UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The lemming cycle at Baker Lake, N.W.T., during 1959-61 Krebs, Charles J. 1962

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1962_A1 K7 L3.pdf [ 11.32MB ]
Metadata
JSON: 831-1.0302572.json
JSON-LD: 831-1.0302572-ld.json
RDF/XML (Pretty): 831-1.0302572-rdf.xml
RDF/JSON: 831-1.0302572-rdf.json
Turtle: 831-1.0302572-turtle.txt
N-Triples: 831-1.0302572-rdf-ntriples.txt
Original Record: 831-1.0302572-source.json
Full Text
831-1.0302572-fulltext.txt
Citation
831-1.0302572.ris

Full Text

THE LEMMING CYCLE AT BAKER LAKE, N.W.T., DURING 1959-61 by CHARLES J . KREBS B.S., U n i v e r s i t y of Minnesota, 1 °57 M.A., Un i v e r s i t y of B r i t i s h Columbia, 195? A Thesis Submitted i n P a r t i a l F u l f i l m e n t of The Requirements f o r the Degree of Doctor of Philosophy In the Department of Zoology We accept t h i s t h e s i s as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA A p r i l , 1962 In presenting t h i s t h e s i s i n p a r t i a l f u l f i l m e n t of the requirements f o r an advanced degree at the U n i v e r s i t y of B r i t i s h Columbia, I agree tha t the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r reference and study. I f u r t h e r agree that permission f o r extensive copying of t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the Head of my Department or by h i s representatives. It i s understood that copying or p u b l i c a t i o n of t h i s t h e s i s f o r f i n a n c i a l gain s h a l l not be allowed without my written permission. Department of Zoology  The U n i v e r s i t y of B r i t i s h Columbia, Vancouver 8, Canada. Date A p r i l 1. 1962 PUBLICATIONS Krebs, C. J. 1961 Population dynamics of the Mackenzie Delta rein-deer herd, 1938-1958. Arctic 14 (2): 91-100. F A C U L T Y O F G R A D U A T E S T U D I E S mi P R O G R A M M E O F T H E F INAL O R A L E X A M I N A T I O N FOR THE DEGREE OF D O C T O R OF PH I LOSOPHY of CHARLES JOSEPH KREBS B.S., University of Minnesota, 1957 M.A., University of British Columbia, 1959 IN ROOM 3332, BIOLOGICAL SCIENCES BUILDING THURSDAY, APRIL 19th, 1962 AT 3:00 P.M. COMMITTEE IN CHARGE Chairman: F. H. SOWARD J. F. BENDELL Z. A. MELZAK D. H. CHITTY W. B. SCHOFIELD I. McT. COWAN T. M. C. TAYLOR H. B. HAWTHORN N. J. WILIMOVSKY W. S. HOAR A. J. WOOD External Examiner: D. G. CHAPMAN University of Washington THE LEMMING CYCLE AT BAKER LAKE, N.W.T., DURING 1959-61 A three year study covering one cycle in numbers of the brown lemming (Lemmus trimucrotiatus) and the varying lemming (Dicros-tonyx groenlandicus) has been carried out at Baker Lake, Keewatin, N.W.T. An attempt was made to describe the events of the cycle in detail by snap trapping and live trapping techniques and by detailed autopsies on about 3 >00 animals, and from this to determine what explanations would fit the observations. Increase began from very low numbers in the summer of 1959 with tremendous population growth occurring over the winter of 1959-60. Little further increase occurred in the peak summer of I960. A great decline occurred over the winter of 1960-61, and this decline continued through the summer of 1961 in the Main Study Area. Two changes in reproduction were associated with this cycle— changes in the length of the breeding season and in the weight at sexual maturity. Winter breeding occurred only in the period of increase, and a shortened summer breeding season occurred in the peak and to some degree in the decline. Young male Lemmus matured in the summer of increase but not in the peak or in the decline; young females matured in the increase and decline summers but not in the peak. The major change in mortality was a very high juvenile mortality in the summer of the decline. Changes in the quality of the animals were manifested not only by these reproductive and mortality changes but also by a 20-30% increase in mean body weights of the adults in the peak summer compared to the increase or decline summers. The role of the extrinsic factors is considered. There was no widespread destruction of the habitat, or any evidence of qualitative or quantitative food shortage in the animals of the decline. Neither predators, disease, nor parasites seemed to be the cause of the observed changes in mortality. The role of the instrinsic factors is also considered. Summer adrenal and spleen weights showed no clear relationship to the cycle. The amount of wounding on skins showed strong seasonal and yearly changes and was not a simple function of density. It was concluded from these observations that the lemming cycle could not be adequately explained by the conventional extrinsic factors such as food supply, but rather that it is essentially a self; regulatory phenomenon. The stress hypothesis proposed by Christian was also rejected as an adequate explanation of these events. The idea of Chitty that populations change in quality during changes in density was fully supported by these results. Behavioral changes in the population may constitute the crux of the lemming cycle, and Chitty's suggestion that these fluctuations may represent a genetic polymorphism involving aggressive behavior is rot refuted by these data. GRADUATE STUDIES Field of Study: Zoology Population Dynamics P. A. Larkin Wildlife Management L McT. Cowan Economic Entomology K. Graham Comparative Vertebrate Embryology P. Ford Quantitative Methods in Zoology P. A. Larkin Other Studies: Animal Growth and Nutrition A. J. Wood and J. Biely Plant Ecology V. J. Krajina Synoptic Oceanography G. L. Pickard Dynamic Oceanography G. L. Pickard Biogeography R. W. Pillsbury ABSTRACT A three year study covering one cycle i n numbers of the brown and varying lemmings has been c a r r i e d out at Baker Lake, Keewatin, N.W.T. An attempt was made t o describe the events of the cycle i n d e t a i l by snap trapping and l i v e trapping techniques and by d e t a i l e d autopsies on about 3U00 animals and from t h i s to determine what explanations would f i t the observations. Increase began from very low numbers i n the summer of 1°59 with tremendous population growth occurring over the winter of 1959-60. L i t t l e f u r t h e r increase occurred i n the peak summer of i960. A great decline occurred over the winter of l°60-6l, and t h i s decline continued through the summer of l°6l on the Main Study Area. Two changes i n reproduction were associated with t h i s c y c l e -changes i n the length of the breeding season and i n the weight at sexual maturity. Winter breeding occurred only i n the period of increase, and a shortened summer breeding season occurred i n the peak and to some degree i n the d e c l i n e . Young male Lemmus matured i n the summer of increase but not i n the peak or i n the decline j young females matured i n the increase and decline summers but not i n the peak. The major change i n m o r t a l i t y was a very high juvenile m o r t a l i t y i n the summer of the d e c l i n e . Changes i n the q u a l i t y of the animals were manifested not only by these reproductive and m o r t a l i t y changes but also by a 20-30% increase i n mean body weights of the adults i n the peak summer compared to the increase or decline summers. The r o l e of the e x t r i n s i c f a c t o r s i s considered. There was no widespread destruction of the habitat, or any evidence of quantitative or q u a l i t a t i v e food shortage i n the animals of the d e c l i n e . Neither predators, disease, nor p a r a s i t e s seemed to be the cause of the observed changes i n mo r t a l i t y . i i The r o l e of the i n t r i n s i c f a c t o r s i s also considered. Summer adrenal and spleen weights showed no c l e a r r e l a t i o n s h i p to the c y c l e . The amount of wounding on skins showed strong seasonal and y e a r l y changes and was not a simple f u n c t i o n of density. I t was concluded from these observations that the lemming cycle could not be adequately explained by the conventional e x t r i n s i c f a c t o r s such as food supply, but rather that i t i s e s s e n t i a l l y a s e l f - r e g u l a t o r y phenomenon. The stress hypothesis proposed by C h r i s t i a n was also r e j e c t e d as an adequate explanation of these events. The i d e a of C h i t t y that populations change i n q u a l i t y during changes i n density was f u l l y supported by these r e s u l t s . Behavioral changes i n the population may c o n s t i t u t e the crux of the lemming cy c l e , and C h i t t y ' s suggestion that these f l u c t u a t i o n s may represent a genetic polymorphism i n v o l v i n g aggressive behavior i s not refuted by these data. i i i TABLE OF CONTENTS Page INTRODUCTION 1 DESCRIPTION OF AREA AND CLIMATE 3 VEGETATION AND HABITATS 8 POPULATION DENSITY . . , „ * . . , . . . , Ik Methods i l l Live Trapping ll+ Snap Trapping 17 Other Census Methods 19 Results 20 Live Trapping 20 Snap Trapping . * 2i| Other Census Methods 30 Summary and Conclusions 30 REPRODUCTION 33 Methods 33 Results 36 Length of Breeding Season 36 L i t t e r Size . . . . . . . . h3 Pregnancy Rates $1 Age at Reproductive Maturity 57 Embryo Rates 6I4. Summary and Conclusions 68 MORTALITY 69 Methods 69 Results 69 i v Page Prenatal M o r t a l i t y 69 Post-natal M o r t a l i t y 73 Adults 73 Juveniles 83 Summary and Conclusions . • 89 MOVEMENTS AND MIGRATIONS 90 Methods 90, Results 90 L o c a l Movements 90 Migrations 91 Summary and Conclusions 98 CHANGES IN EXTRINSIC FACTORS 100 Weather 100 Predators 101 Disease and Parasites 103 Food IOI4. Summary and Conclusions . U O CHANGES IN INTRINSIC FACTORS I l l Methods I l l Age Determination I l l Mean Body Weights 112 Organ Weights l i l t Results 116 Body Weight D i s t r i b u t i o n s 116 Mean Body Weights 125 Organ Weights 129 Fat Changes 137 V Page S o c i a l Relationships 138 Summary and Conclusions li|3 BISCUSSION . . . 1U5 H i s t o r i c a l Approaches and Background Hi5 Reproduction • II48 M o r t a l i t y 152 Migrations iSk Weather and Synchrony 158 Predators . . . . . . . . 161 Disease and Parasites l 6 l Body Weight Changes 162 Three Current Hypotheses . . . . . . . . . . . 165 Food Supply hypothesis l6£ C h r i s t i a n ' s Stress Hypothesis 167 Chi t t y ' s Hypothesis . . . . . 170 Conclusions . * 17U SUMMARY 175 LITERATURE CITED 177 v i FIGURES Page Figure 1. Map showing the l o c a t i o n of Baker Lake, the Main Study Area, and the p e r i p h e r a l sampling areas k Figure 2. Location of the l i v e trapping quadrats of the Main Study Area 15 Figure 3. • Summer habitat d i s t r i b u t i o n i n Dicrostonyx and Le minus on the Main Study Area 29 Figure U. Generalized density changes 1959-61. Numbers indic a t e r e l a t i v e changes i n numbers f o r each species based on l i v e trapping 31 Figure 5« Components of reproduction i n polyestrous mammals 37 Figure 6. Generalized annual chronology of generations and l i t t e r s f o r Lemmus and Dicrostonyx 113 Figure 7. Body weight d i s t r i b u t i o n s f o r Lemmus males, July 16-31, I960 and 1961.. . . . . « B Figure 8. P i t e l k a ' s food supply hypothesis 166 Figure 9» C h r i s t i a n ' s s t r e s s hypothesis 169 Figure 10. C h i t t y 1 s hypothesis . 172: v i i TABLES Pa Table 1. Temperature and precipitation data during l°5>9-6l relative to the mean values for 1950-60 as recorded by the Baker Lake Meteorological Station . . . . . . . . . . 5 Table 2. Depth of snow on ground during early winter . . . 7 Table 3. Area covered by the principal habitat types on the Main Study Area . 10 Table h» Dominant plant species i n the habitats of the Main Study Area, New Lake, and Prince Mver . . . 11 Table 5« Dominant plant species in the habitats of the Aberdeen Lake area 12 Table 6. Dominant plant species i n the moss habitats on the islands of Baker Lake and the south bank of the The Ion mouth .13 Table 7» Numbers of Lemmus on Quadrat # 1 during 1959-61 21 Table 8. Numbers of Dicrostonyx on Quadrat # -3 i n 1960-61 22 Table 9. lemmus snap trapping indices, 1959-61 25 Table 10. Dicrostonyx snap trapping indices, 1959-61 . . . 27 Table 11. Timing of summer breeding periods i n Lemmus females, 1959-61 38 Table 12. Timing of summer breeding periods i n Dicrostonyx females, 1959-61 39 Table 13. Length of the summer breeding seasons of Lemmus and Dicrostonyx, Main Study Area, 1959-61. UO Table lli* Winter and spring breeding of Lemmus and Dicrostonyx, 1958-61 \x2 Table 15.- Number of corpora lutea i n Lemmus females, Main Study Area, summers 1959-61 h$ Table 16. Number of corpora lutea i n Dicrostonyx females, Main Study Area, summers 1959-61 . . . U6 Table 17. Number of embryos i n Lammus females, Main Study Area, summers 1959-61 i;7 v i i i Page Table 18. Number of embryos i n Dicrostonyx females, Main Study Area, summers 1959-61 U8 Table 19. Number of corpora l u t e a and embryos i n Lemmus females at Aberdeen Lake, summers 1960-61 I4.9 Table 20. Number of corpora l u t e a and embryos i n Dicrostonyx females at Aberdeen Lake, summers 1960-61 . . . . 50 Table 21. Crude pregnancy rates per 16 days per female >20.5 g, Lemmus, summers 1959-61 52 Table 22. Crude pregnancy rates per 15 days per female >30.5 g, Dicrostonyx, summers 1959-61 . . . . . . 5U Table 23. Weight at reproductive maturity i n Lemmus males, summers 1959-61 58 Table 21;. Weight at reproductive maturity i n Lemmus females, summers 1959-61 . 59 Table 25. Weight at reproductive maturity i n Dicrostonyx males, summers 1959-61 60 Table 26. Weight at reproductive" maturity i n Dicrostonyx females, summers 1959-61 61 Table 27. Median body weights at maturity f o r Lemmus and Dicrostonyx males and females, 1959-61 . . . . . . 63 Table 28. Crude embryo rates per 16 days per female >20.5 g, Lemmus, 1959-61 65 Table 29. Crude embryo rates per 15 days per female >30.5 g, Dicrostonyx, 1959-61 66 Table 30. P a r t i a l p r e n a t a l m o r t a l i t y data f o r Lemmus females, summers 1959-61, Main Study Area . . . . 71 Table 31. P a r t i a l p r e n a t a l m o r t a l i t y data f o r Dicrostonyx females, summers 1959-61, Main Study Area . . . . 72 Table 32. Minimum s u r v i v a l rate estimates f o r Lammus summer i960 75 Table 33. Minimum s u r v i v a l rate estimates f o r Lemmus summer 1961, Quadrat #1 . . . . 76 Table 3k* Minimum s u r v i v a l rate estimates f o r Mcrostonyx, summer i960, Quadrat # 2 . . . . 77 Table 35. Minimum s u r v i v a l rate estimates f o r Dicrostonyx, summer I96I, Quadrat #3 78 ix Page Table 36. Mnimum s u r v i v a l rates f o r lemmus converted to a 28 day base 79 Table 37. Minimum s u r v i v a l rates f o r Dicrostonyx converted to a 28 day base 80 Table 38. S u r v i v a l estimates f o r j u v e n i l e Lemmus on Quadrat # 1, summer I960 85 Table 39. S u r v i v a l estimates f o r juvenile Lemmus on Quadrat # 1, summer 1961 86 Table UO. S u r v i v a l estimates f o r juvenile Dicrostonyx on Quadrat # 2 , summer i960 87 Table I4.I. S u r v i v a l estimates f o r juvenile Dicrostonyx on Quadrat # 3 , summer 1961 88 Table Jj.2. Length of every movement recorded w i t h i n periods of l i v e trapping f o r Lemmus, summer i960, Quadrat # 2 . 91 Table U3. Length of every movement recorded w i t h i n periods of l i v e trapping f o r Lemmus, summer 1961, Quadrat # 1 . . 92 Table I4J4.. Length of every movement recorded within periods of l i v e trapping f o r Bicrostonyx, summer I960, Quadrats # 2 and # 3 93 Table U5» Length of every movement recorded w i t h i n periods of Live trapping f o r Dicrostonyx, summer 1961, Quadrat # 3 and v i c i n i t y 9k Table 1+6. Standing forage i n grams per 0.5 sq. meter dry weight at the end of summer 105 Table lj.7« Estimate of percentage forage u t i l i z a t i o n i n the spring of the de c l i n e , June 1961 108 Table 1+8. Body weight d i s t r i b u t i o n s f o r Lemmus males on the Main Study Area, 1959-61 117 Table 1+9 • Body weight d i s t r i b u t i o n s f o r Dicrostonyx males on the Main Study Area, 1959-61 118 Table 50. Body weight d i s t r i b u t i o n s f o r Lemmus males at Aberdeen Lake, 1960-61 121 Table 51. Body weight d i s t r i b u t i o n s f o r Dictostonyx males at Aberdeen Lake, 1960-61 122 Table 52. Body weight d i s t r i b u t i o n s f o r Lemmus males on the outlying areas, summer 1961 . . . . 123 Table 53. Mean body weights f o r Lemmus males of the winter and spring generations, summers 1959-61 . , 126 X Page Table 5U» Mean body weights f o r Dicrostonyx males of the winter and spring generationsj summers 1959-61 . . . . . 128 Table 55* Standardized mean organ weights (milligrams) and f a t index f o r Lemmus males, 1959-61 . 131 Table 56. Standardized mean organ weights (milligrams) and f a t index f o r Lammus females, 1959-61 . 132 Table 57. Standardized mean organ weights (milligrams) and f a t index f o r Dicrostonyx males, 1959-61 . . . . . . . . . 133 Table 58. Standardized mean organ weights (milligrams) and f a t index f o r Dicrostonyx females, 1959-61 13k Table 59• Amount of wounding shown on skins of Lemmus males from the Main Study Area, 1959-61 U 4 O Table 60. Amount of wounding shown on skins of Dicrostonyx males from the Main Study Area, 1959-61 Ihl x i AC MOWLEDGMENT S I wish to express my gratitude t o Sr . Dennis C h i t t y and Dr. Ian MoT. Cowan f o r t h e i r stimulation and help throughout t h i s study. I am also indebted to Mr. Andrew H. Macpherson and the Canadian W i l d l i f e Service f o r information and assistance i n the f i e l d work. Support f o r the f i e l d work was provided by the A r c t i c I n s t i t u t e of North America. I was supported during 1959-61 by the National Research Council of Canada, and during 1961-62 by the National Science Foundation of the U.S. I t i s a great presumption to put ray name to t h i s manuscript because many of the ideas were c o l l e c t e d from discussions with f r i e n d s and associates both at the Bureau of Animal Population, Oxford, and at the Department of Zoology, U.B.C. Thanks are also due to my wife who served as a f i e l d a s s i s t a n t f o r two summers. F i n a l l y , I want to express my thanks to the people of Baker Lake whose cooperation has made t h i s study most pleasant. INTRODUCTION A l l animal populations f l u c t u a t e i n numbers. In some these f l u c t u a t i o n s are small, i n others l a r g e . In some again these f l u c t u a t i o n s are i r r e g u l a r , i n others they tend to be regular. Some small mammals i n p a r t i c u l a r show f l u c t u a t i o n s which are large i n magnitude and r e l a t i v e l y regular i n occurrence, and these f l u c t u a t i o n s are r e f e r r e d to as "cycles"• Vie are concerned here with a well-known example of these f l u c t u a t i o n s — the lemming cy c l e of the tundra. Two species of lemmings i n h a b i t the c e n t r a l Canadian a r c t i c , the brown lemming (Lemmus trimucronatus) and the varying lemming (Dicrostonyx groenlandicus). Both are small f u r r y rodents with very short t a i l s and ears, and u s u a l l y weight 60-100 grams when f u l l y grown. The brown lemming remains brown a l l year round, but the varying lemming i s white i n winter and grey i n summer. Both species are a c t i v e throughout the year, burrowing under the snow i n the winter and occupying burrows dug i n the ground during the summer. Their food consists of green plants i n summer and dormant buds and roots i n winter. In summer the varying lemming tends to occupy the d r i e r h a b i t a t s and the brown lemming the wetter s i t e s . There i s an annual overturn of population, no i n d i v i d u a l s l i v i n g more than one year. Breeding may occur at any time of the year and young animals may mature at 3-i| weeks of age i n both species. The gestation period i s 19-21 days and the l i t t e r size v a r i e s seasonally between three and nine. Thus these species have a tremendous p o t e n t i a l rate of increase. The objective of t h i s research program was to study the population dynamics of the brown lemming and the varying lemming over a f u l l cycle i n numbers at Baker Lake, Keewatin, N.W.T. i n the Canadian 2 Barren Grounds. The f i r s t purpose of t h i s study was to describe the events of the lemming cycle of the Barren Grounds. The second purpose was to explain these events i n a comprehensive theory. The basic question d i r e c t i n g my approach i s t h i s : what are the necessary and s u f f i c i e n t conditions f o r the occurrence of a lemming c y c l e . The f i r s t purpose (description) has now been r e a l i z e d f o r one c y c l e . The second purpose (explanation) i s not yet r e a l i z e d , but the r e s u l t s suggest which of the current explanations are inadequate. A hypothesis i s considered which i s not inconsistent with the observed events and the information needed from future work i s noted. The plan of t h i s paper i s as f o l l o w s . A f t e r a b r i e f d e s c r i p t i o n of the study area, the habitats occupied by the two species of lemmings w i l l be described and then population d e n s i t y changes w i l l be considered. Then reproduction, mortality, and movements w i l l be assessed i n r e l a t i o n t o the density changes. Changes i n e x t r i n s i c and i n t r i n s i c f a c t o r s which a f f e c t the population w i l l then be considered. F i n a l l y , a d i s c u s s i o n of the e n t i r e work w i l l be given with an attempt to integrate these r e s u l t s with contemporary ideas. DESCRIPTION OF AREA AND CLIMATE The Baker Lake settlement i s on the northwest corner of Baker Lake near the mouth of the Thelon River i n the east c e n t r a l Barren Grounds. The whole area l i e s i n the Canadian S h i e l d . The t e r r a i n i n general i s f l a t to gently r o l l i n g , mostly covered with g l a c i a l d r i f t with few outcrops of bedrock showing. Lakes dot the landscape, occupying perhaps 30$ of the t o t a l area. Drainage i s poor and even l i g h t summer r a i n s can cause considerable l o c a l f l o o d i n g . This study was c a r r i e d out i n the area shown i n Figure 1. The Main Study Area occupies about 3 sq. miles j u s t north of the Baker Lake settlement. A l l intensive work was c a r r i e d out on t h i s area. Other areas marked on the map are ou t l y i n g areas sampled once or twice during each summer. In addi t i o n some sampling was c a r r i e d out at the Canadian W i l d l i f e Service camp on Aberdeen Lake (6U° 3 7 ' N, 99° Uk% W), about 113) miles west of Baker Lake. The weather f o r Baker Lake i s summarized i n Table 1 f o r 1959-61, and the mean values f o r ten years' records are given f o r comparison. Summer weather v a r i e d considerably between the d i f f e r e n t years. The summer of 195° was c o l d and wet, while the summer of I960 was warm and rather dry. The summer of 1961 was intermediate, coo l to warm and again dry. The spring phenologies of the three years were quite d i f f e r e n t . Spring I960 was the t a r l i e s t . The spring events of 1959 began about 8-20 days behind those of spring I960, and the spring events of 1961 were 2-6 days behind those of I960. This i s r e f l e c t e d i n the dates at which Baker Lake was e n t i r e l y f ree of ices 31 J u l y 1959, 12 J u l y I960, and 17 J u l y 1961. There were corresponding d i f f e r e n c e s i n the time of onset of summer breeding i n lemmings. 4 FIGURE 1. Map showing the l o c a t i o n of Baker Lake, the Main Study Area, and the p e r i p h e r a l sampling areas. TABLE 1. Temperature and p r e c i p i t a t i o n data during 1959-61 and the mean values f o r 1950-60 as recorded by the Baker Lake Meteorological Station. JAN. FEB. MAR. APR. MAY JUNE JULY AUG. SEPT. OCT. NOV. DEC. "WHOLE YR Mean Monthly Temperature (°F.) 1959 -22 -28 -18 -2 15 35 50 46 38 14 -4 -6 +9.8 I960 -26 -25 -22 +1 28 46 5u 52 38 19 -7 4.14 +12.0 1961 -31 -21 -25 +2 16 Ul 53 46 Mean 1950-60 -28 -27 -15 +2 22 39 52 50 38 18 -4 -19 +10.7 T o t a l R a i n f a l l (in.) 1959 T r * 0.89 1.78 2.71 1.07 Tr T r 6.45 I960 - - - - Tr 0.I4 1.28 1.11 1.76 1.60 T r 5.89 1961 - - - - Tr 0.37 0.2U 1.98 Mean 1950-60 - - - T r 0.22 0.73 1.55 1.72 1.16 0,20 T r - 5.58 T o t a l Snowfall (in.) 1959 3.2 0.8 l .U 1.5 6.1 Tr 1.0 8.7 2.1 7.6 I960 1.6 1.0 2.3 4.9 0.8 _ 0.8 7.3 1.5 1.7 30.0 59-1961 0.6 4.4 3.1 4.2 1.0 1.0 - Tr 25.6 ;5o-Mean 1950-60 1.8 1.8 2.3 3.6 1.8 0.6 - Tr 0.9 3.8 3.6 2.9 23.1 # Tr » trace 6 Events over the autumn freeze-up are also of considerable importance f o r lemmings. The 1959 freeze-up was characterized by a lack of f r e e z i n g r a i n and a quick buildup of a p r o t e c t i v e snow cover, thus nnnimizing ground i c i n g and burrow f l o o d i n g . By contrast the I960 freeze-up was accompanied by very wet conditions, o s c i l l a t i n g f r e e z i n g -thawing, and a lack of a good snow cover u n t i l mid-December. This contrast between 1959 and i960 i s brought out i n Table 2. To sum up, the summer of 1959 was cold and wet but was followed by a quick, dry freeze-up and an i d e a l winter. The summer of i960 was warm and dry but was followed by a slow, wet freeze-up and a bad e a r l y winter. The summer of 1961 was warm and dry a l s o . 7 TABLE 2 . Depth of snow on ground during e a r l y winter. Depth of Snow (inches) Oct. 1 Oct. 1 5 Nov. 1 Nov. 1 5 Dec. 1 1 9 5 9 - 6 0 0 2 9 9 1 1 1 9 6 0 - 61 Tr# 2 * 2 : 3» Mean 1 9 5 0 - 6 0 0 1 3 5 7 * Trace TEGETATION AND HABITATS Not much i s known about p l a n t associations i n the Canadian A r c t i c . Nevertheless, some scheme of habitat c l a s s i f i c a t i o n was needed i n t h i s study. What follows i s an attempt to d i v i d e the vegetation of the Baker Lake area i n t o habitats which i n p r i n c i p l e might be applied t o the e n t i r e Barren Grounds. One of the most s t r i k i n g c h a r a c t e r i s t i c s of a r c t i c vegetation i s i t s extreme v a r i a b i l i t y from one small area to the next. This produces a correspondingly great i n t e r s p e r s i o n of habitats and gives the impression of one great continuum of vegetation rather than d i s t i n c t h a b i t a t s . Nevertheless, there are d i s t i n c t habitats which can be recognized even though t r a n s i t i o n s are very common. Three c r i t e r i a were used t o d i s t i n g u i s h h a b i t a t s . As a f i r s t approximation, the dominating influence seems to be water, and thus we can recognize a vegetation continuum from dry to wet. From t h i s perspective the l i c h e n s occur i n dry areas, the heath (Ericaceae) plants i n moderately dry areas, and the sedges and mosses i n wet areas. I have introduced a f u r t h e r f a c t o r i n t o the habitat c l a s s i f i c a t i o n , that of microtopographical r e l i e f . This involves hummocks (low rounded mounds,' 10-18" t a l l , 1-2' i n diameter) and tussocks (thick clumps of Eriophorum, u s u a l l y about 10" t a l l and l e s s than one foot i n diameter). Both these structures occur i n wetter areas. T h i s microtopographical r e l i e f i s important f o r lemmings. The type of habitat at each trapping s t a t i o n was recorded. A l l habitat c l a s s i f i c a t i o n i n t h i s study was done s u b j e c t i v e l y by looking at the vegetation, drainage, and microtopography. A subjective a p p r a i s a l of the two or three dominant species w i t h i n a f i v e f o o t radius of the trapping s t a t i o n was also made f o r most s t a t i o n s . There was not enough time to do 9 anything more o b j e c t i v e . The p r i n c i p a l h a b i t a t s are l i s t e d i n Table 3 which also gives the area covered by each on the Main Study Area. These f i g u r e s were obtained from a combination a e r i a l photo and f o o t survey of the area. The habitats found on the Main Study Area are s i m i l a r to those at New Lake, Prince River, and Aberdeen Lake. The dominant plant species found i n these h a b i t a t s are s i m i l a r on the f i r s t three of these four areas, and Table k gives these data. The Aberdeen Lake area shows somewhat d i f f e r e n t species dominating the same habitat s , and these data are given i n Table 5. The c h i e f difference i s that there are no Dryas i n t e g r i f o l i a and Cassiope tetragona at Aberdeen Lake, but Ledum groenlandicum and Eriophorum are more abundant there. A s l i g h t l y d i f f e r e n t s e r i e s of habitats occurs on the i s l a n d s at the west end of Baker Lake and on the sandplain along the south bank of the Thelon River mouth. These habitats are characterized by a dominance of mosses, as shown i n Table 6. In summary, because there i s no s a t i s f a c t o r y b o t a n i c a l c l a s s i f i c a t i o n of tundra plant associations 11 habitat types have been established to describe the vegetation of the Baker Lake area. The c r i t e r i a used to d i s t i n g u i s h h a b i t a t s were drainage, p l a n t s and microtopographical r e l i e f . The p l a n t species found i n these habitats vary s l i g h t l y from area to area. 10 TABLE 3 . Area covered by the p r i n c i p a l habitat types on the Main Study Area* HABITAT TYPE NUMBER OF ACRES % OF TOTAL LAND SURFACE ROCK and ROCK-LICHEN LICHEN LICHEN HEATH HEATH and HEATH HUMMOCK HEATH SEDGE and HEATH SEDGE HUMMOCK % SEDGE TUSSOCK SEDGE HUMMOCK SEDGE MARSH 29k 53 7U2 5 506 3 290 67 15.0 2.7 38.0 0.2 25.9 0.1 H i . ? 3 . 3 T o t a l area cover typed (exclusive of lakes) 1958 (3.05 sq. miles) 100.1 * Heath-sedge hummock contributes about 30% of these f i g u r e s , heath sedge the remaining %10m 11 TABLE I*. Dominant p l a n t species i n the habi t a t s of the Main Study Area, New Lake, and Prince R i v e r . Figures i n the table r e f e r to the frequency with which the plant species was recorded as dominant at trapping s t a t i o n s i n the given h a b i t a t . SPECIES LICHEN HEATH HEATH HEATH HUM. HEATH SEDGE HEATH SEDGE RTJMM. SEDGE SEDGE TUSSOCK HUM. SEDGE MARSH Lichens 91.5% * - 18.2$ 31.5$ 2iul# 7.9$ 7.0$ -Dryas i n t e g r i f o l i a 28 .8 - 9.1 143.1 32.5 < 2 . 6 20.U mm Betula glandulosa 28 .8 66.7 81.8 26.2 U8.0 kl.k 2U.6 1.0 Vaccinium uliginosum 26.3 72.2 100.0 8.5 2 6 . 8 7 .9 6.3 -Cassiope tetragone 23 .8 27 .8 U5.5 3.8 - - 0 .7 -Ledum groenlandicum 21.3 5.6 9.1 0 .8 1 . 6 2 . 6 2.1 mm Empetrum nigrum 25.0 38 .9 27.3 2.3 1 . 6 mm 1.U -Arctostaphylos rubra 18 .8 5.6 - - - 2 . 6 - -Rhododendron lapponicum 2.5 mm mm 0 .8 - 2 . 6 - -Rubus chamaemorus - - - - - 2 . 6 - -S a l i x spp. 2.5 5.6 - 8.5 U.9 5.3 U.2 1 .9 Eriophorum spp. - - - U.6 18.7 100.0 U3.0 61.0 Carex spp. - mm - 8 6 . 9 77.2 26.3 2M 9U.3 Juncus spp. - - - 3U.6 15. h - 28.2 50.5 Mosses 23.8 61.1 9.1 U7.7 U8.8 U7.U 38.7 30.5 T o t a l Number of Stations 80 18 11 330 123 38 H i 2 105 # The two most common dominants f o r each habitat are underlined. TABIE 5 * Dominant p l a n t species i n the habitats of the Aberdeen Lake area. Figures i n the table r e f e r to the frequency with which the plant species was recorded as dominant at trapping s t a t i o n s i n the given h a b i t a t . SPECIES LICHEN HE. HEATH H. Lichens 9 6 . 9 % * Ledum groenlandicum 1 0 0 . 0 Betula glandolusa Vaccinium v i t i s - i d e a I | 3 . 8 Empetrum nigrum 3 . 1 Rubus charaaemorus 6 . 3 Eriophorum spp. 3 . 1 Carex spp. Mosses 6 . 3 Grasses 1 2 . 5 T o t a l Number of Stations 3 2 TH-SEDGE and SEDGE SEDGE SEDGE S. HUMMOCK TUSSOCK HUMMOCK MARSH 2$.7% 11.855 9.1$ -91.it 58.8 7 2 . 7 mm 5.7 - U.6 mn 28.6 - - -mm 1U.3 1 7 . 6 9.1 -U5.7 9U.1 68.2 75 .0 3 1 . U Ul.2 59.1 95.8 UO.O U7.1 63.6 7 0 . 8 l l .U mm 9.1 12.5 35 17 22 2k * The two most common dominants f o r each habitat type are underlined. 15 TABLE 6. Bominant p l a n t species i n the moss habi t a t s on the i s l a n d s of Baker Lake and the south bank of the Thelon mouth. Figures i n the table r e f e r to the frequency with which the plant species was recorded as dominant at trapping s t a t i o n s i n the given h a b i t a t . SPECIES LICHEN HEATH MOSS HEATH MOSS MOSS SEDGE Lichens 100.0% # - -Dryas i n t e g r i f o l i a 9-5 58.8 27.3 -Vaccinium uliginosum 9.5 5.9 U.5 2.0 Vaccinium v i t i s - i d e a 52.U - U.5 -Betula glandulosa 23.8 11.8 27.3 11.8 Empetrum nigrum . 19.0 26.5 U.5 -Ledum groenlandicum U2.9 5.9 - 2.0 Salxx spp. - Hi.7 59.1 6U.7 Car ex spp. - 2.9 31.8 82 .U Mosses Hi.3 22i2 100.0 96.1 Grasses 9.5 8.8 13.6 29.U Other species - 2.9 U.5 2.0 T o t a l Number of Stations 21 3h 22 51 # The two most common dominants f o r each habitat type are underlined. POPULATION DENSITY The f i r s t requirement of a l l population work must be moderately accurate information about changes i n density. To determine trends i n population density I have used one census method, l i v e trapping, and four other methods, snap trapping, v i s u a l estimates, trace i n d i c e s f o r feces, and dropping boards. Of these f i v e methods only l i v e trapping provides a d i r e c t census of numbers. The other four merely give an index more or l e s s p r o p o r t i o n a l to a c t u a l density. METHODS Live Trapping Live trapping i s the best technique f o r estimating density because i t gives a d i r e c t count. A considerable amount of e f f o r t was expended i n a l i v e trapping program i n t h i s study, but various d i f f i c u l t i e s plagued the r e s u l t s . The most serious problem was trap-deathsj t h i s was not solved u n t i l l ? 6 l and even then not e n t i r e l y . The basic technique was not f i x e d u n t i l 1961 and consequently the data f o r 1959 and I960 are l e s s complete. Three quadrats f o r l i v e trapping were established during the course of t h i s study (Figure 2). Their s p e c i f i c a t i o n s are as follows: Area Length of one Trap spacing No.of traps (acres) side (feet) (feet) Quadrat #1 1.9 288 18 289 Quadrat # 2 31.3 700 50 225 Quadrat # 3 3.5 350 $0 6k Quadrat # 1 was set up i n 1959. Quadrat # 2 was established i n I960 when i t became apparent that movements and home ranges were f a r l a r g e r than could be measured by Quadrat # 1. Quadrat # 3 was also set up i n i960 15 FIGURE 2 . Location of the l i v e trapping quadrats of the Main Study Area. Abbreviations r e f e r to habitat types ( r » rock, lb. » l i c h e n heath, hh =» heath hummock, hs =» heath sedge and heath sedge hummock, sh «* sedge hummock and sm = sedge marsh). 17 weight. The cohort of adult animals present at the beginning of the summer breeding are r e f e r r e d to as the winter generation. Since summer breeding tends to occur synchronously throughout the population, summer young appear i n waves during the summer. These summer l i t t e r s are r e f e r r e d to as follows: Y^ summer young = f i r s t summer L i t t e r j Y^1 summer young = second summer l i t t e r j and Y-j* summer young = t h i r d summer l i t t e r . These groups of young are r e a d i l y separated by body weights u n t i l l a t e • summer when growth rates d e c l i n e . A f u l l d i scussion of aging problems i s given i n a l a t e r section (see Body Weights). Snap Trapping Snap trapping was done both systematically and non-systematically. The systematic l i n e s were set up as Type B l i n e s of the North American Census of Small Mammals (Calhoun, 19U8), i . e . 