The Scottish Naturalist

Founded 1871

A Journal of Scottish Natural History

Editorial Committee:

J.A. Gibson
John Hamilton
John C. Smyth
A. Rodger Waterston
Reproduced with the permission of
THE SCOTTISH NATURAL HISTORY LIBRARY
Foremount House Kilbarchan, Renfrewshire PA10 2EZ
Copyright: May be used for private research only

Due to the size of this text file, the many figures and graphics are individually available
Simply click the underlined links.

The Scottish Naturalist, now published by the Scottish Natural History Library, is an independent journal primarily devoted to the study of Scottish natural history. It was founded in 1871 by Dr. F. Buchanan White, of Perthshire, and in 1988 completed one hundred years of publication. For a summary of the record of publication, see the inside back cover.

Although the journal's main interests have always centred on the history and distribution of Scottish fauna and flora, it is prepared to publish contributions on the many aspects of Scottish natural science embraced by its title, including Zoology, Botany, Geology, History, Geography, Archaeology, and the Environment.

All papers and notes for publication, or books for review, should be sent to the Editors at the Scottish Natural History Library, Foremount House, Kilbarchan, Renfrewshire PA10 2EZ.

Contributions should be clearly written; whenever possible they should be typed, double‑spaced, on one side of the paper, with adequate margins, and should try to conform to the general style and arrangement of papers and notes in the current number of the journal. Maps, diagrams and graphs should be drawn in black ink on white unlined paper. Photographs should be on glossy paper. Proofs of all contributions will be sent to authors and should be returned without delay.

Authors of papers, but not of short notes, will receive thirty reprints in covers free of charge. Additional reprints may be ordered, at cost, when the proofs are returned.

The Scottish Naturalist is usually published three times a year. The standard annual subscription is £25.00, which should be sent to the Editors at the Library address. Members of recognised natural history organisations, however, can receive the Scottish Naturalist at a greatly reduced subscription; for details apply to the Editors.

THE SCOTTISH NATURALIST
Founded 1871

A Journal of Scottish Natural History
With which is incorporated The Annals of Scottish Natural History
and The Western Naturalist

Record of Publication

The Scottish Naturalist and Journal of the Perthshire
Society of Natural Science 1871

The Scottish Naturalist 1872-1891

The Annals of Scottish Natural History 1892-1911

The Scottish Naturalist 191-1939, 1948-1957, 1961-1964

The Western Naturalist 1972-1982

The Scottish Naturalist 1983-date

Published by The Scottish Natural History Library

LOCH  NESS HABITATS OBSERVED BY SONAR

AND UNDERWATER TELEVISION

By

ADRIAN J. SHINE and DAVID S. MARTIN

Loch Ness and Morar Project

International Society of Cryptozoology

Society for the History of Natural History

Symposium on

The Loch Ness Monster

Royal Museum of Scotland, Chambers Street, Edinburgh

25th July 1987

Reprinted from The Scottish Naturalist

1988, pages 111-199

LOCH NESS HABITATS OBSERVED BY SONAR AND UNDERWATER

TELEVISION

By ADRIAN J. SHINE and DAVID S. MARTIN

Loch Ness and Morar Project

Contents

Part A: LOCH NESS

1. MORPHOMETRY

Basin Form

Loch, Ness is the greatest body of fresh water in the British Isles, having a volume (74.52m³ x 10⁸) in excess of all the lakes and reservoirs of England and Wales combined (Smith and Lyle, 1979). It is a glaciated tectonic lake extending for 35km within the north eastern section of the Great Glen. The remarkably regular basin has a mean width of about 1.5km, with steep sides sloping to a flat bed, interrupted only by a rise opposite the Foyers River, which divides the two deep basins of 220m depth. The maximum depth of 230m is considerably deeper than the seas around our shores, and places Loch Ness second only to Loch Morar (310m) among British lakes. Fault line origins, however, give Loch Ness the greatest mean depth, of 132m. The catchment area is 1,775km², mostly of hard rock yielding few nutrients.  The loch is drained by the River Ness to the north. Figure 1a (8K) maps the main river inputs.

The above depths were determined by Sir John Murray's Bathymetrical Survey of the Scottish Fresh‑Water Lochs (Murray and Pullar, 1903-08, 1908a and 1910) using a wire sounding machine, but they were disputed by a sonar depth of 297m recorded by Vickers in 1969 (Eastaugh, 1970). During the 1980s, careful searches of the reported area, a quarter mile south of Urquhart Castle, were made by the Loch Ness and Morar Project using a Kelvin Hughes M.S.48 hydrographic echo‑sounder, calibrated on site in accordance with survey practice, but no depths much in excess of 220m were found.

