Aquatic Toxicology, 14 (1989) 331-344
Behavior of tadpoles of the bullfrog, Rana catesbeiana, in response to sublethal lead exposure Craig W. Steele, Shari Strickler-Shaw and Douglas H. Taylor Department of Zoology, Miami University, Oxford, OH, U.S.A. (Received 31 August 1988; revision received 8 December 1988; accepted 30 December 1988)
Bullfrog, Rana catesbeiana, tadpoles were exposed to 0 (control), 500, 625,750 or 1000 t~g Pb. 1- 1 (as lead nitrate) for six days (144 h). Preference/avoidance responses to plumes of Pb-contaminated water and spontaneous locomotor activity were assessed for non-exposed and Pb-exposed tadpoles. No significant differences were seen in preference/avoidance responses to Pb by either non-exposed or Pb-exposed animals, nor were there significant effects of Pb on spontaneous locomotor activity. There was, however, significantly greater variability in activity of bullfrog tadpoles ( P < 0.025; folded F-test) exposed to 500, 625, 750 or 1000 t~g P b . l 1, as compared to control animals. Key words: Lead; Amphibian; Rana catesbeiana; Behavior; Heavy metal; Activity; Tadpole
The behavioral toxicity of lead (Pb) has been studied extensively, with much of the research concerned with effects of Pb on discrimination learning of mammalian vertebrates (Cory-Schlecta and Weiss, 1985). However, Pb also produces behavioral alterations in non-mammalian vertebrates. For example, Weir and Hine (1970) showed that sublethal exposure to lead impairs learning behavior in goldfish, Carassius auratus. In addition, exposure of green frog, Rana clamitans, tadpoles to 750 or 1000/~g Pb'l-1 produces deficiencies in both acquisition and retention of learned responses by the animals (Eby, 1986; Strickler-Shaw and Taylor, 1989), and increases the variability of their locomotor activity (Taylor et al., 1989). Exposure of bullfrog, Rana catesbeiana, tadpoles to 625,750 or 1000/zg Pb" 1- 1 also produces deficiencies in both learning acquisition and retention, while exposure of bullfrog tadpoles to 500 ttg Pb'l-1 interferes with learning acquisition (Strickler-Shaw, 1988).
Correspondence to." D.H. Taylor, Department of Zoology, Miami University, Oxford, OH 45056, U.S.A. 0166-445X/89/$03.50 © 1989 Elsevier Science Publishers B.V. (Biomedical Division)
Lead is a c o m m o n contaminant of aquatic systems as it is introduced into receiving waters by a variety of sources, including industrial effluents, automobile exhaust, and in r u n o f f f r o m agricultural and sludge fields (Birdsall et al., 1986). Ranids are a significant c o m p o n e n t of m a n y ecological communities, serving as important prey species for a variety of animals. It is important, therefore, to determine if ranid tadpoles can detect and avoid Pb-polluted waters, especially since sublethal exposure to Pb has been shown to have deleterious effects on behavior of both green frog tadpoles and bullfrog tadpoles, as discussed above. Bullfrog tadpoles were selected for study not only because of their importance in m a n y ecosystems and because sublethal exposure to Pb inhibits their learning and m e m o r y , but also because they are readily available, and easy to maintain and to work with in the laboratory. An objective of the current study was to determine if bullfrog tadpoles could detect and avoid concentrations of Pb known to affect their behavior. Additionally, we wished to examine the appropriateness (for tadpoles) of the E P A - r e c o m m e n d e d m a x i m u m 'safe' concentration of Pb in the water column. Thus, this study was restricted to assessing the behavioral effects of sublethal exposure of the tadpoles to Pb from the water only, rather than from sediments or other routes of exposure. Assessments of the preference/avoidance responses of bullfrog tadpoles found no significant preference for or avoidance of 500, 625,750 or 1000/zg P b . 1- ~ by either unexposed (control) or Pb-exposed animals. There were also no significant effects of exposure to these concentrations of Pb on spontaneous locomotor activity. However, exposure to any of the concentrations of Pb used in this study produced significant increases in the variability in locomotor activity in Pb-exposed groups of tadpoles, as compared to control animals. MATERIALS AND M E T H O D S
