ence," being a complete record of every letter passing in and out of the office.

An idea of the extent to which these records are used may be had when it is mentioned that 800,000 pages of such records are filled each year in the executive departments at Washington. It is believed that with the installation of modern filing systems the necessity for these book registers will disappear.

THE PRESS COPY VERSUS THE CARBON COPY.

By the almost universal practice of business concerns the carbon copy has supplanted the press copy as a record of outgoing correspondence. A number of offices in the Government have adopted the same course and use the carbon copy exclusively, but most of them still adhere to the press-copy method.

In studying the question of the relative economy and efficiency of the carbon and press copy as a record of outgoing correspondence, the following considerations have more or less importance:

1. Permanency.

2. Economy.

3. Adaptability.

(a) As facilitating dispatch of mail.

(b) From standpoint of appearance and neatness.

(c) From standpoint of saving time of executives and correspondence clerks.

1. PERMANENCY.

After a careful consideration of the reports of tests of inks, typewriter ribbons, and carbon papers, made by chemists, both in Government employ and elsewhere, it is believed that an indelible copy can not be obtained from a typewriter ribbon, since copying inks are more or less fugitive and many press copies if exposed to the light for any length of time become dim or fade entirely away. Even under the best conditions the press copy fades after 40 to 50 years, whereas the carbon copy makes a permanent nonfading record. The effect of this fading quality of the press-copy record has not yet displayed itself with sufficient force to be a pronounced factor, since the typewriter has been in use only about 35 or 40 years.

2. ECONOMY.

Consideration of the question from the standpoint of economy is somewhat difficult, since in nearly every office a different method is pursued in press copying letters. In the executive departments, however, the part of the salaries proportionate to the time spent by employees in press copying letters amounts to $31,080.46 per annum, according to the statements made in answer to the inquiry of the commission on the subject of handling and filing correspondence.

The average 500-page press copy book, according to the Government schedule, costs $1.33. It is a conservative estimate that onehalf dozen press-copy cloths costing 28 cents a dozen are used up in copying 1,000 letters. Therefore the average cost for materials per 1,000 letters is $2.80.

On the other hand, it is estimated that 65 sheets of carbon paper will suffice to make 1,000 carbon copies. The cost of 65 sheets, according to the Government schedule, is 22 cents. One thousand sheets of manila paper, without letterhead printed thereon, can be obtained at a cost of 34 cents. The total of these two items is 56 cents a thousand copies, as compared with a cost of $2.80 for press copy books and copy cloths, a saving in cost for material in favor of the carbon-copy method of $2.24 a thousand. Applying this rate to the 14,000,000 communications which, it appears from the reports submitted by the executive departments upon the subject of handling and filing correspondence, are press copied annually, the amount of the saving to be effected is over $31,000. There are very few offices where the carbon copy is not taken as well as press copy; so that as a matter of fact the press copy is an additional record which in most cases could be abolished without the substituted expense of taking the carbon copy. In those offices where, due to some peculiar conditions, a chronological record of outgoing correspondence is imperative, an additional carbon copy could be taken and preserved in chronological order.

By discontinuing press copying the space now occupied by presscopy outfits becomes available for other purposes. The amount of space varies, but it is a conservative estimate that an average of 100 square feet of working space is consumed by the presscopying outfit in each office, not to mention the space required for filing the bound press-copy books. This storage space is constantly growing on account of the accumulation of press-copy books. When it is considered that space in Government buildings is worth on the average more than 50 cents a square foot, this factor becomes of importance.

3. ADAPTABILITY.

(a) From standpoint of facilitating dispatch of mail.-By the use of carbon copies the original is ready to mail the minute it is signed, so that mail signed late in the day may be dispatched the same day without requiring overtime work on the part of any employee, while papers signed earlier may catch an earlier mail or be earlier transmitted to another office. This is a distinct increase in efficiency.

(b) From standpoint of appearance and neatness.-In making carbon copies a record-ink ribbon and a record-ink carbon may be used, giving a clear, neat, clean appearance to both original and file copies. A copying ribbon gives a smeary appearance, the ink rubs off on the hands before copying, it is more difficult to erase in making corrections, and unless press copying is carefully done the ink runs or spreads when the letter is copied. The wetting of the paper gives it a wrinkled appearance, the creases become set, and the paper has a tendency to curl or fold itself up after being unfolded for reading. Envelopes coming in contact with recently copied letters often seal themselves and must be destroyed and new ones addressed.

(c) From standpoint of saving time of executives and correspondence clerks. Another factor to consider is the time and convenience of officers and correspondence clerks who desire to read or consult a file of papers containing either carbon or press copies. The carbon copy is legible, smooth, of sufficient weight and stiffness to be readily handled, whereas the press copy is quite often illegible, frequently indistinct, is liable to crumple on account of the flimsy nature of the

paper which necessarily must be used, and will not stand up in the files

From the standpoint of filing convenience the carbon copy is practically indispensable if vertical files are used, the press copy being so flimsy that it can not well be handled unless mounted on a paper backing

THE DICTATION MACHINE.

