So far as the color reactions are concerned, it should be noted that in many instances rosin essences and rosin oils behave like pine tar oil. This is not surprising since products of the dry distillation of turpentine are contained in pine tar oil as well as in the distillate of colophony.

The Herzfeld test1) with sulphurous acid is carried out in the following manner: In a test tube the oil to be examined is shaken with an equal volume of a solution of sulphurous acid. In the presence of pine tar oil the oily layer is colored yellowish green. It is claimed that the presence of 10 p. c. of pine tar oil in turpentine oil can thus be detected.

Another test is recommended by the same author. A small piece of potassium hydroxide is covered with the oil to be tested. The presence of pine tar oil is revealed by the brown coloration of the potassium hydroxide which soon results. H. Wolff2) has modified this test by shaking 0,5 to 1 cc. of potassa solution (d 1,3) with the oil, heating the mixture on a water bath for from 2 to 5 minutes and then adding 3 cc. of water to effect a separation of the emulsion. Pine tar oil colors the aqueous layer brown, turpentine oil produces only a slight coloration or none at all.

Wolff mentions two other pine tar oil tests:

5 cc. of oil are boiled with 5 drops of nitrobenzene and, after the addition of 2 cc. of 25 p. c. hydrochloric acid, are boiled for another 10 seconds. Pine tar oil is colored brown, the hydrochloric acid is colored brown to black (Lyon's reaction with concentrated hydrochloric acid). Turpentine oil produces a much lighter color.

If to a mixture of 4 cc. each of ferric chloride solution (1 : 2500) and of potassium ferricyanide solution (1 : 500) 2 or at most 10 drops of the oil to be tested are added and the mixture thoroughly shaken, pine tar oil rapidly produces a decided precipitation of Prussian blue, whereas turpentine oil produces a noticeable precipitate only after hours.

1) Zeitschr. f. off. Chem. 10 (1904), 382; Chem. Zentralbl. 1904, II. 1770. 2) Farben Ztg. 17 (1911), 21, 78; Chem. Ztg. Repert. 36 (1912), 64.

For the like purpose C. Piest 1) proposes the following reactions: 5 cc. of acetic acid anhydride are shaken with 5 cc. of turpentine oil in a test tube. While shaking and cooling, 10 drops of concentrated hydrochloric acid are added. After thorough cooling another 5 drops of concentrated hydrochloric acid are added. The liquid again becomes warmer and a clear solution results. When treated thus turpentine oil remains water white, pine tar oil turns black.

Old turpentine oils should in all cases be distilled before being tested.

The test suggested by Valenta2) is as follows: If equal volumes of a 1 p. c. gold chloride solution and turpentine oil are shaken in a test tube, heated in a waterbath for a minute then removed and again shaken, the turpentine oil reveals a separation of gold in the oily layer only. The solution itself is not decolorized. The pine tar oils, whether rectified or not, likewise pinolin, decolorize the gold solution completely. Pinolin decolorizes the solution most rapidly.

In order to distinguish between pine tar oil and turpentine oil, Utz1) employs the following test: Equal volumes of the oil and tin chloride solution, prepared according to the German Pharmacopoeia, are mixed. Austrian turpentine oil colors the reagent yellow, but itself remains colorless; Greek oil imparts an orange color to the reagent, and is itself colored yellow; American turpentine oil also colors the reagent orange and itself turns yellow. All varieties of pine tar oil, however, colored the tin chloride solution a raspberry red, the oil itself mostly yellow. In some instances the oil was likewise colored raspberry red. In a few instances the well known brown to black coloration was produced. Positive results were also obtained with mixtures of pine tar oil and turpentine oil.

Detection of Various Hydrocarbons. Benzene and its homologues, "solvent naphtha", "Schwerbenzol", all reduce the bromine value (normally 220 to 240) of the turpentine oil. According to Herzfeld)4 these additions are recognized by treating 10 cc. of oil with 30 cc. of concentrated sulphuric acid while cooling the mixture. Pure oils are thereby dissolved leaving not more than 1 cc. of residue. If a larger residue remains which can be brought into solution by subsequent shaking with fuming sulphuric acid, the presence of benzene hydrocarbons is probable. According to Marcusson1) this method is not reliable. While it is true that, in the cold, benzene and its immediate homologues are attacked but little by concentrated sulphuric acid, complete sulphonation results readily in the presence of large amounts of turpentine oil, so that the residue of adulterated oils is no greater than that of some pure oils. Herzfeld-), however is of the opinion that his method can be used if the mixture is well cooled and less sulphuric acid is used. Nevertheless'), Marcusson, after another test, adheres to his original opinion. The conflict of opinion concerning the utility of the sulphuric acid method may possibly be explained here, as on a previous occasion (comp. p. 39), in that the several analysts employed sulphuric acid of different strength in their tests.

1) Chem. Ztg. 36 (1912), 198.

2) Ibidem 29 (1905), 807.

3) Chem. Zentralbl. 1905, I. 1673.

4) Zeitschr. f. off. Chem. 9 (1903), 454; Chem. Zentralbl. 1909, I. 549.

The nitric acid method described on p. 37 is recommended by Marcusson4) as the best for the detection of benzene hydrocarbons. The nitric acid solution resulting is utilized for testing for benzene, toluene, xylene etc. their nitrocompounds being dissolved therein.

"Into a l/2 1 measuring flask, the neck of which is calibrated into 1/10 cc. for 10 cc. and which contains 150 cc. of water, the nitric acid solution is poured. The mixture is heated for 1/4 hour on a water bath in order to render the reaction products of the turpentine oil as water soluble as possible. After cooling, the solution is set aside for several hours, if necessary over night, in order that it may become clear. If reddish-brown oily drops (nitro compounds) have separated either on the bottom or at the surface, benzene hydrocarbons were present. If, however, only small amounts of resinous matter, usually floating on the surface, are observed, benzene hydrocarbons need not be suspected.

1) Chem. Ztg. 33 (1909), 966.

2) Ibidem 33 (1909), 1081; 34 (1910), 885.

3) Ibidem 34 (1910), 285.

4) Ibidem 36 (1912), 413, 421.