North Island. Besides being dug out of the ground, a good deal of it is gathered fresh from the standing trees, where it has oozed out of the bark. When dug from the ground it is scraped free of earth and shipped away in that condition, no attempt having been made to deal with it here to my know-ledge, except by the syndicate mentioned above.

I have looked up the “Transactions of the New Zealand Institute,” 1869 to 1900, and found next to nothing about the chemical properties of this resin. I found a good many articles on the behaviour of other resins in the back numbers of the “Journal of the Chemical Society,” of which the years from 1870 to 1872 and 1882 to 1900 were accessible to me; but, unfortunately, the number which a dictionary of chemistry gave as containing an investigation of kauri-resin was not to be found.

Among the books which I have most used are Ostwald's “Physico-Chemical Measurements”; Allen's “Commercial Organic Analysis,” 1886; “Lubrication and Lubricants,” Archbutt and Deeley, 1900; “Destructive Distillation,” Mills, 1886; and Sutton's “Volumetric Analysis.”

There are two main varieties of kauri-resin—the fresh resin as it hardens on the tree, which is generally white or transparent, and the fossil resin. The appearance of the latter ranges from transparent to opaque, from white to yellow. I used for my distillation the purest fossil resin, obtained from carefully scraped lumps roughly powdered. For finding the bromine-absorptions I used the method of Mills,^* first making sure of the results by finding the absorption of a known oil (olive). My reason for using the bromine method rather than the iodine was that a solution of bromine in carbon-disulphide is not so complicated to make up as the solution of iodine-chloride. It is doubtful, besides, whether any results could be obtained from a resin in a reasonable time with the iodine solution. Mills has determined the absorption of kauri-resin to be 108–2. He has not stated the variety of resin which he used, but I got values much higher than this for all varieties. Mills has also noticed that hydrobromic acid is always formed, but gives no figures for it. I have given figures in most cases. The method I adopted was to decant the aqueous part of the solution after the reduction with Na2S2O8, wash the CS with water two or three times, adding the water to the part first poured off, then add methyl orange and titrate with decinormal NaOH. After this was done I added phenolphthalein and titrated with decinormal NaOH for organic acids. The values found were nearly the same with very

different weights of resin, so it is evident that these acids are soluble to a limited but definite degree in water, as I used nearly the same volume each time for the determination of HI. After the bromine had been absorbed, and before titrating with Na2S2O8, I added excess of KI. After the starch blue had been bleached with Na₂S₂O₃ it returned on standing in the light, and this went on slowly, in some cases for many days. When kept in the dark there was practically no change. This residual action amounted to more than 0.6 c.c. decinormal Na₂S₂0₈, and was seen both in the resin and in the oils distilled from it. I took some of the reduced solution before the blue had had time to reappear, added litmus, and neutralised with decinormal NaOH. Then I blew through it, but instead of turning red it only turned purple, the red of the litmus being masked by the blue of the starch iodide. This shows that it is not only HI, but the excess of KI also, which is oxidized by some constituent of the resin, with liberation of I.

The following is the way in which I derived the figures for the third and fourth columns:—

Suppose × grams of Br are taken.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Let P = part of resin unaffected by Br;

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

n H = H atoms replaceable byn Br;

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

y Br = weight of Br which can be added to compound:

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Pn H + x Br = P(n Br) — (y Br) + n Br + (x — 2n — y)Br.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

y Br = x Br — 2n Br — (x — 2n — y)Br

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

(x — 2n — y)Br is the excess of Br determined as free I. n Br is the Br determined as HI.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

To get the total absorption I substract the weight of Br calculated from Na₂S₂O₃ from the weight originally taken and calculate Br/Weight of resin × 100/1.

From the weight of Br used in getting the total percentage absorption I now take twice the weight of Br calculated as HBr, and reckon out the percentage absorption of the remainder with the weight of resin taken. By subtracting the latter percentage from the former I get the percentage of Br replacing H in the resin and appearing as free HI. I used all ordinary Precautions to keep the solution dry, such as heating and drying the flasks before adding Br, and keeping the standard Br solution and resin-oils over CaCl₂.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

It is evident that there is no uncombined carbon in the clear resin, and yet come is always left after distillation. A good deal of water is given off, only a negligible quantity of which can consist of hygroscopic water, since the resin is all but impervious to water and not hygroscopic, for I ground some to an exceedingly fine powder and weighed—

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

It thus contains a very small proportion of water, and, when perfectly dry, absorbs moisture very slowly, even when in a finely powdered condition. So all the water must be due

to combined O and H. At the beginning of the distillation the conditions must favour the formation of water, since most of the water is given off just before 100 and up to about 200. After this the proportion of water in the distillate grows negligible. This loss of combined water at an early stage is evidently responsible for the fact that we get a very small proportion of oxidized products in the distilled oils, although the resin itself is a very highly oxidized substance, approximating to the molecular constitution C_2n H_8n O_n by a combustion. The escaping gases are thus oxidized at or before the moment of liberation, water being separated. This view accounts for the fact that the first runnings of the distillate consist of more unsaturated carbon-compounds, while the later portions are more saturated.^* The few bromine numbers I have obtained for the oils bear this out. The total absorptions of bromine decrease as the boiling-points of the distillates rise, but the absorption-figure in the second column ^† decreases still more remarkably, for the substituted bromine calculated as HI remains nearly constant through the oils experimented with, which, of course, makes the rise in bromine-absorption less abrupt when we take the first column in the table than when we take the second. The first products, even if saturated, would be deprived of some of their H to form water and an unsaturated compound. As the distillation proceeds the O has nearly all been got rid of as water, so there is now no reason why unsaturated compounds should be produced rather than saturated. This accounts for the large drop in the bromine-absorption observed by Mills in the case of ordinary resin-oils between the last of the spirit-like distillate and the beginning of the medium oil. This is the point where the proportion of water in succeeding fractions dwindles almost to nothing. Even now, however, it is probable that the formation of unsaturated compounds is favoured by the small proportion of H left relatively to C. We started with the composition C_2n H_3n O_n; nearly all the O has been removed as H₂O, so we are left with C_2n H_n as the basis from which to form the remaining distillate. So at high temperatures the bromine-absorption of the distillate hardly decreases, and remains stationary in the last runnings. We should expect an increase, if anything, at the end—that is, if all the resin is to come over as oil. But if the distillation is pushed to the end there is a residue of pure C, so we must suppose that the unsaturated bodies which would have to be formed to bring over all the C in the combined state are not stable at such high