"Science, Faculty of"@en . "Zoology, Department of"@en . "DSpace"@en . "UBCV"@en . "Northcote, Thomas Gordon"@en . "2012-02-27T21:15:53Z"@en . "1952"@en . "Master of Arts - MA"@en . "University of British Columbia"@en . "Transformation of counts and weights of bottom fauna to either logarithims or square roots was required before application of statistical analyses. Preliminary analyses indicated that variability associated with relatively restricted sampling in a large lake prevented reliable evaluation of the bottom fauna while variability evident in extensive sampling from a large lake was not so great as to prevent reasonably precise estimation of abundance. Further analyses showed that the degree of variability was affected by regional location, depth, changes in abundance of the fauna, and qualitative composition of the fauna. Examination of factors contributing to sampling variability showed that operation of the Ekman-Birge dredge, distinction and delimitation of sampling zones in respect to depth and bottom substrate, diurnal and seasonal changes, and the size of samples all were of importance. The use of Ekman-Birge dredge with more powerful jaws in conjunction with more rigorous horizontal and vertical stratification of sampling was suggested as a means of reducing extreme variability in sampling. A 70 per cent sodium silicate solution was found to provide an effective separation of bottom organisms from certain types of substrate. Predation by fish was suggested as responsible for the significant littoral minimum evident in abundance of bottom organisms in Hatzic lake."@en . "https://circle.library.ubc.ca/rest/handle/2429/40966?expand=metadata"@en . "AN ANALYSIS OF VARIATION IN QUANTITATIVE SAMPLING OF BOTTOM FAUNA IN LAKES by THOMAS:GORDON NORTHCOTE A. THESIS SUBMITTED IN PARTIAL FULFILMENT OF . THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS i n the Department of ZOOLOGY We accept t h i s thesis as conforming to the standard required from candidates for the degree of MASTER OF ARTS. Members of the Department of Zoology THE UNIVERSITY OF BRITISH COLUMBIA A p r i l , 1952 ABSTRACT Transformation of counts and weights of bottom fauna to either logarithims or square roots was required before application of s t a t i s t i c a l analyses. Preliminary analyses indicated that v a r i a b i l i t y associated with r e l a t i v e l y r e s t -r i c t e d sampling i n a large lake prevented r e l i a b l e evaluat-ion of the bottom fauna while v a r i a b i l i t y evident i n extensive sampling from a large lake was not so great as to prevent reasonably precise estimation of abundance. Further analyses showed that the degree of v a r i a b i l i t y was affected by regional l o c a t i o n , depth, changes i n abundance of the fauna, and q u a l i t a t i v e composition of the fauna. Examination of factors contributing to sampling v a r i a b i l i t y showed that operation of the Ekman-Birge dredge, d i s t i n c t i o n and d e l i m i t -ation of sampling zones i n respect to depth and bottom sub-s t r a t e , diurnal and seasonal changes, and the s i z e of samples a l l were of importance. The use of Ekman-Birge dredge with more powerful jaws i n conjunction with more rigorous horizon-t a l and v e r t i c a l s t r a t i f i c a t i o n of sampling was suggested as a means of reducing extreme v a r i a b i l i t y i n sampling. A 70 per cent sodium s i l i c a t e solution was found to provide an e f f e c t i v e separation of bottom organisms from certain types of substrate. Predation by f i s h was suggested as responsible f o r the s i g n i f i c a n t l i t t o r a l minimum evident i n abundance of bottom organisms i n Hatzic lake. TABLE OF CONTENTS Page INTRODUCTION 1 ACKNOWLEDGEMENTS 4 VARIABILITY IN TYPICAL BOTTOM FAUNA DATA 5 STATISTICAL METHODS' 5 KOOTENAY\" LAKE DATA . 5 Numbers of Organisms .. 6 Weights of Organisms 9 Further Analysis of Data 9 Summary of Analyses - 19 GREAT SLAVE LAKE DATA .... 20 Numbers of Organisms 22 Depth d i s t r i b u t i o n 22 Regional d i s t r i b u t i o n 24 Weights of Organisms 27 Depth d i s t r i b u t i o n 27 Qualitative Separation of Fauna .... 30 Oligochaeta 32 Amphipoda 32 Insecta 36 Pelecypoda 36 Intergroup comparisons 40 Summary of Analyses 43 ANALYSIS OF FACTORS CONTRIBUTING TO VARIATION .. 45 OPERATION OF SAMPLING APPARATUS 45 Dredging ., 45 Screening and Separating 53 HETEROGENEITY WITHIN SAMPLING AREA 55 Depth. 55 Substrate 6 l CHANGES WITH TIME 6 6 Diurnal. 6 7 Seasonal 6& Annual 63 i i Page RANDOM SAMPLING 71 Distribution of Fauna ............. 71 Size of Samples 79 Number of Samples 82 GENERAL DISCUSSION .84 SUMMARY AND CONCLUSIONS 89 LITERATURE CITED 93 i i i TABLES Table ' Page I. Mean numbers of organisms i n depth zones from northern and' southern portions of Kootenay lake.' 1 2 I I . Analysis of variance on logarithmic numbers of organisms from northern and southern portions of Kootenay lake 1 2 I I I . Means of logarithmic numbers of organisms from northern portion of Kootenay lake. ... 1 3 IV. Mean number of chironomids from northern and southern portions of Kootenay lake. ... 1 4 V. Analysis of variance on transformed ( ycount 4- 0 . 5 ) numbers of chironomids from northern and southern portions of Kootenay lake 1 5 VI. Mean number of oligochaetes from northern and southern portions of Kootenay lake. .... 1 7 VII. Analysis of variance on logarithmic numbers of oligochaetes from northern and southern portions of Kootenay lake. 1 9 V I I I . Comparisons of v a r i a b i l i t y i n t o t a l numbers of organisms (logarithmic transformation) between depth zones i n Great Slave lake. .. 23 IX. Comparisons of v a r i a b i l i t y i n numbers of organisms (logarithmic transformation^ bet-ween the main lake and east arm of Great Slave lake., 2 9 X. Comparisons of v a r i a b i l i t y i n numbers (log-arithmic transformation) of oligochaetes i n Great Slave lake 33 XI. Comparisons of v a r i a b i l i t y i n numbers (log-arithmic transformation) of amphipods i n Great Slave lake 3 5 Table Page XII. Comparisons of v a r i a b i l i t y i n numbers (logarithmic transformation) of chiro-nomids i n Great Slave lake. 38 X I I I . Comparisons of v a r i a b i l i t y i n numbers (logarithmic transformation) of sphae-r i i d s i n Great Slave lake. 41 XIV. Comparisons of v a r i a b i l i t y between groups of bottom fauna i n Great Slave lake 42 XV. Mean numbers of organisms i n partitioned dredgings i n Hatzic lake 51 XVI. Qualitative separation of mean numbers of organisms i n partitioned dredgings at 2 feet depth i n Hatzic lake. 52 XVII. Mean number of organisms at sampling areas i n 2 foot depth zone, Hatzic lake.. 6 3 XVIII. Comparisons of organic and inorganic material retained i n regional samples at ; 2 feet i n Hatzic lake 66 XIX. Mean numbers of bottom organisms taken i n main portions of Great Slave lake. .. 70 XX. Analysis of variance on t o t a l numbers of bottom fauna taken on three consecutive years i n Great Slave lake. \ 70 XXI. Analysis of variance on three-inch sub-samples from nine-inch p a r t i t i o n a l dredg-ings i n Hatzic lake 75 XXII. Comparison of oligochaete frequency d i s -.. t r i b u t i o n s i n three-inch subsamples to three t h e o r e t i c a l d i s t r i b u t i o n forms. ... 76 XXIII. Comparison of chironomid frequency d i s -t r i b u t i o n s i n three-inch subsamples to three t h e o r e t i c a l d i s t r i b u t i o n forms. .. 78. XXIV. Comparisons of v a r i a b i l i t y between s i x -and nine-inch Ekman samples i n Hatzic \u00E2\u0080\u00A2 lake 80i V FIGURES Figure Page 1. 1948 bottom dredging stations i n Kootenay lake. 7 2. Depth d i s t r i b u t i o n of bottom organisms i n Kootenay lake. Left - mean numbers; right - mean numbers (logarithmic trans-formation) with f i d u c i a l l i m i t s 8 3. Mean weight per dredging (logarithmic trans-formation) of bottom organisms i n Kootenay lake 11 4 . Depth d i s t r i b u t i o n of chironomids i n Kootenay lake 1 6 5. Depth d i s t r i b u t i o n of oligochaetes i n Kootenay lake 18 6 . Mean numbers (logarithmic transformation) of bottom organisms per dredging i n Great Slave lake, 1944-47 21 -7. Great Slave lake 25 8. Mean numbers (logarithmic transformation) of bottom organisms per dredging from'east arm and main portion of Great Slave lake... 2 6 9 . Mean weights (logarithmic transformation) of bottom organisms per dredging from east arm and main portion of Great Slave lake... 28 10. Mean numbers (logarithmic transformation) of oligochaetes per dredging i n Great Slave lake. 31 11. Mean numbers (logarithmic transformation) of amphipods per dredging i n Great Slave lake. ...: 34 12. Mean numbers (logarithmic transformation) of. chironomids per dredging i n Great Slave lake 37 v i Figure Page 13. Mean numbers (logarithmic transformation) of sphaeriids per dredging i n Great Slave lake 39 14. P a r t i t i o n i n g apparatus .. 4# 1 5\u00E2\u0080\u00A2 P o s i t i o n of 1 9 5 1 - 5 2 sampling areas i n Hatzic lake 50 1 6 . Depth d i s t r i b u t i o n of bottom fauna, temperature, and oxygen i n Hatzic lake, September 5 - S , 1 9 5 1 . \u00E2\u0080\u00A2\u00E2\u0080\u00A2 \u00E2\u0080\u00A2 5$ 1 7 . Depth d i s t r i b u t i o n of oligochaetes and chironomids i n Hatzic lake, September 5 - S, 1 9 5 1 . 6 0 1 & . Quality of organic material retained i n bottom sample screenings from four areas i n Hatzic lake 65 1 9 . Theoretical d i s t r i b u t i o n s of 5 0 bottom organisms i n a square yard of bottom 7 2 t INTRODUCTION The problems connected with quantitative sampling of organisms on lake bottoms are i n many ways unique. The invest-i g a t o r i s completely separated from the sampling area and has l i t t l e opportunity of examining the lake bottom d i r e c t l y except i n shallow regions of the l i t t o r a l zone. This condi-t i o n introduces d i f f i c u l t i e s both i n obtaining samples and i n the determination and selection of homogeneous sampling areas. Secondly, the area of bottom that can be included i n a sample i s l i m i t e d i n comparison with sample sizes that can be taken i n t e r r e s t r i a l i nvestigations. The physical nature of bottom deposits i s subject to extreme horizontal and v e r t i c a l v a r i a -t i o n which requires special sampling techniques. B i o l o g i c a l features of bottom organisms present sampling d i f f i c u l t i e s of major importance. A t y p i c a l fauna may be composed of at l e a s t nine phyla each requiring consideration i n sampling procedures. Several \"semiplanktonic\" groups make extensive movements o f f the lake bottom. Other forms are only temporary residents on the bottom during stages i n t h e i r l i f e cycle. In addition, movement of some groups may take place along the lake bottom i n conjunction with seasonal limnological changes i n the lake. Indeed the investigator i s i n a po s i t i o n somewhat analagous to that of estimating numbers of r e p t i l e s , b i r d s , and burrowing mammals over a large area by indiscriminately lowering a cage from a helicopter at night onto mountain tops, forests and deserts. In such a s i t u a t i o n some method must be used to guide i n the in t e r p r e t a t i o n of observed changes i n abundance with space and time. S t a t i s t i c a l treatment of data permits the investigator to attach l i m i t s of confidence to observed changes which he may ei t h e r r e j e c t as explainable by chance v a r i a t i o n or accept as s i g n i f i c a n t . Thus biometrics may become as useful a t o o l to the invest i g a t o r as are h i s quantitative sampling devices. In early studies v a r i a t i o n i n numbers of organisms with space and time were summarized by expression of average numbers within depth zones (Muttkowski, 1918; Juday, 1922). Rawson (1930) and Eggleton (1931) rea l i z e d the necessity of larger numbers of samples i n order to reduce \"sampling error\" but no attempts were made to deal s t a t i s t i c a l l y with v a r i a t i o n . Treatment of bottom fauna by biometrical methods was conducted by M i l l e r (1936) and Deevey (1941). Wohlschlag (1950) compared quantities of l i t t o r a l bottom fauna i n several regions of Wabec lake by s t a t i s t i c a l procedures. Larkin et a l (1950) s t a t i s t i -c a l l y compared weights of bottom fauna from Paul lake dredging data for 1931, 1948, and 1949. In general, however, s t a t i s t i -c a l treatment of bottom fauna data has been r e s t r i c t e d to pre-liminary studies and no attempt has been made to consider i n d e t a i l factors aff e c t i n g the error involved i n sampling 3 methods. A s t a t i s t i c a l study of numerical v a r i a b i l i t y i n bot-tom sampling would be desirable then, not only i n evaluation of sampling methods, but i n refinement of the precision with which the r o l e of bottom fauna i n lake productivity may be assessed. In the present study v a r i a t i o n i n numbers and weights of bottom organisms was considered i n Kootenay lake data of the B r i t i s h Columbia Game Department and i n Great Slave lake data supplied by Dr. D. S. Rawson. More extensive analyses were carried out on data from dredgings taken i n Hatzic lake by the writer i n the f a l l of 1951 and spring of 1952. A l l raw data and analysis were f i l e d with the Fisheries Research Group of the B r i t i s h Columbia Game Department. 