Figure 1b (18K) shows a characteristic echo profile of the loch and contrasts its trench-like profile with that of Loch Morar. Time advances left to right on the record and it is important to note the compression of vertical scale induced by boat speed. The profile illustrated in Figure 2a (19K) results from a faster speed and also serves to illustrate 'side echoes', which in the past have been interpreted as extensions of the side walls continuing down beneath loose sediment for hundreds of metres. As the conical sonar beam profiles 'down' a steep slope it registers not vertical depth but the range of the nearest portion of slope, which is to the side and therefore shallower than true depth. As the loch bed begins to register and the boat draws clear of the wall, the beam's outer edges and side lobes still return echoes from the side, which now appear at a greater depth, thus giving rise to the above impression. A sonar frequency of about 10kHz or less is required for true sediment penetration. Underwater television pictures in Figure 2b (15K) of the steep rocky walls, contrast with the flat silt of the abyssal plain.

There is no foundation for speculations concerning ancient constructions, since there is no evidence that the loch level was ever much lower than it is today; nor of submerged caverns, subterranean connections and the like, since there is no limestone in the area.

Great advances have been made in hydrographic sonars. In September 1987 Simrad U.K. Ltd extended a collaboration while their demonstration vessel R/Y Simson Echo was in the loch. The vessel mounted a multibeam or 'swath' sounder (E.M. 100). Data is recorded continuously from 32 beams, which form a fan athwartships. This greatly reduces the number of survey lines necessary and hence the time required.

On 7th September 1987 a survey was made of Urquhart Bay. A graphic, constructed in Figure 3 (13K), serves to show the bay contours sloping away from the river mouths and then dropping steeply to the main basin floor. The peaks and spikes would be removed by software processing in a full survey and result from second time around returns, side echoes from steep slopes, and from an area of boat moorings. The plot contains over 200,000 data points.

History and Sediments

The three hundred million-year-old fault was glaciated by successive ice ages until approximately 12,000 years ago, when the lake we know today was formed. Meltwater raised the sea level until the land, relieved of the ice burden, rose to bring the loch to its present altitude of 16m. The Loch Ness and Morar Project has extracted sediment cores from Ness and Morar in an attempt to shed light upon the post‑glacial history.

Much of the material remains to be worked but so far, no evidence of a marine transgression has been found at Loch Ness. A core from Loch Morar (9m above sea level) was found by Dr. H.J.B. Birks (pers. comm.) to contain abundant cysts of marine algae, which suggests that the sea could have entered, although the spores could conceivably have been wind borne.

The deep basin floor of Loch Ness is covered by black lake sediment, of which a 4m core only just penetrated to clay, thus suggesting that the organic sediment is of at least that depth. This core has yet to be dated, but a 2m core from the basin floor of Loch Morar has been estimated by Dr. Birks (pers. comm.), on the basis of pollen content, to have penetrated between 5,500 and 6,500 years of post-glacial sediment. This sheds some light upon rates of deposition and, incidentally, upon the likely depth of sediment in Loch Ness. At the seaward end of Loch Ness the sediment consists of clays, an interesting feature of which is that a firm yellow‑brown layer approximately one metre thick overlies a blue‑grey material which is very fluid.

The organic content of the loch sediments has been established to increase with depth (Lee and Collet, 1908). From an analysis of our own profundal substrate samples (J.D. Hamilton, pers. comm.), the percentage composition of the organic matter ranges from 18.3% to 34%, similar to the Bathymetrical Survey results. Like this earlier study, it would also appear that the Urquhart Basin has less organic matter than the Invermoriston Basin, which may be due to the Foyers barrier retaining the organic material within the southern basin.

p118                              The Scottish Naturalist                                          1988

2. THERMAL STRUCTURE

Stratification

In a lake, the onset of stratification in summer consists of the separation of the upper less dense warm water, known as the epilimnion, from the deeper cold water, or hypolimnion, by a region of sharper temperature change, termed the thermocline. Photosynthesis is limited to the epilimnion, where depletion of nutrients cannot be replaced from the hypolimnion until mixing occurs in winter (Figure 4a, 7K).

In eutrophic (richer) lakes, the decay of organic matter descending to the hypolimnion deoxygenates it, to the detriment of deepwater life. In deep oligotrophic (poorer) lakes, such as Loch Ness, these profound effects do not follow, since nutrients are already low in the epilimnion and there is no deoxygenation of the vast hypolimnion, which remains over 80% oxygen saturated. The oligotrophic nature of Loch Ness therefore, offers the compensation of stability in exchange for low productivity. It is spared the seasonal booms and crashes of more productive waters, and a variety of life extends to its deepest regions.

