Relations of G. bulloides and G. glutinata
Only few samples exist in the CLIMAP Holocene data in which both species do not co-occur (Fig. 40). Their relative abundances show some inverse correlation. Globigerina bulloides is more abundant in central upwelling zones and areas of high productivity while G. glutinata is more frequent at their margins and in central ocean areas. This is well expressed in the biogeographic maps of Bé and Hutson (1977) in the area of upwelling in the Arabian Sea offshore from Somalia. The central area is occupied by abundant G. bulloides, while a belt with abundant G. glutinata exists in the marginal upwelling zone (see also Brock et al., 1992). Globigerina bulloides feeds on algal prey (Lee et al., 1966), while G. glutinata has more specific preferences for diatoms (Hemleben et al., 1989). Such different feeding strategies may explain why both species are related to productive environments but tend to occupy different zones, probably related to the phytoplankton bloom succession (dinoflagellates - diatoms).
Relations of G. calida and G. siphonifera
It is difficult to argue about possible taxonomic uncertainties in the counts of G. calida and G. siphonifera and subsequent problems in the interpretation of their relations with the physical environment. CLIMAP micropaleontologists have made serious efforts for quality control of their micropaleontologic data and taxonomic standardisation between the different members of the group. Other species, which are difficult to distinguish in their morphology (e.g. G. falconensis and G. bulloides) have distinctly different adaptations and suggest that the similarities in the ecologic pattern between G. calida and G. siphonifera are real. This problem may suggest to include both species in one taxonomic category and demands for taxonomic research.
Relations of G. rubescens and G. tenella
Globoturborotalita tenella is distinguished from the generally pink-colored G. rubescens by a secondary aperture on the last chamber. Pre-adult stages of G. rubescens and G. tenella are difficult to distinguish in their morphologies and taxonomic discrimination is made more difficult by the existence of a white form of G. rubescens in bottom sediments of temperate regions (Hemleben et al., 1989). Morphologic similarities and the nearly equal relations with the physical environment seen in G. rubescens and G. tenella may suggest ecophenotypes rather than different species. In other species variants are consistently more differentiated in their preferences compared to G. rubescens and G. tenella. Both species require taxonomic and ecologic research.
Relations of G. sacculifer and S. dehiscens
Bé (1965) considered S. dehiscens as a deep-water form of G. sacculifer in a terminal (reproductive) stage. In the laboratory, however, Glbigerinoides sacculifer was observed during gamete release and did not develop the "S. dehiscens" form (Hemleben et al., 1987). Other authors emphasize morphological differences in juvenile stages of the two species (Hemleben et al., 1989). Pattern in the plots of relative abundances vs. physical parameters, however, is very similar for G. sacculifer and S. dehiscens. Both species differ drastically in their relative abundances and comparisons of their relations with the physical environment are difficult. The correlation coefficients of their relative abundances computed with various regression methods are all well below 0.1. This, however, may be caused by the low relative abundance of S. dehiscens (< 5 %) which causes statistical uncertainty due to counting error in the data. Potentially, S. dehiscens may occupy a deep-water habitat with a biogeographic distribution similar to that of G. sacculifer. The possible existence of vertical clines in phenotypes, in contrast to the commonly observed geographic clines in other species, motivates more research on relations of the two species.
G. crassaformis and G. truncatulinoides
The origin of G truncatulinoides as a species, about 2.8 - 2.9 My ago, was analysed in a morphometric study by Lazarus et al. (1995). They suggested a sympatric mode of evolution, in which the differentiation and "geographic isolation" of ancestor (G. crassaformis) and descendant species (G. truncatulinoides) occurs through the occupation of different niches (e.g. depth habitats, seasonally different cycles, etc.) in the same biogeographic region (see discussion by Lazarus et al., 1995). The substantially different specialisations of both species seen in the relations with the physical environment (Figs. 19 and 24) support this view.
Table 1 lists those species which dominate at least one of the 461 samples used in this study. Only six species, however, can be considered as dominant species on a biogeographic scale: N. pachyderma, G. inflata, G. bulloides, G. ruber, G. glutinata, and G. menardii. Broad relations with sea surface temperatures in distinct biogeographic provinces exist for N. pachyderma in the polar and subpolar provinces, G. inflata in the transitional province, G. ruber in the subtropical and tropical province, and G. menardii in the warm tropical province. The latter species is not commonly a dominant species and may reflect selective dissolution (Kipp, 1976). Globigerina bulloides and G. glutinata dominate in productive high latitude environments and areas of upwelling. The biogeographic relations suggest different preferences of the two species for the central and marginal oceanographic and biologic conditions in such areas.
importance of the vertical water structure
Some species show most pronounced relations with the vertical temperature or density gradients, e.g. G. truncatulinoides, G. hirsuta, and T. quinqueloba, among others. On a biogeographic scale, the boundary between water masses with vertical
temperature gradients of more or less than 6 ûC in summer, seems to be the major limit between high and low latitude faunas in planktic foraminifera. This is well seen in the ecologic ranges of e.g. T. quinqueloba (Fig. 39) and N.
pachyderma (Fig. 33), which have their southern limits near this boundary and G. ruber (Fig. 15b), G. menardii (Fig. 22), and P. obliquiloculata (Fig. 36), which have their northern limits at this boundary. The limit corresponds with about
40û latitude in the North Atlantic and about 30û latitude in the Indian Ocean (Fig. 4). Other physical parameters do not show this clear separation between ecologic ranges of low latitude and high latitude faunas.
central cool productive
marginal cool productive