direct combustion of shell material is easier and less time consuming than acidification. In museum collections bivalve shells are traditionally dry stored, whereas soft tissues are preserved in 70% ethanol, sometimes after fixation with 10% formalin. However, often the whole animal is preserved in ethanol and shells are not stored separately. For the application of these preserved specimens in the investigation of past d N values it is essential to know if liquid preservation methods have an effect on the d N values of bivalve shells and if this effect is predictable. The effects of liquid preservation on the d N values of biological tissues have been examined in a variety of For testing the in uence of CaCO 3 content on d N measurements, different mixtures of acetanilide with inorganic pure CaCO 3 were made, containing between 0 and 10.

4 weight % N. Powder calcite samples were loaded into 4 _ 6 mm tin cups and weighed. d N values were measured using an elemental analyzer coupled via a CONFLO III to a ThermoFinnigan Delta V t isotope ratio mass spectrometer. An inline soda lime CO trap was used to scrub MLN8237 CO 2 from the gas stream entering the gas chromatography column of the EA. IAEA N1 was used as a standard, with an accepted value of 0. 4 _ 0. 2% Long term. standard reproducibility is better than 0. 1% for samples nature, even samples between 5 and mg N provided reasonable data. There is also an upper limit to the amount of shell material that can be loaded into the EA, but this was not evaluated here.

This method is robust because calcium carbonate com pletely decomposes around 8258C and the ash combustion Nilotinib in the EA was around 10208C, therefore, all N should be released from the matrix and carried to the IRMS. Moreover, previous studies have used an EA IRMS system to combust Fig. 2 that the narrow and near symmetrical peak shapes are similar for both shell carbonate and synthetic mixtures, which suggests that both matrices are reacting similarly in the EA IRMS. We therefore argue that it is possible to measure carbonates for d C analysis. It is clear from the traces in larger than 30 mg N. d N values are expressed in % vs. atmospheric nitrogen. Pure synthetic CaCO 3 had peaks similar to empty tin cups, empty tin cup 1/4 0. 49 Vs) and therefore did not contribute much to the calculated delta values. The acetanilide standard had a d N value of 2.

12 _ 0. 13% when it was run without PI3K Inhibitors synthetic CaCO 3 and was _2. 02 _ 0. 11% when it was run with 98. 4 to 66. 8% CaCO 3. These values are not significantly different. In addition, during a preliminary trial, we ran 0. 4 mg of the IAEA N1 ammonium sulfate SO 4) standard in. 72 mg CaCO 3 and found no offset from N1 standards run without CaCO3. Our results show that samples with as little as 20 mg N can provide accurate d N values. Prior acidification is not required to eliminate the carbonate matrix to produce accurate results, as has been previously reported. It should be noted that mollusks with very low organic matrix in their shells may require a pre concentration step to reduce the poorer precision of small samples. However, considering the large fractionations associated with nitrogen isotopes in Figure 1.

d N values for acetanilide mixed with 66. 8 to 98. 4 weight % synthetic CaCO 3 powder and pure acetanilide. The solid line represents the mean value of _2. 02% for data above mg N. The error bar represents the 1s of _0. 11%. wileyonlinelibrary. Ion Channel com/journal/rcm Copyright 2011 John Wiley & Sons, Ltd. Rapid Commun. Mass Spectrom. 2011, 25, 675 680 Letter to the Editor tissue is subject to metabolic turnover and is thus repre sentative for a specific time window, see e. g., Paulet et al,. while the shell samples averaged at least 1 year of growth. This makes comparing soft tissues with shell organic matrix difficult. However, as shown in Delong and Thorp, tissues with slower turnover rates, such as the adductor muscle, are better for comparisons with metabolically inactive shells.

Most previous studies that report differences between skeletal d N and soft tissue d N do not take the different amounts of time being averaged into consider ation. Moreover, HSP many studies compare whole body tissue d N data to shell data while it is known that different organs can have quite different d N values, sometimes as much as 5% in the same animal. This may explain why Dtissue shell values for the same species of clam range from 0. 2 to 2. 4%, see ODonnell et al.. Soft tissue d N data from M. edulis specimens collected at three different periods in 2002 from Knokke show significant changes throughout the year, which would be averaged in the shell samples we analyzed. Taking the average of these 25 soft tissue data results in a Dtissue shell value of _1. 5 _ 1. 0%. In the future it is important to compare tissues and shells that represent the same time period.