Varying Isotope Response of Freshwater Molluscs to Environmental Change: Equatorial Africa

Stable oxygen and carbon isotope ratios from modern African freshwater molluscs and water samples were examined to ascertain the reliability of such information as a proxy in the study of paleoclimates and paleoenvironments. In this study, the isotope ratios of the freshwater molluscan shell samples were determined from both the apertures and at interval of accretionary growth of individual shells collected at monthly intervals, from four sites. These ratios were examined to ascertain the extent to which they correlate with temperature, rainfall, water isotope ratios and other ambient condiditons. The samples studied were collected from a tributary of the River Pra at Krobo, in the lJes tern Region, River Densu at Achimo ta, in the Greater-Accra Region and a stream in Sunyani, Brong-Ahafo Region, all in southern Ghana approximately 700 km south of the Sahel, and from the lJinam Gulf of Lake Victoria at Kisumu, Kenya. Molluscan shells and water isotope ratios responded to heavy rainfall, as well as variations in temperature and evaporative effects. Oxygen isotope ratio values can be used to estimate seasonality, and in turn, the life span of the molluscs. For application of this method to fossil molluscan shells for elucidation of paleoclimates, however, additional geological and geographical data are desirable. ACKNO\.TLEDGEMENT Upon completion of this dissertation, I wish to acknowledge those whose earthly paths have crossed mine in a very significant manner while untertaking this study. My sincere gratitude goes to my major Professor, Dr. Paul I. Abell, who initiated this study and offered me the necessary support and encouragement to complete it. To the faculty and staff of the Chemistry department, I offer my sincere gratitude for their interest and assistance whenever the need arose. I am, particularly, thankful to Dr. James L. Fasching, chairman of the department and a member of my committee, and Dr. Leon Thiem, who has expanded my knowledge in water chemistry. The assistance of the Isotope Geochemistry group at the Graduate School of Oceanography is very much appreciated especially those of Joseph Orchardo, Jim Burdett, Dr. Mike Arthur and Dr. Mike Bender. I greatly appreciate the support, concern and encouragement of the members of Dr. Abell's research group who have become an integral part of my life, Dr. Joseph Mcclory, Miguel Muzzio, Joseph Yong, Yi Lin. To all my close associates who have made my stay in Rhode Island a pleasant and fulfilling experience, Kwabena Anyane-Yeboa and registered. especially, Gabriel Fosu, Hsiu-Ling Liu, family, a special note of gratitude is The faithfulness and the relentless efforts of the sample collectors, Leonard Amegashitsi, Richard Oscar Fynn, Rev. Joseph Tsiquaye, Sr. MagaretMary Nimo, Peter Adodo and Ben Amegashitsi has been a source of inspiration, not forgetting those who came to my aid


ACKNO\.TLEDGEMENT
Upon completion of this dissertation, I wish to acknowledge those whose earthly paths have crossed mine in a very significant manner while untertaking this study.
My sincere gratitude goes to my major I greatly appreciate the support, concern and encouragement of the members of Dr. Abell format was adopted for the dissertation presentation according to the guidelines of the Graduate School of the University of Rhode Island.
The dissertation body is composed of two papers and an appendi x .
The first paper examines the oxygen and carbon isotope ratios in the shells of a freshwater gastropod, Melanoides tuberculata as proxies for environmental change in southern Ghana.
In the second paper, the varying isotopic response of modern freshwater molluscs genera, Bellamya, Corbicula, Mutela, and Aspatharia, to environmental changes in the Winam Gulf of Lake Victoria were studied.
The appendi x is a comprehensive dissertation bibliography v   . In the study reported here, the oxygen and carbon isotope ratios of freshwater gastropod shell samples, taken from the apertures (in order to sample the most recent shell growth) of the individual shells collected at monthly intervals from three sites in Ghana were examined to ascertain their correlation with weather, temperature, rainfall, water isotope ratios and other ambient conditions. Also examined were isotope ratios in sequential samples from the growth spiral of gastropods collected from the three sites. The gastropods and water samples studied were callee ted at monthly intervals from a tributary of the River Pra at Krobo in the Wes tern Region, River Densu at Achimo ta in the Greater-Accra Region and a Stream in Sunyani, Brong Ahafo Region, all in Southern Ghana approximately 700 km south of the Sahel.
How this information on modern African freshwater gastropods can be applied to interpret the oxygen isotope ratios of fossil shells is also examined. If clear correlations exist, then we may achieve our ultimate goal of interpreting paleoclimates from shell isotope data.
3 BACKGROUND Despite  suggestion that isotope ratios of carbon and oxygen in naturally occurring carbonates bear the potential for calculation of paleotemperatures, it is only recently that freshwater gastropods have been investigated as paleoclimatic indicators. \.leber and La Rocque (1962) examined the whole-shell isotope ratios of several genera of freshwater gastropods from Plum Lake, \.lisconsin, and one from Lake Erie.
Primarily interested in the carbon isotope ratios, they suggested that a substantial contribution to the carbon of the shell came from organic sources, rather than the dissolved inorganic carbon (DIC) of the water, and that this organic carbon was probably in the form of carbon dioxide released by decay of the isotopically light plant debris. The Lake Erie dwelling species Pleurocera acutum, with the three species living in close association with pond weed had mean carbon isotope ratios of -6.88, -4.50 and -8.85%0 vs PDB standard. It has been established that both marine and freshwater mollusks form their shells at or close to oxygen isotope equilibrium with the water in which they .
dwell (Epstein et al., 1951;, and Keith and \.leber (1964) have studied the 13 C composition of freshwater molluscs by comparing marine and freshwater shells. Studies carried out on the Mackenzie River drainage system by Hitchon and Krouse (1972) concluded that the major source of the dissolved inorganic carbon 13 C is from biogenic HCO - He observed that while temperature changes account for most of the 0 18 0 variations, temperature dependence of o 13 C is small. The portion of o 13 C changes that appeared to correlate with climatic changes was attributed to evaporation-rate changes or photosynthetic activity associated with temperature changes. The work of  was an extension of the investigation of Weber and La Rocque (1962) in which, again, whole shell isotope ratios _ were measured. While this sampling scheme may be satisfactory for the shells from cool temperature regions, where growth is likely to halt completely during the winter months, it may ignore differential growth rates and temporarily variable isotope ratios for shells from subtropical or tropical bodies of water. The use of isotope ratios of the shells of freshwater gastropods as proxies of past African climates have been examined with fossil gastropods of Africa Abell and Nyamweru, 1988;Abell and Williams, 1989). In one of the recent studies which covered much of the African continent , geographical pat terns of the stable oxygen isotope ratios obtained from whole shell ~nalysis of individual modern African freshwater shells were established. These recent studies have revealed that sequential analysis along the growth spirals of the shells gives information on changing environmental conditions during the lifetime of the gastropod. Ve report a detailed study on gastropods collected from three tropical sites with similar rainfall pat terns i n Ghana, Vest Africa. Many freshwater gastropods have life s pans of a bout a year, (Leveque, 1968) and might be expected to preserve useful environmental records over that span of time.

