The reciprocity between coiling direction and dimorphic reproduction in benthic foraminifera

There are various opinions as to what parameter influences the coiling directions in foraminifera. “Do microspheric and megalospheric generations have different coiling ratios?” is an unanswered question in foraminiferal studies. Per view of this, an attempt is made to study the relationship between mode of reproduction (sexual/asexual) and coiling direction (dextral/sinistral) in the benthic foraminiferal species Rotalidium annectens (Parker & Jones). Proloculus (initial chamber) size is taken as an indicator of changes in reproductive behaviour. The present study is based on the observations made on 17722 specimens of this species from 186 samples, obtained from 3 sediment cores (representing a time span of about 9,500 years) from the shallow water region off Karwar, west coast of India. The results indicate an inverse relationship between mean proloculus size and dextrality (% of dextral forms) which is statistically confirmed. The correlation value (r = −0.57) between the two parameters is above the level of significance at 99% level. Therefore, it is inferred that microspheric generation (smaller proloculus) prefers dextral coiling and coiling in benthic foraminifera appears to be influenced by mode of reproduction.


INTRODUCTION
Innature many organisms suchas gastropods, bacterial colonies (Bacillus rnycoides), the spiral vascular conductors of some plants and the spiral distribution of leaves, show differences in coiling direction. Many trochospiral foraminiferal species possess asymmetric tests in which chambers are arranged in a cone-like spiral. This spiral may coil either in an anti-clockwise, sinistral (left handed) or in a clockwise, dextral (right handed) direction.
These changes in coiling direction are the most commonly studied morphological variable exhibited by foraminifera (reviews by Scott, 1974 andKennett, 1976;Hallock & Larson, 1979;Lena, 1981;Vincent & Berger, 1981;Hornibrook, 1981Hornibrook, , 1982Kalia & Chowdhury, 1983;Weaver, 1983;Duprat, 1983;Healy-Williams et al., 1983;Hallock, 1986;Caralp, 1987;Corliss & Chen, 1988;Renzi, 1988;Boyle, 1989;El-Nakhal, 1990;Collins, 1990 and many others) with emphasis on their value as a tool for local stratigraphiccorrelation and/or paleoclimatic studies. For some species arguments have been made in favour of their use as paleotemperature indicators, although a number of contradictions have been observed and they fail to consistently correlate with cold and warm periods (Thiede, 1971;Parker, 1971;Olsson, 1974). For example, higher dextral/sinistralratios in the planktonic foraminiferal species G. truncatulinoides has been considered to be an indicator of higher temperature in the North Atlantic (Bandy, 1960;Takayanagi et al., 1968;BC, 1969) and in the Pacific Ocean (Parker & Berger, 1971) although off Portugal and Morocco mostly dextral G. truncatulinoides were found in glacial sediments (Thiede, 1971). Wollin et al. (1971) had the same opinion. This could be due to the disagreement concerning the temperature conditions of species, as some species appear to be more tolerant of low temperature in the southern hemisphere than in the northern hemisphere (Be, 1969;Boltovskoy, 1969Boltovskoy, ,1970B6 & Tolderlund, 1971). Kennett & Huddlestun (1972) indicated that the same species may show more one response to a particular environment.
Similarly, different species may show different responses in the same region. To quote a few, in Red Sea cores, Herman (1965) found sinistral G. bulloides as an indicator of cold water, whereas, sinistral G. sacculifer indicated warmer conditions. In the North Atlantic, G. quinquelobn showed no preference between dextral and sinistral, G. pachyderm displayed a seasonal alternation in preferred coiling direction, whereas G. truncatulinoides had a distinctly preferred coiling direction regardless of season or temperature. (Tolderlund & BC, 1971).
Discrepancies have also been reported in benthic species such as R. beccarii where sinistral forms were found to be in abundance in cooler water by Longinelli & Tongiorgi (1960). However, observations in other areas have not confirmed this relationship (Boltovskoy & Wright, 1976). Similarly, Brooks (1967) and Malmgren (1984) did not find any clear relationship of coiling direction to the environment. These contradictions lead to various alternative explanations. Lipps (1979) summarised them as (i) salinity, (ii) watermass, (iii) seasonal effect, (iv) test size, (v) water depth, (vi) asexual-sexual generation, (vii) water temperature, (viii) sorting processes, (ix) differential predation, (x) geographically isolated gene pools, (xi) different species, (xii) ice habitat, (xiii) water density, (xiv) paleomagnetism, (xv) evolutionary phenomena, (xvi) reproductive strategy. He further stated that insufficient evidence was available to confirm or disprove these suggestions. However, he selected only two hypotheses for detailed discussion: one was temperature control, as it was used very commonly and the other was the difference in reproductive strategy related to productivity of water. Even the recent attempt by Collins (1990) to study the relationship between temperature and coiling direction by eliminating factors of life cycle stages and possible ontogenetic changes, has yielded only partial success, as dextrally coiled Bulimina marginata and B. aculeata were found strongly associated with warm temperatures but failed to show any consistent relationship of sinistral dominance in cold temperatures. Nevertheless, no significant attempt has been made to study the relationship of coiling direction with reproductive behaviour (asexual-sexual generation) in fcraminifera.
In view of the above, we decided to investigate the relationship between coiling direction and reproduction. For this purpose an attempt is made to study the relationship between coiling direction, expressed in terms of dextra1ity:percentage of dextrally coiled specimens in a population and dimorphism, expressed in terms of mean proloculus (initial chamber) size, a phenomenon related to sexual and asexual reproduction.
This study is based on a large number of specimens of the benthic foraminifera1 species Rotalidium annectens (Parker & Jones) (Figl), obtained from three sediment cores collected off Karwar, central west coast of India. This species has been selected due to the fact that, (i) it exhibits dimorphism (Figs la, 2a, 2b) (Nigam, 1988) and dextral/sinistral coilings (Fig. lb); (ii) its large size and (iii) its abundance in shallow marine sediment. The large size (0.30-1.3Omm) and abundance ensure statistically reliable measurements of proloculus size as well as calculation of mean proloculus size ( M E ) and percentage of coiling ratios.