Experimental animals and handling T w o hundred bullfrog tadpoles (41-80 m m , total length; x = 64.3 mm; SD = 12.57) were captured f r o m a pond in the vicinity of Oxford, O H , as needed from May to August, 1988. The tadpoles were held in c o m m u n i t y aquaria (23 ___1.0°C) under a 12:12 LD photoperiod (lights on, 0800) for 1-3 wk prior to experimentation and were fed freshly-boiled spinach and lettuce ad libitum. No tadpoles had visible limb buds and all were at a comparable premetamorphic stage of development (stage 24 of Roberts, 1962). The tadpoles were divided randomly into 25 experimental groups ( N = 8, each). They were exposed individually for 144 h in 1500-ml Pyrex crystallizing dishes containing 1000 ml of filtered, dechlorinated laboratory water at one of five Pb concentrations: 0 /~g'1-1 (control group, C), 500 /zg'1-1 (low-exposure group, L), 625 #g. 1- ~ (intermediate-exposure group, I), 730/xg- 1 - ~ (medium-exposure group, M), or 1000/xg. 1- ~ (high-exposure group, H). These concentrations of Pb and this ex-
333 posure regimen are identical to those used by Strickler-Shaw (1988) in her study of the effects of Pb on learning and memory of bullfrog tadpoles; they were used here for comparative purposes. The Pb concentrations were chosen a priori with regard to the EPA-recommended maximum 'safe' concentration of 766/~g-l-1 (USEPA, 1980) for water with hardness similar to our laboratory water (see 'Water Quality Analysis'). Thus, 750 #g-1-1 approximates the maximum 'safe' level; 1000 #g-1-1 is approximately 25070 greater; 500 #g. l - 1, 2507o less; and 625 #g. l - 1 is intermediate between the maximum 'safe' concentration and the least experimental concentration. Tadpoles were exposed for 144 h because of a time dependence of 3-6 days for bullfrog and green frog tadpoles exposed to these Pb concentrations to exhibit evidence of behavioral toxicosis in learning and memory testing (Eby, 1986; Strickler-Shaw, 1988; StricklerShaw and Taylor, 1989). Stock solutions (1000 mg Pb. l-1) were prepared daily using 0.1598 g Pb nitrate (Pb[NO3]2) and 100 ml of dechlorinated tap water. Exposure concentrations were made by micropipetting Pb stock solution into 1000 ml of filtered, dechlorinated laboratory water in a volumetric flask and transferred to the crystallizing dishes. Water in the dishes was changed and Pb replenished daily. Tadpoles were fed freshly-boiled spinach or lettuce ad libitum while in the crystallizing dishes, but not fed while in the fluviarium. No deaths occurred in the treatment groups during Pb exposure, justifying the assumption that exposure for 144 h even to 1000/zg Pb" l - 1 is a sublethal exposure. No lethality or LCso data was generated since we were specifically assessing the appropriateness of the recommended maximum 'safe' concentration of Pb in our water and since previous studies had shown that 500 #g Pb-1-1 inhibits learning and m e m o r y of bullfrog and green frog tadpoles, as discussed earlier. Because alterations in such adaptive behavioral attributes as learning and memory occur following sublethal exposure of tadpoles to only 75°7o of the maximum 'safe' concentration, the LCso concentration becomes trivial. We could not justify the destruction of the number of animals required to generate LCso toxicity data, which is of limited ecological value, at best.
Behavioral monitoring system and bioassay The behavioral monitoring system used for data collection consisted of an octagonal fluviarium (the design of which permitted unrestricted choice by experimental animals of up to eight discrete chemical conditions), a water delivery system, and a video-based data acquisition system. With associated statistics, the system facilitated the monitoring and quantitative assessment of preference/avoidance behavior and locomotor activity of aquatic animals in response to chemical stimuli. Immediately following Pb exposure, tadpoles were placed individually into the fluviarium (Fig. 1). The fluviarium (the behavioral arena) was an octagonal Plexiglas tank incompletely partitioned into eight equal radial octants by central and
Fig. I. The octagonalfluviarium: (A) central and peripheral Plexiglas walls; (B) 4.5 cm wide central choice channel; (C) mesh screens; (D) central mixingchambers; (E) central water inlet pipe; (F) water outlet holes in peripheral walls; (G) polyethylenetubing for chemical infusion; (H,1) aeration tubing.
