CC BY
14
АРИДНЫЕ ЭКОСИСТЕМЫ, 2003, том 9, № 18
================== ОТРАСЛЕВЫЕ ПРОБЛЕМЫ ОСВОЕНИЯ ===============
ЗАСУШЛИВЫХ ЗЕМЕЛЬ
УДК 599. 32 + 582. 734. 6
РОЛЬ ГРЫЗУНОВ В РАСПРОСТРАНЕНИИ КОСТОЧЕК ДИКОГО АБРИКОСА PRUNUS ARMENIACA
© 2003 г. Чжан Чжибинь
Институт зоологии Академии наук Китая. 100080, Пекин, Китай
Гибель семенного фонда в лесах часто связывают с отрицательным влиянием грызунов. Тем не менее роль грызунов в процессе лесовозобновления остается не до конца понятной из-за недостатка более детальной информации о количестве семян, изъятых грызунами. В данной работе с использованием нового метода мечения косточек (с помощью небольших оловянных меток с нанесенным на них кодом) были исследованы особенности рассеивания и сохранения косточек дикого абрикоса грызунами в гористом районе (40°00' с.ш., 115°30' в.д.) близ Пекина. Для этого эксперимента 10—15 июня 1998 г. были собраны целые созревшие косточки дикого абрикоса. Был заложен трансект (длиной 240 метров), пересекающий 4 склона, и вдоль него на расстоянии 10 м друг от друга были расположены 24 площадки. 19 — 20 июня, 3 июля и 23 октября 1998 г. косточки с прикрепленными к ним оловянными метками в количестве 1440 были разбросаны в пределах этих площадок. После этого по обе стороны трансекта на ширину до 50 м было проведено тщательное обследование но квадратам 2 х 2 м2 для того, чтобы обнаружить помеченные косточки или их фрагменты. Большинство помеченных косточек или их фрагменты были обнаружены на удалении до 20 м от площадок на трансекте. Большая часть помеченных косточек или их фрагменты оказались рассеянными в направлении центра и нижней части склона. Благодаря своим поведенческим особенностям грызуны делали запасы из косточек. Они предпочитали зарывать косточки в землю в траве на открытых участках, а поедали их в укрытиях в густом кустарнике. Как выяснилось, наличие травянистого покрова положительно сказывается на процессе прорастания абрикоса из косточек, зарытых в землю на глубину 5 см, в то время как, корреляция с густотой кустарникового яруса была отрицательная. Закапывание грызунами косточек в землю на открытых травянистых участках способствовало возобновлению дикого абрикоса, что свидетельствует о существовании взаимной коэволюции между абрикосом и грызунами. Высказано также предположение, что поедание грызунами косточек дикого абрикоса в укрытии под кустами помогает им избежать хищников.
HOW RODENTS PLANT SEEDS OF WILD APRICOT PRUNUS ARMENIACA
© 2003. ZHIBIN ZHANG
Institute of Zoology, Chinese Academy of Sciences Beijing 100080, P.R China
Prior studies suggest that rodents are significantly associated with the seed-fate in many forests. Seeds falien or placed on the soil surface disappear very quickly due to the removal by vertebrates,especially by rodents (Shaw, 1968 a; 1968 b; Kanazawa, Nishikata, 1976; Kikuzawa, 1988; Herrera, 1995). Rodents consumed most seeds and few of them can become seedlings (Sork, 1984; Miyaki, Kikuzawa, 1988; Herrera, 1995). Rodents are also recognized as an important agent for the regeneration of some forests because they disperse and bury seeds (Jensen,Nielsen, 1986).
Although it is relatively easy to quantify seed removal', it is difficult to know what happens to all the seeds that are removed (Sork, 1983; 1984). Therefore, the role of rodents in forest regeneration has not been well understood due to lack of more detailed information on the seeds removed by rodents. Recently, some useful techniques have been employed to study seed-fate by labeling experimental seeds with radioactive (Jensen, Nielsen, 1986; Vander Wall, 1997), fluorescent pigments (Longland, Clements,
1995), metals (Sork, 1984) and threads (Forget, Milleron, 1991). These methods provided more information on the survival, dispersal, food cache and microenvironments of the tagged seeds. The disadvantage of the present techniques is that the tagged seeds are usually not distinguished from each other. In this study, we used anew method to study the fate of a wild apricot nearby Beijing, China by labeling seeds with small pieces of coded tin-tags.
