Mercury levels in Bonaparte's gulls - Springer Link

Arch. Environ. Contam. Toxicol. 16, 539-549 (1987)

nvironmental ntamination C oTo"

9 1987Springer-VerlagNew YorkInc.

Mercury Levels in Bonaparte's Gulls (Larus philadelphia) During Autumn Molt in the Quoddy Region, New Brunswick, Canada Birgit M. Braune and David E. Gaskin Department of Zoology, University of Guelph, Guelph, Ontario, Canada N1G 2W1

Abstract. No significant between sex differences

were detected in Hg concentrations in primary feathers, pectoral muscle, brain, liver, and kidney tissues of fall migrating juvenile and second-year Bonaparte's gulls (Larus philadelphia) collected in the Quoddy region. Adults showed sexual differences only in the first 5 primary feathers, and in muscle, kidney and brain. Differences in Hg concentrations among age groups were reflected in the primary feathers and body tissues, but as the molt progressed, Hg concentrations decreased as they converged toward a minimum asymptotic Hg level for each tissue. This suggests that the body burden of Hg was reduced through its redistribution from the body tissues into the growing feathers. Mercury concentrations in premolt head feathers (pre-egglaying) did not vary significantly between adult females and males, whereas Hg concentrations in postmolt feathers (post-egg-laying) were significantly lower in females, suggesting that egg-laying was also a route for Hg elimination. After the completion of the molt, the new feathers contained most of the body burden of Hg (93.0% in adults).

The Bonaparte's gull, Larus philadelphia Ord, is a migratory species, which uses the Quoddy region of the Bay of Fundy off southeastern New Brunswick, Canada as a staging ground during late summer-autumn (Braune and Gaskin 1982) resulting in the largest known concentration of this species in eastern Canada (Canadian Wildlife Service 1979). During their stop-over in the Quoddy region, second-year and adult birds undergo a complete postnuptial molt including the sequential molt of the primary feathers (Braune 1987). The juveniles undergo a partial molt of the head and body

feathers (Grant 1982), but retain the primary feathers acquired on the nesting grounds (Dwight 1925). Plumage and other keratinized structures concentrate Hg through firm bonding to the disulphide bonds of the keratin (Crewther et al. 1965). The Hg in feathers is not influenced by a variety of rigorous treatments (Appelquist et al. 1984), and so the feathers are a viable tissue for monitoring a bird's exposure to Hg (Goede and de Bruin 1984; Furness et al. 1986). Prior studies have noted variation in Hg concentration in different primary feathers (Gochfeld 1980; Btihler and Norheim 1981; Furness et al. 1986). Some authors have sampled feathers and/or body tissues from birds before and after molt, and commented on differences in tissue levels (Peterson and Ellarson 1976; Stickel et al. 1977; Osborn 1979; Osborn et al. 1979; Evans and Moon 1981; Parslow et al. 1982; Goede 1985). Other investigations have related Hg levels in the prey to Hg levels in the feathers (Johnels and Westermark 1969; Jensen et aI. 1972; B0hler and Norheim 1981; Lindberg and Mearns 1982; Lindberg and Odsj6 1983). This paper reports the effects of autumn molt on levels and distribution of Hg between feathers and body tissues in juvenile, second-year and adult Bonaparte's gulls in the Quoddy region, using samples collected throughout the entire period of molt.

Materials and Methods

Collection o f Birds and Sampling o f Tissues The Quoddy region, demarcated by latitudes 44~176 N and longitudes 66~176 " W, e n c o m p a s s e s the Western Isles district of Deer Island, plus approximately 40

540 smaller islands and ledges, in the Bay of Fundy off southeastern New Brunswick, Canada. Bonaparte's gulls from the Quoddy region, collected by shotgun (under Scientific Kill Permits issued by Canadian Wildlife Service), totalled 222 over 7 years, and were collectively pooled into 15 ten-day time periods spanning 22 July through 18 December 1978-84. Ten days was the approximate period of time required for completion of growth for each primary feather (Braune 1987). In gulls, the molt of the primary feathers is sequential, starting from the first (innermost) to the tenth (outermost) primary feather. The stage of molt in each specimen was noted, based on the stage of new feather growth of the right wing. The birds were aged by plumage (juvenile, second-year, adult) following Grant (1982), and the sex determined by examination of the gonads. Muscle (pectoralis + supracoracoideus), liver, kidney and brain were removed from the fresh carcass, double-bagged in polyethylene bags, and frozen to - 20~ For five postmolt adult birds in 1983 and the twelve postmolt 1984 birds, the carcass remaining after tissue sampling was completely plucked, doublebagged in p o l y e t h y l e n e bags, and frozen to - 2 0 ~ Fresh muscle and liver samples were rinsed with distilled water to remove any adhering feather barbs, and to clear liver sinuses of much of the blood. Rinsed tissues were blotted dry with paper towels. All tissues, with the exception of feathers, are referred to as body tissues. During 1982-84, paired primary feathers were plucked from both wings, individually bagged in polyethylene bags, and frozen to -20~ Remaining feathers plucked from carcasses during 1983-84 were stored in one polyethylene bag per bird during 1984, and sorted into secondaries, wing coverts, rectrices, abdominal, dorsal and down feathers before freezing during 1983. During 1982, pre- and post-molt head feathers were collected as well. Because of the black hood of adult premolt plumage and the light grey head of postmolt plumage, stage of molt was easily verified.

