Previous Page Table of Contents Next Page



G.G. Savostyanova
Scientific Research Institute of Lake and Rier Fisheries
Leningrad, U.S.S.R.


In the Soviet Union, with its great variety of natural zones, there are conditions eminently suited for trout breeding. At present, trout breeding farms of the U.S.S.R. produce slightly more than 100 t of trout per year. However, practical measures are being taken to extend the network of trout breeding farms and to increase the output of this commodity.

Recently much attention has been paid to selection and pedigree stock breeding in trout raising. Scientists in a number of countries have succeeded in increasing considerably the rate of growth of trout, reducing the time of maturation, and increasing the fertility and resistance to furunculosis (Embody and Hayford, 1925; Leitritz, 1959; Lewis, 1944; Donaldson and Olson, 1955; Donaldson, 1960). All this shows the important role of selection in trout breeding.

By applying advanced methods of raising trout and systematic selection work it has been possible to reduce to a minimum the expenditure on the maintenance of the foundation stock and considerably increase the efficiency of trout raising by increasing the rate of growth enabling a reduction of time necessary to rear the commercial commodity.


The home of the rainbow trout is the United States. In the second half of the nineteenth century, American fish breeders began acclimatizing trout in rivers and breeding them in ponds. When acclimatizing the trout no importance was given to the fact that some species were anadromous and others non-anadromous. In the process of breeding, individual species were crossed many times, which gave rise to the hybrid form known as rainbow trout. Thanks to its high palatability and a higher rate of growth than that of brook trout raised in trout farms, as well as owing to spring spawning and other useful characteristics, the rainbow trout in a short period of time has become dominant in trout culture.

At the end of the last century, rainbow trout were brought to England, Scotland, France, Japan, Australia, New Zealand, South Africa and Germany. Rainbow trout were brought to Russia from Germany in 1890, but in the years of the war they almost completely perished. In 1948, eggs of rainbow trout were again brought to the U.S.S.R. from Germany (Marcel farm). Out of these eggs, a foundation stock was raised in the Ropsha pond fishery in the Leningrad region. Later, eggs, sperm and broodstock from the Ropsha farm were transferred to different fisheries of the Soviet Union. Again trout eggs were brought to the Ropsha farm from the Pylula trout breeding farm in the Estonian SSR.

As rainbow trout came to Russia from Germany, it is worth considering here the data available on the origin of German rainbow trout. Schäperclaus (1961) regards the rainbow trout raised in Germany as a product of crossing the following three forms of salmonids:

  1. Shasta,1 brought to Germany from rivers of California in 1882. This form is non-anadromous. Its spawning period falls in January and February.

  2. Steelhead,1 imported to Germany in 1896. This is an anadromous form with a later spawning period (March-April).

  3. Males of cut-throat1 were brought to Germany in 1889 with the purpose of crossing with the trout that had been brought earlier.

In recent years these three forms crossed with one another many times, and in most cases it is impossible to single out any of them as a pure species.

The new variety of the rainbow trout obtained in Europe has characteristics of all initial forms. This mixed form is the progenitor of the trout which are raised in the U.S.S.R. Thus the present day stock of rainbow trout which we deal with is a result of crossing of trout from different trout breeding farms. The hybrid origin of the rainbow trout makes it rewarding material for selection.

Until now no selective breeding for pedigree stock has been carried out on trout farms of the Soviet Union. One of the main tasks of trout culture in the country at present is the organization of selection work. At primary stages of selection, great importance is attached to the detailed estimation of different pedigree groups of fish and the choice of the most worthwhile groups for further selection.

Eggs of Danish and Czechoslovakian trout and sperm of Czechoslovakian and German (GDR) rainbow trout were brought to the Gostilitsy experimental farm in 1964–67. The sperm was used for fertilizing the eggs of local Gostilitsy trout. Thus five groups of rainbow trout were obtained and characterized; namely, Danish, Czechoslovakian, Gostilitsy (local), hybrids of Gostilitsy and Czechoslovakian trout (G × C) and, hybrids of Gostilitsy and Germany trout (G × Ge).

1 Shasta and steelhead are considered to be different strains of rainbow trout Salmo gairdneri and the cutthroat S.clarki.