20 stati o n s spaced at 50* i n a s t r a i g h t l i n e with three traps per s t a t i o n w i t h i n a f i v e f o o t radius of the stake. With a few exceptions these l i n e s were set up i n p a i r s spaced 100' apart (Calhoun, 19U8, recommends ij.001) and p a r a l l e l . Eight p a i r s of l i n e s were set out on the Main Study Area i n 1959* No new l i n e s were added there i n I960, but four a d d i t i o n a l p a i r s were set out i n 1961. On the ou t l y i n g areas two p a i r s of l i n e s were set out i n 1959, two a d d i t i o n a l p a i r s i n I960, and f i v e and one-half a d d i t i o n a l p a i r s i n 1961. In general on the Main Study Area each l i n e was trapped twice during the summer (June and August). These l i n e s are re f e r r e d to as standard l i n e s because they are retrapped each year. Each single standard l i n e run f o r one p e r i o d (three days) represents 180 standard trap nights and a l l snap trapping i n d i c e s i n t h i s paper are expressed i n numbers of lemmings per 100 standard trap nights. Non-systematic snap trapping was done p r i m a r i l y to obtain specimens f o r autopsy. Stations were placed at i r r e g u l a r i n t e r v a l s wherever 18 there seemed to be any chance of catching a lemming. Two traps were placed around each s t a t i o n . The traps were removed a f t e r three days and the same place was never trapped twice. These l i n e s caught two to three times as many lemmings per trap night as the-standard l i n e s . Museum S p e c i a l traps were used throughout t h i s study and r a i s i n s were used as b a i t . Although i t was probably unnecessary, b a i t was used on a l l snap traps throughout t h i s study u n t i l the matter could be i n v e s t i g a t e d . Some differ e n c e between the two species f o r standard snap trap sampling should be pointed out. Lemmus i s a creature of the wet habitats and uses w e l l defined runways i n moving about. I t s occupied holes are d i f f i c u l t to f i n d because there are so many possible s i t e s . Dicrostonyx, on the other hand, i s more often a creature of the d r i e r habitats and does not move along w e l l defined runways. I t s occupied holes are often easy to f i n d and are marked by a mound of f r e s h l y dug sand or peat. Thus runway trapping i s most e f f e c t i v e f o r catching Lemmus, and burrow or den trapping i s most e f f e c t i v e f o r catching Dicrostonyx. The d i f f e r e n t habitats of the two species introduce a f u r t h e r complication. In the Baker Lake area the wet (Lemmus) habitats are reasonably extensive and the d i s t r i b u t i o n of Lemmus i s f a i r l y uniform over these. But, although the Dicrostonyx ha b i t a t s are equa l l y or even more extensive, areas suitable f o r digging burrows are r e s t r i c t e d and t h i s r e s u l t s i n a contagious type of d i s t r i b u t i o n f o r Dicrostonyx. T h i s complicates sampling considerably, because standard trap l i n e s may completely miss these "colonies" and thus give a biased idea of the a c t u a l density. The r e s u l t i s that Lemmus d e n s i t i e s are estimated be t t e r by standard snap trapping l i n e s than are Dicrostonyx d e n s i t i e s i n the Baker Lake area. Two questions about the snap trapping technique should be considered 19 at t h i s p o i n t . F i r s t , does snap trapping provide a good index of population density? The number of animals caught i n a trap l i n e depends not only on population density but also on the weather, habitat, amount of a c t i v i t y , home range s i z e , and proportion of young animals. For t h i s reason numerical i n d i c e s from snap trapping must be t r e a t e d with caution. Trends i n density are shown quite w e l l , but the a c t u a l numerical value of these trends must remain vague. In p a r t i c u l a r , s t a t i s t i c a l confidence l i m i t s f o r these i n d i c e s are meaningless unless the e f f e c t s of a l l the secondary v a r i a b l e s mentioned above can be neglected. In t h i s study there are independent sources of density estimates, such as l i v e trapping, and these can be compared t o the snap trapping i n d i c e s to see how w e l l these d i f f e r e n t estimates agree. The second question i s whether repeated trapping of the same l i n e s both within and between years has i n t e r f e r e d with the c y c l i c changes we are attempting to observe. This question can be answered i n d i r e c t l y , since new standard l i n e s were set out every year. We can enquire whether any catch d i f f e r e n c e s occurred between new l i n e s i n year x and o l d l i n e s i n the same year, taking i n t o account habitat d i f f e r e n c e s between l i n e s . Comparing the numbers caught i n new and o l d l i n e s , we found no di f f e r e n c e s i n catch e i t h e r i n I960 or i n 196l, and so I conclude that snap trapping the same l i n e s does not a f f e c t the c y c l i c changes we are attempting t o observe. Other i n d i r e c t evidence supports t h i s . The number of traps set o f f without a catch i s almost always equal to or greater than the number of lemmings caught, which suggests that a f a i r number of lemmings escape being trapped even i n the immediate v i c i n i t y of the trap l i n e . In a d d i t i o n , i f we consider the s i z e of the area over which the trap l i n e s are spread and the m o b i l i t y of the lemmings, i t i s c l e a r that only a minute f r a c t i o n of the population i s being removed by trapping. Other Census Methods 20 V i s u a l estimates of density changes were obtained by counting the number of lemmings seen per.hour of walking on the tundra. This i s obviously a crude index of density but i t does provide valuable supplementary information f o r areas where no l i v e trapping was done. Trace i n d i c e s of f r e s h feces were made i n 1°5° and 1°60 by doing l i n e transects through h a b i t a t types, dropping a 3' by l 1 rectangle every ten f e e t , and recording presence or absence of f r e s h green droppings. Again t h i s i s a crude index but i t has the advantage of being very q u i c k l y done. F i n a l l y , dropping boards were used as suggested by Emlen et a l . (1957). This technique was used i n 1959 and I960 but discontinued i n 1961 because i t involved a considerable amount of work and merely duplicated other census information. RESULTS Live Trapping Table 7 gives the numbers of Lemmus on Quadrat # 1 (1.9 acres) i n 1959-61, and Table 8 gives the numbers of Dicrostonyx on Quadrat # 3 (3.5 acres) i n 1959-61. While there were few or no Dicrostonyx on Quadrat # 1 i n any of the years, Quadrat # 3 had a lemmus population of 20 animals i n August I960, but none i n e i t h e r 1959 or 1961. The f i r s t p o i n t that emerges from these t a b l e s i s the great increase i n numbers from 1959 to i960 and subsequent decline i n 1961. We can estimate these changes q u a n t i t a t i v e l y . In Lemmus the increase from August 1959 to June i960 i s 28-fold, and i f we accept the argument from snap trapping given below, that the population before the I960 melt-off was approximately equal to the August i960 population, we have an estimated 58-fold increase over the winter of 1959-60. T h i s must be considered as only a crude estimate of the a c t u a l increase. There was probably a minimal 21 TABLE 7. Numbers of Lemmus on Quadrat # 1 during 1959-61." DATE OF SAMPLING 1959 August 5-10 August 11-23 I960 June 18-20 J u l y 6-8 J u l y 28-30 August 25-27 1961 June 12-18 WINTER GENERATION SUMMER GENERATION* TOTAL ANIMALS 28' 22' 12* 8 2 3 5 1 3 1 2 2 2 8 251 16 V 20 J Y '» 1 1UJ June 19-25 June 26-July 2 J u l y 3-9 J u l y 10-16 J u l y 17-23 J u l y 2U-30 J u l y 31-Aug. 6 - -August 7-13 - 1 August ll|-20 - -August 21-27 - - -August 28-Sept. 1 - 1 Superscripts i n the ta b l e give trap m o r t a l i t i e s . 2 Y-j_ = f i r s t summer l i t t e r , Y-^ ' m second summer l i t t e r , Y^ B = t h i r d summer l i t t e r . 0 1 26 30 ia 58 2 3 5 3 h 2 2 0 1 0 0 1 22 TABLE 8, Numbers of Dicrostonyx on Quadrat # 3 i n 1960-61. .2 DATE OF SAMPLING 1959 J u l y 2U-Aug. 1 August 6-10 I960 August 25-27 1961 June 5-11 June 12-18 June 19-25 June 26-July 2 J u l y 3-9 J u l y 10-16 J u l y 17-23 J u l y 2U-30 J u l y 31-Aug. 6 August 7-13 August lU-20 August 21-27 August 28-Sept. 1 WINTER GENERATION (3) (1) 1 0 1 1 9 $ 1 8 k 1 2 SUMMER GENERATION (3) (2) 13 V 11 h 1 y n 1 TOTAL ANIMALS (6) (3) 38 1 9 11 * 1 8 5 5 3 1 2 0 Superscripts i n the ta b l e give trap m o r t a l i t i e s . ^Y^ = f i r s t summer l i t t e r , Y^ 1 = second summer l i t t e r , Y]_" = t h i r d summer l i t t e r . The 1959 f i g u r e s r e f e r to l i v e trapping done along a l i n e i n the v i c i n i t y of where Quadrat # 3 was set out i n i960. They are thus not s t r i c t l y comparably to the 1960-61 f i g u r e s . 23 25-fold increase i n Lemmus over the winter of 1959-60 and t h i s increase may have been as much as 50-fold. The 1959-60 winter increase of Dicrostonyx cannot be estimated i n the same way, since Quadrat # 3 was not set up u n t i l I960. I f we assume that the May I960 population equalled that i n August I960, and that there were about 3-6 animals on the area i n August 1959 (see Table 8), the estimated increase i s 5-10 f o l d over the winter 1959-60. These crude estimates suggest that Dicrostonyx probably d i d not increase as much as Lemmus over the winter of 1959-60 on the Main Study Area. The Lemmus population at l e a s t doubled i t s numbers i n the summer of I960. From the amount of trap m o r t a l i t y involved i n t h i s estimate and a d d i t i o n a l data from Quadrat # 2, a reasonable estimate of t h i s summer increase i s 2-3' f o l d between 15 June and 31 August i960 i n Lemmus. No estimate can be made f o r Dicrostonyx from l i v e trapping data. F i n a l l y , we can estimate the decline over the winter of 1960-61. The Lemmus population on Quadrat # 1 declined from 58 to 5 between August I960 t o June 1961, a 90-95% decrease. The Dicrostonyx population on Quadrat # 3 declined from 38 to 11 over the same period, a 70-80% decrease. These crude f i g u r e s allow us to conclude that Dicrostonyx probably d i d not decrease as much as Lemmus over the winter of 1960-61 on the Main Study Area. The decline continued i n both species through the summer of 1961 on the l i v e trapping area. There were very few summer young i n the 1961 samples, an important point to which we s h a l l r e turn l a t e r , and there was no recovery of numbers. Although i t i s possible to estimate the number of lemmings per acre, I have not done so because the data are too imprecise. A c t u a l d e n s i t i e s are very much a l o c a l phenomenon and do not help us to understand the c y c l e . 2k Snap Trapping Tables 9 and 10 give;.-the?snap trapping i n d i c e s f o r Lemmus and Dicrostonyx and show i n a general way the great changes from s c a r c i t y i n 1959 to abundance i n I960 and the subsequent decline i n 1961. These changes i n abundance occurred i n both species and on a l l the areas trapped. These data suggest that the Lemmus cycle was more pronounced than the Dicrostonyx c y c l e . The i n d i c e s f o r the Main Study Area change p r o p o r t i o n a l l y as f o l l o w s : 1959 I960 1961 Lemmus 1 50 10 Dicrostonyx 5 hO 15 However, we must beware of comparing Lemmus indices with Dicrostonyx indices because the d i f f e r e n c e s i n biology between these species must a f f e c t the absolute value of these i n d i c e s . So a l l that we can say i s that the data suggest that Lemmus fl u c t u a t e d more strongly than Dicrostonyx. The snap trapping i n d i c e s i n the summer of 1961 are p a r t i c u l a r l y v a r i a b l e (compare, f o r example, Lemmus on Nine Mile Island at 12.22 with Lemmus on the Main Study Area i n August at 0.62). This v a r i a b i l i t y i s due p a r t l y to the f a c t that two d i f f e r e n t types of declines occurred i n 1961. On some areas there was moderate abundance i n spring and a steady decrease through the summer with no recovery (Main Study Area, Prince River, Thelon R i v e r ) . On other areas there was moderate abundance i n spring with some recovery of numbers through the summer (Aberdeen Lake, New Lake, Ten Mile Island, Nine Mile Island, Long Island, Second I s l a n d ) . The s i g n i f i c a n c e of these d i f f e r e n t types of declines ( r e s p e c t i v e l y types G and H according to C h i t t y , 1955) and t h e i r associated c h a r a c t e r i s t i c s w i l l be discussed l a t e r . One f u r t h e r d e t a i l of density changes during the cycle was shown by snap trapping r e s u l t s . There was a sharp drop i n density i n the spring 25 TABLE 9. Lemmus snap trapping i n d i c e s , 1959-61. LOCATION AND TIME PERIOD Main Study Area 1959 June J u l y August September 1-10 I960 June . "^July August 1961 June J u l y August Other Areas 1959 August 12-17 Prince River Ten Mile I s . TheIon River August 26-Sept. 5 New Lake I960 J u l y 13-18 Aberdeen Lake J u l y 20-23 New Lake August 15-18 Prince River Ten Mile I s . TheIon River 1961 J u l y 1-13 New Lake J u l y 26-29 Aberdeen Lake J u l y 17-20 Long Island DRY HABITATS MEDIUM HABITATS WET HABITATS N 2 • LEMMUS3 N LEMMUS N LEMMUS 711 0.0 50U 0.0 50U 0.0 153 0.0 504 0.79 378 0.00 68U 5.56 576 0.0 1017 0.0 1260 0.0 1332 0.15 1377 0.0 1377 0.0 810 0.0 3377 4.87 180 6.11 1377 15.5U 16U7 0.2U 1017 0.30 1773 0.11 2268 0.09 990 0.10 990 0.20 477 0.21 999 6.91 198 6.57 999 21.42 1449 1.10 1764 1.42 1935 0.62 18 0.0 126 0.79 576 2.1)3 - - 2k 0.0 216 1.85 213 0.0 - - 387 0.78 - - - - 1620 1.11 201 0.0 226 3.54 218 18.35 - - 54 33.30 306 27.12 15 6.67 96 23.95 391 17.65 - - 2U 12.50 216 18.06 213 17.37 mm 387 27.13 - - 171 0.0 540 I.48 306 0.0 279 2.15 486 8.44 126 0.0 207 2.90 207 7.73 (continued) TABLE 9 (continued) 26 LOCATION AND TIME PERIOD DRY HABITATS-1 N LEMMUi s 3 MEDIUM HABITJTS N LEMMUS WET HABITATS N LEMMUS 1 9 6 1 (Cont'd) J u l y 2l;-27 Second Island 3 6 9 2 . 9 8 August lU-19 Prince River 18 0 . 0 Ten Mile I s . 126 2 . 3 8 Nine Mile I s . 8 1 1 2 , 3 U Thelon River 1 5 3 0 . 0 September 1 - 1 0 New Lake 2 7 0.0 6 . 2 5 9 0 0 . 0 Ii32 3.01 3 . 7 0 180 7.78 - 90 12.22 2 7 3 . 7 0 360 3.33 2 9 7 0 . 6 7 6 9 6 3.02 Bry Habitats = l i c h e n heath, heath, heath hummock and moss heath. Medium Habitats « heath sedge and heath sedge hummocked. Wet Habitats = sedge hummock, sedge tussock, sedge marsh, moss, and moss sedge. 2 N = number of standard trap n i g h t s . ^LEMMUS - number of Lemmus caught per 1 0 0 standard trap n i g h t s . 27 TABLE 10. Dicrostonyx snap trapping i n d i c e s , 1959-61 LOCATION AND TIME DRY HABITATS 1 MEDIUM HABITATS MET HABITATS PERIOD o N 2 DICRO.J N DICRO. N DICRO. Main Study Area 1959 June 711 0.1+2 1332 0.68 2268 0.26 J u l y 501; 0.20 1377 0.51 990 0.20 August 5ol+ 0.20 1377 0.29 990 0.0 September 1-10 153 0.0 810 0.12 U77 0.63 I960 June 5ol+ 2.98 1377 1.31 999 0.30 J u l y 378 10.05 180 2.22 198 0.51 August 681+ 7.31 1377 1.09 999 o.l+o 1961 June 576 1.01+ 161+7 1.1+0 11+1+9 0.1+8 J u l y 1017 2.56 1017 0.88 1761+ 0.62 August 1260 0.79 1773 0.85 1935 0.56 Other Areas 1959 August 12-17 TheIon River 213 0.0 - - 387 0.52 I960 J u l y 13-18 Aberdeen Lake 201 1U.U3 226 6.61+ 218 2.29 August 15-18 TheIon River 213 U.69 - - 387 0.26 1961 J u l y 26-29 Aberdeen Lake 306 5.56 279 6.1+5 1+86 2.06 August 11+-19 Thelon River 153 0.0 27 0.0 360 0.0 ^ Dry Habitats = l i c h e n heath, heath, heath hummock, and moss heath. Medium Habitats •> heath sedge and heath sedge hummock. Wet Habitats • sedge hummock, sedge tussock, sedge marsh, moss and moss sedge. 2 N • number of standard trap nights. 3 -DICRO. = number of Dicrostonyx caught per 100 standard trap n i g h t s . 28 of I960 i n Lemmus j u s t as the snow was melting and summer breeding began. This drop was r e g i s t e r e d i n the standard trap l i n e estimates as follows: equal l i n e s E, F June 1-h ^August I-I4. (melt-off) 2/3 decrease equal • 1 r June U-Ijl 2 to 3 f o l d increase ^August k-l$ s i x other l i n e s ( a f t e r the meltoff) Given t h i s set of r e l a t i o n s h i p s , we estimate a 67% decline i n density of Lemmus over the melt-off, but t h i s i s probably an overestimate because of increased movements of animals during t h i s period (thus increasing trap l i n e catches). Perhaps a 30% m o r t a l i t y estimate i s c l o s e r to the t r u t h . This spring decline occurred i n spite of the absence of b i r d predators and only sparse populations of mammalian predators. Whether t h i s spring decline also occurred i n Dicrostonyx could not be determined. The spring trapping data r e f l e c t a change i n habitat d i s t r i b u t i o n between the two species over the c y c l e . I f we d i v i d e the habitats i n t o dry, medium, and wet (as i n Tables 9 and 10) and p l o t the percentage of the t o t a l numbers caught i n each type of habitat, we obtain the r e s u l t s shown i n Figure 3. There i s an inverse r e l a t i o n s h i p between Dicrostonyx and Lemmus such that the species which i s most abundant occupies the greatest range of h a b i t a t s . Thus Lemmus g r e a t l y expanded i t s habitat spectrum i n the peak summer of i960, while Dicrostonyx contracted i t s habitat spectrum although i t also increased considerably i n numbers. These changes complicate somewhat the i n t e r p r e t a t i o n of density changes observed i n a single habitat, because a given number of animals spread over many habitats w i l l obviously be l e s s dense than the same number i n one habitat only. The explanation of these changes i n habitat segregation probably l i e s i n some form of i n t e r s p e c i f i c 2 9 FIGURE 3» Summer habitat d i s t r i b u t i o n i n Dicrostonyx and lemmus on the Main Study Area. Ordinate i s the percentage of t o t a l numbers caught i n each type of h a b i t a t . L E M M U S 30 i n t e r f e r e n c e , but we have no d i r e c t evidence that t h i s i s the case. Other Census Methods V i s u a l estimates were obtained f o r Lemmus as follows: 1959 - 0.U3 Lemmus seen per 100 hours walking (based on 1;65 hours) 1960 - 85.0 11 « " 11 » » ( » 9 316 n ) 1961 - 0.51 n » " 11 0 » ( » •» 393 " ) These estimates apply only to the summer. During the spring melt-off and the f a l l freeze-up lemmings may become much more noticeable. The extent of the i960 c y c l i c high may be i n d i c a t e d from v i s u a l reports of lemming abundance as follows: May — C h e s t e r f i e l d I n l e t , Rankin I n l e t , C o r al Harbor, Eskimo Point; J u l y — G a r r y Lake, Beverly Lakej August — Chantry I n l e t j and September ~ Repulse Bay, Ferguson Lake. I t i s apparent from these reports that the i960 high occurred over at l e a s t an area 500 miles by 1+00 miles of the c e n t r a l a r c t i c , thus showing that the cycle at Baker Lake was not merely a l o c a l e f f e c t . Data obtained from trace i n d i c e s and dropping boards w i l l not be presented here because they add nothing new t o the observations above. SUMMARY AND CONCLUSIONS Figure k summarizes the density changes i n Lemmus and Dicrostonyx over 1959-61. 1959 Summer: This was a summer of very low numbers of both species, with Dicrostonyx somewhat more abundant than Lemmus. I t was evident by September that some increase had occurred but numbers were s t i l l very low. 1 1959-60 Winter: Tremendous population growth occurred over t h i s winter i n both species, the crude estimates of t h i s increase being 25-50 f o l d i n Lemmus and 5-10 f o l d i n Dicrostonyx from September 1959 to May I960. 51 FIGURE k» Generalized density changes, 1959-61. Numbers i n d i c a t e r e l a t i v e changes i n numbers f o r each species based on l i v e trapping. See t e x t f o r d e t a i l s . 1/5 cr UJ CD o _ J TYPE H TYPE G SUMMER ' WINTER ' SUMMER 1 WINTER ' SUMMER 1 1959 I960 1961 L E M M U S <l 1 28 58 5 1 DICROSTONYX 7 4? 20? 38 1 1 2 32 1960 Summer: The spring population of Lemmus declined considerably when the snow melted and summer breeding began. This m o r t a l i t y was probably between 67% and 30%. By August the Lemmus population had r i s e n 2-3 f o l d from i t s lowest point i n June and was then s l i g h t l y above the spring density. The Dicrostonyx population also increased during t h i s summer, but i t i s not known whether they showed the same drop i n numbers at the melt-off. Densities were highest i n t h i s cycle during August I960. 1960-61 Winter: A severe decrease i n population density occurred over t h i s winter, estimated at 90-95% i n Lemmus and 70-80% i n Dicrostonsoc from August i960 to June 196l. 1961 Summer: There were two patterns found i n t h i s summer of d e c l i n e . On the Main Study Area and two outly i n g areas the decline continued i n both species through the summer with no recovery (Type G d e c l i n e j C h i t t y , 1955)• On f i v e other outlying areas p a r t i a l recovery occurred through the summer (Type H d e c l i n e ) . By the end of t h i s summer on the Main Study Area d e n s i t i e s i n both species were about equal to those at the s t a r t of the study. Reports were received that lemming numbers were also high within a large area of the c e n t r a l a r c t i c i n i960. REPRODUCTION Population density changes because of reproduction, mortality, or migration. In t h i s section we s h a l l deal with the f i r s t of these primary population f a c t o r s . METHODS Reproductive data can only be obtained from dead animals, and since most of these were obtained by snap trapping we must hope that snap trapping samples the population randomly. The d i f f i c u l t i e s of t h i s assumption are p a r t l y avoided i n the ana l y s i s which follows by t r e a t i n g separately each generation, the d i f f e r e n t summer l i t t e r s , and the various time periods. For example, to lump o l d adult and summer young females together f o r an analysis would tax the assumption that t h i s group i s sampled randomly, whereas i f we t r e a t o l d and young females separately the assumption that sampling i s random within each group i s probably v a l i d . Complete autopsies were performed on almost a l l animals trapped^ skins and s k u l l s were saved and the following data were recorded: A l l specimens: date, species, sex, xraight, t o t a l length, hind foot length, f a t index, adrenal weight, spleen weight, lens weight, stomach weight, l o c a t i o n and habitat where caught. Males only: t e s t e s p o s i t i o n and weight, epididymis tubules v i s i b l e or not, size of seminal v e s i c l e s . Females only: Whether l a c t a t i n g or not, vagina perforate or not, s i z e of uterus, number of p l a c e n t a l scars, number, size and age of embryos, number of corpora l u t e a and corpora albicans i n each ovary, combined weight of uterus and embryos. Males were judged as fecund or non-fecund by whether or not the epididymis tubules were v i s i b l e t o the naked eye (Jameson, 1°5>0). There was almost no ambiguity i n determining t h i s , but i n the few doubtful cases accessory data on the size of the seminal v e s i c l e s and the weight and p o s i t i o n of the tes t e s were u t i l i z e d . 3U Females were classed as mature or immature by the presence or absence of corpora l u t e a i n the ovaries. This c r i t e r i o n i s more r e f i n e d than the c r i t e r i o n of perforate or non-perforate vagina ( L e s l i e , Venables, and Venables, 1952). Females were classed as pregnant i f the uterus showed macroscopically v i s i b l e swellings. The gestation period of Lemmus has been measured i n only a few cases. Thompson (1°55 a) gives 20 and 20| days f o r two i n d i v i d u a l s , and i n the present study two pregnancies were timed at 21 and 21^  days. Thus an approximate gestation of 21 days i s indica t e d f o r Lemmus. For Dicrostonyx Manning (195k) gives 19-21 days f o r two cases, and Quay and Quay (1956) give 21 days as a maximum f o r f i v e observations. Thus an approximate gestation of 19-21 days i s in d i c a t e d f o r Dicrostonyx. Assuming that both species of lemmings follow, i n general, the type of development shown by laboratory r a t s and mice, we may estimate that pregnancy becomes macroscopically v i s i b l e on the s i x t h day a f t e r impregnation. Embryos were aged i n the following way i n order to calculate back to the date of insemination. B i r t h weights were determined to average 3.3 grams i n Lemmus (Thompson, 1955 aj t h i s study) and about 3.0 grams i n Dicrostonyx (Quay and Quay, 1956). Laboratory mouse embryo growth curves f o r weight and crown-rump length (Enzmann, 1935) were converted to the gestation period and b i r t h weight of each species of lemming, and ta b l e s of expected weight and crown-rump length f o r each day of gestation were constructed. As a fu r t h e r check anatomical changes associated with development i n the r a t (Henneberg, 1937) were adapted i n the same way to the lemmings. The use to which these aging data are put i s such that accuracy only w i t h i n + 2 days i s necessary, and thus the assumptions made here are not r e a l l y c r i t i c a l f o r the r e s u l t s which f o l l o w . 35 P l a c e n t a l scars are formed at the implantation s i t e s of embryos and show up as areas of black pigmentation on the mesometrial side of the uterus (Conaway, 1955)* Although these scars were counted, the only use made of these data was i n the c l a s s i f i c a t i o n of females as n u l l i p a r o u s (no embryos or p l a c e n t a l s c a r s ) , Primiparous (embryos or one set of p l a c e n t a l scars present,) or multiparous (embryos and p l a c e n t a l scars present, or two or more sets of s c a r s ) . These scars tend to fade with age, but t h i s causes few problems i n animals of short l i f e span l i k e lemmings. Corpora a l b i c a n t i a (degenerate corpora lutea) were also counted; but, as with the p l a c e n t a l scars, the only use made of these' data was to c l a s s i f y females as n u l l i p a r o u s , primiparous, or multiparous. Corpora l u t e a were counted i n the ovaries of pregnant females with the a i d of a binocular d i s s e c t i n g microscope. These structures show up very c l e a r l y i n the small ovary of a lemming, p a r t i c u l a r l y i n animals f r e s h l y dead. Ovaries preserved i n formalin are much more d i f f i c u l t to count without d e t a i l e d h i s t o l o g i c a l work. In order to assess ovulation rate we must assume that each corpus luteum represents one ovulated egg and thus that there are no polyovular f o l l i c l e s or accessory corpora l u t e a formed. There i s almost no experimental or h i s t o l o g i c a l evidence on lemmings f o r these p o i n t s . Quay (I960) found very few (about 0,1%) binuclear and t r i n u c l e a r p r i m o r d i a l f o l l i c l e s i n Dicrostonyx, and t h i s suggests that polyovular f o l l i c l e s are not important i n t h i s species. In general corpora l u t e a counts agree with embryo counts f o r both speciesj only very r a r e l y are there fewer corpora l u t e a than embryos, and r a r e l y more than one to three more corpora l u t e a than embryos. U n t i l f u r t h e r studies are made, the a n a l y s i s which follows must r e s t on the unproven assumption that corpora l u t e a counts accurately and c o n s i s t e n t l y measure ovulation r a t e . There i s no reason yet to doubt t h i s assumption. 36 Resorbing embryos were recognized because they were smaller than normal embryos. Obviously these s i z e d i f f e r e n c e s are eas i e r to detect i n l a r g e r embryos l a t e i n pregnancy, and t h i s introduces some uncertainty i n assessing one aspect of prenatal m o r t a l i t y . In c a l c u l a t i n g l i t t e r size and embryo rates only l i v e embryos were counted. Prenatal m o r t a l i t y i s discussed i n the section on m o r t a l i t y . RESULTS Reproduction i s a complex v a r i a b l e which may be broken down i n t o several components. Figure 5 gives a schematic analysis of the components of reproduction i n polyestrous mammals, and i n the remainder of t h i s section we w i l l attempt t o assess some of these p a r t i c u l a r components. Length of Breeding Season Summer breeding i n lemmings begins when the snow melts i n spring and t h i s tends to synchronize breeding periods f o r the r e s t of the summer. Almost a l l mature females (the winter generation) are impregnated w i t h i n a 5-10 day period at the melt-offj 20-21 days l a t e r t h i s l i t t e r i s dropped (the Y-^  summer young). Post-partum breeding i s very common i n both species, and thus 3 weeks l a t e r a second l i t t e r i s dropped (the Y-^ ' summer young). A t h i r d l i t t e r (Y^** young) and a f o u r t h l i t t e r may be produced, but by l a t e summer the o r i g i n a l synchrony breaks down. This synchronous breeding tendency makes i t possible to t r e a t summer reproduction i n terms of b i o l o g i c a l periods rather than chronological ones. Tables 11 and 12 give the timing of summer breeding periods i n Lemmus and Dicrostonyx. The length of the summer breeding seasons of 1959-61 i n Lemmus and Dicrostonyx on the Main Study Area are given i n Table 13. The beginning of breeding i n every case coincides with the melting of the snow, and i t i s v a r i a t i o n s i n the end of the summer breeding season that must be accounted f o r here. In 1959 there was no evidence that breeding ceased i n the f a l l 57 FIGURE 5» Components of reproduction i n polyestrous mammals. TOTAL YEARLY REPRODUCTION MONTHLY EMBRYO RATES LENGTH OF BREEDING SEASON LITTER SIZE PREGNANCY RATE (during breeding season) 58 TABLE 11. Timing of summer breeding periods i n Lemmus females, 1959-61. Dates given are insemination dates; to obtain periods of b i r t h add 21 days. YEAR I Winter Generation 1 9 5 9 June 12-20 1960 May 29-June 10 1961 June 5- lU Y^ Summer Young 1 9 5 9 I960 1961 Y j • Summer Young 1 9 5 9 I960 1961 PERIOD i i n i i f J u l y 3-11 July 2U-Aug. 6 ? June 16-30 J u l y 8-19 NO BREEDING June 26-July 6 J u l y 18-28 August 6-? J u l y 2I+-Aug. 6 ? J u l y 8-19 NO BREEDING J u l y 18-28 August 7-? August 19-31 NO BREEDING August 8-? 59 TABLE 12. Timing of summer breeding periods i n Dicrostonyx females, 1959-61. Dates given are insemination dates; to obtain periods of b i r t h add 20 days. TEAR PERIOD I I I I I IV WINTER GENERA! ION 1959 June 18-28 1960 May 31-June 8 1961 June 2-LU J u l y 10-18 June 22-28 June 28-July 10 August 1-10 ? J u l y 16-2U J u l y 21- ? NO BREEDING NO BREEDING Y x SUMMER YOUNG 1959 I960 1961 NO BREEDING J u l y 17-21 NO BREEDING ? NO BREEDING NO BREEDING Y x « SUMMER YOUNG 1959 - - NO BREEDING ? 1960 - - - NO BREEDING 1961 . - NO BREEDING 4o TABLE 13. length of the summer breeding seasons of Lemmus and Dicrostonyx, Main Study Area, 1959-61. TEAR LENGTH IN DAYS 1 TIME PERIOD2 IEMMITS 1959 80 1960 70 1961 8U June 12. - September 15 + May 29 - August 9 June 5 - August 28 DICROSTONYX 1959 73 1960 7k 1961 69 June 18 - August 30 + ? May 31 - August 13 June 2 - August 10 Only June - August counted i n t h i s f i g u r e . F i r s t insemination date t o l a s t b i r t h date. Ul i n Lemmus, as pregnant specimens were s t i l l being obtained i n the f i r s t h a l f of September when I l e f t . Whether Dicrostonyx behaved i n the same way i s not known because only three small young and one l a c t a t i n g adult female were caught a f t e r the end of August. In I960 summer breeding stopped at the end of J u l y , no Lemmus being impregnated a f t e r J u l y 20 or Dicrostonyx a f t e r J u l y 25. In 1961 breeding seemed to have stopped by mid-August i n Dicrostonyx and by the end of August i n lemmus. There were very few mature females l e f t i n e i t h e r species by August and no mature males were caught on the Main Study Area a f t e r August 1 i n Lemmus or August 3 i n Dicrostonyx. Under such circumstances i t i s rather d i f f i c u l t to pinpoint the end of summer breeding i n 1961, and some care must be exercised i n i n t e r p r e t i n g these f i g u r e s . At Aberdeen Lake changes i n the length of the summer breeding season of Dicrostonyx seemed even more s t r i k i n g than the changes on the Main Study Area. In I960 widespread evidence was obtained that breeding was stopping by J u l y 15, and i n 1961 by J u l y 25. The p r e c i s e end of the breeding season at Aberdeen Lake cannot be given f o r e i t h e r year because of no August data. Thus breeding seemed to be c u r t a i l e d i n both summers but s l i g h t l y e a r l i e r i n I960 than i n 1961. Lemmus at Aberdeen Lake behaved l i k e those on the Main Study Area. The extent of winter and spring breeding ( i . e . breeding under the snow) i n both species i s given i n Table lU. The data f o r 1958-59 are based on body weight d i s t r i b u t i o n s of June 1959» A few young Dicrostonyx were found which must have been born during the spring, but no young Lemmus. Both species bred extensively i n the winter of 1959-60. Pregnant female Lemmus were obtained i n A p r i l and May, and breeding males i n December, February, and A p r i l . Pregnant female Dicrostonyx were obtained on November 18, January 17, and March 2U, and breeding males i n November, January, 42 TABLE H i . Winter and spring breeding of Lemmus and Dicrostonyx, 1958-61, YEAR WINTER BREEDING"' SPRING BREEDING LEMMUS; 1958- 59 1959- 60 1960- 61 none ? extensive none none ? extensive some DICROSTONYX 1958- 59 1959- 60 1960- 6 1 none 1 extensive none some some some 1 Winter - September to A p r i l 15 o Spring - A p r i l 16 to May 31 1*3 February, March, A p r i l , and May. Since only a few winter specimens were obtained from t h i s winter (57 Dicrostonyx and 21 Lemmus), these data give only a q u a l i t a t i v e idea of winter breeding. In the winter of 1960-61, on the other hand, there was no breeding detected i n e i t h e r species (based on 65 Dicrostonyx and 21+5 Lemmus c o l l e c t e d throughout the winter). Spring breeding d i d occur i n 1961, and pregnant female Dicrostonyx were obtained on A p r i l 16 and May 3. Although no pregnant female Lemmus were obtained, females with f r e s h p l a c e n t a l scars and young animals were caught i n l a t e May. In summary, the major changes i n the length of the breeding season over the lemming cycle were: ( l ) extensive winter breeding only i n the phase of increase (1959-60); and (2) shortening of the summer breeding season both of the peak year (I960) and of the decline (1961). These e f f e c t s occurred i n both species. L i t t e r Size L i t t e r size at b i r t h i s a function of the ovulation rate and the prenatal m o r t a l i t y r a t e s . An approximation to l i t t e r size i s obtained by counting embryos i n pregnant females. We need to f i n d out whether there are any changes i n number of corpora l u t e a or number of embryos per pregnant female over the lemming c y c l e . There are at l e a s t eight i n t e r r e l a t e d v a r i a b l e s that may a f f e c t l i t t e r s i z e : season, food supply, body weight, age, p a r i t y , l a c t a t i o n , population density and s o c i a l structure, and p h y s i o l o g i c a l and genetic changes i n c o n s t i t u t i o n . Thus to say that l i t t e r size d i f f e r s between year x and year or group a and group b, i s to say very l i t t l e . I t i s necessary to correct f o r as many of these v a r i a b l e s as possible i n an assessment of l i t t e r s i z e changes and to compare only groups of s i m i l a r composition. These f a c t s have not always been appreciated by workers assessing reproduction and much confusion has thus r e s u l t e d . A preliminary" a n a l y s i s of the data i n d i c a t e d that body weight per se (indeptendent of p a r i t y and season) had no e f f e c t on corpora l u t e a or embryo counts, and t h i s v a r i a b l e was deleted from the f i n a l a n a l y s i s . Tables 1$ and 16 give the number of corpora l u t e a of pregnant Lemmus and Dicrostonyx females from the Main Study Area, and from these data we are l e d to the fo l l o w i n g conclusions: (1) Ovulation rate i n both species changes seasonally, d e c l i n i n g from higher values at the s t a r t of the summer t o lower values i n the l a t e summer. (2) Primiparous females tend t o have lower ovulation rates than multiparous females i n both species but the di f f e r e n c e s are s l i g h t , i n so f a r as can be generalized from the few samples which contain both groups. (3) Primiparous summer young have s i g n i f i c a n t l y lower ovulation rates than multiparous winter generation adults breeding at the same time. (U) F i n a l l y , and most important f o r our purposes, there are no s i g n i f i c a n t d i f f e r e n c e s i n ovulation rates of e i t h e r species between the years, when we compare s i m i l a r groups of animals. Tables 17 and 18 give the number of embryos of pregnant Lemmus and Dicrostonyx females from the Main Study Area. P r e c i s e l y the same four conclusions drawn from the corpora l u t e a data can be applied to these embryo data. Since a l l these data p e r t a i n only to the Main Study Area, i t i s reasonable to enquire whether these r e s u l t s are l o c a l or general. Fortunately data are ava i l a b l e from Aberdeen Lake, 115 miles west of Baker Lake, f o r I960 and 1961. Table 19 gives the corpora l u t e a and embryo counts f o r Lemmus at Aberdeen Lake i n i960 and 1961, and Table 20 the same data f o r Dicrostonyx. These data show the seasonal change observed above i n ovulation rate and l i t t e r s i z e . The Lemmus do not show any differ e n c e between ovulation rate or l i t t e r s i z e i n summer young and winter generation adults, contrary to what was observed above. F i n a l l y , there are no s i g n i f i c a n t 4 5 TABIE 15. Number of Corpora Lutea i n Lemmus females, Main Study-Area, Summers 1969-61. GROUP I PERIOD I I PERIOD I I I PERIOD IV PERIOD N MEAN SE N MEAN SE; N MEAN SE. N MEAN SE WINTER GENERATION 1959 Primiparous 12 6.92 ±.36 -Multiparous - - 3 7.33 +.67 I960 Primiparous 1$ 7.80 ±.39 - NOT Multiparous 10 8.10 ±.28 18 7.72 ±.21* 13 6.62 ±.33 BREEDING 1961 Primiparous 8 7.25 ±.31 -Multiparous - 5 8.20 +.37 8 6.75 ±31 1 7.00 Y± SUMMER YOUNG 1959 Primiparous - - 7 5.57 +.37 Multiparous - -I960 Primiparous - - 10 5.20 +.36 NOT Multiparous - - - BREEDING 1961 Primiparous - - 13 5.31 ±.29 Multiparous - -Y1 1 SUMMER YOUNG 1959 Primiparous - - - 5 3.80 +.37 Multiparous - - - -i960 Primiparous - - NOT Multiparous - BREEDING 1961 Primiparous - - — 1 U.00 Multiparous -46 TABLE 16. Number of corpora l u t e a i n Dicrostonyx females, Main Study Area, summers 1959-1961. GROUP I PERIOD I I PERIOD I I I PERIOD N MEAN SE N MEAN SE N MEAN SE TOTTER GENERATION 1959 Primiparous 9 7.00 ±0.50 Multiparous - 2 6,50 ±0.50 I 5.00 I960 Primiparous 9 7.00 +0.U7 1 6.00 Multiparous 2 9.50 +1.50 9 7.33 +0.80 6 5.67 +0.8U mm 1961 Primiparous 23 6.96 ±0.35 Multiparous 1 7.00 7 8.1(3 ±0.72 1 9.00 SUMMER YOUNG 1959 Primiparous - - -I960 Primiparous -1961 Primiparous - 2 U.50 ±0.50 4 7 TABLE 17, Number of embryos i n Lemmus females, Main Study Area, summers 1959-61. GROUP I PERIOD I I PERIOD I I I PERIOD I? PERIOD N MEAN SE. N MEAN SE N MEAN SE N MEAN SE WINTER GENERATION 1959 Primiparous 12 6.33 ±.36 - - -Multiparous - 3 6.67 ±.67 I960 Primiparous 15 7.27 ±.37 - - NOT Multiparous 10 7.50 ±.3U 18 7.11 ±.28 13 6.23 ±.28 BREEDING 1961 Primiparous 8 7.00 *.38 -Multiparous - 5 7.80 ±.U9 8 6.75 ±.31 1 6.00 Y x SUMMER YOUNG 1959 Primiparous - — 7 5.U3 ±.U8 -Multiparous -I960 Primiparous - 10 5»00 ±.30 NOT Multiparous - BREEDING 1961 Primiparous - - 13 U.92 +.33 Multiparous - - -Y x « SUMMER YOUNG 1959 Primiparous - - — 5 3.80 +.37 Multiparous -I960 Primiparous - - - NOT Multiparous - - - BREEDING 1961 Primiparous - - - 1 U.00 Multiparous 48 TABLE 18. Number of embryos i n Dicrostonyx females, Main Study Area, summers 1959-61. GROUP I PERIOD I I PERIOD I I I PERIOD N MEAN SE N MEAN SE N MEAN SE WINTER GENERATION 1959 Primiparous 9 6.11 ±0.U6 Multiparous - 2 6.00 ±1*00 1 1.00 I960 Primiparous 9 6.11 ±0.51 1 6.00 Multiparous 2 6.00 ±2.00 9 5.00 ±0.58 6 U.67 ±0.i;9 1961 Primiparous 23 5.61 ±0.ljl - -Multiparous 1 1.00 7 5.29 ±0.61+ 1 8.00 Y± SUMMER I0UNG 1959 Primiparous -I960 Primiparous - - -1961 Primiparous - - 2. 