The low forests or shrubs of wild apricot are widely distributed in the mountainous areas at latitudes of 700-1200 meters in the suburb of Beijing, China (Chen, 1997). It is atypical pioneer plant species, which easily invades sunny deforested areas of dry and poor soil, and avoids the shade or high-density forest canopy. The seed of apricot weighs about 1.47 g (n = 112).The seed cover is very hard and only rodents are able to open it. The seeds of apricot ripe in summer, usually in mid-June. New seedlings will sprout out in early May of the following year. Previous observation indicated that regeneration of apricot depended heavily on rodent burying because surface-sown seeds of apricot germinated poorly (Zhang et al.,1998).But it is little known how rodents disperse and preserve the seeds.
This study, by tracking the tin-tagged seeds of apricot, aims to investigate (1) the dispersal distance and pattern of the tagged seeds; (2) the microenvironment of the seeds or their fragments; and (3) the size of food cache.
METHODS
Study site
The experiment was carried out in a mountainous area nearby the Liyuanling Village, Qijiazhuang County, Mentougou District of Beijing (fig. 1). The study site is located at a latitude and
Fig. 1 The study site and illustration of the 24 plots (A-F, 1-18) along a transect crossing 4 slopes (25-35°) for seed release test. Рис. 1. Обследованная территория и положение 24 участков (A-F, 1-18) вдоль трансекта, пересекающего 4 склона (25-35°),на которых проводилась оценка рассеивания косточек.
longitude of N 40°00',Е 115°30',an elevation of 1200 m, about 120-km northwest of Beijing. The villagers immigrated out 12 years ago under the help of the Beijing government for the nature conservation and poverty-alleviation of mountainous residents. The study area is very degenerated due to extensive cutting and goat grazing for almost a century. Oak (Q. Liaotungensis), wild walnut (Juglans mandshurica), wild apricot (Prunus armeniaca), Vitex negundo and Prunus davidiana shrubs are
commonly found. Under shrublands, Elymus excelsus, Poa spp., Elsholtzia stauntoni are common grasses. Laxix principis-rupprechtii and Pinus tabulaeformis are planted trees by local forestation farm in small areas.
Seed-release test
Intact and ripe seeds of wild apricot were collected on June 10-15,1998 for seed-release test.Tiny holes were drilled at the bottom sides of seeds without destroying the inside kernels, and tied with small and light tags about 4-cm long and 1-cm wide by using thin metal-strings of 3-cm long. The tags are made of Coke tin and coded by using asharpen metal-pen or hard printing blocks with numbers.The tin-tagged seeds are easy to find after being dispersed by rodents. If buried in soil by rodents, the tin-tags are left on surface.
A 240-m transect across 4 slopes was located for seed-release (fig. 1.). Along this transect, 24 plots (Plot A-F,Plot 1-18) were located with 10-m apart. A total of 1440 seeds were released at the 24 plots on June 19, July 3,and October 23,1998. After seed-placement, both sides of the transect of 50-m wide were extensively checked by scanning every quadrate of 2 x 2 square meters with roughly equal efforts to find the tagged seeds or their fragments. The checking dates were on June 21-22, June 26-27, July 4-5,July 11-12,October 24-25.October 31 and November 1 of 1998,and May 15-16, June 20-21, July 4-5 of 1999 respectively.
Four categories of seed-states were defined for the tin-tagged seeds or their fragments (fig. 2a):
Up
SE SU - SU ■ SI ■ SO NG -> G
Fig. 2. Definition of slope-directions, microenvironments and seed states, (a) seed states, (b) slope-directions to which the tagged seeds or their fragments were dispersed, (c) microenvironments of seeds or their fragments. Рис. 2. Определение направлений на склоне, особенностей мест нахождения косточек или их фрагментов и положения косточек: а) положение косточек; Ь) направление на склоне, по которому были рассеяны помеченные косточки; с) места нахождения косточек или их фрагментов.
Intact and buried (IB).The tagged seed is intact and buried in soil with tin-tag out of the soil.
фрагментов.
Intact and on surface (IS).The seed is intact and littered on surface.
Eaten up ^.There is agnawing hole opened by arodent on the tagged seed. The inside kernel of the tagged seed was taken away. Only the seed cover is littered on surface with tin-tag attached on.
Cut off (C).The seed is cut off from the tin-tag. Only the tin-tag is littered on surface. The seed is gone and its fate is unclear.
The distance of the tagged seeds or their fragments to their original releasing plots was measured. Five categories of slope directions of seeds or their fragments to their original releasing plots were defined and shown in fig.2b.
