These figures alone are sufficient to show what variable quantity is the puer employed in a tannery. The mineral matter is composed as follows:-- --------------------+----------------------+-------+------ Soluble in water {|NaCl and Na_{2}SO_{4} | 1·37} | 4·00 {|Sodium phosphate | 2·65} | | | | {|Earthy phosphates |80·37} | Insoluble in water {|Ferric phosphate | 2·09} |94·93 {|CaSO_{4} | 4·52} | {|Silicic acid | 7·94} | --------------------+----------------------+-------+------ The ammonia compounds combine with the lime remaining in the skin, but the constitution of these compounds was unknown. The puer was examined for ammonia by Schlösing’s method, 50 c.c. being mixed with 50 c.c. milk of lime placed under an air-tight bell-glass, together with an open vessel containing a measured quantity of standard acid, for 36 hours. No NH_{3} was given off.[115] [115] This is not invariably the case: other samples of puer examined have given off NH_{3} when similarly treated, but only in small quantity. A similar instance has been observed by Rideal and Orchard in the liquefaction and decomposition of gelatin by B. fluorescens liquefaciens (The Analyst, October 1897), where the quantity of ammonia produced was insignificant, amounting even after 16 days’ incubation to only 0·168 grm. of N per 100 c.c., corresponding to 0·204 grm. of ammonia. The mixture of dung and Ca(OH)_{2} was filtered and a perfectly clear filtrate of a golden-yellow colour obtained. This contained free amines, calcium salts of volatile and non-volatile acids. It was distilled until the distillate was no longer alkaline, and made up to 500 c.c. 50 c.c. required for neutralisation 5·7 c.c. of (N/100) H_{2}SO_{4}, equivalent to 0·1938 grm. of NH_{3} per litre of original puer, being the amines set free by Ca(OH)_{2} in cold solution. A further 50 c.c. was distilled with NaOH, the amines given off neutralised 8 c.c. of (N/1) HCl, equivalent to 2·72 grm. of NH_{3} per litre. The neutralised liquid was evaporated to dryness; the residue, consisting of amine hydrochlorates, amounted 0·27 grm. A qualitative test showed both primary and secondary amines to be present. The action of these amine hydrochlorides was tried upon skin, pieces of limed sheep grain previously washed in water being used. The solution was maintained at 35°C. In the first experiment 0·27 grm. was dissolved in 100 c.c. of water. In two hours the skin was considerably reduced, but had not the touch of a “puered” skin. It tanned well and was a good colour, showing that the lime had been removed, but the leather was not sufficiently soft.