The chemistry of sediments is not really part of this study, but often "contamination with sediment" is often appealed to to explain especially andesite variations. However, the compositions of most sediments is so odd that any contamination should be obvious, e.g., if the high CO₂ of andesite volcanoes is due to globigerina ooze being carried down the subduction plane, then we would not only expect andesites to also have very high Ca but also high Sr. As most oceanic ridge basalts have a heavy coating of globigerina ooze and calcareous sediment ponds occur between flows and are trapped between pillows and under later flows, sediments must be carried along with them. As marine sediments can vary from highly calcic oozes, to high K pelites, to high silica diatomaceous oozes, it is a real problems as to why island arc andesites are not highly potassic in one area, highly silicic in another and so on. Sediments, being in some sort of chemical equilibrium with sea water are often highly homogenous. A student, (Mme Francine Payette) once carried out a study of Appalchian red and green shales from the Gaspe peninsula and found they were very constant in composition and marked by very high K, Rb, Ba, Th and low Ti etc, in fact though not complete the data look very like potassic andesites. However, the shales were probably derived from a granitic terrain.

|__| Appalachian shales.

Other marine pelites have two end members, illitic clay and silica from diatoms as diatomaceous ooze. These two component mixed give straight line variation diagrams and look at first sight to be igneous, but as silica increases, the K, Rb, Ba decreases, the opposite to igneous rocks. Pacific sediments from the Mohole project, (Murata & Erde, 1964) aso show very high Ba ( 40 - 100 x EMORB) and very low Ti, (0.15 - 0.8 x EMORB). The ODP Leg 123/765 drill hole from the continental shelf of NW Australia, brought up a mixture of globigerina ooze, some only slightly altered basalts, and pelitic clay. All of these highly variable samples show a negative Nb anomaly. Globigerina samples have up to 80% Ca and 3000ppm Sr and no K, compared to the pelites with 4-6% K, 120ppm Rb and less than 200 Sr.

|__| Leg 123 rocks. A number of other ODP cores have rocks with negative Nb, including Kerguelen Plateau (Storey et al,1992).

The Nicobar Rise also has depleted Nb and high Ba. Are these rocks sediment contaminated? Or are they incipient island arcs?

|__| Leg 116/717C located 400 miles south of the island of Ceylon also has depleted Nb and high Rb.

Pelitic greywackes from the Torlesse Terrain in Canterbury , New Zealand, (Roser and Korsch, 1985) have a fingerprint almost identical with a potassic andesite, with negative Nb, low Ti and P, but high Rb, Pb, K etc. As greywackes are commonly formed of andesitic lithic fragments and pelitic mud derived from them this is perhaps not surprising. Do all pelites even from the deep ocean basins have the same fingerprint? The data we have cannot yet answer this, but may exist in the many sedimentary petrology journals we do not usually monitor. However, volcanic rocks in a subterrestrial environment do not retain their parental signature. Once in Guadeloupe in the West Indies we noticed the local natives could barely walk and appeared to suffer from rickets. We analysed the local lateritic soils derived from andesitic ash a few thousand years old, and found that tropical plants had lowered the silica, and warm rains had removed virtually all the Mg, Ca, K, Na, P, Rb, Ba, and REE. What was left was a lateritic mixture of gamma Fe₂O₃ and kaolinite. Time at this point does not allow a full investigation but there is a lot of work to be done.

---

Shales, Marls, Mudstones, Claystones, Siltstones and Carbonates for the USGS PLUTO Database

Thanks to Dr Jeff Grossman of the USGS we have been able to get about 4000 analysed samples of the above, mainly of US continental origin though some are marine. The average composition approximates a basaltic andesite, but weathering processed has enhanced the compositions into the three main components.

1. Carbonates. Two trends are seen one towards pure calcite, the other toward magnesian dolomite. As a result the dolomitic trend correlates with the MgO as dolomite has about 21% MgO to 30% CaO.
2. Illitic Clays. These tend to approach pure illite with 45 – 55% silica, 20 -35% Al2O3 and 6 – 10% K2O. The montmorrilonites, (smectites) are considerably less potassic with 45 -53% silica, usually 15 – 25% Al2O3, minor Mg, Ca and <1% K2O.
   Kaolinite is not common and is mainly 45 – 46% silica and about 40% alumina.
3. High silica rocks may be due to either biogenic silica in marine clays, or to detrital quartz in siltstones.

	Var diagram against MgO. This shows little except the prominent dolomitic rocks, all others have low MgO.
	Variation Diagram against alumina. Both Fe and Ti correlate with Al2O3. The K2O reaches a maximum of about 10% showing that some of these rocks are pure illitic clays. Extreme carbonates are low in Al2O3.
	Variation diagram / silica. This shows a strong antithetical relationship with carbonates. AL2O3, Fe2O3, TiO2 and K2O show a maximum near 50% silica coinciding with illite.
	Alumina vs K2O. These seem to show very positive upper and lower bounds, which MAY mean two phases are present.

The overall picture for fine grained sediments looks simple in overview, there may be many complexities in detail, for example a group of phosphatic sandstones have up to 10% P2O5. The next step is to isolate the groups and see which trace element follow the main groups.

However this is a text on IGNEOUS geochemistry so we may not be able to devote the time for a complete coverage, but at least we can show the distributions and main trends.