5
Amongst the several proxies available for the study of environmental changes none provide sharp definitions of those changes.
Some of the paleoclimatic proxies that have been and are currently being used include the distribution of pollen grains in sediments and the deposition of Coleoptera beetles in sediments to give a rough indication of prevailing climatic conditions. Also the deposition of fine layers of windblown dust, loess, can serve as a paleoclimatic proxy, while the oceans with their accumulation of sand, mud, gravel, plant and animal fossils also provide geological paleoclimatic indicators (Schneider 1984). The inability to correlate these proxies with short term seasonal changes is the major limitation. That is, on the time scale of hundreds or thousands of years, many proxies give us much the same information about major climatic changes. But animal habitability, including human, of an area depends as much or more on the annual range of temperatures and the length and severity of annual dry seasons as on average conditions. The short life span of most species of freshwater gastropods and their isotopic ratio sensi ti vi ty to their environment makes them unique proxies to the range of annual climatic conditions; i.e., seasonality.
The technique described herein involves the comparison of the relative amounts of the stable isotopes of oxygen and carbon using their two most common isotopes, with natural abundances of 16 0, 99.76%; 18 0, 0.20% and 12 C, 99%; 13 C, 0.01% respectively. For example, the stable oxygen isotope measurements give the relative differences between the sample and the laboratory standard ratios of 18 0/ 1 6 0 from the following formula :

%0
A negative 0 18 0 value means there is less 18 0 in the sample than in the standard against which it is compared and the reverse if 0 18 0 is positive. It is also known that an increase in 0 18 0 of only 1 per mil in calcium carbonate shells corresponds to a change in water temperature of about 4.0°C to 5.0°C (Craig, 1953). But temperature is not the only variable over the life span of a mollusc; fresh inputs of rainwater and/or evaporation of the host body of water can also affect oxygen isotope ratios.
The short life spans of most species of freshwater gastropods and · their isotopic ratio sensi ti vi ty to the i r environment makes them unique proxies to the range of annual climatic conditions.
Much work is still being done to understand the life cycle or reproductive strategies of the freshwater gastropods (Brown, 1980;Leveque, 1968), but it is known that some correlations exist between isotopic ratios of modern shells and the environment in which they grow  .
Obviously, an understanding of how the oxygen isotope ratios of freshwater gastropod shells vary with the environment in which they grow could facilitate the application of the isotopic ratio measurements to continental paleoclimatology.
Why African climates ?
The geographical limitations of this study were dictated by both practical and theoretical reasons: . 7 (l) The African continent provides a wide range of modern climates for comparisons with past climates.
(2) The latitudinal range of the continental land mass, from approximately 30°N to 30°S gives a useful range of climatic conditions, without encountering extremely low tempera tu res where molluscs may be dormant over large portions of the year.
(3) Initial surveys  were made possible by the availability of good museum collections of gastropod shells collected, at least in part, because of public health problems associated with freshwater gastropods.
(4) It was also hoped that one consequence of this study would be application to paleoclimatic influences on the development of early man, whose original home was Africa.
(5) Finally, past African climates are of interest because we would like to know if the expansion of the Sahara desert in Africa is a cyclical process or not. Such knowledge would help in the formulation of land use policies based on expectations of probable changes rather than merely hoping for the best.
In addition to these ·primary reasons for selection of Africa as our study area, knowledge of ancient climatic variations could provide answers to some questions raised in the evolutionary process and in geology.
The geographical conditions of past water basins could be deciphered, and the validity of the assumptions linking environmental change to the process of evolutionary change might be confi r med.
According to current climatic theory, global weather patterns exist because different parts of the globe receive and absorb varying amounts 8 of solar insolation; upper air currents and ocean currents are mechanisms for restoring heat balance (Budyko, 1978; (1975) and Dansgaard (1964). 'We hope to apply these factors in the discussion of our results from southern Ghana, as we expect seasonal variations in the 0 18 0 values from the sites.
The modern sites were selected on the basis of differences in ·

THE SITES AND THEIR ENVIRONMENTS
The three sites in Ghana, which were selected for this study all lie between the Kwahu-'Wenchi Plateau and the Gulf of Guinea (Fig. 1).
The sites consist of two rivers and a stream. River Pra was sampled at Krobo, in the 'Western Region, Sunyani stream at Sunyani, in the Brong-Ahafo Region and River Densu at Achimota in the Greater Accra Region.

SCALE
Vegetation zones of Ghana.