2.50 ±1.50 4 9 TABLE 19. Number of corpora l u t e a and embryos i n lemmus females at Aberdeen Lake, summers 1960-61. GROUP I PERIOD 1 I I PERIOD I I I PERIOD N MEAN SE' N MEAN SE N MEAN SE NUMBER OF CORPORA Winter Generation I960 Primiparous Multiparous 1961 Primiparous Multiparous Y^ Summer Young I960 Primiparous 1961 Primiparous NUMBER OF EMBRYOS Winter Generation I960 Primiparous Multiparous 1961 Primiparous Multiparous Y-|_ Summer Young I 9 6 0 Primiparous - 3 6 . 0 0 ±0.U7 1961 Primiparous - - 1 6 . 0 0 Insemination dates r I Period - June k-93 I960; June 10-12:, 1961; I I Period June 21+-27, 1960j I I I Period - J u l y 6-10, I96O5 J u l y 10-20, 1961. LUTEA k 9.00 tO.Ul 5 7.20 +1.07 h 5.50 ±0.91 1 9.00 3 6.00 ±0.00 3 6.00 ±0.1+7 1 6.00 U 8.75 +0.63 5 7.00 ±1.00 k 5.50 ±0.91 1 9.00 3 6.00 ±0.00 50 TABLE 20. Number of corpora l u t e a and embryos i n Dicrostonyx females at Aberdeen Lake, summers l °60-6l. GROUP I PERIOD 1 I I PERIOD I I I PERIOD N MEAN SE N MEAN SE N MEAN SE NUMBER OF CORPORA LUTEA Winter Generation I960 Primiparous 7 7.00 ±0.72 Multiparous - 9 5.67 ±0.61; 1961 Primiparous - - -Multiparous - 2 U.00 ±1.00 1 5.00 NUMBER OF EMBRYOS Winter Generation I960 Primiparous 7 5.57 *0.57 Multiparous - 9 U.00 £0.65 1961 Primiparous U 5.50 ±0.61; - -Multiparous - 8 U.00 ±0.38 2 2.50 ±1.50 1 Insemination dates* I Period ~ June 3-9, 1960j June 5-lk, 1961$ 11 Period — June 21-July 9, I960: June 26-July 7, 1961$ I I I Period J u l y 16-20, 1961. 51 d i f f e r e n c e s between the years i n e i t h e r variable f o r e i t h e r species. The seasonal trend i n l i t t e r s i z e c a r r i e s through i n t o the winter, as f a r as our meager winter records i n d i c a t e . L i t t e r s i z e s of the pregnant females obtained i n winter are l i s t e d below. Date Number of Embryos Lemmus Dicrostonyx: November 18, 1959 3 January 17, I960 3 March 2U, I960 3 A p r i l 25, I960 U May 21, i960 5? May 2U, I960 3 A p r i l 16, 1961 3 May 3, 1961 h In conclusion, there seemed to be no s i g n i f i c a n t change over the cycle i n e i t h e r ovulation rate or l i t t e r s i z e i n Lemmus or Dicrostonyx. There was a seasonal trend i n these v a r i a b l e s independent of the cycle i n numbers. Pregnancy Rates Given a summer breeding season of a c e r t a i n length, we may enquire what proportion of mature females i s pregnant at various times i n t h i s breeding season and subsequently whether there are diff e r e n c e s between years i n t h i s v a r i a b l e . The ana l y s i s of pregnancy rates used here follows that of L e s l i e et a l . (1952). Table 21 gives the crude (observed) pregnancy rates f o r Lemmus and Table 22 the rates f o r Dicrostonyx from a l l areas during t h i s study. Since animals i n very e a r l y pregnancy w i l l not be c l a s s i f i e d as pregnant macroscopically, these crude pregnancy rates tend to underestimate the ac t u a l pregnancy rates such that a 0.750 crude pregnancy rate ( i . e . 15/20) f o r Dicrostonyx and a 0.762 pregnancy rate ( i . e . 16/21) f o r Lemmus w i l l be equivalent to every female i n the population being pregnant a l l the time. These rates are expressed per female >20.5 grams f o r Lemmus and >30.5 grams 52 TABLE 21. Crude pregnancy rates per 1 6 1 days per female >20.5 grams, lemmus, summers 1959-61. LOCATION MD TIME PERIOD WINTER GENERATION SUMMER YOUNG N PREG. RATE Y 1 LITTER N PREG. BITE Y » LITTER N PREG. RATE MAIN STUDY AREA 1959 June 18-30 U+ 0.711+ J u l y 3 0.667 August 1+ 0.500 I960 May 16-31 H+O 0.007 June 1-15 31 0.516 June 16-30 21 0.667 J u l y 1-15 16 0.812 Ju l y 16-31 26 0.538 August 1-15 20 0.050 August 16-31 1+ 0.000 1961 May 16-31 6 0.000 June 1-15 20 0.150 June 16-30 5 0.800 J u l y 1-15 2 1.000 Ju l y 16-31 6 0.333 August 1-15 3 0.333 August 16-31 -To t a l l i t t e r Production f o r June, J u l y and August 1959 2.81+1 1960 2.613 1961 2.996 ABERDEEN LffiKE AREA I960 May 27-June 2 9 0.000 June 15-16 it 1.000 J u l y 10-18 12 0.750 1961 June 2-5 3 0.000 June 22 1 1.000 J u l y 26-29 7 0.1+29 1 12 1 U7 1+9 13 3 1 8 1.000 0.583 0.000 0.231+ 0.020 0.000 1.000 0.000 1.568 0.253 1.1+69 0.375 1.000 n 20 0.000 0.000 0.000 0.000 0.813 0.000 0.000 0.200 Estimated p o r t i o n of the 21 day gestation period f o r which pregnancy can be recognized macroscopically. 55 TABLE 21 (continued). Lemmus crude pregnancy r a t e s . LOCJfflON AND TIME WINTER GENERATION SUMMER YOUNG PERIOD Y± LITTER T • LITTER N PREG. RATE N PREG. RATE N PREG. RATE OTHER AREAS 1961 Long Island J u l y 17-20 Second Island J u l y 21+-27 Prince River August lU-17 Nine Mile I s . August LU-19 Ten Mile I s . August lii-19 Thelon River August U4.-I9 0,250 0.667 mm 1.000 1.000 6 6 1 2 3 O.667 0.833 0.000 0.000 0.333 U 1 2 U 0.000 0.000 0.000 0.000 5 4 TABLE 22. Crude pregnancy rates per 15"*" days per female >30.5 grams, Dicrostonyx, summers 1959-61. LOCATION AND TIME PERIOD WINTER GENERATION N PREG. RATE T SUMMER YOUNG !f PREG. RATE MAIN STUDY AREA 1959 June 15-30 3 J u l y 9 August 3 I960 May 16-31 6 June 1-15 22. June 16-30 5 J u l y 1-15 10 J u l y 16-31 6 August 1-15 7 August 16-31 10 1961 May 16-31 3 June 1-15 50 June 16-30 25 J u l y 1-15 10 J u l y 16-31 5 August 1-15 h August 16-31 1+ TOTAL LITTER PROD-UCTION FOR JUNE, JULY AND AUGUST 1959 I960 1961 ABERDEEN LAKE AREA I960 May 27-June 2 4 June 15-16 7 J u l y 10-18 26 1961 May 28-June 7 5 June 15-22 11 J u l y 10-19 17 J u l y 26-29 17 0.667 0.889 0.333 0.000 O.364 1.000 1.000 0.833 0.H|3 0.000 0.000 0.060 0.760 0.600 0.800 0.250 0.000 3.190 3.394 2-524 0.000 1.000 0.3U7 0.200 O.364 0.412: 0.176 2 5 13 0.000 0.000 0.000 14 1 0.214 0.000 ? 0.000 0.228 0.000 Estimated p o r t i o n of the 20 day gestation p e r i o d f o r which pregnancy can be recognized macroscopically. f o r Dicrostonyx because these are the weights above which a majority of females can be mature under good conditions* F i n a l l y , these data are given i n terms of crude pregnancy rates instead of standardized pregnancy rat e s ( L e s l i e e t a l . 1952) because a f t e r a complete standardization of the data there was hardly any change i n the rates and consequently there was no need to include the standardized r a t e s . I f we examine these data (Tables 21, 22) we see that there i s a general r i s e i n the pregnancy rate from zero i n May to high values by June 15 and a subsequent decline i n August. We are not i n t e r e s t e d here i n the timing of t h i s r i s e and f a l l because t h i s has been t r e a t e d under the previous section on the length of the breeding season. What we are int e r e s t e d i n i s the period of midsummer when breeding i s neither s t a r t i n g up nor beginning to stop, and we wish to enquire whether there are s i g n i f i c a n t d i fferences between the years i n the rates during t h i s p e riod. Pregnancy rates during midsummer (June l 5 - J u l y 31) were compared b f o r the Main Study Area and f o r Aberdeen Lake i n JBbth Lemmus and Dicrostonyx. Chi-square t e s t s (Snedecor, 1956, p 228) were made with the following r e s u l t s f o r the winter generation animals: S i g n i f i c a n t Differences Lemmus Dicrostonyx ( l ) Between years: Main Study Area, 1959-61 Aberdeen Lake, 1960-61 - * (2) Between areas: Main Study Area vs. Aberdeen Lake 1960 -1961 -( - * P >.10: * = P <.025, >,01; ** = P <.005) While Lemmus showed no di f f e r e n c e s i n pregnancy rates whatever e i t h e r over 56 the cycle or between d i f f e r e n t areas, Dicrostonyx showed a s i g n i f i c a n t lowering of the midsummer pregnancy rate i n the decline (l°6l) at Aberdeen Lake but no d i f f e r e n c e s on the Main Study Area. Furthermore, i n each year i960 and 1961 the pregnancy rates were s i g n i f i c a n t l y lower at Aberdeen Lake compared with Baker Lake. Looking more c l o s e l y at the Aberdeen Lake Dicrostonyx data, we see that these differences can be explained by a curtailment of breeding i n e a r l y J u l y , i . e . t h a t the summer breeding season was shortened d r a s t i c a l l y i n I960 and 1961. The one point that i s then l e f t to be explained i s the mid-June 1961 r a t e of O.36I4., but t h i s may be due to the l a t e spring phenology of 1961 such that these animals were j u s t beginning to breed. Leaving the winter generation adults and looking at the pregnancy rates f o r the summer young, we f i n d considerable v a r i a t i o n between years. However, the pertinent f a c t o r s involved here are changes i n the length of the breeding season and r e l a t e d changes i n the age at sexual maturity of these young; the former was treated above, the l a t t e r w i l l be dealt with below. Consequently, a s t a t i s t i c a l analysis was not done on the data f o r these summer young. One aspect of Tables 21 and 22 has not been discussed yet, the f i g u r e s f o r t o t a l l i t t e r production. These are obtained i n the manner described by L e s l i e e t a l . (1952) by applying the observed rates to the length of t h e i r p a r t i c u l a r time p e r i o d and summing the r e s u l t s . Unfortunately i t i s not possible to attach s t a t i s t i c a l confidence l i m i t s to these numbers because they are sums of weighted averages. These f i g u r e s are hypothetical i n that they indicate the number of l i t t e r s an average mature female would produce i f she l i v e d over the en t i r e period between June 1 and August 31 f o r the winter generation, or i n the case of the summer young over the p e r i o d bejsween reaching a mature weight and the end of August. Furthermore, these 57 production f i g u r e s are rather a r b i t r a r i l y l i m i t e d to June 1 to August 31 because most of the sampling was done at t h i s time. For Lemmus there i s a s l i g h t depression of t o t a l l i t t e r production i n the peak summer of I960, but t h i s i s small i n view of the f a c t that t h i s summer had a shortened breeding season. For Dicrostonyx, on the other hand, there i s an apparent increase i n t o t a l l i t t e r production i n the peak summer, a f a c t that seems to clash with the previous observation that t h i s summer was characterized by a shortened breeding season. This anomaly i s explained i n part by the f a c t that i n I960 summer breeding began e a r l i e r than i n eit h e r 1959 or 1961 and i n part by random sampling v a r i a t i o n s i n the observed pregnancy r a t e s . In conclusion, on the Main Study Area the midsummer pregnancy rates d i d not change s i g n i f i c a n t l y from year to year i n e i t h e r species of lemming. The same was true at Aberdeen Lake f o r Lemmus but not f o r Dicrostonyx which seemed to show depressed pregnancy rates i n the summer of the d e c l i n e . A l l other observed changes i n pregnancy rates were r e f l e c t i o n s on changes i n the length of the breeding season or the age at sexual maturity. Age at Reproductive Maturity The age at which reproduction begins i s of the utmost importance i n determining the i n t r i n s i c rate of increase of a population (Cole, 195U a). Since we do not know the age of the specimens obtained i n t h i s study, i t i s necessary to use body weight as an index of age. As mentioned above, the c r i t e r i a of maturity were the presence of corpora l u t e a i n females and of v i s i b l e epididymal tubules i n males. Tables 23-26 give the percentage of mature animals i n the various weight classes f o r Lemmus and Dicrostonyx males and females. The method of L e s k e , Perry, and Watson (19U5) was used to determine the median body weight at maturity f o r the various groups. In b r i e f t h i s technique involves TABLE 23. Weight at reproductive maturity i n Lemmus males, summers 1959-61. WEIGHT CLASS (g) WINTER GENERATION 1959 I960 1961 1959 SUMMER GENERATION YOUNG I960 1961 1959 Y x» YOUNG I960 1961 11-20.5 g. mm 0* (28) 0 (2) - 0 (9) 0 (11) 0 (2) 0 (12) 0 (5) 21-30.5 mm 10 (20) 0 (1) 0 (2) 0 (10) 0 (12) 50 (6) 0 (16) 0 (27) 31-U0.5 100 (5) 58 (210 - 80 (*) 0 (65) 0 (9) 100 (1) 0 (28) 0 (3) Ui-50.5 100 (5) 81 (3D 100 (6) - 0 (19) - - 0 (2) -5i-6o.5 100 (9) 98 (59) 100 (12) - 0 (5) - - - -61-70.5 100 (2) 99 (98) 100 (lit) - mm - - - -71-80.5 100 (3) 96 (65) 100 (9) - - - - - -81-90.5 100 (1) 96 (72) 100 (1) - mm* - - - -91-100.5 - 97 (29) mm - mm mm mm - -ioi-no.5 - 100 (10) 100 (1) mm - - - mm -Percentage mature; sample s i z e i n parentheses. TABLE 2lu Weight at reproductive maturity i n Lemmus females, summers 1959-61. WEIGHT WINTER GENERATION CLASS (g) 1959 i960 1961 1959 11-20,5 - 0 * (19) 0 (1) mm 21-30.5 100 (1) 0 (12) 0 (2) 100 (1) 31-U0.5 50 (6) 5 (21) 50 (6) 100 ( 8 ) Ul-50.5 100 (6) 21 ( U 8 ) 6U (11) 100 (2) 51-60.5 100 (U) 52 (52) 100 (16) -61-70.5 100 (U) 92 U8) 100 (1U) -71-80.5 100 (1) 100 (25) 8 8 ( 8 ) -81-90.5 - 100 (26) 100 (1) -91-100.5 mm 100 (16) - -101-110.5 - 100 (3) - -* Percentage maturej sample size i n SUMMER GENERATION Y x YOUNG Y ^ YOUNG I960 1 9 6 1 1959 I 9 6 0 1 9 6 1 33: ( 8 ) 0 (18) 0 (2) 0 (7) 0 (3) UO ( 1 0 ) 75 ( 1 6 ) 1 0 0 (3) 0 (U) 32 (19) 39 ( 6 1 ) 1 0 0 (7) 1 0 0 (2) 0 (25) 1 0 0 ( 1 ) 8 0 (15) 1 0 0 (2) - 0 ( 1 ) -1 0 0 (7) 1 0 0 (15 - - -1 0 0 (2) - - - -ses. TABLE 25. Weight at reproductive maturity i n Dicrostonyx males, summers 1959-61. WEIGHT WINTER GENERATION SUMMER GENERATION GLASS (g) YOUNG Y « YOUNG 1959 I960 1961 1959 I960 1961 1959 I960 1961 11-20.5 - 0 * (7) 0 (3) 0 (U) 0 0 (2) (2) 0 (1) 0 (5) 21-30.5 50 (2) 0 (U) 0 (12) - 0 (3) 0 0 (7) (3) 0 (U) 0 (3) 31-40.5 78 (9) Uo (5) 71 (1U) 0 (1) 0 (3) 0 (1) 0 (6) 0 (1) Ul-50.5 100 (7) 75 (U) 69 (13) mm 0 (12) mm mm 0 (2) 51-60.5 86 (7) 77 (13) 97 (31) mm 0 (3) - - -61-70.5 - 83 (12) 96 (27) - - - - -71-80.5 - 80 (15) 100 (10) - -mm mm - -81-90.5 - 100 (7) 100 (5) - - - - -91-100.5 - 100 (3) 100 (2) - - - - -101-110.5 - 67 (3) mm - - mm mm - -•a Percentage mature j sample siz e i n parentheses. TABLE 26, Weight at reproductive maturity i n Dicrostonyx females, summers 1959-61, WEIGHT GLASS (g) WINTER GENERATION SUMMER GENERATION Y1 YOUNG Y-L' YOUNG 1959 I960 1961 1959 I960 1961 1959 i960 1961 11-20.5 0 * (1) 0 (6) 0 (11) 0 (2) 0 (1) 7 (15) 0 (1) 0 (10) 0 (3) 21-30.5 25 (U) 0 (3) 9 (11) 0 (1) 0 (2) 33 (3) 0 (7) 0 (6) 0 (2) 31-1+0.5 100 (3) 0 (U) 10 (10) - 0 (9) 100 (1) - 0 (7) 0 (1) ia-5o.5 100 (5) 60 (5) 61+ (25) - 12 (17) mm - - -51-60.5 100 (5) 89 (9) 80 (25) - 0 (2) - - - -61-70.5 100 (3) 100 (19) 93 (28) - - - - - -71-80.5 - 100 (11) 78 (7) - - - - - -81-90.5 - 100 (9) 100 (2) - - mm - - -91-100.5 - 100 (U) 100 (1) - - - - - -101-110.5 mm 100 (1) 100 (1) mm - - - - -* Percentage mature; sample siz e i n parentheses. 62 converting the weight data i n t o logarithms and percent mature data into p r o b i t s , f i t t i n g a s t r a i g h t l i n e to t h i s , and then c a l c u l a t i n g the $0% point and i t s standard e r r o r . The r e s u l t s are summarized i n Table 27. Some of the data were not s u f f i c i e n t to cal c u l a t e the median body weight at maturity and i n these classes the upper or lower l i m i t s possible f o r the median were in d i c a t e d . Data from the Main Study Area f o r the whole summer were grouped i n t h i s a n a l y s i s , but i n the actu a l c a l c u l a t i o n s the winter generation r e s u l t s are based mainly on the May and June samples and the summer generation r e s u l t s on J u l y and August information. These data show s t r i k i n g changes i n the median body weight at maturity between the d i f f e r e n t years. In every case i n the peak summer of 1960 there was an increase i n the median body weight at maturity. In the 1961 summer of decline three patterns could be found: ( l ) median weights remained the same as I960, as i n winter generation Dicrostonyx of both sexes and Lemmus males.; (2) median weights declined to a p o s i t i o n intermediate between 1959 and i960 l e v e l s , as i n the winter generation Lemmus females; and (3) median weights declined to the same l e v e l s as 1959, as i n the summer young Lemmus and Dicrostonyx females. Missing from t h i s c l a s s i f i c a t i o n are the summer young males of both species because none of these became sexually mature.in e i t h e r the summer of I960 or the summer of 1961. The summer of 1959 seems to represent the most r a p i d rates of maturation found i n both species. Thus Lemmus females were mature at 20-25 grams and males at 25-35 grams, representing roughly 3-U weeks and U-5 weeks of age r e s p e c t i v e l y . Dicrostonyx males and females were mature at about 30 grams, representing roughly U-5 weeks of age. In neither of the other years were these r a p i d rates of maturation found, with the exception of the 1961 young females, and i t i s these deviations from the possible rates of maturation that must be explained. 65 TABLE 27. Median body weights at maturity f o r Lsmmus and Dicrostonyx males and females, 1959-61. GROUP £ND YEAR MALES FEMALES. MEDIAN 95% CONF. MEDIAN 95% CONF. WEIGHT # LIMITS WEIGHT LIMITS LEMMUS WINTER GENERATION 1959 <30.5 - <26 -1960 36.7 3U.2 - 39.U 51.6 U9.8 - 53.6 1961 31-Ul - I4I.2 37.U - k5.h Yn SUMMER YOUNG 1959 33.8 21+.6 - U6.6 < 26 1960 > 61 - 29.1 26.6 - 31.8 1961 > Ul - 21.3 20.1 - 22.6 Y-,' SUMMER YOUNG 1959 26.5 - 20-25 1960 > 51 - > hx 1961 > Ul - 2U.U 22.6 - 26.U DICROSTONYX WINTER GENERJSTION 195J9 28.3 23.2 - 3U.5 30.5 23.2 - Uo.3 1960 U3.3 38.5 - U8.8 U9.9 1*3.5 - 57.3 1961 38.8 36.7 - Ul.O U9.3 U6.9 - 51.7 Y i SUMMER GENERATION 1959 ? ? -1960 > 51 - > Ul 1961 > 31 - 2U.0 22.3 - 27.1 * Weights i n grams. 6U To sum up, there are s t r i k i n g changes i n the median weight at sexual maturity over the c y c l e . These changes consist i n a general increase i n the median weight at maturity i n the peak summer i n a l l sexes and generations, and a complete lack of maturation of a l l summer young males i n both the peak summer and the summer of d e c l i n e . In only one summer out of three d i d e i t h e r species show maximal rates of maturation. Embryo Rates I t i s convenient to have one f i g u r e which sums up most of the components of reproduction to give some assessment of t o t a l p r o d u c t i v i t y . This can be done by the use of embryo rates, f o l l o w i n g the method of L e s l i e et a l . (1°£2). i f w e observe a sample of N mature females, P of which are pregnant and which contain a t o t a l of E l i v e : embryos, we have f o r the embryo rate E/N = P/N X E/P that i s , the embryo rate i s the proportion of mature females pregnant m u l t i p l i e d by the mean l i t t e r s i z e . Since the pregnancy rate r e f l e c t s changes i n the length of the breeding season, changes i n the weight at maturity, and changes i n the proportion pregnant during the breeding season, t h i s equation e f f e c t i v e l y sums up most of the components of reproduction ( c f . Figure $ ) , Tables 28 and 29 give the crude embryo rates f o r Lemmus and Dicrostonyx. Crude rates are given here instead of standardized rates because there i s very l i t t l e d i f f e r e n c e between the two. T o t a l embryo , production was obtained i n the same way as t o t a l l i t t e r production above, by applying the observed rates to the length of t h e i r p a r t i c u l a r time p e r i o d and summing the r e s u l t s . These production f i g u r e s are hypothetical; f o r the winter generation they give the number of embryos a mature female would produce i f she l i v e d throughout the whole summer and bred at the 6 5 TABLE 28. Crude embryo rat e s per 16^  days per female >20.5 grams, Lemmus, 1959-61. LOCATION AND TIME PERIOD WINTER GENERATION YOUNG SUMMER GENERATION Y x« YOUNG N EMBRYO RATE N EMBRYO RATE N EMBRYO RATE MAIN STUDY AREA 1959 June 15-30 llj. U.57 J u l y 3 4.67 August 4 3.50 I960 May 16-31 I4O 0.02 June 1-15 31 3. 94 June 16-30 21 U.86 J u l y 1-15 16 5.69 J u l y 16-31 26. 3.46 August 1-15 20 0.25 August 16-31 4 0.00 1961 1 12 1 47 U9 13 6.00 3.17 0.00 1.17 0.10 0.00 T o t a l Embryo Producation f o r June, J u l y and August 1959 I960 1961 ABERDEEN LAKE I 9 6 0 May 2 7 - J u n e 2 9 June 15-16 4 J u l y 1 0 - 1 8 1 2 1 9 6 1 June 2-5 3 June 2 2 1 J u l y 2 6 - 2 9 7 19.5U 17.27 22.39 0.00 8.75 4.75 0.00 9.00 2.57 8.76 1.27 5.88 2.25 3.80 11 20 0.00 0.00 May 16-31 6 0.00 - _ _ June 1-15 20 0.90 - mm - _ June 16-30 5 6.00 _ mm _ J u l y 1-15 2 8.50 - mm - _ J u l y 16-31 6 2.16 3 4.00 - -August 1-15 3 2.00 — - 3 0.00 August 16-31 - - 1 0.00 6 0.00 3.09 0.00 0.00 1.20 Estimated p o r t i o n of the 21 day gestation p e r i o d f o r which pregnancy can be recognized macroscopicaHy. 66 TABLE 29. Crude embryo rates per 1$^ days per female >30.5 grams Dicrostonyx, 1959-61. LOCATION JND TIME PERIOD MALTS STUDY AREA 1959 June 15-30 J u l y August I960 May 16-31 June 1-15 June 16-30 J u l y 1-15 J u l y 16-31 August 1-15 August 16-31 1961 May 16-31 June 1-15 June 16-30 J u l y 1-15 J u l y 16-31 August 1-15 August 16-31 TOTAL EMBRYO PROD. June - August 1959 I960 1961 ABERDEEN LAKE AREA I960 May 27-June 2 June 15-16 J u l y 10-18 1961 May 28-June 7 June 15-22 J u l y 10-19 Ju l y 26-29 WINTER GENERATION N EMBRYO RATE: SUMMER GENERATION N EMBRYO RATE 3 9 3 6 22 5 10 6 7 10 3 50 25 10 5 ii U U 7 26 5 11 17 17 li.00 5.22 0.33 0.00 2.23 6.U0 5.10 1+.00 0.57 0.00 0.00 0.U8 3.96 2.70 U.00 2.00 0.00 15.U8 18.56 13.U1 0.00 5.57 1.28 0.60 2.00 1.59 0.59 2 5 13 lU 1 0.00 0.00 0.00 0.71 0.00 ? 0.00 0.76 0.00 Estimated p o r t i o n of the 20 day gestation period f o r which pregnancy can be recognized macroscopically. 67 observed rates, and f o r the summer generation the number of embryos a female would produce from the time of reaching minimum mature weight to the end of August. The most s t r i k i n g changes i n t o t a l reproductive e f f o r t occur i n Lemmus. I f we assume f o r the moment that female m o r t a l i t y rates were the same i n a l l years, the r a t i o of t o t a l reproductive output f o r the summers of 1959-61 i s 31.1+ - 18.5 - 28.3 embryos, or approximately 1.70 - 1.00 -1.53. As we have seen, these d i f f e r e n c e s a r i s e p r i n c i p a l l y because of changes i n the length of the breeding season and i n the age at maturity i n the young. In Dicrostonyx, on the other hand, t o t a l embryo production was apparently highest i n the peak summer of I960. This contrast with Lemmus i s brought about i n part by b i o l o g i c a l d i f f e r e n c e s ( i . e . young Dicrostonyx mature at an older age than lemmus and Y-j_' young Dicrostonyx never seem to mature i n t h e i r f i r s t summer; the breeding season of 1961 i n Dicrostonyx was apparently shorter than that of i960) and i n part by s t a t i s t i c a l d i f f i c u l t i e s ( i . e . 1959 and I96I Dicrostonyx sampling i n August and September was i n s u f f i c i e n t to determine accurately i f and when breeding stopped or young matured). For these reasons I do not place too much confidence i n the t o t a l embryo production f i g u r e s of 1959 and I96I f o r Dicrostonyx. There i s a suggestion of decreased p r o d u c t i v i t y during the decline on the Main Study Area; the data from Aberdeen Lake also suggest t h i s . T o t a l embryo production f o r the e n t i r e year cannot be determined because there are i n s u f f i c i e n t winter and spring data, and i t i s necessary to bear i n mind that the above f i g u r e s are only summer production. Comparison of t o t a l summer production of Lemmus with that of Dicrostonyx shows that the p o t e n t i a l rate of increase of Lemmus i s about 68 twice as great as that of IDicrostonyx under good conditions. To sum up, t o t a l embryo production during the summer i n Lemmus was high i n 1959 and 1961 and low i n I960, but i n Dicrostonyx seemed to be lower i n 1961 than i n i960. SUMMARY AND CONCLUSIONS (1) There was no s i g n i f i c a n t change i n l i t t e r s i z e or midsummer pregnancy rates over the c y c l e . (2) There were two changes i n the length of the breeding season: f i r s t , extensive winter breeding occurred only i n the winter of increasej and second, there was a shortened summer breeding season i n the peak summer and also to some degree i n the d e c l i n e . (3) The median weight at sexual maturity was higher i n the peak summer i n a l l groups and remained high i n most groups i n the summer of the decline (except f o r young females). Young Lemmus males d i d not mature i n e i t h e r the peak or the decl i n e , whereas young females d i d mature i n the decline but not i n the peak summer. MORTALITY The second major f a c t o r which causes changes i n population d e n s i t y i s m o r t a l i t y . T h i s f a c t o r begins i t s operation at ovulation and may be conveniently subdivided i n t o pre-natal m o r t a l i t y and post-natal m o r t a l i t y . The purpose of t h i s section i s to assess the importance of these components. METHODS Prenatal m o r t a l i t y i s assessed by comparing counts made of corpora l u t e a , implantation s i t e s , and l i v i n g embryos. The methods of c o l l e c t i n g these data were discussed i n the section on reproduction. Data on post-natal m o r t a l i t y were obtained from l i v e trapping. The methods used i n l i v e trapping were discussed i n the section on population density. RESULTS Prenatal M o r t a l i t y Prenatal m o r t a l i t y was assessed as f a r as possible by the methods of Brambell and M i l l s (1947, 1 9 U 8 ) . Prenatal m o r t a l i t y may be subdivided as follows: 1. P a r t i a l p r e n a t a l l o s s (at le a s t one embryo survives u n t i l p a r t u r i t i o n ) a. pre-implantation b. post-implantation 2. T o t a l l i t t e r l o s s a. pre-implantation b. post-implantation P a r t i a l pre-implantation m o r t a l i t y i s estimated from discrepancies between corpora l u t e a counts and implantation s i t e counts. From these data we estimate the amount of l o s s of ova i n l i t t e r s that survive implantation. P a r t i a l 70 post-implantation m o r t a l i t y i s estimated from discrepancies between the number of implantation s i t e s and the number of l i v i n g embryos i n ithe uterus. From these data we estimate the number of implanted embryos which f a i l to survive, and t h i s may incltide whole l i t t e r s i n the process of being l o s t . This estimate i s always an underestimate because the females counted come from varying stages between implantation and b i r t h . I d e a l l y counts should be made only on females i n the very l a t e stages of pregnancy, but too few were obtained i n t h i s study. Tables 30 and 31 summarize these data on p a r t i a l p renatal m o r t a l i t y i n Lemmus and Bicrostonyx. A l l data from each summer were grouped to obtain these estimates. Differences between the years were tested by chi-square (Snedecor, 1956, p 228) and a l l found to be n o n - s i g n i f i c a n t . V a r i a t i o n between years was s l i g h t ; i n Lemmus the t o t a l l o s s of ova amounted to k-9% and i n Dicrostonyx to 23-26$. Dicrostonyx s u f f e r s considerably more p a r t i a l p r e n a t a l l o s s than does Lemmus. No information on the l o s s of whole l i t t e r s before or during implantation i s given by the above a n a l y s i s . No l i t t e r s were found undergoing complete resorption i n middle or l a t e pregnancy i n t h i s study. However, i n d i r e c t evidence suggests that under some conditions i n Lemmus complete res o r p t i o n of l i t t e r s j u s t a f t e r implantation does occur e s p e c i a l l y i n young animals. In l a t e J u l y I960 young Lemmus 25-35 grams i n weight with very f a i n t p l a c e n t a l scars, small corpora a l b i c a n t i a , and no active mammary t i s s u e began to appear i n the samples. Since i t was quite impossible f o r these animals to have weaned a l i t t e r already (they were only k-5 weeks old) and since the scars were so f a i n t , a reasonable i n t e r p r e t a t i o n i s that these animals l o s t t h e i r e n t i r e l i t t e r s j u s t a f t e r implantation. Although some of these animals were probably missed during autopsy because of the very small s i z e of these scars and corpora, a minimal assessment of the frequency of 71 TABLE 30. P a r t i a l p r e n a t a l m o r t a l i t y data f o r Lemmus females, summers 19$°-6lj Main Study Area. TYPE OF LOSS 1959 I960 l?6l Pre-implantation l o s s % l i t t e r s showing 16.7 31.3 lk .7 l o s s N° 18 N - 67 N - 3k % ova l o s t 3.9 5.3 2*3 N =» 102 H - U76 H • 216 Post-implantation l o s s % l i t t e r s showing 5.6 17»9 8,8 l o s s N - 18 N - 67 N - 3k % embryos resorbing 2*0 U»2 1»9 N « ?8 N = U5l N - 211 72 TABIE 31. P a r t i a l p r e n a t a l m o r t a l i t y data f o r Dicrostonyx females, summers 1959-61, Main Study Area. TYPE OF LOSS 195° I960 1961 Pre-implantation l o s s % l i t t e r s showing l o s s % ova l o s t Post-implantation l o s s % l i t t e r s showing l o s s % embryos resorbing 6 3 . 6 N - 1 1 1 6 . 9 N - 7 7 18.2 N - 11 7 . 8 N - 6k kk.o N = 25 19.7 N = 178 20.0 N = 25 6.3 N = ll|3 55*9 N = 3k 1 8 . 2 N - 21+2 32.3 N « 3U 9.1 N - 198 73 t h i s t o t a l l i t t e r l o s s may be made from the snap trapping samples. No Dicrostonyx have yet been seen with these c h a r a c t e r i s t i c s . No winter generation Lemmus showing these p e c u l i a r i t i e s have been found, and thus the condition appears to be found only i n Lemmus summer young. The following samples could have contained t h i s type of young Lemmus: No. young females No. mature No. showing > 20.5 grams (with corpora evidence of lutea) t o t a l l i t t e r l o s s Main Study Area 1959 August-Sept. 10 16 16 0 I960 J u l y 16-31 la 15 6 August 1-15 52 16 6 1961 J u l y 16-31 2 2 0 August 1-15 3 1 1 August 16-31 10 U 3 I f these data are approximately c o r r e c t , we reach the conclusion that of a l l the summer young which matured i n I960 about 3$-h0% l o s t t h e i r e n t i r e l i t t e r s j ust a f t e r implantation, and i n 196l about 50-75$ suffered t o t a l l o s s of t h e i r l i t t e r s . Thus these data show a sharp contrast to the p a r t i a l p r e n a t a l l o s s data given above by suggesting a considerable increase i n t o t a l l i t t e r l o s s i n Lemmus summer young over the c y c l e . • To sum up our assessment of pr e n a t a l mortality: p a r t i a l p r e n a t a l m o r t a l i t y i n Lemmus and Dicrostonyx showed no r e l a t i o n s h i p to the c y c l i c density changes. T o t a l l i t t e r l o s s a f t e r implantation d i d not seem t o occur i n Dicrostonyx or i n adult Lemmus, but i n Lemmus summer young i t seemed to be high i n the peak summer and i n the summer of the decl i n e . T o t a l l i t t e r l o s s before implantation could not be assessed i n t h i s study. Post-Natal M o r t a l i t y (a) Adults: Adult m o r t a l i t y here includes a l l winter m o r t a l i t y as w e l l as 7k the summer m o r t a l i t y of winter generation animals. The s p e c i f i c conclusions made here apply to the l i v e trapping area i n p a r t i c u l a r and probably Type G declines i n general. Some general observations on adult m o r t a l i t y may be made from snap trapping records. Unfortunately snap trapping data cannot give v a l i d quantitative estimates of m o r t a l i t y rates but q u a l i t a t i v e observations may be made. There i s an annual overturn i n population. Adults of the winter generation, which comprise the ent i r e population at the s t a r t of the summer breeding, are gradually replaced through the summer by t h e i r own young, and by l a t e August and September there are very few o l d adults l e f t . This i s r e f l e c t e d i n the snap trapping samples as follows: % of winter generation adults i n snap trap samples Lemmus Dicrostonyx 1959-61 combined June 1-15 100 100 June 16-30 100 100 J u l y 1-15 79 93 J u l y 16-31 60 U2 August 1-15 23 35 August 16-31 11 17 September 1-15 2 8 Quantitative observations on adult m o r t a l i t y may be made from the l i v e trapping data. No m o r t a l i t y estimates were obtained i n 1959. Tables 32 and 33 give the minimum s u r v i v a l rates f o r Lemmus i n i960 and 1961, and Tables 3k and 35 f o r Dicrostonyx i n i960 and 1961. Minimum s u r v i v a l rates are obtained by marking a cohort of animals at time t and determining the number known to be a l i v e at time t + w (Chitty, 1952). These rates always underestimate the true s u r v i v a l rate and consequently care must be exercised i n i n t e r p r e t i n g them. To f a c i l i t a t e d i r e c t comparisons the observed minimum s u r v i v a l rates were converted l o g a r i t h m i c a l l y to a standard 28 day period, and these data are given i n Table 36 (Lemmus) and 37 (Dicrostonyx). 