I'm having some trouble with how to determine the representative mass value of a prepared soil sample for calculating its $\ce{Cl-}$ concentration in ppm (mg/kg).
Recently I carried out a gravimetric analysis for determining the concentration of $\ce{Cl-}$ in a soil sample by the addition of $\ce{AgNO3}$. The sample was prepared from 10 g of soil, which were filtered with 50 mL of distilled water. Then, an aliquot of 15 mL was taken from the filtrate. The aliquot was centrifuged in order to separate the solid phase from the liquid phase. The liquid was transfered to a volumetric flask and diluted to 50 ml with distilled water. This final volume was the one used for the analysis. The gravimetric processes were developed normally and at the end it turned out that the produced mass of the AgCl precipitate was 0.003 g and by stoichiometry I calculated its content of $\ce{Cl-}$ (which resulted to be 0.00074204 g).
Since I want the $\ce{Cl-}$ concentration in terms of mg/kg, I expressed the mass of the aformentioned ion in mg but my conflict comes when trying to establish the mass of the used soil sample (in kg) because of the dilutions. As far as I'm concerned, the preparation of the sample could be mathematically arranged as follows: $$\frac{\pu{10 g}\,\text{soil}(\frac{\pu{1 kg}}{1000\,g})}{\pu{50 ml}\,\text{filtrate}}\cdot\frac{\pu{15 ml}\,\text{aliquot}}{\pu{50 ml}\,\text{dilution}}=\pu{6e-5 kg ml-1}$$
but as you can see, the dimensional analysis doesn't result in just kg. I don't really know where the problem is, but I intuit that either I'm missing something or adding more of something, or I'm just having an erronous mathematical concept about the sample preparation.
It will help me a lot if you can help me to identify my mistake.