S a m p l e Preparation and H g Analysis Tissue samples were thawed and the outer exposed layer of tissue cut away where possible in order to minimize effects of potential contamination, and dehydration due to frozen storage. A 0.5-1.0 g (liver, kidney) or a 1.0-2.0 g (muscle, brain) sample was weighed to the nearest 0.01 g in a tared, acid-rinsed 125 ml flask. Carcasses were ground up using a hand grinder (Spong, Model No. 201), and 2.0-3.0 g subsamples prepared as described for the other body tissues. Three to five replicate samples were analyzed, and the results averaged for each carcass in order to minimize the effects of sample heterogeneity. The ten pairs of primary feathers from each bird were analyzed separately. Pannekoek et al. (1974) suggested repeated acetone-carbon tetrachloride washes to remove surface oils, dirt and blood from the feathers. Pairs (left and right) of the larger outer primary feathers (feather nos. 7 or 9) from ten different birds were used to compare the feather laundering technique of Pannekoek et al. (1974) with that of simply rinsing the feathers with water. One primary feather of each pair was laundered according to the method detailed in Pannekoek et al. (1974), and the other was thoroughly rinsed with warm tap water. Both sets of feathers were dried at 70~ for one hour. A paired t-test

B . M . Braune and D. E. Gaskin (Bailey 1959) was used to evaluate any difference in Hg concentration between feathers prepared by the two laundering techniques. There was no significant difference (N = 10, t = 0.59, P > 0.05) between Hg concentration in feathers laundered by the method of Pannekoek et al. (1974) and the tap water rinse. Therefore, the simpler tap water rinse followed by drying at 70~ for one hr was the method employed in this study. Feathers (excluding primaries) plucked from 1984 carcasses were ground twice (Thomas-Wiley Laboratory Mill, Model No. 4) through a 2 mm grate for each sample in order to gain a homogeneous mixture. Ground samples were dried at 70~ for one hr as were different feather groups from the five plucked 1983 birds, and head feathers from the 1982 birds. A 0.2 g subsample of feathers was weighed out as described for body tissue samples. Total Hg (inorganic + organic) was analyzed by cold vapor atomic absorption spectrophotometry by the method described in Gaskin et al. (1972). Mercury recovery was 98-105%. Results are expressed as txg total Hg/g wet weight for body tissues, and as ~g total Hg/g dry weight for feathers.

A c c u r a c y and Precision o f r i g Concentration Data Frozen storage of tissue samples, if accompanied by severe dehydration, might increase Hg concentration on a wet weight basis (Bodaly et al. 1984). Therefore, all tissues were analyzed for Hg within six months of removal from the carcass, and to account for the percent weight loss o f tissues, Hg concentration data were multiplied by correction factors as follows: muscle 0.973, liver 0.955, kidney 0.903, brain 0.945, remaining carcass 0.999 (Braune 1985). Precision and reproducibility were monitored throughout the study with standards and replicate samples. To assess the precision of the Hg analyses over the years, a standard reference material (U.S. National Bureau of Standards SRM 1566--oyster tissue) was analyzed for Hg on seven occasions at intervals of several months during 1981-83. The mean total Hg concentration of the analyzed oyster tissue was 0.057 --- 0.014 ~g/g compared with the value of 0.057 --- 0.015 Ixg/g given by the U.S. National Bureau of Standards.

Statistical A n a l y s e s o f F e a t h e r Groups a n d Parts A paired t-test (Bailey 1959) was used to test for differences in Hg concentration between left and right primary feathers in 37 pairs of feathers across the primary feather sequence (primary feather no. 1: N = 6, no. 4: N = 11, no. 7: N = 10, no. 8: N = 1, no. 10: N = 9). To test for differences in Hg concentration between feather parts, the vane was scraped from the shaft of six primary feathers with a scalpel, and the two components analyzed separately. The resulting sets of data were compared by a paired t-test (Bailey 1959). The vane was scraped from another 25 primary feathers and the shaft cut so as to separate the quill from the rachis. An analysis of variance (Bailey t959) was used to test for differences in Hg concentration among feather vane, quill and rachis, and also among feather groups (secondaries, wing coverts, primaries, rectrices, abdominal, dorsal and down feathers). For all statistical analyses, probability values greater than 0.05 were considered non-significant.

Mercury in Gulls Feathers other than premolt and fully-developed postmolt were not included in the statistical analyses. Since primary feathers are renewed only once per year, and Hg levels are fixed during growth (Lindberg e t al. 1983), premolt feathers of second-year birds were included with the feathers of juveniles leaving only new, fully-developed postmolt feathers for statistical analyses of feathers of second-year birds. Both premolt and postmolt primary feathers of adults were pooled for statistical analyses. Total Hg content per primary feather was calculated by multiplying feather weight by Hg concentration.