75 TABLE 32. Minimum s u r v i v a l r a t e estimates f o r Lemmus, summer I960. TIME PERIOD WINTER GENERATION SUMMER GENERATION N * N t Nt+1— M.S.R. N h t-&.» M.S.! Quadrat # 1 June 18 - 20 19 10 2 — 0.63 J u l y 6-8 16 5 3 8 2 0 0.50 0.25 J u l y 28-30 9 6 0 20 5 0 0.67 0.25 August 25-27 8 — 50 — — Quadrat # 2 June 29 - 1$ 8 0 1 0 1 J u l y 2 o.53 1.00 J u l y 20 - 23 1U 0 hS 17 0 0.36 0.38 August U - 6 6 mm - 69 - -•H- N = number released, N-fc » number known to be a l i v e next time, Kt+1-0B number included from l a t e r sampling, M.S.R. « minimum s u r v i v a l rate over the time perio d shown. 76 TABLE 33• Minimum s u r v i v a l rate estimates f o r Lemmus, summer 1961, Quadrat # 1, TIME PERIOD WINTER GENERATION SUMMER GENERATION N # \ Nt+l-oo M.S.R., N June 5 - 2 2: 0 mm 1,00 June 12 - U 1 2 0.75 June 19 - 5 0 mm 1.00 June 26 - u 3 1 1.00 J u l y 3 - 3 2 I -1.00 J u l y 10 - 3 0 1 1 0.33 J u l y 17 - 3 1 0 0.33 J u l y 2U - 2 0 0 -0.00 J u l y 31 - - mm August 7 - - 1 August Iii - - -August 21 - - -August 28 - _ _ Sept. 1 N A oo M.S.R. 0.00 0.00 * N » number released, = number known t o be a l i v e next time, N t + l - a f numb© 1 , included from l a t e r sampling, M.S.R. » minimum s u r v i v a l rate over the time p e r i o d shown. 77 TABLE 3u» Minimum s u r v i v a l r a t e estimates f o r Dicrostonyx, summer I960, Quadrat # 2. TIME PERIOD N * June 29 - 6 J u l y 2 J u l y 20 - 23 8 August U - 6 8 •* See Table 33 f o r explanation of symbols. WINTER GENERATION SUMMER GENERATION N t Nt +l-«> M ' S ' R - N N t ^t+l-00 M.S.R. k 1 0.83 U 0 5 2. 0 0.50 O./4O TABLE 35 • Minimum s u r v i v a l r a t e estimates f o r Dicrostonyx, summer 1961, Quadrat # 3 . TIME PERIOD WINTER GENERATION SUMMER GENERATION N * N t N t + l - c o M.S.R. N Nt Nt+l-oo M .S.: June 5 - 2 0 1 o.5o June 12 - 16 7 k 0.69 June 19 - Hi 6 5 -0.79 June 26 - 12 2 9 mm 0.92 J u l y 3 - 11 9 2 1.00 J u l y 10 - 11 5 3 -0.73 J u l y 17 - 8 k 2 0.75 J u l y 2li - 6 2 1 2 0 0 0.50 0.00 J u l y 31 - 3 2 0 5 1 2-i 0.67 0.60 August 7 - 2 0 0 k 2 2 0.00 1.00 August Ik - - 6 2 1 o.5o August 21 - mm 6 2 0 0.33 August 28 - mm 2 - » Sept. 1 # See Table 33 f o r explanation of symbols 79 TABLE 36. Minimum s u r v i v a l rates f o r Lemmus converted to a 28 day base. O r i g i n a l data i n ta b l e s 32 and 33. MINIMUM SURVIVAL RATE PER 28 DAIS WINTER GENERATION SUMMER GENERATION I960 Quadrat # 1 June 18-July 6 J u l y 6-July 28 J u l y 28-Aug. 25 Quadrat # 2 June 29-July 20 J u l y 20-Aug. 1; 1961 . Quadrat #1 June 5-20 June 21-July I4. J u l y 5-18 Ju l y 19-31 Aug. 1-31 0.U9 O.I4I 0.68 0.17 0.26 0.ii3 0.15 1.00 0.17 0.56 1.00 0.11 0.00 0.00 0.00 0.00 80 TABLE 37. Minimum s u r v i v a l rates f o r Dicrostonyx converted to a 28 day base. O r i g i n a l data i n t a b l e s 3k and 35 • MINIMUM SURVIVAL RATE PER 28 DAYS WINTER GENERATION SUMMER GENERATION I960 Quadrat # 2 June 29-July 20 J u l y 20-Aug. k 1961 Quadrat # 3 June 5-20 June 21-July k J u l y 5-18 J u l y 19-31 Aug. 1-15 Aug. 16-31 0.78 0.27 0.13 0.53 0.53 O.lU 0.00 0.18 0.00 0.36 0.03 81 Considering only the winter generation, we see f i r s t that over-a l l summer s u r v i v a l appears to have been better i n 1°60 than i n l°6l f o r both species. In l°6l a f t e r mid-July s u r v i v a l seems t o decrease moderately i n Dicrostonyx and considerably i n Lemmus, r e s u l t i n g i n a complete absence of adults be e a r l y to mid-August. These differences between i960 and 1961 seem to be r e a l , although i t i s impossible to estimate t h e i r magnitude from these data. Overwinter m o r t a l i t y cannot be estimated f o r 1959-60 because breeding was occurring, but we can obtain a block estimate f o r the I96O-61 winter because no breeding occurred. As was shown above i n discussing density changes, there was a 90-95% decrease i n Lemmus and a 70-80% decrease i n Dicrostonyx over the period from August I960 to June 1961. There was no breeding over t h i s period (the few animals born i n May are excluded from these estimates) and no major movements occurred. As an approximation we may enquire what mean monthly m o r t a l i t y rate would produce the observed declines over t h i s 10 month period with no recruitment or migration. For Dicrostonyx an 11-15% monthly m o r t a l i t y would produce a 70-80% decline over t h i s period, and f o r Lemmus a 20-25% monthly m o r t a l i t y would produce the observed 90-95% reduction. There i s some i n d i r e c t evidence that the winter m o r t a l i t y rate i n 1960-61 was not constant i n Lemmus but may have been so i n Dicrostonyx. The l o c a l Eskimos brought i n a l l lemmings they found during the winter, and these were recorded as "caught a l i v e " or "found dead". There was a sharp drop i n the number of l i v e Lemmus found by the Eskimos about December 15-31 and thereafter almost a l l specimens were found dead. This same change d i d not seem to occur i n Dicrostonyx. Figures obtained were as follows: Proportion of winter specimens caught a l i v e before December 31 a f t e r January 1 Lemmus 30 of 98 3 of 99 82 Proportion of winter specimens caught a l i v e before December 31 a f t e r January 1 Dicrostonyx 9 of 20 8 of 23 These data suggest a period of increased m o r t a l i t y f o r Lemmus sometime around December. We can introduce these data i n t o our model by adding one month with an increased m o r t a l i t y rate of 5>0$, a l l other months having a constant r a t e . Then a 15-20$ monthly m o r t a l i t y with one month increased to $0% p r e d i c t s a decline i n Lemmus s i m i l a r to that observed. The i n t e r e s t i n g t h i n g to note i s that the magnitude of t h i s increased m o r t a l i t y rate during one month has very l i t t l e e f f e c t on the f i n a l predicted decline; f o r example, 20$ monthly m o r t a l i t y = 89% decline over 10 mo. 20$ " " plus one month at 33% - 93$ " " " K 20$ ,r « " " " " 50$ - 93$ » " " " The reasons f o r t h i s apparent mid-winter sharp decline i n Lemmus are not known. The above hypothetical model suggests, however, that unless t h i s increased m o r t a l i t y extended over a considerable length of time or was exce p t i o n a l l y severe i t need have l i t t l e e f f e c t on spring d e n s i t i e s i n 196l. The conclusion to t h i s d i s c u s s i o n on winter m o r t a l i t y i n 1960-61 i s that although the decline i n both species over t h i s time i s very great numerically, the average monthly m o r t a l i t y rates which could produce the observed declines are reasonable f o r small mammals ( L e s l i e , C h i t t y , and Chitty,AGoLley, 196l)j indeed i f anything they seem to be low rather than high. In summary, m o r t a l i t y of the winter generation appeared s l i g h t l y higher i n the summer of 1961 than i n the summer of i960. Winter m o r t a l i t y rates during 1960-61 were moderate to low but produced a great numerical 83 decline because of the lack of breeding over t h i s 9-10 month period. Nothing i s known of summer m o r t a l i t y i n 1°5° or winter m o r t a l i t y i n 195>9-60. The data are not s u f f i c i e n t to investigate possible d i f f e r e n c e s i n m o r t a l i t y between the sexes, (b) Juveniles: Juvenile m o r t a l i t y r e f e r s to the m o r t a l i t y of summer young during the summer of b i r t h , and thus includes birth-weaning m o r t a l i t y and e a r l y post-weaning m o r t a l i t y . For a general idea of changes i n juvenile m o r t a l i t y we may return to Tables 32-37 f o r the summer generation data. These tables show f o r Lemmus that apparently no young survived on the l i v e trapping area i n 1961, while at l e a s t some survived i n I960 on the same area. For Dicrostonyx the i960 data are not very extensive, but i n I96I there was apparently no s u r v i v a l of young u n t i l a f t e r August 1 and even then s u r v i v a l was not very good ( f i r s t l i t t e r young should have been i n the traps by 1$ J u l y 1961). These data suggest that juvenile s u r v i v a l was poor i n both species during the summer of the d e c l i n e . A more r e f i n e d estimate of t h i s m o r t a l i t y may be made as follows. Knowing from the reproductive data given p r e v i o u s l y the mean timing of breeding periods and the mean l i t t e r s i z e , and knowing from l i v e trapping the number of adult females l i v i n g on the quadrat at the various times, we may estimate the number of young born on the quadrat f o r each breeding p e r i o d . At a subsequent trapping p e r i o d ( l a t e enough a f t e r weaning to ensure a l l the young being trappable) we get a t a l l y of how many of these young are a l i v e on the quadrat, and by comparing t h i s with the c a l c u l a t e d number born we can estimate the juvenile m o r t a l i t y rate d i r e c t l y . We assume i n t h i s a n a l y s i s ( l ) that the females breed at the average rates determined previously, (2) that a l l females have t h e i r l i t t e r s on or 81* adjacent to the quadrat, (3) that a l l the young on the quadrat have been caught, as w e l l as a l l the females, and (U) that there i s no net immigration or emigration of young. Assumptions ( l ) , (2), and (3) are probably v a l i d , and assumption (1+) could not be evaluated. These c a l c u l a t i o n s were done f o r both species i n I960 and 1961 and the r e s u l t s are presented i n Tables 38-ij.l. These s u r v i v a l estimates are a composite of birth-weaning m o r t a l i t y and a va r i a b l e length of e a r l y post-weaning m o r t a l i t y , and hence some caution must be exercised i n comparing the s u r v i v a l rates converted to the standard 28 day base. These data show very low s u r v i v a l rates of summer young i n the decli n e . There was some f u r t h e r suggestion that the second Dicrostonyx l i t t e r survived b e t t e r than the f i r s t l i t t e r . This suggestion i s confirmed i n the snap trapping data i n which the l a t e August samples of both species are dominated by I-,1 young with almost no young and only a few young, (since the breeding adults are dying out through t h e summer, one would expect to get many Y-^  young, fewer Y^' and very few Y^"). There i s no s t r i k i n g d i f f e r e n t i a l m o r t a l i t y between the sexes i n these data. Extensive snap trapping data support these r e s u l t s obtained from l i v e trapping and render improbable any suggestion that these di f f e r e n c e s between years are due to emigration of young from the l i v e trapping area. Birth-weaning m o r t a l i t y could not be separated from e a r l y post-weaning m o r t a l i t y i n these estimates. I f much l o s s occurred at b i r t h or s h o r t l y a f t e r , p a r t i c u l a r l y losses of whole l i t t e r s , t h i s should show up i n a regression of active mammary t i s s u e i n breeding females. However, there was no difference macroscopically between l a c t a t i n g females i n i960 and 1961. During the breeding season of both years v i r t u a l l y every female showed act i v e mammary t i s s u e , and there was no evidence that l a c t a t i o n had stopped i n any of the females such as occurs at the end of the breeding season. 85 TABLE 38, S u r v i v a l estimates f o r juvenile Lemmus on Quadrat # 1, summer I960. Mean Date at which weaning i s complete No* adult females a l i v e then Mean l i t t e r s i z e Calculated No. of young born Date of subsequent trapping No. of these j u v e n i l e s i n traps then Estimated s u r v i v a l rate from b i r t h t o trapping PERIOD OF BREEDING • I I I I I I J u l y 9 J u l y 28 August 18 7.37 51.5 25 8 7*11 56.9 20 k ad., 6 yg. 6.23 ad., 5.00 yg, 39*9 J u l y 28-30 Aug. 25-27 August 25-27 1U 0.U9 0.35 0.35 (per 3k days) (per k3 days) (per 22 days) Estimated s u r v i v a l rate converted to 28 day base 0.56 0.50 0.25 Pregnancy rate of young » o.50; continuous breeding assumed f o r a d u l t s . 86 TABLE 39. S u r v i v a l estimates f o r juvenile Lemmus on Quadrat # 1, summer 1961. PERIOD OF BREEDING I I I . I l l Mean date at which J u l y Ik August k August 27 weaning i s complete No. adult females 1 1 0 a l i v e then Mean l i t t e r s i z e 7.00 7.80 6.75 Calculated No. of 7.00 7.80 0 young born Date of subsequent J u l y 13-15 Aug. 10-12 Aug. 29-31 trapping No. of these j u v e n i l e s 2 1 1 i n traps then Estimated s u r v i v a l 0.29 0.13 rate from b i r t h t o (per I4 days) (per 21 days) trapping time Estimated s u r v i v a l 0.09 rate converted to a 28 day base 0.07 87 TABLE UO. S u r v i v a l estimates f o r ju v e n i l e Dicrostonyx on Quadrat # 2, summer I960. PERIOD OF BREEDING I I I Mean date at which J u l y 7 J u l y 29 weaning i s complete No. adult females 3 3 a l i v e then Mean l i t t e r size 6.11 5.00 Calculated No. of 18.3 15.0 young born Date of subsequent J u l y 20-23 trapping No. of these j u v e n i l e s 7 -i n traps then Estimated s u r v i v a l 0.38 ? rate from b i r t h to (per 28 days) trapping time Estimated s u r v i v a l 0.38 ? rate converted to a 28 day base 88 TABLE Ip.. S u r v i v a l estimates f o r ju v e n i l e Dicrostonyx on Quadrat # 3, summer 1961. Mean date at which weaning i s complete No. of adult females a l i v e then Mean l i t t e r s i z e Calculated No. of young born Date of subsequent trapping No. of these j u v e n i l e s i n traps then Estimated s u r v i v a l rate from b i r t h t o trapping time PERIOD OF BREEDING I I I J u l y 12 August 7 5.61 39.3 5.29 10.6 J u l y 27-29 Aug. 3 - 5 I I I August 27 8.00 0 Aug. 2U-26 0.05 0.38 ? (per 29 days) (per I4 days) Estimated s u r v i v a l rate converted t o 28 day base 0.05 0.I4 89 This i n d i r e c t evidence suggests that the l o s s of whole l i t t e r s at b i r t h or i n ea r l y suckling stages i s not the cause of the observed poor s u r v i v a l of j u v e n i l e s . More d i r e c t evidence on t h i s point i s needed. In summary, juvenile m o r t a l i t y between b i r t h and 1-k weeks a f t e r weaning was very high i n the summer of decline on the l i v e trapping area (Type G d e c l i n e ) . Almost no young of the f i r s t l i t t e r seemed to survive and only moderate numbers of the second and t h i r d l i t t e r s . This high m o r t a l i t y was probably not due to the l o s s of whole l i t t e r s at b i r t h or i n the e a r l y suckling stages, but probably occurred j u s t s h o r t l y before or sh o r t l y a f t e r weaning. SUMMARY WD CONCLUSIONS (1) P a r t i a l prenatal m o r t a l i t y showed no change i n e i t h e r species over the c y c l e , but the complete l o s s of l i t t e r s j u s t a f t e r implantation f o r summer young Lemmus seemed to be high i n the peak summer and i n the summer of d e c l i n e . (2) Adult m o r t a l i t y seemed to increase s l i g h t l y i n the summer of the decline i n both species, and winter m o r t a l i t y rates i n 1960-61 were low to moderate i n both species. Nothing i s known about m o r t a l i t y during 1959 or the 1959-60 winter. (3) Juvenile m o r t a l i t y between b i r t h and s h o r t l y a f t e r weaning was very high i n the summer of the decline p a r t i c u l a r l y f o r the f i r s t summer l i t t e r . This conclusion probably applies only to Type G declines- as w i l l be shown i n a l a t e r s e c t i o n . MOVEMENTS AND MIGRATIONS The t h i r d f a c t o r which can cause changes i n population density i s d i s p e r s a l . D i s p e r s a l may take the form of small l o c a l movements or mass movements ("migrations") of the whole population. On small areas d i s p e r s a l can a f f e c t density through immigration or emigration. On large areas immigration u s u a l l y balances emigration and consequently d i s p e r s a l a f f e c t s density only through reproduction or m o r t a l i t y changes. METHODS Almost a l l data on l o c a l movements were obtained by l i v e trapping, and these methods have been discussed i n a previous s e c t i o n . A few movements were obtained by snap trapping animals which had p r e v i o u s l y been l i v e trapped. RESULTS Local Movements The l i v e trapping program used i n t h i s study was not designed p r i m a r i l y to study movements, and consequently the data leave much to be desired. The many problems of measuring home ranges and movements of small mammals have been discussed by Davis (1953), S t i c k e l (195U), and Brown (1956). No attempt to estimate a c t u a l home range si z e s w i l l be made because very few animals were recaptured more than two or three times during any one trapping per i o d of three days; at l e a s t 5-7 recaptures are necessary f o r home range estimates. The appropriate method f o r the lemming data i s to analyze distances between successive captures (Brown, 1956) because t h i s allows us to use animals captured only twice during a trapping p e r i o d . This type of a n a l y s i s i s confined to short term movements wit h i n trapping periods. Tables J+2 and i+3 give the length of every movement recorded w i t h i n trapping periods f o r Lemmus i n I960 and 1961, and Tables UU and U5 give the same information f o r Dicrostonyx. Differences between the years were tested by chi-square (Snedecor, 1956) and both species showed a 91 TABLE U2. Length of every movement recorded w i t h i n periods of l i v e trapping f o r Lemmus, summer I960, Quadrat # 2 . LENGTH OF MOVEMENT WINTER GENERATION Y1 SUMMER YOUNG MALES FEMALES MALES FEMALES < 5o« 8 17 15 12 51-100' 6 13 7 10 101-150' 5 10 U 7 151-200' 3 9 3 6 201-300! 3 1 3 3 301-U00I 5 2 0 2 U01-500! 0 1 1 0 501-700! mm - - -701-900! - - - -N = 30 N o 53 N - 33 N = UO x « Ui3.3' x - io5.l« X - 9U.6' x « 111, 92 TABLE U3. Length of every movement recorded w i t h i n periods of l i v e trapping f o r Lemmus, summer 1961, Quadrat # 1, and v i c i n i t y . LENGTH OF MOVEMENT MALES ^5o« 1 51-100' 2 101-150' 6 151-200' 2 201-300! 3 301-U00! 0 U01-500! 1 501-700' 1 700-900! 1 N • 17 WINTER GENERATION FEMALES I x SUMMER GENERATION MALES FEMALES x - 229.3' 9? TABLE Ui* Length of every movement recorded w i t h i n periods of l i v e trapping f o r Dicrostonyx, summer I960, Quadrats # 2 and # 3« LENGTH OF WINTER GENERATION Y x SUMMER YOUNG MOVEMENT MALES FEMALES MALES FEMALES ^ 50« 15 8 2 0 51-100' 6 3 1 1 101-150' 1 1 1 -151-200! 1 2 - -201-300' 2 2 301-UOO! 2 0 U01-500! - -501-600! - -N = 27 N • 16 N - U N = 1 x - 86 x - 8U x - 62 2 - 71 9 4 TABLE U5« Length of every movement recorded w i t h i n periods of l i v e trapping f o r Dicrostonyx 3 summer 1961, Quadrat # 3 and v i c i n i t y . LENGTH OF MOVEMENT WINTER GENERATION MALES FEMALES Y-L SUMMER YOUNG MALES FEMALES ^ 50' 51-100' ioi-i5o« 151-200! 201-300.' 301-UOO! U01-J00! 501-600! 1 1 5 0 1 0 1 1 N - 10 1 6 2 3 1 0 1 N - 13 N - 1 N = 0 x = 201 x - 112 71 95 s i g n i f i c a n t l y greater number of long movements recorded i n 1961 than i n 1960 (Lemmus winter generation males, P ^.05, >.025; Dicrostonyx winter males, P<.005; and Dicrostonyx winter females, P < .05,>.025). The suggestion i s that the low density of 1961 was accompanied by a greater m o b i l i t y of the adults at l e a s t , compared to i960. No data on movements were obtained i n 1959* Although these r e s u l t s are reasonable, there are several reservations which render t h e i r s i g n i f i c a n c e somewhat questionable. The primary d i f f i c u l t y i s that the spacing of the traps was not i d e n t i c a l i n the two years. In 1961, i n p a r t i c u l a r , l i v e traps were scattered at i r r e g u l a r i n t e r v a l s outside the quadrat boundaries, and t h i s increased the p r o b a b i l i t y of detecting longer movements. Furthermore, many of the i960 data come from Quadrat # 2 and t h i s area had so few lemmings i n 1961 that i t was not trapped. Observed range lengths ( S t i c k e l , 195U) could be estimated f o r only a few Lemmus winter generation males with the f o l l o w i n g r e s u l t s : 1960 N - U x - 286 f e e t ± 63 f e e t ( l SE) 1961 N = 3 x = 631 f e e t +170 f e e t ( l SE) These data conform to the suggestion of greater m o b i l i t y i n the summer of 1961 made above, but again reservations must be made about t h e i r s i g n i f i c a n c e . Observed range lengths could not be estimated f o r any other group except t h i s because only animals having f i v e or more recaptures within one trapping peri o d can be used. Whether lemmings occupy a d e f i n i t e t e r r i t o r y or home range i s not known. The general impression I have gathered from l i v e trapping i s that the males of both species are wide ranging and almost c o n t i n u a l l y on the move. Untagged adult males c o n t i n u a l l y appeared on the l i v e trapping areas through the summer. This e f f e c t was p a r t i c u l a r l y s t r i k i n g i n i960 on 96 the Lemmus quadrat (see Table 7) where l/l+ to l/3 of the adults were inadvertantly k i l l e d each trapping period, and yet the adult population on the quadrat through the summer declined at a very low r a t e . Net immigration almost completely o f f s e t the a r t i f i c i a l m o r t a l i t y . This same observation applies to a l e s s e r degree to the summer young males and females. The adult females of both species seem to move around l e s s than the males, but even so they range over r a t h e r large areas. Thus any complete study of movements under these conditions must involve very large l i v e trapping areas, p o s s i b l y as b i g as 15-20 acres, i n order to be c e r t a i n of recording most of an i n d i v i d u a l ' s movements. Movements of i n d i v i d u a l s from one week or month to the next during the summer w i l l not be analyzed i n d e t a i l because the data are too fragmentary. A few examples w i l l be given to indicate the sort of movements that can occur. Distance between capture Dates of capture points (feet) I960 Lemmus adult male 3700 June 2-July 8 Lemmus Y, female 2600 J u l y 7-28 Lemmus Y£ male 525 Jul y 23-August 1+ 1960-61 (the following are i960 summer young recaptured a l i v e as adults i n June 1961) Lemmus female 21+00 Aug. i960 - June 196l Lemmus female 365 u 1 1 Lemmus female 2500 " u Dicrostonyx female 165 " M The s i g n i f i c a n c e of these movements i s simply not known. On the one hand, they may be extremely abnormal samples biased toward long movements; on the other hand, they may represent the normal sort of movements which go on i n these populations. I am i n c l i n e d to believe more i n the l a t t e r a l t e r n a t i v e a f t e r having seen movements of 500-800 f e e t take place i n l e s s than 21+ hours wit h i n a trapping period (see Tables 1+3 and 1+5)• 97 Migrations Perhaps the one thing most people know about lemmings i s that p e r i o d i c a l l y they a l l march down to the sea and drown themselves. Obviously i f t h i s i s true i t must have a profound e f f e c t on the population dynamics of the lemmings. Local movements of i n d i v i d u a l animals can be very pronounced at c e r t a i n times of the year. At Baker Lake i n the spring of I960 lemmings began to appear i n p a r t i c u l a r areas as the melt-off proceeded, as each l o c a l center of density began to be af f e c t e d by the snow melting. Individuals and 'groups' of Lemmus were reported on the lake i c e i n f r o n t of the settlement on May 26, and the major a c t i v i t y occurred during the night hours ( t w i l i g h t a l l night.at t h i s time of ye a r ) . From 2 AM to U AM on June 2 I observed 2$ lemmings moving i n d i v i d u a l l y on the ,lake:ice i n f r o n t of the settlement. F i f t e e n of these were caught and tagged ( l Dicrostonyx male; 7 Lemmus males; 7 Lemmus females), and a l l were i n breeding condition. None of these animals seemed to do anything on the i c e except move i n a s t r a i g h t l i n e , u s u a l l y toward the nearest land, running at top speed. A l l were very aggressive when caught. I t was not poss i b l e to determine whether the lemmings on the ice came from the opposite side of the lake (3-5 miles) or whether they had moved out from the area of the settlement onto the ic e and then l a t e r moved back again. One of the Lemmus males tagged was l a t e r recovered on the l i v e trapping area f i v e weeks l a t e r a f t e r having moved 3700 f e e t (see above). Most of t h i s movement on the i c e was over by June u, having l a s t e d about 9 days. I never saw any evidence of group movements on the i c e , and never saw even two lemmings moving together. An Eskimo brought i n a bucketful of 70 Lemmus which he k i l l e d on the i c e during the night of May 26, but whether these represented a r e a l group or merely a l o t of i n d i v i d u a l s could not be determined. Very few dead lemmings were found on the i c e . 98 Apparently these spring movements are not common at Baker Lake, Mr. S. Lunan, who was manager of the Hudson Bay Company post at Baker Lake f o r about 30 years ( u n t i l 193>7) t o l d me that only once had he seen lemmings so abundant that they were common on the i c e i n the spring. Many other areas around Baker Lake reported movements of lemmings on the ice i n the spring of I960: C h e s t e r f i e l d I n l e t , Rankin I n l e t , Eskimo Point, Aberdeen Lake, and Schultz Lake. These spring movements are thus quite common i n p a r t i c u l a r years of higher than average p e a k ' d e n s i t i e s . Many of the people l i v i n g i n the North, even the Eskimos, r a r e l y see a l i v e lemming. Thus when spring movements do occur, there i s a tendency to exaggerate t h e i r s i z e . A few tens of lemmings q u i c k l y become a few hundreds i n the mind, and to the next person the number i s i n the thousands. Another l o c a l movement of brown lemmings was reported i n l a t e August I960 by an Eskimo at the east end of Baker Lake. The r e l i a b i l i t y of the observations could not be e s t a b l i s h e d . There are no other records of f a l l movements from the area. No other "migrations" were observed during e i t h e r 1959 or 1961 i n the area. The general conclusion regarding these "migrations" of lemmings i s that they assume a mental status disproportionate to t h e i r b i o l o g i c a l s i g n i f i c a n c e f o r the lemmings. The a c t u a l events are f a r l e s s s t r i k i n g than the legend, and not a l l peak populations even show these events. SUMMARY JND CONCLUSIONS (1) Whether lemmings occupy a d e f i n i t e t e r r i t o r y or home range i s not know. (2) There i s a suggestion that the average distance moved between traps was greater during the decline of 1961 compared to the peak summer of I960 i n adults of both species. 99 (3) I n d i v i d u a l animals can move over large distances; instances of animals having moved 800 f e e t i n l e s s than one day have been recorded. (U) True group movements ("migrations") are very r a r e l y i f ever recorded, but the spring melt-off during the peak year may be accompanied by considerable l o c a l movement of i n d i v i d u a l s . (0) There i s no evidence that "migrations" or even these spring movements are a necessary part of the cycle i n numbers. CHANGES IN EXTRINSIC FACTORS Factors which a f f e c t reproduction and m o r t a l i t y may be broadly c l a s s i f i e d as i n t r i n s i c or e x t r i n s i c f a c t o r s . E x t r i n s i c f a c t o r s include weather, predators, disease, p a r a s i t e s , and food. These f a c t o r s are normally studied as d i s t i n c t and independent v a r i a b l e s which exert an e f f e c t on the population from the outside. They thus represent the f i r s t and simplest l e v e l of enquiry i n t o the causes of population density changes, and we must enquire whether e x t r i n s i c f a c t o r s can adequately explain the observed density changes of lemmings. Weather The winter of 1959-60, when the lemmings increased, began with a dry freeze-up and a quick buildup of snow cover. The winter of 1960-61, when they declined, began with a wet freeze-up and a slow buildup of snow cover u n t i l December. However, because of the d r i f t i n g of the snow and the tendency of lemmings to seek out the more deeply d r i f t e d areas, probably there were some areas i n 1960-61 that were as favorable weather-wise as areas i n 1959-60. Yet no. winter breeding was found i n 1960-61, which suggests that bad winter weather was not s u f f i c i e n t to cause the observed absence of breeding. One of the most s t r i k i n g f a c t s about t h i s 1959-61 lemming cycle was i t s synchrony over a very large area of the c e n t r a l Canadian a r c t i c . This does not appear t o be a simple coincidence. I t was impossible to f i n d a population around Baker Lake which was not at a peak i n i960. I f t h i s synchrony i s more than a mere coincidence, the agent acting over these large areas would most l i k e l y be weather. We do not know i f good winter weather i s the only thing needed f o r an increase i n numbers, or whether some other f a c t o r must also be present. 101 Summer weather seemed to bear no r e l a t i o n s h i p to the c y c l e . The summer of 1959 was very wet and co l d and yet the population was beginning to increase l o c a l l y . The summers of I960 and l°6l were both warm and dry and yet i n i960 the population remained at a peak whereas i n 1961 i t declined. There are a s u f f i c i e n t number of c l i m a t i c v a r i a b l e s that i f we inv e s t i g a t e enough of them we s h a l l s urely f i n d one or more close c o r r e l a t i o n s with t h i s lemming c y c l e . Because of t h i s post hoc c l i m a t i c c o r r e l a t i o n s must always be suspect. Only by r e p l i c a t i n g and d i v e r s i f y i n g our observations on the a s s o c i a t i o n between types of weather and c y c l i c changes can we hope to obtain a b e t t e r i d e a of i t s role u n t i l experimental work can be done. To sum up, favorable deviations from the average winter weather were associated with a large increase i n density, and unfavorable deviations were associated with a decline i n numbers. Summer weather seemed of l i t t l e importance. Predators Avian predators were not very numerous near Baker Lake compared with the numbers reported f o r northern Alaska ( P i t e l k a , Tomich, and T r e i c h e l , 1955)• Only three l o n g - t a i l e d jaegers (Stercorarius longicaudus), three p a r a s i t i c jaegers (S. p a r a s i t i c u s ) , one rough-legged hawk (Buteo lagopus), and one short-eared owl (Asio flammeus) were seen i n 1959» In I960 the f i r s t jaeger appeared on June 7, and b i r d s of prey were s t i l l very scarce during t h i s summer i n spite of the dense lemming populations. No attempt was made to census these b i r d s i n I960. Three p a r a s i t i c jaeger nests were found on the Main Study Area i n I960, and t h i s seemed to represent most i f not a l l of the jaegers nesting on t h i s area. In 1961 avian predators were again scarce. Only one snowy owl, two l o n g - t a i l e d jaegers, and two p a r a s i t i c jaegers were seen on the Main Study Areaj no nests were found. Long-tailed jaegers were much more common during a l l three years on the i s l a n d s i n Baker Lake which support 102 considerable numbers of nesting b i r d s . Only one mammalian predator was at a l l abundant on the Main Study Area — the weasel or ermine (Mustela erminea). Other l a r g e r predators, such as the a r c t i c fox (Alopex lagopus), wolves (Canis lupus), and wolverine (Gulo l u s c u s ) , were v i r t u a l l y absent. Weasels were very scarce i n 195>9 and none was seen; they were s t i l l uncommon i n I960 and only two were caught by the Eskimos. In 1961 weasels were very numerous. One was caught by an Eskimo on 1 February 1961, another on May 17, another on June 29, and from August 7 on weasels were seen everywhere. Over 70 specimens were caught by the end of August and many more i n e a r l y September. Complete autopsies were done on 22 of the August specimens. Of these 21 were males (286—3Ul mm t o t a l length) and only one was a female (261 mm t o t a l length). None was breeding, and almost a l l were moderately f a t . Stomach contents were classed as follows: empty, 7', b i r d feathers and bones, 3j lemming f u r and bones, 2j f i s h ( ? ) , 3', caribou meat (?), hi b e r r i e s and plant matter, 2j u n i d e n t i f i a b l e matter, 1. I t i s c l e a r that not a l l these weasels could have l i v e d on lemmings during the e a r l y summer because of the very sparse lemming population. The date at which weasels began to appear commonly (August 7) coincided with the time when a l l the young b i r d s were f i n a l l y able t o f l y , and t h i s suggests that the weasels may have fed on b i r d s during much of the summer. We must now see whether these weasels could have been responsible f o r the m o r t a l i t y changes of the lemming population. I t seems doubtful whether weasels were having an important e f f e c t on the lemming population of the l i v e trapping area f o r three reasons: ( l ) no weasels were caught i n the l i v e traps u n t i l August k and s i x weasels were caught i n these traps during the r e s t of August. I f weasels were pursueing lemmings on t h i s area during June and J u l y i t seems inconceivable that one or more of them would not have been caught, since the area was covered with l i v e t r a p s . (2) There i s no evidence of high 103 death rates i n the adults during June and July such as would be expected i f weasel predation was common. (3) The survival of the second l i t t e r of summer young (August) was relatively better than that of the f i r s t l i t t e r , even though the weasels should have exerted more predation pressure on this second l i t t e r . There i s thus no evidence that weasel predation did account for the observed mortality changes. Disease and Parasites No detailed studies on disease or parasites were made in this research program, but i n the course of autopsying some 2500-3000 lemmings only eight specimens have been found with any gross abnormalities such as cysts i n the l i v e r and spleen. There was no macroscopic evidence that most of the animals were not healthy. Parasite loads were superficially quite low and there was no evidence of debilitation even i n the few specimens with considerable numbers of stomach and intestinal parasites. About 50 Dicrostonyx were shipped to Toronto and Ottawa i n August I960. Most of these specimens died either on route or just after arrival in spite of rapid transport and apparently adequate food and bedding (Fisher, pers. comm.j Manning and Macpherson, pers. comm.). The question arises whether these animals died because of a latent disease which could be responsible for the decline. There i s no f i e l d evidence to support this view. Certainly there was no spectacular mortality i n either Dicrostonyx or Lemmus during August, September, or October I960. As we have seen previously, the winter mortality in Dicrostonyx over 1960-61 was not excessive for a population i n which no recruitment was occurring. We seem to have the alternative of ascribing most of this winter decline to an epidemic and assuming a l l other mortality factors to be almost negligible, or of placing disease on a par with many other mortality factors which comprise the winter mortality. ioh Furthermore, even i f we could ascribe a l l this winter mortality to disease, we would be l e f t without an explanation for most of the observed changes In reproduction or mortality described previously. There i s thus l i t t l e evidence that disease or parasites can adequately explain the cyclic events. Food Forage production was assessed by clipping the standing crop of green vegetation at the end of each summer (September 1-10) on 15 pairs of Quadrats, one of which was open and the other enclosed. Each open quadrat was paired as closely as possible with an enclosed quadrat to reduce sampling variation. A l l clipped vegetation was dried i n an oven at 225°F to constant weight and a l l weights given here are dry weights. The quadrats were 2 sq. meters in size and one-fourth of this t o t a l was clipped each year. Ten pairs of quadrats were set out in 1959 and the other five i n I960. Each enclosed quadrat was surrounded by 3/8" hardware cloth screening which was buried 8-12tt in the ground and extended 2l[-28u above ground. There was no evidence that any lemmings got inside any of these enclosed quadrats during the period of study. This general approach was the same as that of Thompson (1955 b). Table I4.6 gives the standing crop measurements at the end of the 1959, I960, and 1961 growing seasons on the Main Study Area. These data may be considered in two parts. Quadrats # 1-10 were present during a l l three years; quadrats # 11-15 were installed in i960 and serve as a further check on the 1960-61 changes. In the analysis of these data we are interested in the differences between the pairs of open and enclosed quadrats. Because there were very few lemmings in 1959 we may adopt the 1959 data as our base and relate a l l changes to i t . Two major effects cause deviations from this base — weather effects and lemming effects — , and the problem i s to separate these. This was done in the following way. * * 1 am indebted to Dr. Monte Lloyd, Bureau of Animal Population, for this s t a t i s t i c a l technique. 