Agree with Celie. Also, note that now a day, larvae and embryo are used for toxicity tests as well. They are more susceptible than juveniles to toxins. Use of each life stage has certain advantages and disadvantages.
I am also in agreement with Celie on the reasons for using juvenile fish. However, current guidelines are shifting towards the use of early life stages of fish in accordance with the 3Rs (replacement, refinement and reduction) principle. Also, it affords a high-throughput screening of toxic compounds as well as sediment toxicity.
Some advantages amongst others have been put forward by Seyyed but I believe the advantages are more than the disadvantages.
I'd say it's a matter of what you want to investigate. Do you want to derive a LC50 or a NOEC ? In that case, egg and larvae development or juveniles are sufficient (OECD 203 and 210 guidelines, for example). Celie and co. gave you the reasons. If you want to investigate the toxicity of chemicals on fish reproduction, then working at least on mature fish is mandatory (e.g. OECD 229).
using juvenile fishes may be advantageous for the following reasons
1. in bioasays, smaller sized fish will require less reagents, toxicants, holding media and smaller facility in general which is necessary to make experiments practicable and easier to handle.
2. juveniles are a life stage that are actively growing thus they are very suitable to demonstrate growth inhibition and other indices of cellular toxicity.
3. juveniles are also suitable for chronic toxicity studies because as they mature into adult hood the effects of the toxicant on maturity and growth of reproductive organs can also be easily demonstrated.
Larvae and embryos may be more sensitive than juveniles sometimes, but not always; depends on the chemical. Juveniles are the best choice for all around sensitivity.
Also I added that juveniles are more suitable as their life history should be more suitable and they would under less stress to interfer on bioassays experiments