Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

The fruits available as food to small rodents in two woodland ecosystems

The fruits available as food to small rodents in two woodland ecosystems Smal, C. M. and Fairley, J. S. 1980. The fruits available as food to small rodents in two woodland ecosystems. - Holarct. Ecol. 3: 10—18. The fruit fall from the canopy and the understorey in an oak Quercus petraea (Matlushka) Liebl. wood and in a yew Taxus baccata L. wood at Killarney, Ireland, were measured over a period of approximately 18 months; this included two complete fruit crops in each wood. Fruil represents the greai bulk of the food of the small rodents present: wood mice Apodemus sylvaticus (L.) and bank voles Ctethrionomys glareolus Schreber. From a knowledge of the particular fruits or their parts eaten, the amounts of available food, in kcal ha~' were derived. In each ecosystem the food potentially available to small rodents in one crop was five times that in the other. C. M. Smal and J. S. Fairley, Depl of Zoology, Univ. College, Galway, Ireland. 1. Introduction 2. Study areas Both woods have been described by Kelly (1975); a study plot of 5.76 ha (240 x 240 m) was established in the oak-wood but, for reasons related to study of the small mammals, the plot area in the yew wood was 3.24 ha (180 X 180 m). Eighty-eight per cent of the plot in Reenadinna wood lies on 'karst' limestone pavement, most of which is covered with a thick layer of mosses. The limestone is greatly eroded so that grykes and much larger, usually soil-filled, gulleys and hollows are plentiful. The wood on the pavement is almost all pure, mature, dense yew casting heavy shade, and the understorey is extremely sparse. Hazel Coryhis avellana L. fares best in the earth-filled hollows, where it may replace yew as the dominant species. Ash Fraxinus excelsior L. is also common where soil is more substantial and frequently contributes to the canopy. Holly Ilex aquifolium L. is thinly spread and other species, birch Betula pubescens Ehrh., rowan Sorbus acuparia L., hawthorn Crataegus moncgyna Jacq., sloe Prunus spinosa L. and Rhododendron ponticum L., are only occasional. Cherry Prunus avittm L. forms small stands at one There have been many ecological studies of forest rodents: their population biology and to a lesser extent their bioenergetics and role in energy flow. However there has been little serious attempt to relate fluctuations in numbers with food supply, simply because the latter is rarely known with any accuracy. This paper forms part of a larger study of the role of the wood mouse Apodemus sylvaticus (L.) and bank vole Clethrionomys glareolus Schreber and the regulation of their numbers in the woodland ecosystems of the Bourn Vincent National Park at Killarney, Ireland. Here we have attempted to estimate the fruit fall in an oak Querctis petraea (Matluschka) Liebl. and in ^yc-w Taxus baccata L. wood. Hence, from a knowledge of the fruits eaten by these rodents (Smal and Fairley 1980), we were able to calculate the amounts of fruit, in kcal ha"', available to them as food. As wood mice are seed-eaters (taking little green plant material), as bank voles were much less common, and as the herb layer is scanty, the fall of tree seed undoubtedly represents the main food supply of the small rodents in both woods. Accepted 3 August 1979 ® HOLARCTIC ECOLOGY lll(l.':-9.i:7/811/(lJ(Kl[{l.(W SOLSO/O ID HOLARCTIC ECOLOGY 3:1 (t980) corner. Apart from mosses, the herb layer is only apparent in some of the hollows, and is sparse there. The remainder of the study plot lies on drift and is dominated by mature oak. The understorey of holly and hazel is well-developed and ash is frequent. Epiphytes and llanes grow in profusion, including honeysuckle Lonicera periclymenium L., old man's beard Clematis vitalba L. and, especially, ivy Hedera helix L. The herb layer is more luxuriant while the moss layer is much thinner. Tomies oak wood is situated on the lower slopes of Tomies mountain, which is composed of Old Red Sandstone; the study plot was aligned parallel to the contours. The soil is nowhere substantial and rocks and boulders, sometimes moss-covered, litter the area. Though virtually exclusively of mature, native oak, with very occasional birch trees, the plot was divided into three by two open rides, accounting for 12% of the total area. There is an extensive understorey of holly, the only other shrubs present being rowan scattered along the edges of the rides and a few crab apple Malus sylvestris Mill, on the most elevated edge of the plot. Although lianes and epiphytes are widespread, they do not grow profusely; ivy is again the major species. The ground vegetation is dominated by mosses but there are also considerable areas of bare earth. The field layer is scarce and scattered. The rides are dominated by bracken Pteridium aquilinum (L.) Kuhn. Sika deer Cervtis nippon Temminck are common in the Park and, though they damage yew trees by eating the bark, small areas in the yew wood which have been fenced to exclude them show that the ground vegetation does not increase substantially in their absence; they apparently do not eat mosses. In Tomies, however, the herb layer is more abundant on fenced ground and may thus be suppressed over much of the study plot by grazing. Although tree-seedlings were found on both areas, there was no evidence of regeneration. This was no doubt due to the deer in the oak wood and to heavy shading in the yew wood. Fig. 1. Seed trap in position. blown inside-out. A trap is shown in Fig. I. The efficiency was assumed to be 100%. Both areas were divided into four equal sectors and ten seed-traps were placed in each of these; a large number were needed to estimate seed fall with precision as the distribution of seed in traps is a skewed one. All were positioned accurately using co-ordinates derived from random number tables. If, by chance, a pair of co-ordinates coincided with a tree, a new position was again randomly selected. Traps were operational from August 1975 in the yew wood and from September 1975 in the oak wood. In the latter this was early enough to catch the entire crop. But in Reenadinna some hazel and yew fruits had fallen in the fortnight before erection and so a 0.5 m^ quadrat was used to count all the tree-seeds exactly below where each trap was sited. Only a small error is therefore involved. All collecting bags were replaced once a month on average with slightly more frequent collections during the heaviest fall. Final collections were in May 1977. In the last five months, four traps had to be replaced because the fabric had rotted and all of these were in the sheltered, more humid