105 TABLE 1+6. Standing forage i n grams per 0.5 sq. meter dry weight at the end of summer. QUADRAT 1959 I 960 1961 ENCLOSED OPEN ENCLOSED OPEN ENCLOSED OPEN 1 H+.6 20.0 19.5 27.9 1+1+.7 57.8 2 118.3 70.0 11*9.8 90.5 157.2 105.6 3 36.7 29.9 51.2 37.1 72.8 1+9.1 U 1+8.9 31.9 58.8 36.5 100.8 68.3 5 28.7 26.0 50.9 31.1 72.3 1+2.3 6 28.1 27.0 31.5 1*3.8 36.2 35.1 7 2U.7 1*3.0 35.1 1+2.0 32.8 8 37.1 1*7.6 57.0 53.7 79. h 1+9.5 9 71+.5 73.2 115.8 86.2 109 .3 90.1? 10 1+6.5 37.3 57.1 1*0.9 51+.9 53.5 11 62.2 55.6 75.1 67.1+ 12 1*7.1 53.2 97.6 83.1 13 91.2 101.9 101+.3 95.5 Hi 57.1 63.7 86.1+ 79.9 15 75.3 89.5 93.6 93.2 xl-10 1+5.8 38.8 63.5 1+8.3 77.0 58.5 "xll-l5 66.6 72.8 91.U 83.8 106 The d i f f e r e n c e between each I960 enclosed quadrat and the same quadrat i n 19^9 must be caused only by weather d i f f e r e n c e s . S i m i l a r l y , the difference between each I960 open quadrat and the same quadrat i n 1959 must be caused by the i n t e r a c t i o n of lemming and weather e f f e c t s . But since we know the weather e f f e c t s alone from the enclosed quadrats, we may subtract t h i s element to estimate the lemming e f f e c t s (we assume these two e f f e c t s to be independent and a d d i t i v e ) . We can apply the t - t e s t to these di f f e r e n c e s and thereby t e s t the s i g n i f i c a n c e of these e f f e c t s . The same procedure may be applied to the 1961 data. The weather e f f e c t s are s i g n i f i c a n t between a l l three years (P <»0l), the progression i n the size of the standing crop being 1959 < I960 <196l# Thus i n terms of the quantity of food, more was a v a i l a b l e at the end of the summer of decline than e i t h e r the summer of increase or the peak summer. The r e l a t i v e changes i n standing forage were: 1959 - 100j 1960 - 1391 1961 - 168. Lemmings s i g n i f i c a n t l y depressed the standing crop i n both i960 and 1961 (P <.05, >.01). There i s no d i f f e r e n c e i n the lemming e f f e c t on quadrats # 1-10 between i960 and 1961, and the lemming e f f e c t shows up i n I96I on the new quadrats # 11-15 as would be expected ( P < . 0 l ) . The depressing e f f e c t of the leirmrings on the forage i s very nearly the same i n I960 and 1961 on both sets of quadrats. I f we set t h e o r e t i c a l standing crop at what would occur i n the absence of lemmings, the lemmings are found to depress standing crop by lu.5% i n i960 and 16.k% i n 1961 on quadrats # 1-10, and by lk»l% i n 1961 on Quadrats # 11-15. The l i m i t a t i o n s of these quadrat data must be stressed. These quadrats are not a random sample of the whole area. They are put almost i n v a r i a b l y i n sedge marsh type, i n the greenest, densest vegetation where one might expect high u t i l i z a t i o n from p r i o r knowledge. As such they are not 1 0 7 even a random sample of sedge marsh, and thus the conclusions from such data can s t r i c t l y be applied only to the area on which the ac t u a l quadrats occur. A f u r t h e r d i f f i c u l t y a r ises from the enclosures' subsequent a l t e r i n g of the microclimate of the quadrat. We must assume that these microclimatic changes are n e g l i g i b l e , but t h i s may not be tr u e . These d i f f i c u l t i e s i n i n t e r p r e t i n g quadrat data do not appear to have been appreciated by Thompson (19$5 b ) . I f we locate quadrats i n the best habitats where maximum u t i l i z a t i o n i s expected, we should not be surprised to f i n d high u t i l i z a t i o n and depressed forage production. However, while t h i s does give us an estimate of maximal e f f e c t s , i t t e l l s us very l i t t l e about the r e l a t i o n s h i p of lemmings to t h e i r food supply i n general. Forage u t i l i z a t i o n was estimated i n the spring of 1 9 6 1 by systematic sampling along l i n e t r a n s e c t s . A3' by 1 ' rectangle was dropped every ten fee t along these transects u n t i l the l i n e s ran out of the wetter habi t a t s . The habitat was c l a s s i f i e d at each s t a t i o n . A l l the cut grass and moss was removed from the 3 sq. f e e t and a v i s u a l estimate was made of the proportion of the forage that had been eaten. Transects were done only i n the wetter habitats and u t i l i z a t i o n was estimated separately f o r sedges, mosses, and heaths on each p l o t . No transects were done i n the dry habitats because u t i l i z a t i o n was so low as to be unmeasurable with t h i s technique. A l l these estimates were made before the new season's growth of plants had begun, i . e . when the quantity and q u a l i t y of the food supply was at i t s lowest point f o r the year. Table i i 7 gives these forage u t i l i z a t i o n estimates made i n the spring of the d e c l i n e . No transects were made i n e i t h e r 19$9 or i 9 6 0 because t o t a l u t i l i z a t i o n was too small to be conveniently measured. These data f o r 1 9 6 1 show average u t i l i z a t i o n of 30$ at the most f o r the wetter h a b i t a t s . I t was rather d i f f i c u l t t o estimate the moss u t i l i z a t i o n but t h i s 108 TABIE I4.7. Estimate of percentage forage u t i l i z a t i o n i n the spring of the decline, June 1961, Figures represent a t o t a l of seven d i f f e r e n t l i n e t r a n s e c t s . HABITAT TYPE No. % WITH % WITH % ESTIMATED UTILIZATION QUADRATS RUNWAYS WINTER CUTTINGS SEDGES MOSS HEATH Heath-sedge 52 65.3 55.8 15.2 7.2 U.1 hummock Sedge hummock 171 86.5 67.8 25.3 29.3 2.1 Sedge marsh 53 39.6 26.U 6.0 5.0 -109 was attempted because moss i s a very important food item during the winter. The pattern of forage u t i l i z a t i o n was very spotty i n 1961. Small areas 2-6 f e e t i n diameter would be completely devastated of a l l l i v e plants down to the roots, and these areas were surrounded by untouched vegetation. There was no evidence that the boundaries of these small feeding places coincided with packed snow, i c e , or any vegetation or topographical changes. In no case d i d these devastated areas coalesce over large areas; no place was more than k-S f e e t from r e l a t i v e l y untouched vegetation. I t i s d i f f i c u l t to see how food supply could be short under these conditions. The dry tundra areas were hardly u t i l i z e d at a l l during the winter of 1960-61. Small l o c a l areas were devastated but on the whole u t i l i z a t i o n must have been l e s s than 5%, Since the plants of the dry tundras grow very slowly (heath r e c o l o n i z a t i o n may take 50 years or more), any widespread destruction of t h i s vegetation would be evident f o r decades afterward. The same point may be made about dwarf b i r c h and willows. There was very l i t t l e g i r d l i n g of these shrubs during 1960-61 e i t h e r on the Main Study Area or on the outlying areas where they are more common. A l l the previous points have been concerned with food quantity. Food q u a l i t y may also be important. No attempt was made to analyze the q u a l i t y of the food i n t h i s study. There was no evidence of any obvious d e f i c i e n c y diseases such as occur i n domestic animals having vitamin or mineral shortages (Maynard and L o o s l i , 1956). With a l l the d i f f i c u l t i e s involved i n measuring forage changes d i r e c t l y , i t seems ea s i e r to turn the problem upside down and to look at the animal as a measure of the adequacy of the food supply (Bandy et a l . , 1956). I have used a f a t index to measure t h i s , and these data w i l l be presented i n the next section. Thus there was no evidence of a quantitative shortage of food over 110 t h i s lemming c y c l e . Nor was there any obvious evidence of a d e f i c i e n c y disease associated with changes i n the q u a l i t y of the food. SUMMARY AND CONCLUSIONS (1) Favorable winter weather was associated with the increase i n numbers, and unfavorable winter weather was associated with the d e c l i n e . Summer weather showed no c o r r e l a t i o n with density changes. (2) Avian predators were uncommon throughout the cycle and d i d not appear to p l a y a necessary r o l e i n i t . The weasel or ermine was the only important mammalian predator and these were not common u n t i l 1961, e s p e c i a l l y during August and September. However, there was no evidence that weasel predation accounted f o r the mo r t a l i t y changes observed i n the lemmings. (3) There was no evidence that disease or p a r a s i t e s played any necessary r o l e i n t h i s c y c l e . (1+) Lemmings s i g n i f i c a n t l y reduced the standing crop of forage i n both I960 and 1961 by about 1$%, Forage u t i l i z a t i o n averaged 30% or l e s s i n the wetter habitats j u s t a f t e r the winter of 1960-61, and the dry habitats were scarc e l y touched. There was no evidence of quantitative food shortage, nor any c l e a r suggestion of a d e f i c i e n c y i n food q u a l i t y over the cyc l e . CHANGES IN INTRINSIC FACTORS Changes i n reproduction and mo r t a l i t y may also r e s u l t from changes i n f a c t o r s i n t r i n s i c to the population, as w e l l as i n the e x t r i n s i c f a c t o r s j u s t discussed. Other animals of the same kind may produce behavioral and p h y s i o l o g i c a l changes i n the i n d i v i d u a l organism. The i n t r i n s i c f a c t o r s are behavior and physiology; these may be studied d i r e c t l y i n themselves or i n d i r e c t l y by t h e i r e f f e c t s . In t h i s section we s h a l l analyze some changes which occur over the cycle i n the following properties of i n d i v i d u a l s : weight d i s t r i b u t i o n s and. mean body weights; organ weights; f a t index; and s o c i a l r e l a t i o n s h i p s. METHODS Age Determination Many d i f f e r e n t techniques f o r measuring chronological age have been proposed, but the majority of small mammal workers s t i l l use body weight as a c r i t e r i o n of age (e.g. C h i t t y , 1952; Hoffmann, 1958). Frank and Zimmermann (1957) found that the body weight - age r e l a t i o n s h i p i n Microtus  a r v a l i s was g r e a t l y a f f e c t e d by both inherent v a r i a b i l i t y and seasonal changes i n growth. Body weight i s more a c r i t e r i o n of p h y s i o l o g i c a l age than chronological age, and as such i t i s more u s e f u l f o r our purposes than chronological age would be. An attempt was made to use the lens of the eye as an age i n d i c a t o r (Lord, 1959) i n t h i s study but ana l y s i s showed that lens weight was normally pr o p o r t i o n a l to body weight. Whatever caused the body weight to change a l s o caused the lens weight to change, and so no a d d i t i o n a l information accrued from weighing lenses (one exception to t h i s i s discussed below). Body weights were used rather than t o t a l lengths because there i s much l e s s v a r i a b i l i t y both within and between workers when using measurements than when using t o t a l length measurements. 112 Mean Body Weights Figure 6 gives a generalized chronology and c l a s s i f i c a t i o n of the l i t t e r s and generations of both species of lemmings and i s necessary f o r the dis c u s s i o n that follows. This basic pattern v a r i e s s l i g h t l y i n the d i f f e r e n t phases of the c y c l e . The winter generation consists e i t h e r of overwintered animals (1959 ? and lp6l) or of animals born during the f a l l and winter (i960). The spring generation appears each year before the snow melts but i s not very large numerically; i t e s s e n t i a l l y behaves l i k e the winter generation during the summer. The summer l i t t e r s f o llow i n rapid sequence; i t i s probable that some adult females produce only two l i t t e r s and others four l i t t e r s , and t h i s may vary with the c y c l i c phase, but the general pattern i s about three summer l i t t e r s per adult female. By f a l l only summer born young are l e f t and these form next year's winter generation. Summer young females may breed i n t h e i r f i r s t summer and add a f u r t h e r generation to the f a l l population, but t h i s complication has been l e f t out of t h i s diagram. In computing mean body weights we would l i k e to follow d i s c r e t e generations so that the r e s u l t i n g means have a c l e a r b i o l o g i c a l s i g n i f i c a n c e , rather than being a mere s t a t i s t i c a l c o l l e c t i o n of data from diverse groups of animals. There i s no problem i n separating summer-born animals from winter or spring animals, but the d i f f i c u l t y a r i s e s i n t r y i n g to keep the spring generation (born April-May) separate from the winter generation. In Lemmus t h i s d i f f i c u l t y arose only f o r the May 16-31 and June 1-15, I960 samples. Since breeding occurred throughout t h i s winter i t was somewhat a r b i t r a r y to d i s t i n g u i s h a winter generation and a spring generation, but t h i s was done f o r the above two samples on the b a s i s of breeding vs. non-breeding animals, the breeding animals being r e f e r r e d to as the winter generation. These spring animals i n Lemmus are absorbed i n t o the r e s t of the winter generation adults by the end of June and cannot be recognized as a d i s t i n c t element of the samples a f t e r then. 115 FIGURE 6. Generalized annual chronology of generations and l i t t e r s f o r Lemmus and Dicrostonyx. T — I — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — r i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — r JAN. MAR. MAY JUL. SEPT. NOV. JAN. MAR. MAY Ilk In Dicrostonyx the problem i s much more d i f f i c u l t . The spring generation appears i n a l l three years and p e r s i s t s as. a d i s t i n c t e n t i t y even i n t o August. Data on body weight, t o t a l length, lens weight, and reproductive condition were u t i l i z e d i n t r y i n g t o separate the winter from the spring born animals. In 1961 the two groups were e a s i e s t to d i s t i n g u i s h because although the body weights overlapped there was a gap i n the lens weights between the winter generation (born i n the summer of I960) and the spring generation (born April-May 1961). For example, i n the June 1-15 sample winter animals had lens weights over 6.0 mg while spring animals had lens weights of l e s s than l+.O mg. In i960 gaps i n the body weight d i s t r i b u t i o n s were u t i l i z e d as break p o i n t s . In 1959 gaps i n body weight and t o t a l length d i s t r i b u t i o n s were mainly u t i l i z e d f o r separating these groups. While there i s a considerable subjective element involved i n these separations ( p a r t i c u l a r l y f o r 1959) I believe the r e s u l t s are b i o l o g i c a l l y more meaningful than they would be i f these two groups were mixed. Organ Weights A l l organs were preserved i n 10% n e u t r a l formalin and weighed i n the winter a f t e r c o l l e c t i o n e i t h e r on an e l e c t r i c balance or on a t o r s i o n balance. Organs weighing more than 200 mg were weighed to the nearest 5 mgj organs weighing l e s s than 200 mg were u s u a l l y weighed to the nearest 0.1 mg. A l l organs were cleaned of surface f a t under a binocular microscope and r o l l e d dry on f i l t e r paper before weighing. Repeated weighings i n d i c a t e d an accuracy of ± 3% i n normal weighings. Some of the l a r g e r organs x^ere weighed f r e s h i n the f i e l d during 1961. The expression of organ weights normally used by p h y s i o l o g i s t s as w e l l as most e c o l o g i s t s i s that of milligrams of organ weight per gram of body weight. However, few workers have heeded the advice of Chester Jones (1957, p 6-7) that such figures'may be misleading when d i f f e r e n t body weight 115 groups are compared. There are only two circumstances under which the above expression may be used v a l i d l y : ( l ) i f a l l the animals compared are of very s i m i l a r body weights; or (2) i f the arithmetic regression of organ weight on body weight i s l i n e a r and passes through the o r i g i n . I know of no instance i n which the l a t t e r i s true, and the former i s not true i n t h i s study. The problem, however, s t i l l remains of co r r e c t i n g f o r d i f f e r e n c e s i n body weight and obtaining a measurement of organ weight which i s independent of the p a r t i c u l a r body weights i n the sample. This d i f f i c u l t y was overcome by C h i t t y (l?6l) by using standardized means ( H i l l , 1959)* These means are calculated as fo l l o w s . A l l the data are grouped and mean organ weights f o r each 10 g weight c l a s s were determined, as w e l l as a grand mean f o r the whole data. The standardized mean i s then obtained by the formula: S = 0/E X G where G = grand mean of the whole data 0 « observed sample mean E = expected sample mean S = standardized mean f o r the sample The observed and expected sample means are ca l c u l a t e d i n the same way as i n chi-square problems. One d i f f i c u l t y of using standardized means i s that confidence l i m i t s cannot be placed on them and no s i g n i f i c a n c e t e s t s may be applied. The technique used by C h r i s t i a n and Davis (1956) and apparently i n a l l of C h r i s t i a n ' s work i s somewhat s i m i l a r to the standardized mean method but the f i n a l r e s u l t s are expressed i n percentages ( i . e . by s u b s t i t u t i n g X 100 f o r X G i n the above equation the r e s u l t s would be expressed i n percentages) The difference i s that he does not weight the means of the component body weight groups i n r e l a t i o n to t h e i r sample s i z e , i . e . a weight group with only one animal i n i t contributes as much t o the mean as a weight group with 25 116 animals i n i t . These means are thus l e s s r e l i a b l e than true standardized means which are used i n t h i s study. RESULTS Body Weight D i s t r i b u t i o n s Much information can be learned from age or weight d i s t r i b u t i o n s (Bodenheimer, 1938j L e s l i e and Ranson, 19u0). Tables 1+8 and 1+9 give the weight d i s t r i b u t i o n s f o r Lemmus and Dicrostonyx males during 1959-61 on the Main Study Area. The data f o r the females are not given here because they are very s i m i l a r to that f o r the males. A l l the weight data discussed here were obtained from the snap trapping samplesj weight data from l i v e trapping are not presented but they show the same changes described here. Some care must be taken i n comparing weight d i s t r i b u t i o n s between the years because 1959 was b i o l o g i c a l l y 2-3 weeks behind I960, and 1961 was about 1 week behind I960. Several points are shown by these data. F i r s t , the peak summer of 1960 was characterized by higher adult body weights than e i t h e r 1959 or 1961. There were very few Lemmus above 76 g i n e i t h e r 1959 or 1961, but i n i960 a majority of the adults were above t h i s weight. In Dicrostonyx the dif f e r e n c e between I960 and 1961 was not so w e l l marked but the same tendency was shown. Second, i f we consider the winter data, Lemmus did not appear to increase i n weight through the winter whereas the Dicrostonyx weight d i s t r i b u t i o n s suggest that they d i d increase i n weight at l e a s t slowly during the winter. A sudden spurt of growth seems to occur i n May f o r each species. Third, there was a gap i n the 1961 summer weight d i s t r i b u t i o n s where the e a r l y summer young should be. Again t h i s was shown more c l e a r l y by Lemmus than by Dicrostonyx. The l e f t part of Figure 7 shows g r a p h i c a l l y the l a t e J u l y Lemmus male weight d i s t r i b u t i o n s f o r i960 and I 9 6 I on the Main Study Area and i l l u s t r a t e s two of these points, i . e . the higher body weights of i960 TABLE 1+8. Body weight d i s t r i b u t i o n s f o r Lemmus males on the Main Study Area, 1959-61. Dotted l i n e s separate summer generation from winter and spring generations. WEIGHT JAN. FEB. MAR. APR. MAY MAY JUNE JUNE JULY JULY AUG. AUG. SEPT. OCT. NOV.. DEC. CLASS 1-15 16-31 1-15 16-30 1-15 16-31 1-15 16-31 (grams) 1959 ••*• — '• \ r -> 11- 3 16- 1 21- X 3 1 2 26- X 3 2 1 31- 3 X 2 1 1 36- 2 X ^ 2 1 ui- 1 1 1+6- 3 1 51- 2 1 mm N 1 56- 3 1 2 61- 1 mm X 66- 1 X X 71- 3 76- mm 81- 1 W T W T~ l i -16-21-26-31-1960 1 1 7 21 11 7 10 k 5 7 \ 18 7 2 3 11 35 5 3 6 lit Ul- - 11 3 —a^c: m * v vv 9 9 -1 — t — iaL 11 1+6- 1 11 6 1 - N 1 2 7 1 5 51- - 22 5 1 - \ 2 3 1 1 56- 1 20 10 1 _ mm 61- 29 10 3 1 X — _ X _ 1 •M 66- 1+2 5 5 2 1 - N *•» * 2? 7 It 1 1 3-, 1 2 81-86-91-96-101-106-25 9 9 1+ 1 5 3 2 1 1 3 6 1+ 1 1 2 1 3 3 1 1 2 3 2 5 1+ 2 3 3 1 X - N 297 6T 5T W W IOT ^9" 29" 17 3T 1961 11- 1 1 1 16- - 1 V v 1 1 2 21- 1 X x— 2 6 26- mm _ X — X, — 3 6 31- 3 1 1 _ X X — 3 36- 6 2 1 _ X _ x 1+1- 6 6 1 1 1 3 1 1+6- 2 1 2 1 X X, x 51- 1 1+ 1 X X 56- 1 2 1+ 1 X X, 61- — 2 2 2 66- ** 2 1 2 3 1 2 2 76- 1 3 81- 1 86-91-96-101-IT 12~ T __ W i i ~ 1~ T I T TABLE 49. Body weight d i s t r i b u t i o n s f o r Dicrostonyx males on the Main Study Area, 1959-61. Dotted l i n e s separate summer generation from winter and spring generations. WEIGHT JAN. FEB. MAR. APR. MAY MAY JUNE JUNE JULY JULY AUG. AUG. SEPT. CLASS 1-15 16-31 1-15 16-30 1-15 16-31 1-15 16-31 (grams) OCT. NOV. DEC, 1959 11-16-21-26-31-36^. Ul-46-51-56-61-66-71-1 1 _3_ 1 1 3 3 •s. 1 2. _3_ 2 2 2 1 1 2 1 1 1 10 I960 11-16-21-26-31-1 5 2 1 1 2 2 2 3 1 v 1 \ \ -x- 2 1 - \ 1 1 1 1 1 2 2 2 6 hi- 1 l _ - 2 1 i k U6- - U - - _ 2 N— 2 3 1 l 2 51- - 3 1 3 1 2 \ — X - 2 2 I 2 56- — k - 3 1 2 - \ — 1 3 61- 1 2 1 2 1 - —* •BB IM> 66- 1 3 3 1 - 2 - 2 mm-s- 1 71- — 1 - 3 3 3 1 -76- 1 - 1 2 - - 1 1 -81- 1 - 2 2 - M -86- - 1 mm 1 1 -91- - 2 - 1 -96- - - -101- 1 2 -106- 1 T 20" IB" 2T r r w ~9~ W T "8~ 1961 11- 1 1 1 16- - 2 N X. 1 5 2 21- - 1* 1 mm 7 26- - 6 1 1 6 2 31- 1 2 1 2 v 1 2 36- 1 1 3 _ u — 1 hl- 1 1 1 7 - 1 l v -46- 1 1 1 2 1 1 1 — 51- 1 1 6 2 3 mm _ 56- 2 2 _ 10 4 2 1 3 61- - — - 7 2 2 2 3 66- - 1 _ 8 1 1 1 -71- 1 2 2 2 _ 1 76- _ U — 1 81- 1 2 _ 86- 1 2 91- 1 1 96-T T T T 58" IT TT 20" w 119 FIGURE 7. Body weight d i s t r i b u t i o n s f o r Lemmus males, J u l y 16-31, I960 and 1961. Winter, spring, and summer in d i c a t e generations. i ooH to < 80H I o > o 6 0 H 40H 20H o-WINTER AND SPRING SUMMER N= 40 WINTER WINTER SPRING SUMMER ( _ 2 0 % _ | N = 8 T N= IO —I A L L DATA FROM JULY 1 6 - 3 1 I 9 6 0 MAIN STUDY AREA 1961 MAIN STUDY AREA TYPE G DECLINE 1961 SECOND ISLAND TYPE H DECLINE 120 and the missing summer young of 1961 f o r the Main Study Area. I t i s i n s t r u c t i v e to compare the body weight d i s t r i b u t i o n s f o r Aberdeen Lake during I960 and 1961 with those of the Main Study Area. Tables $0 and $1 give the data f o r Lemmus and Dicrostonyx males at Aberdeen Lake. The f i r s t point to notice about these data i s that there i s very l i t t l e d i fference between the I960 and 1961 d i s t r i b u t i o n s i n e i t h e r species, contrary to the r e s u l t found on the Main Study Area. High body weights are found i n both years and there does not seem to be a missing group of summer young i n 1961. This diff e r e n c e between Aberdeen Lake and the Main Study Area does not appear u n t i l 1961, as the I960 d i s t r i b u t i o n s on the two areas are very s i m i l a r . With these two d i f f e r i n g patterns i n mind l e t us look at the weight d i s t r i b u t i o n s found on the other out l y i n g areas i n 1961. These data are given i n Table $2. Only data f o r Lemmus males are given; Dicrostonyx i s very sparse on a l l these trapping areas. New Lake, Lower TheIon River, Ten Mile Island, and the Prince River were sampled i n I960 also, but these data are not given here because they are v i r t u a l l y i d e n t i c a l with that p r e v i o u s l y given f o r the Main Study Area i n I960. These 1961 data are based on small numbers of animals, but i f we compare these samples with the corresponding ones from the Main Study Area we f i n d some s t r i k i n g d i f f e r e n c e s . New Lake, Long Island, Second Island, Ten Mile Island, and Nine Mile Island shoxir the weight d i s t r i b u t i o n pattern found at Aberdeen Lake and not that found on the Main Study Area. The Prince River shows the Main Study Area pattern. These two d i f f e r e n t patterns are shown g r a p h i c a l l y i n the r i g h t p art of Figure 7« We can summarize these r e l a t i o n s h i p s i n the following way: 121 TABLE 50. Body weight d i s t r i b u t i o n s f o r Lemmus males at Aberdeen Lake, 1960-61. WEIGHT. CLASS 11-16-21-26-31-ia-U6-51-56-61-66-81-86-91-96-101-106-MAY 27-JUNE 2 1 1 2 1 I960 JUNE 15-16 JULY 10-18 1 6 N 1 V \ — \ 1 1 T 1 2 3 2 2 ~~2~ 3 2 1 2 SB-JUNE 1-5 1 1 1961 JUNE 22- JULY J u l y 10 26-29 2 7 \ \ —v-1 1 ~2~ 1 2 2T" 1 2 2 TABLE J?l. Body weight d i s t r i b u t i o n s f o r Dicrostonyx males at Aberdeen Lake, 1960-61. Dotted l i n e s separate summer generation from winter and spring generations. WEIGHT I960 1961 CLASS MAY 2 7 - JUNE JULY MAY 28- JUNE JULY JULY JUNE 2 1 5 - 1 6 1 0 - 1 8 JUNE 7 13-22 10-19 2 6 - 2 9 1 1 - 1 16- v 3 s — 1 2 1 - 1 N 2 1 — N 1 2 6 - 1 _ N 2 31- wm N — N mm — \ 2 36- 2 N x - - -Ul- 1 5 1 - - 3 v U 6 - _ i 3 1 2 2 5 1 - mm 2 2 - - - 1 5 6 - mm 2 aw 2 mm 2 6 1 - _ 1 1 2 2 66- wm mm 2 MB 2 - 1 71- 1 - 1 3 - 1 1 7 6 - 2 1 2 - 1 2 2 81- mm — 3 - 1 - 1 8 6 - 1 1 mm - 2 91- 1 2 - 1 2 9 6 - mm 1 -1 0 1 - _ 1 1 0 6 - 1 1 1 1 - 1 9 5 2 9 7 1 1 8 26 125 TABLE 52. Body weight d i s t r i b u t i o n s f o r Lemmus males on the o u t l y i n g areas, summer 1961. Dotted l i n e s separate summer generation from winter and spring generations. WEIGHT CLASS 11-16-21-26-31-26= NEW LAKE J u l y U-12 LONG ISLAND J u l y 17-20 SECOND ISLAND J u l y 21+-27 1 U Ul-46-51-56-61-66-13-2 mm 1 3 10" 76-81-86-1 U "cT 1 2 WEIGHT CLASS 11-16-21-26-31-i6r Ul-U6-51-56-61-66-71= 76-81-86-91-LOWER THELON R. August lU-19 1 3 h TEN MILE IS. Aug. lU-19 k 4 1 NINE MILE IS. Aug. 14-19 2 2 PRDJCE R. Aug. 14-17 1 1 3 1 11 12U 1959 Weights Recruit. Ad. d* Y young I960 1961 Weights Recruit Weights Recruit. Ad. ci* Y^ young Ad. cr* Y^ young 1 . Main Study Area * Low + High Low -2 . Aberdeen Lake * High + High * 3. New Lake High + High + ? u. TheIon River High + ? -5 . Ten Mile I s . High + High 6. Prince River High + Low -7 . Nine Mile I s . High + 8 . Long Island High + 9 . Second Island High + ( * Dicrostonyx and Lemmus. Others r e f e r t o Lemmus only.) In the summer of decline those areas which show recruitment are undergoing a Type H decline ( s l i g h t recovery) by d e f i n i t i o n , and those areas showing no recruitment of these e a r l y young are undergoing a Type G decline (no recovery). Thus we reach two conclusions which apply to both species: ( l ) that Type H declines were associated with high body weights and Type G declines with low body weights; and (2) that the adult body weight change was associated with population phenomena and was not simply a side e f f e c t i r r e l e v a n t to the c y c l e . I t i s c l e a r from the 1959 data that low body weights per se are not s u f f i c i e n t to cause a l a c k of recruitment of young, but that something else must also be necessary. I t i s pertinent to enquire what dif f e r e n c e s there are between the areas showing no recovery of numbers i n 1961 and the areas showing some recovery. There i s no apparent r e l a t i o n s h i p with e i t h e r the q u a l i t y of the habitat or the population density of the area i n I960. This i s i l l u s t r a t e d i n the following t a b l e : TheIon River New Lake Main Study Area Ten Mile I s . Type of Vegetation t h i c k t h i c k sparse sparse Density i n I960 very high very high mod. high mod. high Type of Decline no recovery some recovery no recovery some recovery 125 It i a also c l e a r that weather cannot be the only cause because opposite e f f e c t s were found within 1-2 miles of each other. While we cannot rule out other e x t r i n s i c e f f e c t s such as disease, t h i s difference i n the qu a l i t y of the i n d i v i d u a l s as measured by body weight may be caused by differences i n the i n t r i n s i o f a c t o rs o f the various populations independent o f the absolute density. There i s no information from t h i s study to t e s t t h i s suggestion. F i n a l l y , a l l the four i s l a n d populations sampled were undergoing Type H declines. The s i g n i f i c a n c e of t h i s i s not understood. To sum up the r e s u l t s of analyzing body weight d i s t r i b u t i o n s , we have seen that the peak summer was characterized by adults of high body weight, and that two patterns appeared i n the decline* ( l ) low body weights and no recruitment of young, i n Type G declines; and (2) high body weights and recruitment of young, i n Type H declines. Mean Body Weights We may quantify the observation that high body weights were associated with the peak summer and Type H declines by computing mean body weights f o r the adults. Tables 55 and 54 give the mean body weights f o r the winter and spring generations o f Lemmus and Dicrostonyx males f o r 1959-61. We are mainly concerned here with the winter generation. The Lemmus data (Table 55) are very c l e a r . On the Main Study Area the peak summer of I960 showed mean body weights about 28% greater than 1959 and 20% greater than 196*1. In both cases the diff e r e n c e s are c l e a r l y s i g n i f i c a n t . For Aberdeen Lake the i960 and I96I data are not s i g n i f i c a n t l y d i f f e r e n t , high body weights occurring i n both years. The other areas sampled i n I96I a l l have high mean body weights except f o r the Prince River. The Dicrostonyx data (Table 54) are not so c l e a r . On the Main Study Area the peak summer of i960 showed mean body weights about 4o$ greater than 1959 and 11% greater than I96I. The question a r i s e s whether the l a t t e r f i g u r e i s s t a t i s t i c a l l y s i g n i f i c a n t . A non-parametric ranking t e s t 126 TABLE 53. Mean body weights f o r Lemmus males of the winter and spring generations, summers 1959-61. LOCATION AND TIME PERIOD Main Study Area 1959 June 16-30 J u l y 1-15 J u l y 16-31 August 1-15 August 16-31 I960 May 16-31 June 1-15 June 16-30 J u l y 1-15 J u l y 16-31 August 1-15 August 16-31 1961 May 16-31 June 1-15 June 16-30 J u l y 1-15 J u l y 16-31 August 1-15 August 16-31 N WINTER GENERATION WEIGHT S.E. 10 a a 3 2U1 57 35 15 10 1U 7 2 22 11 2 7 50.21 56.60 mm 69.10 71.30 66.92 67.20 79.33 81.77 83.79 91.06 82.61 76.60 62.58 61.56 67.80 62.8a ±1.83 ±3.07 mm ±4.40 *7.7U ±0.95 ±1.83 ±2.52 ±4.53 ±2.50 ±2.20 ±3.30 +1.80 ±3.28 ±2.10 ±1.30 ±3.50 SPRING GENERATION N WEIGHT S.E. 5 Ml 1 56 8 2 1 31.38 mm 49.40 22.23 36.00 ±1.00 ±0.88 ±3.51 19.U0 19.00 ±4.80 Other Areas I960 Aberdeen Lake May 29-June 2 7 66.9U +7.56 June 15-16 3 66.4O +4.57 J u l y 10-18 20 75.83 +4.00 1961 Aberdeen Lake June 1-5 2 70.45 ±1.35 June 22-July 10 4 85.17 ±4.78 J u l y 26-29 8 79.80 ±2.96 New Lake J u l y 4-12 8 75.99 ±3.14 127 TABLE 53. (continued) Lemmus male mean body weights. LOCATION JSND WINTER GENERATION SPRING GENERATION TIME PERIOD N WEIGHT S.E. N WEIGHT S.E. Other Areas (cont'd) 1961 Long Island J u l y 17-20 6 79.97 ±3.14 mm mm " -Second I s l a n d J u l y 2U-27 5 8I.O4 ±1.52 mm mm -Lower The Ion R. Aug. lU-19 mm mm - - -Ten Mile I s . Aug. 14-19 2 75.75 ±5.74 - -Nine Mile I s . Aug. 1U-19 1 84.4O - - -Prince River Aug. 14-17 h 65.03 •1.1*5 - -128 TABLE 51*. Mean body weights f o r Dicrostonyx males of the winter and spring generations, summers 1959-61. LOCATION AND TIME PERIOD Main Study Area 1959 June 15-30 Ju l y 1-15 J u l y 16-31 August 1-15 August 16-31 I960 May 16-31 June 1-15 June 16-30 J u l y 1-15 J u l y 16-31 August 1-15 August 16-31 1961 Aberdeen Lake I960 WINTER GENERATION N WEIGHT S.E. 5 2 2 1 10 17 8 12 2 3 1 50.00 51.70 52.95 51.1*0 77.50 71.02 67.68 67.58 7U.95 70.73 106.00 May 1-15 3 59.77 May 16-31 5 66.1+0 June 1-15 36 67.33 June 16-30 12 70.32 J u l y 1-15 8 61.71 J u l y 16-31 1* 62.95 August 1-15 5 63.1+6 August 16-31 - mm +2.22 ±2.60 +0.1+5 t i i . l l ±3.79 ±3.72 ±3.1*9 ±0.75 +I*.l6 ±3.57 +6.78 +1.58 ±3.60 +2.95 ±2.70 ±2.59 N SPRING GENERATION WEIGHT S.E. 7 1* k 1 8 8 1 1* 1* 32 7 9 2 3 33.60 37.10 la. 80 U2.00 18.63 29.72 39.20 1+5.20 33.22 35.9U 1+0.3U 39.28 1*5.95 53.03 May 27-June 2 8 81+.70 ±7.71 1 23.10 June 15-16 2 81.75 ±1*.85 3 50.77 July 10-18 12 76.90 +2.9U 12 l*i*.l*7 1961 May 28-June 2 5 75.51* ±6.11+ 2 32.00 June 13-22 9 75.58 +5.53 2 38.85 July 10-19 5 70.76 ±3.60 2 1+8.55 July 26-29 10 80.90 ±3.28 10 1*9.53 +1.98 ±1.68 ±3.00 ±1.13 ±2.56 •2.08 ±6.95 +2.16 ±3.90 +2.62 ±0.1*5 ±6.25 +1.1*9 +1.55 ±9 .60 +8.35 +1.05 ±2.1*6 129 (Snedecor, 1 9 5 6 , p 115) was applied to the i 9 6 0 vs. 1961 data f o r the winter generation and gave a s i g n i f i c a n t difference ( P < . 0 5 , > . 0 l ) . For Aberdeen Lake the i 9 6 0 and 1961 data are not s i g n i f i c a n t l y d i f f e r e n t , as i n Lemmus. Growth of adult animals i s mostly complete by e a r l y June i n both species. I n d i v i d u a l adults captured i n the l i v e trapping program from June to August show growth rates averaging about 0 . 