MATERIALS AND METHODS
The three cores were collected during three different cruises in the Arabian Sea off Karwar near the mouth of the Kali river (a) the first core, GV 3713 (at 14" 53.1"; 73" 57.9'E), 1.16m long and 20m deep was collected during the 150th cruise of R.V. Gaveshani. The core was sampled at 2cm intervals ( at a water depth of 22m. Only the portion below 4.50m in this core was utilized and sampled at 5cm intervals (Fig. 3).
These cores represent recent Holocene time as one sample (300-305cm) from core SK 27B/8 and two samples (455-460 and 600-605cm) from core SK 44/13 (dated by I4C method using Accelerator Massspectrometer) show an age of 3,510+60; 6,200+90 and 8910 +160 years BP respectively. 198 samples from the three different cores were washed througha 60pmsieve and ovendried. Theresultingspecimens were kept in dorsal view and the direction of progression of new chambers noted. In a dextrally coiled specimen, new chambers are added in a clockwise manner, while anticlockwise addition gives a sinistrally coiled specimen. The ratio of these two forms can be counted in any assemblage.
Features associated with reproduction need more attention. it is well known that the shape, size and proloculus size of tests of foraminifera belonging to the species are different. This dimorphism is related to reproduction and the two forms are known as megalospheric and microspheric. Size is the simplest to measure and some workers (Thiede, 1971;Steuenvald & Clark, 1972;Vella, 1974) noted a tendency in several planktonic species for the coiling direction to be related to the size of the specimens. However, dimorphism in planktonic foraminifera has yet to be demonstrated.
In benthic foraminifera, at least in the case of Rotalidium annectens (Parker & Jones), proloculus size can be easily measured. Moreover, the size of the proloculus and the direction of coiling (which is decided by addition of a few chambers soon after formation of the first chamber) will remain unchanged with the growth of specimens, and hence will be independent of size. Therefore, for the present study the proloculus size is taken as a factor representing the modes of reproduction.
Computations for correlation coefficients and regression equations between dextrality and mean proloculus size were carried out on a ND 520 Computer at the Computer Centre of the National Institute of Oceanography. The levels of significance were determined from Table 7 of Fisher & Yates (1963).
The total number of specimens measured for mean proloculus size at different levels of cores are given in Tables 1 -3. However, a few samples showing extremely poor (less than 20 specimens) occurrence of Rotalidium annectens were excluded from the present analysis due to paucity of the material. The final summary of data is given in Table 4.

