Dixon et al. 1984, Tyree and Yang 1990) but the
loss of hydraulic conductivity due to cavitation may result in
further decreases in water potential leading to runaway xylem
cavitation (Sperry 1995). However, a balance seems to exist
between use of water from cavitated elements and loss of hy-
draulic conductivity due to cavitation (Sellin 1991a, Èermák
and Nadezhdina 1998, Sperry et al. 1998, Domec and Gartner
2001).
There is increasing evidence that in large trees sapwood hy-
draulic capacity maybe vastly more than sufficient to meet
transpiration needs under favorable conditions. In Quercus
robur L. and
Laurus azorica (Seub.) Franco, it was found that
only about 2% of all stem xylem conducting elements were
theoretically needed to supply water for rapid transpiration
(Krejzar and Kravka 1998, Èermák et al. 2001 and Morales et
al. 2001), whereas almost 100% of all conducting xylem ele-
ments must function in petioles. Theoretically, the majority of
conducting elements in stems could be embolized without sig-
nificantly affecting stem hydraulic conductivity. For Laurus
azorica, stem vessels represented the largest store of free wa-
ter. Because the sapwood does not conduct water uniformly
with depth (Swanson 1971, Èermák et al. 1984, 1992, 2004,
Phillips et al. 1996, Èermák and Nadezhdina 1998, Jimenez et
al. 2000, Nadezhdina et al. 2001), and does not dehydrate uni-
formly with depth (Èermák and Nadezhdina 1998) it should
become a more important source of water as sapwood depth
and volume increase. Under such circumstances, the less-con-
ducting part of the sapwood can serve as a source of stored wa-
ter, while having a minimal impact on hydraulic conductivity.
If such stored water is used more than once, cavitated elements
must be refilled. Domec and Gartner (2002) suggested that the
latewood, because of the smaller diameter of its conductive el-
ements and is greater resistances to flow, may cavitate first and
provide water to the transpiration stream. Loss of latewood
would not severely impact whole sapwood water conduction.
Waring and Running (1978) hypothesized that refilling of
cavitated xylem conduits occurs over winter. However, more
recent evidence suggests that refilling of cavitated elements
may occur diurnally (Zwieniecki and Holbrook 1998, Hol-
brook et al. 2002). Irvine and Grace (1997) noted a linear rela-
tionship between xylem dimension and xylem water potential,
suggesting water loss of individual xylem elements and di-
mensional changes as a result of the loss. They hypothesized
that the conducting xylem could lose volume without cavi-
tating. Second, their study suggests that water can be released
in this way within the normal range in plant water potentials.
Recently, using snap-freezing of roots and stems and cryo-
scanning electromicroscopy, Shane and McCully (1999) and
McCully (1998, 1999) observed that as many as 60% of the
stem or root conducting vessels may be cavitated. They also
observed rapid refilling of cavitated vessels, thereby offering a
much more dynamic view of the role and behavior of water in
conducting elements. Both the refilling of cavitated elements
and the release of water by conducting elements without cavi-
tation needs further study.
194
ÈERMÁK, KUÈERA, BAUERLE, PHILLIPS AND HINCKLEY
TREE PHYSIOLOGY VOLUME 27, 2007
Table 4. Comparison of estimates of water used from storage in various tree species.
Species
Height (m)
Age (year)
Quantity (dm
3
day
–1
) (%)
Reference
Acer saccharum Marsh.
> 9
30–100
Not given (17%)
Tyree et al. 1991
Anacardium excelsum (Bertero and Balb. ex. Kunth) Skeels.
35
Mature
54 (14%)
Goldstein et al. 1998
Carya illinoensis Wangenh.
4
5
4 (3%)
Steinberg et al. 1990
Cecropia longipes Pittier.
18
Mature
4.0 (9%)
Goldstein et al. 1998
Ficus insipida Willd.
30
Mature
25 (15%)
Goldstein et al. 1998
Larix decidua Mill.
18
70
18 (4%)
Schulze et a l. 1985
Luehea seemannii Triana.
29
Mature
16 (12%)
Goldstein et al. 1998
Malus pumila Mill.
2.5
9
0.7 (20%)
1
Landsberg et al. 1976
Nothofagus fusca (Hook F.) Oerst.
34
300–400
5–10 (4–8%)
Köstner et al. 1992
Picea abies (L.) Karst.
30
80
9 (14%)
Schulze et al. 1985
Picea abies (L.) Karst.
0.6–1.2
4–6
(2–15%)
Zweifel et al. 2000
Pinus maritima Mill.
24
64
10–13
Loustau et al. 199
Pinus sylvestris L.
15
41
20–30 (30–50%)
Waring et al. 1979
Prunus avium L.
6
15
1 (min 4%)
Èermák et al. 1976
Pseudotsuga m.1373
57
450+
40–70 (20–30%)
This study
Pseudotsuga m. 091
65
450+
22–74 (29–17%)
Phillips et al. 2003
Pseudotsuga menziesii (Mirb.) Franco.
19,15, 6
40
1.8, 1.0, 0.1 (5%)
Lassoie 1979
Quercus robur L.
32
100
10–31 (15–22%)
Èermák et al. 1982
Salix fragilis L.
10
30
3% (1% stem vol)
Èermák et al. 1984
Schefflera morototoni (Aubl.) Maguire, Steyerm. and Frodin.
20
Mature
0.9 (2.5%)
Tyree et al. 1991
Spondias mombin L.
23
Mature
8.7 (11%)
Goldstein et al. 1998
Thuja occidentalis L.
10
Not given
2.5 (22%)
Tyree 1988
1
Estimated as 2 h of transpiration divided by a nominal 10 h.
Downloaded from https://academic.oup.com/treephys/article-abstract/27/2/181/1664618
by guest
on 25 July 2018