2 % per day i n both i 9 6 0 and 1961 f o r both species. I t thus appears that the c r i t i c a l growth pe r i o d f o r the adults i s A p r i l and May, before the snow melts or the new season's plant growth begins. In summary, an analysis of mean body weights f o r the winter generation confirms q u a n t i t a t i v e l y the p r i o r observation that high body weights (x « 70-85 g) were found i n the peak summer on a l l areas and i n Type H d e c l i n e s i n 1 9 6 l . Lower mean body weights (x = 5 0 - 6 5 g ) p r e v a i l at the other times f o r both species. Most of the growth which produces these d i f f e r e n c e s occurs i n A p r i l and May before the snow melts, and adult growth rates during the summer are low. Organ Weights Over 6000 organs from about 2I4OO lemmings were weighed i n the course of t h i s study i n an attempt to f i n d a p h y s i o l o g i c a l index which i s cor r e l a t e d with the pr e v i o u s l y described population processes. The idea that c e r t a i n p h y s i o l o g i c a l changes i n i n d i v i d u a l s cause profound changes i n population processes i s very widely held, p a r t i c u l a r l y because of the work of C h r i s t i a n ( 1 9 5 0 , 1957, 1 9 6 1 ) . The assumption i s that each i n d i v i d u a l has a c e r t a i n i n t e r n a l p h y s i o l o g i c a l state which can be conveniently measured by weighing one or more of several i n t e r n a l organs such as the t e s t e s , adrenals, and spleen. The fu r t h e r assumption i s made that t h i s p h y s i o l o g i c a l state causes changes i n population processes. Thus we have 130 diagrammatically: 'physiological state" \ causes changes i n population processes (e.g. reproduction) r e f l e c t e d by changes i n organ weights Now i t i s of course possible that the r e a l " p h y s i o l o g i c a l state" i s not measured by these organ weights. But the point here i s that a l l the upholders of these " p h y s i o l o g i c a l " t heories do r e l y on organ weights and have based t h e i r supposed confirmations on organ weights. Hence we may begin by using t h e i r assumption. Lemmus males and females, and Tables 57 and 58 give the same data f o r Dicrostonyx males and females. The organs included are: f o r the males — t e s t e s , adrenals, and spleen; and f o r the females — adrenals and spleen. A l l t e s t e s and adrenal weights given are paired weights. Fat i n d i c e s are also given and w i l l be discussed l a t e r . Four separate standardizations were performed f o r each organ: males and females, and winter and summer generations. For t h i s reason comparisons should be made down the columns only and not across the rows of these t a b l e s . To correct to some extent f o r v a r i a t i o n due t o reproductive status,".I have included only fecund males and pregnant or l a c t a t i n g females f o r the winter generation f i g u r e s , and only non-fecund males and nullip a r o u s females f o r the summer generation f i g u r e s . The groups omitted by these r e s t r i c t i o n s are small and discontinuous i n time. enquire whether population density changes were associated with adrenal weight changes. I t i s d i f f i c u l t to see any consistent r e l a t i o n s h i p between these adrenal data and the population changes. There i s a seasonal change Tables 55 and 56 give the mean standardized organ weights f o r The f i r s t organ we may consider i s the adrenal gland, and we may TABLE 55. Standardized mean organ weights (milligrams) and f a t index f o r Lemmus males, 1959-61 LOCATION AND TIME PERIOD Main Study Area 1959 June 16-30 July-August Sept.-Oct. December I 9 6 0 F e b . - A p r i l May 1 6-31 June 1 - 1 5 June 16-30 J u l y 1 - 1 5 J u l y 16-31 August 1 - 1 5 August 1 6-31 Sept. 1 5 - 3 0 Oct. 2 7 - N o v . December 1 9 6 1 January February March A p r i l May 1 - 1 5 May 16-31 June 1 -15 June 1 6 - 3 0 J u l y 1 - 1 5 J u l y 1 6 - 3 1 August 1 - 1 5 August 16-31 Other Areas I960 Aberdeen Lake May 27-June 2 J u l y 10-18 1 9 6 1 - J i - r Aberdeen Lake J u l y 2 7 - 2 9 Long Island J u l y 17-20 Second Island J u l y 2U-27 Prince River August H t - 1 7 Nine Mile Island August lU-19 Ten Mile Island August 1U-19 TheIon River August lk-19 ADRENALS TESTES SPLEEN FAT INDEX W# S* ¥ S W S> W S N WT. N WT. N WT. N WT. N WT. N WT. N IND. N IND. 12 25.2 - ia 6a6 ia 231 _ _ 15 1 . 7 _ Mi a 26.0 - - a 521 - - a 2a3 - - a 1.8 - -7 2a.8 11 12.0 7 361 11 86 7 353 11 63 7 2.0 11 1 . 7 — - 2 10.2 - Mi - Mi 1 58 M - 3 1.9 2 16.9 mm - 2 186 - - 2 83 - - 3 2.a 3 2.a mm 3 11.0 2 15a mm 1 18 Ml 3 1.0 88 23.2 - - 89 5a5 - 88 66 Mi - 207 2.1 56 26.2 - - 58 512 - 57 109 - - 57 i.a _ 35 27.5 - - 31 5io - Ml 35 199 - Mi 35 1.2 — • 2 3 28.8 8 1 1 .9 23 a9o 7 30 23 2a9 6 33 22 1.2 9 1.5 ia 2U.8 38 7.3 ia 380 31 15 ia aoi 33 7 7 ia 1.1 33 1.3 ia 25.0 86 7.8 ia 305 87 16 ia 2 8 a 85 72 13 1.5 9a 1.8 - - 50 9 . 7 - - 50 15 - _ a9 56 mm - 52 2.0 - - 27 7.6 mm - 27 13 mm - 27 69 mm. Ml 27 2 . 7 - - 17 16.1 - - 1 7 15 - mm 17 38 — 17 i.a - - 32 1 6 . 3 - - 32- 20 - - 31 35 - - 31 1.5 - - 19 13.7 - M* 19 17 - - 19 30 _ _ 19 1.5 - - 12 11.5 - - 12 15 - 12 31 Mi Mi 12 1.2 - - a 13.8 - Mi a ia - - a aa _ - a 1.8 - Ml 3 i a . i - - 3 31 - - 3 aa — 3 2.0 12 a3 — 13 1.9 2 25.3 - mm 2 635 _ 2 iao _ _ 2 a.2 22 2a.a - - 22 63a _ 22 96 22 2.5 _, 8 21.8 - - i i 576 — — 11 156 — — 11 1.1 _ _ 10 26.7 - - 10 556 1 a3 10 267 1 a5 10 1.3 1 1.5 7 25.7 1 11.8 7 a37 9 35 6 516 1 59 7 1.0 1 1.2 - - 7 n.a - - 7 ia - Ml 7 90 _ _ 7 i.a - - 17 12.5 Ml - 17 10 - - 16 7a - - 17 2.1 7 22.5 - mm 7 666 7 86 _ _ 7 2.3 M i >0 2a.7 8 8.2 20 512 8 17 20 265 7 as 20 1.3 8 1.6 8 23.6 ia 9.7 8 535 ia 21 8 ao3 ia 60 8 1.2 ia 1.7 6 2a.8 3 7.2 6 a86 3 25 6 375 2 52 6 0.9 3 i.a 5 25.7 5 8.a 5 576 5 a3 5 a6a 5 55 5 0.9 5 i.a a 2a.3 - 3 12.8 a 399 3 9 a 67a 3 6a a 1.3 3 1.9 1 22.1 a 7.9 1 a22 a 35 1 231 a 79 1 1.7 3 1.9 2 22.7 9 u.a 2 538 9 32 2 213 9 62 2 0.9 8 1.5 - - 8 11.2 - - 8 12 _ 8 73 _ M 8 1 . 8 •»• W = winter generation (fecund animals o n l y ) . # S = summer generation (non-fecund animals o n l y ) . TABLE 56. Standardized mean organ weights (milligrams) and f a t index f o r lemmas females, 1959-LOCATION AND TIME PERIOD ADRENALS SPLEEN FAT INDEX WINTER * SUMMER *• WINTER SUMMER WINTER SUMMER N WT. N WT. N WT. N WT. N IND. N IND. Main Study Area 1959 June 16-30 u 3 8 . 0 6 21*3 - - 9 1.8 - -J u l y 2 3 2 . 3 - - 2 259 - - 2 2.1* mm -August 1* 29.0 3 1 0 . 7 5 2 6 7 3 1*7.6 - mm h 1.7 Oct.-Dec. - - 3 H*.o - - 1 2 1 . 9 - - k l*.o I 9 6 0 A p r i l 2 3 0 . 0 mm - 2 119 - - - - 1 2.3 May 16 -31 3 25.3 - - 3 8 1 - mm 1*5 1.7 - -June 1-15 1 8 3 2 . 7 - - 18 1 6 6 — — 2 1 1.1 - •m. June 1 6 - 3 0 2 1 3 8 . 1 — - 2 1 130 - — 1 1 1.1* - -J u l y 1-15 1 6 3 6 . 5 8 1 2 . 3 1 6 198 8 53.9 - - 9 1.7 J u l y 1 6 - 3 1 2 1 3 2 . 7 26 7 . 1 21* 285 25 1 0 8 . 0 2 1.3 26 1.8 August 1-15 3 2 . 2 1*1 7.2 11* 2 1 1 1*0 59.6 6 1.3 1*0 1.9 August 16-31 - mm 3 7 1 0 . 2 - - 3 7 6 9 . 1 * 2 1.3 3 7 2.2 Sept. 15-30 - - 15 6.8 - - 15 50.1 - - 15 2.5 Oct.-Nov. - - 18 13.1 _ - 1 8 ll * . 3 - 18 1.7 December - - 26 15.U - - 2 6 2 7 . 7 mm - 2 6 1.7 1 9 6 1 January - - 32 13.3 - - 3 2 22.7 - - 3 2 1.5 February - - 8 15.1* - - 8 3U.8 - - 8 1.1 March - - 6 1 0 . 9 - - 6 2 0 . 3 _ mm 6 2.0 A p r i l - - 3 15.3 - - I* 26.0 mm - I* 1.1* May 1-15 - - 11 2 0 . 1 * - - 11 1*1*.2 - - 13 1.3 May 16-31 - - - ' - - - - - 1* 2.5 m June 1-15 2 7 . 3 - — 1* 11*9 mm mm 8 1.8 _ mm June 1 6 - 3 0 5 3 7 . 3 - - 5 188 1* 1.8 _ J u l y 1-15 2 29.7 mm - 2 215 — _ Mft _ _ J u l y 16-31 6 33.2 - - 6 1*66 mm _ _ _ _ August 1-15 3 26.5 1* 11.8 3 171* 1* 6 2 . 1 _ _ 1* 1.7 August 16-31 - - 8 15.6 - - 8 70.1 - - 8 2.0 Other Areas I960 Aberdeen Lake May 27-June 2 - -June 15-16 U 35.1 J u l y 10-18 12 28.6 7 8.9 1961 Aberdeen Lake J u l y 26-29 7 30.3 16 10.7 Long Island J u l y 17-20 h 28.0 5 9.5 Second Island J u l y 2l*-27 3 30.9 2 10.5 Prince River August U*-17 - 2 1U.7 Nine Mile I s . August ll*-19 1 22.3 1 8.1 Ten Mile I s . August ll*-19 1 27.6 3 13.3 TheIon River August ll!.-19 - - I* 10.8 1* 12 7 1* 3 157 179 371* 201 189 11* 5 2 2 1 3 1* 32.8 77.7 1*9.9 156.1 51.0 85.8 31*. 9 73.6 5 1* 1 1.5 1.0 1.5 16 5 2 2 1 3 1* 1.6 1.8 1.7 1.6 2.1* 1.6 1.9 2.0 * Winter Generation : adrenals and spleen - pregnant or l a c t a t i n g only f a t index - pregnant or parous only (not l a c t a t i n g ) Summer Generation : nul l i p a r o u s animals only. TABLE 57. Standardized mean organ weights (milligrams) and f a t index f o r Dicrostonyx males, 1959-61 LOCATION AND TIME PERIOD A D R E N M J S W " # S # TESTES S P L E E N F A T I N D E X W w N WT. N WT. N WT. N WT. N WT. N WT. Main Study Area 1959 June 15-30 6 I4 .8 - - 6 261 - - 6 31.5 - -J u l y 12 12.1* - - 12 273 - - 12 55.3 - -August 2 13.8 6 9.1 2 202 6 51 2 37.6 6 25.3 November 2 11.1 3 6.2 1 89 2 a5 18.8 3 23.7 I960 Jan.-Mar. - - - - 3 69 2 17 - - mm -A p r i l 8 1U.7 5 8.3 11 115 9 38 a 61.3 - mm May 16-31 2 12.1 - - 2 210. - - 2 31.3 -June 1-15 1U 13.7 - - Iii 22U - - i a a9.2 -June 16-30 9 15.5 - mm 9 196 - - 9 60.8 — — J u l y 1-15 13 1U.9 1 7.1 13 230 1 a i 13 77.a 1 ao.8 J u l y 16-31 2 12.7 10 8.1 2 139 ,11 20 2 73.1 9 39.a August 1-15 3 33.6. 7 8.8 1 132 7 23 3 132.3 7 as.s August 16-31 - - 21* 9.1* 1 100 2a 22 1 121..0 2a 59.9 Sept. 15-30 - mm 9 7.3 - - 9 25 - _ 9 ai.7 November - - 2 10.8 - - 2 19 _ 2 37.0 December - - 8 11.2 - - - 8 37 - - 8 a2.3 1961 January - mm 5 9.3 - - 5 3a _ _ 5 33.1 February - - 1 7.6 - - 1 36 — 1 2a.9 March - - 3 13.6 - - 3 56 _ _ 3 28.2 May 1-15 1 13 .1* - — 1 12U _ _ 1 22.6 _ _ May 16-31 7 15.3 — — 7 I69 — 7 35.6 M M June 1-15 U7 16 .0 - - 1*7 217 mm MM a6 U9.0 _ Junel6-30 16 16.6 _ - 16 186 mm _ 16 69.8 _ _ J u l y 1-15 16 1U.6 - - 16 235 — - 16 98.a mm MM J u l y 16-31 6 12.3 2 7.6 6 253 3 a6 6 65.a 3 37.5 August 1-15 7 11.8 12 9.3 7 2Ul 12 27 6 112.0 12 38 .6 August 16-31 - - i a 9.U - - 1U 25 - - i a 3l*.9 w N IND. N IND. 7 12 2 1 3 11 2 i a 9 12 2 3 1 1 7 a7 18 16 6 7 1.5 2.0 1.9 1.2 2.6 1.9 2.3 1.1 1.0 1.2 1.2 1.7 1.8 2.6 3.0 2.a i . a 1.3 i . a 1.2 6 2 2 9 1 11 7 2a 9 2 8 5 1 3 3 11 ia 1.6 2.a 2.7 2.6 1.6 1.2 1.7 2.2 2.5 2.a 1.6 1.7 0.9 1.0 2.0 1.6 1.7 Other Areas I960 Aberdeen Lake May 27-June 2 7 15.2 - - 8 250 - - 8 3U.2 - 8 2.3 -June 15-16 5 12.3 - - 5 22a - - 5 a7.3 - 5 1.7 -J u l y 10-18 11 12.7 5 7.3 9 178 5 32 9 109.1 a 87.3 9 1.2 5 1.7 1961 Aberdeen Lake J u l y 26-29 5 15.6 5 7.3 5 160 5 16 5 91.2 5 1*3.8 5 I . a 5 1.8 -* W » winter generation (fecund animals o n l y ) . S = summer generation (non-fecund animals o n l y ) . TABIE 58. Standardized mean organ weights (milligrams) and f a t index f o r Dicrostonyx females, 1959-61. LOCATION AND TIME PERIOD ADRENALS WINTER # SUMMER # SPLEEN FAT INDEX WINTER SUMMER WINTER SUMMER N WT. N WT. N WT. N WT. N IND. N IND, Main Study Area 1959 June 15-30 3 27.3 - mm 3 108 mm mm u 1.1 - -J u l y 9 22.5 2 8.3 9 102 2 Ui 6 2.2 2 2.5 August U 17.7 8 9.0 U 71 8 2U mm 8 2.0 Oct. - Nov. 1 12.8 - mm 1 65 - - 3 2.9 2 1.8 I960 Jan.-March 2 17.0 - - 2 20 - - 3 2.1 6 3.1 Apr.-May 15 - - - - - - - - 2 0.9 11 1.8 May 16-31 1 11.9 - - 1 25 - - - - 7 1.7 June 1-15 9 29.U - mm 10 73 mm _ 17 1.3 - -June 16-30 5 25.6 mm - 5 165 - 5 1.6 M l _ J u l y 1-15 10 23.6 - - 10 79 - _ 1 1.8 — J u l y 16-31 6 25.3 15 7.3 6 71 15 58 2 1.3 15 l.U August 1-15 5 2U.0 H i 8.8 U 73 lU ia 2 1.2 13 1.7 August 16-31 1 16.0 25 8.7 1 22 26 U8 9 1.6 27 2.3 Sept.,. 15-30 - - 6 6.1 _ 6 29 U 1.7 6 2.5 Oct. - Nov. - - 5 9.2 - - 5 33 _ _ 5 1.3 December - - 5 7.6 - - 5 33 - - 5 1.6 1961 January - - 2 7.1 - 2 3U — - 3 1.5 February - - 3 10.U - - 3 28 - — 2 0.9 March - - — - — - - - - - 1 3.7 A p r i l - - 2 10.1 - - 2 52 - - U 2.1 May 1-15 1 11.2 - - 1 16 - - 1 1.7 2 1.6 May 16-31 mm - - - - M l - - 1 1.6 - -June 1-15 7 25.9 - - 5 69 - — 22 1.6 — -June 16-30 19 23.5 - - 20 82 - 22 1.7 - -J u l y 1-15 9 22.2 U 7.3 8 106 U 56 5 l.U u 2.0 J u l y 16-31 5 16.9 11 7.0 5 56 10 Ul 2 1.2 11 1.6 August 1-15 3 20.2 5 8.0 3 68 5 UU 2 0.8 5 1.6 August 16-31 3 15.7 10 9.0 3 85 10 35 1 1.8 9 1.5 Other Areas I960 Aberdeen Lake May 27-June 2 - - - - U 1.9 - -June'15-16 7 25.0 - - 7 97 7 1.1 J u l y 10-18 21 21.2 9 6.6 21 93 8 6U 11 1.1 9 1.6 1961 Aberdeen Lake J u l y 26-29 9 21.3 10 7.3 9 95 11 U9 9 1.5 13 1.8 •* Winter Generation ( i n c l . Spring Gen.) ? adrenals and spleen - pregnant or l a c t a t i n g only. : f a t index - pregnant or parous only (not l a c t a t i n g ) Summer Generation: nulli p a r o u s animals only. 135 i n adrenal weights, r i s i n g to a peak i n June or July, s i m i l a r to that described by C h i t t y (1961) f o r Microtus a g r e s t i s . At one point or another almost a l l the means overlap and thus i t i s not possible to say c a t e g o r i c a l l y that any one summer showed higher or lower adrenal weights than another summer. However, some years tend to be higher or lower than others, and we may broadly c l a s s i f y the years as follows: Summer Winter 1959 I960 1961 1960-61 Lemmus Adult males Low High Low Adult females Low 1 High Low Young males High • Low High High Young females Low ? Low High High Dicrostonyx Adult males Low Low High Adult females High High Low Young males Low Low Low High Young females Low Low Low High The mean dif f e r e n c e between "high" and "low" adrenal weights was $-lk% f o r the summer adults, and 23-33% f o r the Lemmus summer young. High adrenal weights were found i n a l l groups i n the winter of 1960-61, but unfortunately comparative data from the previous winter are not a v a i l a b l e . There i s a cl e a r r e l a t i o n s h i p of summer adrenal weights to the cycle i f we look at single groups such as the Lemmus adult males. These r e l a t i o n s h i p s , however, are not consistent between groups, as can be seen, f o r example, by comparing adults and young. Two conclusions follow from these data: (1) summer adrenal weights do not show a consistent r e l a t i o n s h i p to the phase of the cycle; and (2) winter adrenal weights seemed high i n I96O-6I r e l a t i v e to the summer weights. The second organ to be considered i s the t e s t i s . There i s some c o r r e l a t i o n between tes t e s weights and density changes, which shows up as 136 follows: Slimmer 195° I960 1961 Lemmus Adult males Young males High High Low Low High Low Dicrostonyx Adult males Young males High High Low Low Low Low The mean diff e r e n c e s between "high" and "low" testes weights was lk-21% f o r the summer adults and U9-78$ f o r the summer young. Testes weights were highest i n both species during 1959. The peak summer was characterized by low tes t e s weights i n a l l groups, but i n the decline Lemmus adults showed a d i f f e r e n t trend from the others. These r e s u l t s f o r the young agree with those discussed p r e v i o u s l y regarding reproduction. Young male Lemmus d i d not mature e i t h e r i n I960 or 196l. Young male Dicrostonyx d i d not ever seem to mature i n t h e i r f i r s t summer; nonetheless they seem to show the same type of changes i n testes weights ( c f . August weights) as do Lemmus young. Because t h i s i n h i b i t i o n of gonadal development i n young male lemmings occurred i n the summer of decline as w e l l as i n the peak summer, density per se cannot be the f a c t o r d i r e c t l y involved here, but rather the important variable must be capable of acting at very low d e n s i t i e s i n the d e c l i n e . The f i n a l organ to be considered i s the spleen. The spleen i n lemmings v a r i e s considerably i n s i z e , weighing from 5-600 mg i n Dicrostonyx and 5-1200 mg i n Lemmus. The very heaviest spleens are found only i n midsummer animals (l a t e July-August) and there i s thus a. very strong seasonal v a r i a t i o n i n average weights. There was l i t t l e d i fference between the d i f f e r e n t years, and the spleen weights showed no c l e a r r e l a t i o n s h i p to the cycle i n numbers. The s t r i k i n g seasonal change i n spleen size i s not understood but may be 137 associated with blood p a r a s i t e s transmitted by mosquitoes during the midsummer in s e c t season (Baker and C h i t t y , ms.) or by mosquito or other ectoparasite b i t e s d i r e c t l y (Chitty and Phipps, I960). This hypothesis i s consistent with the observation that Lemmus has a greater spleen enlargement than Dicrostonyx. because Lemmus l i v e s i n the wetter places where mosquitoes are more abundant. To sum up, summer adrenal and spleen weights showed no c l e a r r e l a t i o n to the cycle i n numbers. Winter adrenal weights were high i n 1960-61 but comparative data from 1959-60 are l a c k i n g . Testes weights tended to change systematically over the c y c l e , being high i n 1959, low i n I960, and somewhat v a r i a b l e i n 1961. A l l these organs showed a seasonal cycle of weight changes independent of the population c y c l e . Fat Changes The amount of f a t stored by lemmings may be used as another index of general p h y s i o l o g i c a l condition. This was assessed by an a r b i t r a r y f a t index scale of 1-5, 1 being the value f o r an animal with no f a t and 5 f o r a very f a t animal. This index was estimated purely s u b j e c t i v e l y by observing the amount of f a t on the skin, between the shoulders, around the hind legs, and around the v i s c e r a and gonads. Animals with no f a t to be seen (except around the gonads where i t i s almost always present) were always classed as f a t index 1. The skins of animals of f a t 3 or more were always greasy and had to be wiped or scraped a f t e r drying. These data on f a t changes were analyzed i n the same way as the organ weight data. Fat index data f o r Lemmus and Dicrostonyx males and females are given i n Tables 55-58 along with the organ weight data. The f a t index shows a seasonal v a r i a t i o n , being at i t s lowest i n midsummer when breeding i s intense and highest i n the winter ( p a r t i c u l a r l y i n f a l l and s p r i n g ) . I f we compare the spring and summer data of the d i f f e r e n t years, there does not 138 seem to be any diff e r e n c e between years i n t h i s index. In p a r t i c u l a r , the spring and summer of 1961 have f a t i n d i c e s equal to or greater than e i t h e r I960 or 1959• There i s thus no i n d i c a t i o n of undernourished animals i n the spring of the d e c l i n e . Low f a t indi c e s may have p r e v a i l e d i n the winter of 1960-61, but data from the previous winter are not s u f f i c i e n t f o r a good comparison. S o c i a l Relationships Very l i t t l e i s p r e s e n t l y known about s o c i a l r e l a t i o n s h i p s i n n a t u r a l populations of c y c l i c rodents. Some i n d i r e c t evidence of s o c i a l r e l a t i o n s h i p s and a few d i r e c t observations w i l l be b r i e f l y presented here with the c l e a r understanding that they are very inadequate. In e a r l y June I960 both species of lemmings were extremely h o s t i l e i n behavior, at l e a s t towards humans. On several occasions while walking across the tundra, I encountered loud squeaking lemmings(both species). Often they were heard squeaking long before one could a c t u a l l y see them ( i n one p a r t i c u l a r instance squeaking began when I was 20 f e e t away). As mentioned previously, lemmings caught on the i c e i n t h i s spring of I960 were also very aggressive. This type of behavior contrasts sharply with t h e i r behavior l a t e r i n the summer when they t r y to hide as soon as one approaches. C o l l e t t (1895) and Curry-Lindahl (1961) also report t h i s curious behavior f o r the Norwegian lemming. The s i g n i f i c a n c e of these observations i s not known. Some crude measure of aggressive behavior may be obtained by the incidence of wounding i n a population (Southwick, 1958). A l l lemming skins c o l l e c t e d i n t h i s study were examined and c l a s s i f i e d on an a r b i t r a r y scale as follows: no recent wounds or obvious scars showing on in s i d e of skin; l i g h t wounds; moderate wounds; or severe wounds. Skins were selected and set out as standards f o r each of these four categories, and these were 139 constantly used.for comparisons. A l l t h i s c l a s s i f i c a t i o n was done i n a two week period at the end of the study so as to minimize the subjective element. Tables 59 and 60 give the incidence of wounding shown on skins f o r Lemmus and Dicrostonyx males over the c y c l e . For both species 1959 showed the lowest o v e r - a l l amount of wounding both f o r old and young animals. The I960 adults of both species showed a high incidence of x^ounding i n l a t e June and July, and t h i s declined by August when breeding had ceased. The i960 young showed a considerable amount of wounding as they moved under the snow i n the f a l l of I960. The 1961 adults showed the highest wounding percentages f o r a l l years i n la t e J u l y and e a r l y August. Unlike I960 however, there appeared to be very l i t t l e wounding u n t i l e a r l y J u l y i n 1961, a time j u s t a f t e r the f i r s t summer young had been born and the females were breeding again. The s i g n i f i c a n c e of these year to year and seasonal differences are not understood, and much more d e t a i l e d work must be done on these p o i n t s . However, these crude data do i l l u s t r a t e three general points: ( l ) there i s a considerable seasonal v a r i a t i o n i n the amount of f i g h t i n g which causes sk i n wounds; (2) t h i s f i g h t i n g was not a simple f u n c t i o n of density because at c e r t a i n times i n the summer of the decline wounding was more extensive than i n the peak summerj (3) both Lemmus and Dicrostonyx showed the same general pattern, although there was l e s s wounding shown on Dicrostonyx skins than on Lemmus skins. Maturation of summer young males during t h e i r f i r s t summer was associated with a considerable amount of wounding. Only one sample of Lemmus summer young contained both mature and immature animals, and the data on these are as follows: l4o TABLE 59• Amount of wounding shown on skins of Lemmus males from the Main Study Area, 1959-61. TIME PERIOD WINTER GENERATION SUMMER GENERATION N % % % N % % % LIGHT MOD. HEAVY LIGHT MOD. HEAV3 1959 28.6 June 16-30 7 mm - - - - -J u l y 5 20.0 - - - - - -August 7 28.6 U2.9 - 11 9.1 - -I960 May 16-31 138 21.7 0.7 - - - - -June 1-15 65 liO.o 1.5 - - - mm June 16-30 3h . 58.8 5.9 5.9 mm - mm -J u l y 1-15 25 U8.0 8.0 12.0 8 - - -J u l y 16-31 1U 50.0 21.il - 36 2.8 - -August 1-15 1U 28.6 7.1 - 85 25.9 1.2 1.2 August 16-31 mm *M - mm 55 36.1, 1.8 -Sept. 15-30 - mm - mm 25 28.0 8.0 -Oct.-Nov. - - - mm 16 62.5 — -December - - - - 30 46.7 10.0 3.3 1961 January - - - mm 18 27.8 - 5.5 February- - - - - 12 8.3 25.0 -March - - - - k 25.0 mm -A p r i l - - - - 2 - - mm May 1-15 - - - - 1 - - mm May 16-31 k 25.0 - - mm mm - -June 1-15 21 19.1 - - - - mm -June 16-30 12 16.7 - — - - - — J u l y 1-15 8 25.0 12.5 mm - — -J u l y 16-31 18 66.7 5.6 - 10 - - -August 1-15 - - - - 7 28.6 - -August 16-31 - - - - 18 38.9 - -Winter Generation - mature animals only. Summer Generation - immature animals only 141 TABLE 60. Amount of wounding shown on skins of Dicrostonyx males from the Main Study Area, 1959-61. TIME PERIOD WINTER GENE! RATION 1 SUMMER GENERATION N % % % N % % LIGHT MOD. HEAVY LIGHT MOD. 1959 June 15-30 11 9.1 9.1 «• mm mm J u l y 13 15.1* - - 3 - -August 2 5o.o . - - 6 - -I960 May 6 - - - — -June 1-15 28 28.6 10.7 _ _ _ June 16-30 9 1*1*. 1* 11.2 _ _ — — J u l y 1-15 16 37.5 - - 1 mm -J u l y 16-31 1* 25.0 - - 11 9.1 mm August 1-15 1* - - - 8 25.0 mm August 16-31 - - - - 22 18.2 aM Sept. 15-30 - - - - 9 hh.k tea Oct.-Nov. - - - 2 50.0 December - - - - 9 22.2 11.1 1961 January - mm - - 5 60.0 M i Feb.-March - - - 3 _ mm May 9 - 11.1 _ _ _ mm June 1-15 65 20.0 — — mm _ June 16-30 17 23.5 - mm mm mm- _ J u l y 1-15 17 1*7.1 - — - _ J u l y 16-31 6 66.7 16.7 - 2 _ mm August 1-15 8 50.0 mm - 12 8.1* mm August 16-31 - - - - 13 7.7 mm Winter Generation - a l l mature. Summer Generation - a l l immature. 11+2 Sample % showing Mean body Range of body Size wounds weight weights August 1959 Immature males 11 9.1 21.7 g 13.5 - 30.8 Mature males 8 75.0 33.5 g 27.8 - 38.5 These differences i n wounding are s i g n i f i c a n t (chi-square f o r independence = 6.01+, df 1, P <.05,>.0l). Since the mean body weights of these samples d i f f e r considerably, some of the diffe r e n c e i n wounding may be explained on t h i s b a s i s . However, t h i s i s probably not the entire explanation because i n I960 and 1961 none of the samples of immature animals with as high or higher body weights showed as much wounding as t h i s mature sample from 1959* This suggests a reason why there seems to be a great increase i n the amount of f i g h t i n g i n the August 1959 Lemmus. I t also suggests a fu n c t i o n f o r the observed i n h i b i t i o n of maturation of males i n the peak and decline summers, i . e . i t prevents a considerable amount of f i g h t i n g i n these populations. Summer young Lemmus females d i d not seem to show t h i s d i f f e r e n c e : J u l y i960 Immature females N = 18 % showing wounds =0.0 Mature females N = 11+ % showing wounds = 0.0 However more p o s i t i v e evidence i s needed on t h i s p o i n t . I n d i r e c t evidence from snap trapping suggests some sort of antagonism between o l d and young Lemmus during the summer of I960. Given the August I960 snap trapping data, we may make the n u l l hypothesis th a t the proportion of adult and young Lemmus i n the habitat types i s the same. The relevant data are as follows: Lichen- Heath-sedge and Sedge Sedge Totals heath h-s hummock Hummock Marsh August adults 0 21 20 9 50 « young 23 188 102 53 366 H43 I f young animals tend to disperse from the densely occupied habitats to the l e s s densely occupied ones, the proportion of young i n the poorer habitats ( i . e . dry f o r Lemmus) should be greater than that of adults, and conversely f o r the better h a b i t a t s . These data were tested by chi-square and the n u l l hypothesis was not rejected, although the r e s u l t i s close (chi-square = 6 . 4 0 , df 3, P< .10, >.0$). Thus although the data suggest fewer young i n the sedge hummock and more young i n the lichen-heath habitats, compared to the adults, the d i f f e r e n c e s are not s t a t i s t i c a l l y s i g n i f i c a n t . I t i s c l e a r that the data a v a i l a b l e on s o c i a l r e l a t i o n s h i p s i s very meagre and almost a l l i n d i r e c t . We know that at times lemmings were very aggressive, that there were large changes i n the amount of wounding shown on skins, and that there may have been some antagonism between old and young Lemmus. Taken together t h i s i s enough to point out that behavioral changes represent the l a r g e s t gap i n our knowledge of the i n t r i n s i c f a c t o r s operating i n the c y c l e . SUMMARY AND CONCLUSIONS (1) High body weights were associated with a l l peak populations and Type H d e c l i n e s . These high weights were about 20-30% greater on the average than the normal low body weights found during the p e r i o d of low numbers and Type G d e c l i n e s . (2) Midsummer weight d i s t r i b u t i o n s f o r Type G declines showed a conspicuous gap where the f i r s t summer young should have been. (3) Organ weights d i d not appear to give us any i n s i g h t i n t o the causes of the c y c l e . Summer adrenal weights and spleen weights d i d not show any consistent r e l a t i o n s h i p to the density changes. Winter adrenal x*eights during 1960-61 seemed to be high i n both species. Testes weights showed a f a i r l y consistent r e l a t i o n to the cycle s i m i l a r to the changes described under reproduction. (10 Spring and summer f a t indices showed no r e l a t i o n to the cycle, and lemmings i n the spring of the decline were as f a t as or f a t t e r than animals from the other years, thus e l i m i n a t i n g any doubt about the quantity of food av a i l a b l e during the d e c l i n e . Winter f a t indices f o r 1960-61 may have been lower than those f o r the previous winter but not enough data are a v a i l a b l e f o r an adequate comparison. (5) I n t r a s p e c i f i c s t r i f e , as measured by the wounds on the skin, showed strong seasonal and year to year changes. This s t r i f e was not a simple function of density because the highest amount of wounding was recorded i n the summer of the d e c l i n e . The changes i n the amount of s t r i f e shown by the skins are not understood. (6) Sexually mature young Lemmus males suffered more wounding than immature young males i n the August 1959 sample. DISCUSSION We. have now presented the r e s u l t s of t h i s study on lemming cycles and must integrate these r e s u l t s with contemporary ideas and studies by other workers. The amount of l i t e r a t u r e published about "c y c l e s " i s t r u l y voluminous, but the proportion of t h i s which presents o r i g i n a l thought or s o l i d evidence i s very low indeed. The pattern followed i n t h i s discussion w i l l be as follo w s . A f t e r a b r i e f h i s t o r i c a l review and some methodological dis c u s s i o n we s h a l l consider the main,changes discussed i n the previous sections and integrate these with the r e s u l t s of other workers. F i n a l l y , we s h a l l consider the current theories about microtine cycles and t h e i r status i n the l i g h t of these data from lemmings. H i s t o r i c a l Approaches and Background The h i s t o r y of "c y c l e s " i s not very long i n terms of years, but we can recognize two general approaches to the problem. The o r i g i n a l observation (Elton, 19210 was that animal populations f l u c t u a t e d i n size and i n some species there appeared to be some r e g u l a r i t y to t h i s change. Given these i n i t i a l data, some workers emphasized the r e g u l a r i t y of the cycles and concentrated much e f f o r t on an attempt to determine the precise p e r i o d of these cycles f o r each species. An example from t h i s group i s Siivonen (19U8). Another approach emphasized " c y c l e s " as a p a r t i c u l a r problem of population regulation and concentrated study on the f a c t o r s operating on the population to cause these increases and d e c l i n e s . A blending of both these approaches i s i l l u s t r a t e d by the work of E l t o n (192U, 1931, 19U2). The f i r s t approach was challenged by Palmgren (19U9) and Cole ( 1 9 5 1 , 195U b, 1958) who demonstrated that " c y c l e s " s i m i l a r i n length to those found i n nature could be interpreted as e s s e n t i a l l y random f l u c t u a t i o n s with 1U6 some s e r i a l c o r r e l a t i o n between successive years. I t i s e s s e n t i a l to understand Cole's argument or we r i s k a complete misunderstanding of what he has shown. Given a set of " c y c l i c 1 1 data on population size f o r any animal, Cole has shown that you can produce a s i m i l a r " c y c l e " i n random numbers by introducing some s e r i a l c o r r e l a t i o n . Now t h i s does not prove anything. I t suggests that, given only these data, we could i n t e r p r e t the " c y c l e " as a random f l u c t u a t i o n , and t h i s would be the simplest i n t e r p r e t a t i o n i f no other data were a v a i l a b l e . In other words, i f we wish to understand " c y c l e s " we must study something more than changes i n numbers. Cole (1958) statest "¥e should seek to understand the causes of each case of population growth and decline instead of looking f o r some hypothetical and c r y p t i c phenomenon capable of generating c y c l e s " . The second approach i s the one noitf emphasized by a majority of workers on c y c l e s . Attention has turned away from the p e r i o d i c i t y and toward the population aspects of c y c l e s . A supposition of t h i s approach i s that the problem of c y c l i c length w i l l be solved once the mechanism of these cycles i s understood. In t h i s study I have followed the second approach. Much of the d i f f i c u l t y of t a l k i n g about " c y c l e s " a r i s e s because several meanings are given to the term, and f a i l u r e to d i s t i n g u i s h between them (e.g. Slobodkin, 1961) gives r i s e to much confusion. .We must therefore attempt to d e l i m i t the p a r t i c u l a r phenomenon to be discussed here from a l l other " c y c l e s " . .Chitty (1952, I960) has discussed t h i s problem and claims that a s p e c i f i c type of cycle may be recognized i n microtine rodents. Using t h i s approach, we may adopt the following d e f i n i t i o n f o r the p a r t i c u l a r type of cycle studied here: i n t h i s paper a cycle i s defined as a t y p i c a l l y 3-U year  -fluctuation i n numbers i n microtine rodents characterized by high body weights  of adults i n the peak summer. I do not propose t h i s as a d e f i n i t i o n everyone i s supposed to accept, but I am merely s t a t i n g the way I s h a l l use the word III? cycle i n t h i s paper. C h i t t y (1?60) defined the problem somewhat more widely and includes t h i s d e f i n i t i o n as only a p a r t i c u l a r instance of the more general problem of why populations f a i l to maintain a high rate of increase. I s h a l l assume, u n t i l there i s evidence to the contrary, that these cycles (as defined above) are a sing l e c l a s s of events and have a common explanation. The two f a c t s ( l ) that they are us u a l l y 3-U year cycles and (2) that high body weights seem to be always associated with peak populations present a strong argument f o r t h i s working hypothesis. Furthermore, t h i s i s a sound methodological approach to the problem i n the present state of knowledge. One a l t e r n a t i v e i s to begin with the assumption that a l l these rodent cycles have a d i f f e r e n t explanation. This implies that each cycle i s unique, l o c a l event and that successive cycles i n the same l o c a l i t y or d i f f e r e n t l o c a l i t i e s cannot be compared, and consequently t h i s makes i t impossible to t e s t hypotheses or to p r e d i c t future phenomena. Another a l t e r n a t i v e i s to d i s t i n g u i s h a l i m i t e d number of d i f f e r e n t types of cycles based on, f o r example, groups of species or c l i m a t i c zones. I have not used t h i s approach because I do not f e e l i t i s the most f r u i t f u l one i n the present state of knowledge. I am thus i n t e r e s t e d p r i m a r i l y i n the things common to a l l cycles and only secondarily i n those things r e s t r i c t e d to a given area or circumstance. However, we must recognize that there i s no guarantee that t h i s i s a s i n g l e c l a s s of events. I t i s possible that the c l a s s i s la r g e r than we have indic a t e d , perhaps inc l u d i n g the gallinaceous b i r d s and the snowshoe hare. I do not wish to argue with those who wish to make the c l a s s l a r g e r , but i t does not seem to me to be prudent to extend the c l a s s beyond the l i m i t s set by the body weight c h a r a c t e r i s t i c u n t i l more evidence becomes a v a i l a b l e . But I do object to a r e s t r i c t i o n of the clas s to include l e s s than that given above. However, i f one believes the body weight c h a r a c t e r i s t i c to be unimportant, one can define the problem d i f f e r e n t l y . 