RESULTS
This study is based on a total of 17722 specimens from 186 sediment samples from three cores representing a time span of about 9,500 years. Out of 17722 specimens, 11.26% of all specimens possessed dextralcoiling. The general range of proloculus size is 0.025 to 0.125mm. The mean proloculus size of various samples shows a range of 0.040 to 0.068mm. It is important to notice that in each core the average mean proloculus size of specimens showing dextral coiling is invariably smaller than those coiled sinistrally (Table5). This shows that dextrally coiled specimens are associated with smaller proloculus, which is a characteristic of sexually formed microspheric forms. Similar results are obtained by plotting the down core variations of MPS and percentage of dextral specimens (Figs 4, 5, and 6). These curves show the absolute as well as five point moving averages at every data point. In each core the majority of the prominent peaks in the curves of percentage of dextral forms can be correlated with troughs in mps curves. This further indicated that dextrality in benthic foraminifera Fig. 5. Down core variations in mean proloculus size and percentage of dextral forms in core SK 44/15. Line joining the squares is profile of raw data, whereas line joining the black circles is 5 point moving average. is inversely correlated with mean size of proloculus.
We have computed the correlation coefficient (r) values between the percentage of dextral forms and mean proloculus size for every core ( Table 4). The results show a consistent inverse relationship and all the (r) values are above the level of significance at 99% level (calculated as per Table 7 of Fisher & Yates, 1963). The collective plotting of MI' S and percentage of dextral forms from all the three cores (Fig. 7) also exhibits a significant inverse relationship (r = -0.57) at 99% level of significance (r = <0.25). Boltovskoy & Wright (1976), while listing the significant unanswered or poorly stated questions, have raised the question "Do microspheric and megalospheric generations have different coiling ratios?". This suggests a genetic control for coiling direction.

DISCUSSION
It was already noticed by earlier workers (Thiede, 1971;Tolderlund & Be, 1971;Steuerwald & Clark, 1972;Vella, 1974) that coiling direction may have its origin in reproductive MPS (mm) 0 =L Fig. 6. Down core variations in mean proloculus size and percentage of dextral forms in core BV 3713. Line joining the squares is profile of raw data, whereas line joining the black circles is 5 point moving average. strategies. Unfortunately, the ratio of megalospheric and microspheric forms is rarely determined in planktonic foraminifera (Thiede, 1971) as it is not very easy to differentiate these forms even if dimorphism exists. On the other hand, thisis less problematicinbenthic foraminifera asmany benthic species show definite proof of dimorphism which can be quantified by measuring proloculus size.
However, the examination of the relationship between reproductive mode and coiling direction in benthic foraminifera1 species has been more limited, perhaps due to the small proportion of benthic foraminifera in deep sea sediments.
The results of the present work indicate very clearly that mean proloculus size is inversely proportional to percentage of dextral forms (Figs 4, 5 and 6). This relationship has also .." . been confirmed through statistical analysis of the data for regression equation and correlation coefficient (r) ( Table 4; Fig. 7). It further suggests that microspheric generations have a tendency for dextral coiling, whereas sinistral coiling is favoured by megalospheric generations. Our conclusions are in agreement with the results of culture experiments onRosalina floridana by Lee et al. (1963). They found that dextrally and sinistrally coiled forms occurred in both generations (microspheric / megalospheric) but sinistral coiling was found in 63 out of the 66 measured megalotypic (megalospheric) individuals, whereas the reverse situation occurs in the microtypic (microspheric) generations in which 28 out of the 34 measured specimens coiled dextrally. These observations were on specimens from a common environment. A similar relationship was also noticed by Myers (1940), who found that microspheric tests coiled dextrally, but megalospheric tests coiled sinistrally in Discorbis patelliforrnis.
In view of the foregoing account, it may be summarized that coiling direction in benthic foraminifera1 species, including R. annectens (Parker & Jones) shows relationship with reproduction.
It is likely that reproductive mode (or even proloculus size independent of reproductive phase) is affected by environmental fluctuations. Over the Indian region, the increase in precipitation around 9,000 years BP (Prell et al., 1990), 6,000 years BP (Singh et al., 1972), 4,000 and 3,500 years BP (Nigam & Khare, 1992) must have contributed to the lowering of the salinity in coastal areas. In our results, these periods of low salinity are marked by relatively higher MPS values at approximately 3.00m (ca. 3,500 years BP) and 4.00m (ca. 4,000 years BP) depth incoreSK 27B/8 (Fig. 4 ) and around4.6m (ca. 6,000 years BP) and 6.00m (ca. 9,000 years BP) in core SK 441 13 (Fig. 5).
Similarly, another period of climatic aridity around 2,000 years BP has been noticed elsewhere (Bryson & Swain, 1981) which is reflected in this study where a marked low value of MPS can be seen around 1.30m down the core which corresponds more or less to the dry phase. This is in agreement with Nigam & Rao (1987) who noticed an inverse relationship between salinity and MPS of R. annectens in the coastal Arabian Sea. The observation that coiling ratio also varies in an inverse fashion, through the study section may imply an independent response to the same environmental signal.
The present study may be taken as a strong signal of the existence of a possible relationship between reproductive mode (or proloculus size) and coiling directions in benthic foraminifera. Since this study is based on a single species, for more general results the relationship should be tested in many other benthic species from different areas and also in culture experiments.