The great body of water does not fall below a uniform 5C in winter, and consequently there is no inverse stratification and no freezing. Indeed the influence of the loch's stability extends to the area around the shoreline, where the release of heat ensures that snow seldom lies for long. On 10th January 1987, for example, the air temperature was 4.4C after a very cold spell of overnight temperatures down to -14C. The water temperature was 6.4C to a depth of at least 50m. Of course the loch is still relatively cold from a biological point of view and many of its inhabitants are relicts from glacial times.

Physical Studies

The biology of Loch Ness may have been neglected until recently, but physical studies have been particularly rewarding. The steepness of the 'walls', length, and regularity of the basin, together with the loch's orientation N.E-S.W. in line with the prevailing winds, all tend to simplify and amplify thermal effects. The absence of salinity complications has also facilitated observations in support of hypotheses afterwards extended to oceanography.

It was during early investigations at Loch Ness that the first internal temperature seiches were discovered (Watson, 1904; Wedderburn, 1907). These arise when wind, blowing warm surface water to the leeward end, tilts the isotherms lengthwise in this direction. The cessation of wind stress permits recovery of the isotherms but, because of momentum, they overshoot and oscillate for some days (Figures 4a, 4b, 17K and 6a, 6b. 6c, 6d, approx 12K each)). L.H. Mortimer (1955) of the Freshwater Biological Association showed the effect of the earth's rotation on the currents accompanying seiches, which leads to a deflection to the right and hence a cross-loch tilting of the isotherms (Figure 8 , 15K). Mortimer's observations of asymmetry in the seiche led Dr. S.A. Thorpe of the National Institute of Oceanography to establish the presence of a 'surge' wave front characteristic (Thorpe, Hall and Crofts, 1972). Thorpe (1977) has since conducted other work, including the mechanisms of mixing and turbulence.

Factors affecting Sonar

The Project considers it important to understand the nature of thermal characteristics likely to affect sonar work. For example, sonar beams propagating horizontally are refracted by temperature gradients and it is also important to recognize direct echoes caused by them. Conversely, we wish to discover the extent to which thermoclines can themselves be monitored by sonar, and their relationship to 'scattering layers'. Lastly, we are seeking to establish the role played by physical factors in the distribution of biomass. The thermocline movements may affect the range of vertical migrations, and internal seiches can cause horizontal transport.

Observations will first be made on the build‑up of stratification, and then the role of sonar discussed. From 19th June to 7th October 1983 temperature measurements were made from a mooring to the north of the deep basin (220m) off our station just south of Achnahannet (see map in Figure 1a). The readings were too infrequent and irregular to give a valid continuous record, and were complicated by our position near the node of the main longitudinal seiche, which doubles the frequency, and towards the northern side of the basin, where we were subject to the transverse seiche. An example of daily records (Figure 5a, 7K) reveals large changes in the depth and degree of stratification. Nevertheless, the averages of readings per week (Figure 5b, 6K) serve to show progressive deepening and convergence of the isotherms towards autumn.

Further background is provided by the record from the fixed station (see map in Figure 1a) employed in 1984. This was moored towards the south end of the loch, in the centre of the deep basin, about 4.5km north of Fort Augustus, thus simplifying interpretation, and readings were taken at least four times per day. The result is shown in Figures 6a, 6b, 6c, 6d. Winds were measured with a simple hand-held device (Ventimeter) and currents, where recorded, are merely observations of drag on the temperature probe wire. A detailed discussion of the record will not be attempted in this review but the following comments may be helpful.

The Record

The record begins on 22nd July (Figure 6a, 12K), as N.E. winds mix and deepen the 17C epilimnion, until 26th July when the wind changes to S.W. As the water is transported to the N.E., successive isotherms break surface, exposing colder layers which will mix into the warm water to leeward (Mortimer, 1952). The epilimnion begins to recover as the winds moderate on 28th July and with calmer weather, the upper isotherms generally deepen, although possibly checked by S.W. winds from 31st July to lst August, until they are forced down by strong N.E. winds (Figure 6b, 14K) from lst to 3rd August. As a result of mixing, the epilimnion is now cooler (14C) and deeper. Stratification becomes marked.

The disturbances generate a seiche in the lower isotherms with a period of approximately 54 hours characterized by a rise in temperature of 18 hours and a fall of 36 hours. The wind cycle appears to have various amplifying and dampening effects on the seiche until 13th August (Figure 6c, 11K) when calmer weather accompanies a loss of regularity and amplitude. From 15th August a consistent 15C isotherm re‑enters the record and an epilimnion is ultimately deepened by N.E. winds, from 23rd to 24th August, to bring about a situation similar to that at the beginning (Figure 6d, 10K).