1U8 I suggest therefore that these cycles as defined above seem to represent a si n g l e c l a s s of events and have a common explanation. Thus a single explanation may be sought f o r lemming cycles at Baker Lake, i n Alaska, Scandinavia, and Russia, and vole cycles i n England and e/SQ/ewhere. This i s e s s e n t i a l l y the same b e l i e f expressed by C h i t t y ( 1 9 5 2 ) . Reproduction Several authors have described winter breeding i n lemmings. Thompson (1955 a) working i n northern Alaska on Lammus trimucronatus found that winter breeding occurred only during the period of increase which he claimed occupied two winters, although evidence f o r breeding during the second winter i s not very conclusive (as we have seen, lemmings may breed under the snow every spring). Dunaeva and Kucheruk (19IP-) found winter breeding i n both Dicrostonyx torquatus and Lemmus s i b i r i c u s i n Russia during the period of increase. Sutton and Hamilton (1932) found winter breeding i n both Dicrostonyx groenlandicus and Lemmus trimucronatus on Southampton Island during the period of increase. Nasimovich, Novikov, and Semenov-Tyan-Shanskii (19U8) believed that winter breeding of the Norwegian Lemming was l i m i t e d to the phase of increase. Recently, Curry-Lindahl (1961) and Koponen, Kokkonen, and K a l e l a (1961) reported probable winter breeding i n the Norwegian lemming during the period of increase. Thus i t i s c l e a r that the only reports of winter breeding i n lemmings are from the period of increase. However, during the p e r i o d of low numbers i t would be very d i f f i c u l t to detect winter breeding. There i s equally good evidence that the summer breeding season i n the peak year i s shortened i n lemmings, compared with the increase or decline summers. Thompson (1955 a) reported t h i s f o r L. trimucronatus i n Alaska. Dunaeva and Kucheruk (19U1) reported that breeding had ceased by August i n the peak summer f o r D. torquatus. Nasimovich et a l . (19U8) and I l l 0 K a l e l a (1961) both found t h i s shortened summer breeding season i n peak populations of L. Lemmus* Wildhagen (1953) d i d not report e i t h e r winter breeding or a shortened summer breeding season i n the peak year f o r Lo lemmus i n Norway; h i s samples however are very scattered and discontinuous. The available data on l i t t e r - s i z e changes and pregnancy-rate changes over the lemming cycle are very scarce. Thompson (1955 a) reported no change i n l i t t e r - s i z e , and his data seem to indicate no diff e r e n c e i n midsummer pregnancy rates over the c y c l e . Unfortunately the data are presented i n such a way that no s t a t i s t i c a l assessment or d e t a i l e d comparisons may be made. His data seem to agree with what was found i n t h i s study, and even hi s own data f a i l to bear out his conclusion that reproduction proceeded at a reduced l e v e l i n summers of low population density but reached great peaks of i n t e n s i t y i n the summers of high d e n s i t i e s . There i s a l s o l i t t l e information from these other.lemming studies on the question of changes i n the age or weight at sexual maturity over the c y c l e . Nasiraovich et a l . (191*8) state that most of the summer young females do not mature i n the peak year; nothing comparable i s said about young males. Wildhagen (1953) states that both male and female Lemmus lemmus become fecund during t h e i r f i r s t summer i n the peak year, but h i s c r i t e r i o n of maturity f o r the males has been questioned by Newson (pers. comm.), and furthermore h i s samples are very discontinuous. Considering other c y c l i c microtines besides lemmings, we f i n d a close p a r a l l e l i n Kalela's (1957) study of Clethrionomys rufocanus i n F i n n i s h Lapland. From his data on the reproduction of t h i s c y c l i c vole he concluded: (1) i n peak populations nearly a l l the summer young males and some of the summer young females f a i l e d to mature; (2) a shortened summer breeding season occurred i n the peak and decline years; and (3) there was no change i n l i t t e r s i z e over the c y c l e . The s i m i l a r i t y of these r e s u l t s to those given p r e v i o u s l y f o r t h i s study i s quite impressive. 150 C h i t t y (1952) reported a shortened summer breeding season i n the peak year f o r Microtias a g r e s t i s i n England. Godfrey (1953) suggested that a delay i n reaching maturity f o r M. agrestis young may only occur i n years of peak population. Stein (1957) found no change i n l i t t e r size over the cycle f o r M. a r v a l i s and a decrease i n the percentage of young females maturing during the peak summers. Adams, B e l l , and Moore (quoted by C h r i s t i a n , 1961) found i n M. montanus that breeding ceased e a r l y i n the peak summer and apparently most of the summer young males d i d not mature e i t h e r i n the peak or the d e c l i n e . Zejda (1961) reported a shortened peak breeding season and a f a i l u r e of summer young to mature i n the peak summer f o r Clethrionomys glareolus i n C z e c h o s l o v a k i a . This series of p o s i t i v e instances suggests that cycles of the type defined previously are associated with a f a i r l y s p e c i f i c set of reproductive changes. I t i s important to look f o r negative instances to see how f a r t h i s g e n e r a l i z a t i o n holds. Hamilton (1937 a, 19Ul) reported an accelerated breeding rate, increased l i t t e r s i z e , and longer reproductive season i n increasing and peak populations of M. pennsylvanicus. No s t a t i s t i c a l data were given f o r the l i t t e r size changes so i t i s not possible to t e l l i f they are s i g n i f i c a n t . Also, Hamilton does not discount possible body weight or p a r i t y e f f e c t s and h i s increased l i t t e r sizes might be explained by the heavier animals i n h i s high populations (Hamilton, 1937 b ) . He found winter breeding only i n the peak year and no curtailment of the peak summer breeding season. His data also show an increase i n the amount of post-partum breeding i n the peak year. Hamilton's observations are at complete variance with those described above f o r lemmings, and they have never been repeated. Hoffmann (1958) studied reproduction and mortality i n M. montanus and M. c a l i f o r n i c u s . He defined the phases of the cycle i n terms of changes 151 i n f a l l population d e n s i t i e s , and a completely d i f f e r e n t pattern i s seen i f we consider changes i n h i s spring d e n s i t i e s , which C h i t t y and C h i t t y (I960 a) considered to be the i n d i c a t o r of c y c l i c phase. From t h i s point of view he has no data f o r the p e r i o d of increase i n spring d e n s i t i e s f o r e i t h e r species or f o r the period of decline f o r M. c a l i f o r n i c u s , and h i s data e s s e n t i a l l y r e f e r to populations at peak phase only. He found not "change i n age at maturity or incidence of post-partum breeding over the period studied and only minor changes i n l i t t e r size and ovulation rate. He concluded that reproductive changes were a r e l a t i v e l y minor part of the cycle and that the important changes must have been i n m o r t a l i t y . This i s the exact a n t i t h e s i s of Hamilton's conclusions, but part of t h i s apparent c o n f l i c t of views may a r i s e because Hamilton's data cover the period of increase and Hoffmann's do not. Much more c r i t i c a l data on reproduction i n r e l a t i o n to c y c l i c events i s needed. There i s c l e a r evidence from the more no r t h e r l y lemmings and voles that at l e a s t some cycles are accompanied by s t r i k i n g changes i n the length of the breeding season and age at maturity. We must now ask whether t h i s i s a u n i v e r s a l c h a r a c t e r i s t i c of these c y c l e s . Hamilton's (193? a) and Hoffmann's (1958) data suggest that i t i s not and that other patterns are p o s s i b l e . I f t h i s i s true, how and why do these patterns d i f f e r from one another? I t i s pertinent to enquire what could have been the cause of the reproductive changes observed i n t h i s study. Let us f i r s t consider e x t r i n s i c f a c t o r s . I t seems u n l i k e l y that changes i n the food supply were the d i r e c t cause of these reproductive changes. Maynard and L o o s l i (1956, p387) point out that the n u t r i t i v e requirements of breeding females are greater than those f o r males and yet i n t h i s study males were affected much more than females ( c f . Table 27), which suggests that the f a c t o r s involved are not n u t r i t i o n a l . M i l d winter weather may have been necessary f o r the extensive winter breeding 15a. to occur, but there was no evidence that any c l i m a t i c f a c t o r could have caused the midsummer breeding changes found i n the peak summer or the summer of de c l i n e . I f we t u r n to the i n t r i n s i c f a c t o r s , there i s no evidence that these reproductive changes were a function of density per se because they p e r s i s t e d i n t o the decline i n some cases and af f e c t e d the sexes d i f f e r e n t i a l l y . There i s also no evidence that these reproductive changes were caused by stress as defined by C h r i s t i a n ( 1 9 5 9 ) j and, although one can obtain reproductive changes by str e s s i n g animals, there are other ways to do t h i s as xrell (e.g. Parkes and Bruce, 1961), and we are thereby no c l o s e r to knowing what happens i n the f i e l d . Nevertheless, t h i s i s not to say that the reproductive changes observed i n t h i s study do not have a p h y s i o l o g i c a l explanation. I conclude that these reproductive changes were not caused p r i m a r i l y by e x t r i n s i c f a c t o r s or by stress or density per se, but rather were caused by some i n t r i n s i c change i n the population, probably associated with i n t r a s p e c i f i c s t r i f e . To sum up, at l e a s t some cycles are accompanied by a set of s p e c i f i c reproductive changes i n v o l v i n g winter breeding during the increase, a shortened summer breeding season at the peak, and a lack of maturation i n young males and to some extent i n young females during the peak summer. The a v a i l a b l e evidence suggests that while t h i s i s a common pattern i t may not be found i n a l l cases, and i t i s important to seek information on contrary instances such as described by Hamilton (1937 a ) . The reproductive changes described here cannot be explained by e x t r i n s i c f a c t o r s but seem to be caused by i n t r i n s i c changes i n the population. M o r t a l i t y Very l i t t l e work has been done on quantitative m o r t a l i t y assessments f o r c y c l i c microtines. This i s an important point because there i s a tendency to disregard v a r i a b l e s which have not been studied i n t e n s i v e l y , or else to p o s i t reasonable but u n v e r i f i e d explanations f o r the c y c l i c m o r t a l i t y which 153 would not be tenable i f quantitative data were a v a i l a b l e . P a r t i a l prenatal m o r t a l i t y does not seem to play a necessary part i n the c y c l e . K a l e l a (1957) reported no obvious change i n prenatal m o r t a l i t y f o r Glethrionomys rufocanus, and Hoffmann (1958) found only a s l i g h t change i n p a r t i a l prenatal m o r t a l i t y between peak and d e c l i n i n g populations of M. montanus. This agrees with the r e s u l t s of t h i s study. Information on t o t a l l i t t e r l o s s i s almost completely l a c k i n g f o r c y c l i c microtines because t h i s type of l o s s i s d i f f i c u l t to measure. I'fe may conclude that probably p a r t i a l prenatal m o r t a l i t y does not change over the cycl e , but whether there i s some change i n t o t a l l i t t e r l o s s e s , p a r t i c u l a r l y among young animals (as shown i n t h i s study), i s not yet known. T o t a l l i t t e r losses among adults are probably not s i g n i f i c a n t (Hoffmann, 1958; t h i s study). Hoffmann (1958) found that weanling and juvenile m o r t a l i t y increased considerably i n M. montanus during a decline, and he suggested that t h i s change was the key to the d e c l i n e . Godfrey (1955) found that high m o r t a l i t y of juveniles was associated with the decline of two M. a g r e s t i s populations, and juvenile male m o r t a l i t y was also high i n the peak summer. El t o n , Davis, and Findlay (1935) have recorded another instance of high juvenile m o r t a l i t y i n a decline of M. a g r e s t i s . C h i t t y (1952) found that high juvenile m o r t a l i t y was associated with peak and d e c l i n i n g populations of the same species. The r e s u l t s of t h i s study agree with those on Microtus and indicat e that a high juvenile m o r t a l i t y rate occurred at l e a s t i n a l l the d e c l i n i n g populations which showed no recovery (G). Juvenile m o r t a l i t y i n those declines which showed some recovery (H) must be l e s s than i n Type G declines, but no quantitative data are a v a i l a b l e . Some workers found high juvenile m o r t a l i t y also i n the peak summer, but there was no suggestion of 151* t h i s i n the present study, and t h i s c h a r a c t e r i s t i c may not be a constant feature of the c y c l e . C h i t t y (1952) reported increased adult m o r t a l i t y i n the spring of the decline, and the data of Godfrey (1955) suggest the same thing. This increased m o r t a l i t y however may be confined to a short period i n the spring when breeding begins ( C h i t t y and Ch i t t y , I960 a ) . Very few extensive measurements of adult m o r t a l i t y rates have been made ( L e s l i e et a l . , 1953; C h i t t y and C h i t t y , I960 a), and we must be c a r e f u l not to extrapolate too much from observations on midsummer adult m o r t a l i t y such as were made i n t h i s study. We may conclude from the above data that juvenile m o r t a l i t y changes are most important over the cycle and exceed any changes i n adult m o r t a l i t y which may occur. This i s not an unusual s i t u a t i o n , f o r studies on non-cyclic mice by Bendell (1959) and Martin (1956) also pointed to the importance of juv e n i l e m o r t a l i t y i n determining density changes, and Lack (195U) concluded that i n a l l animals the death rate i s higher i n the juve n i l e s than i n the adul t s . Migrations Migrations of lemmings have been reported from Scandinavia i n p a r t i c u l a r but also from various parts of North America. In view of the preoccupation of many people with these migrations i t may be p r o f i t a b l e to enquire how these migrations d i f f e r from the spring unrest and wandering found at Baker Lake i n I960. Thompson (1955 c) has described a brown lemming emigration at Point Barrow, Alaska that seems to resemble c l o s e l y my observations given p r e v i o u s l y . For about s i x days at the beginning of June of the peak summer, when the snow was melting and summer breeding had just begun, i n d i v i d u a l lemmings moved haphazardly through the camp and out onto the sea i c e . Only 155 a small percentage of the t o t a l population took part i n t h i s emigration. Thompson remarked that t h i s emigration was very d i f f e r e n t from the mass migrations of Scandinavian lemmings. There i s no good evidence f o r the North American a r c t i c that any other type of lemming movements occurs besides that described by Thompson. The report of Gavin (19U5) of a ten day mass migration of brown lemmings at a density of one per sq. yard i s hardly c r e d i b l e . There i s no question that lemmings do move i n d i v i d u a l l y on sea i c e , lakes, and the land during the spring melt-off i n peak years and that they may move quite long distances on i c e . There i s no question that one may see ten or f i f t e e n lemmings at a time on the bare patches of ground during the melt-off, and that sled dogs may engorge themselves on lemmings while t r a v e l l i n g across country. But i t i s a complete myth to extrapolate such events, as Gavin (lQU5) d i d , into a s o l i d mass of lemmings marching i n a p a r t i c u l a r d i r e c t i o n f o r days on end. I have been s e r i o u s l y t o l d by people at Baker Lake that during the I960 spring there were " m i l l i o n s " of lemmings marching across the tundra toward Hudson Bay, and that there were "thousands" of lemmings a l l over the lake i c e when i n f a c t fewer than 50 lemmings were a c t u a l l y seen by the persons involved. I therefore r e j e c t the suggestion that mass migrations of lemmings occur i n North America. Let us now look at the Scandinavian lemming migrations. C o l l e t t (l895» 1911j summarized by E l t o n , 19U2) has given one of the most extensive de s c r i p t i o n s of these movements. The evidence f o r migrations which he gives seems to be as follows: ( l ) lemmings are found i n the lowlands i n great numbers during some years; (2) i n d i v i d u a l lemmings may appear on c i t y s t r e e t s , swimming i n the ocean, or other abnormal places during the peak years; and (3) various observers report "migrating swarms". But i t has never been shown that lemmings do not inhabit the woodland and lowland zones as a normal habitat even i n low years, and yet t h i s i s a c r i t i c a l point regarding whether 156 or not a migration i s necessary to account f o r the presence of these animals. C o l l e t t {189$, p 17) states that one r a r e l y sees lemmings even i n the best ha b i t a t s during normal years, and yet the bulk of the evidence that lemmings do not inhabit the xiroodland and lowland zones i s that they are never seen there. Again there i s no doubt that i n d i v i d u a l leiranings do move in t o abnormal places during a peak year such as C o l l e t t describes. C o l l e t t (1895), portraying the type of movement, stat e s : "They are not sociable i n the sense of several i n d i v i d u a l s d e l i b e r a t e l y j o i n i n g company f o r long distances... Therefore they seldom, i f ever, advance i n close ranks as generally depicted i n drawings..." (p U3) and again (1911): "They migrate c h i e f l y by night, but also p a r t l y by daylight, always s i n g l y or some few near together, never i n close formation.." (trans.) I f t h i s i s true, then how do we decide when a "migration" i s occurring? There i s not a single quantitative observation on the extent of these movements. K a l e l a (19U9) states that the Norwegian lemming extended i t s range by more than 100 kilometers over three subsequent c y c l e s , but there i s no evidence why the simpler explanation of permanent low density populations i n the "invaded" areas i s not acceptable. Nasimovich et a l . (19U8) recorded the f o l l o w i n g observations on Norwegian lemmings: "In spring....the lemmings ran s i n g l y on the i c e , never forming groups, and only i n a few cases were more than three animals seen simultaneously ... On an excursion on the ice of ( a lake ) from end to end (about 18 km) 20-32 running lemmings were counted. Thus the spring migrations observed by us are f a r d i f f e r e n t from the picture of mass 'flows' described by other w r i t e r s . . . " ( t r a n s l a t i o n page 27) Recently Kalela (1961) and Koponen e t a l . (1961) have discussed lemming migrations i n northern Lapland. They d i s t i n g u i s h (1) spring migrations which go on f o r about 1 week, and (2) f a l l migrations which may go on f o r 2-3 months. Koponen et a l . (1961) have described a spring migration on lake i c e which was extremely s i m i l a r to that described above f o r t h i s study and that described by Thompson (1955 c ) . Each lemming on the i c e moved 157 independently, and only very small numbers of lemmings were involved. The beginning of these movements coincided with the s t a r t of the spring breeding season and seemed to be associated with a seasonal change of habitat. K a l e l a (1961) has described f a l l migrations associated with a seasonal change of h a bitat. Not a single animal was marked and recaptured i n t h i s work, and a l l the evidence f o r migratory movements consists of the f a c t s that ( l ) lemmings were found i n farmyards and other unusual places at the peak, and (2) no lemmings were seen i n an area at one time and at a l a t e r date some lemmings were trapped there. No one doubts that i n d i v i d u a l lemmings do wander i n t o strange places at times of peak d e n s i t i e s , and no evidence i s presented why l o c a l reproduction along with movements of several hundred meters at the most between seasonal habitats would not be enough to account f o r a l l the changes observed by K a l e l a (1961). I r e j e c t h i s claim that f a l l migrations occur i n the Norwegian lemming because the data presented have a much simpler explanation. I t i s indeed s u r p r i s i n g to f i n d that there i s no objective evidence f o r mass migrations of the Norwegian lemming. The spring "migrations" described seem to be no d i f f e r e n t from the l o c a l movements found by Thompson (1955 c) and by t h i s study. There i s no good evidence that the f a l l "migrations" are anything but l o c a l movements of i n d i v i d u a l s at high d e n s i t i e s . No oriented long distance movements of groups have ever been demonstrated. U n t i l evidence to the contrary becomes a v a i l a b l e , i t seems best to regard mass lemming migrations as a f i c t i o n and to confine our attention to the i n d i v i d u a l movements found sometimes at peak d e n s i t i e s . Thompson (1955 c) states that the mass unrest i n the spring of the peak year at Point Barrow was probably caused by changes i n available food and cover and seemed to have a very minor e f f e c t on l o c a l population d e n s i t i e s . He does not consider the f a c t that t h i s unrest marks the onset 158 of summer breeding. The sudden environmental changes associated with the melt-off may not even by a necessary cause of t h i s unrest, because the same type of s h u f f l e i s also found i n voles at the s t a r t of summer breeding ( C h i t t y and Phipps, 196l). I t would be d i f f i c u l t of course to f i n d a n a t u r a l s i t u a t i o n i n the lemmings to t e s t t h i s hypothesis that the onset of summer breeding i s a s u f f i c i e n t cause of the mass unrest observed, and so we must l e t the matter r e s t f o r the moment. To sum up, there i s no convincing evidence that mass migrations of lemmings occur e i t h e r i n North America or i n Scandinavia. The descriptions published f o r spring "migrations" can be explained as small l o c a l emigrations of i n d i v i d u a l s such as described by Thompson (1955 c ) . These l o c a l emigrations may be caused by the onset of summer breeding a c t i v i t y i n dense populations perhaps coupled with the strong environmental changes during the melt-off. The f a l l "migrations" seem to be nothing but l o c a l movements of i n d i v i d u a l s caused i n part by high d e n s i t i e s . Weather and Synchrony The problem of synchrony of cycles over large areas of country has long i n t r i g u e d workers. I do not propose to discuss any of the cosmic theories that have at one time or another been put forward to explain synchrony. Weather seems to be the only reasonable v a r i a b l e which could account f o r t h i s synchrony. Thus we must attempt to discover what e f f e c t ordinary weather phenomena — deep snow covers, warm springs, wet summers, etc. — have on reproduction and m o r t a l i t y of c y c l i c animals. As C h i t t y (1952) has pointed out, i f everything about a cycle i n numbers was explained by an i n t r a s p e c i f i c process (or, f o r that matter, by the food supply hypothesis), we would expect non-synchronous f l u c t u a t i o n s , which i s not what we observe at a l l . Thus weather must pl a y a necessary part i n the c y c l e . I t seems c l e a r at the other extreme that weather changes 159 cannot be a s u f f i c i e n t cause of the c y c l i c increase or decline because we would not normally get 3-4 year cycles i f t h i s were true (assuming there are no weather cycles of t h i s p e r i o d i c i t y ) . We conclude that weather changes must be a necessary cause of increase or decline or both i n these c y c l i c species but cannot be the e n t i r e cause. Shelford (19U3) concluded that Dicrostonyx populations at C h u r c h i l l tended to increase with average or above average snowfall which gave p r o t e c t i o n over the e n t i r e winter and with warm temperatures i n J u l y and August, and tended to decline over cold winters with l i t t l e snow. I t i s c l e a r from h i s data, however, that weather changes alone were probably not responsible f o r the increases or declines observed because some favorable winters were not accompanied by increases and at l e a s t one favorable winter was followed by a d e c l i n e . C o l l e t t (1895) pointed out that spring and summer weather had l i t t l e e f f e c t on L. lemmus populations. The suggestion, therefore, f o r lemmings i s that summer weather normally has l i t t l e or no e f f e c t on the cyc l e , but that winter weather may be a p a r t i a l cause of the increases and de c l i n e s . C h i t t y (1952, i960) has reported instances of asynchrony i n populations very close to each other and concluded from t h i s that bad weather alone was not s u f f i c i e n t to cause a decline i n Microtus a g r e s t i s populations. There are few other reports of populations i n the same l o c a l i t y f l u c t u a t i n g out of phase such as C h i t t y found. P i t e l k a (1961) reports some instances from northern Alaska of asynchrony, but i t seems c l e a r that i t i s not easy to f i n d these, and the areas involved seem to come back into phase rather q u i c k l y . P i t e l k a (1957 b) has discussed some aspects of re g i o n a l synchrony i n northern Alaska. He has concluded from the ava i l a b l e data that the short term cycle i s not a normal c h a r a c t e r i s t i c of tundra microtines everywhere, 160 and that cyclic fluctuations among several microtines i n the same area are not typically i n phase. Now i t i s probably true that very strong cycles such as occur at Point Barroxtf do not occur i n the areas more toward the interior of Alaska, but this does not mean that the same phenomenon may not be occurring there to a lesser extent. In other words, the absolute densities at the "peak" and the "low" may be very different from area to area (this i s one problem to be explained) while the same cyclic process may occur i n a l l these areas (and this i s another problem). To map the extent of cyclic "highs" by means of airplane observations on the abundance of predators and drifted winter-cut vegetation, such as was done for northern Alaska, seems to me to miss the whole point at issue. If we applied this same technique to the Canadian Barren Grounds we would conclude some very misleading things about cyclic "highs" (i.e. that the lemming cycle was confined to a very small part of the t o t a l area, i n habitats of very thick marshy vegetation), and yet the vegetation of the Foothills sector of northern Alaska (Britton, 1957) i s rather similar to that of the Barren Grounds. Even casual observations on the ground can be very misleading i n these respects. It seems premature to decide whether cycles are or are not characteristic of tundra microtines everywhere. However, we must look for instances of non-cyclic populations as i t would be most interesting to compare this type of population with a normal cyclic population. A second problem regarding synchrony treated by Pitelka (1957 b) i s whether sympatric microtines cycle in phase. Elton (191+2, p U39) stated that both species of lemmings probably fluctuated i n phase. Watson (1957) believed that when Lemmus and Dicrostonyx were sympatric they tended to fluctuate i n phase, although this synchrony was never exact. The data given i n this paper support this belief. No intensive work has yet been done on areas where three or more cyclic microtines commonly occur. 161 The role of weather i n cyclic fluctuations remains very poorly understood, and this generalization probably applies to almost a l l animal populations (Andrewartha and Birch, 195U). We cannot study a natural population i n the absence of weather and we have not yet learned to set up laboratory populations which are comparable to f i e l d populations. It i s certainly possible to ascribe almost a l l population-changes to weather changes by ad hoc hypotheses (e.g. Schindler, l ° 6 o ) , but this hardly furthers our understanding of the changes. To sum up, weather must be regarded as a necessary cause of the cyclic increase or decline because (l) i f weather was not necessary, cycles would not tend to be synchronous, and (2) i f weather was sufficient to cause the increase or decline, cycles would not tend to be 3-h years i n length. To date the only plausible explanation of this role of weather i s that of Chitty (1952, 1955 b, I960) which i s illustrated i n a model by Leslie ( 1 9 5 9 ) . Predators Very few workers today support the idea that the cycle i s caused by predators (Lack, W9$hy p 213; Pitelka et a l . , 1 9 5 5 ) . There i s no doubt that under certain conditions predators do k i l l many lemmings, and Pitelka (1959) believes that they may dampen the fluctuations of the lemmings at least in northern Alaska. A sharp spring decline i n Lemmus occurred in i 9 6 0 i n this study i n the vir t u a l absence of predators; similar spring declines were described by Thompson (1955 a) and Pitelka (1957 a) and attributed to predators. While i t i s reasonable to suppose that predation might alter the length and pattern of the cycle, mere association of events must be viewed c r i t i c a l l y . Disease and Parasites Elton (19U2, p 201) and Chitty (195U, I960) have shown that disease cannot be regarded as a sufficient condition for a decline in numbers. Disease 162 i s believed to be a l o c a l f a c t o r of v a r i a b l e i n t e n s i t y and occurrence and not an e s s e n t i a l part of the c y c l i c process. Nothing from t h i s study opposes these ideas. Body Weight Changes High body weights i n the peak summer have been described!, by C h i t t y (19$2) f o r Microtus a g r e s t i s , Zimmermann (19$$) and Stein (1957) f o r M. a r v a l i s , Thompson (1955 a) f o r Lemmus trimucronatus, K a l e l a (1957) f o r Clethrionomys rufocanus. S t e i n (1956) and Zejda (196l) f o r C. glareolus, and by t h i s study f o r Lemmus trimucronatus and Dicrostonyx groenlandicus. I t i s important to enquire why t h i s weight change occurs. A change i n mean body weight of adults may be produced i n two general ways: ( l ) by a change i n the growth rate of i n d i v i d u a l s ; or (2) by a change i n the normal age c l a s s structure of the population. The f i r s t would be a r e a l e f f e c t , the second a s t a t i s t i c a l e f f e c t . Zimmermann (1955) believed that these weight changes d i d not represent mere changes i n the proportions of the age classes but were p a r t l y caused by changes i n the growth rate of i n d i v i d u a l s . He states that probably the same e x t r i n s i c f a c t o r s cause the density changes and the growth changes (e.g. favorable weather). Stein (1956) found that the lower age groups were missing from peak and d e c l i n i n g populations of Clethrionomys glareolus i n Germany and t h i s caused the mean body weight i n the spring to be greater i n peak populations. He believed that these changes were not produced by e x t r i n s i c v a r i a b l e s which caused a change i n growth rate, but rather that they were due to a s e l e c t i v e e l i m i n a t i o n of the younger animals by some form of i n t r a s p e c i f i c s t r i f e ( i . e . that the e f f e c t was s t a t i s t i c a l ) . Zejda (1961) offered a d i f f e r e n t i n t e r p r e t a t i o n of Stein's r e s u l t s based on h i s won work on C. glareolus, i . e . that these lower age groups were missing because of the shortened reproductive season i n the peak year. 163 Thus populations descended from normal spring to f a l l breeding seasons would have on the average normal body weights, but those descended from c u r t a i l e d summer breeding seasons would include only the l a r g e r spring animals and not the smaller f a l l animals, and those from which the spring l i t t e r s are eliminated would produce on the average below-normal size animals. I t i s d i f f i c u l t to reconcile any of these hypotheses with the r e s u l t s found i n t h i s study. Presumably lemmings born during the winter should have the lowest growth rates, and yet i t was these animals that formed the bulk of the high weight adults of summer 1°60. As we have seen, adult lemmings do not grow much a f t e r the f i r s t week of June. Thus we have the anomalous s i t u a t i o n i n which the 1°60 adult animals were produced at a time of the year when there i s no vegetative growth and had reached t h e i r high weights before any summer plant growth began. Furthermore, a 3-4 year c y c l i c a l change of e x t r i n s i c f a c t o r s would be needed to v e r i f y Zimmermann's hypothesis f o r c y c l i c a l species. These considerations seem t o r u l e out Zimmermann's explanation as s u f f i c i e n t . Also, the c u r t a i l e d summer breeding season of 1°60 produced small adults the following year on some areas (Type G decline) and large adults on other areas (Type H d e c l i n e s ) . This seems to rule out the i n t e r e s t i n g hypothesis of Zejda. F i n a l l y , the hypothesis of S t e i n i s excluded because the youngest adults were also the l a r g e s t (winter generation I960), and the greatest amount of s t r i f e seemed to produce smaller animals (winter generation 1961) not l a r g e r ones. Furthermore, the large animals of the peak lemming populations are bigger animals i n every way than those from d e c l i n i n g or low populations. These large animals occupy weight classes which are not even approached i n the low years, and thus t h i s change represents more than a s t a t i s t i c a l change of proportions w i t h i n c e r t a i n size groups. 16U C h i t t y and C h i t t y (I960 b) found that weather d i f f e r e n c e s could not account f o r the d i f f e r e n c e s i n growth since opposite e f f e c t s were observed on two areas subject to the same weather w i t h i n one season. They also found that high population density at the time of poor growth was not an adequate explanation. There i s no i n d i c a t i o n that age d i f f e r e n c e s are responsible f o r the observed d i f f e r e n c e s ; indeed i n the present study the low weight animals of the Type G decl i n e s were at l e a s t 3-4 months older on the average than the high weight animals of the peak. S t e i n (1956) seems to have been the f i r s t to postulate that t h i s s i z e change associated with density changes might involve genetic changes i n the population. Newson and C h i t t y (1962) found that some voles from d e c l i n i n g populations would grow i f brought i n t o the lab but none grew i n the f i e l d . This demonstrated that the i n t r i n s i c condition of the animals was not a s u f f i c i e n t explanation of low body weights during the decline, and hence that some environmental v a r i a b l e must be involved, probably some aspect of behavior. The conclusion which emerges from t h i s discussion i s that the weight changes associated with c y c l i c f l u c t u a t i o n s represent a change i n the growth rates of i n d i v i d u a l s and thus a change i n population q u a l i t y . Me do not know why growth rates should change over the c y c l e . One way to change the growth rate of laboratory animals i s to modify the d i e t (e.g. Osborne and Mendel, 1926), but i t i s possible to change growth rates i n other ways as w e l l (e.g. MacArthur, 1949; Crowcroft and Rowe, 196l) and so we cannot conclude what f a c t o r s are n e c e s s a r i l y involved i n these weight changes u n t i l f u r t h e r study i s made. I t i s possible that these body weight changes discussed here are not a single class of events but are produced by several d i f f e r e n t f a c t o r s . But, u n t i l we have evidence to the contrary, I believe that we should look f o r a common explanation. 