Statisticai! Analyses of rig Concentration Data There were no significant differences among years for Hg concentrations in muscle, brain or primary feathers of adult Bonaparte's gulls (Braune 1985) and, therefore, data for each tissue were pooled over the years. Although Bonaparte's gulls passed through the Quoddy region in 'waves' (Braune 1987), the birds have been treated as having a continued presence in the area during July-December, since the overall trends of Hg concentrations in the tissues masked any minor differences among waves (Brmme 1985). Normal probability plots (SAS Institute 1982) of Hg data grouped by tissue type, and bird age and sex, were evaluated and the data found to be acceptable for use in single-factor analysis of variance (ANOVA) (Sokal and Rohlf 1969). ANOVAs (SAS Institute 1982) were used to test for sexual differences in Hg concentrations in primary feathers of juvenile, second-year and adult Bonaparte's gulls. Differences in Hg concentrations in primary feathers were tested among Bonaparte's gulls of different ages (juvenile, second-year, adult) using ANOVAs followed by GT2 tests for unequal sample sizes (Sokal and Rohlf 1969; SAS Institute 1982). Based on results of tests for sexual differences, juveniles and second-year birds were not separated by sex. Three sets of analyses among age groups were carried out: juveniles vs second-year birds vs 1) adults (sexes pooled), 2) adult females, 3) adult males. ANOVAs were also used to test for differences in Hg concentration and in Hg content of each individual primary feather for each of the three age groups. ANOVAs were used to test for differences between sexes in Hg concentrations in muscle, liver, kidney and brain for each age and time period. Differences among time periods of Hg concentrations in these same four tissues of juvenile, second-year and adult gulls, with sexes pooled and sexes separate, were tested using ANOVAs followed by GT2 tests. Differences in Hg concentrations of tissues among Bonaparte's gulls of different ages were similarly analyzed. Based on results of tests for sexual differences, juveniles and second-year birds were not separated by sex. Three sets of analyses among age groups were carried out: juveniles vs second-year birds vs 1) adults (sexes pooled), 2) adult females, 3) adult males. Correlation coefficients were calculated from regression analyses (SAS Institute 1982) carried out for mean Hg concentrations of all feathers (including primary feathers) as calculated in Braune (1985), in relationship to Hg concentrations in each of the ten primary feathers. Mercury c o n c e n t r a t i o n s in head feathers of second-year birds and adult females and males were

541 Table 1. Mean Hg concentrations Qxg/g dry weight) in feather parts and feather groups of Bonaparte's gulls collected from the Quoddy region, New Brunswick, Canada

Feather parts quill rachis vane Feather groups secondaries wing coverts primaries rectrices dorsal abdominal down

N

~

SD

25 25 25

2.348 3.289 3.512

2.1104 2.8219 2.0701

3 3 3 3 3 3 3

1.879 2.313 2.537 2.783 3.631 5.237 5.725

1.0619 1.0679 1.2027 1.0442 0.9947 0.8074 2.6393

tested for differences between pre- and post-molt Hg concentrations using t-tests for small sample sizes (Bailey 1959).

Results

Feathers T h e r e was no significant difference b e t w e e n Hg c o n c e n t r a t i o n s in p a i r e d left and right p r i m a r y feathers (N = 37, t = 0.78, P > 0.05), nor were there significant differences between Hg concentrations in feather shaft and vane (N = 6, t = 0.29, P > 0.05), or among feather quill, rachis and vane (N = 75, F = 1.71, P > 0.05). The vane, however, contained a slightly higher mean Hg concentration than the rachis or the quill (Table 1). There were significant differences among feathers from different parts of the body (N = 21, F = 3.50, 0.025 > P > 0.01) with higher mean Hg concentrations in the abdominal and d o w n feathers than in the remaining contour feathers (Table 1). There was a significant difference (N = 15, P = 0.030) in Hg concentration in primary feather no. t between sexes of juveniles but, overall, there were no significant d i f f e r e n c e s b e t w e e n sexes in Hg levels o f j u v e n i l e s , and o f s e c o n d - y e a r birds. Adults, however, s h o w e d significant differences between sexes for Hg in primary feather nos. t - 5 (N = 6 2 - 6 8 , 0.01 > P > 0.001) with males having the higher Hg concentrations (Figure 1). T h e r e w e r e significant differences among age groups in Hg concentrations in primary feather nos. 1 - 5 (N = 85-89, P < 0.002) and 9 - 1 0 (N = 87-89, P ~< 0.03) for comparisons of juveniles and second-

542

B . M . Braune and D. E. Gaskin

15 1 29 14 13 12

28

30

11

"--- 10 133

9

2~

31

30

24

36 1 0 ~

1

2

3

4

Feather

5

6

7

8

9

34 10

10

No.

Fig. 1. Mean Hg concentrations (ixg/g dry weight) in individual primary feathers of adult male [], adult female ~N, second-year [] and juvenile [] Bonaparte's gulls collected from the Quoddy region, New Brunswick, Canada. Sample sizes are given above the standard error bars

year birds with adult females, and in primary feather nos. 1-6 (N = 82-84, P = 0.0001 for nos. 1-5, and GT2 test significant between juveniles and adult males only for no. 6) and 8-10 (N = 70-84, P < 0.03) for comparisons with adult males. GT2 tests showed that adult females did not have significantly different Hg levels from second-year birds in any of the cases, whereas adult males were significantly different from second-year birds only for primary feather nos. 1-4 (P < 0.05). Juveniles were significantly different from second-year birds and adult females and males for primary feather nos. 1-5 and 10 (GT2 tests). The only case for which there was no significant difference in Hg levels among age groups was primary feather no. 7. The relationships among age groups for total Hg content per feather for each of the 10 primary feathers (Figure 2) follows the same pattern as for Hg concentrations (Figure 1). Mercury concentration per primary feather decreased significantly for feather nos. 1 through 10 for second-year (N = 191, P = 0.0001) and adult (female: N = 354, male: N = 294, total: N = 648, P = 0.0001 in each case)

birds but not for juveniles. Total Hg content of primary feathers remained relatively constant over the 10 feathers for second-year and adult female birds, but significantly decreased in adult males (N = 294, P = 0.0001), and increased in juveniles (N = 318, P = 0.0001).