4 ACKNOWLEDGEMENTS , This study was carried out under the supervision of Dr. P . A. Larkin to whom the w r i t e r i s greatly indebted f o r h i s guidance and c r i t i c i s m throughout. The kindness of Dr. D. S. Rawson i n supplying data from dredgings taken i n Great Slave lake i s g r a t e f u l l y ack-nowledged. The w r i t e r i s indebted to Dr. S. W. Nash, Department of Mathematics, for development of many of the s t a t i s t i c a l analyses, and to Dr. K. Graham, Department of Zoology, for supplying d i s t r i b u t i o n formulae. F i e l d work was made possible by the Fisheries Research Group of the B r i t i s h Columbia Game Department. The genero-s i t y of Dr. E. C. Black, Faculty of Medicine, i n giving the wr i t e r use of his laboratory i s sincerely appreciated. The wr i t e r wishes to extend special thanks to his parents, r e l a t i v e s and fellow students for t h e i r assistance i n c o l -l e c t i o n of samples. 5 VARIABILITY IN TYPICAL BOTTOM FAUNA DATA STATISTICAL METHODS The s t a t i s t i c a l analyses applied i n t h i s study were methods suggested by Snedecor ( 1 9 4 6 ) and Dr. S. W. Nash, Department of Mathematics, University of B r i t i s h Columbia. A l l counts were transformed to logarithms where the variance tended to be proportional to the mean before application of s t a t i s t i c a l t e s t s . When numerous zero counts were present the logarithm of the count plus one, l o g (count 4 - 1 ) , was used to di s t i n g u i s h between counts of zero and unity. Where no proportionality was evident between variances and means, but counts i n general were l e s s than 5 0 per dredging, trans-formation was made to the square root of the count.. I f numerous counts less than ten were present the square root of the count plus 0.5,^(count 0 . 5 ) ,was used. In analysis of the weight data transformation to logarithms was made where the variance and mean tended to be proportional. KOOTENAY LAKE DATA Ninety-two samples were taken with a nine-inch Ekman dredge at twelve stations on Kootenay lake during the 6 summer of 1949 (Figure 1). A l l samples were washed success-i v e l y through two screens having respectively 10 and 30 meshes per l i n e a r inch. The blotted wet weight of organisms was obtained on an a n a l y t i c a l balance. Numbers of Organisms The d i s t r i b u t i o n of a l l organisms with depth i s given i n Figure 2. Ostensibly,\"the depth d i s t r i b u t i o n may be i n t e r -preted i n the following manner. The decrease i n number of organisms per dredging i n the 11 to 30 metre zone as compared with numbers at 0*to 10 metres possibly represents a sub-l i t t o r a l minimum. The profundal fauna reaches a maximum i n the 6 l to 90 metre zone and declines i n the lower profundal zone. Although there i s considerable evidence to support the existence of s u b l i t t o r a l minima, at lea s t i n lakes of the \"eutrophic\" type (Rawson 1930, Deevey 1941), no such i n t e r p r e t a t i o n may be applied to the s u b l i t t o r a l zone of Kootenay lake when s t a t i s t i c a l l i m i t s of confidence are placed on mean numbers of organisms i n the f i v e depth zones (Figure 2). I t i s obvious (Figure 2) that there would be no j u s t -i f i c a t i o n s t a t i s t i c a l l y f or separating means of the f i r s t three depth zones. On the basis of the samples taken no s i g n i f i c a n t differences are evident i n the t o t a l number of organisms per dredging from 0 to 60 metre zone throughout 7 8 Figure 2. Depth d i s t r i b u t i o n of bottom organisms i n Kootenay lake. Left - mean numbers; ri g h t - mean numbers (logarithmic trans-formation) with f i d u c i a l l i m i t s . 9 Kootenay lake. The smallest and largest means are not s i g -n i f i c a n t l y d i f f e r e n t (p = >.05); only means at 11 to 30 and 91 to 150 metre zones are d i s t i n c t (p = <.01). When actual counts are considered (Figure 2), the mean of the 91 to 150 metre zone, 111.27, i s over four times larger than that of the 11 to 30 metre zone, 27.60 organisms per dredging. Weights of Organisms In Figure 3 logarithmic mean weights of organisms per dredging are shown f o r f i v e depth zones i n Kootenay lake. The same general trend previously shown f o r numbers i s e v i -dent; likewise the f i d u c i a l l i m i t s of mean weights are extre-mely high. The average weight of organisms per dredging i n the 11 to 30 metre zone i s s i g n i f i c a n t l y lower than that at the 91 to 150 metre zone; the mean at 31 to 60 metres i s s i g -n i f i c a n t l y lower than weights at either 61 to 90 or 91 to 150 metres. F i d u c i a l l i m i t s associated with weights appear to be proportionately as large as those of counts. Further Analysis of Data Two factors possibly contributing to variations are immediately apparent. F i r s t , dredgings from a l l regions of the lake have been lumped i n t o one group, neglecting changes 1 0 that may occur i n substrate, physical and chemical features i n the lake, or ecological relationships of the fauna at d i f -ferent depths. Second, a l l organisms have been grouped together regardless of taxonomic or ecological status. Kootenay lake may be divided into three general eco-l o g i c a l regions, a northern portion, a southern portion and the west arm. (Figure 1). The limnology of the north end of the lake i s markedly affected by the cold Lardeau r i v e r and the rocky, precipitous nature of the lake basin. The warm, s i l t laden Kootenay r i v e r flowing into the south end of the lake probably affects the general productivity of the bottom i n that region. The west arm outlet of the lake i s charac-t e r i z e d by i t s shallow depth and d i s t i n c t current. The one dredging series taken i n that area i s included, f o r purposes of t h i s examination, with, those from the northern end of the lake. Separation of dredging data i n t o northern and southern portions i s given i n Table I . These data, a f t e r a l o g a r i t h -mic transformation, are treated by analysis of variance, mul-t i p l e c l a s s i f i c a t i o n , using the appropriate correction for unequal subclass numbers. Results of the analysis are shown i n Table I I . The northern and southern regions are decidedly d i f f e r e n t , numbers of organisms per dredging from the south end of the lake being consistently higher than those from the north end i n a l l depth zones. Figure 3 Mean weight per dredging (logarithmic trans-formation) of bottom organisms i n Kootenay lake. 12 Table I. Mean numbers of organisms in depth zones from northern and southern portions of Kootenay lake. Depth zones i n metres -0 - 1 0 1 1 - 3 0 3 1 - 6 0 6 1 - 9 0 9 1 - 1 5 0 Northern Number of dredgings 6 12 15 6 17 Mean number of . organisms 26.0 21.0 39.4 52.5 24.8 Southern Number of dredgings 5 3 5 Mean number of organisms 81.4 54.0 116.2 Table II. Analysis of variance on logarithmic numbers of organisms from northern and southern portions portions of Kootenay lake. Source of Variation Degrees Freedom of Sum of Squares Mean Square F Between locations 1 6.1173 6.1173 17.0019xx Between depths 4 1.8134 .4533 1.2599 Discrepancy 82 29.5077 .3598 xxSignificant at p - 0.01 7 16 187.6 203.1 13 Separation of data into two portions reduces the num-ber of samples within the five depth zones; thus data from the southern portion of Kootenay lake do not warrant further analysis. Logarithmic means from the northern portion of the lake have high standard errors (Table III). No significant differences may be demonstrated between means from the various depth zones in that region. Table III. Means of logarithmic numbers of organisms from northern portion of Kootenay lake. Depth zones in metres 0 - 1 0 . 1 1 - 3 0 3 1 -\u00E2\u0080\u00A2 6 0 6 1 - 9 0 91 - 1 5 0 Number of samples 6 Logarithmic mean number of organisms 1.0832 Standard error of mean 1 .2844 1 2 1 5 1 . 1 3 7 5 1 . 0 6 3 0 t . 1 3 1 7 1 . 2 0 1 2 6 17 1.3148 1.3210 t.3279 i.0697 Lumping of all organisms into \"bottom fauna\" possi-bility contributes to variation evident in total counts within depth zones. Oligochaetes and chironomid larvae represent about 90 per cent of the total fauna from each of the five depth zones, thus separation of these two should indicate any major effect of grouping. Total counts of chironomid larvae 14 from northern and southern portions of Kootenay lake are given in Table IV. An analysis of variance was carried out on data transformed by ]/(count f 0.5) to determine the effects of depth and region of sample on the number of chir-onomids per dredging. Analysis showed that the effect of locality was insignificant (Table V) whereas depth had a significant effect on the density of chironomids per dredg-ing. Chironomid counts from northern and southern parts of the lake were lumped in following analyses. Table IV. Mean number of chironomids from northern and southern portions of Kootenay lake. Depth zones in metres 0-15 16-30 31 - 60 61 - 90 91 -150 Northern Number of dredgings 13 5 15 6 17 Mean number of chironomids 18.77 5.00 8.93 7.33 13.88 Southern Number of dredgings , 5 3 5 7 16 Mean number of chironomids 57.21 6.00 9.00 6.71 7.87 15 Table V. Analysis of variance on transformed ( V count 4- 0.5) numbers of chironomids from northern and southern portions of Kootenay lake. Source of v a r i a t i o n Degrees of freedom Sum of Squares Mean Square F Between locations 1 2.81 2.81 .90 Between depths 4 64.99 16.24 5.69 xx Discrepancy 82 234.05 2.85 xxSignificant at p = 0.01 The depth d i s t r i b u t i o n of chironomids throughout Kootenay lake i s shown i n Figure 4. The v a r i a t i o n about the means i n each depth zone does not appear to be large when chironomids are considered.separately. The average count i n the 0T15 metre zone i s s i g n i f i c a n t l y higher than that of other depth zones except the deepest, 91 - 1 5 0 metres. There i s no s i g n i f i c a n t difference i n the mean counts at 1 6 - 3 0 , 3 1 - 6 0 and 6 1 - 9 0 metre zones; the mean of the 9 1 - 1 5 0 metre zone, however, i s s i g n i f i c a n t l y higher than that of the three previous zones. Mean counts f o r oligochaetes from northern and southern portions of Kootenay lake are shown i n Table VI. These data are treated by an analysis of variance log(count 41) transformation to determine the effects of lo c a t i o n and depth on the number of oligochaetes. Both l o c a l i t y and depth are shown to have very s i g n i f i c a n t effects. (Table V I I ) . 17 Further analyses of oligochaete counts are carried out on data from the northern portion of the lake. The transformed mean numbers of oligochaetes per dredging in the northern region of Kootenay lake are given in Figure 5. Table VI. Mean number of oligochaetes from northern and ' southern portions of Kootenay lake. Depth zones in metres 0-15 16-30 31-60 61-90 91-150 Northern Number of dredgings 13 5 15 6 17 Mean number of oligochaetes 3.31 5.00 25.00 44.50 8.35 Southern Number of dredgings 5 3 5 7 16 Mean number of oligochaetes 12.20 38.00 104.80 178.14 186.06 18 J 1 \u00E2\u0080\u0094I I I 0 -15 1 6 - 3 0 3 1 - 6 0 61 -90 91-150 D E P T H Z O N E S IN M E T R E S Figure 5. Depth distribution^ of oligochaetes i n Kootenay lake-. 6 19 Table VTI. Analysis of variance on logarithmic numbers of oligochaetes from northern and southern portions of Kootenay lake. Source of Degrees Sum of Mean F Variation of freedom Squares Square Between locations 1 9.0321 9.0321 19.68xx Between depths 4 7.5263 1.8816 4.10xx Discrepancy 82 37.6349 0.4590 Fiducial limits of mean counts at most zones are extremely large. It is not possible to show significant differences between mean counts of oligochaetes at any one of the depth zones. Summary of Analyses Several features related to degree and source of variation in counts of bottom organisms have become apparent in analysis of Kootenay lake data. Difference between means must be very large in order to demonstrate significance in data with this order of variability. Standard errors asso-ciated with weight means appeared to be proportionately as large as those of count means. Separation of data into two portions corresponding to rough ecological divisions of the lake did not appreciably reduce variations of total counts 20 w i t h i n depth zones. However, data were not extensive enough to permit complete subdivision on t h i s basis. Probably regional d i v i s i o n was not complete enough;to e f f e c t i v e l y reduce v a r i a t i o n introduced by grouping of samples from d i f -ferent habitats. Separation of broad taxonomic groups within the fauna showed that counts of some groups tended to be more variable; standard errors associated with mean counts of chironomids appeared r e l a t i v e l y lower than those associated with mean counts of oligochaetes. These features w i l l be considered i n analyses of sub-sequent data where a more complete separation of depth zones, regional l o c a t i o n , and faunal composition w i l l be possible. GREAT SLAVE LAKE DATA Data from 502 quantitative dredgings taken on Great Slave lake between 1944 and 1947 were made available by Dr. D. S. Rawson. Most dredgings were taken with a standard nine-inch Ekman dredge although an automatic closing modifi-cation of the Ekman dredge was used for deepwater dredgings (Rawson, 1947c). A few samples were made with a Peterson grab dredge but these were not included i n the analyses. 21 Figure 6. Mean numbers (logarithmic transformation) of bottom organisms per dredging in Great Slave lake, 1944-47. 