yew wood. Altogether there were breakages of one kind or another at 5.8% of trap-visits In the yew wood and 1.8% in the oak wood. The 'catch' at each of these had to be estimated from the previous catch and the overall increase or decrease in all operational traps. Where the previous catch was small (and therefore unreliable) the result from the collection following was employed. After preliminary drying, the fruits were separated from the leaf litter and the other material and all 3. Methods For practical purposes, sampling in Tomies was confined to the wood; it was assumed that seed fall in the rides was zero. The fruit fall was estimated with seed traps. Each trap was essentially a funnel of calico supported on wire, to which it was sewn. The wire frame at the mouth measured 70.7 ± 0.5 cm square, and the trap thus had a catchment area of 0.5 m^. The collecting bag was also of calico with a square wire frame at the mouth, matching that at the base of the funnel, and secured to wire loops by means of cotton tapes. The trap was supported by wooden stakes about 1 m above the ground and a large stone was placed in the bag to prevent it from being HOLARCTIC ECOLOGY 3:1 (198(1) It aborted fruits transferred to the leaf-litter. Although acorn cups were counted, they were also added to the leaf-litter. The remaining fruits were sorted into well-defined categories and counted as follows: YEW ripe whole fruits, ripe arils without seeds, ripe seeds, partially ripe fruits, unripe seeds, pieces of shell (subdivided into those less than half and more than half a complete shell); OAK ripe acorns, unripe acoms, empty shells; HOLLY ripe berries, unripe berries, partially eaten ripe berries, partially eaten unripe berries, seeds; HAZEL ripe nuts, unripe nuts; IVY ripe berries, unripe berries, seeds; ASH fruits; CHERRY whole fruits; ROWAN ripe berries. Because only parts of some fruits are eaten by the mice and voles, the relative weights of these were investigated. The entire collections of acorns and hazel nuts from each trap were separated into shell and endosperm, dried for 48 h at 105°C and weighed. No such measurements were made on rowan berries and cherries as these were few. For the remaining fruits - yew, holly, ivy and ash — random samples were taken from the material collected, the fruits divided into seed, shell and (where relevant) flesh, dried and weighed. All the remaining fruit was also dried and weighed but sufficient was retained from all species to allow calorific values of the various components to be determined. The latter material was dried slowly (at 85°C for 48 h) to reduce loss of any volatile constituents. Calorific values were obtained using a ballistic bomb calorimeter (Gallenkamp CB37O). Samples around 0.5 g were taken for each determination. Fruit or fruit part Yew whole ripe fruits ripe seeds without seeds with partially formed arils unripe seeds pieces of shell greater than half of a whole pieces of shell less than half of a whole Hazel ripe and unripe hazelnuts Ash Crop I (10^ ha"') Crop 2 (10* ha-') i«sftn+937.6 1440 5+539.8 '**"-^-446.6 ,2g 0+273.0 52^-^-211.6 505 5 + " 0 " ,.,, n+268.4 '^^'•«-203.3 1335 5+^5^'' „ , ^ + 283.2 462.0_,75 5 25Q 0+112.8 ' - 90.1 224 " - 97.8 " ^ • 0+'29-^ g 5+113.4 i»4.i_ g^2 5JJP 5 + 282.0 1032 5+536.3 * -415.3 47 0+ 36.1 ^'•"- ^'-^ 1746.0+^j|3 all at>h keys Oak ripe acoms unripe acoms acorn cups 1 0+ 3-0 '•"1.0 2 0+ ^-^ ^•^1.9 42 5 + ^ ' - ^ ^'^•^- 32.8 3* •0+ ^-^ • "2.6 0 0+ 0.0 "•"0.0 12 0+ ^^-^ 4. Results Seasonal variations in fruit fall in the yew wood are shown in Fig. 2. There was a much heavier crop in 1976 than in 1975 for yew, hazel, oak. holly and ash, indeed there was practically no ash seed in the first year. Ivy, however, which fruits early in the year, was more abundant in 1976 than 1977, corresponding respectively to (he 1975 and 1976 crops for other species. Also taken, in insignifieant amounts, were rowan berries, cherries and cherry stones. The fall of yew fruits was by far ihe most abundant and a high proportion of the latter material was of ripe seeds without flesh. This was even more marked with holly berries and in both species was almost certainty entirely caused by the feeding of birds, mainly redwings Turdus iliacus L. and thrushes Turdus sp. which were frequently seen eating them. The seeds pass through the guts unharmed, resulting in substantial Holly ripe drupes unripe drupes seeds (without any fruit flesh) ivy ripe berries unripe berries seeds (without any fruit flesh) Cherry whole fruits stones Rowan ripe berries 3 5+ 10.5 •*-^3.4 + 91.6 347 5+'^^-^ •*^'^-lI8.0 7nftn+'27.8 206.0_ ^^j 19 0 + '^•'* ' ^ • " - U.O 0.0+ g-g ^20 5 + 375.3 3 •0+ ^-O ^ " - 2.6 BA i;+ "^8.0 - 43.4 n n+ 0.0 "•"0.5 0.0 Tab. 1. The crops of fruits or their components in the yew wood, given by number per hectare with 95% confidence limits. The first crop is the total colleclion over the period 1.8.75 lo 29.6.76 (except for cherry: 1.8.75 to 23.5.76) and the second crop that over the remainitig period to 16.6.77. Where data were limiicd, confidence limits arc not given. 2 5+ 3.2 ^•^- 1.8 3i u _ • 0+ ^^ 2.9 2 5+ 2.4 ^•^- '-5 HOLARCTIC ECOLOGV 3:1 (19gO) Q. " 0.5 m rg *th I. jg. J\ A /\ Fig. 2. Seasonal variation in the fall of fruit in the yew wood, given as the number of items falling per day per seed trap. A. Black squares, entire ripe yew fruits; white squares and broken line, ripe yew seeds without arils; black dots, pieces of shell, B. Black triangles, yew fruits wilh partially formed arils; white triangles, unripe yew fruits. C. White squares, ash keys; black triangles ivy berries; white triangles, ivy seeds. D. Black dots and broken line, acorns; black triangles, hazel nuts; black squares, holly berries; white squares, holly seeds. numb( 197G 1976 HOLARCTIC ECOLOOY 3:1 (1980) Fig. 3. Seasonal variation in the fall of fruit in the oak wood, given as Ihe number of items falling per day per seed trap. A. Black dots, acoms; black triangles, ivy berries; white triangles, ivy seeds. B. black squares, holly berries; white squares, holly seeds. catches in seed traps below roosting sites. Similarly, more ivy seeds are caught than complete berries because of the depredations of wood pigeons Columba palumbus L. Cherries too were sometimes eaten by birds. Many pieces of shell from yew seeds were collected and their history is not clear. But as several seeds were found which were beginning to split irregularly and contained no endosperm, it is likely that the pieces were derived from such empty shells. Most hazel nuts did not contain endosperm. The variations in fniit fall in the oak wood are given In Fig. 3. As at Reenadinna, the holly and ivy crops were