3.1 Growth

For studying the rate of growth in the first year of life, fingerlings and yearlings of different groups of rainbow trout were experimentally raised together. The indices of growth in the first year of life showed some superiority of the Danish trout. However, its comparison with other groups is difficult due to the difference in weight at the time of planting and the effect this fact had on further rate of growth. At the time of planting fry for combined raising, the mean weight of the Danish trout always exceeded that of fish of the other groups. Therefore, it was impossible to obtain precise comparative data on characteristics of different trout in the first year of life.

Difference in the rate of growth is more distinct in the second year of life. Observations spread over three years showed clearly that Danish trout, stocked in the second year of life (of weight close to that of the other groups of fish), gave the maximum gain in weight (Table I). The Gostilitsy group showed the slowest growth.

Along with combined rearing of different groups in ponds, two-year-old specimens of the same groups were raised separately in live boxes (cages) with three replications. When reared separately in cages, specimens of different groups differed in weight to a lesser extent that when raised together in ponds. However, the Danish trout registered the highest rate of growth and survival (Table II).

The Danish trout has proved somewhat better also in feed conversion as compared to other groups. The Gostilitsy trout takes the last place in all indices. For separate and combined culture, the largest specimens were chosen. Selection in different groups was done with different stocking densities. In spite of the fact that in the Gostilitsy group selection was more strict, the Danish trout retained its superiority.

Based on the data obtained when taking stock of spawners the fish culturists had the opportunity to trace the growth of trout of different groups during their first three years of life. Stocks of the Danish trout overtake those of other groups in the rate of growth, the difference between them gradually increasing, and by the time they are four-year-old they gain 0.5–0.7 over the others (Table III).

Coefficient of condition of different groups of rainbow trout varies from 1.11 to 1.47. The parents of different groups have the same coefficient of condition. Also, it was observed that the coefficient of condition of four-year-old individuals of all groups of trout is somewhat lower.

3.2 Maturation and fecundity

Some females reach maturity at the age of two. Mature specimens of this age were observed among Danish, Czechoslovakian and hybrid (G × C) trout. None of the Gostilitsy trout were mature at the age of two. Females of the Danish trout, considerably excelling other females in the rate of growth, also differ from them in fecundity and diameter and weight of eggs. Statistical analysis (one-factor dispersion analysis) has disclosed close connection between the weight of fish and their fecundity and between the weight and the size of eggs. It proves that weight is the most important criterion in selection. Consequently, for breeding purposes it is necessary to choose fast growing fish which will prove not only more fecund but will have larger eggs.

Quality of sperm of different trout was ascertained by the following characteristics: duration of the active state of sperm and concentration of sperm in relation to volume of milt. Attempts to determine the percentage of non-viable sperm gave no results. The fish breeders failed to apply the method of colouring live sperm with 5 percent solution of eosin, as successfully done by Zhukinsky (1965) in studying the sperm of roach and carp. Smears showed no difference in the colour of live and dead sperm.

Table I

Gain in weight in combined rearing of two-year-old trout of different groups in 1965–67

Groups of troutNumber of individualsMean weight (g)
When stockedWhen harvested
Danish1 90025    256
Czechoslovakian2 39524    176
Hybrids (G × C)   31023    174
Danish4 86439.7173
Gostilitsy11 114  35.6136
Hybrids (G × C)   87038.2134
Danish2 52446.2296
Czechoslovakian1 04247.5254
Gostilitsy   17644.0201

Table II

Gain in weight in separate rearing of two-year-olds of different groups of rainbow trout

GroupNumber of individualsMean weight (g)
When stockedWhen harvestedPercent survival

Table III

Characteristics of females of different groups of trout

AgeMean weight
Number of eggsMean diameter of eggs (mm)Mean weight of eggs (mg)
25171 8603 6503.7326.44
31 172   3 0414 9004.6646.53
42 070   4 6766 700--
23721 5002 9003.6720.59
38852 5673 9964.5044.62
41 522   3 9385 300--
  Hybrids (G × C)  
24142 1002 8003.7223.33
37602 5483 9004.3039.94
31 012   2 7443 600--
41 245   3 0646 6004.7750.03

Comparing two-year-old males, a slightly lower activity of sperm was observed in the Czechoslovakian trout. No significant difference was found in the other indices. Among three-year-old males, the highest activity of sperm was seen in the Danish trout. A marked difference was observed between the Danish (79.5 cm) and the Gostilitsy (57.4 cm) trouts. A slight reduction in the period of activity and lower concentration of sperm were observed in three-year-old rainbow trout when compared with two-year-olds. On the other hand, the volume of milt of the older fish doubled. It was not possible to establish any connection between the activity or concentration of sperm, volume and milt and the weight of fish, or the coefficient of condition within an age group.