Body Tissues Male and female juvenile and second-year Bonaparte's gulls did not have significant differences in Hg concentrations in muscle, liver, kidney and brain analyzed by time period. Adult Bonaparte's gulls, however, showed significant differences between sexes in Hg concentrations in muscle (N = 8-23, 0.04 > P > 0.005, 5/10 time periods), kidney (N = 8, P = 0.032, 1/10 time periods), and brain (N -- 21, P = 0.04, 2/10 time periods). Tissues of adult males contained higher Hg concentrations in all significant cases except during mid-November (time period 12). Tissue Hg concentrations of adult males

Mercury in Gulls

543

.6

ii

Ii

1

2

3

4 Feather

5

6

7

8

lr!

9

t0

No.

Fig. 2. Mean total Hg content (txg Hg dry weight) in individual primary feathers of adult male N, adult female ~ , second-year [] and juvenile ~ Bonaparte's gulls collected from the Quoddy region, New Brunswick, Canada. Sample sizes are the same as those given above the standard error bars in Figure 1

were not, however, consistently higher than those of females when all time periods were considered. Therefore, most results involving tissue Hg concentrations of adults are presented both with sexes pooled and sexes separate. Mercury concentrations in muscle and kidney of juveniles varied significantly with time period (N = 46, P = 0.014, and N = 47, P = 0.0048, respectively). None of the tissues analyzed from secondyear birds varied significantly with time period. Mercury concentrations in body tissues analyzed for adults, however, did vary significantly with time period, both with sexes pooled (N = 137-145, P = 0.0001 in each case) and sexes separate (N = 54-85, P < 0.001). The general trend was for an overall decrease in tissue Hg concentration over time periods until mid-November (time period 12) in muscle, kidney and brain, and late October (time period 10) in liver, followed by an increase in tissue Hg concentration of second-year and adult birds after that time (Figure 3). No data were available

for juveniles after mid-November (time period 12). GT2 tests showed that the decrease in Hg concentrations in muscle, liver, kidney and brain of adults from time periods 1 and/or 2 to 10 and/or 12 were significant (P < 0.05), but the following increase was significant only for Hg concentration in liver (time periods 10 vs 14). There were significant differences among age groups in Hg concentrations in muscle (N = 10-39, 0.01 > P >/- 0.0004, 5/9 time periods), liver (N = 8-40, 0.03 > P >~ 0.0001, 5/9 time periods), kidney (N = 8-40, 0.04 > P t> 0.0001, 6/9 time periods), and brain (N = 6-22, 0.04 > P I> 0.0001, 5/9 time periods) with most differences occurring during late July-late September (time periods 1-6). After that time, Hg concentrations in tissues of second-year and adult birds approximately paralleled each other, while Hg concentrations in tissues of juveniles decreased to converge with those of second-year and adult birds by mid-November (time period 12) (Figure 3).

544

B.M. Braune and D. E. Gaskin

.32 A.

B

.2~

Brain

36

924

32

,2C

28

r~

~

7

40

Muscle

.16

14

~

13

.12

16

.24

.2o 24

'..,.,

.08

I

...............i .... "i-"

.04

"4.,.

,o

.... ~ ,

21

16

y

~2

/.-

--

,2

I0

14

..... 'I

;7io

I

...............

1

2

3

July

4

5

6

Aug

7

8

Sept Time

9

10

11

Oct

12

13

14

Nov

15 Dec

Period

8

1.4

C

1

Liver

2

July

3

4

5

Aug

6

7

8

Sepl Time

1.2

9

I0

11

Oct

12

13

14

Nov

15 Dec

Period

8

~ ! 1.0

D.

Kidney

1.0

i 6

9~

..

..

z/

~,

.6

/

2 .4

1 July

2

3 Aug

4

5

6

7

Sep[ Time

8

9 Oct

10

11

12

~

.4

13

Nov

14

15 Dec

Period

1 July

2

3 Aug

4

5

6

7

8

Sepl Time

9 Oct

10

11

12 Nov

13

t4

15 Dec

Period

Fig. 3. Mean Hg concentrations (~g/g wet weight) in: A) muscle, B) brain, C) liver, and D) kidney of juvenile (e .......... e), second-year (A ..... A) and adult (11 I ) Bonaparte's gulls collected from the Quoddy region, New Brunswick, Canada. Sample sizes are given above the standard error bars

Distribution and Interrelationships of Tissue Hg Concentrations The weight of all feathers including primary feathers of adults accounted for only 12.1 _+ 0.85% of the total body weight (minus stomach contents) but 93.0% of the body burden of Hg (Table 2). The distribution of the body burden of Hg (excluding feathers) among the various organs of postmolt adult birds showed that although the liver accounted for only 5.4 -+ 1.67% of the body weight (minus stomach contents and feathers), it contained 36.4% of the Hg in the body (Table 2). Regression analyses relating the mean Hg con-

centration of all feathers including the primary feathers with the Hg concentrations of each of the 10 primary feathers showed that the highest correlations occurred with primary feather nos. 5 - 7 , even though relationships with feather nos. 4 - 1 0 all had significantly positive correlation coefficients (Table 3). Mean Hg concentration for all feathers including the primary feathers was lowest for juveniles and highest for adults (Table 4). There was no significant difference (N = 8, t = 0.86) between mean Hg concentrations of all feathers of adult females and males. Mercury concentrations in head feathers of second-year birds and of adult males did not vary significantly between pre- and post-molt states,

Mercury in Gulls

545

Table 2. Percent distribution of b o d y burden of total Hg in postmolt juveniles, second-year and adult Bonaparte's gulls collected during