16
The area where collections were made are characterized predominantly by Pre-Cambrian rocks and intruding granites referred to as the Birrimian and the Tarkwaian (Boateng, 1967). This indicates the absence of major carbonate sediments such as limestone and dolomites.
The region between the Kwahu-Venchi plateau and the Gulf of Guinea is a gentle slope from the scarplands to the coastal plains.
In this region, one finds ridges and valleys of Rivers Pra, Ankobra, Tano, Amisa and Densu. River Pra has its source close to Mpraeso in the Kwahu scarp with its tributary rivers like Offin, the Anum, and the Birim, it is the largest river system in the southwest of Ghana. It joins the sea near Shama in the Vestern Region with rapids at Krobo. River Densu derives its source from the Akwapim-Togo Ranges. The soils at the three sites are forest ochrosols which are usually red or reddish brown on the summits and upper slopes of hills, orange brown or brown on the middle slopes, and yellow-brown on the lower slopes. They are better drained and less acidic than the oxysols. (Boateng, 1967) The lower valleys experience the same rainfall pat tern as the Kwahu Plateau except in the· western section near Axim and Half Assini where the maximum rainfall in the country occurs and therefore is the wettest area in the Ghana, with permanent equatorial rainforest. ( Figure   3) The eastern section which includes Achimota in Accra has very little rainfall between September and November and, therefore, lies in the dry part of Ghana with coastal thicket and grassland. Vegetative cover at both Sunyani and Krobo consists of moist semi-deciduous forest which forms a transition zone between the evergreen rainforest and the . 1-7 savannah. This area is very humid with tall trees creating a "canopy" effect which decreases gradually away from the evergreen to the savannah grassland.
In some of the moist areas, ferns and orchids grow on the branches of trees and some swamps harbour raffia and oil palms, mahogany, and ebony. Other crops that fluorish in this type of vegetation are cocoa, cocoyam, plantain, maize and sugarcane. (Varley and Yhite, 1958 (Oboli, 1963). These episodes are characterized as follows: Dec. -Feb. Dry, "Harmattan" winds off Sahara desert, Temperatures range from 2s 0 c· to 32°C with high evaporation in day, strong radiational cooling at night. Rainfall is lowest.
Har. -Jul. Principal rainfall with maximum in June or July Cloudy. Approximately 10 gastropods were collected each month from the bottom of the river (usually in a ~luster), allowed to die and · then dried before shipping ·to the laboratory. The water sample was collected in a 25 mL septum capped vial, which was lowered down into the river at the pr oximity of the gastropods and then opened to fill up.
It was recapped, labeled and kept sealed for analysis in the laboratory.
Vater temperatures were also measured monthly over the same pe r iod at the collection site.
Individual gastropods from the set of 10 collected each month at the three sites were accepted as representative. We had already The processing of the freshwater gastropod shells for the purpose of evolving their C0 2 content into ampoules for stable isotope ratio determination was by the procedure of .  (Grossman, 1962). calc1 e Monthly water samples were analyzed according to the procedure of .
This involved the equilibration of approximately 24 cc (STP) of C0 2 with 2.5g of water in a thermostated bath for approximately 7 days and the analysis of approximately 3cc aliquot of the gas on a V. G. Micromass 602-D mass spectrometer. To facilitate rapid equilibration the pH may be adjusted to 6 or lower (Hills and Urey, 1940). The water sample was inserted into a 25cc round bottom flask provided with ground glass joints and a stopcock so as to be connectable to a vacuum manifold . The water was frozen in liquid nitrogen-2-propanol slush bath and the air quickly pumped away. The ice was melted and warmed to room temperature to release gas which was trapped during the initial freezing. The water was then refrozen and pumped for a minute to remove the remaining non-condens i ble gases. The ice was melted again and commercial cylinder C0 2 , 99. 8% purity, was All mass spectrometric analyses we r e r e ported re l ative to the PDB C02 standard gas . The probable error was ± 0.01%. Replicate samples measured on the same day were us ually r epr oducib l e t o ± 0 . 05%.
Standards (usually NBS-20), we r e always r un t he s ame day as any unknowns were run .
As part of the procedure for sampling, sample collectors at the e asked to furnish information on environmental changes sites wer l 'allY at the habitat of the . freshwater gastropods.
We Krobo Site Table 1 gives the days and rainfall amounts during the period of collection at an airforce meteorological station at Takoradi. The rainfall pattern at this station is representative of rainfall in southern Ghana within which all the collection sites fall except that the amount of rainfall will be slightly lower at Sunyani and Achimota.
Helanoides tuberculata and Bellamya unicolor shells were collected from December 9, 1985 to October 12, 1986. Water samples were also collected at the same period. The water temperatures are summarized in Figure 4.
The following are comments recorded verbatum from the collector's notes: September rains -river flooding again.

Achimota
The Achimota Melanoides tuberculata shells were collected monthly between Nov. 22, 1985-Nov. 16, 1986, with the exception of Jan. 1986 when collection was not possible. Yater samples were also collected for the same period . Water temperatures are summarized in Figure 4.