The five slope direction, including Up Slope (Up), Down Slope (Down),Middle Slope (Middle), Up-Middle Slope (Up-Middle), Down-Middle Slope (Up-Middle),Up/Down, were identified according to the area in which seeds were located. Up/Down means the seeds or their fragments were moved across the top of a hill or a valley.
Four categories of microenvironments of the tagged seeds or their fragments were defined as follows (fig.2c):
Naked ground (NG). The tagged seeds or their fragments were buried in or littered on naked ground.
Grass (G). The tagged seeds or their fragments were buried in or littered on open grasses.
Under shrubs (SU).The tagged seeds or their fragments were buried in or littered on ground under shrub cover.
Shrub edge (SE).The tagged seeds or their fragments were buried in or littered on ground under the edge of shrub cover.
The surrounding environments of the tagged seeds or their fragments within 1 square meter were recorded. The environmental criteria included grass cover, grass height, shrub cover and shrub height.
Statistics
SPSS for Windows was employed for statistic analysis. Non-parametric Chi-square test was used for identifying if four different seed states (IB, IS, C, and E) were distributed evenly at five different slope-directions (UP, Down, Middle, Up-Middle, Down-Middle) or four categories of microenvironments (G, NG,SE and SU). Pearson correlation test was used for identifying the significance of correlation between seedling numbers or heights of apricot and grass or shrub cover or height.
RESULTS
Dispersal distance
Fig.3 showed that the dispersal-distances of seeds by rodents were not very far. Most of them were found within 20 meters from their original plots. A large proportion of the tagged seeds was found within 5 meters. The mean dispersal distance of tagged seeds or their fragments were given in table 1. The
80
70
Sh Г>0
Ъ
п 50
о 10
:л
а> 30
Е 20
К)
0
□ Ili
■ F,
□ IS
Л
П
30. I
35. 0
\ •>!;
10. I 15. 1 20. I 25. 1 15.0 20.0 25.0 30.0
I) i si ¿nice group fut)
Fig. 3. Dispersal distance of the tagged seeds or their fragments from plots where they were released.
Рис. 3. Расстояния между местами нахождения помеченных косточек или их фрагментов, рассеянных
грызунами,и теми участками,где они были разложены.
dispersal distances of seed fragments С and seeds IS were obviously shorter than that of seeds or fragments of IB and E.
Table 1. The dispersal distance (m) of the tagged seeds or their fragments (IB,IS,E,C) by rodents.
Таблица 1. Расстояние между помеченными косточками или их фрагментами, рассеянными грызунами (IB,IS.
Е,С*).
Seed state Состояние и положение косточек Mean Среднее расстояние Std Dev. Стандартное отклонение Minimum Минимальное расстояние Maximum Максимальное отклонение N число
IB 10.97 17.38 0.30 ПО 114
IS 6.9 8.37 0.00 45 31
Е 12.53 39.50 0.20 250 39
С 5.62 15.92 0.00 150 97
* IB (intact and buried/ целые и закопаны) — помеченные косточки не повреждены и остались в земле, а метка на поверхности; IS (intact and on surface/ целые и на поверхности) — помеченные косточки не повреждены, но оказались разбросаны на поверхности земли; Е (eaten up/ съеденные) - ядра косточек были съедены грызунами, и только скорлупки, с прикрепленными к ним метками, остались разбросанными на поверхности земли; С (cut off/ отсоединенные) — были обнаружены только метки, которые оказались отсоединены от косточек, косточки были потеряны.
Slope-directions
The tagged seeds or their fragments were not evenly distributed in the five slope directions (for C, X2 = 34.437,df - 4,/J = 0.000; for IB,X2 = 28.569, df = 4,p = 0.000; for IS,X2 = ] 5.929, df = 4, p = 0.03) except for С with an approximate significance level (X2 = 8.5, df = 4, p = 0.075). As shown in Fig.4, rodents preferred to move the tagged seeds to the directions of Down-slope or Middle-slope.
Slope direction
Fig. 4. Proportion of the tagged seeds or their fragments dispersed to different slope-directions from plots where they were released. Рис. 4. Соотношение между количеством помеченных косточек или их фрагментов, рассеянных по различным направлениям на склоне, и их количеством в тех местах, где они были разложены.
Cache size and cache line It was found that most seeds or their fragments were preserved singly by rodents (fig.5, table 2).