165 In summary, we know that high body weights are associated with peak populations of several d i f f e r e n t types of microtines. ¥e do not understand why animals do not grow during c e r t a i n phases of the c y c l e , except that these differences seem to be a r e s u l t of i n t r a s p e c i f i c i n t e r a c t i o n s . The relevant change i s i n the growth rates of i n d i v i d u a l animals, and genetic changes may be involved. Three Current Hypotheses I would now l i k e to consider three current hypotheses which attempt to e x p lain these c y c l i c a l f l u c t u a t i o n s . (1) Food Supply Hypothesis; The hypothesis supported by P i t e l k a from h i s work and that of Thompson at Point Barrow, Alaska on the brown lemming i s shown i n Figure 8. The essence of the change involved i n the decline i s a q u a l i t a t i v e and q u a n t i t a t i v e change i n the forage; predation does not seem to be an e s s e n t i a l p art of the decline ( P i t e l k a e t a l . , 1955)• We may enquire whether t h i s hypothesis f i t s the observations of t h i s study. F i r s t , there was no extensive forage u t i l i z a t i o n at Baker Lake and t h i s would seem to c r i p p l e t h i s hypothesis at the s t a r t (Table U?)« Second, there was no evidence of st a r v a t i o n i n animals a l i v e i n the spring of the decline (Tables 55-58). However, t h i s i s apparently not an objection to the hypothesis because Thompson (1955 a) reported no evidence of malnut r i t i o n i n Point Barrow lemmings i n the spring of the decline e i t h e r . We are l e f t with the q u a l i t a t i v e forage change as the supposed cause of the d e c l i n e . Yet there was no evidence of d e f i c i e n c y diseases i n the young of the decline; indeed, the whole d i f f i c u l t y i s to account f o r the l o s s of very normal looking young. Thus the d e f i c i e n c i e s must be such that they are not noticeable macroscopically. They must prevent the young males from maturing i n the decline but allow the young females to mature. Furthermore, they must 166 FIGURE 8. P i t e l k a * s food supply hypothesis. Peak numbers Very extensive forage u t i l i z a t i o n Quantitative food d e f i c i e n c y ("starvation") Q u a l i t a t i v e food d e f i c i e n c y ( i . e . phosphate d e f i c i e n c y ?) 'Weakening of animals (e.g. decline i n reproductive rate) I Decrease i n population density Depletion of cover Predation Cessation of heavy forage u t i l i z a t i o n 1 Q u a l i t a t i v e and quantitative recovery of vegetation increase i n population density peak numbers e t c . 167 account f o r an increasing s u r v i v a l rate i n the l a t e r summer l i t t e r s of the decline compared to the f i r s t l i t t e r . Such e f f e c t s seem highly u n l i k e l y to be the r e s u l t of q u a l i t a t i v e forage changes. For these reasons I r e j e c t the food supply hypothesis as an adequate explanation of the Baker Lake lemming c y c l e . The events to be explained are mainly i n t r i n s i c changes i n v o l v i n g both reproduction and mo r t a l i t y and are of such a general nature that ad hoc hypotheses regarding e x t r i n s i c f a c t o r s i n l o c a l s i t u a t i o n s are e n t i r e l y i n s u f f i c i e n t . Yet the same e f f e c t of the lemmings on the standing crop of forage as were reported by Thompson (1955 b) at Point Barrow were also found i n t h i s study. There i s no doubt that lemmings exert a strong e f f e c t on the vegetation but t h i s i s hardly evidence f o r the above hypothesis. A l l the evidence f o r t h i s hypothesis consists of an observed association between lemming declines and extensive forage u t i l i z a t i o n (Thompson, 1955 b; P i t e l k a , 1957 a), and u n t i l more conclusive evidence i s a v a i l a b l e i t i s necessary to remain s k e p t i c a l of t h i s i n t e r p r e t a t i o n . Rausch (1950) states that there was nothing to indic a t e that the decline i n numbers i n 19U9 at Point Barrow r e s u l t e d from s t a r v a t i o n . C h i t t y ( 1 9 5 2 , I960) has presented h i s reasons f o r r e j e c t i n g the food supply hypothesis as an explanation of Microtus a g r e s t i s c y c l e s . K a l e l a (1957) came to the same conclusion f o r Clethrionomys rufocanus. Nasimovich et a l . (19U8) state that the food supply was not responsible f o r Lemmus lemmus f l u c t u a t i o n s . (2) C h r i s t i a n ' s Stress Hypothesis: The idea that cycles were caused by stre s s and that declines could be associated with changes i n a d r e n a l - p i t u i t a r y functions and shock disease was proposed by C h r i s t i a n (1950) from the basic work of Selye (19U6). The basic hypothesis has not changed much since then, with the exception of the added e f f e c t s of stress on l a t e r generations ( C h r i s t i a n and Lamunyan, 1 9 5 7 ) , 168 and Figure 9 outlines the stress hypothesis summarized i n Christian (1961). A long series of papers presents the evidence for this idea (Christian, 1955 a, 1955 b, 1956, 1957, 1959, 1961, and others). Me must distinguish a general and a specific aspect of Christian's ideas. His general thesis i s that a l l mammals limit their own densities by a combination of behavioral and physiological changes. His specific thesis i s that the mechanism of this limitation involves the General Adaptation Syndrome and i s purely phenotypic. I shall not discuss the applicability of this scheme to a l l mammals, but w i l l limit my discussion to the cyclic rodents. There are two conditions which must be f u l f i l l e d to verify this specific hypothesis: (1) there must be increased adrenal ac t i v i t y and decreased reproductive act i v i t y at high densities; (2) this increased adrenal activity must cause an increased death rate. It i s not sufficient merely to find increased adrenal activity at high densities and to claim that the hypothesis has been confirmed. Christian has amassed a large amount of data to support his hypothesis. The sheer bulk of data from at least partly controlled laboratory studies i s considered by some to be the strongest point of this hypothesis, but Chitty ( I 9 6 0 ) considers this the weakest point. There i s no evidence that these laboratory situations correspond to anything that goes on i n nature, and thus the extrapolation from the lab to the f i e l d i s not j u s t i f i e d . No consistent relationship between summer adrenal weights and the phases of the cycle was found in this study (Tables 55-58), and Chitty (196l) reported the same result from Microtus agrestis. Data from M. montanus given by Christian (196l) shows no relationship between adrenal weights and 169 FIGURE 9. C h r i s t i a n ' s s t r e s s hypothesis. The system i s purely phenotypic and operates through the general adaptation syndrome. Increase i n population density Increased s o c i a l pressure Stress (General Adaptation Syndrome) Decrease i n population density v Decrease i n s o c i a l pressure v Increase i n population d e n s i t y V e t c . 170 population s i z e , contrary to what C h r i s t i a n says. Given these data, we seem to have two choices. We can r e j e c t C h r i s t i a n ' s hypothesis, or we can save the hypothesis by saying that adrenal weights are not always a v a l i d index of the General Adaptation Syndrome. I f we accept the second a l t e r n a t i v e we must also question the majority of the evidence i n favor of the hypothesis, since i t i s mostly based oh adrenal weights. Neither horn of the dilemma i s very favorable to the current hypothesis. Munday (l Q6l) has c r i t i c a l l y reviewed the evidence that stress may explain c y c l i c declines and has concluded that there i s as yet no evidence that normal stressors can induce disease i n e i t h e r normal i n d i v i d u a l s or succeeding generations. Turner (i960, p 265) has concluded much the same t h i n g . C h i t t y (1959) has shed considerable doubt on the existence of shock disease i n nature. There i s l i t t l e conclusive evidence of any c o r r e l a t i o n i n n a t u r a l populations between adrenal hypertrophy and a regression of reproductive function, and f i n a l l y the idea that stress has an e f f e c t on subsequent generations has received l i t t l e support (Munday, 196l). To sum up, there i s no evidence from adrenal weights that stress (sensu Selye) played any major role i n t h i s lemming cycle and thus C h r i s t i a n ' s s p e c i f i c hypothesis was rejected as an explanation. (3) Chitty's Hypothesis; C h i t t y (1952) found that i n t r a s p e c i f i c s t r i f e during the peak summer produced l i t t l e e f f e c t on the adults but rather the progeny of these animals appeared t o be le s s v i a b l e . He emphasized the i n d i r e c t e f f e c t on the progeny rather than the d i r e c t e f f e c t s on the adults, and pointed out that C h r i s t i a n ' s views (1950) could not explain the long continued declines which may occur (Chitty, 1955 a ) . C h i t t y (1952, 1957, I960) proposed the following concept: that mutual antagonism associated with high breeding d e n s i t i e s brings about a change i n the p r o p e r t i e s of the contemporary population, and of the subsequent 171 generations, which become l e s s r e s i s t a n t to the normal sources of m o r t a l i t y . I t i s important to d i s t i n g u i s h t h i s concept of a change i n q u a l i t y of the population from the explanation (mechanism) of t h i s concept (Conant, 1951, p 106). C h i t t y (l°60) reviewed the evidence f o r t h i s concept and concluded that there was no evidence against i t , although a mechanism had not yet been demonstrated. C h r i s t i a n ' s (l°6l) general ideas have much i n common with t h i s concept. The relevant changes produced by mutual antagonism might involve two mechanisms: ( l ) changes i n maternal ptrysiology which are transmitted to the o f f s p r i n g ( i . e . s i m i l a r to the stress hypothesis of C h r i s t i a n ) ; or (2) changes i n the genetic composition of the population by s e l e c t i o n . The f i r s t explanation was t e s t e d extensively i n the laboratory by C h i t t y and r e j e c t e d as an adequate explanation because, although s t r i k i n g e f f e c t s could be produced i n the adults by mutual antagonism, t h e i r o f f s p r i n g d i d not show the changes i n q u a l i t y found i n n a t u r a l populations (Chitty, 1957, i960). A t t e n t i o n was thus turned to the second p o s s i b i l i t y , genetic changes. The f i r s t p o s s i b i l i t y i n v e s t i g a t e d was hereditary splenic anemia (Dawson, 1956; C h i t t y , 1957)- This has been r e j e c t e d as an explanation of the r e c u r r i n g declines by Newson and C h i t t y (1962). C h i t t y has thus modified h i s views regarding the mechanism involved, while r e t a i n i n g the primacy of mutual interference as a necessary agent i n these d e c l i n e s . His current view on the mechanism of the cycle i s shown i n Figure 10 ( C h i t t y , pers. comm.). Evidence from t h i s study f u l l y supports the general concept that populations change i n q u a l i t y during changes i n abundance. Peak populations showed these q u a l i t a t i v e differences by (1) high body weights and (2) reproductive changes which c a r r i e d over i n t o the d e c l i n e . The high j u v e n i l e m o r t a l i t y during the decline could not be predicted from the population density 172 FIGURE 10. C h i t t y 1 s hypothesis. The system i s p a r t l y genetic and p r i m a r i l y behavioral. Increase i n numbers Low Numbers V No interference Increase i n numbers e t c . 173 at the time. F i n a l l y , the d i f f e r e n t types of declines could not be e x p l i t o a d by di f f e r e n c e s i n e x t r i n s i c f a c t o r s . There i s no d i r e c t evidence from t h i s study to t e s t the mechanism proposed by C h i t t y (Figure 1 0 ) . We have seen that a considerable amount of wounding occurs i n these lamming populations (Tables H > 9 - 6 0 ) . Adult males may range over very large areas (Tables U2-U5) and the most probably hypothesis about the disappearing young of the decline i s that these adult males k i l l them. None of these points i s good evidence f o r t h i s mechanism, and nei t h e r are they good evidence against t h i s mechanism. Bir c h ( i 9 6 0 ) has discussed the f a c t that n a t u r a l s e l e c t i o n acts to bring the act u a l rate of increase r to a maximum. I f t h i s i s the case, we may assume that i n peak and d e c l i n i n g lemming populations there i s some s u r v i v a l value i n a f a i l u r e to mature ( f o r males at l e a s t ) . I t i s very important t o determine whether t h i s change i n maturation i s genotypic or phenotypic; as yet we do not know. In e i t h e r case, i t would seem possible that a very high rate of s e l e c t i o n against the early-maturing young could occur because of .increased f i g h t i n g associated with sexual maturation, and we could thereby get a complete change i n q u a l i t y of the population over a very short period of time at high d e n s i t i e s . I f the maturation change were phenotypic, i t would be secondary i n importance to aggressive behavior. On the other hand, i f t h i s change x i r e r e genotypic, i t could be of primary importance i n the c y c l e . One of the i n t e r e s t i n g points that has come from t h i s lemming work i s the s i m i l a r i t y between t h i s lemming cycle and the cycles i n Microtus a g r e s t i s described i n d e t a i l by C h i t t y . That s i m i l a r types of events should occur i n two such d i f f e r e n t e c o l o g i c a l s i t u a t i o n s argues 17U quite strongly f o r a u n i f i e d view of c y c l i c processes. To sum up, C h i t t y ' s concept of a change i n q u a l i t y of the population during changes i n density i s f u l l y supported by t h i s study. His view that the mechanism involved i s behavioral and genetic i s not refuted by my data which suggest that behavioral changes may constitute the crux of the lemming c y c l e . Conclusions I have attempted to give a semi-complete d e s c r i p t i o n of a single lemming cycle, and with t h i s s i n g l e observation on a very complex na t u r a l event have attempted to examine the current ideas on population c y c l e s . The wider our horizon of f a c t s has become, the l e s s and l e s s adequate seem the conventional ideas. As long as we s t i c k to small parts of l o c a l problems and seek only confirmatory evidence we s h a l l be content with ad hoc explanations and conventional ideas. We have t r i e d to penetrate to the core of the phenomenon studied. We have seen that e x t r i n s i c f a c t o r s could not explain the c y c l e . Of the i n t r i n s i c f a c t o r s we discarded the purely p h y s i o l o g i c a l ideas because the i n t e r a c t i o n s of i n d i v i d u a l s which could produce p h y s i o l o g i c a l changes were severe enough that a t t e n t i o n was turned d i r e c t l y to the underlying behavior and possible s e l e c t i v e forces that might r e s u l t . Future work on the mechanism of c y c l i c a l f l u c t u a t i o n s should consider the r o l e of behavior i n f a r more d e t a i l than has been done i n the past. The suggestion of C h i t t y that these f l u c t u a t i o n s may represent a genetic polymorphism deserves considerable a t t e n t i o n . The problem remains f a r from being solved. SUMMARY 1. A three year study covering one cycle i n numbers of the brown lemming (Lemmus trimucronatus) and the varying lemming (Dicrostonyx groenlandicus) has been c a r r i e d out at Baker Lake, N.W.T. i n an attempt to describe a lemming cycle from the Canadian Barren Grounds and to see what explanations would f i t the observed events. 2. Increase began from very low numbers i n 1959 and tremendous population growth occurred over the winter of 1959-60. L i t t l e f u r t h e r increase occurred i n the peak summer of i960. A great decline occurred over the winter of I96O-6I, and t h i s decline continued through the summer of 1961 on the Main Study Area. This cycle was synchronous i n both species over a wide zone of the c e n t r a l a r c t i c . 3. Two major changes i n reproduction occurred over the c y c l e . A lengthened summer breeding season and winter breeding occurred during the increase i n 1959-60, but no winter breeding and a shortened summer breeding season characterized the peak and de c l i n e . Young male Lemmus d i d not mature i n e i t h e r the peak or decline summers, nor d i d young females i n the peak. No changes i n midsummer pregnancy rates or l i t t e r size occurred. k» P a r t i a l p r e n a t a l m o r t a l i t y d i d not change over the c y c l e . Adult m o r t a l i t y •may have been s l i g h t l y higher i n the summer of decline than i n the peak summer. Juvenile m o r t a l i t y was very high i n the summer of decline, p a r t i c u l a r l y f o r the f i r s t summer l i t t e r . 5. Spring movements of i n d i v i d u a l lemmings on the ice were found i n the peak year. The existence of mass lemming migrations i s questioned both f o r North America and Scandinavia. There i s no good evidence of any oriented long distance group movements of lemmings. 6, Favorable f a l l and winter weather was associated with the increase, and unfavorable f a l l and e a r l y winter weather was associated with the declineo 7. Avian predators were uncommon throughout the cyc l e . The weasel was the only common mammalian predator but could not have accounted f o r the observed m o r t a l i t y changes. Diseases and p a r a s i t e s d i d not seem to p l a y any s i g n i f i c a n t r o l e i n the c y c l e . 8. Lemmings reduced the forage crop by about 1$% i n the peak and de c l i n e . Forage u t i l i z a t i o n averaged 30% or l e s s i n the wet habitats and was n e g l i g i b l e i n the dry habitats a f t e r the c r i t i c a l winter of 1960-61. There was no evidence of quantitative food shortage nor any suggestion of d e f i c i e n c i e s i n food q u a l i t y over the cyc l e . Lemmings i n the spring of the decline were as f a t as usual. 9. High body weights (20-30% above 'normal') were associated with a l l peak populations. 10. Organ weights (adrenals, spleen) d i d not give any clue to what was causing the c y c l e . Summer adrenal weights showed no consistent r e l a t i o n s h i p to the density changes. 11. I n t r a s p e c i f i c s t r i f e , as measured by wounds on skins, showed strong seasonal and ye a r l y changes which were not a simple f u n c t i o n of density. 12. Three current hypotheses were considered i n the l i g h t of these data. The Food Supply Hypothesis of P i t e l k a was rej e c t e d as an adequate explanation. The Stress Hypothesis of C h r i s t i a n was also r e j e c t e d . Chitty's general concept that populations change i n q u a l i t y during changes i n density i s supported by t h i s study. C h i t t y ' s s p e c i f i c explanation that these cycles represent a genetic polymorphism i n v o l v i n g aggressive behavior was not refuted by these data, which suggest that behavioral changes may constitute the crux of the c y c l e . LITERATURE CITED Andrewartha, H.G. and L.C. B i r c h . 19$k» The d i s t r i b u t i o n and abundance of animals. Univ. of Chicago Press, Chicago. 782 pp. Baker, J.R. and D. C h i t t y . (ms). The incidence of protozoan blood p a r a s i t e s i n a w i l d population of Microtus a g r e s t i s i n England, unpublished ms. Bandy, P.J., I . McT. Cowan, W.D. K i t t s , and A.J. Wood. 1956. A method f o r the assessment of the n u t r i t i o n a l status of w i l d ungulates. Can. J . Zool. 3U« U 8 - 5 2 . Bendell, J.F. 1 9 5 9 . Food as a c o n t r o l of a population of white-footed mice, Peromyscus leucopus noveboracensis ( F i s c h e r ) . Can. J . Zool. 37: 173 - 2 0 9 . B i r c h , L.C. I960. The genetic f a c t o r i n population ecology. Amer. Nat. 9U: 5-24. Bodenheimer, F.S. 1 9 3 8 . Problems of animal ecology. Clarendon Press, Oxford. Brambell, F.W.R. and I.H. M i l l s . 1947• Studies on s t e r i l i t y and prenatal m o r t a l i t y i n w i l d r a b b i t s . Part I I I . The loss of ova before implantation. J . Exp. B i o l . 21+: 1 9 2 - 2 1 0 . Brambell, F.W.R. and I.H. M i l l s . I948. Studies on s t e r i l i t y and pr e n a t a l m o r t a l i t y i n w i l d r a b b i t s . Part IV. The l o s s of embryos a f t e r implantation. J . Exp. B i o l . 2 5 : 21+1-269. B r i t t o n , M.E. 1957. Vegetation of the a r c t i c tundra, pp. 26-61 i n A r c t i c Biology, Biology Colloquium, Oregon State College, 1957. Brown, L.E. 1 9 5 6 . Movements of some B r i t i s h small mammals. J . Anim. E c o l . 25:- 54-71. Calhoun, J.B. 19U8. North American Census of Small Mammals. Release No. 1. Announcement of Program. Rodent Ecology Project, Johns Hopkins Univ., Baltimore. 9 pp. mimeo. Chester Jones, I. 1957. The adrenal cortex. U n i v e r s i t y Press, Cambridge. 316 pp. Ch i t t y , D. 1 9 5 2 . M o r t a l i t y among voles (Microtus agrestis) at Lake Vyrnwy, Montgomeryshire i n 1 9 3 6 - 9 . P h i l . Trans. Roy. Soc. London, Ser. B, 236: 5 0 5 - 5 5 2 . C h i t t y , D. 1 9 5 U « Tuberculosis among w i l d voles: with a discussion of other p a t h o l o g i c a l conditions among c e r t a i n mammals and b i r d s . Ecology 3 5 : 227-237. C h i t t y , D. 1955 a. Adverse e f f e c t s of population density upon the v i a b i l i t y of l a t e r generations, pp. 57-67 i n the Numbers of Man and Animals, ed. by J.B. Cragg and N.W. P i r i e . O l i v e r and Boyd Ltd., Edinburgh. 178 C h i t t y , D. 1955 b. Allgemeine GedandengSnge liber die Dichteschwankungen be i der Erdmaus (Microtus a g r e s t i s ) . Z e i t . Saugetierkunde 20: 55-60. Ch i t t y , D. 1957. S e l f - r e g u l a t i o n of numbers through changes i n v i a b i l i t y . Cold Spring Harbor Symp. Quant. B i o l . 22: 277-280. Chi t t y , D. 1959. A note on shock disease. Ecology 1+0: 728-731. C h i t t y , D. i960. Population processes i n the vole and t h e i r relevance to general theory. Can. J . Zool. 38: 99-113. Ch i t t y , D. and H. C h i t t y . I960 a. Population trends among the voles at Lake Vyrnwy, 1932-60. Symposium theriologicum,Brno, i960. C h i t t y , D. and E. Phipps. I960. The e f f e c t of f l e a s on spleen si z e i n vole s . J . P h y s i o l . 151: 27-28. C h i t t y , D. and E. Phipps. 1961. A d e c l i n i n g vole population. J . Anim. E c o l . 30: 1+90. Ch i t t y , H. 1961. Variations i n the weight of the adrenal glands of the f i e l d vole, Microtus a g r e s t i s . J . Endocrin. 22: 387-393. Ch i t t y , H. and D. C h i t t y . i960 b. Body weight i n r e l a t i o n to population phase i n Microtus a g r e s t i s . Symposium theriologicum, Brno, I960. C h r i s t i a n , J . J . 1950. The adreno-pituitary system and population cycles i n mammals. J . Mammal. 31: 21+7-259. C h r i s t i a n , J . J . 1955 a. E f f e c t of population size on the adrenal glands and reproductive organs of male mice i n populations of f i x e d s i z e . Amer. J . P h y s i o l . 182: 292-300. C h r i s t i a n , J . J . 1955 b. E f f e c t of population size on the weights of the reproductive organs of male white mice. Amer. J . P h y s i o l . 181: 1+77-1+80. C h r i s t i a n , J . J . 1956. Adrenal and reproductive responses to population size i n mice from f r e e l y growing populations. Ecology 37: 258-273. C h r i s t i a n , J . J . 1957. A review of the endocrine responses i n r a t s and mice to increasing population size i n c l u d i n g delayed e f f e c t s on of f s p r i n g . Naval Med. Res. Inst., Lect. Rev. Ser. No. 57-2: 1+1+3-1+62. C h r i s t i a n , J . J . 1959. The r o l e s of endocrine and behavioral f a c t o r s i n the growth of mammalian populations, pp. 71-97 of Comparative Endocrinology, ed. by A. Gorbman. J . Wiley and Sons, Inc., New York. C h r i s t i a n , J . J . 1961. Phenomena associated with population density. Proc. Nat. Acad. S c i . 1+7: 1+28-1+1+9. C h r i s t i a n , J . J . and D.E. Davis. 1956. The r e l a t i o n s h i p between adrenal weight and population status of urban Norway r a t s . J . Mammal. 37: 1+75-U86. 179 C h r i s t i a n , J.J. and C D . Lemunyan. 1957. Adverse e f f e c t s of crowding on reproduction and l a c t a t i o n of mice and two generations of t h e i r progeny. Naval Med. Res. Inst. Res. Per. V o l . 15: 925-936. Cole, L.C. 1 9 5 1 . Population cycles and random o s c i l l a t i o n s . J. W i l d l . Mgmt. 15s 233 - 2 5 2 . Cole, L.C. 195U a. The population consequences of l i f e h i s t o r y phenomena. Quart. Rev. B i o l . 2 9 : 103-137. Cole, L.C, 195U b. Some features of random population c y c l e s . J. W i l d l . Mgmt. 18: 2-21; Cole, L.C. 1 9 5 8 . Population f l u c t u a t i o n s . Proc. Tenth Inte r . Cong. Ent. v o l . 2 , 1956: 639-6U7. C o l l e t t , R. 1 8 9 5 . Myodes lemmus, i t s habits and migrations i n Norway. C h r i s t i a n i a Videnskabs-Selekabs Forhandlinger 1895 No. 3. 62 pp. C o l l e t t , R. 1911. Norges Pattedyr. K r i s t i a n i a . 7hh pp. Conant, J.B. 1 9 5 l « On understanding science. Mentor Books, New York.' lUU pp. Conaway, C H . 1955* Embryo resorption and p l a c e n t a l scar formation i n the r a t . J. Mammal. 36: 5 1 6 - 5 3 2 . Crowcroft, P. and F.P. Rowe. 1961. The weights of w i l d house mice Mus Musculus l i v i n g i n confined colonies. Proc. Zool. Soc. London 136 (2) : 177 - 1 8 5 . Curry-Lindahl, K. 1961. Fjallammeln (Lemmus lemmus) i SVerige under i 9 6 0 . Fakta och t e o r i e r . Fauna och F l o r a 55. T-27. Davis, D.E. 1 9 5 3 . Analysis of home range from recapture data. J. Mammal. 31*: 3 5 2 - 3 5 8 . Dawson, J. 1 9 5 6 . Splenic hypertrophy i n voles. Nature 1?8: 1183-1181;. Dunaeva, T.N. and V.V. Kucheruk. 1 9 u l . M a t e r i a l on the ecology of the t e r r e s t r i a l vertebrates of the tundra of South Yamal. M a t e r i a l on the fauna and f l o r a of the U.S.S.R. No. k (19), pp. 1-80. ( i n Russian) 1921;. Periodic f l u c t u a t i o n s i n the numbers of animals: t h e i r causes and e f f e c t s . B r i t . J. Exp. B i o l . 2 : 1 1 9 - 1 6 3 . 1 9 3 1 . The study of epidemic diseases among w i l d animals. J. Hyg. 3 1 : U35-Ji56. I9I4.2. Voles, mice and lemmings. Clarendon Press, Oxford. 1|96 pp. D.H.S. Davis, and CM. Fi n d l a y . 1 9 3 5 . An epidemic among voles (Microtus agrestis) on the S c o t t i s h border i n the spring of 1931).. J. Anim. E c o l . 4: 277-288 E l t o n , C. E l t o n , C. El t o n , C. E l t o n , C. 180 Emlen, J.T. J r . , R.L. Hine, W.A. F u l l e r , and P. Alfonso. 1957- Dropping boards f o r population studies of small mammals. J . W i l d l . Mgmt. 2 1 : 300 - 3 H * . Enzmann, E.V. 1 9 3 5 . Intrauterine growth of albino mice i n normal and i n delayed pregnancy. Anat. Rec. 62: 31-1*5* Frank, F. and K. Zimmermann. 1957* Uber die Beziehungen zwischen Lebensalter und morphologischen Merkmalen b e i der Feldmaus, Microtus  a r v a l i s ( P a l l a s ) . Zool. Jahr. Syst. 8 5 * 283-300. Gavin, A. 191*5. Notes on mammals observed i n the Perry River d i s t r i c t , Queen Maud Sea. J . Mammal. 2 6 : 2 2 6 - 2 3 0 . Godfrey, G.K. 1953. Factors a f f e c t i n g the s u r v i v a l , movements, and i n t r a s p e c i f i c r e l a t i o n s during e a r l y l i f e i n populations of small mammals with p a r t i c u l a r reference to the v o l e , unpublished D. P h i l , t h e s i s , Bureau of Animal Population, Oxford. 127 pp. Godfrey, G.K. 1955* Observations on the nature of the decline i n numbers of two Microtus populations. J . Mammal. 3 6 : 209-211*. Golley, F.B. 1 9 6 1 . Interaction of n a t a l i t y , m o r t a l i t y and movement during one annual cycle i n a Microtus population. Amer. Mid i . Nat. 6 6 : 1 5 2 - 1 5 9 . Hamilton, W.J. J r . 1937 a. The biology of microtine c y c l e s . J . A g r i c . Res. 51*: 7 7 9 - 7 9 0 . Hamilton, W.J. J r . 1937 b. Growth and l i f e span of the f i e l d mouse. Amer. Nat. 71: 5 0 0 - 5 0 7 . Hamilton, W.J. J r . 19l*l. Reproduction of the f i e l d mouse Microtus pennsylvanicus (Ord). C o r n e l l Univ. A g r i c . Exp. Sta. Mem. 2 3 7 . 23 pp. Henneberg, B. 1937* Normentafeln zur Entwicklungsgeschichte der Wanderratte (Rattus norvegicus Erxleben). Normentaf. Wirbe l t . No. 1 5 , 160 pp. Jena. H i l l , A.B. 1959. P r i n c i p l e s of medical s t a t i s t i c s . The Lancet Ltd., London. 311* pp. Hoffmann, R.S. 1 9 5 8 . The r o l e of reproduction and m o r t a l i t y i n population f l u c t u a t i o n s of voles (Microtus). E c o l . Monographs 28: 7 9 - 1 0 9 . Jameson, E.W. J r . 1 9 5 0 . Determining fecundity i n male small mammals. J . Mammal. 3 1 : 1+33-1*36. K a l e l a , 0 . 19i*9» Uber Fjeldlemming-Invasionen und andere i r r e g u l a r s Tierwanderungen. Ann. Zool. Soc. 'vanamo' 13: 1-90. K a l e l a , 0 . 1957. Regulation of reproduction rate i n subarctic populations of the vole Clethrionomys rufocanus (Sund.). Ann. Acad. S c i . Fenn. Ser. A, IV, 31*: 1 - 6 0 . 181 K a l e l a , 0. l°6l. Seasonal change of habitat i n the Norwegian lemming, lemmus lemmus ( L . ) . Ann. Acad. S c i . Fenn. Ser. A, IV, 55: 1-72. Koponen, T., A. Kokkonen, and 0. K a l e l a . 1961. On a case of spring migration i n the Norwegian lemming. Ann. Acad. S c i . Fenn. Ser. A. IV, 5 2 : 1-30. Lack, D . 195U* The n a t u r a l r e g u l a t i o n of animal numbers. Clarendon Press, Oxford. 3U3 pp. L e s l i e , P.H. 1959• The properties of a c e r t a i n l a g type of population growth and the influence of an external random f a c t o r on a number of such populations. P h y s i o l . Zool. 3 2 : l 5 l - l 5 9 » L e s l i e , P.H., D. C h i t t y , and H. C h i t t y . 1 9 5 3 . The estimation of population parameters from data obtained by means of the capture-recapture method. I I I . An example of the p r a c t i c a l a p p l i c a t i o n s of the method. Biometrika U O : 137 - 1 6 9 . L e s l i e , P.H., J.S. Perry, and J.S. Watson. 1 9 U 5 « The determination of the median body-weight at which female r a t s reach maturity. Proc. Zool. Soc. London 115: 473-U88. L e s l i e , P.H. and R.M. Ranson. 1940. The mortality, f e r t i l i t y , and rate of n a t u r a l increase of the vole (Microtus agrestis) as observed i n the laboratory. J . Anim. E c o l . 9: 27-527 L e s l i e , P.H., U.M. Venables, and L.S.V. Venables. 1952. The f e r t i l i t y and population structure of the brown r a t (Rattus norvegicus) i n c orn-ricks and some other h a b i t a t s . Proc. Zool. Soc. London 122: 187-238. Lord, R.B. 1 9 5 9 . The lens as an i n d i c a t o r of age i n c o t t o n t a i l r a b b i t s . J . W i l d l . Mgmt. 23: 358-360. MacArthur, J.W. 1949. S e l e c t i o n f o r small and large body size i n the house mouse. Genetics 34: 194-209. Manning, T.H. 195U* Remarks on the reproduction, sex r a t i o , and l i f e expectancy of the varying lemming, Dicrostonyx groenlandicus, i n nature and c a p t i v i t y . A r c t i c 7: 36-48. Martin, E.P. 1 9 5 6 . A population study of the p r a i r i e vole (Microtus  ochrogaster) i n northeastern Kansas. Univ. Kansas Mus. Nat. H i s t . Pub. V o l . 8, No. 6: 361-106. Maynard, L.A. and J.K. L o o s l i , 1 9 5 6 . Animal N u t r i t i o n . Fourth Ed. McGraw H i l l , Inc. New York. 484 pp. Munday, K.A. 1961. Aspects of stress phenomena, pp. 168-189 i n Mechanisms i n B i o l o g i c a l Competition, Symp. Soc. Exp. B i o l . No. XV. Cambridge, at the U n i v e r s i t y Press. Nasimovich, A., G. Novikov, and 0. Semenov-Tyan-Shanskii. 1 9 4 8 . The Norwegian lemming: i t s ecology and r o l e i n the nature complex of the Lapland Reserve. Materialy po gryzunam 19U8, 3: 203-262. ( i n Russian) 182 Newson, J. and D. Chitty. 1962. Haemoglobin levels, growth and survival i n two Microtus populations. Ecology, i n press. Osborne, T.B. and L.B. Mendel. 1926. The relation of the rate of growth to diet. J. B i o l . Chem 69: 661-673. Palmgren, P. 19U9* Some remarks on the short-term fluctuations in the numbers of northern birds and mammals. Oikos 1: llU-121. Parkes, A.S. and H.M. Bruce. 1961. Olfactory stimuli in mammalian reproduction. Science 13U: 10U9-105U* Pitelka, F.A. 1907 a. Some aspects of population structure i n the short-term cycle of the brown lemming i n northern Alaska. Cold Spring Harbor Symp. Quant. B i o l . 22: 237-251. Pitelka, F.A. 1957 b. Some characteristics of microtine cycles in the Arctic, pp. 73-88 i n Arctic Biology, Biology Colloquium, Oregon State College, 1957. Pitelka, F.A. 1959. Population studies of lemmings and lemming predators i n northern Alaska. XV tn International Cong. Zool., Sect. X, Paper 5« 3 pp. Pitelka, F.A. I96I. Ecology of lemmings and other microtines i n northern Alaska. Progress report for 1960-61 submitted to Arctic Inst, of North America. 18 pp. mimeo. Pitelka, F.A., P.Q. Tomich, and G.W. Treichel. 1955. Ecological relations of jaegers and owls as lemming predators near Barrow,Alaska. Ecol. Monographs 25: 85-117. Quay, W.B. i960. The reproductive organs of the collared lemming under diverse temperature and light conditions. J. Mammal. Ul: 7U-89* Quay, W.B. and J.F. Quay. 1956. The requirements and biology of the collared lemming, Dicrostonyx torquatus Pallas, 1778, i n captivity. Saugerierkundliche Mitteilungen U: 17U-180. Rausch, R. 1950. Observations on a cyclic decline of lemmings (Lemmus) on the arctic coast of Alaska during the spring of 19U9* Arctic 3: 166-177. Schindler, U. i960. Auftreten und Bekampfung der Rbtelmaus (Clethrionomys  glareolus Schreber) im Durrejahr 1959 i n Nordwestdeutschland. Z. angew. Zool. U7: U31-UU7. Selye, H. 19U6. The general adaptation syndrome and the diseases of adaptation. J. C l i n . Endocrin. 6: 117-230. Shelford, V.E. 19U3. The abundance of the collared lemming (Dicrostonyx  groenlandicus (Tr.) var. richardsoni Mer.) i n the Churchill area, 1929-to 19U0. Ecology 2k: U72-U8U. 183 Siivonen, L. 19U8. Structure of s h o r t - c y c l i c f l u c t u a t i o n s i n numbers of mammals and b i r d s i n the northern p a r t s of the northern hemisphere. F i n n i s h Foundation f o r Game Preservation, Papers on Game Research No. 1, 166 pp. Slobodkin, L.B. l°6l. Growth and re g u l a t i o n of animal populations. Holt, Rinehart, and Winston, New York. 18U pp. Snedecor, G.W. 1956. S t a t i s t i c a l Methods, F i f t h Ed. Iowa State College Press, Ames. 53U pp. Southwick, C H . 1958. Population c h a r a c t e r i s t i c s of house mice l i v i n g i n E nglish corn r i c k s : density r e l a t i o n s h i p s . Proc. Zool. Soc. London 131: 163-175. Stein, G.H.W. 1956. Naturliche Auslese b e i der RBtelmaus. Z.Saugetierkunde 21: 8U-100. S t e i n , G.H.W. 1957. M a t e r i a l i e n zur Kenntnis der Feldmaus, Microtus  a r v a l i s P. Z. Saugetierkunde 22: 117-135. S t i c k e l , L.F. 195U» A comparison of c e r t a i n methods of measuring ranges of small animals. J . Mammal. 35s 1-15* Sutton, G.M. and W.J. Hamilton J r . 1932. The mammals of Southampton Island. Mem. Carnegie Museum 12 (2): 1-111. Thompson, D.C 1955 a. The ecology and population dynamics of the brown lemming (Lemmus trimucronatus) at Point Barrow, Alaska, unpublished Ph.D. t h e s i s , Department of Zoology, U n i v e r s i t y of Missouri. 138 pp. Thompson, D.Q. 1955 b. The role of food and cover i n population f l u c t u a t i o n s of the brown leniming at Point Barrow, Alaska. Trans. 20th No. Amer. W i l d l . .Conf., 166-176. Thompson, D.Q. 1955 c. The 1953 lemming emigration at Point Barrow, Alaska. A r c t i c 8: 37-U5. Turner, C D . i960. General Endocrinology. Third Ed. W.B. Saunders Co., Ph i l a d e l p h i a . 511 pp. Watson, A. 1956. E c o l o g i c a l notes on the lemmings Lemmus trimucronatus and Dicrostonyx groenlandicus i n B a f f i n Island. J . Anim. E c o l . 25: 289-302. Wildhagen, A. 1953. On the reproduction of voles and lemming i n Norway. Statens V i l t u n d e r s ^ k e l s e r 1953, l - 6 l . Zejda, J . 196l. Age structure i n populations of the Bank Vole, Clethrionomys  glareolus Schreber 1780. Zoologicke L i s t y 10: 21+9-261+. Zimmermann, K. 1955* KBrpergrBsse und Bestandsdichte b e i Feldmlusen (Microtus a r v a l i s ) . Z. Saugetierk. 20: III+-II8. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0302572/manifest

Comment

Related Items