much larger in 1976. In contrast acorn mast was scanty in 1976 but substantial in 1975; many aborted acoms fell in 1976, though few in the previous year. Small rodents fed on unripe holly berries in the canopy. For example, counts from two consecutive collections on 1 September and 1 October 1976 - showed that 34% and 48% respectively of unripe benies had been affected. The purpose seemed to be to gain access to the seeds. Part of the unripe fruit flesh was removed (which is unpalatable to both mice and voles - Smal and Fairley 14 1980), but only the largest seeds were attacked. As bank voles are disinclined to open holly seeds (Smal and Fairley 1980), it may be assumed that wood mice were responsible. Although ripe berries are quite edible, judging by the seed traps they were practically never taken in the canopy. It is likely that, by the time they were ready, other more accessible and palatable food was available, in the form of acorns. (Rodents did not feed on holly in the canopy in the yew wood: yew fruits fall much earlier than acorns). There was no evidence of feeding in the seed traps. Any of this would certainly have included acoms but partially eaten acoms or discarded shells with tooth marks were never found. Flocks of at least 250 woodpigeons were present in Tomies in October 1975. feeding on acoms. As at Reenadinna, ripe ivy and holly berries were taken by birds too. The study covered two complete crops in both areas, although the period of study was actually less than two full years. In the oak wood the crops were readily separated by segregating collections up to 20 June 1976 from the remainder. The corresponding period ended on 29 June 1976 in the yew wood. A single exception is HOLARCT-IC ECOLOGY 3:t (1980) Tab. 2. The crops of fruits or their components in the oak wood, given by the number per hectare with 95% confidence limits. The first crop is the total collection from 1.9.75 to 20.6.76, and the second crop that over the remaining [leriod up to 18.5.77. Fruit or fruit part Oak ripe acoms unripe acorns.. acorn cups . . . . Holly ripe drupes . . . . unripe drupes (total) seeds with no fruit flesh (eaten plus uneaten).. total seeds obtained from unripe drupes uneaten seeds obtained from unripe drupes . unripe fruit flesh expressed as equivalents in whole fruits Ivy ripe berries . . . unripe berries . seeds (with no fruit flesh) 23.6 + 6.1 42.2 20.0 147.2 + 68.2 54.1 Crop 1 10' ha-') Crop 2 ,+'0.0 7.9 + + - + - + - 39.0 - 26.3 + 4I2.I 282.1 - 219.1 + 1535.8 2005.1 -1069.5 + 1591.1 1082.7 - 843.4 825.6 + 1213.3 - 643.1 46.4 + 1437.9 + 1542.6 - 949.3 + + are shown in Tabs 1 and 2; they emphasize the substantial differences between 1975 and 1976 crops. The calorific contents of the components of the six major fruits are given in Tab. 3. (The term 'endosperm' is used to include the embryo and cotyledons). Ash content was not accounted for but would have increased estimates fractionally. The values for oak, haze! and ash are in agreement with those of Grodzinski and Sawicka-Kapusta (1970) and Drozdz (1967); the small differences are not unexpected through variations in climate, habitat and genetic constitution. Values for yew, holly and ivy fruits have apparently not been determined before; that for yew exceeds the calorific content of any species recorded to date and is probably a result of the high fat content in the seed. The total energy content (biocontent — Newbould 1967) was calculated for the crops of fruits or their parts from the calorific values and mean weights of their various components; the 95% confidence limits of the latter were almost always well under 20% of the mean. These are given in Tabs 4 and 5 and, as several categories (From Tabs 1 and 2) are combined, no confidence limits are included. A few assumptions were made. Values for holly were used for cherry and rowan. For yew, the calorific value of unripe seeds was taken to Tab. 3. Calorific values of fruit flesh, seed shells and endosperm of the main species of fruits found in the seed traps. Sample No. of tests Calorific content (kcal g-') Mean 95% Cl. 5.07 5.19 5.41 8.41 4.45 4.62 4.89 5.02 5.00 6.31 6.26 6.72 5.02 7.79 4.64 6.21 ±0.19 5.9 + + 255.2 169.1 - 133.3 + 7.4 0.0 + - Yew 33.3 + 0.0 + - shells from ripe seeds shells from seeds with partially formed arils. endosperm Oak ±0.25 ±0.19 ±0.60 ±0.26 ±0.03 ±0.20 +0.68 ±0.07 ±0.23 ±0.12 322.4 + 316.8 - 202.0 + - endosperm Holly Ihe cherry with the first crop arbitrarily assumed to have terminated on 23 May 1976. The catch of fruit from each species in each crop were calculated, the mean per trap and hence the total fall in each crop. As there was a clear tendency for a small number of traps to carry a large catch, a problem arose in obtaining confidence limits to the means. Most of the sets of data were skewed and found to correspond to a gamma distribution. Although, for statistical reasons, it was only possible to check this with a chi-squared test on 32 sets, only four of these differed from a gamma distribition at the 5% level. For simplicity therefore, all sets of data were assumed to have a gamma distribution. The crops per hectare and their 95% confidence limits 3:1 (1980) flesh of unripe drupes endosperm Ivy flesh from ripe berries . . . endosperm (from ripe berries) Hazel endosperm Ash ±0.31 ±0.23 ±0.37 ±0.31 Tab. 4. The weight and calorific content, per crop, of the components of each fruit falling to the ground in the yew wood. Component Yew ripe aril alt shell (including partially formed arils but excluding shell fragments) shell fragments endosperm Total Oak acorn shells endosperm Total Holly ripe flesh of drupes unripe flesh of drupes all shells endosperm Total Ivy ripe flesh of berries unripe flesh of berries endosperm Total Hazel hazelnul shells endosperm Total Ash wings (shells) endosperm Total Cherry flesh shells endosperm Total Rowan flesh shells endosperm Total Weight (kg ha-') Crop 1 Crop 2 Bicontent (10^ kcal ha"') Crop 1 Crop 2 QAt o.n be the same as for shells (with arils) of seeds with partially formed coats. For shell fragtnents: 1 greater than half a whole -I- 1 less than half = 1 shell; remaining fragments less than half -^3 = whole shells. The mean weight of shell here was given by averaging values from shells from ripe seeds with arils and those without. For holly the mean weight of shell and endosperm from unripe fruits were assumed to be equal to those from ripe ones. Values for unripe ivy berries were similarly assumed to be equivalent to those of ripe berries. The 'losses' to small mammals of seed and flesh of unripe holly were estimated from counts of seeds opened (all of which were assumed to contain endosperm) and assessment of