Looking into the past – the reaction of three grouse species to climate change over the last million years using whole genome sequences

Tracking past population fluctuations can give insight into current levels of genetic variation present within species. Analysing population dynamics over larger timescales can be aligned to known climatic changes to determine the response of species to varying environments. Here, we applied the Pairwise Sequentially Markovian Coalescent (psmc ) model to infer past population dynamics of three widespread grouse species; black grouse, willow grouse and rock ptarmigan. This allowed the tracking of the effective population size (N_e) of all three species beyond 1 Mya, revealing that (i) early Pleistocene cooling (~2.5 Mya) caused an increase in the willow grouse and rock ptarmigan populations, (ii) the mid‐Brunhes event (~430 kya) and following climatic oscillations decreased the N_e of willow grouse and rock ptarmigan, but increased the N_e of black grouse and (iii) all three species reacted differently to the last glacial maximum (LGM) – black grouse increased prior to it, rock ptarmigan experienced a severe bottleneck and willow grouse was maintained at large population size. We postulate that the varying psmc signal throughout the LGM depicts only the local history of the species. Nevertheless, the large population fluctuations in willow grouse and rock ptarmigan indicate that both species are opportunistic breeders while black grouse tracks the climatic changes more slowly and is maintained at lower N_e. Our results highlight the usefulness of the psmc approach in investigating species’ reaction to climate change in the deep past, but also that caution should be taken in drawing general conclusions about the recent past.     

Where is the pulse to have the finger on? A retrospective analysis of two decades of Alpine Galliforms (Aves: Galliformes) census and game bag data in Italy

Information on the abundance of the Italian populations of black grouse (Lyrurus tetrix), Alpine rock ptarmigan (Lagopus muta helvetica) and Alpine rock partridge (Alectoris graeca saxatilis) rely only on extrapolations of local data to the national scale, since there is no national standardized survey. Consequently, their status is virtually unknown. We performed a first-ever assessment of a medium-term (1996–2014) population trend of these species using and comparing post-breeding count and bag data at hunting district scale. These data were collected from various authorities in charge of wildlife management and allowed us to test the influence of hunting policies on the estimated trends. Rock partridge showed a stable trend with numbers fluctuating between years, while there was evidence of a severe decline for rock ptarmigan. No general conclusion could be drawn for the black grouse, as we detected lack of consistency of count and bag data. Counts were greatly overdispersed as a result of an uneven count effort among hunting districts. Adding the game management authority as model covariate resulted in more robust trend estimations, suggesting a significant effect of different policies that emerged also as similar hunting pressure across species within authorities. Hunting effort variation over the time was instead negligible. Species-specific game management bias is discussed. Our results highlight the need for a survey scheme or guidelines to be applied uniformly at a national scale.     

Past and potential future population dynamics of three grouse species using ecological and whole genome coalescent modeling

Studying demographic history of species provides insight into how the past has shaped the current levels of overall biodiversity and genetic composition of species, but also how these species may react to future perturbations. Here we investigated the demographic history of the willow grouse (Lagopus lagopus), rock ptarmigan (Lagopus muta), and black grouse (Tetrao tetrix) through the Late Pleistocene using two complementary methods and whole genome data. Species distribution modeling (SDM) allowed us to estimate the total range size during the Last Interglacial (LIG) and Last Glacial Maximum (LGM) as well as to indicate potential population subdivisions. Pairwise Sequentially Markovian Coalescent (PSMC) allowed us to assess fluctuations in effective population size across the same period. Additionally, we used SDM to forecast the effect of future climate change on the three species over the next 50 years. We found that SDM predicts the largest range size for the cold‐adapted willow grouse and rock ptarmigan during the LGM. PSMC captured intraspecific population dynamics within the last glacial period, such that the willow grouse and rock ptarmigan showed multiple bottlenecks signifying recolonization events following the termination of the LGM. We also see signals of population subdivision during the last glacial period in the black grouse, but more data are needed to strengthen this hypothesis. All three species are likely to experience range contractions under future warming, with the strongest effect on willow grouse and rock ptarmigan due to their limited potential for northward expansion. Overall, by combining these two modeling approaches, we have provided a multifaceted examination of the biogeography of these species and how they have responded to climate change in the past. These results help us understand how cold‐adapted species may respond to future climate changes.     