WATER AND SHELL 0-18 AT RIVER DENSU, ACHIMOTA
Oct   . It is known that fractionation of precipitation is very significant at higher elevations.
The latitude at which the rain fell should influence the oxygen isotope ratios significantly.
Temperature plays a very vital role in the determination of the & 18 0 of both the shells and the water in which they grew. Just as air temperature alters the oxygen isotope ratios of precipitation so do -air and water temperature influence the precipitation of aragonitic shells.
The paleoclimatic scale of Craig (1965) is based on this principle. In most modern environments studied, the air temperatures correlate in parallel with the water temperatures. Evaporation which is also temperature dependent and engendered by dry wind may concentrate a body of water thereby, enriching its isotope ratios as well as that of the molluscs. In environments with high humidity, rainfall and moderating temperatures, evaporation is reduced and therefore, the oxygen isotope ratios of the gastropod shells are less enriched.
sampling the oxygen isotope ratios in that rainwater in Gastropods, body of water, will generally provide a faithful record of their host . otope ratio, particulary if the gastropods lived in large deep that is . .
bodies of water, resistant to evaporative changes or in areas where rainfall is persistent through much of the year and humidity is high enough to discourage evaporative fractionation. The deep lakes, Tanganyika and Malawi, are the obvious examples of those bodies of water large enough to resist short term evaporative change, while the rivers and lakes in the tropical forest of Central and West Africa are going to be characterized by low evaporation . On the other hand, there are a number of areas where control of 0 18 0 will be largely by evaporation. The shallower East African and Southern Saharan lakes will be maintained by seasonal rainfall in their source areas, but will be subject to continuous high evaporation rates. Lake Turkana, for example, fed largely (80%) by seasonal rainfall in the highlands of Ethiopia, has a climate of relatively unchanging temperature, but the lake level, and the 0 18 0 of the gastropod shells oscillates with this seasonal imput. In all these arid locations, the average value of 0 18 0 will be positive, reflecting the prevailing evaporative condition.
The combination of temperature and evaporation are enough to offset the latitudinal effects of several permil. These general climatic trends and their manifestation in the oxygen isotope ratios of gastropod shells has been reported by . Latitudinal effects produce 0 18 0 values near -4 to -5%0 at the equator, and ex tending to 8 to -9%o at the northern and southern extremes of Africa. Evaporative situations in East Africa are clearly delineated with 0 18 0 values near 63 +1 to +2%0 • The regional anomaly in the Transvaal area of South Africa is also obvious· Abell' s isotopic ratio map supplies the norms for ratios and climatic patterns of modern Africa, but it also isotope supplies examples of regional effects which can be used for the · of paleo i'sotope ratios Some general observat1'ons have interpretation -.
emerged from the recent studies of 6 18 0 values and their correlation with environmental conditions. One of such observations is that in cool, high and regular rainfa'l areas, where 6 18 0 of shell is -3 to -S%o, in situ evaporation is probably much less important than temperature in controlling 6 18 0. But where rainfall is light, and conditions are arid and hot, rn most circumstances temperature variation is small, and the 6 18 0 values are predominantly governed by evaporation.
For example, at both Lake Malawi and Lake Victoria (Winam Gulf) there is minimal temperature change with the seasons, but 6 18 0 of the shells varies considerably (Amegashi tsi, Abell & Ochumba, unpublished). There will be exceptions to these generalities, but they make a starting point.
In an attempt to apply these interpretations to paleoclimates we must not lose sight of the importance of the correlation between amount effect or rainfall and vegetation. The amount effect which according to Dansgaard (1964) is engendered by the deep cooling of air in heavy frequent rainfall, with minimum possible post-precipitation enrichments through evaporation, has been found to correlate well with vegetation cover (Tucker et al., 1985. With respect to the African cl' . imate, it has been found that areas with enough rainfall, and rainfall suff · · lciently well distributed throughout the year to ensure permanent show 0 18 0 values vegetation cover, ranging from -3. 5 to -0. 9% 0 or range Abell and Nyamweru, 1988;Abell and Villiams, 1989).
The above factors that influence the 0 18 0 in s hells will not affect the cS 13 C in the same way as they do the 0 18 0, due to the various sources of carbon available to the molluscs.
Changes in o 13 C may be to varying utilization of DIC or attributed photosynthetic activity, both of which may be changes in the associated with .
ture changes . There are other factors which also tempera . .
contribute. to the changes in o 13 C and limit the use of o 13 C as a climatic indicator. Carbon in shells of freshwater systems may come from dissolved inorganic carbon or allochthonous and autochthonous organic matter. They can also ingest limestone particles but the major species has been observed to be bicarbonate ion (Hem, 1970). Other species that are present in water in relative amounts are C0 2 (aq), HC0 3and co/- (Hitchon and Krouse 1972). Their concentrations in a given system vary with temperature, pH and Pc 02 of the atmosphere.
Ve know that long residence time of rainfall in a given body of water could cause enrichment in the 13 C through exchange with atmospheric co2.
In the study of the stable isotope mass-balance of DIC during blue-green algae bloom in a softwater lake,  have found an enrichment factor of 13%0 for HC0 3 -with respect to C0 2 (g) for a solution at pH 9.5 and suggested that a chemical enhancement of C02(g) at high pH may also cause enrichment close to this value. They have attributed the anomalous o 13 C they were reporting in the softwater lake to the fractionation induced by photosynthetic algae when dissolved C02 concentration is high in surface freshwater. The o 13 C of the algae depleted by 13%0 with respect to the aqueous co 2 when its is concentration is above that of p atm and up to 20%0 when the concentration of aqueous C0 2 is far greater than that of the atmosphere 1978; . In another recent (Rau, Tanaka et al.(1986)  matter.
An identical difference was observed between the o 13 C of organic matter and pedogenic carbonate from palaeosols of Pleistocene to late Miocene age in Northern Pakistan. These findings are in agreement with isotopic equilibrium between C0 2 (g), · HC0 3 -(aq), co/and aqueous and solid carbonate specie~ in a soil system controlled by diffusive mass transfer of soil C0 2 derived from irreversible oxidation of soil organic matter.
An enrichment of up to 2%o of the o 13 C of soil carbonate in low respiration rate soils at a depth of SOcm has been reported by Quade et al.(1989). This enrichment has been attributed to diffusional mixing of atmospheric C0 2 and plant derive d As a result of fractionation effect, the C 3 and C 4 biomass in Car bonate would have -12 and 2%o repectively at 25°C. soil It has been observed by  that shells with depleted O f s13c, will indicate incorporation of substantial fra~tion of values de rived from photosynthetically produced organic material such as carbon algae or plant detritus. Parallelism observed between carbon and oxygen in most of the analyses may reflect changing organic productivity with the changing supply of nutrients introduced by fresh water imput.

REFLECTION OF ENVIRONMENTAL CHANGES IN SHELL AND ~ATER ISOTOPE RATIOS
The plots in Figures 6, and 7 show S 18 0 values for both shells (aperture samples) and water corrected to PDB and temperature corrected to 2s 0 c. It can be observed that there is a correlation between the 6 18 0 values of the gastropod shells and those of the water samples, and that the shell isotope ratios lag behind water isotope ratios by about 1 month. Deviations from an exact correlation are to be expected, given that an instantaneous sampling of a constantly changing water supply may deviate from the time-averaged isotope ratio of a finite shell aperturegrowth that may have taken place over several weeks.
The parallelism between the 0 18 0 plots of the water and the shells, which is a good indicator as to the possibility of applying oxygen isotope ratios in paleoclimatology, is in accordance with an earlier observation by  that the oxygen isotope constant of calcite is dependent on both the 0 18 0 and temperature of ambient water.
At Krobo, the Collector's monthly remarks on water level of the River Pra wh1'ch were requested as part of this research project, and the climatic conditions during the period of water and gastropod prevailing correspond very well with the mean monthly rainfall values collection bu lated in Table 1 and also plotted in Figure 17 (Griffiths, 1972). ta . .
V t and central Equatorial Africa where rainfall is high and well Jn es d out over the year, such that there is permanent vegetative cover, sprea 61 a 0 values in gastropod shells have been observed to be between -3.5 to -O. 9 per mil .
Our values fall close to this range with the average monthly 0 18 0 for the one year period being -3. 51 for the water samples and -3.60 for the gastropod shells, giving a 0.11 per mil average difference between the shells and the water samples. Without any fractionation associated with land masses, the expected values of the g1so at this latitude should have been approximately -1.0 or -2.0 . Here, this difference is attributable to an amount effect, which is substantiated by the os -ow values of -0.52 from the water and -0.54 from the shell. This is in consonance with the observation of Tucker et al., (1985) and  that the latitudinal effect is overwhelmed by the amount effect in tropical rainforests. (Figure 3) The seasonality information summarized in Table   2 can assist in the separation of the amoun~ effect from the temperature effect in the tropics. This is accomplished by taking the difference between 1: Os! the unweighted mean o value for summer and ow, the unweighted mean o value for winter. A negative os -ow indicates the predominance of amount effect in the tropics.   Table 3.
The average temperature recorded in Table 3    1 ~180 values correlate better with precipitation than the water the shel o 6 uo values· Air temperature correlations with shell 0 18 0 and water go od as the precipitation correlation with shell 0 18 0 and 6 uo are as water 51so.
The correlation between the shell oxygen isotope ratio 1 S a nd the water oxygen isotope ratios, 0.68, is also listed in va ue Table 4.
The climatic conditions prevailing in the Southern rainforest region of Ghana consist of two rainy periods in March-July and September-October. The September-October rains in Ghana are followed by the Harmattan winds from the Sahara desert, which are characterized by high daytime temperatures with low humidity and dry cool nights. The Barmattan is usually intense during December and January  however, during the sampling period, the highest intensity of the  t correlations exist between shell 0 18 0 and water 0 18 0, air Significan d Precl . p1' tat ion however the correlat1' on between shell tempera tu re an ' ' 6 u 0 and water tempera tu re is even better (0.74). This is understandable because the microenvironment of the gastropod is greatly influenced by the water tempera tu re. For all practical purposes, the correlation between water 0 18 0, shell 0 18 0 and water temperature are the same 0.75.
Since it does not rain heavily all year round, the precipitation influences the 0 18 0 variation by approximately 44% leaving the major influence, 55%, of the 0 18 0 values to water temperature variation which improves considerably with the slippage (65%).