peripheral walls, leaving a central, octagonal choice channel 4.5 cm wide in which animals could move without restriction through all eight octants (Scarfe et al., 1985; Steele, 1986; Taylor et at., 1989). The inner and outer circumferences of the choice channel were limited by stainless steel mesh. Additional stainless steel mesh in the central and peripheral portions of the fluviarium further compartmentalized each octant. The center-most compartment of each octant served as a mixing chamber where infused chemicals were mixed by aeration with inflowing laboratory water; this water then flowed centrifugally to establish a chemically-treated plume of water in the associated radial octant. The additional mesh of the mixing chambers also aided in lamination of the water flow through the radial octants. The integrity of chemical plumes established in this manner has been verified with dye studies, and the results indicate little or no lateral mixing of water between adjacent octants (Scarfe et al., 1985; Steele, 1986). After a 10-min adjustment period in the arena, the behavior of a tadpole was video-taped for an additional 10 min (all monitoring was conducted between approximately 0800 to 1200 h daily). This 10-min period was divided into two 5-min experimental periods. The inflow during the first 5 min consisted of filtered, dechlorinated laboratory water only (inflow rate, 1 1/min). During the second 5 min, Pb was infused using a peristaltic infusion pump (Scientific Industries, Inc., Model 403) into one of the eight central mixing chambers from a 1000 mg-1- 1 stock solu-
tion using polyethylene tubing (0.76 m m ID) at an infusion rate appropriate to produce one of the five Pb-exposure concentrations. Lead stock solution was prepared daily, immediately prior to experimentation. Eight replicate experiments were conducted for each combination of concentrations o f pre-exposure to P b and Pb infusion into the fluviarium. The octants selected for Pb infusion were selected pseudo-randomly, such that each octant was used once in each experimental series. Experimental series are identified by a two-letter designation which indicates the Pb-exposure concentration (C, L, I, M, H) to which the animals were exposed in the crystallizing dishes and the concentration o f Pb presented to tadpoles while in the fluviarium (C, L, I, M, H). For example, tadpoles in series CC were pre-exposed to 0/~g Pb" 1- 1 in the dishes and presented with 0/~g Pb- 1- 1 in the fluviarium, tadpoles in series IL were pre-exposed to 625/zg P b . I - 1 in the dishes and presented with 500 #g Pb" 1-1 in the fluviarium, and so forth. Data analyses Following an experiment, the video tape was viewed and sequential counts were made o f the location (octant) and the number o f octant crossings (locomotor activity) o f tadpoles in the fluviarium for each experiment. The location of a tadpole in the fluviarium was determined for each second of a 10-min experiment and summed for each min of an experiment. The number of octant crossings per min were counted for each animal, then summed for each tadpole for each 5-min period. For each experimental series, a non-parametric rank test, the Friedman's Xzsquare test (Schefler, 1979), was used to compare the distribution of the counts of tadpoles in the eight octants o f the fluviarium for each minute of an experiment with a r a n d o m distribution. Significant values of the test statistic occur only for 1-min periods when an octant receives the same relative ranking a m o n g replicates within an experimental series. For a r a n d o m distribution, the expected percentage of counts in each radial octant during an experiment is 12.5%. Counts of tadpoles in each radial area greater than or less than expected can be correlated with the rankings. This correlation is necessary because, although values of the test statistic are based on the rankings, the statistic indicates only whether a distribution of counts is rand o m or is non-random; if the distribution is non-random, the statistic does not indicate which octant(s) are contributing to the non-randomness or whether a nonr a n d o m distribution is the result of greater or lesser counts in octants receiving Pb infusion. This analysis can distinguish between attraction, avoidance, or no apparent response, i.e. randomness in the distribution of counts of tadpoles in the fluviarium for a specific 1-min period (Scarfe et al., 1985; Steele, 1986; Taylor et al., in press). One-way analysis of variance (ANOVA) was used to compare activity for each 5-min period, separately, a m o n g all experimental groups (Steel and Torrie, 1980). Next, activity data were pooled for all tadpoles exposed to the same concentration
of Pb prior to placement into the fluviarium. ANOVA was then used to compare activity for each of the two 5-min experimental periods (pre-infusion, infusion), separately, among the five Pb-exposure conditions. The homogeneity of the variances of the pooled activity data for each Pb-exposure condition was also tested for each 5-min period using the test procedure described in Steel and Torrie (1980), with H0 that the variances were equal among all Pb-exposure conditions. If H0 was rejected, differences in the variance of locomotor activity (pooled according to Pb exposure) were examined a posteriori using the folded form of the F statistic (SAS, 1982), with H0 that the variances were equal for each comparison.