Searching for mechanisms of synchrony in spatially structured gamebird populations

1.  Time series data on five species of gamebird from the Dolomitic Alps were used to examine the relative importance of dispersal and common stochastic events in causing synchrony between spatially structured populations.
2.  Cross-correlation analysis of detrended time series was used to describe the spatial pattern of fluctuations in abundance, while standardized time series were used to describe both fluctuations and the trend in abundance. There were large variations in synchrony both within and between species and only weak negative relationships with distance.
3.  Species in neighbouring habitats were more likely to be in synchrony than species separated by several habitats. Species with similar density-dependent structure were more likely to be in synchrony.
4.  In order to estimate the relative importance of dispersal and environmental stochasticity, we modelled the spatial dynamics of each species using two different approaches. First, we used estimating functions and bootstrapping of time series data to calculate the relative importance of dispersal and stochastic effects for each species. Second, we estimated the intensity of environmental stochasticity from climatic records during the breeding season and then modelled the dispersal rate and dispersal distance for each species. The two models exhibited similar results for rock ptarmigan, black grouse, hazel grouse and rock partridge, while contrasting patterns were observed for capercaillie.
5.  The results suggest that environmental stochasticity plays the dominant role in synchronizing the fluctuations of these galliform species, although there will also be some dispersal between populations.     

Small-scale habitat use of black grouse (Tetrao tetrix L.) and rock ptarmigan (Lagopus muta helvetica Thienemann) in the Austrian Alps

We investigated the small-scale habitat use of two grouse species, black grouse (Tetrao tetrix L.) and rock ptarmigan (Lagopus muta helvetica Thienemann) in a study area in the Austrian Central Alps in summer. To build habitat suitability models, we applied multiple logistic regression using presence–absence data from fieldwork as the response variable and a set of habitat characteristics as explanatory variables, respectively. To gain a better understanding of the mechanisms that drive habitat selection, we tested for two-way interaction terms before excluding any variables from the initial variable set. Four explanatory variables significantly contributed to the black grouse model: dwarf shrub cover, dwarf shrub height, patchiness and ant hills. The final model for rock ptarmigan contained three explanatory variables: dwarf shrub cover, rock cover and dwarf shrub height. Most notably, the interaction terms dwarf shrub cover × patchiness in the black grouse model and dwarf shrub cover × dwarf shrub height, rock cover × dwarf shrub height in the rock ptarmigan model point out trade-off mechanisms between food, cover and overview providing features. Thus, our models do not only identify the parameters that mainly drive habitat selection, but also deepen our understanding about the causal relationships between these factors. Therefore, the information gained in this study allows for a deduction of appropriate habitat management strategies and supports conservation efforts of local stakeholders.     

Territory occupancy rate of goshawk and gyrfalcon: no evidence of delayed numerical response to grouse numbers

Two recent studies on territory occupancy rates of goshawk Accipiter gentilis and gyrfalcon Falco rusticolus report a 2–3-year-delayed numerical response to grouse numbers, which is a requirement for a hypothesis of predator-generated grouse cycles. The time lags were assumed to reflect the average age of sexual maturity in the raptor species. In southern Norway, however, subadult (two-year-old) goshawk hens occupied only 18–25% of territories where occupancy was not recorded in the preceding year, and there was no significant relationship between the proportion of subadults among recruits and grouse indices two years earlier. We argue that territory occupancy rates are not appropriate indices of total raptor population levels, but rather reflect the proportion of territorial pairs that attempt to nest. Because this depends on the body condition of the hens, fluctuations in other important winter resident prey species (most important for the goshawk) and winter weather (most important for the gyrfalcon) should also be addressed. During 1988–2006, the annual proportion of goshawk territories with recorded nesting attempts in southern Norway was most closely related to the preceding autumn’s population indices of black grouse Tetrao tetrix and mountain hare Lepus timidus, whereas the annual proportion of gyrfalcon territories with observations of falcons or with confirmed breeding attempts in central Norway were best explained by population indices of willow grouse Lagopus lagopus and ptarmigan L. mutus from the previous autumn, and by December temperatures. Hence, our studies do not support the predation hypothesis for grouse cycles.     

