Faunal composition

Africa has over 2000 known species of indigenous freshwater fishes. Table 2.1 lists the families they represent, giving approximate numbers of species, although these are only provisional because in some families, especially the Cichlidae, many new species have yet to be described. The table also shows whether families are endemic to Africa, or also found in the Neotropical or Asian tropics or in marine communities. Table 2.2 compares numbers of all species and of cichlids in the main river systems and their associated lakes. Examples of African fishes are shown in Fig. 2.1.

The most striking feature of the African freshwater fish fauna is the high degree of endemism. Eighteen of the families are endemic to Africa and this endemism at family level occurs amongst the less–advanced (pre–Acanthopterygian) fishes: lungfish Protopteridae, brachiopterygian Polypteridae, clupeomorph Denticipitidae and Congothrissidae, osteoglossomorph Pantodontidae, Mormyridae and Gymnarchidae, anotophysan Kneriidae (now taken to include Cromeriidae and Grasseichthyidae (Greenwood, 1975; Nelson, 1984)) and Phractolaemidae, and otophysan characiform Hepsetidae, Distichodontidae, Citharinidae, siluriform Amphiliidae, Mochokidae and Malapteruridae, and cyprinodontiform Aplocheilidae. At generic level almost all genera are endemic to Africa, with the exception of some euryhaline fishes of marine origin and seven genera shared with the Oriental region, viz. the cyprinid Barbus, Garra, Labeo, Raiamasand loach Neomacheilus, the catfish Clarias, also Channa.

Production in the pelagic zone of tropical seas is very patchy, greatest in upwelling areas and convergence zones, often with much variation seasonally and from year to year. The resources are exploited by fishes in two main ways: (1) by small species with short cycles (such as clupeoids) which multiply fast, so producing many mouths to exploit planktonic food when it is abundant; (2) in larger species (such as tuna) by migrating long distances in search of food. Adaptations to life in the pelagic zone include schooling which helps the fishes to keep together when moving in the vastness of the open ocean, as well as providing protection from predation in the small species. Physostome swimbladders enable the fishes to migrate vertically with ease (which physoclistous fishes cannot do). The uniform silvery colouration of the openwater fishes is linked with an environment that is visually uniform except for light intensities varying with depth and time of day. The acoustico–lateralis system of clupeoids, linked with the swimbladder, enhances sensitivity to water vibrations and so aids synchronous swimming (Blaxter & Hunter, 1982).

Clupeoid fishes of upwelling areas

Clupeoids are among the world's most important food fishes; in 1971 they made up 35% of world landings, and the fourfold increase in catches since 1958 was mainly due to increased landings of clupeoids from low–latitude waters, particularly those of the Peruvian anchovy fishery.

The aim of fisheries management, to obtain the maximum (or optimum) sustained yield of fish from a water body, involves removing fishes equivalent to the amount of fish flesh produced each year (the production) without making inroads into the capital (standing stock or biomass). When comparing data from different sources, a clear distinction has to be made between biomass (stock of fish present at any one time), biological production (a rate) and the catch (yield). The yield represents only a proportion of the production and can range from almost the entire production in fishponds, to a very small proportion where fishing conditions are difficult, or where there are few fishermen, or where natural mortalities are great. As very few estimates of biological production have yet been made in natural waters (freshwater or sea), the yield (expressed as kg ha−1), is often used as an index of production, and in fisheries literature this yield or catch, i.e. the proportion of the production cropped by man, is often loosely called the ‘production’.

Comparative yields

Tropical fishes from freshwaters and the sea within easy reach of land have been fished from time immemorial by subsistence fishermen, and such fishing is still very important in many parts of the tropics. It is only in the last half century that fishing has been heavily mechanized. In the 1950s and 1960s the world catch rose very fast, more than trebling from around 20 million tonnes in 1948 to a record 71.3 million tonnes in 1979.

The high endemism in the African Great Lakes makes it clear that many fish populations respond to the lacustrine conditions in such a way as to become recognizable as distinct species, or even distinct genera. A species is generally taken to be a group of interbreeding natural populations which is reproductively isolated from other such groups. Generic distinctions rely more on the degree of difference as interpreted by various taxonomists. Within these lakes morphological differences within the family Cichlidae far exceed those between whole families of fishes in the sea. These lakes provide some of the best natural laboratories in which to study the ecological and behavioural changes accompanying the evolution of new species. A great deal has been written about speciation in these cichlids, much of it speculative, some of it controversial; see, for example, papers prepared for a recent symposium on the evolution of fish species flocks (Echelle & Kornfield, 1984). Earlier reviews by Fryer & lies (1972) and Greenwood (1974, 1981) provided detailed analyses.