Water quality analyses Water quality parameters of laboratory water and of local pond water from which tadpoles were collected were determined using appropriate chemical tests (Hach Chemical Company) and atomic absorption analyses (Varian Spectra AA-20). Representative quality characteristics ( _+1 SD of the measurements; N = 11, each) of the filtered, dechlorinated laboratory water follow: pH 7.21 (_+ 0.23); conductivity, 727 (___42.7)/zmhos; total hardness, 340.5 (+ 18.3) mg. 1- 1 as CaCO~; total alkalinity, 285.2 (+ 31.4) mg" 1- 1 as CaCO3; free chlorine and total chlorine, below detection limits (<0.1 mg'l-~); nitrate, 0.75 (+0.96) mg.1-1; nitrite, 0.008 (+0.009) mg-1-l; ammonia, 0.30 (+0.07) mg-1-1; calcium, 88.1 (+6.7) mg.l-1; copper, 0.002 (+0.0001) mg-l-1; magnesium, 29.5 (+2.1) mg.l-1; and lead, 0.6 (+0.1) #g. 1- 1. The chemical state of metals affects the ability of aquatic animals to perceive them (Hartwell et al., 1987), however, determination of the chemical state of Pb (and of the other metals present) and its subsequent potential for complexation with organic and inorganic constituents of the water was beyond the scope of this study. Comparative water quality characteristics (_+ 1 SD; N = 7, each) of the pond water are: pH 7.89 (+0.39); conductivity, 1058 (+_ 189.8) /~mhos; total hardness, 232.6 (+ 54.3) mg.1-1 as CaCO3; total alkalinity, 146.4 (___36.3) mg-l-1 as CaCO3; free chlorine and total chlorine, below detection limits (<0.1 mg'l-1); nitrate, 0.75 (+0.50) mg.1-1; nitrite, 0.035 (+0.044) mg-l-1; ammonia, 0.42 (___0.12) mg.1-1; calcium, 54.9 (+ 15.3)mg.1-1; copper, 0.049 (+0.0005)mg'l-1; magnesium, 25.8 (+2.7) mg.l-1; and lead, 3.0 (+ 1.5)/zg.l-1 RESULTS AND DISCUSSION
Friedman's X2 tests (Table I) indicated no consistent preference or avoidance of Pb by bullfrog tadpoles and no consistent preference of the animals for any areas of the arena during the pre-infusion periods of experiments. Although there were some significant values, these are no more than would be expected by chance and are interpreted as misleading indicators. No significant values of the test statistic were obtained during the Pb-infusion periods. These statistical results are confirmed observationally by the lack of response of tadpoles encountering the plumes of Pb-
337 TABLE I Values of the Friedman to Pb-enriched water. Experimental series* CC CL CI CM CH LC LL LI LM LH IC IL II IM IH MC ML MI MM MH HC HL HI HM HH
~(2 statistic during experiments on the preference/avoidance of bullfrog tadpoles
Minutes since experiment began Pre-infusion period
9.12 4.02 8.18 7.81 3.88 12.71 2.95 16.54 b'P 3.03 6.84 6.09 5.44 11.49 5.13 3.32 12.03 12.29 8.54 5.99 5.40 5.34 4.21 4.94 10.28 9.97
5.51 2.99 17.16 b'A 5.45 4.10 8.07 11.83 8.99 9.83 2.72 4.52 10.17 9.50 5.05 6.03 11.65 4.21 10.10 4.84 6.09 12.06 3.14 5.18 19.23 a'P 4.45
9.21 5.40 11.81 9.98 7.72 11.98 5.77 5.65 2.02 10.10 4.68 4.43 3.38 4.12 4.35 4.77 12.56 4.12 7.83 14.12 ~'A 10.25 3.93 6.18 4.28 5.28
7.84 6.40 9.96 8.49 7.45 11.52 6.44 5.50 2.36 6.81 4.55 9.21 8.34 6.95 6.18 5.26 l 1.49 4.68 3.32 13.47 7.38 12.88 15.51 c'P 6.62 6.33
6.72 8.07 9.02 5.67 7.37 3.79 17.95 b'A 10.75 4.75 7.93 9.96 7.53 5.16 8.18 5.68 3.90 9.88 8.07 5.77 5.64 9.39 12.37 4.00 6.51 4.33
5.42 6.10 4.96 5.48 6.87 7.99 5.54 3.54 6.81 6.54 6.62 4.15 3.31 4.89 7.14 4.58 1.81 6.38 3.58 6.38 7.78 5.84 8.10 5.55 6.14 9.52 4.05 11.13 8.34 4.59 5.63 10.65 12.66 9.40 2.91 7.20 7.00 4.84 6.27 3.26 5.90 6.33 4.85 4.47 8.11 3.64 5.67 5.10 4.31 2.44
10.83 9.58 11.38 2.64 7.39 9.44 9.52 2.80 2.69 6.87 4.88 2.26 4.26 5.62 6.10 6.16 5.23 4.79 11.75 10.15 5.64 9.12 8.03 4.89 3.15
8.40 7.26 8.13 7.79 7.63 4.51 5.91 2.37 8.41 6.37 5.68 4.52 5.64 5.78 4.05 3.70 2.73 5.70 6.20 7.20 3.72 1.63 5.12 7.65 8.92
7.24 3.50 6.51 5.63 7.25 8.26 5.61 4.13 3.39 5.96 3.99 8.72 2.59 10.42 6.10 5.62 7.06 5.08 6.96 4.79 10.32 9.95 11.84 7.58 4.57
*First letter = concentration of prior exposure to Pb; second letter = concentration of Pb infused into fluviarium during 2nd 5-min experimental period: C = 0 ttg Pb- 1- l; L = 500/~g Pb. 1- ~; 1 = 625/~g Pb. l - a; M = 750/zg Pb" I - ~; H = 1000 p.g Pb" l - 1. a p < 0 . 0 1 ; bp<0.025; cp<0.05; ' P ' = Preference for and 'A' = Avoidance of the 'experimental' octant.