Grouse dynamics and harvesting in Kainuu,northeastern Finland

We study the dynamics of the capercaillie, black grouse, hazel grouse and willow grouse in Kainuu game management district in northeastern Finland in the years 1989–2004. It appears that the 6–7 year periodicity that prevailed in this region from 1960s up to 1980s has now vanished in all species. The grouse data are modelled using a linear autoregressive model with lag terms for population dynamics including grouse harvest as annual bag and an index of winter severity (winter‐time area of Baltic Sea ice cover). We use the Akaike information criterion for selecting the best model for each species; first order lag is forced to the models. It turns out that a term is needed for harvesting (with a negative coefficient) in models for all species. For the capercaillie and the hazel grouse second order lag was included, for the black grouse and the willow grouse first order lag suffices. The willow grouse is the only species where the index of winter strength (with a negative coefficient) is needed in the model.     

Coupling in goshawk and grouse population dynamics in Finland

Different prey species can vary in their significance to a particular predator. In the simplest case, the total available density or biomass of a guild of several prey species might be most relevant to the predator, but behavioural and ecological traits of different prey species can alter the picture. We studied the population dynamics of a predator–prey setting in Finland by fitting first-order log-linear vector autoregressive models to long-term count data from active breeding sites of the northern goshawk (Accipiter gentilis; 1986–2009), and to three of its main prey species (1983–2010): hazel grouse (Bonasa bonasia), black grouse (Tetrao tetrix) and capercaillie (T. urogallus), which belong to the same forest grouse guild and show synchronous fluctuations. Our focus was on modelling the relative significance of prey species and estimating the tightness of predator–prey coupling in order to explain the observed population dynamics, simultaneously accounting for effects of density dependence, winter severity and spatial correlation. We established nine competing candidate models, where different combinations of grouse species affect goshawk dynamics with lags of 1–3 years. Effects of goshawk on grouse were investigated using one model for each grouse species. The most parsimonious model for goshawk indicated separate density effects of hazel grouse and black grouse, and different effects with lags of 1 and 3 years. Capercaillie showed no effects on goshawk populations, while the effect of goshawk on grouse was clearly negative only in capercaillie. Winter severity had significant adverse effects on goshawk and hazel grouse populations. In combination, large-scale goshawk–grouse population dynamics are coupled, but there are no clear mutual effects for any of the individual guild members. In a broader context, our study suggests that pooling data on closely related, synchronously fluctuating prey species can result in the loss of relevant information, rather than increased model parsimony.     

Are indirect measures of abundance a useful index of population density? The case of red grouse harvesting

Indirect measures of population abundance, such as harvest data, are often used to make inference on long term population dynamics when direct data are either not available or are logistically difficult to obtain. However, when harvesting records are used, a common concern is that they may not reflect actual population abundance. We investigated the extent to which harvest data reflected changes in population density of the red grouse Lagopus lagopus scoticus in Great Britain. We used 92 independently managed populations over the period 1977–2000 and examined the temporal and spatial variability of the hunting records and independently obtained count data from each of these managed estates. Three different analyses support the conclusion that grouse hunting records are a reliable indicator of grouse abundance: 1) the number of red grouse shot in autumn showed a tendency to be linearly related to the density of individuals counted in the summer prior to the harvesting, 2) the relationships between the variance and the mean in the harvesting and corresponding count data, calculated over different populations at the same time, or the same locations at different times, were not statistically distinguishable, 3) similar direct and delayed density dependence patterns were observed in hunting records and count data. Our results suggest that red grouse hunting time series are a good proxy for population abundance.     

Goshawk predation during winter, spring and summer in a boreal forest area of central Sweden

Predation by goshawks was studied in a central Swedish boreal forest area. Data were collected in winter (January–February) 1977-81 by tracking radio-tagged goshawks, and in the breeding season (April–July) by collecting prey remains at the nest. In the breeding season birds dominated the prey, amounting to 86% of prey number and 91% of prey biomass. Wood pigeon Columba palumbus , black grouse Tetrao tetrix , hooded crow Corvus corone cornix and jay Garrulus glandarius accounted for more than 50% of the prey animals, whereas capercaillie Tetrao urogallus and black grouse accounted for more than 50% of prey biomass. There was no functional response to black grouse density fluctuations. Every year goshawks killed significantly more females than males of both capercaillie and black grouse, due to high vulnerability of the grouse hens while laying and incubating. It was estimated that during spring and early summer goshawk predation removed 25% of the female, and 14% of the male black grouse population. In winter squirrel was the dominating prey, both in terms of number (79%) and weight (56%). The proportion of squirrel in the diet was equally high both in winters of low and high squirrel density. The high proportion of squirrel in the winterdiet, as compared to the breeding season, is believed to be due to squirrels having to accept an increased predation risk in winter, in order to feed efficiently enough.     