missing flesh. As these were unquestionably available to small mammals, they have been included in Tab. 5. From feeding experiments (Smal and Fairley 1980) it had been determined that all endosperm and flesh was palatable to small mammals with the following exceptions; unripe holly flesh and all parts of ivy fruit to both HOLARCTIC ECOLOGY 3:1 (1980) Tab. 5. The weight and calorific content, per crop, of the components of each fruit falling to the ground in Ihe oak wood. Weight (kg ha"') Crop 1 Crop 2 Oak acorn shells endosperm Total Holly ripe flesh of drupes unripe flesh of drupes all shells endosperm Total Bicontent (10^ kcal ha"') Crop 1 Crop 2 10.14 28.30 38.44 9.76 40.43 6.13 8.41 IM 8.05 60.08 82.44 0.25 0.00 1.02 1.27 176.46 492.36 668,82 4,97 Ivy ripe flesh of berries unripe flesh of berries endosperm Total Tab. 6. The total biocontent of fruit flesh and endosperm falling to the ground in the two study areas and the fractions available as food to mice and voles. Component Yew wood Crop 1 Biocontenl (10^ kcal ha"') Oak wood Crop 2 Crop 1 12.79 591.26 604.05 497.34 44.24 0.00 541.57 89.66% Crop 2 51.74 112.88 164.62 38.06 77.70 0.00 115.76 70.32% Total fruit flesh Total seed endosperm Total Total palatable to both species . . . Total palatable to mice but not voles (holly endosperm) Total palatable to voles but not mice (ash endosperm) Total palatable to small mammals Percentage palatable of total 55.00 284.66 339.66 202,03 5.99 0.05 208.07 61.26% 98.46% wood mice and bank voles, holly seeds to voles and ash keys to mice. Small quantities of ash may be eaten by wood mice and similarly the flesh of ivy might be eaten by voles in minor amounts but the error involved is negligible. In this light. Tab. 6 gives the total biocontent of fruit falling to the ground (including the unripe holly eaten in the canopy) available as food to both wood mice and bank voles. The amounts of fruit flesh eaten by birds was calculated from numbers of the naked seeds in the traps of yew, holly, ivy and cheny. From observations on feeding activity, the fruits involved were invariably ripe and therefore appropriate mean weights and calorific contents of fruit flesh could be applied with confidence. The results are given in Tab. 7 along with the data for holly seeds removed by small mammals in the canopy. 2 HOLARCTtC ECOLOGY 3:1 (1980) 5. Discussion Although there are obvious sampling and statistical errors inherent in this study, and the fruit fall in Tomies wood is likely to be underestimated through the unknown extent of feeding on acorns by woodpigeons, the figures for fruit fall potentially available as food to small mammals for both woods are undoubtedly close approximations to the true values. As fruit represents the main food of small rodents in both woods, it is clear that there is considerable variation in the food supply from year to year. At Reenadinna, that from the second crop was five times that in the first and at Tomies the situation was almost the reverse. It seems reasonable to infer that such marked fluctuations in the food supply of bank voles Tab. 7. Estimates of the weight and caloriHc value of the fruit removed by birds and small mammals in the canopy in both study areas. Component Weight (kg ha-') Crop 1 Crop 2 Bicontent (10^ kcal ha"') Crop 1 Crop 2 Yew ripe arils eaten by birds in the yew wood percentage of lotal ripe aril Holly ripe hoily drupe flesh eaten by birds in the yew wood percentage of total ripe flesh ripe holly drupe flesh eaten by birds in the oak wood percentage of total ripe flesh unripe holly drupe flesh removed by small mammals percentage of total unripe flesh holly seed endosperm eaten by small mammals percentage of total endosperm Ivy ripe berry flesh eaten by woodpigeons in the yew wood percentage of total eaten by woodpigeons in the oak wood percentage of total Cherry ripe cherry flesh eaten by birds in the yew wood percentage of total ripe flesh 2.69 35.1% 15.52 43.3% 13.65 35.1% 78.68 43.3% 2.28 96.4% 16.22 94.1 % 0.01 3.6% 0.00 0.0% 3.92 98.9% 22.62 91.9% 3.23 40.1% 1.64 13.3% 11.15 96.4% 79.37 94.1% 0.03 3.6% 0.00 0.0% 19.18 98.9% 110.68 91.9% 16.20 40.1% 10.32 13.3% 6.58 58.8% 5.30 82.7% 0.90 67.8% 0.59 70.4% 41.20 58.8% 33.16 82.7% 5.61 67.8% 3.70 70.4% 0.02 100.07o 0.11 83.3% 0.11 100.0% 0.53 83.3% and wood mice in woodland are quite usual, yet comparable variation in peak densities of the animals is apparently rare. In a later publication we will examine the effect of abundance of food on population densities of mice and voles in these woods. This study exposes the dangers inherent in explaining fluctuations in the numbers of woodland rodents on the basis of casual observations on seed fall, especially that of the main tree species in the canopy alone. Fall of fruit is locally exceedingly variable (hence the gamma distribution); some species are unpalatable and, of those which are, not all parts are eaten; calorific values of fruits differ considerably; in some years the crop from the understorey is of more importance than that from the canopy (for example in 1976 in Tomies wood); a large proportion of fruits may contain no endosperm (for example hazelnuts); and a significant part of the crop may be eaten by small mammals in the canopy or removed there by birds. Acknowledgements - We are indebted to E. Thomson and the staff of tbe Bourn Vincent Memorial Park for assistance in producing seed traps. We would also like to thank Professor J. N. R. Grainger for the use of a bomb calorimeter and Mr P. Stafford for assistance in its operation. Helpful advice is gratefully acknowledged from Dr J. Shields on application of the gamma distribution and Mr M, Shaw on construction and use of seed traps. This investigation was financed by the Office of Public Works, Dublin. References Drozdz, A. 1967. Food preference, food digestibility and natural food supply of small rodents. - In: Petrusewicz, K. (ed.), Secondary productivity of terrestrial ecosystems. Warsaw. Grodziiiski, W. and Sawicka-Kapusta, K. 1970. Energy values of tree seeds eaten by small mammals. - Oikos 21: 52-58. Kelly, D. 1975. Native woodland in western Ireland with special reference to the region of Killamey. - Ph. D. thesis. University of Dublin. Newbould, P. J. 1967. Methods of estimating the primary production of forests. - Blackwell. Oxford. Smai, C. M. and Fairley, J. S. 1980. Food of wood mice and bank voles in oak and yew woods at Killamey, Ireland. - J, Zool., Lond. (in press). HOIj^RCnC ECOLOGY 3:1 (1980) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Ecography Wiley

The fruits available as food to small rodents in two woodland ecosystems

Ecography , Volume 3 (1) – Jan 1, 1980

Loading next page...