Fig. 5. The frequency of the cache size (numbers of seeds or their fragments were preserved together). IB+C, IB+IS, and IB+IS+E indicate that different seed states (e.g.IB,E,C and IS) were found in one food cache. Рис. 5. Частота встреч сделанных грызунами скрытых запасов разных размеров (число косточек или их фрагментов,спрятанных в одном месте). IB+C, IB+IS, IB+IS+E сочетания косточек, их фрагментов и меток, найденных в одном месте.
Table 2. The frequency of seed cache size of rodents. IB+C, IB+IS and IB+IS+E indicate the food cache is comprised of more than two seed states. Таблица 2. Частота, с которой встречаются сделанные грызунами скрытые запасы косточек разных размеров. IB+C, IB+IS, IB+IS+E показывают запасы, состоящие из различных сочетаний косточек,их фрагментов и меток.
Seed state Состояние и положение косточек
1
Cache size Размер скрытых запасов
3
4
10
IB
IS
Е
С
IB+C
IB bis
IB+IS+E
101
30 33 74
2
1 1 1
1* 1*
2
5
1
1
Seeds or fragments coming from different plots.
Косточки или их фрагменты с разных участков площадки наблюдения.
For larger food cache size,seeds usually were taken from different plots (table 2). Nine food cache lines, where seeds were preserved in aroughly direct line,were discovered (table 3).Most cache lines were in the direction of Down-slope or Middle-slope. Seeds of some cache lines were taken from several different plots.
Microenvironments of seeds or their fragments
Seeds or fragments were not evenly distributed in microenvironments of G,NG,SE and SU (for IB, X
= 108.808,df = 3,p = 0.000; for C, X2 = 12.122,df - 3,p = 0.007; for E, X1 = 46.512,df = 3,p =
0.000) except for IS (X = 1.645,df = 3,p ~ 0.649). As shown in Fig. 6, most of IB were preserved in open grasses (G),and very few were found under shrub cover (SU).In contrast,most of E was found on surface under shrub cover (SU) and very few of them were found in the other microenvironments (i.e. NG,SE,G). There was slightly higher probability of С being found at G and NG.There was not much difference of preserving conditions for IS.
Table 3. The aspects of nine cache lines observed. Numbers in parenthesis indicate the numbers of seeds with different seed states (IB,IS,E,C) or coming from different plots (A-FJ-18).
Таблица 3. Характеристики сделанных грызунами скрытых запасов на 9 исследованных трансектах. В скобках указан размер сочетаний косточек, их фрагментов и меток (IB, IS,E,C) на разных участках (A-F, 1-18).
Line No. Numbers of seeds Length (m) Slope Seed states Plots
Номер Количество Протяженность Положение Состояние и Участки
трансекта косточек трансекта на склоне положение косточек
1 12 20 Down IB(10),C(2) 16(8), 17(4)
2 4 4.5 Down Ш(4) 15(4)
3 5 8.4 Middle C(4),IB(1) 18(5)
4 4 2.9 Down-Middle IB(4) 10(4)
5 7 15.4 Down E(3),C(3),IB(1) B(5),A(I),!(!)
6 4 8.4 Up/Down IB(4) 11 (2), 12(2)
7 5 14.2 Down IB(5) 13(5)
8 5 14.7 Up IB(4),C(1) 14(3),4(2)
9 7 13 Middle IB(7) 3(5),2(2)
Down — в нижней. Middle — в средней, Down-Middle — ниже средней, Up — в верхней части склона, Up/Down означает, что косточки или их фрагменты были перенесены через вершину холма или через долину.
Table 4 The surrounding environmental characteristics (Mcan+SD) of the tagged seeds or fragments (TB,E, IS,C). Таблица 4. Характеристики мест нахождения помеченных косточек или их фрагментов (IB, IS,E,C).
IB Е IS С
Grass Cover (%) 61.32 29.62 29.5 35.20
Травяной покров (%) ±30.85 ±29.83 ±34.11 ±32.07
Grass Height (cm) 15.05 10.46 11.82 10.38
Высота травяного покрова (см) ± 10.64 ±7.90 ±8.96 ±8.64
Shrub Cover (%) 30.44 58.28 38.93 27.84
Заросли кустарников (%) ±27.29 ±30.24 ±27.67 ±29.76
Shrub Height (cm) 56.62 78.77 56.25 60
Высота кустарников (см) ±65.29 ±62.82 ±44.34 ±70.7
N= НО 39 28 74
Table 5 Correlation coefficients between seedling numbers or height of apricot and the surrounding vegetation (grass and shrub). Таблица 5. Коэффициенты корреляции между числом или высотой проростков абрикоса и окружающей растительностью (трава и кустарники).
