Previous Page Table of Contents Next Page


H.J. Egglishaw
DAFS, Freshwater Fisheries Laboratory, Pitlochry, Scotland


The number of salmon Salmon salar L. spawning in the upper River Tummel has always been regarded as low and from 1953 to 1981 an average of only 153 adults have migrated into it each year. The river is 79 km long, including 48 km through lochs, and it is fed by numerous tributaries, which together drain an area of 820 km2. There are major hydro-electric installations, with dams and diversions regulating the flow and affecting the successful migrations of smolts and adults. In 1979 a potential productivity of 165 000 smolts per year was determined. At present 10 000–15 000 smolts migrate each year, the majority being two or three years old and between 10 and 15 cm long. A programme of planting six streams with unfed fry commenced in April 1980, eggs being obtained from impounded adult fish.

Instantaneous mortality rates (M per day × 100) of the 0+ age class during the first growing season were low at 0.68–1.25 for plantings at 2.0–3.5/m2, and were usually 1.2–1.4 at plantings of 5/m2. Presence of 1+ salmon did not increase mortality of 0+ salmon. Mortality rates in the year following the first growing season were, at 0.14–0.39, low when compared with those of natural populations. The presence of large numbers of 1+ salmon in 1981 did not decrease numbers of trout. Mean daily growth rates of 0+ salmon were generally above 2 percent, with rates being lower in 1981 than in 1980. At the end of the growing season in 1981, 11 400 of an estimated 13 516 1+ salmon present in the streams were larger than 90 mm, and survivors from these will probably migrate as smolts in 1982.


Le nombre de saumons, Salmo salar L., frayant dans le cours supérieur de la Tummel, l'un des principaux affluents du Tay, a toujours été considéré comme faible; entre 1953 et 1981, seuls 153 adults en moyenne y ont migré chaque année. La Tummel a 79 km de long, dont 48 km à travers des lacs, et elle est alimentée par de nombreux affluents qui baignent une superficie totale de 820 km2. Il y a sur la Tummel plusieurs grands ouvrages hydro-électriques, avec des barrages et des dérivations qui règlent le débit des eaux et qui gênent la migration des tacons et des adultes.

En 1979, on a étudié l'écologie du saumon dans le réseau formé par le cours supérieur de la Tummel et par ses affluents. Compte tenu des résultats des recherches sur les populations naturelles de saumons et des expériences de repeuplement, on a évalué la productivité potentielle à 165 000 tacons par an. On estime à 10 000–15 000 le nombre des tacons descendant chaque année la rivière; la majorité d'entre eux ont deux ou trois ans et une longueur de 10 à 15 cm. En avril 1980, on a lancé un programme de repeuplement de certains cours d'eau du réseau avec du frai. Les oeufs utilisés sont prélevés sur des adultes pris au piège. L'auteur donne des détails sur les techniques utilisées, sur les taux de survie et de croissance du frais et sur l'efficacité du projet telle qu'on peut la juger à l'heure actuelle.


The number of salmon Salmo salar L. spawning in the upper River Tummel, a major branch of the River Tay in Scotland, has always been regarded as low, and in recent time (1953–81) an average of only 153 adults have migrated through a fish counter into Loch Tummel each year (see Fig. 1). The river from its headwaters on the River Ba down to its junction with the River Garry is 79 km long, including 48 km through lochs, and it is fed by numerous tributaries, which together drain an area of 820 km2. The low number of adults has been attributed at different times to various causes: to falls near the entrance to the river making access difficult, the several hydro-electric installations, predation of juveniles by pike Esox lucius L. and trout Salmo trutta L., loss of direction by smolts migrating through the large lochs, poor spawning areas and inadequate food resources for fry.

During 1979 a survey of the river and its main tributaries was carried out to assess the potentialities for increasing smolt production and subsequent adult runs through a programme of fry planting, designed on the basis of research carried out in previous years (Egglishaw and Shackley, 1973, 1977, 1980). The survey included an examination of physical, chemical and biological characteristics of the river and details are given in a report entitled “The stock of salmon in the upper River Tummel”. Fry planting commenced in the spring of 1980 and growth and survival of the fish was monitored.

This paper summarizes the observations made during 1979, including some earlier work, and describes the current state of the smolt enhancement programme.


The arrangement of the lochs and streams in the drainage basin of the River Tummel is shown in Fig. 1. Streams to the west of Dunalastair flow over much harder and less soluble rocks (granodiorite, psammitic granulite) than those to the east of Dunalastair (black schist and limestone out-crops) and their alkalinity (ppm as CaCO3) expresses this difference (range 4.4–18.4 in summer and 0.2–4.6 in high water during the winter for streams in the west, compared with 28.0–60.0 in the summer and 8.4–28.0 in winter for streams in the east). Most of the land in the valley is moorland, usually of heather, but with rough grassland and bog near streams and rivers. Large areas of the moorland have been planted with conifers, and semi-natural woodlands, both coniferous and deciduous, also occur. There is little arable farming. The total population of the valley is about 700. Tourists are attracted to the area by the scenery and for recreational activities.