enriched water, i.e. there were no observable detection responses, or obvious behavioral stress. This lack of avoidance of concentrations of Pb which have been shown to inhibit learning and retention in both green frog tadpoles (Strickler-Shaw and Taylor, 1989) and bullfrog tadpoles (Strickler-Shaw, 1988) could be of ecological importance not only for the tadpoles (learning and/or recognition of predators and conspecifics could be affected since predator and conspecific recognition by tadpoles is apparently based, at least in part, on the learning of waterborne chemical cues (Waldman, 1982; Petranka et al., 1987)), but also for the animals following metamorphosis. However, this speculation remains unsupported since
338 behavioral alterations in adult frogs resulting from sublethal exposure to pollutants as tadpoles have not been assessed. Lead accumulates in both bullfrog and green frog tadpoles (Birdsall et al., 1986; Strickler-Shaw, 1988). Total tissue concentrations for tadpoles inhabiting roadside drainage ditches ranged from 0.7-270 mg. kg - 1 dry weight for bullfrog tadpoles and from 4.8-240 m g ' k g - 1 dry weight for green frog tadpoles (Birdsall et al., 1986). Strickler-Shaw (1988) exposed bullfrog tadpoles to the same concentrations of Pb as those used in the current study, under identical conditions; her study was also conducted at the same time as the current study and the animals shared the same food sources. Her results of total tissue analyses for Pb content of the tadpoles are reprinted in Table II. These data are important because comparison of the Pb content of control tadpoles with experimentals indicates that negligible Pb uptake occurred from laboratory water and food. There were no detectable effects of Pb exposure on spontaneous locomotor activity during the pre-infusion or Pb-infusion periods of the experiments (Tables III and IV), as indicated by A N O V A . A m o n g all experimental groups, F = 0 . 9 2 9 , mean square error (MSE)= 540.67, P = 0 . 5 6 3 , for the first 5 min (no Pb infusion) and F = 0.960, MSE = 410.09, P = 0.521, for the second 5 min (Pb infusion), with df= 24, 175 for both tests; among experimental groups pooled according to Pb-exposure, F = 1.396, MSE = 531.77, P = 0.237 for the first 5 min and F = 0.937, MSE -- 408.63, P = 0.443 for the second 5 min, with df= 4, 195 for both tests. Results of the preference/avoidance and locomotor activity analyses indicated no apparent increased sensitivity of the tadpoles to Pb-enriched water due to previous Pb exposure. However, the variances in the activity counts (as indicated by the SD) were consistently greater in all the Pb-exposure groups when compared to the control groups (Tables IH and IV). Tests of the homogeneity of the variances indicated significant differences among the Pb-exposure conditions for each 5-min period (X2= 35.74, o~=0.05, for the first 5 min; X2=22.77, a =0.05, for the second 5 min; df= 4 for each test). TABLE II Results of Duncan's multiple comparison tests for measured lead content of bullfrog tadpole tissues as determined by atomic absorption spectrophotometry(N= 6 tadpoles/exposure group). Pb-exposure conc (/zg Pb-l-1)
Mean tissue conc* (#g Pb.g-1 wet wt.)