 
/lp/wiley/the-fruits-available-as-food-to-small-rodents-in-two-woodland-rQxwhrX9pX

References (4)

Publisher
Wiley
Copyright
Copyright © 1980 Wiley Subscription Services, Inc., A Wiley Company
ISSN
0906-7590
eISSN
1600-0587
DOI
10.1111/j.1600-0587.1980.tb00703.x
Publisher site
See Article on Publisher Site

Abstract

Smal, C. M. and Fairley, J. S. 1980. The fruits available as food to small rodents in two woodland ecosystems. - Holarct. Ecol. 3: 10—18. The fruit fall from the canopy and the understorey in an oak Quercus petraea (Matlushka) Liebl. wood and in a yew Taxus baccata L. wood at Killarney, Ireland, were measured over a period of approximately 18 months; this included two complete fruit crops in each wood. Fruil represents the greai bulk of the food of the small rodents present: wood mice Apodemus sylvaticus (L.) and bank voles Ctethrionomys glareolus Schreber. From a knowledge of the particular fruits or their parts eaten, the amounts of available food, in kcal ha~' were derived. In each ecosystem the food potentially available to small rodents in one crop was five times that in the other. C. M. Smal and J. S. Fairley, Depl of Zoology, Univ. College, Galway, Ireland. 1. Introduction 2. Study areas Both woods have been described by Kelly (1975); a study plot of 5.76 ha (240 x 240 m) was established in the oak-wood but, for reasons related to study of the small mammals, the plot area in the yew wood was 3.24 ha (180 X 180 m). Eighty-eight per cent of the plot in Reenadinna wood lies on 'karst' limestone pavement, most of which is covered with a thick layer of mosses. The limestone is greatly eroded so that grykes and much larger, usually soil-filled, gulleys and hollows are plentiful. The wood on the pavement is almost all pure, mature, dense yew casting heavy shade, and the understorey is extremely sparse. Hazel Coryhis avellana L. fares best in the earth-filled hollows, where it may replace yew as the dominant species. Ash Fraxinus excelsior L. is also common where soil is more substantial and frequently contributes to the canopy. Holly Ilex aquifolium L. is thinly spread and other species, birch Betula pubescens Ehrh., rowan Sorbus acuparia L., hawthorn Crataegus moncgyna Jacq., sloe Prunus spinosa L. and Rhododendron ponticum L., are only occasional. Cherry Prunus avittm L. forms small stands at one There have been many ecological studies of forest rodents: their population biology and to a lesser extent their bioenergetics and role in energy flow. However there has been little serious attempt to relate fluctuations in numbers with food supply, simply because the latter is rarely known with any accuracy. This paper forms part of a larger study of the role of the wood mouse Apodemus sylvaticus (L.) and bank vole Clethrionomys glareolus Schreber and the regulation of their numbers in the woodland ecosystems of the Bourn Vincent National Park at Killarney, Ireland. Here we have attempted to estimate the fruit fall in an oak Querctis petraea (Matluschka) Liebl. and in ^yc-w Taxus baccata L. wood. Hence, from a knowledge of the fruits eaten by these rodents (Smal and Fairley 1980), we were able to calculate the amounts of fruit, in kcal ha"', available to them as food. As wood mice are seed-eaters (taking little green plant material), as bank voles were much less common, and as the herb layer is scanty, the fall of tree seed undoubtedly represents the main food supply of the small rodents in both woods. Accepted 3 August 1979 ® HOLARCTIC ECOLOGY lll(l.':-9.i:7/811/(lJ(Kl[{l.(W SOLSO/O ID HOLARCTIC ECOLOGY 3:1 (t980) corner. Apart from mosses, the herb layer is only apparent in some of the hollows, and is sparse there. The remainder of the study plot lies on drift and is dominated by mature oak. The understorey of holly and hazel is well-developed and ash is frequent. Epiphytes and llanes grow in profusion, including honeysuckle Lonicera periclymenium L., old man's beard Clematis vitalba L. and, especially, ivy Hedera helix L. The herb layer is more luxuriant while the moss layer is much thinner. Tomies oak wood is situated on the lower slopes of Tomies mountain, which is composed of Old Red Sandstone; the study plot was aligned parallel to the contours. The soil is nowhere substantial and rocks and boulders, sometimes moss-covered, litter the area. Though virtually exclusively of mature, native oak, with very occasional birch trees, the plot was divided into three by two open rides, accounting for 12% of the total area. There is an extensive understorey of holly, the only other shrubs present being rowan scattered along the edges of the rides and a few crab apple Malus sylvestris Mill, on the most elevated edge of the plot. Although lianes and epiphytes are widespread, they do not grow profusely; ivy is again the major species. The ground vegetation is dominated by mosses but there are also considerable areas of bare earth. The field layer is scarce and scattered. The rides are dominated by bracken Pteridium aquilinum (L.) Kuhn. Sika deer Cervtis nippon Temminck are common in the Park and, though they damage yew trees by eating the bark, small areas in the yew wood which have been fenced to exclude them show that the ground vegetation does not increase substantially in their absence; they apparently do not eat mosses. In Tomies, however, the herb layer is more abundant on fenced ground and may thus be suppressed over much of the study plot by grazing. Although tree-seedlings were found on both areas, there was no evidence of regeneration. This was no doubt due to the deer in the oak wood and to heavy shading in the yew wood. Fig. 1. Seed trap in position. blown inside-out. A trap is shown in Fig. I. The efficiency was assumed to be 100%. Both areas were divided into four equal sectors and ten seed-traps were placed in each of these; a large number were needed to estimate seed fall with precision as the distribution of seed in traps is a skewed one. All were positioned accurately using co-ordinates derived from random number tables. If, by chance, a pair of co-ordinates coincided with a tree, a new position was again randomly selected. Traps were operational from August 1975 in the yew wood and from September 1975 in the oak wood. In the latter this was early enough to catch the entire crop. But in Reenadinna some hazel and yew fruits had fallen in the fortnight before erection and so a 0.5 m^ quadrat was used to count all the tree-seeds exactly below where each trap was sited. Only a small error is therefore involved. All collecting bags were replaced once a month on average with slightly more frequent collections during the heaviest fall. Final collections were in May 1977. In the last five