Since 1927 there have been several dams, weirs, diversion tunnels and aquaducts constructed for hydro-electric generation. The developments resulted in the formation of new reservoirs and the extension of natural lochs flooded spawning areas, as well as created habitats suitable for pike and trout. Fish passes have been provided at five barriers (Fig. 1) and these offer possibilities for monitoring upstream movement of adult salmon.

The salmon smolt populations of the upper River Tummel have been the subject of a number of investigations since 1959. The principal migration of smolts through the fish ladder at the east end of Loch Tummel takes place from March to early June. The majority of smolts are two or three years old and between 10 and 15 cm long. During the period 1967–73, 16 424 were tagged and, of these, 353 were later recaptured as adults. Most were caught off West Greenland (42 percent) and in the River Tay (53 percent). The fish caught in the River Tay had spent one to three winters at sea.

Trout were present in all 17 streams surveyed in 1979 and salmon were present in four. Since two of the latter (Kynachan, Kinardochy) are stocked by the Tay District Salmon Fisheries Board, a proportion of the fish found could have resulted from artificial planting. The salmon in the two other streams (Frenich, Inverhadden) must be of natural origin. No salmon were found in the streams running into Loch Rannoch or in tributary streams further west. The densities of 1+ age class salmon in the four streams in which they were present ranged from 18 to 35 per 100 m2, which is similar to densities found in the highly productive Shelligan Burn at the same time of year in the period 1966 to 1979 (i.e., 11–41 per 100 m2) (Egglishaw and Shackley, 1977). In sections of the River Tummel which have a low, stable flow, high densities of salmon were found (53 and 61 0+ fish per 100 m2) but sections which fluctuate in flow depending on electricity generation requirements had lower densities (3 and 8 0+ fish per 100 m2). Only low numbers of salmon were found in the River Gaur and none in the River Ba.

The survey indicated that the production of salmon smolts in the upper Tummel Valley system is much lower than its total production potential. In 1975–77 and 1980 and 1981 and estimated 10 000–15 000 migrating smolts reached Clunie Dam (Struthers, 1978, 1978a). The total production potential of the upper River Tummel was estimated to be 165 000 smolts per year (Egglishaw et al., 1981).


3.1 Fry planting

The brood stock used to supply the eggs was collected from the River Tummel. Fish were trapped and held until they were ready for stripping (Struthers, 1982). Eggs were counted and kept in a hatchery until required. The unfed fry, with remains of the yolk sac still visible, were planted out as described by Egglishaw et al. (1982). Details of the streams stocked, areas and planting densities, with dates in 1980 and 1981 are given in Table 1. In 1980, 67 136 of the fry were planted at a density of 2.0–2.1/m2 evenly distributed in three of the more acid streams draining the granodiorite around Loch Eigheach. The remainder were planted at densities of 3.5–6.2/m2 in the streams draining mainly from the slightly richer psammitic granulite into Loch Rannoch. In 1981 planting densities in all streams were 4.6–5.3/m2.

3.2 Fish population and growth estimates

At the end of each growing season the numbers of salmon (and trout) present in sections of an area stocked and in sections downstream were estimated by isolating them with stopnets and catching the fish enclosed with electric stunning apparatus (Egglishaw, 1967, 1970). Each estimate (Leslie/DeLury method) was based on four fishings (units of effort) during which between 75 and 95 percent of the fish present were caught in the first two fishings. The lengths of all fish were measured to the nearest millimetre from the tip of the jaws to the fork of the tail. Weights of fish were read from curves relating known fresh weights and lengths. The numbers of survivors at the end of the growing season were estimated. For this purpose the number of salmon in a section not fished was determined from the product of the area of the section and the mean density of salmon in adjoining upstream and downstream sections (Egglishaw and Shackley, 1977). Survival rates to the time of fishing were determined. As the period between planting and electro-fishing at the end of the growing season varied, it is more appropriate to compare instantaneous mortality rates, M. These were calculated according to the formula:

M = (logeN1 - logeN2) / (t2 - t1)

where N1 is the number of fry stocked at the beginning of the period at t1 days and N2 is the number of fish present at electrofishing at the end of the period at t2 days.


