|
,;514
<br />COPEIA, 1985, NO. 3
<br />see a greater frequency of spawning activity and
<br />a higher percentage of consecutive-year spawn
<br />ing (i.e., increased reproductive effort) among
<br />older fishes.
<br />We here present evidence indicating that fe
<br />males spawn on a less than annual basis, pro-
<br />ducing a biased sex ratio in the spawning mi-
<br />gration of white suckers in the Mill Rive
<br />watershed, Amherst, Massachusetts. We also test
<br />the hypothesis that the frequency of spawning
<br />among females is directly related to increasing
<br />age.
<br />METHODS
<br />This study was done on the Lake Warner
<br />watershed in Amherst-Hadley, Massachusetts
<br />(Hampshire County). Lake Warner is a eutro-
<br />phic, 24 ha mill pond supporting a diverse fish
<br />fauna. Maximum depth is 2.5 m; mean depth is
<br />1 m. Each spring, ripe suckers migrate approx-
<br />imately 3.75 km up the lake's tributary, the Mill
<br />River, to the only spawning ground in the wa-
<br />tershed. A small check dam stops their up-
<br />stream progress except during high-flood pe-
<br />riods. The area 0.7 km downstream from this
<br />check dam is the only section of the watershed
<br />that provides extensive substrate and riffle-pool
<br />configurations appropriate for white sucker
<br />spawning as described by Breder and Rosen
<br />(1966). A small amount of spawning activity has
<br />been noted at several very restricted (<20 m in
<br />length) areas of gravel substrate downstream
<br />from this stream section. However, nearly all
<br />white sucker spawning in the watershed is be-
<br />lieved to take place in the uppermost 0.7 km
<br />section downstream from the check dam.
<br />During the spawning seasons of 10 April-5
<br />May 1979 and 10 April-21 May 1980, white
<br />suckers were collected throughout the spawn-
<br />ing area with a DC, backpack electroshocker.
<br />All captured suckers were measured and tagged
<br />with numbered anchor tags inserted in the dor-
<br />sal pterygiophores to the right of the dorsal fin.
<br />We removed the first three pectoral fin rays
<br />from all fishes in 1979 for aging, while in 1980
<br />fin rays were taken from every fifth individual
<br />of each sex. Estimates of the number of spawn-
<br />ing suckers were made for each year using the
<br />multiple census Schumacher-Eschmeyer esti-
<br />mator (Ricker, 1975). Chi-square contingency
<br />tables (Bagenal, 1978) were used to: 1) detect
<br />any differences in recapture percentage be-
<br />tween male and female white suckers of 12 size-
<br />classes; 2) evaluate the extent of random mixing
<br />d of tagged and untagged fishes; and 3) test
<br />whether all size-classes were equally vulnerable
<br />to capture.
<br />From 13 June-16 November 1979 and 6
<br />June-14 November 1980, adult suckers were
<br />captured in Lake Warner with 76 and 102 mm
<br />(stretched mesh) gillnets. Gillnets were set ran-
<br />domly throughout the lake except for areas of
<br />thick submergent or emergent macrophytes. All
<br />fishes were measured and tagged, and fin rays
<br />were taken from a subsample of fishes for aging.
<br />Estimates of the size of the adult population
<br />were made using the Schumacher-Eschmeyer
<br />estimator.
<br />Gillnets are notably selective gear. We as-
<br />sessed gillnet selectivity by the direct methods
<br />of Cucin and Regier (1966) and Hamley and
<br />Regier (1973). Direct estimates compare the size
<br />distribution of fishes captured in two or more
<br />mesh sizes to the size distribution of a known
<br />population (Hamley, 1975). In our study the
<br />known population consisted of fishes previously
<br />measured, tagged and released. Chi-square con-
<br />tingency tables (Bagenal 1978) were used to de-
<br />tect any differences in recapture percentages
<br />between the two mesh sizes for 12 size-classes
<br />of fishes, and among the 12 size-classes for each
<br />mesh size of net.
<br />The sex of spawning white suckers was easily
<br />determined by the presence of tubercles on the
<br />anal fin and lower lobe of the caudal fin of males,
<br />plus the presence of ripe gametes. Dimorphism
<br />of the anal fin (Reighard, 1920; Spoor, 1935)
<br />was used to sex fish at other times. Sex ratios
<br />from the 1979 and 1980 spawning runs and
<br />summer-fall populations were applied to the
<br />population estimates of these time periods to
<br />estimate abundance of each sex. Age-length keys
<br />were constructed for males and females from
<br />all fishes that had been aged. These keys were
<br />used to determine the relative frequency of each
<br />age-class of each seem or spawning runs and sum-
<br />mer-fall populations. stlmates of absolute
<br />numbers of fishes in eac age of each sex for
<br />the spawning runs and the su? I?mer populations
<br />were made by applying the frequencies to the
<br />population estimates of each sex?\To estimate
<br />percentage participation in the 1981 spawning
<br />run for each age of each sex, the estimated size
<br />of each year-class of each sex in the run was
<br />divided by the estimated number of individuals
<br />from corresponding year-classes in Lake War-
<br />ner, summer-fall 1979. We determined the
<br />percentage of fishes captured in the 1980
<br />spawning run that had been tagged during the
<br />QUINN AND ROSS-WHITE SUCKER SPAWNING 615
<br />TABLE 1. AGE AND SEX COMPOSITION OF WHITE SUCKER SPAWNING RUNS IN THE MILL RIVER, 1979 AND
<br />1980.