The process continues throughout the summer as calms produce transitory shallow epilimnia which are soon mixed in by winds. Mass transports caused by seiches spread the heat deeper by turbulence. The most effective mixing is accomplished by the autumnal equinoctial gales, which since the loch is no longer gaining heat, are able to produce an epilimnion of near uniform temperature. It is from this simple situation that our description of the sonar contribution begins.

Figure 7 (19K) shows one of the 252 sonar profiles made with Simrad EY-M (narrow beam 11 degrees) and Skipper 603 (wide beam 33 ^degrees) echo-sounders off the Achnahannet station in 1983. The record for 29th September has an epilimnion at an almost uniform 11.3C with the thermocline at 40m. Two temperature graphs serve to show the depth and degree of stratification to be related to the echo trace. A striking illustration of transverse tilt is provided by Figure 8, (15K), where the isotherms conform to the tilt registered on the trace, although it would appear that three layers are involved.

An example of mixing occurs during a short period of high winds from 2nd to 4th August (Figure 9a, 59K). It is not clear whether a seiche was in progress (although the trace shows tilts suggestive of currents), but the temperature graphs before and after the episode show a cooling and deepening of the epilimnion. Prior to the autumn situation the gradients are complex, and another example (Figure 9b, 67K) shows the destruction, or at least removal, of a secondary thermocline at about 10m by S.W. winds from 28th to 30th August.

During the fixed station experiment of 1984, a Lowrance Eagle Mach 2 was run every hour for at least five minutes from 31st July to 24th August and in addition, almost always when temperature probes were being taken. In Figures 6e (28K) and 6f (31K) examples of the records are attached beneath their relevant positions and often show the thermocline. From the foregoing it might seem that sonar is quite reliable in resolving thermal structures but a considerable degree of caution is required in use and interpretation.

Possible Errors

Monitoring of open water 'scattering' layers as early as 1982 had highlighted the problems of 'second time around' returns, which can confuse or obscure the record. With range set to less than water depth, echoes will return from the lake bed and be received during the 'listening' phase of subsequent pulses. If the set range is sufficient to include the lake bed, second time around echoes can still be received, having been inter‑reflected between the bed and water surface. They appear on the chart as diffuse layers, and are accentuated by the high gain settings necessary for thermocline detection. These returns are indicated on some of the charts, and in practice limit the choices of effective range settings. Such returns can be identified since they change range with alterations in pulse repetition frequency, i.e. range settings.

Another cause for confusion lies in the frequent association of biological 'scatterers', such as fish and plankton (Schroder, 1962), with the thermocline, at least by day. Detritus has also been suggested, but we have never found detectable traces in samples.

In Figure 10 (31K), a comparison of simultaneous runs with wide and narrow beam sounders clearly shows the contribution of fish to the wider beamed sounder's return. In Figures 6e and 6f, during the fixed station work, many of the daytime returns from depths of around 30m are strengthened by the layer of fish. For further discussion see sections on plankton and fish.

p126                              The Scottish Naturalist                                          1988

The records sometimes show inconsistencies between the degree of stratification and the strength of echo trace. It is particularly noticeable that, on the narrow‑beam Simrad EY‑M, thermal gradient traces often appear patchy. An examination of Figure 8, (15K) for the southern profile shows discontinuities and traces to be reasonably consistent, but to the north, strong traces register from lesser gradients. This may be partly accounted for due to the thermocline being divided into 'layers' of weak gradient, several metres thick, divided by much thinner 'sheets' with steep gradients (Woods, 1968; Simpson, 1970). Our one‑metre temperature sampling interval would not resolve microstructure adequately. We believe, however, that the explanation lies in turbulence, to which echo‑sounders are particularly sensitive, as shown by traces throughout the paper recording the raising and lowering of samplers and illustrated vividly in Figures 9a and 9b showing mixing. In the above example the transverse tilts, by definition, imply strong currents moving in opposite directions. Echoes are often pronounced near the side walls where greater turbulence is to be expected.

Shear

Shear is actually associated with most thermoclines and waves within them sporadically produce inversions known as Kelvin-Helmholtz billows (Woods, 1968; Thorpe, 1974). We believe this turbulence to be the main factor in regulating the strength of echo returns, possibly by creating multiple interfaces of small size but high gradient. In support of this, attention is drawn to the 1984 fixed station record in Figure 6f (31K) for the period 3rd to 7th August, which includes the sharpest thermocline observed. The records for the night of 3rd August show the gradient strongly after the fish have left it and while the isotherms are rising. However, by the time of the temperature profile at 10.00 hrs the interface echo was weak or undetectable despite the intensity of stratification. It is at this point that the isotherms are at rest and the currents therefore at their slackest. A conclusive thermocline trace does not re‑enter the record until 5th August at 22.00 hrs, when the isotherms are once again in significant motion.