0 500 625 750 1000
0.29 6.07 10.82 11.34 20.96
(0.37) a (5.57) ab (7.24) bc (5.57) bc (14.17) c
*SD in parentheses; values with different letters are significantlydifferent (~ = 0.05); values with the same letter are not significantly different (a > 0.05). Reprinted, with permission, from Strickler-Shaw (1988).
339 TABLE III Mean number of octant crossings per 5-min period for each experimental series of bullfrog tadpoles (N= 8 tadpoles per group). Experimental series a
CC CL CI CM CH LC LL LI LM LH IC IL II IM IH MC ML MI MM MH HC HL HI HM HH
Mean no. of crossings 1st 5 min (no Pb infusion)
2nd 5 min (Pb infusion)
50.8 52.4 50.9 60.6 50.4 57.9 59.8 60.1 47.8 49.1 61.1 54.6 52.5 55.3 46.5 51.8 59.5 44.1 45.4 48.0 60.0 48.8 57.7 60.8 51.3
(8.33) (9.68) ( 15.94) (10.59) (7.39) (28.11) (13.09) (27.27) (16.54) (20.91) (26.97) (14.13) (18.71) (22.61) (14.21) (17.40) (31.31) (22.49) (16.21) (10.31) (25.77) (19.07) (28.84) (21.70) (25.01)
46.5 46.5 49.6 59.1 43.5 49.1 55.9 53.0 41.0 43.9 59.5 44.5 57.9 55.2 46.6 51.1 42.0 42.4 37.9 53.6 50.6 50.1 52.4 50.9 44.8
(6.26) (10.43) (8.25) (10.53) (7.64) (17.00) (19.13) (27.07) (10.55) (21.50) (31.57) (18.98) (31.06) (22.50) (14.64) (12.61) (15.66) (9.45) (15.94) (18.90) (19.75) (I 3.98) (22.34) (18.16) (23.03)
aFirst letter = concentration of prior exposure to Pb; second letter = concentration of Pb infused into fluviarium during 2nd 5-min experimental period: C = 0 #g Pb' 1- ~; L = 500/~g Pb' 1- ~; I = 625/~g Pb. 1- ~; M = 750 #g Pb. 1- ~; H = 1000/~g Pb" 1- ~.
R e s u l t s o f t h e f o l d e d F tests i n d i c a t e d s i g n i f i c a n t d i f f e r e n c e s in t h e v a r i a n c e o f l o c o m o t o r a c t i v i t y b e t w e e n t a d p o l e s e x p o s e d to 0 #g P b . l - 1 ( c o n t r o l s ) a n d t h o s e exp o s e d t o 500, 6 2 5 , 7 5 0 o r 1000 #g P b - l - 1, f o r b o t h 5 - m i n p e r i o d s o f t h e e x p e r i m e n t s ( T a b l e V). T h e s e f i n d i n g s a r e s i m i l a r to t h o s e o f T a y l o r et al. (1989) f o r g r e e n f r o g t a d p o l e s ( f o l l o w i n g e x p o s u r e t o 750 o r 1000 #g P b . 1 - 1 b u t n o t t o 500 #g P b - l - 1). I n c r e a s e d o r d e c r e a s e d v a r i a b i l i t y in a c t i v i t y o f a n i m a l s f o l l o w i n g s u b l e t h a l e x p o s u r e t o a p o l l u t a n t is i n t e r p r e t e d as i n d i c a t i o n o f i n c r e a s e d stress a n d o f b e h a v i o r a l t o x i c i ty ( C o r y - S c h l e c t a a n d W e i s s , 1985; Steele, 1989). C h a n g e s in t h e v a r i a b i l i t y o f locomotor activity following treatment have also been found by Cory-Schlecta and W e i s s (1985) a n d C o r y - S c h l e c t a et al. (1985) f o r rats e x p o s e d t o l e a d , b y Steele (1989)
T A B L E IV
Mean number of octant crossings per 5-min period for each experimental g r o u p o f b u l l f r o g t a d p o l e s p o o l e d according t o P b - e x p o s u r e c o n d i t i o n ( N = 40 for each pooled g r o u p ) .