22 Numbers of Organisms Depth d i s t r i b u t i o n : Counts of organisms -are transformed to logarithms i n order to apply appropriate s t a t i s t i c a l treatments. Means and associated f i d u c i a l l i m i t s (p = 0.05) of t o t a l numbers within depth zones of Slave lake are shown i n Figure 6. A marked decrease i n numbers of organisms from l i t t o r a l to profundal depths i s apparent. This decrease, takes place i n two stages. F i r s t , a s i g n i f i c a n t decrease i n abundance i s evident between the upper (0-5 metres) and lower (6-20 metres) l i t t o r a l regions. No s i g n i f i c a n t changes occur between 20 to 80 metres. Below 80 metres the mean numbers decrease r a p i d l y ; means being s i g n i f i c a n t l y lower at each succeeding depth zone up to 220 metres. Beyond 220 metres no s i g n i f i c a n t differences may be demonstrated between means. I t i s apparent that f i d u c i a l l i m i t s about each of the means up to the 81 to 140 metre depth zone are smaller than those associated with means beyond 140 metres (Figure 6). The size of f i d u c i a l l i m i t s at any one zone, however, i s not a s a t i s f a c t o r y measure of v a r i a b i l i t y i n numbers f o r comparison with v a r i a b i l i t i e s at other zones. Conversion of counts to logarithms permits comparison of standard deviations by \"z\" values of Fisher (1948), although means may not approximate each other. The \"z\" values calculated i n comparisons of v a r i a b i l i t y between depths are given i n Table V I I I . I t i s Table V I I I . Comparisons of v a r i a b i l i t y i n t o t a l numbers of organisms (logarithmic transformation) between depth zones i n Great Slave lake. Depth zones i n metres 0 - 5 6 - 2 0 2 1 - 4 Q 41 - 80 81-140 141-220 2 2 1 - 3 0 0 3 0 0 - 5 0 0 5 0 0 - 6 0 2 Number of samples 7 8 151 106 9 8 6 9 3 3 1 9 25 1 7 Mean number of organisms (logarithmic) 1 . 9 5 6 6 1 . 7 0 9 5 1 . 7 7 9 7 1 . 7 8 7 6 1 . 4 2 4 4 1 . 2 1 0 4 1 . 0 2 2 9 1.0384 1 . 1 9 0 6 Standard deviation . 5 2 7 8 . 6 0 6 9 . 4 0 4 9 . 3 1 5 9 . 4 4 2 4 . 4 7 2 2 . 3 5 5 4 .3770 . 3 7 7 6 \"z\" values com-pared with 0 - 5 metre zone .1398 .2647* .5128** .1765 .1154 . 3 9 9 5 * * .3364* .3350* \" z \" values com-pared with 6 - 2 0 metre zone .4048** .6528** .3155** .2507* . 5 3 5 3 * * .4776** .4774** ^ S i g n i f i c a n t at p = 0 . 0 5 . S i g n i f i c a n t at p = 0 . 0 1 . 24 evident that t o t a l counts of organisms below 20 metres are, f o r the most part, less variable than those i n the 0 to 5 metre zone. A l l series below 20 metres are significantly-l e s s variable than that i n the 6 to 20 metre zone. The var-i a b i l i t y of samples, i s thus greater i n the l i t t o r a l and sub-l i t t o r a l zone, than i n the deeper portions of Great Slave lake; Changes were noted i n abundance of bottom organisms i n two depth regions, 0 to 20 and 81 to 220 metres (Figure 6 ) . Samples from the, f i r s t region (0 to 21 metres) were shown to exhibit more v a r i a t i o n than i n most other depth zones. Samples from the second region (81 to 220 metres) could not be demon-strated as s i g n i f i c a n t l y less variable than those from the 0 to 5 metre zone (Table V I I I ) . Apparently numbers of bottom organisms were more variable i n regions where s i g n i f i c a n t changes i n abundance occurred. , Regional d i s t r i b u t i o n : In the previous section degree of v a r i a t i o n i n num-bers of bottom organisms was discussed i n r e l a t i o n to depth, i . e . v a r i a t i o n on a v e r t i c a l plane was considered; v a r i a t i o n also could occur h o r i z o n t a l l y i n any one depth zone, between di f f e r e n t portions or ecological d i v i s i o n s within a lake. Such v a r i a t i o n was demonstrated i n analysis of the Kootenay lake data. Great Slave lake may be divided i n t o two portions (Figure 7), the \"east arm\" and the main lake,on the basis of' geomorphological features of the basin (Larkin, 1948). For 26 gure 8. Mean numbers (logarithmic transformation) of bottom organisms per dredging from east arm and main portion of Great Slave lake. 27 purposes of comparison with the \"east arm\", dredgings taken i n the \"north arm\", and the mouth of Slave r i v e r were not included with those taken i n the main lake. Figure 8 gives mean numbers of organisms ( l o g a r i t h -mic transformation) f o r the east arm and main lake. Marked differences between the two portions of the lake are evident. Between 0 to 20 metres mean numbers i n the east arm are s i g n i f i c a n t l y higher than those i n the main lake. No differences, can be shown between the two portions i n the 21 to 40 metre are zone. Below t h i s depth, however, mean numbers* s i g n i f i c a n t l y lower i n the east arm. Comparisons of v a r i a b i l i t y between samples taken i n the east arm and main lake are summarized i n Table IX. Num-bers of organisms are s i g n i f i c a n t l y more variable i n the east arm at 0 to 5, 6 to 20, and 41 to 80 metre depth zones. The v a r i a b i l i t y of samples at 81 to 140 metres i n the main lake i s not s i g n i f i c a n t l y lower than those i n the east arm. Sample within the 6 to 20 metre zone are more variable i n the main lake 1 than i n the east arm. Weights of Organisms Depth d i s t r i b u t i o n : Total weights of organisms per dredging are transfor med to logarithms before application of s t a t i s t i c a l procedure Mean logarithmic weights for the two portions of Great Slave 28 Figure 9. Mean weights (logarithmic transformation) of bottom organisms per dredging from east arm and main portion of Great Slave -L eu\u00C2\u00A36 \u00E2\u0080\u00A2 2 9 Table IX. Comparisons of v a r i a b i l i t y i n numbers of organisms (logarithmic transformation) between the main lake A and east arm of Great Slave lake. Depth zones 0 - 5 6 - 2 0 21-40 41-80 81-140 i n metres East arm Number of samples 25 3 0 3 1 2 9 3 9 Mean number of organisms (log-arithmic) 2.2082 1.8730 1 . 6 1 3 0 1 . 6 4 2 8 1 . 2 3 6 8 .'Standard deviation . 5 7 1 0 . 4 1 5 9 . 5 2 8 6 . 3 2 7 9 . 4 2 5 7 Main lake Number of samples 3 8 8 4 4 3 4 5 2 6 Mean number of organisms (log-arithmic) 1.7641 1 . 5 8 5 8 1.8097 1.8080 1 . 6 2 6 1 Standard deviation . 3 9 3 9 . 6 2 3 8 . 3 1 3 7 . 2 2 2 0 . 3 3 1 7 \"z\" values between \u00E2\u0080\u00A2 regions . 3 6 4 6 * . 3 9 8 7 ^ * . 4 0 7 4 * * . 3 9 0 2 * * . 2 4 6 9 S i g n i f i c a n t at p - 0 . 0 5 . S i g n i f i c a n t at p = 0 . 0 1 lake are given i n Figure 9 . The same general trend evident i n counts (Figure 7 ) i s also apparent i n weights of organisms at each depth zone. In Great Slave data,, as with Kootenay lake data, there appears to be l i t t l e difference i n v a r i a b i l i t y 30 between counts and weights. Qualitative Separation of Fauna Changes in abundance and variability of numbers of bottom organisms have been considered in relation to depth and regional location in Great Slave lake. It is of importance, however, to determine which groups within the .fauna contribute most significantly to changes in abundance and variability in numbers. Qualitatively, the bottom fauna of Great Slave lake may be divided into nematodes, oligochaetes, ostracods, amphi-pods, chironomids, gastropods, sphaeriids, and a miscellaneous group including such forms as hydracarinids. These divisions are obviously not all at the same phylogenetic level nor are they of comparable quantitative importance. In the following section nematodes, ostracods, and gastropods will not be considered. Analyses will concern a separation of oligochaetes, amphipods, chironomids, and sphaeriids, which constitute over 90 per cent of the total fauna both numerically and gravimetrically. These four groups represent the basic profundal fauna of most lakes (Welch, 1935). Figure 10. Mean numbers (logarithmic transformation) of oligochaetes per dredging in Great Slave lake. 32 Oligochaeta: The depth d i s t r i b u t i o n of oligochaetes i n the main lake and east arm of Great Slave lake are quite d i s t i n c t (Figure 10). In the 0 to 5 metre zone mean numbers are s i g -n i f i c a n t l y higher i n the east arm than i n the main lake. With further increase i n depth means i n the east arm decrease rap i d l y , while those i n the main lake show a gradual increase. Mean numbers of oligochaetes beyond 21 metres i n the main lake are s i g n i f i c a n t l y higher than those i n the east arm. Examination of \"z\" values (Table X) shows that numbers of oligochaetes i n the east arm are s i g n i f i c a n t l y l e s s variable at depths below 20 metres than i n l i t t o r a l and s u b l i t t o r a l regions. No s i g n i f i c a n t differences i n v a r i a b i l i t y with depth are evident i n the main lake. Numbers of oligochaetes are s i g n i f i c a n t l y more variable at 0 to 5 metres i n the east arm than i n the main lake while at 21 to 40 and 81 to 140 metre r zones, numbers from the main lake are more variable. Amphipoda: The amphipods i n Great Slave lake are represented almost exclusively by Pontoporeia a f f i n i s whose d i s t r i b u t i o n has been extensively studied (Larkin, 1948). Numbers of amphi-pods are s i g n i f i c a n t l y lower i n the 6 to 20 metre zone of the east arm.than i n the main lake (Figure 11). Beyond 40 metres, however, the group i s more abundant i n the main lake, Numbers of amphipods are s i g n i f i c a n t l y more variable i n the 0 to 5 metre zone than i n a l l zones below 20 metres i n both portions Table X. Comparisons of v a r i a b i l i t y i n numbers (logarithmic transformation) of oligochaetes i n Great Slave lake. Depth zones i n metres 0 - 5 6 - 2 0 2 1 - 4 0 4 1 - 8 0 81 - 1 4 0 \u00C2\u00A3 A Number of samples 2 5 3 0 3 1 2 9 3 9 S T Mean Number of oligochaetes (logarithmic) . 9 9 4 5 . 5 2 9 7 .1928 . 2 9 2 5 . 2 9 6 5 A R M Standard deviation . 6 8 6 3 . 4 9 8 2 . 3 3 5 9 . 3 5 8 2 .3490 \"z\" values compared to 0 - 5 metre zone . 3 1 9 9 . 7 1 4 4 S * . 6 5 0 2 * * . 6 7 6 0 \u00C2\u00ABz\" values compared to 6 - 2 0 metre zone .3941 .3300* . 3 5 5 5 * M Number of samples 3 8 8 4 4 3 4 5 2 6 A I N L Mean number of o l i -gochaetes (logarithmic) . 3 3 8 3 . 4 1 0 8 . 4 9 7 9 . 6 3 5 2 . 6 3 9 7 Standard deviation . 4 1 6 6 . 4 3 0 4 . 4 5 9 7 . 3 7 9 2 . 5 3 8 2 A K E \"z\" values compared to 0 - 5 metre zone . 0 3 2 4 . 0 9 8 0 . 0 9 3 5 . 2 5 6 2 n z \" values between regions . 4 9 8 9 X K .1458 . 3 1 3 3 * . 0 5 6 4 . 4 3 1 1 X s S i g n i f i c a n t at p = 0 . 0 5 . ^ S i g n i f i c a n t at p a 0 . 0 1 . 3-4 _ l l I 1 1 \u00E2\u0080\u00A2 0 - 5 6 - 2 0 2 1 - 4 0 , 4 1 - 8 0 8 1 - 1 4 0 D E P T H Z O N E S IN M E T R E S Figure 11. Mean numbers (logarithmic trans= formation) of amphipods per dredging i n Great Slave lake. Table XI. Comparisons of v a r i a b i l i t y i n numbers (logarithmic transformation) of amphipods i n Great Slave lake. Depth zones i n metres 0 - 5 6-20 21 - 40 41 - 80 81 - 140 E Number of samples 2 5 30 31 29 39 A S Mean number of amphi-pods (logarithmic) 1 . 4 7 4 2 1.6742 1.4810 1.3269 .8453 1 A Standard deviation . 9 4 0 3 .4511 .5507 .5019 . 4 6 2 6 R M \"z\" values compared to 0-5 metre zone .7343** .5347** .6275** .7095** \"z\" values compared to 6-20 metre zone - .1997 .1053 . 0 2 4 7 M Number of samples 38 84 43 45 26 A I Mean number of amphi-pods (logarithmic) 1.4005 1.3800 1.6623 1.6599 1.3566 N L Standard deviation .6058 .7365 .3450 .2848 .5595 A K \"z\" values compared to 0 - 5 metre zone .1956 , XX . 5 6 3 0 .7547** .0797 . E \"z\" values compared to 6 - 20 metre zone XX * 7 5 8 5 . 9 5 0 1 * * . 2 7 4 6 * \" z \" values between regions .4395s .4904** . 4 6 7 5 * . 5 6 6 4 * * .1898 ^ S i g n i f i c a n t at p = 0 . 0 5 . ^ S i g n i f i c a n t , at p = 0.01. 3 6 of the lake (Table XI). For the most part numbers of amphipods exhibit greater v a r i a b i l i t y i n the east arm than i n the main lake, exceptions being the 6 to 20 metre zone where main lake numbers were more variable and the 81 to 140 metre zone where no s i g n i f i c a n t difference i n v a r i a b i l i t y could be demonstrated. Insecta (chironomidae): The abundance of chironomids decreases s i g n i f i c a n t l y between 0 to 5 metre and 6 to 20 metre depth zones.in both regions of Great Slave lake (Figure 12). Below 20 metres there are marked fluctuations i n the abundance, however, num-bers i n the east arm are s i g n i f i c a n t l y higher than those i n the main lake at a l l depth zones. Although v a r i a b i l i t y i n num-bers of chironomids appears greater i n the 0 to 5 metre zone of the east arm than at a l l other depths t h i s difference i s s i g n i -ficant- only i n the 21 to 40 metre zone (Table X I I ) . Numbers i n the 0 to 5 metre zone i n the main lake are s i g n i f i c a n t l y more variable than at a l l deeper zones. Regional comparisons indicate that there i s no s i g n i f i c a n t difference i n v a r i a b i -l i t y of chironomid counts between the east arm and main lake. Pelecypoda (sphaeriidae): A marked decrease i n abundance of sp,haeriids occur between the l i t t o r a l and profundal zones i n both portions of the lake (Figure 13). Numbers i n the east arm are s i g n i f i -cantly higher than those from the main lake i n 0 to 5 , 6 to 37 0 - 5 6 -20 2 1 - 4 0 4 1 - 8 0 81-140 D E P T H Z O N E S IN M E T R E S Figure 12. Mean numbers (logarithmic transformation) of chironomids per dredging i n Great Slave lake. Table XII. Comparisons of v a r i a b i l i t y i n numbers (logarithmic transformation) of chirono-mids i n Great Slave lake. Depth zones i n metres 0 - 5 6-20 21 - 40 41 - 80 81 - 140 E A S T Number of samples 25 3 0 3 1 29 39 Mean number of chiro-nomids (logarithmic) 1.3113 .7801 .7108 .9645 .6922 A R M Standard deviation .5050 .3820 .3656 .4278 .3919 \"z\" values compared to 0-5 metre zone .2791 .3228s .1655 .2531 M A I N Number of samples 38 84 43 45 2 6 Mean number of chiro-nomids (logarithmic) .6017 .3974 .5862 .3846 .4567 L A K Standard deviation .4712 .3630 .3194 .3286 .3093 \u00E2\u0080\u00A2- E w z \" values compared to 0-5 metre zone .2608H .3S86H \u00E2\u0080\u00A23598H .4207 \"z\" values compared to 6-20 metre zone .1275 .0998 .1604 \" z M values between regions .0695 .0507 .1354 .2631 .2366 S i g n i f i c a n t at p - 0.05. 