The foregoing conclusions seem particularly relevant to our observation in October 1985 of a surge front wave of the type described by Thorpe. These waves (Figure 11a, 12K) lead the seiche and have a length of approximately one kilometre and a speed, depending on amplitude, of approximately 1.5 km/hr. Figures 11b, 11c, 11d, 11e, 11f, show a continuous echo trace from 18.55 hrs on 11th October to 14.44hrs on 12th October, superimposed with some relevant temperature graphs and  current observations, all taken from a moored position close to that of the 1984 fixed station. S.W. winds on the previous night had reached hurricane strength over parts of Scotland but had calmed to a gentle breeze by the time we took station. The record commences (Figure 11b, 115K) with a shallow 12-15 metre epilimnion of a uniform 10C. At 23.15 hrs (Figure 11c, 76K) the thermocline dipped sharply and a spectacular wave train of decreasing regularity was observed. Temperature probes (Figure 11d, 93K) show a rapid lowering of the isotherms with discontinuities corresponding to the echo trace, and at 01.15 hrs an increasing subsurface current to the S.W. set against the surface drift. By 02.28 hrs the surface water was also moving against the wind and by 08.00 hrs (Figure 11e, 75K) was rapid enough to cause difficulty with temperature probes and plankton sampling, because the wires assumed unacceptable angles. At 13.07 hrs (Figure 11f, 86K) the current was easing, and at 14.30 hrs was slack. By this time the S.W. wind had freshened and the surface drift had reverted to N.E.

Obviously the entire event was attended by a high degree of shear as the epilimnion slid beneath our station and it will be noted that the surge front wave crest registers particularly strongly. We speculate that the crest comprises turbulence similar to that observed by Woods (1968) on thermocline sheet waves, although of a greater amplitude. The two initial waves are of 23m and 45m amplitude. The trace from 06.30 hrs to 08.15 hrs (Figure 11e), when the strongest currents were observed, is interpreted as consisting of Kelvin-Helmholtz billows (Thorpe, 1988). The cause of the near surface turbulent effects seen from about 08.15 hrs is less obvious but coincides with the influx of slightly warmer surface water and is probably also associated with shear.

The plankton hauls (Figure 11g, 10K) show a mass influx at 02.08 hrs, and it is evident that fish are also being transported, sometimes in very localized concentrations (see sections on zooplankton and fish).

Conclusion

In conclusion, it may be said that echo‑sounders have considerable value in the detection of general thermal conditions, but do so by registering degrees of turbulence rather than gradient, even though the two are often associated. Future experiments will include induced artificial turbulence, which may then propagate across potentially unstable structures, thus revealing their positions (see Figure 12, 12K).

3. BIOLOGY

A measure of the neglect shown towards Loch Ness may be judged from the fact that the first basic observations to be made there since the Bathymetrical Survey in the early years of this century were made over fifty years later, by teams investigating the 'Monster' controversy. Student expeditions from Birmingham and Cambridge Universities made plankton hauls, while Mackal and Love (1970) of the Loch Ness Investigation Bureau took water samples (Figure 13a).

The situation changed, however, when from 1977 to 1980 a comparative study of five major lochs, including Ness and Morar, was undertaken by the Institute of Terrestrial Ecology. Wide‑ranging papers were edited by Maitland (1981) in the book The Ecology of Scotland's Largest Lochs. The Project's general biological objectives have therefore concentrated upon areas not covered by the comparative study, such as plankton and fish migrations. A special study has been made of the abyssal fauna.

Phytoplankton

Primary productivity is low. The high latitude and frequent cloud reduce sunlight to a short growing season. Photosynthesis is further limited, by suspended peat, to a shallow photic zone. Rooted plants are restricted to a depth of about 6m around the shoreline. The hard rocks of a steep catchment yield few nutrients to fast flowing rivers and streams entering the loch. Acidity will tend to slow bacterial decay of organic particles and hence the release of their nutrients (Figure 13a, 12K).

Thus the phytoplankton crop is low, and was found by the Cambridge expedition to have a diatom population of 200/1 as compared to 5,000/1 for Lake Windermere (Baker, 1962). Bailey-Watts and Duncan (1981) found the plankton in Loch Ness to be dominated by chrysoflagellates, with a total cell count of 568/1 at the time of the chlorophyll 'a' peak in August. Some of the species are shown in Figure 13b (19K).

Zooplankton

The cladoceran grazers of the phytoplankton consist of Diaphanosoma brachyurum, Holopedium gibberum, Daphnia hyalina and Bosmina coregoni. The larger predators are Polyphemus pediculus, Bythotrephes longimanus and Leptodora kindti. Copepods are represented by Diaptomus gracilis, the most numerous species, and Cyclops strenuus abyssorum. See Figure 13c (16K).