Short- and long-term population dynamical consequences of asymmetric climate change in black grouse

Temporal asymmetry in patterns of regional climate change may jeopardize the match between the proximate and ultimate cues of the timing of breeding. The consequences on short- and long-term population dynamics and trends as well as the underlying mechanisms are, however, often unknown. Using long-term data from Finland, we demonstrate that black grouse (Tetrao tetrix) have responded to spring warming by advancing both egg-laying and hatching. However, early summer (the time of hatching) has not advanced, and chicks have to face colder post-hatching conditions. Demonstrating that these conditions are critical to post-hatching survival, we show that chicks are increasingly suffering higher mortality because they hatch too early. Consequently, breeding success and population size has severely declined over the past four decades. Finally, we modelled the impact of this particular climate change scenario on population dynamics and show that the mismatch can further explain the observed collapse of cyclic fluctuations. Because the evolutionary response of grouse is lagging behind the novel selective pressures, seasonally asymmetric climate change is likely to constitute an important determinant of future short- and long-term changes in the dynamics of black grouse populations.     

Sex and age influence intestinal parasite burden in three boreal grouse species

Parasite infections are often more common in male vertebrates than in females. Sexual selection leading to dimorphism can cause sexual differences in immune defence, behaviour or body size. Possible proximate explanations for male bias in parasitism are the immunosuppressive effect of male sex hormones and the large body size of males which increases the likelihood of being parasitized. To evaluate these hypotheses, we studied the prevalence and abundance of intestinal helminth parasites in three boreal grouse species, the capercaillie Tetrao urogallus , the black grouse Tetrao tetrix , and the hazel grouse Bonasa bonasia . The first two are sexually dimorphic polygynous species while the latter species is sexually monomorphic and monogamous. We found a male-bias in the prevalence and abundance of the nematode Ascaridia compar . The bias was most pronounced in the polygynous black grouse and capercaillie. In the monogamous hazel grouse, there was a slight male-bias in occurrence of ascarids, but no bias in abundance. In juvenile grouse, the male-bias was larger than in adult grouse. No sexual bias was found in regard to the cestodes ( Skrjabinia cesticillus , Paroniella urogalli and Hymenolepis spp.). However, age was a factor in cestode prevalence: juvenile grouse were more commonly infected than adults. Differences in growth rates and body size are potential factors that may lead to male-biased parasitism in these grouse species, and their impact requires further studies.     

Cycles in voles and small game in relation to variations in plant production indices in Northern Sweden

Population dynamics for voles (Cricetidae), Tengmalm's owl (Aegolius funereus (L.)), red fox (Vulpes vulpes (L.)) willow grouse (Lagopus lagopus (L.)), black grouse (Lyrurus tetrix (L.)), capercaillie (Tetrao urogallus L.), hazel hen (Tetrastes bonasia (L.)), mountain hare (Lepus timidus L.) and tularemia (Francisella tularensis (McCoy & Chapin)) and game bird recruitment were studied by index methods in northern Sweden. In addition contemporary temperature records and spruce (Picea abies (L.) Karst.) and pine (Pinus silvestris L.) cone crops (as indices for plant production) and the occurrence of forest damage, caused by voles eating bark, were studied.During 1970–80 two synchronous 4-year cycles were observed for voles, predators (Tengmalm's owl and red fox) and their alternative prey species (grouse and mountain hare). In grouse the change of numbers was correlated with that of recruitment. Autumn vole numbers peaked about a year before the other species and extensive forest damage occurred at winter peak densities of voles. These population fluctuations are consistent with a predator-prey model for their regulation. In short the model suggests that vole-food plant interactions trigger the cycle of voles, that voles generate the cycle of predators and that these in turn synchronize alternative prey populations to the others at vole declines.For voles, grouse and red fox the amplitude was higher in the first cycle compared to the second one whilst the opposite was true for the mountain hare. Although temperature and cone crops showed large interannual variations they still implied that herbivore food conditions were better during the former cycle. Hence, the reduction of the amplitude of the vole cycle may be explained by inter-cyclic differences in plant food conditions, implying food shortage (as indicated by bark-eating) at different population levels. The similar decrease of grouse and red fox populations may also be explained by deteriorated food conditions and/or for the fox by an outbreak of sarcoptic mange (Sarcoptes scabiae var. vulpes). The increased amplitude of the mountain hare cycle was part of a long-term rise in numbers after a tularemia epidemic in 1967. This is interpreted as a recovery, probably towards the generally higher pre-epidemic population level.     