The main task facing trout breeding in the U.S.S.R. is to increase the productivity of trout ponds to 50 or more tons per ha. In the past, attention has been paid mainly to the problem of feed and improvement of existing ponds. Now it is necessary to begin work on selection aimed at breeding new strains with higher rate of growth, higher fertility and resistance to diseases.

Efficiency of selection, to a great extent, will depend on the choice of up-to-date methods of selection and the way they are applied. There are two main methods of artificial selection: mass selection and selection for relatives. Individual selection for relatives is used on a wide scale in cattle breeding. This method allows evaluation of the sires by their genotype. Mass selection is by no means unimportant, but allows individual specimens to be evaluated by phenotype only.

Fish culturists practising selection of fish differ in opinion as to the efficiency of mass selection and selection for relatives. For instance, Moav and Wohlfarth (1960, 1963, 1967), Nakamura and Kasahara (1955, 1956, 1957) and Kryjazheva (1966) are of the opinion that carp selection should be based on progeny testing. Schäperclaus (1961) recommends giving more attention to mass selection. Kirpichnikov (1966) recommends that both methods should be combined to increase the efficiency of selection. American and Polish salmonid-breeders used only the mass method of selection.

The rainbow trout culturists of the U.S.S.R. chose the mass method as the chief means of improving the hereditary traits of rainbow trout at the first stages of selection. The main criterion for selection was weight; additional criteria were defectlessness and the state of health.

4.1 Selection for weight in different age groups

Mass selection for weight is relatively easy to carry out, and its severity and intensity may be considerably increased. The selection of rainbow trout for weight dealt mostly with one- and two-year-old fish. The largest specimens were chosen from each group for further rearing. Severity of selection among one-year-old individuals in different groups of trout varied considerably (Table IV).

Selection in the Gostilitsy group was carried out with great severity both in 1966 and 1967, as this group was slower than the others in growth, and in order to stock equal weights the trout breeders had to select this group with greater strictness.

Some authors dealing with carp (Schäperclaus, 1955; Stegman, 1965a; 1965b; Kirpichnikov, 1966) have shown differences in the rate of growth of fingerlings in different years of life, i.e., in two- and three-year-old fish. Though large fingerlings as a rule retain their advantage in the second year of life, part of the fish selected in the autumn begin to grow slower, and they join the middle weight group. Likewise, part of the fish that were not selected by the end of the second growing period catch up with the champions (largest in previous growth periods). From 10 to 50 percent of fish selected in the first year as champions in weight cease to be so in the second and third years of life. The same is true of rainbow trout. Therefore, the trout breeders tried to select mostly from two-year-old trout, as it is at the age of two years that rainbow trout become commercially valuable. Selection was strictest among two-year-old rainbow trout (Table V).

Table IV

Selection among yearlings of different groups of rainbow trout

GroupNumber of fishSeverity of selection (percent)Mean weight (g)Selectional difference (g)
before selectionafter selection
Danish4 71936.036.549.513.0
Gostilitsy11 000  20.017.435.618.2
Hybrids (G × C)   91180.036.037.5  1.5
Danish9 25034.
Czechoslovakian3 71923.024.144.820.7
Gostilitsy1 728  5.710.831.720.2

Table V

Selection among two-year-old fish of different groups

GroupNumber of fishSeverity of selection (percent)Mean weight (g)Selectional difference (g)
before selectionafter selection
Danish 225  5.0173388215
Gostilitsy  57  0.7136277141
Hybrids (G × C)  11  1.0134291157
Danish7 349  20.2296392  96
Czechoslovakian  7810.1254246  91
Gostilitsy  6052.6201233  32

To preserve a sufficient number of specimens for further selection purposes, selection was carried out with different severity, depending on the available amount of material. Thus for a number of years, selection in the Danish group was carried out with less severity than in the other groups.