Zooplankton of most open waters appear adapted to avoid predation by transparency, and at Lochs Ness and Morar are also adapted to low productivity, by remaining relatively small and by producing fewer but larger eggs than their counterparts in more productive waters. Food seems to be stored in some copepod specimens as oil globules. Peak numbers have been found in October (Maitland, 1981: 144). Project work was conducted in collaboration with Dr. A. Duncan of Royal Holloway College in September 1983, with the aim of determining vertical and horizontal distribution. A Clarke‑Bumpus collector was used to take horizontal hauls, which have yet to be worked, and a 31‑litre Patalas sampler was used for the vertical work.

The vertical hauls are shown in Figures 14a (49K) and 14b (40K) with a summary in 14c (16K). From 12.00 hrs on 12th September 1983 to 05.00 hrs the following morning, five hauls were made, in conjunction with temperature profiles and echo‑soundings, from the mooring off the Achnahannet station. The results show a clear concentration of plankton at the thermocline at 13.00 hrs which involved most species. Thereafter, a migration takes place towards the surface, particularly in the case of Cyclops and Diaptomus. Total numbers increase towards midnight, possibly due to horizontal transport in response to the rising of the isotherms. Some of the peak densities are noted and have no convincing connection to the echo traces, which, although complicated by second time around returns and reflections from the mooring, appear to bear more relationship to the temperature profile.

There has been considerable research on acoustic scattering by marine zooplankton. Theoretically even a 200kHz echo‑sounder of high sensitivity could detect layers of organisms as small as one millimetre, and euphausiids of 15‑22 mm have been assessed by multifrequency methods (Greenlaw, 1979). In fresh water, where plankton sizes are generally much smaller, convincing records have been made at 200kHz of Chaoborus larvae in the absence of thermoclines (Northcote, 1964). However, these insect members of the zooplankton have a length of 9mm-12mm together with paired air sacs at either end, contributing to much higher scattering strength than from the more typical cladocera and copepods. Chaoborus is absent from Loch Ness, and the largest of the more abundant herbivores measure up to 2mm. 

Schroder (1962) reported the detection of one millimetre freshwater plankton at densities of as little as 2/litre, although the sonar frequency was not stated. Examination of the traces suggests that plankton horizons presented cannot be distinguished from the thermocline records. Indeed, the prime conclusion of the paper is that the "zooplankton in Lake

Constance is mostly found by the echo‑sounder in layers with sharp gradients (thermoclines) where it is also to be found during vertical migration". At our frequency of 50kHz (33 ^degree beam) no correlations were evident at concentrations of 10/litre. The frequent concentration of zooplankton in thermoclines must also inevitably associate them with the temporal and localized shear to which echo-sounders are so sensitive. It would seem difficult to design thresholds for echo integration of zooplankton volume scattering where thermal gradients exist.

An example of horizontal transport appears to occur in the strong currents noted during the surge shown in Figure 11d (93K). Four sets of net samples were made, all but the first to a depth of 36m in 6m stages. A closing net of 28cm diameter was used. The results indicate a mass influx of plankton as the currents strengthened.

To conclude, we have observed zooplankton in Loch Ness concentrated at the thermocline by day and exhibiting a vertical migration at night. It may be of interest that this migration has not been observed in Loch Morar, where the water is much clearer but where there are far fewer midwater fish (Figure 14c, 16K); here the plankton remain in the top 15m of water by day. In Loch Ness considerable horizontal transport of plankton occurs, particularly during a seiche. We are not convinced that, at the concentrations observed, our echo‑sounders have yet made a contribution to recording zooplankton (50kHz‑200kHz) and draw attention to possible confusion owing to the presence of turbulence. Layers of small fish may also give misleading (diffuse) returns, particularly at depth and on wide‑beamed sounders. Diagnosis is possible if a vertical migration takes place (see Figure 20a, 13K).

Littoral and Sublittoral Benthos

The steep and stony shores of Loch Ness provide a restricted habitat with considerable exposure to wave action (Figure15a, (7K). The fauna is therefore similar in some respects to that of fast‑flowing streams, in that the organisms must retain their positions among stones (Figure 15b, 11K), rather than on the silts, muds and plant life of more lentic waters.

A comprehensive survey of the animals dwelling within the first 50cm of water, conducted by the Institute of Terrestrial Ecology (Maitland, 1981) confirms that the community is dominated by insect nymphs. Stonefly (Plecoptera) make up 30% and mayfly (Ephemeroptera) 18% of the fauna, some of which are adapted to retain position by the possession of grasping claws or present a low profile to the water by a dorso‑ventral flattening of their bodies. A common mayfly, Ameletus inopinatus, is generally confined to streams above 300m or to lochs much further north, so its presence here suggests a preference for the lower temperatures found in deep, windy lakes, even at the surface, owing to regulating effects of mixing.