39 Figure 13. Mean numbers (logarithmic transformation) of sphaeriids per dredging i n Great Slave lake. 40 20 and 41 to 80 metre zones. Comparison of v a r i a b i l i t y between depth zones indicate that numbers at 0 to 5 metres are more variable than at any succeeding depth zone i n the east arm (Table X I I I ) , Numbers i n the 0 to 5 metre zone of the main lake are s i g n i f i c a n t l y more variable than those below 40 metres. There i s no difference i n v a r i a b i l i t y of sphaeriids between the east arm and the main lake except at 0 to 5 metres where num-bers are more variable i n the east arm. Intergroup comparisons: Comparisons of v a r i a b i l i t y are made between o l i g o -chaetes, amphipods, chironomids and sphaeriids from the two major portions of Great Slave lake (Table XIV). Only compari-sons showing s i g n i f i c a n t differences are given although a l l combinations were tested i n the analysis. In the 0 to 5 metre zone of the east arm and the main lake amphipods and sphaeriids exhibit the greatest v a r i a b i l i t y i n numbers. No s i g n i f i c a n t differences i n v a r i a b i l i t y are evident between east arm groups within 6 to 20 metres, while i n the main lake amphipods and sphaeriids are again the most variable groups. At 21 to 40 metres amphipods and sphaeriids are s t i l l the most variable forms i n the east arm. Sphaeriids and oligochaetes appear to be most variable at t h i s depth i n the main lake. Amphipods exhibit the most v a r i a b i l i t y i n the 41 to 80 metre zone of the east arm, as do oligochaetes i n that zone of the main lake. Below 80 metres amphipods are the most Table XIII. Comparisons of v a r i a b i l i t y i n numbers (logarithmic transformation) of Sphaeriids i n Great Slave lake. Depth zones i n metres 0 - 5 6-20 21 - 40 41 - 80 81 - 140 E Number of samples 25 30 31 2 9 39 . A S Mean number of sphae-r i i d s (logarithmic) 1.3594 1.1044 . 5 7 6 0 .4599 . 1 6 7 7 T A Standard deviation .8124 . 4 6 3 7 . 4 6 2 5 .3931 .2254 R M \"z't values compared to 0-5 metre zones . 5 6 0 7 * * . 5 6 3 0 * * . 7 2 6 1 * * . 9 5 7 0 * * \"z\" values compared to 6-20 metre zone .0100 .1647 , 7 2 1 2 X X M A Number of samples 38 84 43 45 2 6 I N Mean number of sphaeriids (logarithmic) .7769 .5054 .5620 .2046 .2051 L A Standard deviation .5690 .5079 .4700 .3074 *2869 K . E \"z , T values compared to 0-5 metre zone .1133 .1914 X X .6157 .6846** \" z \" values between regions .3563* .0907 .0159 .2453 .2405 ^ S i g n i f i c a n t at p = 0.05. X X \u00E2\u0080\u00A2 ' S i g n i f i c a n t at p = 0.01 Table XIV. Comparisons of v a r i a b i l i t y between groups of bottom fauna i n Great Slave lake. Standard \" Depth zones i n metres deviations of groups 0 - 5 6-20 21-40 41 - 80 81 - 140 Oligochaetes .6863 .4982 .3359 .3582 .3490 E' Amphipods .9403 .4511 .5507 .5019 .4626 a s t A r m Chironomids .5050 .3820 .3656 .4278, .3919 Sphaeriieds .8124 .4637 .4625 .3931 .2254 \"z\" between A-C \"z\" between A-0 \" z n between A-S \"z\" between S-C \"z\" between 0-S .5944** .4756* .4094* .4941** .3192* . 3 3 7 2 * .6956** .5533** .4369** M a Oligochaetes .4166 .4304 .4597 .3792 .5382 Amphipods .6058 . 7 3 6 5 .3450 2^848 .5595 i Chironomids .4712 .3630 .3194 .3286 .3093 n Sphaeriids .5690 .5079 .4700 .3074 ... .2869 L a k e \"z\" between A-C \"z\" between A-0 \"z'.' between A-S '-'z\" between S-C \"z\" between 0-C \"z\" between 0-S .3743* .3119 .3716* .3357* .3039* .3359* . 3 6 3 9 * .2852* .X3\u00E2\u0082\u00AC .5928 .6678** .5538** .6286** S i g n i f i c a n t at p = 0.05. ^ S i g n i f i c a n t at p = 0.01. 43 variable group i n the east arm while amphipods and oligochaetes are s i g n i f i c a n t l y more variable than either chironomids or sphaeriids i n the main lake. From t'he previous analysis i t i s evident \"that amphi-pods are by f a r the most variable group at nearly a l l depth.' zones i n both portions of Great Slave lake. Sphaeriids are the next most variable group between depths of 0 to 40 metres i n the east arm and main lake. Oligochaetes generally are more variable than chironomids at a l l depth zones. .Below 40 metres they become the second most variable group, i n some cases being even more variable than amphipods. Chironomids show less v a r i a b i l i t y i n numbers than any other group when a l l depth zones are considered. Summary of Analyses Analysis of -Great Slave lake data was i n i t i a t e d to determine the degree of v a r i a t i o n i n numbers of bottom organ-isms i n an extensive series of dredgings taken from a large lake. Variation i n numbers was f i r s t considered f o r a l l organisms i n r e l a t i o n to depth, and was found to be greatest i n l i t t o r a l and s u b l i t t o r a l zones. Variation was shown to be greater i n depth regions where marked changes i n abundance of bottom organisms were taking place. Comparison between dredgings taken i n d i f f e r e n t 4 4 portions of Slave lake showed that there were s i g n i f i c a n t d i f -ferences i n v a r i a b i l i t y of samples between l o c a l i t i e s as w e l l as s i g n i f i c a n t changes i n t o t a l abundance of organisms. Qualitative separation of t o t a l numbers indicated that oligochaetes, amphipods, and sphaeriids were more variable i n the l i t t o r a l and s u b l i t t o r a l zones than at greater depths. Likewise these three groups appeared to be more variable i n the east arm than i n the main lake. For chironomids, however, increased v a r i a b i l i t y within the l i t t o r a l zones was indicated only i n one region, and no s i g n i f i c a n t differences i n v a r i a -b i l i t y were noted between locations. Marked differences i n v a r i a b i l i t y were evident bet-ween the four groups of bottom organisms. Amphipods were the most variable, numerically, followed by sphaeriids and then oligochaetes. Chironomids appeared to be the least variable of a l l groups examined. The v a r i a b i l i t y of numbers and weights of bottom organ-isms i n Slave lake data i s not excessive when compared to that i n Kootenay lake data. Marked changes i n abundance of organ-isms with depth and location may be evaluated with reasonable confidence thus permitting a r e l i a b l e estimation of t o t a l bot-tom fauna production. I t i s evident, however, that many dredgings are required from a large lake before a r e l i a b l e measure of abundance and v a r i a b i l i t y may be obtained to guide further sampling procedures or analyses. 45 ANALYSIS OF FACTORS CONTRIBUTING TO VARIATION In the previous section v a r i a b i l i t y i n numbers within ' series of representative bottom samples was examined and the extent of t h i s v a r i a t i o n was considered i n respect to depth, l o c a l i t y and type of organisms. An analysis of the factors contributing to the v a r i a b i l i t y i s made, considering i n part the previous data, data collected by the w r i t e r and that a v a i l -able i n the l i t e r a t u r e . OPERATION OF SAMPLING APPARATUS Sampling from a population introduces sources of error inherent i n the design and operation of the sampling apparatus. This error may not be random. Dredging The f i r s t t r u l y quantitative bottom dredge was i n t r o -duced by Ekman i n 1911. Peterson designed a grab dredge i n the same year which was applied to marine sampling at f i r s t , but l a t e r became of value i n sampling p a r t i c u l a r bottom types i n lakes. After the Birge modification of the Ekman dredge 46 i n 1922, the l a t t e r was widely used throughout North America i n most quantitative studies. Today, supplemented at times with the Peterson grab, i t has become a standard piece of quantitative limnological equipment for sampling macroscopic bottom fauna. In addition to obvious sources of error i n the opera-t i o n of an Ekman dredge such as incomplete closure of the jaws, sp i l l a g e over the top, and leakage through the sides, there are other mechanical features evident i n both construction and operation of the dredge which may introduce a non-random error into sampling. Penetration into bottom sediments may be of importance depending upon the v e r t i c a l d i s t r i b u t i o n of bottom organisms. Lenz (1931), by construction of v e r t i c a l sectioning modification of the Ekman-Birge dredge, investigated the v e r t i -c a l d i s t r i b u t i o n of macroscopic organisms in, the bottom sedi-ments of seven German lakes. He showed that the macroscopic fauna was not l i m i t e d to the upper sediments but extended to depths of more than 20 cm. Lenz demonstrated that there were -differences i n the v e r t i c a l d i s t r i b u t i o n of various components of bottom fauna. Small chironomids were generally found i n the uppermost layers while large chironomid larvae burrowed in t o deeper layers. T u b i f i c i d s were scattered throughout a l l la y e r s . Lenz considered that the Ekman dredge was too short and l i g h t f o r sampling many types of bottom; under best of conditions penetration r a r e l y exceeded 24 cm. With frequent penetration of bottom organisms to 47 depths beyond 20 cm. i t i s possible that the effectiveness of the dredge may vary with r e l a t i v e l y small fluctuations i n the consistency of bottom sediments and the penetration of bottom organisms into these sediments. In addition, the curved jaws do not sample equal v e r t i c a l sections w i t h i n edges of the dredge (Figure 14). In penetrations of 20 cm. or more, effects of unequal v e r t i c a l sampling are probably i n s i g n i f i c a n t . . Greater variations i n t h i s factor may be expected when penetra-t i o n of the dredge i s low. However, i t should be noted that a p o s i t i v e c o r r e l a t i o n may exis t between the penetration of the dredge and bottom organisms i n t o the substrate. In t h i s case low penetrations of)the dredge may not s i g n i f i c a n t l y affect complete sampling of fauna i n the area enclosed. A p a r t i t i o n i n g device was constructed to determine the importance of t h i s source of v a r i a b i l i t y i n sampling. A nine-inch Ekman dredge was part i t i o n e d i n t o nine three-inch squares by i n s e r t i o n of a metal l a t t i c e (Figure 14). Separa-t i o n of the nine subsamples was effected by opening the dredge into a'specially constructed tray (Figure 14). Three s l i d i n g panels on the bottom of the device permitted separate removal and screening of each three-inch subsample. The t o t a l nine square inch area sampled by the par-t i t i o n e d dredge may be divided into three rectangular s t r i p s , two along the sides of the dredge and one central portion beneath the closing margin of the jaws. Each of these s t r i p s sample an equal area of bottom, 27 square inches, but because 4 - 8 Figure 14. P a r t i t i o n i n g apparatus. Upper: Standard nine-inch Ekman-Birge dredge pa r t i t i o n e d by metal l a t t i c e into three-inch subsamples. Lower: Tray f o r separating three-inch subsamples from pa r t i t i o n e d dredge. 49 of curved jaws on the dredge do not take an equal volume of bottom material. Regardless of dredge penetration beyond a depth of two inches, an edge section samples approximately 22 cubic inches l e s s than the centre section. By t h i s d i v i s i o n of the sampling area i t i s possible to determine whether the difference i n volumes sampled by marginal and central portions of the dredge i s r e f l e c t e d i n the numbers of organisms taken i n those; sections. Ten dredgings were taken within a c i r c l e of 20 feet i n diameter at each of four depth regions i n Hatzic lake (Figure 15). At depths of 45 and 25 feet penetration into the bottom sediment was generally over s i x inches. Substrate i n these regions was composed of a soft grey-brown mud contain-ing small amounts of organic d e t r i t u s . At 15 feet penetrations ranged between four to s i x inches. In t h i s area the bottom substrate was s i m i l a r to that i n deeper regions, although greater amounts of organic d e t r i t u s were present. Bottom de-posits at depths of two feet were composed of firmer mud than that i n deeper regions and penetrations ranged between two to three inches. Mean numbers of organisms taken i n the three portions of the dredge are given f o r the four depth regions (Table XV). There i s no s i g n i f i c a n t difference i n numbers of organisms taken by centre or edge p a r t i t i o n s of the dredge at depths of 45, 25, and 15 feet .At .'