Pyrenean ptarmigans decline under climatic and human influences through the Holocene

In Europe, the Quaternary is characterized by climatic fluctuations known to have led to many cycles of contraction and expansion of species geographical ranges. In addition, during the Holocene, historical changes in human occupation such as colonization or abandonment of traditional land uses can also affect habitats. These climatically or anthropically induced geographic range changes are expected to produce considerable effective population size change, measurable in terms of genetic diversity and organization. The rock ptarmigan (Lagopus muta) is a small-bodied grouse occurring throughout Northern hemispheric arctic and alpine tundra. This species is not considered threatened at a continental scale, but the populations in the Pyrenees are of concern because of their small population size, geographical isolation and low genetic diversity. Here, we used 11 microsatellites to investigate genetic variations and differentiations and infer the overall demographic history of Pyrenean rock ptarmigan populations. The low genetic variability found in these populations has been previously thought to be the result of a bottleneck that occurred following the last glacial maximum (i.e., 10 000 years ago) or more recently (i.e., during the last 200 years). Our results clearly indicate a major bottleneck affecting the populations in the last tenth of the Holocene. We discuss how this decline can be explained by a combination of unfavorable and successive events that increased the degree of habitat fragmentation.     

How reliable are harvesting data for analyses of spatio‐temporal population dynamics?

Harvest data are commonly used as proxy for count data, especially in studies of long‐term temporal and spatial patterns of population fluctuations. However, usually the concurrence of the conclusions based on different types of data is impossible to verify due to the lack of count data. Here, we use annual (1964–2004) harvest and population census data for capercaillie, black grouse and hazel grouse from 14 game management districts covering Finland, and demonstrate some mismatch in the information that these data sets provide. Overall, linear regressions of annual harvest against population count give a reasonable fit, but the slopes are less than 1 in every species. Harvest bags have been proportionally larger in north and eastern Finland than in southwestern Finland, with marked species‐specific differences. Considering population variation, the CV% in the census data (30–50%) is consistently smaller than it is in the harvest data (60–70%). Most importantly, conclusions on the spatio‐temporal patterns of the population dynamics are different if based on harvest rather than count data. In capercaillie, synchrony decreases faster with distance according to the harvest data, while in black grouse and hazel grouse the census data show the steeper decline. In addition, the autocorrelation coefficients in the census time series are higher in capercaillie and black grouse than in harvest data, but in hazel grouse the opposite is true. Finally, the parameter estimates for a second order autoregressive model using different data sets differ, and these differences are species‐specific. Despite the fact that annual harvest is a positive and linear function of annual grouse population density, the pattern of population dynamics derived from the bag data is different from that shown by the census data. This result urges caution in using wildlife bag data as reliable indices of population dynamics. deceased August 2008.     

Patterns in grouse and Woodcock Scolopax rusticola hunting yields from central Norway 1901–24 do not support the alternative prey hypothesis for grouse cycles

According to the alternative prey hypothesis, autumn populations of ground-nesting game birds fluctuate in synchrony with vole numbers because generalist predators that mainly eat voles switch to alternative prey, such as eggs and chicks, when vole numbers decline. In hunting statistics from Nord-Trøndelag, central Norway, 1901–24, annual fluctuations in the number of Willow Grouse Lagopus lagopus and Western Capercaillie Tetrao urogallus , but not of Woodcock Scolopax rusticola , were positively related to vole numbers in the current year. Both Woodcock and grouse indices were related to hunting indices of Goshawk Accipiter gentilis and to weather variables assumed to influence the birds' survival or reproduction, suggesting that the indices actually reflected local population levels. Synchronous vole and grouse fluctuations are consistent with the alternative prey hypothesis (although predator densities were low in the early 1900s), but the asynchronous Woodcock fluctuations refute the hypothesis. Rather, because the Woodcock does not feed on plants utilized by voles and grouse, I suggest that food quality is the ultimate factor for the synchrony in vole and grouse numbers in Norway.     

Was high grouse hag in early 20th century Norway due to a program for extermination of small game predators?