According to Kirpichnikov (1966) maximum severity of selection should be 1:500 or 1:1 000 (from 0.1 to 0.2 percent). Such a coefficient of severity can be easily obtained when selecting fish with high fecundity. With lower fecundity, as in the case of rainbow trout, selection may be carried out at the ratio of 1:20 (5 percent). Selection with lower severity will result in abrupt decrease in intensity.

It is well known that the efficiency of mass selection depends on the intensity of selection (i), standard deviation (σ) and heritability of traits (h2).

As no data have been obtained on weight heritability of different groups of rainbow trout, it is not possible to say in which group mass selection will be most effective.

As the Gostilitsy trout was undoubtedly subject to intensive inbreeding, weight heritability of this group cannot be more than, but in fact must be lower than, that of the Danish trout. Also, as the Danish trout has somewhat higher variation as compared with the Gostilitsy trout, it may be assumed that the effectiveness of selection among Danish trout will be higher.

The Danish trout is undoubtedly suited for further selection. Equally important is breeding a hybrid stock by crossing the various groups of rainbow trout. Work on these lines has been initiated by crossing Danish, Gostilitsy and Czechoslovakian trout. An equal number of females was taken from each group and their eggs were mixed and fertilized by mixed sperm obtained from an equal number of males taken from these groups. The use of this heterogeneous group will increase abruptly the effectiveness of the next selection.

4.2 Histophysiological control

Selection in trout breeding is accompanied by intensive feeding, which often results in fatty degeneration of the liver. As this disease shows no clinical symptoms for a long time and, more often than not, occurs among quickly growing specimens, it is necessary to exercise thorough histological control of the condition of the livers of the rainbow trout under selection. Also, when comparing different selection groups of trout, it is important to determine their liability to fatty and cirrhotic degeneration of liver so as to carry out further selection by this trait.

When comparing different groups of rainbow trout no cirrhotic degeneration was seen during the entire period of observations, though the conditions of the liver varied in the different groups. A state very close to normal as well as a state very far from normal were observed. In the former state the parenchyma of liver looked compact and was free of fat interstices, while in the latter, parenchymal preparations had many large fat deposits and the liver tissue looked quite lacy. Thus, all groups of rainbow trout showed that they are susceptible to pathological obesity of liver, the susceptibility being higher in the Danish trout (Faktorovich, 1968).

In all groups of trout that were compared, a number of specimens showed a state of liver close to normal. The largest percentage of such specimens (24.5 percent) was observed among Gostilitsy trout. Presence of fish with nearly normal liver is evidence of the fact that in each group there are specimens resistant to liver disease. It is these specimens that should be the founders of a new strain-group possessing high economic characteristics and resistance to liver disease. It is therefore necessary to select spawners according to the condition of the liver. Such selection proved possible by the use of the method of live histological analysis of liver worked out by Faktorovich (1968). This method consists of a biopsy by the extraction of a bit of liver from the fish by puncturing the body wall.

4.3 Marking

Of great importance in selection work is tagging or marking fish. Precise estimation of different groups of trout necessitates combined rearing of fish, which is possible only if they are marked.

Of the various methods generally used for marking fish, fin clipping, use of hanging tags, branding, and injection of dyes have all been tried with success in marking trout. The use of attached tags has not proved successful.

Fin clipping in different combinations can be recommended for mass marking of trout weighing one gram or more.

Hanging tags are expedient for individual tagging of rainbow trout. They should be attached under the ventral fins to ensure retention. If tagging is carefully carried out, 100 percent of tagged fish retain their tags during the two years of observation.

Branding rainbow trout has shown that they can quite fairly endure the process of branding and its traumatic effects. Applying brands as a method of individual tagging greatly facilitates selection and stock breeding of trout.

Dichlorotriazene (M - procion) dyes were suggested by M.N. Melnikova for marking trout. These dyes are used in industry for dyeing textiles and have never been used for marking animals before.

For marking rainbow trout the following dyes were used: bright red, orange, green, and sky-blue. Marks with dichlorotriazene dyes not only last a long time but have no adverse side effects on the fish and so can be used for mass marking of trout. The variety of dyes allows them to be applied for simultaneous marking of several groups, which is of great practical application in selection.


The first results of rainbow trout selection in northwestern U.S.S.R. have shown the great importance of organized selection and stock breeding for raising the effectiveness of trout breeding farms. Selection allowed the formation of an improved foundation stock of different groups of trout.