Grass Grass Shrub Shrub
Cover Height Cover Height
(%) (cm) (%) (cm)
Ехр.1 Seedling numbers Число проростков (см) R .621 .171 -.645 -.205
Эксперимент 1 Р ,055# .636 .044* .571
N 10 10 10 10
Seedling height (cm) Высота проростков (см) R .159 .592 -.206 -.460
Р .660 .071# .567 .181
N 10 10 10 10
Ехр.П Seedling numbers Число проростков R .592 .117 -.369 .003
Эксперимент 2 Р .043* .718 .238 .992
N 12 12 12 12
Seedling height (cm) Высота проростков (см) R .842 .560 -.155 .377
Р .001" .058# .631 .227
N 12 12 12 12
* Correlation is significant at the 0.05 level (2-tailed).** Correlation is significant at the 0.01 level (2-tailed).
# Correlation is approximately significant at the 0.05 level (2-tailed).
* Корреляция значительна на уровне 0,05 (2-х сторонний тест);
** Корреляция значительна на уровне 0,01 (2-х сторонний тест);
# Корреляция приблизительно соответствует уровню 0,05 (2-х сторонний тест).
-SO
a is □ с a i;
■ i н
m
si;
si:
Micro environments
Fig. 6. Proportion of different seed states (i.e. IB, IS,C,E) were preserved in different microenvironments (i.e.G,NG, SE,SU).Pиc. 6. Соотношение косточек,их фрагментов и меток в разном состоянии и положении (IB,IS,E,C), спрятанных в различных местообитаниях (G,NG,SE,SU). Помеченные косточки или их фрагменты были закопаны или разбросаны в/на:
G (grass) — открытые участки с травяным покрытием; NG (naked ground) — голая земля; SU (under shrubs) - под кустами; SE (shrub edge) - на границе с зарослями кустарников.
The surrounding microenvironments of IB, E, IS and С were given in table 4. This result further supported the above conclusion that the tagged seeds were buried in the microenvironments with more grasses and less shrubs, while they were often eaten up under dense shrubs.
DISCUSSION
The home range of field mouse (A. Speciosus) was estimated to be 341-1620 square meters with maximum moving distance of 35-m by using radio telemetry method (Kikuzawa, 1988). W.P. Xia and Z.Long (1978) reported that the home ranges of male and female striped mouse (A. agrarius) were 1034 ± 70.1 and 769.1 ± 56.9 square meters, and the moving distances for male and female were 53.4 ± 2.4 meters and 45.4 ± 2.6 meters respectively when population density was 28.9 mice/ha. S.J. Yang and G.Y.Zhu (1989) reported that the home ranges of male and female striped mouse were 2271 ± 204 and 1841 ± 183 square meters, and the moving distances for male and female were 88.4 ± 4.9 meters and 82.1 ±5.1 meters respectively when population density was lower (10.1 mice/ha) than that reported by W.P.Xiaand Z.Long (1978).The home range and moving distance of white-bellied rat should be larger than that of mice because its body mass is much bigger than that of these mouse species. The dispersal distances of the tagged seeds were obviously shorter than the normal dispersal distance of the dominant rodent species. This implied that rodents did not take foods to their original nests. This also conforms to many other observations that rodent preserved seeds in very short distance (Vander Wall, Balda, 1977).
The reason why rodents tended to preserve seeds to the down-slope or middle-slope directions might be because this would save energy when hoarding seeds. Many studies reported that animals tended to preserve food with minimum expenditure of energy (Stapanion, Smith, 1984; Elliot, 1988; Clarkson et al., 1986). However, there was still a relatively small proportion of the tagged-seeds being dispersed to the up-slope direction although it would take rodents more energy.