October-November (time periods 8 and 12) from the Quoddy region, New Brunswick, Canada % Body

burden b

% Body Hg ~

Age Juvenile N = 2

Second-year N = 2 Adult N = 12c

g SD ~ SD g SD

LIV

KID

MUS

BRN

CARC

BOD

FEATH

38.3 9.90 32.3 4.67 36.4 6.57

6.2 0.14 3.6 1.34 5.5 1.53

6.1 2.26 5.5 0.57 8.8 3.65

1.4 0.21 0.7 0.07 0.9 0.25

48,1 8.02 58,0 6,59 48.4 8,19

12,1 3.32 9.1 1.98 7.0 2.24

88.0 3.32 90.9 1.98 93.0 2.24

% distribution of Hg content in body tissues excluding feathers. L I V - l i v e r , K I D - k i d n e y , M U S - p e c t o r a l m u s c l e , B R N - b r a i n , CARC-remaining carcass after sample tissues were removed. Total percentages may not equal 100% due to rounding errors b % distribution of body burden of total Hg including all feathers. BOD-body, FEATH-feathers c N = 8 for calculations involving feathers

Table 3. Correlation coefficients for the relationships between H g concentrations of individual primary feathers ( F 1 - F 1 0 ) and mean Hg concentrations of all feathers including primary feathers (AF) of adult Bonaparte's gulls collected from the Quoddy region, New Brunswick, Canada

Table 4. M e a n H g c o n c e n t r a t i o n s (~g/g dry weight) for all

Relationship

N

ra

Age

N

2

SD

AF AF AF AF AF AF AF AF AF AF

8 8 8 8 8 8 8 8 8 8

0.32 0.27 0.49 0.83* 0.94** 0.95** 0.95** 0.91" 0.88* 0.91"

Juvenile Second-year

2 2 8 4 4

1,976 2.493 4.137 4.824 3.450

0.0658 0,6562 2.2235 2.9923 1.1511

vs vs vs vs vs vs vs vs vs vs

FI F2 F3 F4 F5 F6 F7 F8 F9 F10

a Significant correlation coefficients indicated as follows: *0.01 > P > 0.001, **P < 0.001

whereas the mean Hg concentration in postmolt head featJhers of adult females was significantly lower than in premolt feathers (Table 5). Mercury concentrations in premolt head feathers of adult females and males did not vary significantly (N = 9, t = 0.05), whereas the mean Hg concentration in postmolt feathers of adult females was significantly lower than in males (N = 32, t = 3.53, 0.002 > P >

0.001).

Discussion

Comparison of Feather Groups and Parts As emerging feathers are keratinized, they lose all vascular and nervous connections and become physiologically isolated from the rest of the bird

feathers including primary feathers of juvenile, second-year and adult B o n a p a r t e ' s gulls collected during m i d - N o v e m b e r (time period 12) 1984 from the Quoddy region, N e w Brunswick, Canada

Adult Adult female Adult male

(Voitkevich 1966). Mercury bonded into the feather keratin therefore reflects the amount of Hg in the blood when the individual feather was formed (Westermark et al. 1975; Scanlon et al. t980). Since right and left primary feathers of the same number are formed approximately simultaneously, it was to be expected that they show similar Hg concentrations. As demonstrated in the present study, right or left primary feathers of the same number can be substituted for analysis of Hg concentration. Although there were no significant differences among Hg concentrations in feather quill, rachis and vane, the Hg concentrations in the vane, and also the rachis, were slightly higher than in the quill (Table 1), which agrees with the findings of Berg et al. (1966), Bfickstr6m (1969), and Doi and Fukuyama (1983). Since the vane of a primary feather is more highly mineralized than the shaft (Kelsall 1970), there is, perhaps, greater opportunity for the substitution of Hg for other elements, such as Zn, which are essential for feather synthesis (Evans and Moon 1981). The abdominal and down feathers had significantly higher Hg concentrations than the dorsal, tail and wing feathers (Table 1), in agreement with the findings of Doi and Fukuyama (1983) for a va-

546

B . M . Braune and D. E. Gaskin

Table 5. Mercury concentrations (ixg/g dry weight) in pre- and post-molt head feathers of Bonaparte's gulls collected during late J u l y early December 1982 from the Quoddy region, New Brunswick, Canada. T- or d-values are given for comparison between Hg concentrations in pre- and post-molt head feathers Premolt

Postmolt t- or

Age

N

2

SD

N

2

SD

d-valuesa

Juvenile Second-year Adult female Adult male

-6 6 3

-3.930 5.960 6.051

-2.1130 3.2268 1.4424

10 13 14 18

3.406 3.099 2.493 5.100

1.3036 2.6278 1.1354 2.8571

-0.68 2.56" 0.56

a Significance levels indicated as follows: * 0.05 > P > 0.02

riety of seabirds including the Asiatic common gull (L. canus kamschatschensis), and with the findings of Frank et al. (1983) for the common loon (Gavia immer). Furness et al. (1986) found a different ranking of feather groups in a great skua (Catharacta skua) and a black-legged kittiwake (Rissa tridactyla), probably because the birds were collected after the partial spring molt of head and neck feathers. The molt of the primaries provides a rough gauge by which the progress of autumn molt may be measured, with the molt of the rest of the plumage taking place mainly within the period when the primaries are being renewed (Grant 1982). Since Hg concentrations progressively decreased from the first to the last primary feathers developed (Figure 1), as the Bonaparte's gull gradually redistributed its body burden of Hg into the feathers (Figure 3), the higher Hg concentrations in the abdominal and down feathers suggests that the new growth of these feathers was initiated before that of the dorsal feathers. The high correlations between mean Hg concentration in all feathers and Hg concentrations in primary feather nos. 5 - 7 (Table 3) indicate that the peak period of feather renewal coincided with the growth of these primary feathers which occurred about late August-mid-September (Braune 1987). This also agrees well with the progression of the body feather molt as monitored by the transition from the black hood of adult summer plumage to the white head feathers of the winter plumage (Braune 1987).