Other benthos include triclads, various nematode and oligochaete worms, the gastropod Lymnaea peregra, the crustaceans Asellus spp. (found in more sheltered areas) and ostracods, water mites, and caddis (Trichoptera) and chironomid larvae (Figure 15c, 25k). Typical Project hauls at Achnahannet and the Horseshoe Scree ‑ additional to Maitland's list ‑ are included in Figure 15d (10K).

The steeply sloping walls of Loch Ness provide a narrow and ill‑defined sublittoral zone, but hauls from 20m‑30m depths at Urquhart Bay as well as Borlum Bay, in sandy sediments, have revealed the organisms listed in Figure 15e (10K). These littoral and sublittoral communities, together with terrestrial insects falling into the water, provide the food for the inshore fish populations.

Profundal Fauna

In contrast to the turbulence and variety of physical conditions among the stones of the shoreline, the fine and relatively rich silts of the abyssal regions offer remarkable stability. In an environment of great hydrostatic pressure, constant darkness, and a scarcely changing low temperature of 5.6C, high oxygen levels (over 80% saturation), permit surprising variety in the profundal community of the 200m deep basin floors.

The first samples were taken during the Bathymetrical Survey (Murray, 1904: 442). It would appear however, that no further collections have been made until the present work. For a description of the sediments and a literature review of the benthos see Maitland (1981: 205). The Project has made a particular study of the profundal fauna at both Ness and Morar and a few comments are appropriate here.

A variety of qualitative and quantitative collection techniques have been used, including dredges, grabs and a static 'colonization experiment'. At Loch Ness, over thirty species have been recorded, some of which are doubtless casual occurrences, with an average density of 295 individuals per sq m. Recognized characteristic profundal fauna is present, such as oligochaete worms, chironomid larvae (non‑biting midges) and Pisidium spp. (pea mussels) but, numerically, ostracods predominate, comprising 62.6% of the community. The dominant bivalve mollusc, Pisidium conventus, is more normally found in arctic streams, but at our latitudes is confined to the cold water of deep lakes or to high altitudes. It is considered to be an ice-age relict species in Loch Lomond (Hunter and Slack, 1958). The dominant chironomid, Sergentia coracina has been similarly described in southern Sweden (Brundin, 1949). The community also has its predators, such as chironomids of Procladius spp. and the large copepod Acanthocyclops viridis, which are more cosmopolitan in distribution although apparently absent from the littoral zone of Loch Ness.

Some interesting occasional records include caseless caddis and the flatworm Phagocata woodworthi, which has been recently discovered in the littoral near a sewage outfall (Reynoldson, Smith and Maitland, 1981). This North American triclad is speculated to have been introduced on equipment imported in the search for the Loch Ness Monster. Another native of North America, the amphipod Crangonyx pseudogracilis, has been spreading northwards since its discovery in the London area (Crawford, 1937), and its occurrence during the present work is the first record for a Scottish loch. It could well have been responsible for small fast moving images first seen on underwater television in 1981. It may be significant that Crangonyx was captured only in the passive colonization substrate. Fish, again first noted during the television work of 1981, have now been netted and identified as Charr Salvelinus alpinus. Figure 16a (24K) shows some of the members of the abyssal fauna.

The role of the echo‑sounder during this work was limited but aside from enabling the quick location of the depths to be sampled, sonar can often aid the control of sampling equipment. In particular, it was noted that the Ekman grab penetrated too deeply to enclose the vital top 5cm of sediment. With the echo‑sounder, sampling was quicker and more effective since the sampler could be allowed to free fall to within a few metres of the bed and then be gently lowered. Figure 16b (10K), shows a grab being deployed, and also a method for the recovery of long term experiments, in this case our colonization substrate sampler, by the location and grappling of a buoyant rope laid horizontally and with the end weighted.

In general, it seems that the depths offer a refuge, not only for ice‑age relicts but also for a variety of widely occurring 'pond life' unsuited to the turbulence and predation pressures of the shoreline. It will now be necessary to examine the rocky side walls to define the limits of the two communities and to see if there is also a third.

Fish

As the ice retreated, the loch was colonized from the sea by coldwater salmonids, which entered fresh water only to spawn. Thus the Salmon Salmo salar still migrates via the River Ness, as does the Sea Trout Salmo trutta. However, with the passage of time and increasing sea temperatures another variety of Salmo trutta, the Brown Trout, and the Charr Salvelinus alpinus now spend their whole life cycles in the loch. It seems possible that new populations of Charr may have entered during the period of the Loch Lomond re-advance (Greer, pers. comm.).