A Norwegian program for extermination of (small game) predators (NPEP) was run during 1900‐ 1914, This initiative is believed to have caused larger small game stocks and more regional synchrony in rodents. To investigate the effectiveness of the NPEP time series of predators bountied (1885–1914), rodent dynamics (1885–1914), ptarmigan hunting index (1885–1914 and 1900/1‐1914), and willow ptarmigan and berry export statistics (1885–1914) were analyzed for three different regions: south, east and central Norway. In south and east Norway there were higher ptarmigan export in the period 1907–1914 than the years before, but not in central Norway. There was not bountied higher number of red fox. eagle or goshawk in any of the three regions when comparing years a) before and after 1900. and b) when comparing the periods 1900–1906 and 1907–1914. This suggests that predator removal was not the cause of increased ptarmigan export. The ptarmigan hunting index and rodent index for south and east Norway were correlated in the period 1885 1914, while the ptarmigan export from east Norway was correlated with berry export from the year before. However, for central Norway the rodent index was not correlated with the hunting index. There were cross correlation between berry and ptarmigan export with lag from minus one to nine years for south and east Norway, The NAO (North Atlantic Oscillation) ‐ an indication of the winter weather variation ‐ had higher values during 1900–1914 than 1885–1899, indicating moister and warmer winters in the last period. This analysis indicates that NPEP generally did not increase predator removal. The results suggest, however, that it was a series of years with high rodent density, that caused the increase in ptarmigan populations In south and cast Norway, which, in turn, may have been caused by favourable weather conditions leading to among others good berry crops. Conclusions based on old data series must, however, be drawn with caution.     

State‐dependent dynamics of cycles in red grouse abundance

Fluctuating populations are frequently demonstrated to co‐vary in abundance over space, but the dynamics of coupling between populations that gives rise to this synchrony are poorly understood. Synchrony may arise through coupling that is weak and continuous, but in populations that cycle with a characteristic period, synchrony can be maintained through stronger coupling that acts only intermittently. Here, we apply a discrete Markov model that describes the state of a population trajectory to be in one of four possible states. The Markov model reveals the nature of the coupling that gives rise to the weakly synchronous cycles of red grouse abundance. Using time‐series data from 287 populations across the species range in the UK, we show that grouse populations appear mostly uncoupled through time, but that approximately one year in six, “collective forcing events” occur, where populations in a region are forced into synchrony to a significantly greater degree than would be expected if their dynamics proceeded independently. In the absence of these events, synchrony between populations dissipates within ~3 yr. Smaller, low abundance populations tend to make the less probable phase shifts required to synchronize with nearby high abundance populations, suggesting that these low abundance populations are more susceptible to the perturbations responsible for phase shifts than larger populations.     

Synchrony, scale and temporal dynamics of rock partridge (Alectoris graeca saxatilis) populations in the Dolomites

1. Harvesting records of rock partridge ( Alectoris graeca saxatilis ) were examined first to identify the presence of cycles in a species with a southern European distribution and then to examine synchrony between populations at a range of scales.
2. Hunting records from 1965 to 1994 were obtained from 210 hunting areas and analysed at three spatial scales: subpopulation, population and metapopulation. Rock partridge exhibited cyclic fluctuations in about 40% of the time series with a period of 4–7 years. The results did not change with spatial scale. The density-dependent structure of the populations showed that most populations exhibited damped oscillations.
3. The proportion of populations that were in synchrony increased with scale from the population to metapopulation level. There was no decline in synchrony with distance but a large variation between populations irrespective of distance.
4. The populations clustered into dry and wet habitats, with those in the dry habitat being more cyclic. We suggest the lack of spatial synchrony with distance but greater synchrony within habitats may reflect the influence of stochastic events operating on populations with different density dependence structures.     

Diet of the golden eagle Aquila chrysaetos during the breeding season in Sweden

During a five-year period, 1975–1979, a total of 2881 prey individuals of 65 prey species were collected at 162 golden eagle nests from northern Sweden and from the island of Gotland. In northern Sweden birds are taken in higher numbers than mammals but calculated as weight the two categories are of equal importance, The main prey during the breeding season are capercaillie, black grouse, willow grouse, ptarmigan, mountain hare and reindeer fawns which together form 91% of the total food biomass. The capercaillie and the black grouse are taken more in the southern part of the coniferous region than in the northern. In contrast, in northern areas, reindeer fawns are more preyed upon than in the South, Ptarmigan and willow grouse are the most commonly captured prey species in mountain areas. The total number of reindeer fawns taken (dead and/or alive) by the Swedish golden eagle population during one summer is estimated at 600 individuals. On Gotland the golden eagles take mammals more often than in its northern distribution area. Rabbit and hedgehog arc the most important species.