SUNYANI
The water 0 18 0 was responsive to the effect of temperature as can be seen in Figure 6. At a temperature of about 20°C during the Harmattan months (Nov-Jan) 0 18 0 of the water was -5%o but became enriched by 1%o with a 2°C increase in temperature . The oxygen isotope ratios remained almost constant until a further temperature increase in July and August, when it was enriched by another 1%0 • Therefore, for a temperature range of 3°C, a 2%0 enrichment was observed.
It should be noted that this collection site was in a fas t f lowing body of water and in this high rainfall region, changes in o 18 0H 20 may occur too r "dl api Y to be recorded by slow shell accretion. The robust shells , recorded a 0 18 0 range of 2%o throughout the collection period. however, d .
to Dansgaard (1964) a one degree change in temperature should Accor ing an average change in the oxygen isotope ratio of 0.7%o of water. bring The shell oxygen isotope ratios predicted for this general area  is recorded as -4 while at this location we according to find an average shell isotope ratio of -5.4.
The depletion may be attributed to altitudinal effect and limited evaporation due to vegetation cover at the collection site. According to Dansgaard (1964) and  the altitudinal effect will cause a depletion in the 0 180 and it should be noted that Sunyani is at an altitude of 310m.
A lapse rate of 0.5%o per 3°C is reported   This may be attributed to altitudinal influence, a predominant factor at Sunyani which was not incorporated as a variable in the regression plot.
Ve know that Sunyani is at an altitude of 310m and there is "canopy " f the forest at the sampling site inhibiting evaporation. This effect o unt for the depletion of the 0 18 0 at this site compared with maY acco f Krobo and Achimota. The amount effect is recorded in the shell that o .
h 'n Table 2. &lBO as s own i Ve notice that there is less correlation between shell & 18 0 and 1s 0 The 0  There appears to be no correlation between the o 13 C and the & 18 0 of the shell. The average o 13 C value of -ll.6%0 in the gastropod shell may be 1 tion of an incorporation of bicarbonate (from atmospheric C0 2 a ref ec "')) and metabolic carbon of C 3 or C 4 plants. If the latter is the (.-7too Se then the shell isotope ratio is a result of a mass balance between ca ' . bicarbonate at (-0%0) and C3 (--26.5%0) or C4 (--12.5%0) (Vogel et al., 19 78) plants. Of the two possible explanations, the latter is tenable based on results of aquaria experiments in our laboratory. In aquaria exposed to atmospheric C0 2 and gastropods fed on fish food, with o 13 C of for a month, we have observed that the o 13 C incorporated into the -20%0 shell was -9.1%0· The shell o 18 0 was 30% influenced by the water temperature and 24% by relative humidity. The reason why the influence of water temperature on the water 0 18 0 is not as high may be attributed to the overwhelming effect of the amount effect as indicated in Table 2. There appears to be an unaccountable factor influencing the 0 18 0 of the shells a t Achimo ta (Figure 13). and water water siso. Amount effect is recorded by the Sequential analyses of whole Melanoides tubercula ta shells from each of the sites collected in December 1985, February 1986 and April 1986 show how major seasonal changes like the peak of the rainy season and the drought periods are preserved in the oxygen isotope ratios of the shells (Figures 15-19).
In December 1985 shells (Figure 17) from the 3 sites all recorded 6 u 0 of -4% 0 which is the expected value for this area according to . For better correlation, samples from a bigger body of wate r may be required or a relatively stable body of water like a pond. · water movement may account for some of the lag observed Differences in dBO and 51soshell. between 0 water .
Alth~ugh the sequential shell analyses data from the three sites ( Figures 15, 17, 18 and 19) are compared for the following months, December 1985, February 1986 and April 1986, it should be noted that the samples being compared within a given month were not collected on the same day. In some cases, they were collected approximately two or three weeks apart. Despite this discrepancy, shells collected within a given month from the three sites all do record the climatic changes prior to the time of sampling.
The 0  It is highly desirable to know the age of the shell in order it with other historical environmental information. Fossil to correlate shells may undergo recrystallization from aragonite to calcite, a engendered by local weathering conditions. In order tb extract process na lity information from a fossil shell, a sequential sampling along seaso the accretionary growth pattern of the individual shells is necessary, while the average oxygen isotope ratios of whole shells will be useful in the characterization of major regional climatic trends.
In addition to shell isotope data, some environmental information Amegashitsi, Achimota; and Peter Adodo, Accra.
85 C mbridge, England, pp157-177 press, a Beadle, L. c.' (1974 Epstein, S. , Allen, H. A. and Urey, H. c. (1950). Improvements in mass spectrometers for the measurement of small differences in isotope abundance ratios. Rev. Sci. Insr. 21, 724-730. Mills, G. A., and Urey, H. C. (1940 In these studies we wanted to be sure we understood the correlations between the oxygen isotope ratios of the shells and the isotope ratios i < g ing from several months upwards to a year (Leveque, 1968). span ran . monthly samples, we hoped that they would provide us with taking . overlapping sequences of environmental information as we analyzed for e ratios of carbon and oxygen along the growth spiral of the shell isotop from apex to aperture. Furthermore, we assumed that the carbon isotope ratios might provide information on the existence of "vital effects", in sources, whose carbon and oxygen isotope ratios bear the potentiality of 101 calculations, were carried out by Keith et al ( 1960, paleotemperature 19 65) after such a suggestion by . This technique has 1964, P lied by other workers (Epstein et al., 1951Mook, 1971, been ap 1977 ; Burchardt, 1977) and it has been established that both marine and h t er molluscs form their shells at, or close to, oxygen isotope f res wa equilibrium with the water in which they dwell (Epstein et al., 1951;.  has demonstrated that seasonal temperatures are recorded in the growth bands of both modern and fossil clams by carrying out an oxygen isotope study of the Pismo clam (Tivela stultorem) of the coast of California. A micro-sampling technique along a section cut from the shell was employed in this study.
Similarly,  and   in shells of different species .of gastropods and pelecypods grown at the same locality, however, the o 13 C variations were rather more substantial (up to 4%0 ) .  studied "vital effects" using stable isotope ratios in modern benthic foraminifera.
The investigation of the isotope ratios of the shells of African freshwater molluscs as proxies of past climates was initiated several years ago with the examination of whole shell oxygen isotope ratios of the "fossil" molluscs of Lake Turkana . This was followed examination of whole shell oxygen isotope ratios of gastropod by an f rom some 80 sites covering much of Africa . Since shells table isotope ratios in the shells of the freshwater gastropod then, s .