.two feet, however, there are s i g n i f i -cantly fewer organisms i n the centre p a r t i t i o n than the second Figure 15. P o s i t i o n of 1951-52 sampling areas i n Hatzic lake. 51 edge partition. Although significance cannot be demonstrated for the greater number of organisms in the f i r s t partition, the mean number in the latter does not dif f e r significantly from that in the second edge partition. These are good indica-tion then, that the centre of the dredge i s taking fewer organ-isms per dredging than eitiher edge at this depth. If the dredge was taking a complete sample the reverse situation might be expected i n penetrations of the ordery. for i t has been shown that the centre portion encloses a larger volume than the edges. It was noted when sampling, however, that the jaws of the dredge did not completely close on release of the tripping mechanism, u n t i l the dredge began to be pulled out of the mud. Thus the centre section of the partitioned dredge actually may have been sampling less volume of bottom material than either edge sections. Table XV.. Mean numbers of organisms in partitioned dredgings in Hatzic lake. Depth Penetration M e a n number of organisms of of Edge Centre Edge samples dredge Partition 1 Partition Partition 2 (feet) (inches) 45 +6 22.9 23.2 25.0 25 +6 18.7 18.6 17.8 15 4-6 21.9 26.2 23.9 2 2-3 33.6 29.2 , 37.8 52 Table XVI. Qualitative separation of mean numbers of organ-isms i n partitioned dredgings at 2 feet depth i n Hatzic lake. Organisms Mean numbers of organisms Edge P a r t i t i o n 1 Centre P a r t i t i o n Edge P a r t i t i o n 2 Oligochaetes 18 .9 1 4 . 5 1 9 . 5 Chironomids 9 . 1 1 0 . 2 1 1 . 4 Ceratopogonids 4 . 4 3 . 5 6 . 0 Miscellaneous 1 . 2 1 . 0 0 . 9 Total 3 3 . 6 2 9 . 2 37.8 Qualitative d i v i s i o n of t o t a l numbers of organisms show that chironomids and the miscellaneous group including hydracarinids, nematods, and mysids exhibit l i t t l e difference between p a r t i t i o n s (Table XVI). Lenz (1931) demonstrates that small chironomid larvae are generally found i n the f i r s t few centimeters of bottom material. Furthermore hydrarinids and mysids are not found below the surface of the bottom substrate. Undoubtedly deposits immediately adjacent to the surface are equally sampled by edge and centre p a r t i t i o n s of the dredge and therefore, as has been shown, l i t t l e difference i n numbers of these forms would be expected between p a r t i t i o n s . Cerato-pogonids and oligochaetes, the l a t t e r e s pecially, contribute; most to the observed difference between p a r t i t i o n s . As Lenz 53 (1931) has shown, oligochaetes penetrate r e l a t i v e l y deep i n t o bottom sediments and thus fewer of these forms would be expec-ted i n the centre p a r t i t i o n where incomplete sampling of the lower l e v e l s of substrate occurs. Several sources of non-random v a r i a t i o n are apparent i n mechanical construction and operation of the Ekman dredge. Regardless of the closing mechanism, penetration of the dredge i n r e l a t i o n to v e r t i c a l d i s t r i b u t i o n of the bottom fauna may introduce a non-random source of v a r i a t i o n . When the closing mechanism of the dredge i s considered i t i s evident that the dredge may not equally sample organisms i n the area which i t encloses. The degree of va r i a t i o n related to t h i s factor depends upon penetration of the dredge, depth at which i t s jaws completely close, and penetration of the organisms into the substrate. In regions of fi r m substrate, often associated with shallow depths, mechanical features of the dredge may i n t r o -duce a s i g n i f i c a n t error into sampling procedures. Screening and Separating Removal of animals from bottom material, involving the use of screens, introduces a source of v a r i a t i o n related to the siz e and shape of the organisms. The number of organ-isms retained i n t h i s separation depends upon the mesh size of the f i n a l screen. Rawson (1930) used a f i n a l screen of approximately 37 meshes per l i n e a r inch i n studies on Lake Simcoe. A f i n a l screen of 30 to 42 meshes per l i n e a r inch i s 54 commonly used in ;recent studies of macroscopic bottom fauna (Deevey 1941, B a l l 1948, and Wohlschlag 1950). Where very fin e screens are used removal of organisms by picking becomes a laborious, time-consumming task because of large amounts of retained organic and inorganic material. A 30-to 40-mesh per l i n e a r inch sdreen probably takes a l l forms which contribute s i g n i f i c a n t l y to the t o t a l weight of organisms i n a sample. Finer screens not only increase error i n picking but also reduce the number of samples which may be taken within a given time. Some types of bottom material retained i n screens may be separated from organisms by density f l o a t a t i o n processes. A sodium chloride f l o a t a t i o n separation described by Lyman (1943) was found to be unsatisfactory. Where large amounts of inorganic materials are retained i n screenings an e f f e c t i v e separation may be made with a 70 per cent sodium s i l i c a t e s o l -ution. A l l organisms including sphaeriids and gastropods f l o a t to the surface i n t h i s solution where they may be separated from the denser inorganic materials. The method i s not p r a c t i -c a l , however, when large amounts of plant d e t r i t u s are included with screenings. Most e f f i c i e n t separation of organisms from bottom materials i s probably made by the use of a screen with 30 to 40 meshes per l i n e a r inch and subsequent picking of retained material while a l i v e , supplemented by a sodium s i l i c a t e f l o a t -ation where p r a c t i c a l . Variation related to separation 55 processes i s probably greatest where large amounts of organic material are,retained by the screen. This s i t u a t i o n generally occurs i n samples taken i n or adjacent to the l i t t o r a l zone where concentrations of organic detritus are high. HETEROGENEITY WITHIN SAMPLING AREA Several i n t e r r e l a t e d factors operate to produce a heterogeneous and variable sampling area on a lake bottom. Re-cognition of these factors and some appreciation of t h e i r effect on bottom fauna populations i s of major importance i n design of e f f i c i e n t sampling procedures. Depth The i n t e r r e l a t i o n s of temperature, oxygen supply, l i g h t penetrations, bottom deposition, and indeed most physi-c a l , chemical and b i o l o g i c a l factors with depth r e s u l t i n a complex v e r t i c a l zonation of bottom organisms. Most studies on d i s t r i b u t i o n of bottom organisms have re a l i z e d the import-ance of depth and have divided the bottom into several depth zones of various si z e s . The d i s t i n c t i o n of these zones, how-ever, has been somewhat ar b i t r a r y and related only i n part to changes occurring i n the physical, chemical, and b i o l o g i c a l aspects of the environment with depth. Generally lake bottoms have been divided into three 56 major zones, the l i t t o r a l , s u b l i t t o r a l , and profundal. Eggleton (1931) defines the l i t t o r a l zone as that portion bet-ween the shoreline and the outward l i m i t of rooted aquatic vegetation, the profundal as that portion from the greatest depth up to the average upper l i m i t of the hypolimnion; and the s u b l i t t o r a l as the remaining portion between the l i t t o r a l and profundal. Obviously the p o s i t i o n of these zones w i l l vary not only between lakes but also w i t h i n any given lake depending on the position of the thermocline and development of rooted aquatics. ' Di s t i n c t i o n of boundaries may be of primary concern i n development of sampling procedures. Neyman (1934) has shown that the most e f f i c i e n t sampling of a heterogeneous \"universe\" i s accomplished by taking a number of random samples from each zone proportional both to the area of the zone, and the standard deviation within that zone. An experimental plot was set up i n Hatzic lake i n order to examine changes i n numbers of bottom fauna which took place between, close depth i n t e r v a l s on a f a i r l y constant bot-tom type. The plot was 50 feet i n width and extended 6 2 6 feet off-shore to -\u00E2\u0080\u00A2:<*.:maximum depth i n the lake pi 45 feet (Figure 15). A t o t a l of 150 six-inch Ekman dsedgings was taken i n t h i s p l o t between September 5 to 8, 1951. Dredgings were taken i n series of f i v e at any one depth across the p l o t , with a maximum depth i n t e r v a l of two feet between s e r i e s . A l l samples were washed through a 30-mesh per l i n e a r inch screen. 57 Average numbers of organisms at each depth zone i n the plot are given i n Figure 1 6 . A marked decrease occurs i n the abundance of organisms between the shore and nine feet; numbers between zero to two feet are s i g n i f i c a n t l y higher than those i n the two and one half to nine feet depth zone. Between nine and 13 feet a very marked increase i n numbers of organisms i s evident, followed by a gradual decline i n abundance with further increase i n depth. Rapid changes i n numbers of organisms with depth i n t r o -duces an important consideration i n design of sampling proce-dures and subsequent analysis of dredging data. I t i s seen that threefold changes i n abundance of organism may take place wi t h i n a depth i n t e r v a l of two feet (Figure 1 6 ) ; also l e s s marked but s t i l l s i g n i f i c a n t changes may occur between depth i n t e r v a l s . o f less than two feet. Unless r e l a t i v e l y large num-bers of samples are taken at close depth i n t e r v a l s , those changes w i l l be obscured. Further, i f samples are lumped into r e l a t i v e l y large, more or less arbitrary depth zones, these changes w i l l be attributed to sampling v a r i a b i l i t y . Analysis of the data from the Hatzic lake plot would indicate that the v a r i a b i l i t y of numbers of bottom organisms i n any given area may be greatly increased by i n s u f f i c i e n t sampling of a l l depths, or by grouping of counts i n t o a r b i t r a r y or too i n c l u s i v e depth zones. I t would appear then that large numbers of samples may be required at r e l a t i v e l y closer depth i n t e r v a l s than have been taken i n most studies. 2 ON H c t d a> \u00C2\u00A3 CO CD 3 -cu 3 CD p. 01 C+ cr o B X - (D M vO H p. W-O O o c+ ct O S 3 ct N H< O 05 \u00E2\u0080\u0094 O R G A N I S M S \u00E2\u0080\u0094 T E M P E R A T U R E \u00E2\u0080\u0094 O X Y G E N 125 i o o 2 < a. 75 5 0 2 5 < z LJ o > X o H Z Ld u tr u a. 10 15 2 0 2 5 D E P T H IN F E E T 3 0 3 5 4 0 4 5 59 The possible causes for low abundance of bottom organ-isms between depths of two and one half to nine feet are worthy of consideration. Chironomids and oligochaetes are the most important groups of fauna i n t h i s area. Although nema-tods, mysids, ceratopogonids, and hydracarinids are present, they occur only r a r e l y i n the dredgings and do not contribute s i g n i f i c a n t l y to the t o t a l fauna. Numbers of both chironomids and oligochaetes are at a minimum between two and one h a l f and nine feet (Figure 17). The postulation of a s u b l i t t o r a l minimum i n the s t r i c t sense was not v a l i d as t h i s region was s t i l l well within the 3-ittoral zone. The consideration that the minimum could r e s u l t from f i s h predatioh seemed plausible. Lundbeck, i n 1926, according to Deevey (1941), suggested that predation by f i s h may be responsible f o r s u b l i t t o r a l minima apparent i n German lakes. B a l l (1948) showed that f i s h populations may e f f e c t -i v e l y reduce the volume of invertebrates per unit area i n the l i t t o r a l zone. Coarse-scaled suckers (Catastomus macrocheilus), Carp (Cyprinus carpio). Chub (Mylocheilus caurinus). Squawfish (Ptychocheilus oregoriensis). Catfish(Amelmrus nebulosus), Calico Bass,- (Pomoxis nigro-maculatus) and Sculpins, (Cottus asper) are present i n Hatzic lake. A l l these species feed upon bottom organisms, some almost exclusively; thus predation by f i s h may be an important source of depletion of a bottom fauna population. 60 Figure 17. Depth d i s t r i b u t i o n of oligochaetes and chironomids i n Hatzic lake. September 5 - 8, 1951. 61 The d i s t r i b u t i o n of temperature and dissolved oxygen i n the lake at t h i s time is' of importance. Fish are probably r e s t r i c t e d from the deeper waters of Hatzic lake due to the serious oxygen depletion present (Figure 1 6 ) . High tempera-tures (Figure 1 6 ) as well as other unfavourable factors such as exposure to predation probably prohibits f i s h from extensive feeding i n the extreme shallow water zone (zero to two f e e t ) . Thus predation by f i s h may be a major factor i n the marked decrease i n abundance of bottom organisms at t h i s depth i n t e r -v a l . Substrate In the previous section v a r i a t i o n i n abundance was considered from the shoreline to deeper waters. Variations occurring between d i f f e r e n t areas within a depth zone may be examined now. In the analysis of Kootenay and Great Slave lake data s i g n i f i c a n t differences were noted i n abundance of bottom or-ganisms between major portions of the lakes. Although physi-cal-chemical features of the water as well as other ecological factors were undoubtedly concerned i n these horizontal d i f f e r -ences, changes i n the nature of bottom substrate were likewise of importance. Baker (1913), Adamstone and Harkness (1923), Rawson (1930), Krecker and Lancaster (1933)j and Roelofs (1944) have correlated changes i n abundance of bottom organisms with the type of substrate. 