Other species gaining access by sea migrations are the Three‑spined Stickleback Gasterosteus aculeatus, the Brook Lamprey Lampetra planeri and the Eel Anguilla anguilla, which has an opposite life cycle to the salmonids in that it matures in the loch after having been spawned in the Sargasso Sea.

The exclusively freshwater 'coarse fish', which have been spreading northwards since the ice age (Maitland, 1977), are represented only by the Pike Esox lucius and the Minnow Phoxinus phoxinus. The Minnow was not recorded by Maitland (1981); it was first observed by the Project in 1985 at Bona Narrows just downstream of the loch proper, but in June 1987 it was also recorded in sheltered water at Fort Augustus.

In collaboration with members of the Ness Fisheries Board, together with Dr. A. Duncan and Mr. R. Greer, various netting experiments have been carried out. Underwater television (UW/TV) has so far made contributions in the littoral and abyssal zones, while sonar has been most extensively used in the pelagic. Much work remains to be done on the ageing and electrophoresis of the specimens.

The littoral is dominated by brown trout to a depth of about 20m, below which charr prevail. The really shallow and sheltered water harbours minnows, three‑spined sticklebacks and salmonid parr. Eels live on the loch bed and range widely along the shoreline, being most concentrated off river mouths. Work on the growth rate of the Loch Ness eels has been conducted by Mackal and Frake (Mackal 1976: 319‑330).

Benthos is doubtless an important food source and many of the brown trout seem particularly specialized, e.g. a 21.7cm fish contained 143 snails (Lymnaea peregra). In summer, however, a substantial proportion of the trout diet consists of a terrestrial input of adult insects. Figure 17a (27K) presents some relevant information on the fish of the littoral. Figure 17b (16K) shows underwater television pictures of trout and eels in Urquhart Bay, concentrated in the rich organic leaf deposits at 30m off the mouth of the River Coiltie in early June 1987. As summer progresses. a considerable amount of gas generates in this material and also in other parts of the bay, possibly because of higher water temperatures. In October 1987 only an occasional Eel was observed here. Gas has not been detected in the fine deepwater sediments.

It is along the shoreline that the migrating salmon pass before spawning in rivers, and attention is drawn to the fact that salmon, as the largest fish present, owe very little to the loch's rather poor food chain. After only a few years as small parr they will then mature at sea, and return at weights up to 20kg. There is no fish counter on the River Ness, but two counters installed on the Rivers Garry and Moriston showed less than 900 fish in a peak year of 1975 (Maitland, 1981: 243). Although this does not include the rivers in Urquhart Bay, a comparison may be made with a maximum count of just over 3,000 fish entering Loch Morar in 1966. The picture is rather sad when one recalls the comments of Captain Burt at Inverness in 1758, that the price of salmon was a penny a pound and that "the merest servants who are not at board wages will not make a meal upon salmon if they can get anything else to eat" (Mills, 1980).

We have yet to observe fish by underwater television on the loch 'walls' beneath about 150m. The fish of the profundal are not amenable to detection by sonar because of attenuation and widening beam angle, so the information presented  results from UW/TV (Figure 18a, 7K) and netting experiments at 180m-220m depth on the deep basin floors(Figure 18b, 7K). Future experiments will include lowering deep‑towed narrow‑angle transducers to assess this region.

The presence of deepwater fish was confirmed in 1981 through UW/TV, and in 1982 the first three specimens of charr were netted from 220m. It is always something of a problem to be certain that fish are really caught at depth and not during the long setting and retrieval process, but of the twelve charr caught at this depth, seven contained profundal fauna including Pisidium conventus. Some fish believed to have been caught at 200m had surface zooplankton in the lower gut, thus suggesting that a relatively fast descent takes place (Figure 18c, 15K). Versatility is thought to enable Charr to inhabit relatively sterile lakes and to avoid direct competition with the less adaptable brown trout.

On the other hand, it has been shown that there are two distinct races of charr, a pelagic and benthic form, in Loch Rannoch (Walker, Greer and Gardner, 1988). Electrophoresis has shown genetic differences between the stocks and initial results with our Loch Ness specimens also appear to show two varieties, although the visible differences, larger eyes for example, do not appear obvious at this stage.

The offshore waters to 30m are dominated by charr, which feed upon the larger zooplankton such as Daphnia, Leptodora and Bythotrephes. As part of the fixed station work of 1984, depth‑marked nets of various meshes were suspended beneath the raft to a depth of over 30m, aligned N.E.-S.W. The results in Figure 19a (17K), show a poor catching frequency between 8th and 21st August with charr taken at approximately 10m by night, when it is also evident that trout extend over the surface. In addition to zooplankton, the trout contained a proportion of winged insects (