Melanoides tuberculata, collected from Holocene sites in the Chalbi
-. of north-central Kenya, have been applied in a study of the nature Basin of paleo Lake Chalbi. (Abell and Nyamweru, 1988) In a recent development, carbon and oxygen isotope ratios of ancient freshwater gastropod shells collected from the Afar region of Ethiopia have been utilized in studying the paleoenvironments of that region. (Abell and Villiams, 1989) It was evident from these investigations that much more detailed isotope studies were needed to understand the forces producing the isotope ratios. Shells from natural sites with sequential analyses along the growth spirals and bands of the molluscan shells might be expected to give short sequences which would be interpreted in terms of past environmental events, the temporal length of the sequence depending upon the life span of the mollusc being sampled. This paper reports on one of the first of these studies on molluscs collected at regular intervals from a natural site -the Winam Gulf of Lake Victoria.

Description of Site and Environment
Lake Victoria is one of the largest lakes in the world with a surface area of 75,000 km 2 at an altitude of 1240 m. It lies across the equator between latitudes 0° 30'N and 2° SO'S. The lake is shallow with a maximum depth of 79 m and an average depth of 40m. Water levels are altered by changes both in rainfall and temperature. The Winam Gulf has an average depth of less than 20 m. Lake Victoria is bound along the southern, northwestern and h shores by shallow bays most of which lie in the branch valleys nort ern of the pre-lake river system, which provides good conditions for vegetative growth (macrophytes).  The lake is fed largely by the Kagera River which drains the high rainfall area of the Ruanda and Kigezi Highlands bordering the ~estern Rift, but it is also fed by the Nzoia River in the northeast which has its source in the Kenya Highlands, specifically Mount Elgon and the Cherangani hills.
Smaller seasonal rivers come in from the south and the east while northern and northwestern inflows are mainly from slow flowing swamps. Lake Victoria flows out into Lake Kyoga in the north.
There is little seasonal change in water temperature compared with temperature changes in lakes in the temperate zones. Surface and bottom temperatures differ by not more than about 2°C. The lowering of water temperature between May and July is attributed to the Southeast trade winds which might affect evaporative cooling of the lake, as the intensity and duration of insolation is nearly constant. Relatively clear skies, and radiational cooling in the night is accountable for the lowering of temperatures between May and July, and surface cooling during this period is believed to provide energy for lake mixing. From February to May, the bottom water fluctuates little and becomes slightly acidic due to an increase of dissolved C0 2 •  The frequency of violent storms, coupled with temperature and oxygen data gathered at the center of lake (sampling depth 60cm) during one of such storms by Kitaka (1971), suggest the drawing of lower water to the surface by a cyclonic swirl in the center of the depression.
The climate of Lake Vic tori a, like most of East Africa, is d by subtropical high pressure areas situated North and South influence of the equator. The subtropical air masses move north during northern and south during southern summers. summer The south-east trades from the Indian Ocean dominate the air mass over East Africa in July (See  pp12). Rainfall occurs as a result of the convergence of the northeastlies and the southeastlies at the inter-tropical convergence zone (ITCZ). (Boucher, 1975; September is the driest month of the year while April is the wettest over the East African region. The northern and western margins of Lake Victoria experience heavy rain brought about by the presence of the lake itself.
The vegetative cover around Lake Victoria is mainly moist savannah composed of fairly closely spaced trees with re la ti vely large leaves, and a herbaceous stratum of tall, narrow-leaved grasses. The soil is deep yellow or red and sandier than forest soils.  EXPERIMENTAL An opportunity to examine critically · the relationship between isotope ratios and lake environment is The the above n bivalves species were determined by sampling across the growth • 0 11usca The gastropods to be analyzed were sectioned longitudinally increments.
lo~ speed diamond wafering saw in such a way that if a shell has with a 6 to 8 complete turns in the growth spiral, sampling from opposite edges of each turn (180°) in the sectioned shell gave 12 to 16 milligram-size samples. In the bivalves, samples were taken at short intervals along the increments of the growth bands (Figure 3). The results were corrected to 25 °C, and recalculated to the PDB standard.
Since all the samples analyzed were aragonitic, 0.6%0 was subtracted from the values to compensate for the difference between calcite and aragonite (Grossman, 1962).
Yater samples were analyzed according to the procedure of Epstein and Mayeda (1953). This involved the equilibration of approximately 24 cc (STP) of C0 2 with 2.5 g of water in a thermost a ted ba t h for approximately 7 days and the analysis of a 3 cc aliquot of the gas on a V. G. Micromass 602 -D mass spectrometer . To facilitate rapid · the pH may be adjusted to 6 or lower (Mills and Urey, equilibration 1940) .
The water samples were placed in a 25 cc round bot tom flask . h could be connected to a vacuum manifold. The water was frozen in yhlC 1 . uid nitrogen-2-propanol slush bath and the air quickly pumped away. a iq Wa s melted and warmed to room temperature to release any gas The ice trapped during the initial freezing.
The water was then refrozen and pumped for a minute to remove the remaining non-condensible gases. The ice was melted again and commercial cylinder C0 2 , 99. 8% purity, was introduced into the flask to a pressure of about 73.5 cm Hg. After equilibration in a bath thermostated at 31°C for 7 days with frequent shaking, an 3 cc aliquot of the C0 2 was withdrawn on a vacuum line by freezing into ampoules and analyzed on the V. G. Micromass 602-D mass spectrometer using the method described by McKinney et al. (1950).
All mass spectrometric analyses are reported relative to the PDB C0 2 standard gas. The analyses were calculated from the formula The probable error is ± 0 . 01%. Replicate samples measured on the same day are usually reproducible to± 0.05%0 • Standards (usually NBS-20), were always run the same day as the unknowns.