months, four traps had to be replaced because the fabric had rotted and all of these were in the sheltered, more humid yew wood. Altogether there were breakages of one kind or another at 5.8% of trap-visits In the yew wood and 1.8% in the oak wood. The 'catch' at each of these had to be estimated from the previous catch and the overall increase or decrease in all operational traps. Where the previous catch was small (and therefore unreliable) the result from the collection following was employed. After preliminary drying, the fruits were separated from the leaf litter and the other material and all 3. Methods For practical purposes, sampling in Tomies was confined to the wood; it was assumed that seed fall in the rides was zero. The fruit fall was estimated with seed traps. Each trap was essentially a funnel of calico supported on wire, to which it was sewn. The wire frame at the mouth measured 70.7 ± 0.5 cm square, and the trap thus had a catchment area of 0.5 m^. The collecting bag was also of calico with a square wire frame at the mouth, matching that at the base of the funnel, and secured to wire loops by means of cotton tapes. The trap was supported by wooden stakes about 1 m above the ground and a large stone was placed in the bag to prevent it from being HOLARCTIC ECOLOGY 3:1 (198(1) It aborted fruits transferred to the leaf-litter. Although acorn cups were counted, they were also added to the leaf-litter. The remaining fruits were sorted into well-defined categories and counted as follows: YEW ripe whole fruits, ripe arils without seeds, ripe seeds, partially ripe fruits, unripe seeds, pieces of shell (subdivided into those less than half and more than half a complete shell); OAK ripe acorns, unripe acoms, empty shells; HOLLY ripe berries, unripe berries, partially eaten ripe berries, partially eaten unripe berries, seeds; HAZEL ripe nuts, unripe nuts; IVY ripe berries, unripe berries, seeds; ASH fruits; CHERRY whole fruits; ROWAN ripe berries. Because only parts of some fruits are eaten by the mice and voles, the relative weights of these were investigated. The entire collections of acorns and hazel nuts from each trap were separated into shell and endosperm, dried for 48 h at 105°C and weighed. No such measurements were made on rowan berries and cherries as these were few. For the remaining fruits - yew, holly, ivy and ash — random samples were taken from the material collected, the fruits divided into seed, shell and (where relevant) flesh, dried and weighed. All the remaining fruit was also dried and weighed but sufficient was retained from all species to allow calorific values of the various components to be determined. The latter material was dried slowly (at 85°C for 48 h) to reduce loss of any volatile constituents. Calorific values were obtained using a ballistic bomb calorimeter (Gallenkamp CB37O). Samples around 0.5 g were taken for each determination. Fruit or fruit part Yew whole ripe fruits ripe seeds without seeds with partially formed arils unripe seeds pieces of shell greater than half of a whole pieces of shell less than half of a whole Hazel ripe and unripe hazelnuts Ash Crop I (10^ ha"') Crop 2 (10* ha-') i«sftn+937.6 1440 5+539.8 '**"-^-446.6 ,2g 0+273.0 52^-^-211.6 505 5 + " 0 " ,.,, n+268.4 '^^'•«-203.3 1335 5+^5^'' „ , ^ + 283.2 462.0_,75 5 25Q 0+112.8 ' - 90.1 224 " - 97.8 " ^ • 0+'29-^ g 5+113.4 i»4.i_ g^2 5JJP 5 + 282.0 1032 5+536.3 * -415.3 47 0+ 36.1 ^'•"- ^'-^ 1746.0+^j|3 all at>h keys Oak ripe acoms unripe acoms acorn cups 1 0+ 3-0 '•"1.0 2 0+ ^-^ ^•^1.9 42 5 + ^ ' - ^ ^'^•^- 32.8 3* •0+ ^-^ • "2.6 0 0+ 0.0 "•"0.0 12 0+ ^^-^ 4. Results Seasonal variations in fruit fall in the yew wood are shown in Fig. 2. There was a much heavier crop in 1976 than in 1975 for yew, hazel, oak. holly and ash, indeed there was practically no ash seed in the first year. Ivy, however, which fruits early in the year, was more abundant in 1976 than 1977, corresponding respectively to (he 1975 and 1976 crops for other species. Also taken, in insignifieant amounts, were rowan berries, cherries and cherry stones. The fall of yew fruits was by far ihe most abundant and a high proportion of the latter material was of ripe seeds without flesh. This was even more marked with holly berries and in both species was almost certainty entirely caused by the feeding of birds, mainly redwings Turdus iliacus L. and thrushes Turdus sp. which were frequently seen eating them. The seeds pass through the guts unharmed, resulting in substantial Holly ripe drupes unripe drupes seeds (without any fruit flesh) ivy ripe berries unripe berries seeds (without any fruit flesh) Cherry whole fruits stones Rowan ripe berries 3 5+ 10.5 •*-^3.4 + 91.6 347 5+'^^-^ •*^'^-lI8.0 7nftn+'27.8 206.0_ ^^j 19 0 + '^•'* ' ^ • " - U.O 0.0+ g-g ^20 5 + 375.3 3 •0+ ^-O ^ " - 2.6 BA i;+ "^8.0 - 43.4 n n+ 0.0 "•"0.5 0.0 Tab. 1. The crops of fruits or their components in the yew wood, given by number per hectare with 95% confidence limits. The first crop is the total colleclion over the period 1.8.75 lo 29.6.76 (except for cherry: 1.8.75 to 23.5.76) and the second crop that over the remainitig period to 16.6.77. Where data were limiicd, confidence limits arc not given. 2 5+ 3.2 ^•^- 1.8 3i u _ • 0+ ^^ 2.9 2 5+ 2.4 ^•^- '-5 HOLARCTIC ECOLOGV 3:1 (19gO) Q. " 0.5 m rg *th I. jg. J\ A /\ Fig. 2. Seasonal variation in the fall of fruit in the yew wood, given as the number of items falling per day per seed trap. A. Black squares, entire ripe yew fruits; white squares and broken line, ripe yew seeds without arils; black dots, pieces of shell, B. Black triangles, yew fruits wilh partially formed arils; white triangles, unripe yew fruits. C. White squares, ash keys; black triangles ivy berries; white triangles, ivy seeds. D. Black dots and broken line, acorns; black triangles, hazel nuts; black squares, holly berries; white squares, holly seeds. numb( 197G 1976 HOLARCTIC ECOLOOY 3:1 (1980) Fig. 3. Seasonal variation in the fall of fruit in the oak wood, given as Ihe number of items falling per day per seed trap. A. Black dots, acoms; black triangles, ivy berries; white triangles, ivy seeds. B. black squares, holly berries; white squares, holly seeds. catches in seed traps below roosting sites. Similarly, more ivy seeds are caught than complete berries because of the depredations of wood pigeons Columba palumbus L. Cherries too were sometimes eaten by birds. Many pieces of shell from yew seeds were collected and their history is not clear. But as several seeds were found which were beginning to split irregularly and contained no endosperm, it is likely that the pieces were derived from such empty shells. Most hazel nuts did not contain endosperm. The