There are two different strategies of preserving foods by animals: scatter hoarding and larder hoarding (Hurly, Robertson, 1987). Some animals (e.g. Dipdomys indens, Eatamias spp.) can shift between these two strategies depending on the environments (see review; Jiang, 1996a). In this study, rodents preserved most of seeds by scatter hoarding. The advantage of scatter hoarding lies in reducing the risk of being discovered by other rodent individuals. No single buried-seed was found to move again by rodents to the other places through out of this study, which supports this hypothesis. Unless protected, seeds preserved by larder hoarding are more easily lost than seeds preserved by scatter hoarding. Herrera( 1995) reported that single acorns of cork oak (Querqus suber L.) experienced lower predation and had relatively higher emergence ffl km clumped acorns. The survival of buried acorns was negatively related to acorn density
Dispersal
Food cache size
due to more predation pressure on high density nuts (Stapanian, Smith, 1984). However, some chipmunks, sqoinefe. and birds prefer to preserve seeds in clumps (Jiang, 1996b). These clumped seeds were usually preserved nearby the nests of animals and well protected by the host animals.Z.G.Jiang (1996b) reported that red squirrels preserve seeds of pine in clumps in Canada, and the frequency increased when intruder approaching their food caches.
S.B. Vander Wall (1912) proposed some hypothesis on how animals relocated their preserved foods. Birds mainly rely on vision and cache site memory to find their preserved foods (Cowie et al., 1981; Vander Wall. 1990).while mammals relocate their cached foods mainly relying on odor. Rodents are very capable of finding the buried seeds of oak and apricot, especially when seeds were buried in clumps (Zhang et al.. 1998). Hw er.cache site memory could also be important for rodents to relocate their cached foods (Jiang. 1996c) Burying seeds by scatter hoarding in lines could reduce predation by non-host animals, while help host animals to relocate the preserved seeds by using site memory. These food cache lines might also be formed through rodents burying seeds along their runways if these runways were roughly straight. T.S.Jensen and O.F.Nielsen (1986) reported that acorns of (Q. robur, Q. petrea) were found in the walls or on the floor of rodent runways. However,seeds in some food cache lines were from several different plots in this study, which implied that rodents did not drop seeds along their runways simply.They might need to relocate their runways after carrying seeds from several different sites before putting seeds along their runways.This might had involved cache site memory.
Microenvironments of the tagged seeds or their fragments
There were very few studies on microenvironments of seeds buried by rodents. T.S. Jensen and O.F.Nielsen (1986) reported that radioactive acorns were preferably deposited under Empetrum nigrum mats in the walls of runways because Empetrum nigrum provided space for rodent runways. Our study clearly indicated that the tagged seeds of apricot were buried mostly in open grasses,while eaten-up seeds were mostly found under shrub covers. Here, we propose a hypothesis to explain this phenomenon: seeds buried in open grasses would have more chance of germination, and eating seeds under dense shrubs would reduce risk of predation. Since apricot is highly relying upon sunlight, seeds of apricot buried in open grasses with good sunlight would more likely germinate.The first part of the hypothesis is supported by results of two experiments (table 5,fig. 7),which were carried out in 1997-98 for a different purpose. In the Experiment-I, seedling numbers of apricot was positively correlated with grass cover with an approximate significance level (r= 0.621,p = 0.055, n = 10),but negatively correlated with shrub cover (r = -0.645, p = 0.044, n = 10).Seedling height of apricot was positively correlated with grass height with an approximate significance level (r= 0.592, p = 0.071, n — 10).In the Experiment-II both seedling numbers and height of apricot were positively correlated to grass cover (for seedling numbers ,r = 0.592, p = 0.043, n = 12; for seedling height. r- = 0.842, p = 0.001, n = 12). Seedling height of apricot was positively correlated to grass height with an approximate significance level (r= 0.560, p = 0.058, n = 12).These two experiments clearly indicated that seeds of apricot germinated better in open grasses than in dense shrubs. Therefore, we further hypothesized that this kind of seed hoarding behavior might be the result of mutual co-evolution between rodents and apricots. Rodents plant seeds of apricot for food, and apricot provide food to rodents for expanding distribution and improving regeneration.
Eating seeds under dense shrub covers is obviously an evolutionary strategy of reducing risk of predation because dense ground vegetation usually provides better cover and shelter for rodents and other seed predators. Cover is an important factor favoring foraging activities of mice (Kondo 1980). Increasing plant cover and appearance of rodents (including field mouse and striped field mouse) is usually positively correlated (Kikuzawa, 1988; Wada, 1993). Acorns of Quercus mongolica usually disappear more rapidly from sites with ground cover than barren sites (Kikuzawa, 1988). Acorn predation and hoarding by mice is generally highest under shrubs and in non-mowed grasslands (Kollmann, Schill, 1996). However, J. Herrera (1995) reported no acorn predation under dense shrub, but 52% predation under open shrub. Observations of rodent occurrence differing among microenvironments as a function of their predator behavior are frequently reported (Gill, Marks 1991; Wada, 1993; Schupp, 1995; Wilson, Whelan, 1990; Reichman, 1981). Mice populations are probably higher in tall-grass rather than short-grass fields because of better hiding cover from avian predators (Kollman, Schill, 1996). Since the seed cover of apricot is very hard, rodents have to spend a longer time to open it. Therefore, they need to find safer places to consume the hard seeds of high nutrition. Unlike acorns of oak, none single tagged seed of apricot was consumed at the plots where they were released. Therefore, eating up seeds by rodents under dense shrub cover was likely an evolutionary strategy of avoiding predation.