Hg Accumulation in Feathers Second-year and adult Bonaparte's gulls showed a significant decrease in Hg concentration from primary feather nos. 1 through 10 (Figure 1). Bt~hler and Norheim (1981) found the same pattern in the sparrowhawk (Accipiter nisus) as did Gochfeld

(1980) in the Peruvian booby (Sula variegata), and Furness et al. (1986) in a variety of seabird species. The first developed feathers received the highest Hg concentrations and as the Hg burden in the body tissues gradually decreased (Figure 3), the Hg concentrations in the developing feathers also decreased. All feathers of juveniles are formed at the same time on the nesting ground (Dwight 1925) and, therefore, show similar Hg concentrations (Figure 1). Total Hg content remained relatively constant for all of the primary feathers (Figure 2) because, although Hg concentrations progressively decreased from primary feather nos. 1 through I0, the feather size/weight progressively increased (Braune 1985). There was a significant increase in Hg content from primary feather nos. 1 through 10 in juveniles because feather size increased across the sequence but Hg concentration did not vary significantly among feathers. Total Hg content significantly decreased through the primary feather sequence of adult males, emphasizing the importance of this Hg elimination route in males. Adult males had significantly higher Hg concentrations in primary feather nos. 1 through 5 than adult females, second-year birds and juveniles in order of decreasing Hg concentrations (Figure 1), and Hg levels in adult females as well as in secondyear birds were significantly higher than in juveniles. Second-year birds were not significantly lower in feather Hg concentration than adult females in any of the primary feathers, and they significantly differed from adult males only in primary feather nos. 1-4. Adult females may have reduced their body burden of Hg through egg-laying to the level found in second-year birds. Adult males approached similar Hg levels as those found in the females and second-year birds only after having eliminated large amounts of Hg into the first few developing feathers. Therefore, if Hg elimination via the eggs were omitted, Hg accumulation increased with

Mercury in Gulls

age from juveniles to second-year birds to adults as reflected by the Hg concentrations in the first few primary feathers developed. The reduction of the body burden of Hg due to elimination via the eggs was also reflected in the pattern of Hg concentrations in pre- and post-molt head feathers. Head feathers from the prenuptial molt showed no significant difference in Hg conc e n t r a t i o n b e t w e e n adult females and males, whereas head feathers from the postnuptial molt indicated a lower Hg level in females than in males (Table 5), again indicating reduction of body burden of Hg in females via egg-laying.

Hg Accumulation in the Body Tissues If a bird has reached equilibrium between intake and elimination of Hg during the non-molting season, it is probable that Hg levels in body tissues will decrease during molting (Lindberg and Odsj6 1983). All tissues of adult Bonaparte's gulls showed a progressive decrease in Hg concentration during the period of molt (Figure 3), even though Hg content of the diet remained relatively constant during that time (Braune and Gaskin, in press). Seasonal fluctuations in tissue Hg concentration due to molt has been reported for other species. In mallard ducks (Anas platyrhynchos), tissue Hg concentration decreased with the onset of feather growth, and lowest levels in tissues were correlated with highest concentrations in feathers (Stickel et al. 1977). Tissue Hg concentrations of waders (Charadriiformes: Limicolae) were lowest just after molt (Goede 1985) and increased through the winter (Parslow 1973). Starlings (Sturnus vulgaris) showed the lowest liver Hg concentrations during the molt period (Osborn 1979). Peterson and Ellarson (1976) reported that during the fall, oldsquaws (Clangula hyemalis) reduced their relatively high Arctic Hg levels to lower Lake Michigan levels through normal fe,ather molt. The liver, which plays a major role in the metabolism and elimination of methylmercury ,(Bhatnagar et al. 1982), was the tissue that showed the first signs of Hg accumulation after molt was completed in Bonaparte's gulls (Figure 3C). Second-year birds did not show any significant differences in tissue Hg concentrations over time (Figure 3). By time period 1, the first sampling period, these birds had already started their molt which was advanced, relative to the adults, by 2-3 primary feathers (Braune 1987). Therefore, tissue

547

Hg concentrations did not reflect premolt levels which may have been higher. Juvenile Bonaparte's gulls contained higher tissue Hg concentrations than second-year and adult birds before converging to a minimum asymptotic Hg level by mid-November (time period 12) (Figure 3). March et al. (1983) reported that in juvenile chickens, relatively more Hg is bound into muscle proteins, with a slower turnover than in adults because juvenile muscle is synthesizing new protein as well as replacing degraded protein. Initially, then, tissues of juveniles would contain higher Hg levels than adults. Since the juveniles were undergoing only a partial head and body molt (Grant 1982), the decrease in tissue Hg concentrations was more gradual (Figure 3). Some studies have found no correlation of tissue Hg concentration with age (Peterson and Ellarson 1976; Holt et al. 1979; Evans and Moon 1981; Hutton 1981; Nicholson 1981), but as noted by Peterson and Ellarson (1976), collections were made on the wintering grounds, presumably after autumn molt, when the birds of different ages had possibly converged to minimum asymptotic Hg levels.