<br />1979
<br />Age `c Sex ConmPo- Bt.
<br />3 Male 6.1 104
<br /> Female 7.1 41
<br />4 Male 12.5 215
<br /> Female 9.5 54
<br />5 Male 25.3 433
<br /> Female 26.0 149
<br />6 Male 26.3 450
<br /> Female 29.6 169
<br />7 Male 17.2 294
<br /> Female 16.0 68
<br />>7 Male 12.7 218
<br /> Female 11.8 68
<br /> 1980
<br /> % "Po. Es.
<br />Age Sex Itian no.
<br />3 Male 10.5 188
<br /> Female 13.0 77
<br />4 Male 16.8 298
<br /> Female 14.3 85
<br />5 Male 26.6 473
<br /> Female 27.3 162
<br />6 Male 23.0 409
<br /> Female 21.7 129
<br />7 Male 14.4 256
<br /> Female 14.3 85
<br />>7 Male 8.9 158
<br /> Female 9.3 55
<br />1979 run, and analyzed this sample of consec-
<br />utive-year spawners as to age and sex.
<br />RESULTS
<br />We marked 662 spawning white suckers be-
<br />tween 10 April and 5 May 1979, of which 129
<br />were recaptured. The estimated number of fish-
<br />es in the spawning run was 2,284 (1,930 < N <
<br />2,796, P < 0.05). The sex ratio of captured
<br />suckers was 2.9 males:I female. A total of 579
<br />white suckers were tagged and 67 recaptured
<br />during 10 April-21 May 1980. The estimated
<br />number of fishes in the 1980 spawning run was
<br />2,376 (1,931 < N < 3,087, P < 0.05). The sex
<br />ratio of these fishes was 2.7 males:I female. Five-
<br />and six-year old fishes were the most abundant
<br />age groups for both sexes in the spawning runs
<br />of both years (Table 1).
<br />From 13 June-16 November 1979, 465 adult-
<br />sized white suckers were marked in Lake War-
<br />ner with 13 recaptured. These data yielded an
<br />estimate of 5,824 (4,307 < N < 8,989, P <
<br />0.05) suckers. The sex ratio of all captured fish-
<br />es was 1:1. Thus, approximately three times as
<br />many males as females present in the summer-
<br />fall of 1979 participated in the 1980 spawning
<br />run. A total of 355 white suckers were tagged
<br />in Lake Warner from 6 June-14 November
<br />1980 and nine were recaptured, yielding an es-
<br />timate of 7,214 fishes (4,573 < N < 17,072,
<br />P < 0.05). The sex ratio was 1:1. Although the
<br />confidence limits on the population estimates
<br />were broad, the sex ratio, based upon exami-
<br />nation of all captured fishes, accurately repre-
<br />sented the sex ratio of the population.
<br />Gear selectivity did not bias the population
<br />estimates in a major way. Electrofishing selec-
<br />tivity curves (Cucin and Regier, 1966) revealed
<br />that the probability of recapture was somewhat
<br />related to fish size. Recapture efficiency was
<br />lower for the smallest and largest size-classes.
<br />Highest efficiency was for fish 340-440 mm (to-
<br />tal length), which included 75% of all marked
<br />fishes. Results of the Chi-square contingency
<br />tables indicated that during the 1979 spawning
<br />run, small females (340-360 mm) and larger
<br />males (420-440 mm) were captured more effi-
<br />ciently than other fishes by electrofishing. Dif-
<br />ferential recapture efficiency among size-classes
<br />of females in 1979 was significant due to fre-
<br />quent recapture of small females. There were
<br />no other significant differences among size-
<br />classes or between sexes. The lack of consistent
<br />differences in recapture efficiency suggests that
<br />the capture and recapture probability was sim-
<br />ilar for most fishes. Although late arrivals or
<br />early immigrants could bias estimates of the size
<br />of the spawning runs, analyses of recapture ef-
<br />ficiency suggest that resulting bias was not ma-
<br />jor, and it would be toward overestimation of
<br />the number of spawners. The sex ratio would
<br />not be affected. Relative gillnet selectivity curves
<br />(Cucin and Regier, 1966), curves calculated from
<br />the selectivity equations of Hamley and Regier
<br />(1973), and data from the Chi-square contin-
<br />gency tables indicated that the two mesh-sizes
<br />had similar efficiencies in their overlapping se-
|