By comparing different groups of trout raised in identical conditions in ponds it has been possible to ascertain that the Danish trout has the most valuable piscicultural characteristics, namely high rate of growth, high viability, early maturation and fertility. The least suited for culture is the slowly growing Gostilitsy trout. Retarded growth may be closely connected with inbreeding as well as with the fact that the Gostilitsy farm carried out no selective breeding of this trout. The Danish group can be used as a foundation for further selection of trout. Of great importance in selection of rainbow trout for rate of growth is in vivo examination of the condition of liver of fish under selection.

Selection gave an impetus to forming pedigree foundation stock of different groups of rainbow trout. Spawners of Danish trout proved to be far larger than other trout, and matured in the second year of life. At present they are able to give another selection generation.

One of the vital tasks of future selection of rainbow trout is to determine heritability of some selected characteristics, most important of which is weight heritability.

Another very important consideration is selecting specimens with normal liver. This will allow the breeding of a new strain of rainbow trout possessing high economic characteristics as well as resistance to liver disease when intensively fed.


Donaldson, L.R., 1960 Development of brood stocks of salmonid fishes. Research in Fisheries, Contr. 77:22–3

Donaldson, L.R. and P.R. Olson, 1955 Development of rainbow trout brood stock by selective breeding. Trans.Amer.Fish.Soc., 15:93–101

Embody, G.C. and C.O. Hayford, 1925 The advantage of rearing brook trout fingerlings from selected breeders. Trans.Amer.Fish.Soc., 55(I):135–42

Faktorovich, K.A., 1968 Histophysiological control over the condition of the liver in selected trout. Izv.gosud.nauch.-issled.Inst. ozer.rech.ryb.Khoz., (in press)

Kirpichnikov, V.S., 1966 Efficiency of mass selection and selection for relatives in fish culture, Genetika, (4):16–25

Krjazheva, K.V., 1966 Influence of stocking rate on viability, growth rate and variability of young hybrid carp. Izv.gosud.nauch.-issled.Inst. ozer.rech.ryb.Khoz., 61:80–101

Leitritz, E., 1959 Trout and salmon culture (hatchery methods). Fish.Bull., Sacramento, 107:1–169

Lewis, R.C., 1944 Selective breeding of rainbow trout at Hot Creek hatchery. Fish.Bull., Sacramento, 30:95–7

Moav, R. and G. Wohlfarth, 1960 Genetic improvement of carp. I. Theoretical background, Bamidgeh, 12(1):1–12

Moav, R., 1963 Breeding schemes for the genetic improvement of edible fish. Progr.Rep. 1962, U.S. Fish Wildl.Serv., Israel Govt., 39 p.

Moav, R., 1967 Breeding schemes for genetic improvement of edible fish. Progr.Rep., 1964–65, U.S. Fish Wildl.Serv. Israel Govt., 3–61

Nakamura, N. and S. Kasahara, 1955 On the earliest stage at which the shoot carp appears. Bull.Jap.Soc.sci.Fish., 21(2):73–6

Nakamura, N., 1956 On the effect of particle size and quantity of the food. Bull.Jap.Soc. sci.Fish., 21(9):1022–4

Nakamura, N. and S. Kasahara, 1957 On the result of culturing the modal group, and the growth of carp fry reared individually. Bull.Jap.Soc.sci.Fish., 22(2):647–8

Schaperclaus, W., 1955 Die Bewertung des Karpfens bei der Suchtauslese. Z.Fisch., 4(718):483–520

Schaperclaus, W., 1961 Lehrbuch der Teichwirtschaft. Berlin, Paul Parey, 582 p.

Stegman, K., 1965a Variability of weight gains in carp in the first two years of life. Zesz.nauk.Szk.Glow.Gosp.viejsk., 2:67–92

Stegman, K., 1965b Growth of carp in the second and third year. Zesz.nauk.Szk.Glow.Gosp. viejsk., 2:93–110

Zhukinsky, V.N., 1965 The dependence of the quality of sexual products and vitality of embryos upon the age of the spawners on the progeny in fish, “Ed. V.I. Vladimirov, Kiev, Publ. Akad.Nauk. U.S.S.R. “Naukova Dumka”, 94–124

Previous Page Top of Page Next Page