WSP06488

• File Number: 8240.300.31.J
• Description: San Juan River - Environmental Studies
• Basin: San Juan/Dolores
• Water Division: 7
• Date: 11/1/2000
• Title: The Chronic Toxicity of Dietary and Waterborne Selenium to Adult Colorado Pikeminnow in a Water Quality Simulating that in the San Juan River
• Water Supply Pro - Doc Type: Report/Study

'I.

.,
;
'"j e""-':)
C.J
.i i\)

I~
c...:
....1

". i

,;'" ~

..;

.- -~

" .j

"':

, :

~P:

. )

, ,

There were no differences in AGR-L among treatments (F~I.08, P~0.43) or between

sexes (F=0.35, P~O.56), but AGR-L differed among exposure periods (F~36.45, P<O.OI) and

there was a significant interaction between treatment and period (F~2.04, P~O.OI), which

indicated that changes in AGR-L over time were not the same among treatments, Consequently,

LSD comparisons of AGR-L were made using treatment means pooled across sex, rather than

mean AGR-L for each sex pooled across treatments as is given in Table 7. Results of these

comparisons showed that although the rank order of AGR-L varied among treatments (Appendix

G), the largest AGR-L for fish in each treatment occllrred between days 90 and 120 (range 0.10

to 0.22 mm1d) and the smallest or second smallest AGR-L occurred during the following period

from days 120 to 153 (range -0.20 to -0.10 mm1d; Appendix 0),

Reproductive response

Only 10 of35 females (29%) spawned and spawning success was highly variable among

treatments. The control and treatments 2 and 6 had only one female spawn, treatments 4 and 5

had two females spawn, and the treatment 3 had three females spawn. It was highly unlikely that

any of the females spawned in the exposure tanks because no eggs were ever observed during

daily siphoning and filtering procedures. All males expressed milt during the spawning trials,

but 6 of 19 males (32%) expressed less than 20 ml, compared to at least 50 ml for the other 13

males. Results of the factorial ANOV A tests did not detect any significant treatment effects

(dietary selenium, waterborne selenium, and diet and water interaction) on number of eggs

expressed, egg size (diameter and weight), hatchability, time to hatch, or on survival, growth,

and deformities ofresulting larvae (Appendix H). However, the reproductive data are

problematic due to the lack of replicate spawns for half of the treatments. The Mean Square

error term in the ANOV A tables had only two degrees of freedom and the power of these tests

35