RESULTS
The results of this study have been s ummarized in Figures 4-26 .

---
The environmental factors that influence the isotopic ratios are examined as they will form the basis for the interpretation of our results.
~ctors That Influence Carbon And Oxygen Isotope Ratios The precipitation of calcium carbonate as either calcite or aragonite shows well defined fractionation factors relating the oxygen isotope ratio to the water from which it was precipitated and the temperature of precipitation. The well-known Craig paleotemperature scale (Craig, 1965) is a practical utilization of the temperature control on the oxygen isotope ratio in calcite precipitation. In addition to temperature effects on oxygen isotope ratios, there is a considerable variation derived from the quite different influences on the oxygen isotope ratios in the bodies of fresh water in Africa (covering a range of about 10%0), which may be dependent on latitude, altitude and amount of rainfall (Dansgaard, 1964;, as well as post-precipitative evaporation. The latitudinal, altitudinal and amount effects are usually fairly easy to estimate by interpolation from the isotope map of Africa , but these effects may be augmented or altered by evaporative effects after the rain has fallen.
These evaporative alterations will depend on such factors as regularity of rainfall, humidity, wind speed, air temperature, mass of the water body, efficiency of mixing, etc. This sounds very complicated but the oxygen isotope ratios of gastropod shells from most bodies of water seem to be regulated either by temperature or evaporative effects; in cool, temperate regions, where seasonality is pronounced, temperature controls isotope ratio variations, while in equatorial regions, evaporation seems to be the most important factor in controlling the isotope ratios.
precise quantification of oxygen isotope ratios is not possible, but rough estimates are possible. It should be emphasized that evaporative alterations of isotope ratios are not subtle, i.e. a given shell may incorporate a range of oxygen isotope ratios that covers 2 to 3%0 • Carbon isotope ratios of gastropod shells are more difficult to predict. Temperature effects are insignificant, and the carbon isotope ratios, of course, are not involved in latitudinal, altitudinal or amount effects in precipitation, nor in post-precipitation evaporation.
Nonetheless, there is often a correlation observed between 0 18 0 and o 13 C for shells from a given locality and common age. Scatter diagrams of the two isotope ratios usually have a regression slope of about 2, a long way from the 0.5 one might expect if atomic masses of carbon and oxygen isotopes were governing the formation of calcite or aragoni te.
The origins of these correlations remain obscure, but may be related to changes in nutrients with fresh additions of rainwater to a lake. It has been assumed that "vital" effects might also play an important role in carbon isotope ratios. These are effects that are peculiar to the organism being studied, and presume that carbon other than dissolved inorganic carbon may be contributing to the shell aragonite.
There have been few attempts to investigate the magnitude of vital There is little information on how shell formation takes place, but it is obviously not simply an inorganic precipitation phenomenon, because it produces the thermodynamically unstable aragonite in most species. In a recent study, in which the crystallization of CaC0 3 was controlled under stearic acid monolayers, Mann et al., (1988) observed that in the absence of a monolayer of s tearic acid, rhombohedral calcite crystals were formed as opposed to the va teri te formation in the presence of an organized monolayer. Estimates of the percent carbon derived from metabolic processes as opposed to DIC have been few.  estimate that the marine bivalve ~rcenaria mercenaria, derives about 45% of its shell from metabolic carbon. There have been no estimates from freshwater species.

Reflection Of Environmental Changes In Shell And Yater Isotope Ratios
From the temperature plots in Figure 4, we notice minimal variability in the recorded water tempera tu res, with low temperatures recorded between August and January. The maximum temperature (28.5°C) was recorded for January and March, with the minimum temperature (26.5°C) being recorded in October, November and December (Table 1).
This gives an overall temperature range of 2°C for the sampling period.
Latitudinal effects produce 0 18 0 values near -4 to -5%o at the equator, and extending to -8 to -9%o at the northern and southern extremes of Africa. It is known that evaporative conditions in East Africa are clearly delineated with 0 18 0 values near +1 to +2% 0 • According to the observations of , in cool, high and regular rainfall areas, where 0 18 0 of shell is -3 to -5%o, in situ evaporation is probably much less important than temperature in controlling 0 18 0.
In arid and hot areas, where rainfall is light and variation in temperature is minimal, the 0 18 0 is mainly controlled by evaporation.
At Yinam Gulf where temperature variation is minimal, we notice considerable variation in the 0 18 0 values (Table 2).
According to the rainfall information provided by the meteorological station in Kisumu, and plotted in Figure 5, it rains all year round with highest rainfall in April and May followed by November  In Figure 12, the carbon and oxygen isotope plots parallel each other for the first six months. Not much paralleli s m i s ~bservable in the remaining months of collection. The correlation factor for the carbon and oxygen in this plot is 0.62 (Table 3).
In Figure 13 the Winam Gulf Corbicula carbon and oxygen isotope ratios show parallelism from December 1985 to October 1986 after which there is a lag in the Corbicula plots. This is summarized as 20% correlation between the carbon and the oxygen isotope ratios (Table 3).
In Figure 14, there is no obvious parallelism between the Mutela carbon and oxygen isotope ratio plots. In Figure 15, there is parallelism between the Aspatharia carbon and oxygen isotope ratios with slight noise. The carbon and oxygen correlation factor observed here is 0.35.
The 0 18 0 plots of the Winam Gulf molluscan species are summarized in Figure 16. There is similarity in the isotopic response of all the molluscan shells to environmental changes with slight "noise" here and there which may be ascribed to temporary variations in microhabi tat.
The 0 18 0 values range between -1 to +2%0. This is expected for molluscan species in a body of water near the equator , where there are seasonal differences in rainfall and evaporative influences. Each of the molluscan species will be discussed in turn to examine the extent to which they responded to monthly changes in the Winam Gulf and also how the molluscs collected monthly and analyzed in detail differ from those of the previous month. For Mutela, while some of the 0 18 0 near the aperture closely correspond with the (0 18 0)water of Figure 8, others lag behind. There is, however, a good correlation between their o 13 C plots. It is difficult to estimate the life span of the Mutela from Figures 13-19, because there appears to be an influence of the freshly deposited isotope ratios on the previous isotope ratios deposited. They appear to average out such that previous enrichments or depletions are no longer as pronounced as when they were first deposited. This is to be expected, as not all shell growth takes place at the aperture, but there is an asymptotically thin film added back inside the shell.