4.1 Survival of the salmon

The percentages of fish surviving from the plantings to the time of electrofishing at the end of the growing seasons are given in Table 2. Values of the instantaneous mortality rates, M per day × 100 are also shown. The instantaneous mortality rate of 0+ salmon in the Shelligan Burn remained constant until November (Egglishaw and Shackley, 1977). The values of 0.68 and 0.98 for the populations in Duibhe and in Camghouran, respectively, in 1980 are very low mortality rates for salmon in their first year, and probably can be attributed to the low stocking densities (2.0 and 3.5/m2) and the low trout population (2.44 and 1.01 g/m2 trout of all age-classes, repsectively). Values of M per day × 100 obtained during planting experiments on the Fender Burn (a highly productive tributary of the River Tilt flowing through areas of limestone and black schist) ranged from 0.81 to 1.24 (Egglishaw and Shackley, 1980) and were related to the stocking densities (3 to 30/m2) of the fry. Values of M × 100 for the natural population of salmon in the Shelligan Burn (River Almond) ranged between 0.84 and 1.81 for six years (1966 to 1971). In this stream, too, these rates of loss of salmon in their first growing season were related to their population densities (range 2 to 12 on 1 July) (Egglishaw and Shackley, 1977).

The mortality rates of the 0+ salmon in the different streams in the upper River Tummel in 1981 were remarkably similar, M per day × 100 for five of the seven values ranged between 1.28 and 1.41. The stocking density for all streams was about 5/m2. The presence of 1+ salmon in five of the stocked areas did not seem to increasse mortality, when compared with areas without 1+ salmon (Camghouran upper, Craiganour).

Of 121 128 fry planted in five streams in 1980, 13 516 (11.2 percent) were surviving in the streams as large parr at the end of their second growing season in September 1981. As they were dispersed right down to the mouth of three streams (Camghouran, Aulich, Dall) it is possible that others had entered Loch Rannoch and the estimates given in Table 2 could be minimum rates of survival. Mortality rates (M per day × 100) between the mid dates of electrofishing in 1980 and 1981 (346 to 388 days for different streams) ranged from 0.14 in Dall Burn to 0.39 in Camghouran lower. These are low mortality rates for salmon in their second year and are comparable to rates of usually 0.28 to 0.40 between successive autumn fishings for salmon in the experimental plantings in the Fender Burn. After the first growing season, mortality rates of the natural population of salmon in the Shelligan Burn were, during 1 December to 31 March, 0.51 ± 0.24 (mean and standard error of six years), during 1 April to 31 May, 1.09 ± 0.54 and during 1 June to 31 October 0.50 ± 0.19 (Egglishaw and Shackley, 1977).

The presence of large numbers of 1+ age class salmon in the streams in 1981 did not appear to have affected the numbers and biomass of trout.

4.2 Growth of the 0+ age class

The mean weights of the 0+ salmon at the end of the growing seasons in 1980 and 1981 are given in Table 3. This table also includes the daily growth rates (G) calculated according to the formula:

G = (logeW2 - logeW1)/t2 - t1

where W1 is the mean weight of fry at planting, time t1 days, and W2 is the mean weight at the end of the season, time t2 days.

The highest mean rate, of 2.60 percent per day, occurred in the Duibhe Bheag and generally the rates were above 2 percent per day, with rates being lower in 1981 than 1980. The low mean rate of 1.37 percent per day for fish in Camghouran upper was presumably owing to that site being at a much higher altitude (Table 1) and therefore colder, than the others. Temperature was found to be the most important factor determining growth of salmon in the Shelligan Burn (Egglishaw and Shackley, 1977). In all streams of 1981 trout of 0+ age-class grew slower, and attained smaller mean sizes by the end of the growing season, than they did in 1980.

Samples of 0+ salmon were caught in all streams during the spawning season to determine monthly growth rates in weight. Rates declined from very high levels during May (G per month 90– 210 percent in 1980; 84 to 111 percent in 1981 with Camghouran upper 58 percent) to lower ones during August (22–47 percent) and, throughout the growing season, were generally higher in 1980 than in 1981.

4.3 Growth of the 1+ age class

The salmon planted in 1980 attained mean weights of between 9.45 g and 10.59 g in the five streams stocked. Approximately 11 400 of the 13 516 1+ age class salmon estimated to be present in the streams at the end of the growing season were larger than 90 mm. A high proportion of the survivors from these are likely to migrate downstream as smolts during the spring of 1982. About 7 800 will move out of Loch Eigheach catchment (Table 1) and will have to travel 28 km to reach Dunalastair smolt trap. The remainder, from Loch Rannoch catchment, will move between 8 and 17 km to Dunalastair. A total 484 1+ age class salmon longer than 85 mm were tattooed with a fluorescent mark to indicate the stream of origin.