Experimental series a
Mean n o . o f crossings 1st 5 min ( n o P b infusion)
2nd 5 min (Pb infusion)
aC = 0 #g P b - 1 - ~; L = 5 0 0 / x g P b . 1 - 1; I = 6 2 5 / z g P b " 1- ~; M = 750 #g P b . 1 - ~; H = 1000 #g P b . 1- 1
Results of folded F-tests on the variances in locomotor activity between experimental g r o u p s o f b u l l f r o g tadpoles pooled according t o P b - e x p o s u r e c o n d i t i o n ( N = 4 0 for each pooled g r o u p ) . Group
1st 5 min ( n o P b infusion)
2nd 5 min (Pb infusion)
C vs L
C vs I
C vs M
C vs H
L vs I
L vs M
L vs H
I vs M
1 vs H
M vs H
aC = 0 / ~ g P b ' 1 - l; L = 500 ttg P b . 1 - ~; I = 625 g.g P b . 1 - l; M = 7 5 0 / x g P b " 1 - l; H = 1000 #g P b - 1 - ~.
bCritical values o f F' (df= 39,39): P0.05 = 1.70; P0.025 = 1.89; Po.ol = 2 . 1 2 ; P0.005 = 2 . 3 1 .
for sea catfish, Ariusfelis, exposed to copper, and by Levine et al. (pers. commun.) for lumbricid earthworms exposed to a variety of metals in sewage sludge. Altered variability in response appears to be a more sensitive indicator of behavioral toxicity of Pb than changes in spontaneous locomotor activity, although the ecological implications of changes in the variability in activity are not easy to assess. However, in a normal distribution of a trait, those individuals recorded in the 'tails' of the distribution are usually selected preferentially against (in some cases, they may be preferentially selected for). Increasing the variability of expression of a trait tends to 'flatten' the normal curve, increasing the proportion of a population in the 'tails', thus increasing the proportion of a population that may be selected
against in nature. There is no evidence to support this speculation for tadpoles. Steele (1989), however, has demonstrated that altered variability in activity contributes to disruptions in species-specific diel activity patterns in sea catfish, A. felis, which may cause the fish to be active, or inactive, at inappropriate times. Comparison o f the results of the current study with those o f Taylor et al. (1989) indicates that bullfrog tadpoles are more susceptible than green frog tadpoles to the effects of Pb pollution on variability in locomotor activity. In addition, comparison o f the results obtained by Strickler-Shaw (1988), and Strickler-Shaw and Taylor (1989) indicates that bullfrog tadpoles are also more susceptible than green frog tadpoles to the effects of Pb pollution on learning and memory. Bullfrog tadpoles may, therefore, be a more sensitive indicator of sublethal environmental Pb toxicosis than green frog tadpoles. Similarly to green frog tadpoles (Taylor et al., 1989), bullfrog tadpoles also apparently do not detect Pb, or do not experience sufficient irritability to develop an avoidance response to Pb, at least for the concentrations used here. Confirmation o f a lack o f avoidance o f behaviorally deleterious concentrations of Pb in two such closely-related ranid species suggests that the lack of an avoidance response to Pb may be a phenomenon of ranids in general. However, because the chemical state of metals affects their detectability by aquatic animals (Hartwell et al., 1987), the lack of a detectable avoidance response to Pb by bullfrog tadpoles could also result from chelation or complexation of Pb with constituents of the laboratory water, lowering the nominal ionic concentration o f Pb below the detection or irritability threshold. As indicated by Strickler-Shaw (1988), however, similar exposure of bullfrog tadpoles in the same laboratory water to 625,750 or 1000/zg Pb- 1- 1 produces deficiencies in both learning acquisition and retention, and exposure to 500 #g Pb" 1- 1 interferes with learning acquisition. The concentrations used in the current study were also sufficient to induce significantly greater variability in locomotor activity in the groups of Pb-exposed animals, a subtle behavioral effect. Thus, regardless of any removal of ionic Pb from the water through chelation a n d / o r complexation, sufficient amounts remain to produce behavioral toxicity. The apparent lack of avoidance by tadpoles of water containing sublethal, behavior-altering concentrations of Pb complicates the environmental problem of Pb pollution. Tadpoles may remain in Pb-polluted waters where they will be susceptible to chronic exposures, which is especially important considering the time dependence for development of impaired learning and memory in tadpoles following sublethal Pb exposure (Eby, 1986; Strickler-Shaw, 1988; Strickler-Shaw and Taylor, 1989). The results also suggest an inadequate protection of field populations of ranid tadpoles under the current standard (USEPA, 1980) of 766 #g. 1- 1 as the maximum 'safe' allowable concentration of Pb in water o f the hardness (340 mg CaCO3' 1- 1) used in this study. Concentrations of Pb similar to those used here may similarly affect behavior in