Mutela
By inspection and comparison of 0 18 0 plots in Figure 12 versus Oct. 1986 (Figure 9). The 0 18 0 value of February is an interpolation; no sample was supplied for that month. The shell 0 18 0 values lag behind the water 0 18 0 but the lag is eliminated with a one month shift of the water 0 18 0, whereupon the correlation between water and the shell 0 18 0 values becomes obvious (Figure 9).
The September Aspatharia & 18 0 plot ( Figure 19) does not reflect growth continuity from the previous month's plot. It has a & 18 0 plot similar to the May Aspatharia 0 18 0 plot, probably a manifestation of slow growth. By inspection of the isotope ratio plots in Figure 9 and Figures 13 to 19, we estimate that the Aspatharia is giving us 0 18 0 information spanning a period of 7 to 9 months. The lag in the same molluscan species is not uniform. For example, by inspection of Figure   9 versus Figures 13-19 non-uniform growth rate is observed in all molluscan species. This is, of course, not unexpected in a situation where conditions are not uniform over time.

Corbicula
Correlation between water and shell & 18 0 plots appeared elusive ( Figure   7), but with one month's shift the correlation becomes obvious.
However, the first six month's values appear to lag behind yet another month.
We observe the influence of previously recorded isotopic ratios in the growth bands of the Corbicula by subsequent depositions through inspection of Figure 9 versus Figures 13-19 and also Figures 20-26.
From the same plots, the life span of the Corbicula molluscs analyzed from March to May is estimated to be approximately 4 months and 5-6 months for those from June to August.

Bellamya
The first 6 month's aperture & 18 0 plots appear to lag behind the siso b Water Y two months, while the last five month's plots show a correlation without a shift of the water plots ( Figure 6). This may be 165 attributed to differences in growth rate at various times of the year.
T,Jith the Bellamya gastropods, the 0 18 0 values plotted in Figure 6 appear averaged out when compared to the 0 18 0 values from sequential analyses plotted in Figures 13-19.
Changes in the o 13 C values from sequential analyses (Figures 20-26) are not as pronounced when compared to the aperture o 13 C values in Figure 6. The gastropods are estimated to be providing us with 4 months isotopic ratio information. Although there is correlation of carbon isotope ratios between Corbicula and Aspatharia, there is a month or two lag from March to October (Figure 11). Corbicula and Mutela carbon isotope ratio plots partially parallel each other between January and October.
The exclusive enrichment of Mutela o 13 C may be attributed to a vital effect or microenvironmental difference.
Depleted carbon isotope ratios are recorded in the following months: Jan. Feb., May, Jun. Sep. and Oct. These depletions correspond to months with high rainfall and lower temperatures and, therefore, may possibly be attributed to higher input of dissolved C0 2 •  has attributed such depletions to either evaporative rate changes or changes in the photosynthetic activity associated with temperature changes.
From Table 3 we obser.ve that with the exception of Bellamya, a gastropod, which shows a good correlation between o 13 C and 0 18 0, the other molluscan species, all bivalves, show a poor correlation between their o 13 C and 0 18 0. This may be an indication that the molluscs may not be taking their carbon from the same source or that they are not governed by the same factors governing dist r ibution of ca r bon sour ces when it comes to shell accretion. We know that there are several sources of carbon, which gives the molluscs a choice of either DIC or metabolic carbon or different percentages of each.
Since all the molluscan o 13 C values in this study are all fairly close in their isotopic ratios with respect to o 13 C 8 icarbonate (-0%o), we suggest that the percentage of metabolic carbon incorporated into the shells of the Melanoides tuberculata is minimal or that whatever amount of metabolic carbon available, has become enriched through increased lake productivity, a process which causes enrichment of the o 13 C left behind (Deuser et al., 1968, Calder and Parker, 1973, Herczeq and Fairbanks, 1987. The o 13 C values reported in this study fall within the range (-9 to +2). This range seems to be the o 13 C signature range for African lakes in the study of paleoenvironments because of the high frequency with which this o 13 C range has been reported for molluscan species collected from most African lakes. For example, molluscs from paleo-Lakes Erer, and Besaka in the Afar region of Eth i opia recorded o 13 C values ranging from as high as -5 to -1%o (Abell and Williams, 1989). Unpublished data (Abell and Amegashitsi) on the gastropod shell isotope ratios of Lakes Malawi and Tanganyika, recovered from depths to 250', show carbon isotope ratios +0.5 to -2%o• On the other hand, most of the molluscs in streams and rivers of \lest Africa studied in our laboratory gave a o 13 C value range of -19 to -10% 0. (Figure 27) Obviously, there are quite different factors influencing car bon isotope ratios, with no obvious explanations for these differences. In summary, this project offered a chance to study carbon isotope ratios as a function of shell growth, and yielded some rather unexpected results. We do not pretend to have solved all the problems, but only that we have an emerging picture of the carbon isotope sources of the molluscs.
Upon completion of this study we have come to the following conclusions: 1. In general the 0 18 0 and o 13 C plots paralleled one another. While there is no theoretical reason for this correlation, it is clearly evident in most of the shells studied.
2. There is continuity in growth patterns, which can be traced through monthly sequential analyses plots within the same molluscan species.
This continuity gives a way of utilizing mollusc shell isotope ratios for reconstructing short segments of past climates.
3. Heteroscedasticity of scatter plots of o 13 C and 0 18 0 gives information on the extent of seasonal climatic variations.
4. The o 13 C of the Mutela is different from the other three species.
Since the o 13 C values of the other three molluscan species are more negative, it may well be that their utility of metabolic carbon may be more extensive than that of Mutela.
5. Estimation of life span of molluscs is possible from aperture, water and sequential 0 18 0 plots. For example, the life span of the Bellamya is estimated to be 4 months and that of the Corbicula is estimated to be 6 months. We estimate the life span of Mutela and Aspatharia to be in °172 the range of 7-9 months.
Knowledge of the life span of the molluscan species is important, as it determines the extent of utility of the molluscan shell isotope ratios for the reconstruction of paleoclimates and paleoenvironments. In other words, it provides the information span of the isotope ratios of the shell.
6. The use of the same molluscan species in the study of different paleoenvironments will minimize errors in data analysis, as it tends to cancel out vital effects and micro-habitat effects. Ltd. London. pp326. Brown, D. S., (1980 Cosmochim. Acta 32, 657-660.