Egglishaw, H.J., 1967 The food, growth and population structure of salmon and trout in two streams in the Scottish Highlands. Freshwat.Salm.Fish.Res., 38:1–32

Egglishaw, H.J., 1970 Production of salmon and trout in a stream in Scotland. J.Fish Biol., 2:117–36

Egglishaw, H.J. et al., 1981 The stock of salmon in the upper River Tummel. Working Pap.Freshwat.Fish.Lab.Pitlochry, (81/8)

Egglishaw, H.J., 1982 Principles and practice of stocking streams with salmon eggs and fry. Pitlochry, Freshwater Fisheries Laboratory (mimeo)

Egglishaw, H.J. and P.E. Shackley, 1973 An experiment on faster growth of salmon Salmo salar (L.) in a Scottish stream. J.Fish Biol., 5:197–204

Egglishaw, H.J., 1977 Growth, survival and production of juvenile salmon and trout in a Scottish stream, 1966–75. J.Fish Biol., 11:647–72

Egglishaw, H.J., 1980 Survival and growth of salmon, Salmo salar (L.), planted in a Scottish stream. J.Fish Biol., 16:565–84

Struthers, G., 1978 Investigations on smolt migration in the upper River Tummel in 1976. Pitlochry, Freshwater Fisheries Laboratory (24.FW.78) (mimeo)

Struthers, G., 1978a Investigations on smolt migration in the upper River Tummel in 1977. Pitlochry, Freshwater Fisheries Laboratory (25.FW.78) (mimeo)

Struthers, G., 1982 Cage impoundment of Atlantic salmon broodstock. Pitlochry, Freshwater Fisheries Laboratory (mimeo)

Table 1 The numbers of unfed salmon fry planted in tributary streams of the upper River Tummel in 1980 and 1981

ExtentAltitudeDateNo. of fryDensityExtentDateNo. of fryDensity No./m2
Loch Eigheach
Eigheach3503 092265–28025/46 5752.100---
Duibhe1 10015 143270–28019/530 7772.01 00013 8831 – 8/572 4485.2
Duibhe Bheag1 40014 750300–31515/529 7842.01 30012 45228/4–6/565 5425.3
Loch Rannoch
Camghouran upper 0375–415---1 4009 50324/4–4/549 0495.2
Camghouran lower6005 723206–2301/520 0273.55004 95823/4–4/522 8624.6
Dall4002 846218–2298/521 6306.24002 8461–4/515 0795.3
Aulich7504 875215–23512/522 9104.77004 17329/4–5/520 8885.0
Craiganour 0245–285---8003 89923/4–5/520 1705.2

Table 2 The numbers of 0+ and 1+ age class salmon surviving on the dates shown from the 1980 and 1981 plantings of the tributary streams of the upper River Tummel. The instantaneous mortality rates are given as M per day × 100. Eigheach was not sampled in 1981. Data on numbers and total biomass of trout are included.

StreamDateNo. of 0+% survivingM×100No. of 1+% survivingM×100No./100 m2g/m2 
Eigheach29–30/91 24819.01.060  8611.30
Duibhe25–27/815 71651.10.680  4422.44
Duibhe Bheag13–17/104 46315.01.250  9332.35
Camghouran lower18–24/94 91724.60.980  15201.01
Dall28–29/83 77021.41.380  201084.50
Aulich10–15/92 82812.31.700  26922.21
Duibhe21/8–7/913 80619.11.357 12745.30.214342.10
Duibhe Bheag8/9–14/107 00010.71.411 79040.00.253231.00
Camghouran lower12–15/102 28510.01.371 09922.40.397310.97
Camghouran upper20–23/108 67917.70.990  3311.08
Dall18–20/83 71924.71.282 31361.40.14321464.64
Aulich21–26/85 65427.11.161 18742.00.25631644.30
Craiganour10–16/93 06615.21.380  143566.85

Table 3 Growth rate in weight (mean G per day) of 0+ salmon between planting and the end of the growing season. Weight at planting was 0.13 g and fish are assumed not to grow in weight after 15 September

StreamWeight at end of seasonTimeG×100Weight at end of seasonTimeG×100
(g)(days) (g)(days) 
Duibhe Bheag3.111222.602.031362.02
Camghouran upper   0.871391.37
Camghouran lower2.271372.091.431391.73
Craiganour   1.431391.73
Fig. 1

Fig. 1 The distribution of juvenile salmon in the upper River Tummel and its tributaries in 1979. About 71 ha of stream bed is accessible to the average of 153 adult salmon which enter the upper River Tummel each year. Juvenile salmon were found in only 25 ha. Areas which have been planted with fry are boxed.

Previous Page Top of Page Next Page