Distribution and Implications of Body Burden of Hg In adult Bonaparte's gulls, the new feathers contained 93.0% of the body burden of Hg just after molt (Table 2). The binding of a higher proportion of the body burden of Hg to the plumage than the body tissues has also been reported for ospreys (Pandion haliaetus) (H~ikkinen and H~isfinen 1980) and puffins (FratercuIa aretica) (Parslow et ai. 1972). In the body tissues of chickens, skeletal muscle comprises the major storage site for Hg because, although the Hg concentration is not as high as in the liver and kidney, the greater muscle mass accommodates a greater total accumulation (March et al. 1983). The remaining carcass after sampling probably consists, to a large extent, of muscle. Therefore, over 50% of the total Hg in the body (minus feathers) was stored in the muscle mass (muscle + remaining carcass) of Bonaparte's gulls (Table 2). For its size, however, the liver contained a disproportionately high amount of Hg (Table 2). The strong parallel between decrease in tissue and feather Hg levels over time as reflected by Hg concentrations in the primary feather sequence establishes the functional importance of molt as a mechanism for the reduction of the body burden of Hg through its redistribution into the feathers. Fur-

548

ness et al. (1986) also support the view that as the amount of Hg stored in the body tissues is reduced during molt, so is the amount of Hg available to enter the growing feathers. The data presented in this study suggest that the widely held idea that Hg levels in feathers reflect dietary intake at the time of feather growth be re-evaluated in view of the strong relationship between body tissue and feather Hg concentrations. Acknowledgments. The authors wish to t h a n k Drs. J. B. Sprague and V. G. Thomas of the University of Guelph, Dr. R. Frank of The Ontario Ministry of Agriculture and Food (OMAF) Pesticide Residues Laboratory, Guelph, Ontario, and Dr. M. Gochfeld of Rutgers University, Piscataway, New Jersey, for their advice and constructive criticism on an earlier draft of the manuscript. The cooperation of the OMAF laboratory for the use of their facilities for the mercury analyses is gratefully acknowledged, and in particular, Mr. P. Suda and Mr. K. Stonefield, for their advice regarding technical procedures. We also thank the members of the University of Guelph Cetacean and Seabird Research Group, especially Ms. E M. Mercier and Mr. M. A. Showell, for their help in compilation of data. The work was supported by Natural Sciences and Engineering Research Council of Canada strategic grant no. G0304 (1980-83) awarded to Drs. D. E. Gaskin and R. Frank, as well as grants to Dr. D. E. Gaskin from the Natural Sciences and Engineering Research Council of Canada (operating grant no. A5863), the Canadian National Sportmen's Fund and the Department of Fisheries and Oceans.

References Appelquist H, Asbirk S, Drabaek I (1984) Mercury monitoring: Mercury stability in bird feathers. Mar Pollut Bull 15:22-24 B~.ckstr6m J (1969) Distribution studies of mercuric pesticides in quail and some flesh-water fishes. Acta Pharmacol Toxicol 27(suppl 3): 1-103 Bailey NTJ (1959) Statistical Methods in Biology. Hodder and Stoughton, London, England, 216 pp Berg W, Johnels A, Sj6strand B, Westermark T (1966) Mercury content in feathers of Swedish birds from the past 100 years. Oikos 17:71-83 Bhatnagar MK, Vrablic OE, Yamashiro S (1982) Ultrastructural alterations of the liver of Pekin ducks fed methyl mercurycontaining diets. J Toxicol Environ Health 10:981-1003 Bodaly RA, Hecky RE, Fudge RJP (1984) Increases in fish mercury levels in lakes flooded by the Churchill River diversion, northern Manitoba. Can J Fish Aquat Sci 41:682-691 Braune BM (1985) Total mercury accumulation during autumn moult in Bonaparte's Gulls of the Quoddy region, New Brunswick, Canada. PhD thesis, University of Guelph, Guelph, Ontario, 216 pp (1987) Body morphometrics and molt of Bonaparte's Gulls in the Quoddy region, New Brunswick, Canada. Condor 89:150-157 Braune BM, Gaskin DE (1982) Feeding methods and diving rates of migrating larids off Deer Island, New Brunswick. Can J Zool 60:2190-2197