variations in fniit fall in the oak wood are given In Fig. 3. As at Reenadinna, the holly and ivy crops were much larger in 1976. In contrast acorn mast was scanty in 1976 but substantial in 1975; many aborted acoms fell in 1976, though few in the previous year. Small rodents fed on unripe holly berries in the canopy. For example, counts from two consecutive collections on 1 September and 1 October 1976 - showed that 34% and 48% respectively of unripe benies had been affected. The purpose seemed to be to gain access to the seeds. Part of the unripe fruit flesh was removed (which is unpalatable to both mice and voles - Smal and Fairley 14 1980), but only the largest seeds were attacked. As bank voles are disinclined to open holly seeds (Smal and Fairley 1980), it may be assumed that wood mice were responsible. Although ripe berries are quite edible, judging by the seed traps they were practically never taken in the canopy. It is likely that, by the time they were ready, other more accessible and palatable food was available, in the form of acorns. (Rodents did not feed on holly in the canopy in the yew wood: yew fruits fall much earlier than acorns). There was no evidence of feeding in the seed traps. Any of this would certainly have included acoms but partially eaten acoms or discarded shells with tooth marks were never found. Flocks of at least 250 woodpigeons were present in Tomies in October 1975. feeding on acoms. As at Reenadinna, ripe ivy and holly berries were taken by birds too. The study covered two complete crops in both areas, although the period of study was actually less than two full years. In the oak wood the crops were readily separated by segregating collections up to 20 June 1976 from the remainder. The corresponding period ended on 29 June 1976 in the yew wood. A single exception is HOLARCT-IC ECOLOGY 3:t (1980) Tab. 2. The crops of fruits or their components in the oak wood, given by the number per hectare with 95% confidence limits. The first crop is the total collection from 1.9.75 to 20.6.76, and the second crop that over the remaining [leriod up to 18.5.77. Fruit or fruit part Oak ripe acoms unripe acorns.. acorn cups . . . . Holly ripe drupes . . . . unripe drupes (total) seeds with no fruit flesh (eaten plus uneaten).. total seeds obtained from unripe drupes uneaten seeds obtained from unripe drupes . unripe fruit flesh expressed as equivalents in whole fruits Ivy ripe berries . . . unripe berries . seeds (with no fruit flesh) 23.6 + 6.1 42.2 20.0 147.2 + 68.2 54.1 Crop 1 10' ha-') Crop 2 ,+'0.0 7.9 + + - + - + - 39.0 - 26.3 + 4I2.I 282.1 - 219.1 + 1535.8 2005.1 -1069.5 + 1591.1 1082.7 - 843.4 825.6 + 1213.3 - 643.1 46.4 + 1437.9 + 1542.6 - 949.3 + + are shown in Tabs 1 and 2; they emphasize the substantial differences between 1975 and 1976 crops. The calorific contents of the components of the six major fruits are given in Tab. 3. (The term 'endosperm' is used to include the embryo and cotyledons). Ash content was not accounted for but would have increased estimates fractionally. The values for oak, haze! and ash are in agreement with those of Grodzinski and Sawicka-Kapusta (1970) and Drozdz (1967); the small differences are not unexpected through variations in climate, habitat and genetic constitution. Values for yew, holly and ivy fruits have apparently not been determined before; that for yew exceeds the calorific content of any species recorded to date and is probably a result of the high fat content in the seed. The total energy content (biocontent — Newbould 1967) was calculated for the crops of fruits or their parts from the calorific values and mean weights of their various components; the 95% confidence limits of the latter were almost always well under 20% of the mean. These are given in Tabs 4 and 5 and, as several categories (From Tabs 1 and 2) are combined, no confidence limits are included. A few assumptions were made. Values for holly were used for cherry and rowan. For yew, the calorific value of unripe seeds was taken to Tab. 3. Calorific values of fruit flesh, seed shells and endosperm of the main species of fruits found in the seed traps. Sample No. of tests Calorific content (kcal g-') Mean 95% Cl. 5.07 5.19 5.41 8.41 4.45 4.62 4.89 5.02 5.00 6.31 6.26 6.72 5.02 7.79 4.64 6.21 ±0.19 5.9 + + 255.2 169.1 - 133.3 + 7.4 0.0 + - Yew 33.3 + 0.0 + - shells from ripe seeds shells from seeds with partially formed arils. endosperm Oak ±0.25 ±0.19 ±0.60 ±0.26 ±0.03 ±0.20 +0.68 ±0.07 ±0.23 ±0.12 322.4 + 316.8 - 202.0 + - endosperm Holly Ihe cherry with the first crop arbitrarily assumed to have terminated on 23 May 1976. The catch of fruit from each species in each crop were calculated, the mean per trap and hence the total fall in each crop. As there was a clear tendency for a small number of traps to carry a large catch, a problem arose in obtaining confidence limits to the means. Most of the sets of data were skewed and found to correspond to a gamma distribution. Although, for statistical reasons, it was only possible to check this with a chi-squared test on 32 sets, only four of these differed from a gamma distribition at the 5% level. For simplicity therefore, all sets of data were assumed to have a gamma distribution. The crops per hectare and their 95% confidence limits 3:1 (1980) flesh of unripe drupes endosperm Ivy flesh from ripe berries . . . endosperm (from ripe berries) Hazel endosperm Ash ±0.31 ±0.23 ±0.37 ±0.31 Tab. 4. The weight and calorific content, per crop, of the components of each fruit falling to the ground in the yew wood. Component Yew ripe aril alt shell (including partially formed arils but excluding shell fragments) shell fragments endosperm Total Oak acorn shells endosperm Total Holly ripe flesh of drupes unripe flesh of drupes all shells endosperm Total Ivy ripe flesh of berries unripe flesh of berries endosperm Total Hazel hazelnul shells endosperm Total Ash wings (shells) endosperm Total Cherry flesh shells endosperm Total Rowan flesh shells endosperm Total Weight (kg ha-') Crop 1 Crop 2 Bicontent (10^ kcal ha"') Crop 1 Crop 2 QAt o.n be the same as for shells (with arils) of seeds with partially formed coats. For shell fragtnents: 1 greater than half a whole -I- 1 less than half = 1 shell; remaining fragments less than half -^3 = whole shells. The mean weight of shell here was given by averaging values from shells from ripe seeds with arils and those without. For holly the mean weight of shell and endosperm from unripe fruits were assumed to be equal to those from ripe ones. Values for unripe ivy berries were similarly assumed to be equivalent to those of ripe berries. The 'losses' to small mammals of seed and flesh of unripe holly