6 2 As with changes taking place v e r t i c a l l y , so too with horizontal changes i t i s necessary to delim i t the boundaries within which they occur i f most e f f i c i e n t methods of sampling are to be employed. Recognition of boundaries immediately demands some scheme of c l a s s i f i c a t i o n f o r bottom deposits. In most studies c l a s s i f i c a t i o n of substrate has been based upon arb i t r a r y size and colour d i s t i n c t i o n s of the p a r t i c u l a r matter. Thus \"boulders\", \" f l a t rubble\", \"gravel\", \"sand\", \"blue clay\", \"brown mud\", \"black ooze\" are some of the bottom types d i s t i n -guished. Welch (1935) notes that a sat i s f a c t o r y c l a s s i f i c a t i o n i s yet to be made. L i t t l e i s found i n more recent l i t e r a t u r e to improve the s i t u a t i o n ; Roelofs' (1944) c l a s s i f i c a t i o n i s b a s i c a l l y the same as previous ones. Eight nine-inch Ekman dredgings were taken at each of four areas i n Hatzic lake on October 13, 1951. A l l dredgings were taken at a depth of two feet (Figure 15). Types of bot-tom substrate represented were (a) brown mud i n a region of scattered Ceratophyllum demersum, (b) brown mud adjacent to the mouth of a small creek, (c) sandy mud, (d) barren region of brown mud. I t i s immediately apparent that there are marked differences i n t o t a l numbers of organisms among the four areas (Table XVII). Total counts are transformed to the square root of the count f o r comparisons between means. There are s i g n i -f i c a n t l y fewer organisms at area C, the sandy mud, than i n any of the three other areas. Area B, adjacent to the creek mouth, has s i g n i f i c a n t l y fewer organisms than areas A or D. Thus s i g n i f i c a n t changes may occur i n the abundance of bottom 6 3 Table XvTI. Mean number of organisms at sampling areas i n 2 foot depth zone, Hatzic lake. Sampling area A B G D Number of samples 8 8 8 8 Mean number of organisms 5 2 . 6 2 4 2 . 2 5 1 1 . 5 0 6 4 . 5 0 Transformed means (square root) 7 . 2 7 6 . 4 8 3 . 2 7 7 . 9 5 F i d u c i a l l i m i t s at 0 . 0 5 l e v e l i . 5 7 7 \u00C2\u00B1 . 3 9 3 \u00C2\u00B1 . 7 1 6 \u00C2\u00B1 . 9 9 3 organisms between closely adjacent areas within the same depth zone. These changes would increase the v a r i a b i l i t y of samples drawn at random from a l l areas as opposed to s t r a t i f i e d sampl-in g at each area. Further study was made on the nature' of bottom depo-s i t s within the four areas. A l l material which did not pass through the 30-mesh per l i n e a r inch screen was l a t e r divided i n t o organic and inorganic components by means of a density separation. By measuring penetration of the dredge into the bottom i t was possible to estimate the amount of material passing through the screen, but no information regarding the organic and inorganic content of the l a t t e r was obtained. Comparisons of r e l a t i y e proportions of organic and inorganic material retained i n screenings from the four regions were made (Table XVIII). Relative amounts of organic matter 64 were high i n area B, adjacent to the creek mouth but low i n area C, a region of more compact sandy mud. The q u a l i t a t i v e nature of organic material retained i n screenings of samples from the four areas i s shown i n Figure 1Q. Organic material at area B, adjacent to the creek mouth, i s i n a primary stage of decomposition compared with the more advanced state evident i n material at other areas. The amount of available organic matter i n bottom sub-strates i s one of the important l i m i t i n g factors i n bottom fauna production (Rawson, 1930). Thus the low abundance of organisms at area C may be related to low amounts of organic material i n that region. In addition the firmer nature of the substrate as indicated by decreased penetration of the dredge (Table VIII) may phys i c a l l y l i m i t the penetration of organisms below the immediate surface layers. Although large amounts of organic matter are present at,area B t h i s material i s for the most part i n preliminary stages of decomposition related to recent introduction into the lake. Lower abundance of bottom organisms i n t h i s area com-pared with areas A and D may be a re s u l t of the undecomposed nature of the substrate and s h i f t i n g environmental conditions adjacent to the creek mouth. The greater abundance of bottom organisms at areas A and D may be related to high amounts of organic matter and the advanced state of i t s decomposition. Area B X $ Figure 18. Quality of organic material retained in bottom sample screenings from four areas in Hatzic lake. 66 Table XVIII. Comparisons of organic and inorganic material retained i n regional samples at 2 feet i n Hatzic lake. Sampling area A B C D Average penetration of dredge (inches) 6 7 3 5 Average volume i n t o t a l sample 7136 cc. 8464 cc. 3163 cc. 5808 cc. Nature of f r a c t i o n (0, organic; Inor., inorganic) 0. Inor. 0 Inor. 0. Inor. 0. Inor. Average volume retained by screen (cc.) 360 54 720 8 20 12 141 34 Percentage of t o t a l volume 5.0 0.8 8.5 0.1 0.6 0.4 2.4 0.6 < Ratio of organic/inorganic 6.2 85.0 1.5 4.0 CHANGES WITH TIME Fluctuations i n abundance of bottom organisms asso-ciated with time may be d i u r n a l , seasonal, or annual i n extent. The importance of temporal changes are to be discussed i n r e l a t i o n to t h e i r effect on v a r i a b i l i t y of sampling by consi-deration of data i n the l i t e r a t u r e and from Great Slave lake. 67 Diurnal Diurnal fluctuations i n abundance have been found i n those bottom organisms which are independent of the substrate during some stage i n t h e i r l i f e cycle. Thus such fluctuations have not been reported i n numbers of adult nematodes, o l i g o -chaetes, pelecypods, hirudinians or gastropods. Diurnal f l u c t -uations occur i n cer t a i n groups of crustaceans and insects. Larkin (1948) noted marked diurnal migrations of Mvsis r e l i c t a i n Great Slave lake. Similar migrations were also observed i n the amphipdd Pontoporeia a f f i n i s from the same lake. Rawson (1930) confirmed previous observations that Corethra larvae exhibit nocturnal upward migrations. M i l l e r (1941) showed that the maximum emergence of chironomid adults above the thermocline occurred between four and seven A.M., although no diurnal fluctuations i n emergence were noted below the thermocline. This s i t u a t i o n could r e s u l t i n diurnal f l u c t u a -tions i n abundance of chironomid pupae on the bottom above the thermocline:; however, no information was given by M i l l e r con-cerning the abundance of pupae. Apparently diurnal changes i n the abundance of bottom organisms may be r e s t r i c t e d to a few groups of crustaceans and insect larvae^. However, more study should be made to deter-mine t h e i r extent i n a l l groups of bottom organisms and i t s effect on sampling v a r i a b i l i t y . 68 Seasonal The extent and p e r i o d i c i t y of seasonal fluctuations i n abundance of bottom organisms are closely associated with the l i f e h i s t ory of the species concerned. Eggleton (1931) and Deevey (1941), however, show that seasonal fluctuations i n num-bers of oligochaetes are of minor extent when the v a r i a b i l i t y of samples are considered. The abundance of most insect larvae i s affected by seasonal periods of emergence. Such forms show a d e f i n i t e midsummer minimum when emergence i s greatest and a midwinter maximum in d i c a t i n g growth of the next generation (Rawson, 1930). D i s t i n c t midsummer minimum and midwinter maxi-mum have been demonstrated i n the abundance of Chironomus and Corethra from many lakes (Eggleton, 1931; Deevey, 1941). M i l l e r (1941) shows a close r e l a t i o n between abundance of l a r v a l chiro-nomids and t h e i r periods of emergence. In many of the studies cited above, extensive seasonal changes i n abundance of bottom organisms were noted. Often f i v e f o l d changes i n abundance occurred i n a period of few weeks. Obviously combination: of data collected over a few months even i n a single l o c a l i t y would increase greatly the v a r i a b i l i t y of samples i n that region. Annual Fluctuations of considerable magnitude have been re-ported i n the abundance of bottom organisms from year to year. Aim (1922), found almost twofold changes i n quantity of bottom 69 organisms between two consecutive years. Lundbeck i n 1926, according to Rawson (1930) showed yearly changes of the same order. Eggleton (1937) reported over 3,500 organisms per square metre i n Lake Michigan f o r August, 1931, and less than 800 organisms per square metre i n the same lake i n July, 1932. Larkin et a l (1950) were able to demonstrate s i g n i f i c a n t de-creases i n the mean weight of bottom organisms i n 1931 as opposed to 1949 i n Paul lake. No s i g n i f i c a n t differences could be shown i n the data between the two consecutive years 1948 and 1949. Comparison was made between dredgings taken i n the main portion of Great Slave lake on three consecutive years (Table XIX). Total counts were transformed to logarithms and then treated by an analysis of variance with appropriate cor-rection for unequal subclass numbers. Results of the analysis are given i n Table XX. No s i g n i f i c a n t difference may be demonstrated between years a l -though the effect of depth i s s i g n i f i c a n t , as has been shown previously. Table XIX. Mean numbers of bottom organisms taken i n main portion of Great Slave lake. Years 0 - 5 6 - 1 0 Depth zones i n metres 11 - 20 21 - 40 41 - 60 61 - 140 1944 Number of dredgings Mean number of organisms 8 74.50 9 72.44 15 9 76.40 101.44 8 82.37 7 75.14 1945 Number of dredgings Mean number of organisms 17 89.29 9 79.11 9 13 39.44 51.92 6 69.00 12 48.25 1946 Number of dredgings Mean number of organisms 13 73 .61 15 100.60 25 16 71.56 98.81 10 89.60 12 76.58 Table XX. Analysis of variance on t o t a l numbers of bottom fauna taken on t i v e years i n Great Slave lake. three consecu-Source of Variation Degrees of Freedom Sum Square Mean Square F Depth Years Discrepancy-5 2 197 2.7733 0.3773 42.3679 0.5547 0.1886 0.2150 2.58* 0.88 Sign i f i c a n t at p = 0.05. 71 RANDOM SAMPLING Regardless of v a r i a b i l i t y introduced into a series of samples by mechanical features of the sampling apparatus, heter-ogeneity within sampling area or time and duration of sampling, a certain portion w i l l be associated with a random sampling \"error\" inherent i n a l l sampling methods. The extent of t h i s sampling \"error\" i t s e l f depends upon a number of fa c t o r s . D i s t r i b u t i o n of Fauna The d i s t r i b u t i o n form of a bottom fauna population i s d i r e c t l y r e f l e c t e d i n the variance of a series of samples taken at random from that population. Theoretically, three general types of d i s t r i b u t i o n may be assumed by bottom organ-isms, the random, the contagious or clustered, and the spaced or uniform. In a random d i s t r i b u t i o n (Figure 19) the probabi-l i t y of taking \"n\" (any number) of organisms per sample at one spot i s equal to the p r o b a b i l i t y of taking \"n\" organisms per sample at another spot. In other words, the presence of one organism at a p a r t i c u l a r spot does not affedt the p r o b a b i l i t y of another organism being i n that immediate area. When samples are drawn from a contagious d i s t r i b u t i o n (Figure 19) the Figure 19. Theoret ica l d i s t r i b u t i o n s of 50 bottom organisms i n a square yard of bottom. Upper: random. Centre: contagious. Bottom: spaced. 73 p r o b a b i l i t y of taking \"n\" organisms per sample at one spot i s not equal to that at another spot; the presence of one or-ganism at a p a r t i c u l a r spot increases the p r o b a b i l i t y of another being i n that immediate area. In a spaced or uniform d i s t r i b u t i o n pattern (Figure 19) the presence of one organism at a p a r t i c u l a r spot affects the pr o b a b i l i t y of another being . i n that immediate area i n a negative manner. I f a series of six-inch Ekman samples were taken from a random d i s t r i b u t i o n with s p a t i a l relationships s i m i l a r to Figure 19 the observed d i s t r i b u t i o n would approximate a Poisson d i s t r i b u t i o n i n which the variance equals the mean. I f a series of si x - i n c h samples were taken i n an area of con-tagious d i s t r i b u t i o n similarjuo Figure 19 many blank samples and samples containing large numbers of organisms would be taken as well as intervening numbers r e s u l t i n g from smaller \" c l u s t e r s \" or s p l i t t i n g of groups. In such a d i s t r i b u t i o n the variance would exceed the mean. Similar sampling from a uniform d i s t r i b u t i o n (Figure 19) would give a variance less than the mean. M i l l e r (1936) shows that the bottom fauna as a whole, and Corethra and Chironomus larvae separately, have a random d i s t r i b u t i o n at 20 feet i n Lake Opeongo. No study, however, has been made of the d i s t r i b u t i o n pattern of bottom organisms at a series of depths. Data from nine-inch partitioned dredgings (see \"Dredging\", Page 45) could be used f o r such a study, provided 74 three-inch subsamples were independent. Total counts of organ-isms i n the three subsamples were transformed to ycount * 0.5. These data were treated by analysis of variance to test the effect of rows (three-inch s t r i p s p a r a l l e l i n g closing margin of jaws), columns (three-inch s t r i p s at r i g h t angles to: closing margin of jaws), times (differences between the 10 nine-inch samples at each depth zone), and i n t e r a c t i o n between rows and columns. Results of the analysis (Table XXI) indicated a s i g -n i f i c a n t mean square f o r rows at two feet. I t was noted (see \"Dredging\", Page 45) that the centre row of the par t i t i o n e d dredge took s i g n i f i c a n t l y fewer organisms than edge rows at that depth. Subsamples within dredgings at two feet were not independent and therefore could not be used i n d i s t r i b u t i o n a l studies. No s i g n i f i c a n t mean squares were evident for rows, columns or inte r a c t i o n at 15, 25, and 45 feet depths. S i g n i -f i c a n t mean squares for times at depths of 15 and 25 feet indicated s i g n i f i c a n t v a r i a b i l i t y between dredgings but did not affect independence within the subsamples of a nine-inch dredging. Only three-inch subsamples at depths of 15, 25 and 45 feet were used i n d i s t r i b u t i o n a l l s t u d i e s . Frequency d i s t r i b u t i o n s of oligochaetes were deter-mined i n 90 three-inch subsamples from 10 nine-inch dredgings at depths of 15, 25, and 45 feet i n Hatzic lake. These were compared with three t h e o r e t i c a l d i s t r i b u t i o n forms, the Poisson, negative binomial, and Neyman's contagious calculated from the observed data (Table XXII). Adequate f i t s to each of the 75 Table XXI. Analysis of variance on three-inch subsamples from nine-finch p a r t i t i o n a l dredgings i n Hatzic lake. Source of Var i a t i o n Degrees of Freedom Mean Square at 2 f t . Mean Square at 15 f t . Mean Square at 25 f t . Mean Square at 45 f-Rows 2 1.413 H . 