-

-

B . M . Braune and D. E. Gaskin A mercury budget for the Bonaparte's Gull during autumn moult. Ornis Scand (in press) B t~hler U, Norheim G (1981) The mercury content in feathers of the Sparrowhawk Accipiter nisus in Norway. Fauna norv Ser C, Cinclus 5:43-46 Canadian Wildlife Service, Atlantic Region (1979) Summary of unpublished surveys of waterfowl and seabirds in Passamaquoddy Region, 1973-79. In: Scarratt DJ (ed) Evaluation of recent data relative to potential oil spills in the Passamaquoddy area. Fish Mar Serv Tech Rep 901:61-67 Crewther WG, Fraser RDB, Lennox FG, Lindley H (1965) The chemistry of keratins. In: Anfinsen CB, Anson ML, Edsall JT, Richards FM (eds) Advances in Protein Chemistry, Vol 20, Academic Press, New York, pp 191-303 Doi R, Fukuyama Y (1983) Metal content in feathers of wild and zoo-kept birds from Hokkaido, 1976-78. Bull E n v i r o n Contam Toxicol 31:1-8 Dwight J (1925) The gulls (Laridae) of the world; their plumages, moults, variations, relationships and distribution. Amer Mus Nat Hist Bull 52:63-402 Evans PR, Moon SJ (1981) Heavy metals in shorebirds and their prey in north-east England. In: Say PJ, Whitton BA (eds) Heavy Metals in Northern England: Environmental and Biological Aspects, University of Durham, pp 181 - 190 Frank R, Lumsden H, Barr JF, Braun HE (1983) Residues of organochlorine insecticides, industrial chemicals, and mercury in eggs and in tissues taken from healthy and emaciated common loons, Ontario, Canada, 1968-1980. Arch Environ Contam Toxicol 12:641-654 Furness RW, Muirhead SJ, Woodburn M (1986) Using bird feathers to measure mercury in the environment: Relationships between mercury content and moult. Mar Pollut Bull 17:27-30 Gaskin DE, Ishida K, Frank R (1972) Mercury in harbour porpoises (Phocoena phocoena) from the Bay of Fundy region. J Fish Res Board Can 29:1644-1646 Gochfeld M (1980) Mercury levels in some seabirds of the Humboldt Current, Peru. Environ Pollut 22:197-205 Goede AA (1985) Mercury, selenium, arsenic and zinc in waders from the Dutch Wadden Sea. Environ Pollut A, 37:287-309 Goede AA, de Bruin M (1984) The use of bird feather parts as a monitor for metal pollution. Environ Pollut B, 8:28t-298 Grant PJ (1982) Gulls: A Guide to Identification. T&AD Poyser, Calton, England, 280 pp Hgtkkinen I, H~ts~inen E (1980) Mercury in eggs and nestlings of the osprey (Pandion haliaetus) in Finland and its bioaccumulation from fish. Ann Zool Fennici 17:131 - 139 Holt G, FrCslie A, Norheim G (1979) Mercury, DDE, and PCB in the avian fauna in Norway 1965-1976. Acta Vet Scand Suppl 70:1-28 Hutton M (1981) Accumulation of heavy metals and selenium in three seabird species from the United Kingdom. Environ Pollut A, 26:129-145 Jensen S, Johnels AG, Olsson M, Westermark T (1972) The avifauna of Sweden as indicators of environmental contamination with mercury and chlorinated hydrocarbons. In: Voous KH (ed) Proc X V t h I n t e r n a t Ornithol Congr, EJ Brill, Leiden, Netherlands, pp 455-465 Johnels AG, Westermark T (1969) Mercury contamination of the environment in Sweden. In: Miller MW, Berg GG (eds) Chemical Fallout: Current Research on Persistent Pesticides, Charles C Thomas, Springfield, Illinois, pp 221-241 Kelsall JP (1970) Comparative analysis of feather parts from wild mallards. Can Wildl Serv Progr Notes No 18 -

-

Mercury in Gulls Lindberg P, Mearns R (1982) Occurrence of mercury in feathers from Scottish Peregrines (Falco peregrinus). Bull Environ Contain Toxicol 28:181 - 185 Lindberg P, Odsj6 T (1983) Mercury levels in feathers of peregrine falcon Falco peregrinus compared with total mercury content in some of its prey species in Sweden. Environ Pollut B, 5:297-318 Lindberg P, Odsj6 T, Wikman M (1983) Mercury in feathers of the peregrine falcon Falco peregrinus in Finland. Ornis Fenn 60:28-30 March BE, Poon R, Chu S (1983) The dynamics of ingested methyl mercury in growing and laying chickens. Poult Sci 62:1000-1009 Nicholson JK (1981) The comparative distribution of zinc, cadmium and mercury in selected tissues of the herring gull (Larus argentatus). Comp Biochem Physiol 68C:91-94 Osborn D (1979) Seasonal changes in the fat, protein and metal content of the liver of the starling Sturnus vulgaris. Environ Pollut 19:145-155 Osborn D, Harris ME Nicholson JK (1979) Comparative tissue distribution of mercury, cadmium and zinc in three species of pelagic seabirds. Comp Biochem Physiol 64C:61-67 Pannekoek WJ, Kelsall JR Burton R (1974) Methods of analyzing feathers for elemental content. Fish Mar Serv Tech Rpt 498 Parslow JLF (1973) Mercury in waders from the wash. Environ Pollut 5:295-304

549 Parslow JLF, Jeffries DJ, French MC (1972) Ingested pollutants in puffins and their eggs. Bird Study 19:18-33 Parslow JLF, Thomas GJ, Williams TD (1982) Heavy metals in the livers of waterfowl from the Ouse Washes, England. Environ Pollut 29:317-327 Peterson SR, Ellarson RS (1976) Total mercury residues in livers and eggs of oldsquaws. J Wildl Manage 40:704-709 SAS Institute (1982) SAS User's Guide: Statistics, 1982 Edition. SAS Institute, Cary, North Carolina, 494 pp Scanlon PF, Oderwald RG, Dietrick TJ, Coggin JL (1980) Heavy metal concentrations in feathers of ruffed grouse shot by Virginia hunters. Bull Environ Contam Toxicol 25:947-949 Sokal RR, Rohlf FJ (1969) Biometry. WH Freeman, New York, 859 pp Stickel LF, Stickel WH, McLane MAR, Bruns M (1977) Prolonged retention of methyl mercury by mallard drakes. Bull Environ Contain Toxicol 18:393-400 Voitkevich AA (1966) The feathers and plumage of birds. October House Inc, New York, 335 pp Westermark T, Odsj6 T, Johnels AG (1975) Mercury content of bird feathers before and after Swedish ban on alkyl mercury in agriculture. Ambio 4:87-92

Manuscript received November 24, 1986 and in revised form February 9, 1987.