were estimated from counts of seeds opened (all of which were assumed to contain endosperm) and assessment of missing flesh. As these were unquestionably available to small mammals, they have been included in Tab. 5. From feeding experiments (Smal and Fairley 1980) it had been determined that all endosperm and flesh was palatable to small mammals with the following exceptions; unripe holly flesh and all parts of ivy fruit to both HOLARCTIC ECOLOGY 3:1 (1980) Tab. 5. The weight and calorific content, per crop, of the components of each fruit falling to the ground in Ihe oak wood. Weight (kg ha"') Crop 1 Crop 2 Oak acorn shells endosperm Total Holly ripe flesh of drupes unripe flesh of drupes all shells endosperm Total Bicontent (10^ kcal ha"') Crop 1 Crop 2 10.14 28.30 38.44 9.76 40.43 6.13 8.41 IM 8.05 60.08 82.44 0.25 0.00 1.02 1.27 176.46 492.36 668,82 4,97 Ivy ripe flesh of berries unripe flesh of berries endosperm Total Tab. 6. The total biocontent of fruit flesh and endosperm falling to the ground in the two study areas and the fractions available as food to mice and voles. Component Yew wood Crop 1 Biocontenl (10^ kcal ha"') Oak wood Crop 2 Crop 1 12.79 591.26 604.05 497.34 44.24 0.00 541.57 89.66% Crop 2 51.74 112.88 164.62 38.06 77.70 0.00 115.76 70.32% Total fruit flesh Total seed endosperm Total Total palatable to both species . . . Total palatable to mice but not voles (holly endosperm) Total palatable to voles but not mice (ash endosperm) Total palatable to small mammals Percentage palatable of total 55.00 284.66 339.66 202,03 5.99 0.05 208.07 61.26% 98.46% wood mice and bank voles, holly seeds to voles and ash keys to mice. Small quantities of ash may be eaten by wood mice and similarly the flesh of ivy might be eaten by voles in minor amounts but the error involved is negligible. In this light. Tab. 6 gives the total biocontent of fruit falling to the ground (including the unripe holly eaten in the canopy) available as food to both wood mice and bank voles. The amounts of fruit flesh eaten by birds was calculated from numbers of the naked seeds in the traps of yew, holly, ivy and cheny. From observations on feeding activity, the fruits involved were invariably ripe and therefore appropriate mean weights and calorific contents of fruit flesh could be applied with confidence. The results are given in Tab. 7 along with the data for holly seeds removed by small mammals in the canopy. 2 HOLARCTtC ECOLOGY 3:1 (1980) 5. Discussion Although there are obvious sampling and statistical errors inherent in this study, and the fruit fall in Tomies wood is likely to be underestimated through the unknown extent of feeding on acorns by woodpigeons, the figures for fruit fall potentially available as food to small mammals for both woods are undoubtedly close approximations to the true values. As fruit represents the main food of small rodents in both woods, it is clear that there is considerable variation in the food supply from year to year. At Reenadinna, that from the second crop was five times that in the first and at Tomies the situation was almost the reverse. It seems reasonable to infer that such marked fluctuations in the food supply of bank voles Tab. 7. Estimates of the weight and caloriHc value of the fruit removed by birds and small mammals in the canopy in both study areas. Component Weight (kg ha-') Crop 1 Crop 2 Bicontent (10^ kcal ha"') Crop 1 Crop 2 Yew ripe arils eaten by birds in the yew wood percentage of lotal ripe aril Holly ripe hoily drupe flesh eaten by birds in the yew wood percentage of total ripe flesh ripe holly drupe flesh eaten by birds in the oak wood percentage of total ripe flesh unripe holly drupe flesh removed by small mammals percentage of total unripe flesh holly seed endosperm eaten by small mammals percentage of total endosperm Ivy ripe berry flesh eaten by woodpigeons in the yew wood percentage of total eaten by woodpigeons in the oak wood percentage of total Cherry ripe cherry flesh eaten by birds in the yew wood percentage of total ripe flesh 2.69 35.1% 15.52 43.3% 13.65 35.1% 78.68 43.3% 2.28 96.4% 16.22 94.1 % 0.01 3.6% 0.00 0.0% 3.92 98.9% 22.62 91.9% 3.23 40.1% 1.64 13.3% 11.15 96.4% 79.37 94.1% 0.03 3.6% 0.00 0.0% 19.18 98.9% 110.68 91.9% 16.20 40.1% 10.32 13.3% 6.58 58.8% 5.30 82.7% 0.90 67.8% 0.59 70.4% 41.20 58.8% 33.16 82.7% 5.61 67.8% 3.70 70.4% 0.02 100.07o 0.11 83.3% 0.11 100.0% 0.53 83.3% and wood mice in woodland are quite usual, yet comparable variation in peak densities of the animals is apparently rare. In a later publication we will examine the effect of abundance of food on population densities of mice and voles in these woods. This study exposes the dangers inherent in explaining fluctuations in the numbers of woodland rodents on the basis of casual observations on seed fall, especially that of the main tree species in the canopy alone. Fall of fruit is locally exceedingly variable (hence the gamma distribution); some species are unpalatable and, of those which are, not all parts are eaten; calorific values of fruits differ considerably; in some years the crop from the understorey is of more importance than that from the canopy (for example in 1976 in Tomies wood); a large proportion of fruits may contain no endosperm (for example hazelnuts); and a significant part of the crop may be eaten by small mammals in the canopy or removed there by birds. Acknowledgements - We are indebted to E. Thomson and the staff of tbe Bourn Vincent Memorial Park for assistance in producing seed traps. We would also like to thank Professor J. N. R. Grainger for the use of a bomb calorimeter and Mr P. Stafford for assistance in its operation. Helpful advice is gratefully acknowledged from Dr J. Shields on application of the gamma distribution and Mr M, Shaw on construction and use of seed traps. This investigation was financed by the Office of Public Works, Dublin. References Drozdz, A. 1967. Food preference, food digestibility and natural food supply of small rodents. - In: Petrusewicz, K. (ed.), Secondary productivity of terrestrial ecosystems. Warsaw. Grodziiiski, W. and Sawicka-Kapusta, K. 1970. Energy values of tree seeds eaten by small mammals. - Oikos 21: 52-58. Kelly, D. 1975. Native woodland in western Ireland with special reference to the region of Killamey. - Ph. D. thesis. University of Dublin. Newbould, P. J. 1967. Methods of estimating the primary production of forests. - Blackwell. Oxford. Smai, C. M. and Fairley, J. S. 1980. Food of wood mice and bank voles in oak and yew woods at Killamey, Ireland. - J, Zool., Lond. (in press). HOIj^RCnC ECOLOGY 3:1 (1980)

Journal

EcographyWiley

Published: Jan 1, 1980

There are no references for this article.