518 .051 .164 Columns 2 .829 .050 .087 .388 Times 9 1.383** .983** .756** .270 Rows x columns in t e r a c t i o n 4 .127 .163 .114 .044 Discrepancy 72 .316 .200 . 2 0 6 .261 d i s t r i b u t i o n forms were indicated at a l l depths. Where there was l i t t l e difference i n pr o b a b i l i t y of f i t i t was considered advisable to accept the most simple of the d i s t r i b u t i o n forms as the better f i t . Thus a random d i s t r i b u t i o n of oligochaetes was indicated at depths of 1 5 , 2 5 , and 4 5 f e e t . Frequency d i s t r i b u t i o n s of chironomids from the same dredgings were compared with Poisson, negative binomial, and Neyman's contagious d i s t r i b u t i o n s calculated from the observed data (Table XXIII). S i m i l a r l y a random d i s t r i b u t i o n of chiro-nomid larvae was indicated at a l l three depths. At 25 feet the variance of samples was le s s than the mean suggesting u n i -formity i n d i s t r i b u t i o n . 7 6 Table XXII. Comparison of oligochaete frequency d i s t r i b u t i o n s i n three-inch subsamples to three t h e o r e t i c a l d i s t r i b u t i o n forms. Depth Number/ Dredging Fre- Poisson Negative -jtegatyfrre quency/ Binomial Contagious 0 0), 1 3P 2 7-1 3 12 5 4 9 5 14 F 6 17 e 7 12 e 8 7. t 9 4 10 0) 11 2). 12 2) 5 13 1) Chi-square (pooled)* Degrees-of Freedom \u00E2\u0080\u00A236)0 2.00r*-*\u00C2\u00B0 5.52-10;14 13.97 15.40 15.40 12.45 8.58 5.25 2.89) .28) .63)., , 0 6.32. 10.24 12.91 13.51 12.19 9.74 7.03 4.65 2.86) .89)5.86 \u00E2\u0080\u00A246) .64)o r c 2.8l);i-Zf:) 8.05 15.95 18.48 19.62 17.73 14.20 10.28 6.83 4.18) 2.39K c 1.29)2\u00C2\u00AB52 .66) rmean - 5.51, \u00E2\u0080\u00A2 0 1 4) 5 2 6 2 3 10 5 4 14 5 14 F 6 11 e 7 11 e 8 9 t 9 4 10 2) 11 2)6 12 2) variance 3 . 7 4 8 = 6 . 4 7 4 4 * 2 6 7 1 1 . 6 9 7 2 . 5 3 .39) 2.14) 5.81 10.52 14.28 15.51 14.04 9.53 6.47 3.90 2.12) 1.05)3.64 -47) 3 . 7 2 . 7 0 ) 3 . 0 2 ) 6 . 8 9 1 0 . 9 2 1 3 . 4 8 1 3 . 8 4 1 2 . 2 8 9 . 6 7 6.90 4 . 5 2 2 . 7 5 ) 1 . 5 7 ) 5 . 1 7 . 3 5 ) 3.08)3*80 6.95 10.96 13.56 13.98 12.51 9.94 ?.19 4.79 2.97) 1.74)5.79 1.08) (pooled) 3 6 6 . 0 5 Degrees of freedom 8 /Mean - 5 . 4 3 , variance - 6.719 1.78 7 1 . 3 7 7 77 Table XXII Cont'd. Comparison of oligochaete frequency d i s -t r i b u t i o n s i n three-inch subsamples to three t h e o r e t i c a l d i s t r i b u t i o n forms. Depth Number/ Fre-Dredging quency/ Poisson Negative Binomial Contagious 0 1 2 3 4 5 6 7 8 9 1G 11 12 13 14 3) 0) 6 1 ) 2 ) 8 1 2 1 6 9 1 2 7 5 7 6 ) 1 ) 8 1 ) Chi-square (pooled) 3 6 Degrees of freedom /Mean =7.15 Variance = 8.650 .07) .50)6, 1.80) 4.30) 7;69 11.00 13.11 13.39 11.97 9.51 6.SO 4.42 2.63) 1.45)4. .74) .13) 5.08) 8.19 10.84 12.27 12.21 10.90 8.86 6.62 4 .61 3.01) 82 1.-85)5.92 1.06) .14) 5.06) 8.10-10.69 12; 07 12.04 10.79 8.81 6 . 6 4 4.66 3.07) 1.91)6.10 1.12) 6 . 9 9 8 5.66 7 5.64 7 Tabled Chi-square values at 0.05 p r o b a b i l i t y l e v e l : g degrees of freedom \u00E2\u0080\u00A2= 15.507 7 degrees of freedom - 14.067 78 Table XXIII. Comparison of chironomid frequency d i s t r i b u t i o n s i n three-inch subsamples to three t h e o r e t i c a l d i s t r i b u t i o n forms. Depth Number/ Dredging Frequency Poisson Negative Binomial Negative' Contagious 0 9 8.36 9.25 9.16 1 23 19.86 19.89 19.62 1 2 1 6 2 3 . 6 1 22.32 22.05 5 3 22 18.70 17.46 17,19 4 12 11.12 10.62 10.53 F 5 4 5 . 2 8 5,40 5,31 e 6 3) 2,09) . ) ) e 7 0) .71)3.01 )5.06 . J6.14 t 8 1 ) .21) ) . ) Chi-square (pooled)* Degrees of freedom ^ x \u00C2\u00AB 2.389 Variance - 2.645 3^ 44 5 4.71 4 4; 84 4 2 5 F e 0 1 2 3 4 2 37 9 2 45.73 30.96 10.48 2.36 e t Chi-square (pooled)* Degrees of ^ x \" .678 Variance freedom - .558 1.74 4 4 5 F 0 1 2+ 52 29 9 44.20 31.38 14.32 57.27 17.47 15.26 60.93 11.42 17.65 e e t Chi-square (pooled)* ^ x - .711 variance = 1.657 3.44 10.66 32.60 ^Tabled Chi-square (0.05 p r o b a b i l i t y l e v e l ) Degrees of freedom: 5 - 11.07 4 - 9.49 79 Size of Samples The sample size i n r e l a t i o n to actual s p a t i a l d i s t r i -bution of bottom organisms w i l l determine the observed d i s t r i -bution as determined from a series of dredging samples. The th e o r e t i c a l contagious d i s t r i b u t i o n i l l u s t r a t e d i n Figure 19 may be considered. I f six - i n c h samples are taken from an area with bottom organisms d i s t r i b u t e d i n t h i s manner, the observed frequency d i s t r i b u t i o n w i l l be contagious. I f a random d i s t r i -bution of the \"c l u s t e r s \" i s assumed and 18 square inch samples were taken i n an area with s i m i l a r d i s t r i b u t i o n , the-actual contagious form of d i s t r i b u t i o n may be eliminated i n observed counts. The effect on observed d i s t r i b u t i o n of taking s i x and nine-inch samples i n an area of contagious d i s t r i b u t i o n may be i l l u s t r a t e d by examining counts i n such areas when the squared d i v i s i o n s are superimposed on Figure 19. The variance also may be affected by sample s i z e , being reduced by larger samples where the population d i s t r i b u t i o n form i s random, and sampling i s confined to that population. Thus considering both d i s t r i b u t i o n and variance,,it i s apparent that there may be an optimum sample size to use when sampling bottom organ-isms within a p a r t i c u l a r area. Both six-and nine-inch Ekman dredges have been used i n quantitative sampling of bottom organisms; larger sample sizes being impractical f o r general f i e l d use. In many i n v e s t i -gations, however, l i t t l e consideration has been given to the '80 most suitable sample s i z e . Further, i f no s i g n i f i c a n t d i f -ference i s evident i n v a r i a b i l i t y of samples from six-and nine-inch dredges, the six - i n c h dredge would be more conven-i s n t to use i n may investigations. Comparison was made of v a r i a b i l i t y i n six-and nine-inch Ekman samples taken at depths of 15, 2 5 , and 45 feet i n Hatzic lake. Counts of organisms from s i x - i n c h dredgings were \"weighted\" to adjust f o r the smaller area sampled. Then a l l counts were transformed to logarithms to correct f o r proportionality between means and variances. Standard devia-tions of the respective means were compared by Fisher's \"z\" te s t (Table XXIV). No s i g n i f i c a n t differences i n the v a r i a b i l i t y of s i x -and nine-inch samples at depths of 15 and 25 feet were evident. At 45 feet, however, six - i n c h samples were s i g n i f i c a n t l y more variable than nine-inch samples. At depths of 15 and 25 feet estimates of bottom fauna abundance based on six-and nine-inch samples would approxi-mate each other closely (Table XXIV). At 45 feet, however, an estimate based on s i x - i n c h samples would be considerably lower than that of nine-inch samples. Although the d i s t r i -bution of oligochaetes and chironomids was random at fihis depth (Table XXII and XXIII), possibly the s p a t i a l distance between organisms was great enough to affect numbers and v a r i a b i l i t y w ithin a six-inch sample size more than nine-inch samples. Table XX1Y. Comparisons of v a r i a b i l i t y between six-and nine-inch Ekman samples i n Hatzic lake. Six-inch samples Nine-inch samples Number Mean Weighted Weighted Stan- Number Mean Mean Stan- Fisher's - of num- mean mean dard of num- num- dard \"z\" samples ber (logar- devi- sam- ber ber devi-of ithmic) ation ples of (log- ation organ- organ-- a r i -isms isms thmic) 15 feet 10 33.2 74.6 1.8499 .1571 10 72.1 1.8462 .1091 .3646 25 feet 10 27.4 61.4 1.7723 . 1 2 6 1 10 55.2 1.7308 .0872 . 3 3 2 0 45 feet 10 13.1 29.4 1.4195 .2177 10 71.2 1.8496 .0386 1.7306 Si g n i f i c a n t at p = 0.01. 82 Number of Samples There i s no general agreement i n the l i t e r a t u r e con-cerning the number of samples that should be taken i n order to estimate numbers of bottom organisms even within a l i m i t e d area of a lake. Rawson (1930) indicates on the basis of samples taken at a depth of nine metres that counts of Chironomus plumosus varied by a maximum of 6 4 per cent from a mean of 25, and emphasizes the necessity of \"large numbers of samples\". Eggleton (1931) likewise stresses the importance of numerous samples, combining at leas t f i v e dredgings from one l o c a t i o n , and often including 10, 15, or even 50. M i l l e r (1936) considers that \" f i v e dredgings are a r e l i a b l e sample from one type of bottom\". Deevey (1941) presents data to show that r e l a t i v e l y small numbers of single samples from a lake may give r e l i a b l e estimates of bottom fauna abundance at the 0.05 confidence l e v e l . He shows that the 0.05 confidence l i m i t s of mean; number of organisms from several lakes do not seriously overlap. That significance may be demonstrated between mean numbers obtained by a few single samples from d i f f e r e n t lakes does not affirm the r e l i a b i l i t y of single sampling. Also Deevey points out that these s i g n i f i c a n t d i f -ferences are only evident where the counts per sample are large. The precision with which the mean of a series of bottom samples may be defined, as measured by the standard 83 error, i s not necessarily increased by taking large numbers of samples. Snedecor (1946) points out that greater p r e c i -sion only accompanies increased numbers of samples when sam-p l i n g i s confined to a single homogeneous population; when extended beyond these bounds, precision may decrease with f u r -ther sampling. When heterogeneity within a sampling area may be d i s -tinguished and delimited, the most precise estimate of the population mean may be obtained by taking several small num-bers of samples each proportional to the area and standard deviation of the population within a homogeneous portion. This method of sampling i s more e f f i c i e n t than \"broadcasting\" many single samples over'the area, or taking a few r e s t r i c t e d large samples. 