field populations of bullfrog tadpoles in waters with hardness similar to that of the laboratory water, although the presence of additional constituents in natural waters that are not present in the laboratory water (e.g., humic acids, silt/sediments, particulates, etc.) may ameliorate the ecological impact by reducing the concentration of ionic Pb in the water. However, additional routes of exposure to Pb undoubtedly exist in aquatic systems, e.g. through contact with lead-polluted sediments and food, and Pb bioaccumulates in tadpoles (Birdsall et al., 1986; Strickler-Shaw, 1988). The contribution of these additional routes of exposure to behavior-altering sublethal Pb toxicosis is not reported in the literature and was beyond the scope of the objectives of our study. These additional important aspects of environmental Pb pollution should lead to future comparative investigations of preference/avoidance, bioaccumulation, routes of uptake and metal chelation/complexation. ACKNOWLEDGEMENTS
This research was funded, in part, by an Academic Challenge grant from the State of Ohio to D.H. Taylor et al., and the Department of Zoology, Miami University. REFERENCES Birdsall, C.W., C.E. Grue and A. Anderson, 1986. Lead concentrations in bullfrog Rana catesbeiana and green frog R. clamitans tadpoles inhabiting highway drainages. Envir. Pollut. (Ser. A) 40A, 233-247. Cory-Schlecta, D.A. and B. Weiss, 1985. Alterations in schedule-controlled behavior of rodents correlated with prolonged lead exposure. In: Behavioral pharmacology: the current status, edited by L.S. Selden and R.L. Balster, Alan R. Liss, Inc., St. Louis, MO, pp. 487-501. Cory-Schlecta, D.A., B. Weiss and C. Cox, 1985. Performance and exposure indices of rats exposed to low concentrations of lead. Toxicol. Appl. Pharmacol. 78, 291-299. Eby, J., 1986. Sublethal concentrations of cadmium and lead affect acquisition of discriminate avoidance learning in green frog (Rana clamitans) tadpoles. M.S. thesis, Miami University, Oxford, OH. Hartwell, S.L., D.S. Cherry and J. Cairns, Jr., 1987. Field validation of avoidance of elevated metals by fathead minnows (Pimephales promelas) following in situ acclimation. Envir. Toxicol. Chem. 6, 189-200.
Petranka, J.W., L.B. Kats and A. Sih, 1987. Predator-prey interactions among fish and larval amphibians: use of chemical cues to detect predatory fish. Anim. Behav. 35,420-425. Roberts, R., 1962. Experimental embryology, 3rd ed. Burgess Publishing Co., Minneapolis, MN. SAS, 1982. SAS user's guide: statistics. SAS Institute, Inc., Cary, NC. Scarfe, A.D., C.W. Steele and G.K. Rieke, 1985. Quantitative chemobehavior of fishes: an improved methodology. Envir. Biol. Fish. 13, 183-194. Schefler, W.C., 1979. Statistics for the biological sciences. 2nd Ed. Addison-Wesley Publishing Co., London. Steel, R.G.D. and J.H. Torrie, 1980. Principles and procedures of statistics: a biometrical approach, 2nd ed. McGraw-Hill, New York. Steele, C.W., 1986. Responses of zebrafish, Brachydanio rerio, to behavior-altering chemicals. Ph.D. dissertation, Texas A&M University, College Station, TX. Steele, C.W., 1989. Effects of sublethal exposure to copper on diel activity of sea catfish, Ariusfelis. Hydrobiologia, in press.
343 Strickler-Shaw, S., 1988. The effects of low lead concentrations on acquisition and retention of avoidance learning in tadpoles of the bullfrog, Rana catesbeiana. M.S. thesis, Miami University, Oxford, OH. Strickler-Shaw, S. and D.H. Taylor, 1989. Sublethal exposure to lead inhibits acquisition and retention of discriminate avoidance in green frog (Rana clamitans) tadpoles. Envir. Toxicol. Chem., in press. Taylor, D.H., C.W. Steele and S. Strickler-Shaw, 1989. Responses of green frog (Rana clamitans) tadpoles to lead-polluted water masses. Envir. Toxicol. Chem., in press. U.S. Environmental Protection Agency, 1980. Ambient water quality criteria for lead. USEPA, Washington, D.C. Waldman, B., 1982. Sibling association among schooling toad tadpoles: field evidence and implications. Anim. Behav. 30, 700-713. Weir, P.A. and C.H. Hine, 1970. Effects of various metals on behavior of conditioned goldfish. Arch. Envir. Health 20, 45-51.