1»
f К %
V
л
10 га ni <о ы Ground ЕТ^З сот* г (%) № 1п В 0
и IA и 1»
О
о О
§ с ы О О
о
о L4 -----о о
о
о
У 4D 60 1»
Cri'ini irpf« i L1V г- г (ъ) (с)
ю
„ 8 Ьа
V
1 6 С
«с
S 4
тЭ
V f
<Л L
о
20
<10 60
Shrub cover CI)
so
j
J00 (b)
flrriund (ГНК* СОУОГ i*)
№
Fig, 7. The relationship between seedling numbers or height of wild apricot and grass or shrub vegetation. Seeds of apricot were buried 5-cm deep in soil on October 20,1997, and seedlings were counted and measured on May 20, 1998. In the Experiment-1, ten seeds were buried 5-cm deep at each of 10 plots. Seedling numbers of apricot were positively related to grass cover (a), but negatively to shrub cover (b). In the Experiment-11, twenty-five seeds were buried at each of the 12 plots. Both seedling numbers and height of apricot were positively related to grass covers (c, ё).Рис. 7. Корреляция между числом или высотой проростков дикого абрикоса и окружающей травянистой или кустарниковой растительностью. 20 октября 1997 г. косточки были посажены в землю на глубину 5 см, а 20 мая 1998 г. было подсчитано число проростков и их высота. В первом эксперименте 10 косточек были посажены на глубину 5 см на каждом из 10 участков. На число проростков дикого абрикоса положительное влияние оказало наличие травянистого покрова (а) и негативное - наличие кустарников (Ь). Во втором эксперименте 25 косточек были посажены на каждом из 12 участков. Как на число проростков, так и на их высоту положительное влияние оказало наличие травянистого покрова (c,d).
Acknowledgements
I thank Mr. Wang Fu-Sheng, Mr.Hao Shou-Shen,Mr.Wang Yong-Qing, Ms Wang Shu-Qing,Ms Cao Xiao-Ping and Dr. Zhang Jian-Xu in helping me doing the field survey. The study was supported by the key project (39893360) of China National Natural Science Foundation and key projects (KZ951-B1-106, KZ952-S1 -107) of the Chinese Academy of Sciences.
REFERENCES
1. Chen L. Z. The importance of Dongling Mountain region of warm temperate deciduous broad-leaved forest. In: Chen LZ and Huang JH (eds) Structure and Function of Warm Temperate Forest Ecosystem. Science Press, Beijing. 1997. ppl-9. (in Chinese)
2. Clarkson K., Eden S. F, Sutherland W. J. Density-dependence and magpie food hoarding. Journal of Animal Ecology, 1986, 55:111-121.
3. Come R. J,, Krehs J. R., Sherry D. F. Food storing in marsh tits, Paras palastris. Animal Behavior, 1981,29:12521259.
4. Elliot E. F. Foraging behavior of a central-place forager: Field test of theoretical predictions. American Naturalists, 1988, 131:159-174.
5. Forget P.-M., Milleron T. Evidence for secondary dispersal by rodents in Panama. Oecologia 1991, 87:596-599.
6. Gill D. S., Marks P. L. Tree and shrub seedling colonization of old fields in central New York. Ecological Monographs 1991, 61:183-205.
7. Herrera J. Acorn predation and seedling production in a low-density population of cork oak
(Quercus suberL.). Forest Ecology and Management, 1995, 76:197-201.
8. Hurly T. A., Robertson Scatter hoarding by territorial red squirrels: a test on the optimal density model. Canadian Journal of Zoology, 1987, 65:1247-1252.
9. Jensen T. S., Nielsen O. F. Rodents as seed dispersers in a heath-oak wood succession. Oecologia
1986, 70:214-221.