84 GENERAL DISCUSSION Sampling methods applied to quantitative studies of \" bottom organisms are of necessity d i r e c t l y concerned with v e r t i c a l zonation within a lake. Further, both the type and influence of factors contributing to v a r i a b i l i t y change i n r e l a t i o n to depth. Greatest v a r i a t i o n i n numbers of bottom organisms i s usually observed i n the l i t t o r a l zone. This has been demon-strated s i g n i f i c a n t l y i n the Great Slave lake data. An Ekman dredge does not sample proportionally the enclosed area of bottom at shallow depths i n presence of firmer substrates and lower penetrations of the dredge., Heterogeneity within the sampling area i s also greatest i n the l i t t o r a l zone. Var-i a t i o n may take place v e r t i c a l l y , between the shore and out-ward l i m i t of the l i t t o r a l zone, or h o r i z o n t a l l y between regions within that zone. Threefold changes i n abundance of organisms may occur within a v e r t i c a l distance of two feet or l e s s i n Hatzic lake while over fourfold changes occur between locations at the same depth. Fluctuations with time i n physi-c a l and chemical factors as w e l l as b i o l o g i c a l f actors, such as predation, may affect the l i t t o r a l bottom fauna populations. D i v e r s i t y of environmental conditions within the l i t t o r a l zone probably resul t s i n a greater number of d i f f e r e n t kinds of 85 animals i n that region than i s found i n either s u b l i t t o r a l or profundal zones (Welch, 1935). Many of these forms such as amphipods and sphaeriids may be r e s t r i c t e d almost e n t i r e l y to the l i t t o r a l zone i n certain lake types. I t has been shown previously that amphipods and sphaeriids are more variable than any other group of bottom organisms i n Great Slave lake. The problems of quantitative sampling within the l i t -t o r a l zone are obviously complex. That portion of v a r i a b i l i t y associated with d i f f e r e n t groups of organisms can not be avoided. Quantitative sampling i n the presence of large amounts of aquatic vegetation cannot be accomplished with a standard Ekman dredge ( B a l l , 1948). I t appears from analyses of p a r t i -tioned dredge data that a dredge with more powerful jaws i s required i n certain types of mud bottom where the standard Ekman-Birge dredge has generally been considered s a t i s f a c t o r y . Although no s i g n i f i c a n t differences are noted i n v a r i a b i l i t y of samples from s i x or nine-inch dredges within t h i s region, the nine-inch dredge, because of i t s greater weight and more powerful jaws, may be more s a t i s f a c t o r y for sampling firmer substrates. A large part of sampling v a r i a b i l i t y introduced by extreme v e r t i c a l and horizontal heterogeneity of substrate within the l i t t o r a l zone may be avoided by s t r a t i f i c a t i o n of sampling. Ideally several small samples should be. taken from each homogeneous area, the number i n each portion being pro-portional both to the size of the area, and the standard 86 i deviation of the population within that area. Previous know-ledge regarding variation is not available generally, but approximations through information concerning range are pos-sible (Snedecor, 1946). Thus as sampling may be more variable on certain types of substrate as illustrated in the Hatzic lake data, a larger number of samples should be taken in those regions than in other portions of equal area where less varia-bility was evident. The sublittoral zone is in general a region of tran-sition between the littoral environment and the profundal. It is, by definition, a zone varying both, in position and area. Marked changes in the abundance of bottom organisms have been noted within the sublittoral regions of many lakes. Deevey (1941) summarizes the vertical distribution of bottom fauna into three general types, all showing well defined changes within the limits of the sublittoral zone. Variability of samples may be increased in regions where marked changes in the abundance of bottom organisms are taking place as shown in Great Slave lake data. Thus pronounced changes in abundance of organisms often associated with the sublittoral zone may contribute to variability of sampling within this area. Var-iation introduced in operation of sampling apparatus appears to be of less importance than in littoral areas. Hetero-geneity of substrate is probably less pronounced in the sub-littoral zone, however, when evident, variability from this source may be reduced by sampling stratification as previously 87 suggested. Profundal regions of oligotrophic lakes are probably-subject to less f l u c t u a t i o n i n environmental conditions than either the l i t t o r a l or s u b l i t t o r a l . This feature i s r e f l e c t e d i n the decreased v a r i a b i l i t y of samples taken from profundal regions as shown i n Great Slave Lake analyses. However, de-c l i n e i n numbers of organisms associated with greater depths necessitates a consideration of sample size i n r e l a t i o n to s p a t i a l d i s t r i b u t i o n of the fauna. Six-inch Ekman samples at the greater depths may exhibit more v a r i a b i l i t y than nine-inch samples. Although heterogeneity within bottom substrate i n profundal regions i s lower than that evident either i n l i t -t o r a l or profundal zones i t may not be eliminated from consi-deration. Eggleton (1931) shows that the nature of bottom deposits i n Douglas lake i s not uniform i n depressions within the profundal zone. D i s t i n c t i o n of types of profundal sub-strate i s made on proportions and size of organic and in o r -ganic components. U n t i l a more extensive study i s made on the corr e l a t i o n of types of profundal substrate with abundance of organisms l o c a l v a r i a b i l i t y i n sampling resultant from t h i s factor may be considered to be minor. Seasonal limnological changes i n the profundal and lower s u b l i t t o r a l zones of eutrophic lakes i n conjunction with periods of insect emergence, and reproduction i n a l l forms r e s u l t i n seasonal changes i n abundance of bottom organ-isms. Thus changes with time as w e l l as space are of major 88 importance i n the. design of sampling methods as w e l l as the analysis of subsequent data. 89 SUMMARY AND CONCLUSIONS 1. Specia l s t a t i s t i c a l methods were required i n the treatment of bottom fauna data; (a) where,means were p r o p o r t i o n a l to variances transformation of counts to logarithms was made before a p p l i c a t i o n of tes ts of s i g n i f i c a n c e between means or standard d e v i a t i o n s ; (b) where no such propor-t i o n a l i t y was evident, transformation was made to the square root of the count or to the square root of the count plus 0.5, Y (count + 0.5), i f numerous zero counts were present. 2. Analys is of Kootenay lake data indicated that v a r i a b i l i t y i n numbers and weights associated with r e l a t i v e l y few dredgings scattered throughout a large lake was so great that l i t t l e confidence could be placed i n mean values of the depth zones. 3. Further analysis of Kootenay lake data showed (a) that s i g n i f i c a n t dif ferences i n abundance of organisms occurred between port ions of the lake and (b) that dif ferences i n v a r i a b i l i t y were present between groups of the fauna. 4. Analys is of Great Slave lake data demonstrated that v a r i a -b i l i t y i n numbers and weights associated w i t h extensive sampling was not so great as to prevent i n t e r p r e t a t i o n of 90 marked changes i n mean values with depth. 5. V a r i a b i l i t y i n numbers of bottom organisms was s i g n i f i c a n t l y higher i n the l i t t o r a l and upper s u b l i t t o r a l (0 to 20 metres) zone than i n deeper zones i n Great Slave lake. 6. An increase i n v a r i a b i l i t y appeared to be associated with marked changes i n abundance of bottom organisms i n Great Slave lake. 7. Definite regional changes i n abundance and numerical v a r i a -b i l i t y were noted i n the Great Slave lake data. 8. Qualitative analyses of Great Slave data indicated consi-derable differences i n v a r i a b i l i t y between groups within the fauna; (a) oligochaetes, amphipods, and sphaeriids showed marked changes i n v a r i a b i l i t y with depth and region, while chironomids showed no s i g n i f i c a n t changes with these factors, (b) amphipods exhibited s i g n i f i c a n t l y greater num-e r i c a l v a r i a b i l i t y than other groups, followed by sphae-r i i d s , and then oligochaetes, while chironomids again showed the lea s t v a r i a b i l i t y . 9. Si g n i f i c a n t non-random v a r i a b i l i t y was demonstrated i n the operation of the nine-inch Ekman-Birge dredge on substrates within the l i t t o r a l zone. 10. A 70 per cent sodium s i l i c a t e solution was found to provide an e f f e c t i v e separation of bottom organisms from certain 91 types of substrate. 11. Threefold changes i n abundance of bottom organisms were noted between depth i n t e r v a l s of less than two feet i n Hatzic lake. 12. Evidence was given to suggest predation by f i s h as the major factor responsible for a s i g n i f i c a n t l i t t o r a l mini-mum i n abundance of bottom organisms i n Hatzic lake. 13. S i g n i f i c a n t differences i n abundance and v a r i a b i l i t y of bottom organisms were noted between four areas at the same depth within the l i t t o r a l zone of Hatzic lake. 14. Difference i n abundance of organisms at four areas i n the l i t t o r a l zone of Hatzic lake were related to both the pro-portion and the stage of decomposition of organic matter i n those regions. 15. The effects of change i n abundance of bottom organisms with time on the numerical v a r i a b i l i t y of samples were consi-dered; seasonal changes were regarded as most important when compared with diurnal and annual fl u c t u a t i o n s . 16. No s i g n i f i c a n t changes i n abundance of bottom organisms could be demonstrated between samples taken on three con-secutive years i n the main portion of Great Slave lake. 17. The d i s t r i b u t i o n of oligochaetes and chironomids as indicated 92 by three-inch samples did not depart s i g n i f i c a n t l y from-a random d i s t r i b u t i o n at depths of 15, 25, and 45 feet i n Hatzic lake. 18. Comparisons of v a r i a b i l i t y between six-and nine-inch Ekman dredgings i n Hatzic lake showed no s i g n i f i c a n t d i f -ference at depths of 15 and 25 feet; at 45 feet, however, six-inch dredgings were s i g n i f i c a n t l y more variable than nine-inch dredgings. 19. An Ekman dredge with more powerful jaws was recommended for sampling i n the l i t t o r a l zone. 20. A horizontal and v e r t i c a l s t r a t i f i c a t i o n of sampling was proposed f o r reduction of v a r i a b i l i t y associated with heterogeneity within the sampling area. 93 LITERATURE CITED Adamstone, F. B. and W. J . K. Harkness. 1923. The bottom or-ganisms of Lake Nipigon. Univ. Toronto Stud.; Pub. Ont. Fish. Res. Lab., 1^ : 1 2 3 - 1 7 0 . Aim, G. 1922. Bottenfaunan och Fiskens biologie i Yxtasjon samt jamforande studier over bottenfauna och fiskavkastning i vara sjoar. Meddel f r . Kungl Lantbruksstyrelsen, No.237. Referat von Naumann i n Int. Rev, Bd. XI. S. -352, 1923. Baker, F. C. 1918. The productivity of invertebrate f i s h food on the bottom of Oneida lake with sp e c i a l reference to mollusks. Tech. Pub. No. 4, N.Y. State College of Forestry, Syracuse Univ., Vol. XVII, No. 2. B a l l , Robert C , 1948. Relationship between available f i s h food, feeding habits of f i s h and t o t a l f i s h product-ion i n a Michigan lake. Michigan State Col-lege, Agr. Exp. Sta., Tech. B u l l . No. 206:59 pp. Deevey, E. S, Eggleton, F. E. Fisher, R. A. Juday, C. 1941. Limnological studies i n Connecticut. IV. The quantity and composition of the bottom fauna of 3 6 Connecticut lakes. Ecol. Monogr. 11:413-455. 1931. A limnological study of the profundal bottom fauna of certain fresh-water lakes. Ecol. Mon. 1:231-33,2. 1937. Productivity of the benthic zone i n Lake Michigan. Papers Mich. Acad. S c i . , Arts, and L e t t . 20:593-612. 1948. S t a t i s t i c a l methods for research workers. O l i v e r and Boyd, London. 10th E d i t i o n . 354 pp. 1922. Quantitative studies of the bottom fauna i n the deeper waters of Lake Mendota. Trans. Wise. Acad. S c i . , Arts, L e t t . , Vol. XX. Krecker, F. H. Larking P. A. and L. Y. Lancaster. 1933. Bottom shore fauna of western'Lake E r i e : a population study to a depth of s i x feet. Ecology 1/fc (2):79-93. 1948. Pontoporeia and Mysis i n Athabaska, Great Bear, and Great Slave lakes. B u l l . Fish. Res. Bd. Can., 2#:l-33. 94 L a r k i n , P. A. et a l . 1950. The production of Kamloops trout (Salmo g a i r d n e r i i kamloops, Jordan) i n Paul lake, B r i t i s h Columbia. S c i . Publ. B; C. Game Dept. No. 1. Lenz, F. 1931. Untersuchungen uber die V e r t i k a l v e r t e i l u n g der Bodenfauna im Tiefensediment von Seen. Verh. Int. theor. u. angew. Limnol., Budapest j> : 2 3 2 - 2 6 0 . Lyman, F. E. 1943. A pre-impoundment bottom fauna study of Watts Bar reservoir area (Tennessee). Trans.. Amer. Fis h . Soc. 2?.:52-62. M i l l e r , R. B. (MS) A s t a t i s t i c a l analysis of dredging data and notes on the bottom fauna of f i v e Algonquin Park lakes. Unpublished t h e s i s , University of Toronto Library. 1941. A contribution to the ecology of the Chirono-midae of Costello lake, Algonquin Park, . Ontario, Publ. Ontario Fish. Res. Lab. No.60. Univ. Toronto Stud., B i o l . Ser. No. 49:1-63. Muttkowski, R. A. 1918. The fauna of Lake Mendota. A q u a l i t a -t i v e and quantitative survey with special r e f e r -ence to the insects. Trans. Wis. Acad. S c i . , Arts, and Lett. 1^ :374-482. Neyman, Jerzy. Rawson, D. S. 1934. On the two di f f e r e n t aspects of the. re-presentative method: the method of s t r a t i f i e d sampling and the method of purposive selec t i o n . Jour. Royal Stat. Soc. \u00C2\u00A32:558-625. 1930. The bottom fauna of Lake Simcoe and i t s role i n the ecology of the lake. Univ. Toronto Stud.; Publ. Ontario F i s h . Res. Lab. 0^:1-183; 1947c. An automatic closing Ekman dredge and other equipment for use i n extremely deep water. Spec. Publ. Limn. Soc. Amer., 18:1-8. Roelofs, E. W. 1944. Water s o i l s i n r e l a t i o n to lake pro-. d u c t i v i t y . Tech. B u l l . Michigan Agric. Exp. Sta. 190:1-31. Snedecor, G. W. 1946. S t a t i s t i c a l Methods. Iowa State C o l l . Press, Ames. 485 pp. 95 Welch, P. S. 1935. Limnology. McGraw-Hill, N. Y. 471 pp. Wohlschlag, Donald E. 1950. Vegetation and invertebrate l i f e i n a marl lake. Invest. Indiana Lakes and Streams. \u00C2\u00A3 (10):373-388. "@en . "Thesis/Dissertation"@en . "10.14288/1.0106541"@en . "eng"@en . "Zoology"@en . "Vancouver : University of British Columbia Library"@en . "University of British Columbia"@en . "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en . "Graduate"@en . "An analysis of variation in quantitative sampling of bottom fauna in lakes"@en . "Text"@en . "http://hdl.handle.net/2429/40966"@en .