into the tip served to prevent the sand being drawn through into the vacuum can. (2) A vacuum can of 28 litres' capacity, provided with stop cocks at either end, and a vacuum gauge placed near one end used as an aspirator; also the necessary rubber tubing and friction joint couplings. (3) An exhaust pump. (4) An half-inch L-shaped glass tube for transferring the melted gelatine from the test tube into the aërobioscope. (5) Sand, used as a filter, No. 60 (mesh) one inch in length. It is needless to add that necessary precautions connected with bacteriological technique were observed throughout, such as sterilization, planting of samples, etc.

The samples of air collected in each case were of ten litres. The method of taking the samples of air was as follows: Ten litres were exhausted from the 28-litre aspirator can by the vacuum pump, as registered by the vacuum gauge. The stop cock was turned, the air pump with its rubber attachment was removed and replaced by the sterilized aerobioscope with its proper connections, attaching it to the same ends of the vacuum can. At the same instant the cotton plug in the flared end of the aërobioscope was removed and the stop cock of the aspirator can opened. The action of the partial vacuum in the can drew a current of outer air through the aerobioscope until the vacuum gauge, by running down, showed that ten litres of outer air had passed through the aerobioscope and into the can. The sterilized cotton plug was replaced in the open end of the aërobioscope. The aerobioscope was detached from the can and that end was likewise closed with a sterilized cotton plug. The aerobioscope has caught in its filter and contains all the germs present in ten litres of the air to be tested.

The next step was to transfer to gelatine the germs collected in the sand filter. To effect this the aërobioscope was tilted so that the sand filter and its contents were all shaken down into the cylindrical portion of the aerobioscope. The plug at the larger end was removed and the melted gelatine poured into the cylindrical portion. The cotton plug was replaced. The aërobioscope was gently and carefully agitated until the gelatine and sand were uniformly mixed, care being

At the same time the

taken to insure a complete mixture. motion was not so violent as to cause bubbles in the gelatine. When the mixture was perfect the tube was rapidly rotated in ice water until in perhaps half a minute the contents of the cylinder had solidified in an even film on the inner side of the aerobioscope. The tubes with the planted samples are then laid away in a dark place at the ordinary temperature to await development of germs if any be present.

The chance for error in the aerobioscope method is slight. When the sand is shaken into the cylindrical portion of the aërobioscope it does not necessarily come in contact with the entire inner surface of the drawn-out end in which the filtering substance is placed. Here a very slight error might arise in case some germ or germs failed to be carried into the cylindrical portion of the aërobioscope upon the transferring of the filtering substance. Again, when the culture medium, gelatine, is added to the sand, its introduction into the aerobioscope is not without some possible contamination, but sterile apparatus and good technique reduce the possibility of error to a minimum. With liquefying colonies great care must be taken to prevent their products of liquefaction from obscuring or inhibiting the growth of other colonies.

The planted samples of air should be examined several times a day, and if counts of colonies are taken each time, careful observation will prevent such error. The advantages of this method lie in the simplicity of the apparatus used. The aerobioscope is a single cylindrically shaped piece of glass easily sterilized and manipulated. The aspirator can and suction pump complete the outfit. The counting of colonies on the gelatine film lining the aerobioscope is simplified in proportion to the surface area, which in the Sedgwick-Tucker aerobioscope is ample for a ten-litre sample. By the division of the outer cylindrical surface of the aerobioscope into sections one inch square, the counting of colonies is further facilitated.

In the microscopical examination subsequent to the planting and developing of the samples of air collected, only

quantitative results were considered, the real object of the experiments being to prove or disprove the efficiency of the different systems of ventilation.

Among other buildings, theatres were visited, and the ventilating plants, when present, carefully examined. Two theatres were finally selected for carrying on experimentation. Theatre A, with a good mechanical system of ventilation, and Theatre B, where no regular system was present, save ordinary window ventilation. The tests lasted during two consecutive weeks, samples being taken at evening performances only.

Theatre A. The air supplying this theatre is drawn in through a shaft, extending from the roof to the basement of the building, by means of a large fan. From the bottom of the shaft the air is forced through a system of radiating flues, which distributes the air under the main floor of the theatre and in turn into the audience hall through a system of ventilating holes under the seats. In the roof of the theatre are several large ventilators through which the air makes its exit.

Samples of air were taken (1) at the bottom of the shaft; (2) in the audience hall before the doors were open to the public; (3) same as 2, but at the end of the second act; (4) from the second balcony during the third act; (5) same as 2, immediately after the entertainment.

Theatre B. There is no mechanical system of ventilation in this theatre. The air enters at will through most rudimentary ventilators, windows forming the principal means for obtaining the air. The dome of the theatre is provided with one large central ventilator and two small ones on the extreme right and left.

The locations selected for taking samples corresponded as far as possible with those selected in Theatre A: (1) from the level of the street; (2) in the centre of the theatre before the doors were open to the public; (3) in the aisle of the main floor at the end of the second act; (4) at the top of the theatre overhanging the second balcony; (5) the same as 3 at the close of the performance.

Although moulds were counted and tabulated in every sample of air taken, their consideration is of minor importance in this paper and will consequently be omitted. The following tabulated figures represent the number of individual colonies present in the samples of air taken. Only the mean, the extreme, and the average counts will be given.

The question of good management with reference to the "housekeeping" of theatres is an important point to be considered before drawing conclusions from the above table. There are, however, certain routine details, in taking care of a theatre, shared in common by theatre managers. Provided equal ability was present in these two instances, I think we are justified in making a point in favor of mechanical ventilation.

The amount of dirt introduced into the two buildings in question is practically the same; but by the action of the "forced draft" which is a constant factor in Theatre A, the suspended particles, instead of being allowed to settle night after night and thus accumulate, are carried out of the building by the constant upward current induced by the fans. The above averages show conclusively, then, the superior ventilative advantages of Theatre A over those of Theatre B. In every instance the analyses from Theatre A show a much purer condition of the air than is found in Theatre B.

EXPERIMENTS AT BOSTON PUBLIC LIBRARY.

The Public Library was selected for air tests because of its modern and most complete ventilation plant.

The air is obtained from the ground level, but enters the building from the central grass court, which is sheltered from the streets by the walls of the library, and consequently is comparatively free from dust and germs. The air is next forced through two sets of filtration bags, one on either side of the main fan. Each bag is twenty-five feet long and two and a half. feet in diameter; twelve bags constituting a set. The arrangement of the bags is as follows: The large ventilator flue, receiving air directly from the fans, is divided into twelve circular openings, each two and a half feet in diameter, by an iron frame, and into these openings the mouths of the bags are fitted. The bags are then held in place by being suspended along horizontal wires. The current of air produced by the fans is sufficient to force the air through these bags into the ventilative flues and in turn into the several halls of the library. At the top of the building is a fan which draws air from the building, and thus a constant stream of air is passing from the basement to the ventilator in the roof.

The following table represents only the mean, the extreme, and the average counts of colonies present:

The above figures are self-explanatory. The filter removes practically all the organic life contained in the air, as well as a great amount of dust and other suspended matter. There can be no question, then, as to the sanitary advantages of filtering the air before sending it into buildings.

Samples of air were also taken from Bates Hall, the head of the main stairway, in the lower corridor, and as it passes out of the building directly in front of the exit fan.

The following table shows a marked variation in the quality of the air in different parts of the building. A great source of contamination presents itself at the main entrance, where outside air enters from the constant opening of the doors.