Controversies aroused include the following: (1) Can changes in habits and behaviour, engendered by changed ecological conditions, lead to speciation (as man-made lake studies seem to suggest and as Fryer (1977) held), or do genetically induced mutations, generally accompanied by slight morphological changes, generally occur first on which selection can than act (as held by Liem (1974) and Liem & Osse (1975))?

We have looked at riverine and lacustrine communities with African faunas subject to African conditions. How do studies in other parts of the tropics corroborate, refute, or enlarge these findings?

The Neotropical fish fauna, examples of which are shown in Fig. 6.1, is the most diversified and richest freshwater fish fauna in the world, with more than 2400 species already described. It differs from the African fauna in being derived from fewer basic stocks of fish. It lacks the rather primitive families endemic to Africa and cyprinoids and is richest in characoids and siluroids, developed by spectacular adaptive radiations initiated during South America's long isolation during the Tertiary. In the Amazon, 85% of the fishes are otophysans (compared with 54% in the Zaire), 43% are characoids, 39% are siluroids and 3% are gymnotoids. Roberts (1972) has suggested that the present Amazonian fish fauna with its large number of species might be the product of a million years of evolution from an original stock of two or three hundred founder species, but we do not know phyletic time scales. The numbers of species in families endemic to South America shown in the Appendix are only very provisional, as whole river systems have yet to be explored.

Ecological attributes of tropical fish communities considered in earlier chapters are summarized here and in Table 11.1. There are of course many exceptions to such broad generalizations, but they are made here as a framework to stimulate further study.

1. Phyletic patterns. Phylogenetically different types of fishes dominate the various communities. Riverine communities are dominated by otophysan fishes (primary freshwater fishes), together with other endemic primary freshwater fish families in African rivers; lacustrine communities are dominated by secondary freshwater fishes, cichlids in the littoral zone, clupeids and their centropomid predators in the pelagic zone. In the sea, clupeoids and their acanthopterygian predators abound in upwelling areas, while reef communities are dominated by acanthopterygian fishes able to manoeuvre backwards into crevices, with jaws and teeth capable of exploiting the very varied food resources, finspines and poisons protecting the fish from the numerous piscivores. A diverse group, primarily acanthopterygians (sciaenids, grunts, sparids, snappers) live demersally on the continental shelves. There is also an apparent relationship between phylogenetic position and temporal habits (Hobson, 1974): the more generalized teleosts (such as percomorphs) are typically large mouthed nocturnal or crepuscular carnivores, while many advanced fishes have specialized towards diurnality and feeding on smaller animals and plants. Ontogenetic changes may also occur, for example juveniles feeding diurnally then switching to nocturnal feeding with increased size.

2. […]

We have seen how West African riverine fish communities changed into lacustrine ones in the newly created man–made lakes, also how climatic changes during the 1970s droughts in the Sahelian zone affected fish populations in Lake Chad. How have fish populations responded to the much longer–lasting conditions in the Great Lakes?

Lakes in the Nile drainage with its Nilotic (sudanian) fishes fall into two groups: (1) those with little–modified faunas, showing a low level of endemicity and with few cichlids – Turkana, Albert and Tana; and (2) lakes with a high level of endemism – Victoria, Edward/George – lakes in which cichlids dominate the faunas, as they do in Kivu and Tanganyika in the Zaire drainage and Malawi in the Zambezi drainage (Table 2.2).

Lake Albert (Mobutu) in the western rift valley, a 150 km long, 35 km wide, 56 m deep lake, from which the White Nile flows, has a fish fauna most of which is shared with the Nile. The river has, however, a few species which have not colonized the lake and the lake has a few endemic species: about six endemic cichlid species, one cyprinid species and an offshore–living Lates, L. macrophthalmus, living over the deeper waters.

Lake Chad provided an example of a natural lake in which many of the fish were under the influence of the inflowing rivers, and there was little distinction between riverine and lacustrine fish communities. The creation of a man–made lake behind a hydroelectric or irrigation barrage provides a large–scale experiment on how riverine species become adapted for lacustrine life, how faunas change and how lacustrine communities evolve. In the sudanian area, pre– and postimpoundment studies were made at Kainji, in Nigeria, where a hydroelectric barrage across the Niger, closed in 1968, led to the formation of a 1280 km2 lake with an outflow:storage ratio of 4:1 and a large (10 cm) annual drawdown. Prior to this, the closure of the Volta Dam in Ghana in 1964 had created an 8800 km2 lake, one of the world's largest man–made lakes; here the storage volume was four times that of the outflow and the drawdown only 3 m annually, providing more truly lacustrine conditions. Both lakes were populated with the same species of sudanian fishes from their inflowing rivers, the Niger and Volta, respectively. Lake Nasser/Nubia on the Nile, which formed behind the Aswan Dam closed in 1964, also has many of the same species of fish (Ali, 1984; Latif, 1984), as does Lake Kossou on the Bandama River in the Ivory Coast, formed more recently (Daget, Planquette & Planquette, 1973).