10. Jiang Z. G. Significance of food hoarding by animals. Chinese Journal of Zoology, 1996a, 31 (3) :47-
49. (in Chinese)
11. Jiang Z. G. Strategies of food protection by animals. Chinese Journal of Zoology, 1996b, 31 (5) :52-54. (in Chinese)
12. Jiang Z. G. How animals relocate their cached food? Chinese Journal of Zoology, 1996c, 31 (6) :47-
50. (in Chinese)
13. Kanazawa Y., Nishikaia S. Disappearance of acorns from the floor in Quercus crispula forests. Journal of Japanese Forestry Society, 1976, 58:52-56.
14. Kikuzawa K. Dispersal of Quercus mongolica acorns in a broad-leaved deciduous forest. 1. Disappearance. Forest Ecology and Management, 1988, 25:1-8.
15. Kondo N. Seasonal fluctuation of population size, activity area of Apodemus speciosus (Thomas) in a small stand. Journal of Mammalogy Society of Japan, 1980, 8:129-138. (in Japanese)
16. Kollmann J., Schill H.P. Saptial patterns of dispersal, seed predation and germination during colonization of abandoned grassland by Quercus petraea and Corylus avellana. Vegetatio, 1996, 125:193-205.
17. Longland W. S., C Clements. Use of fluorescent pigments in studies of seed caching by rodents. Journal of Mammalogy, 1995, 76(4): 1260-1266.
18. Miyaki M., Kikuzawa K. Dispersal of Quercus mongolica acorns in a broad-leaved deciduous forest. 2. Scatterhoarding by mice. Forest Ecology and Management, 1988, 25:9-16.
19. Reichman O. J. Factors influencing foraging in desert rodents. In: A.C. Kamil and T.D. Sagent (eds.), Foraging behavior ecological, ethological, and psychological approaches. Garland Press. New York. New York. USA. 1981. P.195-213.
20. Schupp E. W. Seed-seeding conflicts, habitat choice, and patterns of plant recruitment. American Journal of Botany, 1995, 82:399-409.
21. Shaw M. W. Factors affecting the natural regeneration of sessile oak (Quercus petraea) in North Wales. II. A preliminary study of acorn production. Journal of Ecology, 1968a, 56:565-585.
22. Shaw M. W. Factors affecting the natural regeneration of sessile oak (Quercus petraea) in North Wales. II. Acorn losses and germination under field conditions. Journal of Ecology, 1968b, 56:647-660.
23. Sork V. L. Mammalia seed dispersal of pignut hickory during three fruiting seasons. Ecology, 1983, 64:1049-1056.
24. Sork V. L. Examination of seed dispersal and survival in red oak, Quercus rubra (Fagaceae), using metal-tagged acorns. Ecology, 1984, 65(3):1020-1023.
25. Stapanian M. A., Smith C C. Density-dependent survival of scatterhoarded nuts: an experimental approach. Ecology. 1984. 65(5): 1387-1396.
26. Vander Wall S. B. An experiment analysis of cache recovery in Clark nutcracker. Animal Behavior, 1982, 30:84-94.
27. Vander Wall S. B. Food Hoarding in Animals. University of Chicago Press, Chicago. 1990.
28. Vander Wall S. B. Dispersal of single leaf Pinon pine (Pinus monophylla) by seed-caching rodents. Journal of Mammalogy, 1997, 78(I):181-191.
29. Vander Wall S. B., Balda R. P. Co-adaptation of the Clark nutcracker and the pine for efficient seed harvest and dispersal. Ecological Monograph, 1977, 47:89-111.
30. Wada N. Dwaef bamboos affect the regeneration of zoochorous trees by providing habitats to acorn-feeding rodents. Oecologia, 1993, 94:403-407.
31. Wilson M.F., Wlielan C.J. Variation in postdispersal survival of vertebrate-dispersed seeds: effects of density, habitat, location, season, and species. Oikos, 1990, 57: 191-198.
32. Yang S. J., Zhu G. Y. The home range of striped field mouse and shrew in farmland and relationship between these two species. Acta Theirologica Sinica, 1989, 9(3): 186-194. (in Chinese).
33. Xia W. P., Long Z. Some ecological aspects of home range and population demography of striped field mouse in Changyang, Hubei Province. Series Report of Rodent Control and Rodent Biology, 1978, 3:85-94. Science Press. Beijing, (in Chinese).
34. Zhang Z. B, Hao S. S., Wang F. S., Wang S. Q., MengZ.B., WangZ.W. Influence of rodents on regeneration of forests in the mountain regions near Beijing. Bulletin of Russia Academy of Sciences, 1998, 4: 467-470 (in Russia). "
