paper throw little light on the manner in which calcium and sulfur are combined as sulfides.

In the course of the investigation, of which this paper records but a part, many dips were analyzed. Every dip, when analyzed, gave percentages of calcium and sulfur combined as sulfides and as calcium thiosulfate. These figures varied somewhat, but could always be assigned to a solution of various mixtures of monosulfide of calcium, one or more of the polysulfides of calcium, and thiosulfate of calcium.

For the sake of simplicity and for comparison, it is assumed that the dip is a solution of a mixture of monosulfide of calcium, pentasulfide of calcium, and calcium thiosulfate, the relative proportions of which may vary, and of course would vary in a dip exposed to the air, if they decompose at different rates. Some illustrations may be given:

2CaS.18CaSs. 5CaS2O3.

5CaS.6CaS. 3CaS2O3.

It is not believed that so simple a mixture is the true one, but for comparing results in the vat experiment the above possibility is given.

Some interesting facts may be noted by turning to table I. In the first vertical column it will be seen that the specific gravity is constantly and regularly increased. This increase is due, first, obviously to concentration from evaporation of the water; secondly, to the oxidation of the sulfides. In the next column it will be seen that the percentage of the calcium thiosulfate constantly increases, the greatest increase occurring in the first three days of the experiment. The total increase was 0.99 per cent. in the sixteen days, an average of 0.063 per cent. per day. This increase in calcium thiosulfate may be ascribed to two causes: First, to the concentration, as shown in the increase of specific gravity; and second, to the oxidation of polysulfides from the oxygen in the air:

CaSO CaS2O3 + S3.

(The rate of evaporation was found in a later experiment for the same vat and same kind of dip to be 40.16 per cent. for sixteen days; so the thiosulfate from this cause alone would have increased from 0.3 per cent. on the first day to about 0.5 per cent. on the sixteenth day.)

In the third column the percentage of total calcium shows a constant and regular increase. There are at least two independent agents acting here: First, concentration acting to increase the per

centage; and second, the decomposition of the sulfides by atmospheric carbon dioxide throwing out the calcium as calcium carbonate, and thus tending to decrease the percentage.

In the fourth column may be seen the percentages of total sulfur. After the second day this percentage remains practically constant, varying only 0.03 per cent. As with calcium, there are several changes affecting these figures: First, concentration, tending to increase the percentage; second, decomposition of sulfides by carbon dioxide in the air forming, (a) in the case of polysulfides, volatile hydrogen sulfide and insoluble sulfur, and, (b) in the case of monosulfides, volatile hydrogen sulfide only, tending to decrease the percentage; and third, oxidation to polysulfides and thiosulfates, thereby throwing sulfur out of solution. These three agencies, and perhaps others acting in the same and different directions, produce equilibrium in the percentage.

If there were sulfites or sulfates present, we would expect a further decrease in percentage total sulfur, owing to the throwing of sulfur out of solution according to the reactions,

A study of table II will reveal some other interesting facts. The columns representing calcium and sulfur calculated from the percentage of calcium thiosulfate of course show the same rate of increase as the thiosulfate column in table I.

The column giving percentages of calcium in sulfates shows a constant decrease, as would be expected, since the increase of calcium in thiosulfate must have been at the expense of the calcium in the sulfide. The same holds true for the sulfur in sulfides.

On the first day the ratio of calcium in thiosulfate to calcium in sulfides was 0.08 to 0.64, or as 1 to 8; on the sixteenth day the ratio of calcium in thiosulfate to calcium in sulfides had increased to 34 to 52, or more than 5 to 8.

Likewise the ratio on the first day of the sulfur in the thiosulfate to sulfur in the sulfides was 1 to 13, while on the sixteenth day it had increased to nearly 4 to 13. This disagreement between the two ratios would be expected if the explanation regarding the calcium and sulfur in table I be accepted. The ratio of calcium in sulfides to sulfur in sulfides on the first day is 64 to 211, while on the sixteenth day the ratio had changed to 52 to 182, or, in terms of the first, to 64 to 224. This difference in ratio in calcium and

sulfur is easily explained if it be accepted that the dip is a mixture of two or more sulfides decomposing at different rates.

Assuming, for the sake of comparison, as has been stated before, that the dip is a solution of a mixture of monosulfide, pentasulfide, and thiosulfate, then on the first day 2CaS+ 7CaSs + CaS2O3 would correspond to the different percentages, while CaS+ 9CaS5 +5CaS203 would represent the mixture for the percentage given on the sixteenth day.

ON THE DURABILITY OF CEMENT PLASTER.

By E. H. S. BAILEY and W. G. STROMQUIST, University of Kansas, Lawrence.

FORMERLY the only material used for inside plastering was

lime mortar, and upon this was spread a skim coat made from plaster of Paris. Since the discovery of extensive beds of gypsite or gypsum dirt, in various localities in Texas, Indian Territory, Kansas, and Wyoming, this material has been heated to drive off nearly all of the water, and is put upon the market under the names of Agatite, Acme, Alabastine, etc., as a substitute for lime mortar. As this crude gypsum occurs in beds where it can be handled by a grader, loaded on wagons, and hauled directly to the mill, the production of this kind of plaster has become an important industry. While the ordinary gypsum rock must be ground before it can be heated in the kettles, this material needs no preliminary treatment. The composition of the raw gypsite, the finished product and the set plaster is as follows: *

From an examination of the analyses, it will be seen that the crude material loses all but about five per cent. of its water when "boiled," and when water is added to the "plaster," before spreading on the wall, it retains practically the same amount of water as it lost when heated; and in fact it again becomes gypsum, or CaSO4+2H2O.

For practical work, the mason mixes with the cement plaster from three to five parts of sand, and a little "retarder" is used to prevent the too rapid setting. This material has been used for plastering walls for the past fifteen years, and, as far as observed, it seems to last as well as lime mortar.

An opportunity to study the disintegration of this plaster under

⚫ University Geological Survey of Kansas, vol. V, p. 163.

peculiar conditions was recently afforded the author. In the Chemistry building at the State University, which was completed in 1900, there is a room which is plastered on the sides, and in which is situated the fan which ventilates the building. On one side this room is partially open to the air, and cold air is drawn over hot steam-pipes in the cold weather as it enters the room. The air is then drawn through the fan into the flues, by which, after passing over other steam-coils, it is distributed throughout the building. The temperature of the air in the fan room varies with that of the outside air, ranging probably in the winter from 40° F. to 72° F.

It was noticed that the plastering on both sides of the room, above a line extending from the floor on the sides nearest the pipes to the ceiling on the opposite side, had become powdery, and fallen off. In order, if possible, to find the cause of this, several analyses were made of the hard plaster which was still in place in the wall and of that which had fallen off, with the following results in one case :

As the samples are taken from different parts of the wall, they will not agree in composition. An inspection of the analysis, however, shows that in every case the fallen plaster contains less than half as much water as the plaster which is in place on the wall. Comparing this analysis with those previously given, we can also form some estimate of the amount of sand that is used with the original material.

It is generally conceded that the setting of plaster of Paris is due to interlacing of the small crystals of gypsum, which are formed in the moist material, and we are familiar with the fact that if all the water is driven out of the gypsum in heating, the plaster will not readily "set." It is evident, then, that about five per cent. of water is necessary to the stability of the finished wall.