High diversity at low latitudes is a characteristic of both plant and animal communities, and fishes are no exception to this generalization, for among both freshwater and marine tropical fishes greater diversity is shown both within taxa and within communities than at high latitudes. This chapter examines first how species are packed into communities and coexist, and then how these very diverse communities have evolved.

Diversity maintenance

Previous chapters have indicated that diversity is greatest in relatively aseasonal communities, both in freshwaters and in the sea (Table 11.1). Coral reefs carry the most diverse of any fish communities; upwelling areas in the sea have far less diverse faunas than those on reefs. Diversity in rivers becomes less: (1) as the latitude increases and seasonal fluctuations in water level become more marked; (2) with increases in altitude, for example in Andean and Kenyan streams, where this is probably a temperature effect; and (3) towards the headwaters of streams where physicochemical factors, obstructions causing water falls, high speeds of flow, and the size and conditions of dry–season refuges, may be more limiting than food resources.

Fauna lists are very large for many tropical water bodies, but how many species actually share biotopes? The long species lists for a river system could result from each section and the innumerable tributary streams each having its own fauna.

Introduction

Frozen water, in all its forms ranging from snowflakes to icebergs, has intrigued scientists for many centuries. Already Kepler [1] had speculated on the hexagonal shape of snowflakes, wondering what were the forces that transformed round droplets of water into the beautiful stars that he saw in falling snow. Although he did not resort to an atomic picture, he used a concept with formation by the cold of very small, identical particles, which would grow from an octahedral origin. There would thus be three orthogonal growth directions, leading to the formation of six branches in the star, and a hexagon could then form by a flattening of the star along one of the three-fold axes. Kepler discussed the form of different natural objects, ranging from honeycombs over pomegranates to different crystal forms, and he ended by proposing that different fluids would have in-built abilities that would lead to different forms when frozen.

Since then scientists have been equipped with a multitude of tools that allow them to go well beyond visual observation and speculation in the study of nature, but water in all its forms has always been a subject of interest. In this paper we shall deal with one small domain, namely the structure of ice-Ih. Methods of structure analysis originated in the first quarter of this century.

Introduction

The present picture we have of the three-dimensional structure of water in the majority of its existing states is far from satisfactory. Most of the structures in the solid phase – ordinary hexagonal ice-Ih, cubic ice-Ic and the high-pressure ices II to IX – are apparently well characterized, whereby the respective details are understood in the relatively simple terms of a tetrahedrally coordinated hydrogen-bonded lattice. The situation is very different for the liquid, aqueous solutions and amorphous ice. Within these phases, there is no underlying repeating lattice that forms the fundamental structural basis in crystals. The molecules can occupy many alternative positions relative to one another, provided they do not violate the intermolecular forces: hydrogen-bonding, van der Waals contacts, dispersion, etc. Significant angular distortions from the expected tetrahedral coordination occur, but the extent and limitations of these deviations are unclear.

The main method used in analysing the details of molecular structure is diffraction (neutron, X-ray and electron). This has been invaluable for locating the individual relative atomic positions of the molecules in the crystalline state; however, for molecular assemblies which have no repeating units the method is more limited. Diffraction from liquids and amorphous substances yields only structural information about correlations between pairs of atoms, pair correlation function (PCF), which gives the probability of finding an atom within a certain distance of another.

These words, attributed to Rutherford, provide a focus for this article. To what extent can one ‘see’ that most important of all the cations – the hydrated proton? What do different eyes – different experimental techniques – see and how do the views from these differing points of observation differ or complement one another? Our search will encompass the three states of matter – solid, gas and liquid solutions. And our ‘eyes’ will include X-ray and neutron diffraction, nuclear magnetic resonance, pulsed-electron-beam high-pressure mass spectrometery, infrared and Raman spectroscopy and relaxation techniques.

In addition we must consider the ‘mind's eye’: quantum simulations and theoretical inferences. But first we will place our search in historical perspective.

Historical Perspective [2–5]

The recognition that the class of compounds generally recognized as acids had, in common, the element hydrogen can be attributed to Davy and Liebig. In 1838 Liebig defined acids as ‘compounds containing hydrogen, in which hydrogen can be replaced by a metal’ [2]. This concept unified a body of empirical knowledge that had been accumulating: the sour taste; the solvent power; the ability to turn blue vegetable dyes red; the ability to combine with bases (or alkalis) to form salts and water; the existence of acidic compounds which did not contain oxygen, e.g. HF, HC1, HBr etc. (a discovery by Davy (circa 1810) that laid to rest the oxygen theory of acids proposed by Lavoisier).