Guided by the experience of large commercial concerns in the use of the dictation machine, and knowledge of the fact that the installation of such machines in recent years has been extending rapidly, careful consideration has been given to the question of the practicability of using the phonograph for the preparation of correspondence in the Government service.

There appears to exist some prejudice against these machines on the part of stenographers. At least, such has been the experience of commercial concerns now using the phonograph. To cite one example: A large corporation is now using 150 phonographs. Upon the first installation of the machines there was such a strong and universal protest from the stenographers that very shortly after the experiment was begun it was found impossible to continue their use and the machines were removed. A second trial on a large scale was had, but again the prejudice was sufficient to secure the removal of the machines. Upon the third trial the machines were adopted. And the result has proved the wisdom of the change. By the present phonograph method the average cost of a letter is 2.7 cents, whereas by the stenographic method formerly employed the average cost per letter was 4.3 cents. If offices in the Government can attain results to be compared with these, there can be no argument against the installation of such machines throughout the service.

One of the strongest arguments that the dictation machine is an economical aid in the preparation of correspondence lies in the fact that some of the largest railroad companies and industrial corporations in the country are using them, after careful test. It may be assumed they would not use the phonograph unless they found it more efficient and economical.

The dictation machine saves the time not only of the stenographer but of the dictator, and in the latter respect is about the only laborsaving device that affects the departmental head or executive personally. It increases his efficiency and gives him concentration and independence of action. All the other time-saving devices are for assistants with less valuable time and less demand for executive promptness, privacy, and high efficiency. To save the time of the person who directs others is of prime importance, and when the dictation machine accomplishes this and at the same time economizes the time of the transcriber, it becomes valuable as a piece of office machinery.

It is not disputed that the dictation machine saves the time of the stenographer; tests made anywhere prove that, but the saving in the time of the dictator is by indirect process and is not so easily ascertainable and demonstrated. That executives are benefited, however, is not open to serious question. In the first instance, the executives may use their spare moments between their many interruptions to dictate correspondence, orders to assistants, etc. The

experience of one of the high executive officers of a large railroad company was that the phonograph proved a great time saver for his assistants. He had 15 call buttons on his desk, and before he used the phonograph would press a button, calling some important head from a distance to receive a few instructions which are now typewritten in better form and at a minimum expense through the phonograph. This practice also eliminates the uncertainty of verbal instructions. In fact, in many large commercial concerns it has been made the rule that oral instructions will not be permitted.

The commission recently installed in its offices 16 dictation machines for experimental purposes. Careful record was kept of the rate of speed in turning out work by the phonographic as compared with the stenographic method. The statistical result of this test is given in the table below, which shows for a period of 10 weeks the weekly output and the rate of speed by each method.

An examination of this table discloses a constant improvement in speed by the dictation-machine operators, while the rate of output by the stenographic operators remained stationary. At the outset, however, the test disclosed no saving; in fact, the rate of speed of the stenographers was greater than that of the dictation-machine operators. This was due to the fact that the comparison was between experts and novices. The stenographers had had from 1 to 10 years' experience, whereas but one of those engaged in the phonographic test had ever before used a dictation machine. Under such conditions any improvement whatever in speed by the operators of the dictation machine over the stenographers is significant and a promise of a very great improvement as fast as the machine operators become accustomed to the machines and expert in their use. If, at the outset, the novices using the dictation machine can produce as much work as the expert stenographers the economy of the phonograph is demonstrated. The stenographers in the continuation of the test are not likely to increase their speed, but an increase in the output of the phonographic operators is inevitable.

As proof of the above statement, it will be noted from the preceding table that, comparing the output of the first five weeks of the test with that of the last five weeks, the rate of speed of the stenographers decreased 8 per cent, while the rate of the dictation-machine operators increased 73 per cent, and that during the latter period the output

of the dictation-machine operators was at a rate of speed 125 per cent greater than that maintained by the stenographers.

A number of offices in the executive departments have installed dictation machines for the purpose of making a trial.

In connection with these tests careful statistical records are being kept to show the comparative speed at which work is produced by means of the phonographic and stenographic methods. The same difficulties which confronted the commercial concerns are being met with, but as the result of persevering efforts and excellent cooperation from those interested in the subject in the executive departments, fair trials and accurate statements of results are expected.

In one of the large divisions of an executive department 30 machines were installed for the purpose of a test which lasted two months. Before the test was undertaken a complete record of the time spent by stenographers at the desks of correspondence clerks and others in connection with the taking of dictation, together with the time consumed in transcribing the same, was maintained for a month. This record showed that the output of the stenographers of this division during the time actually engaged in stenographic and transcribing work was at the rate of 14.67 words a minute.

During the succeeding seven weeks the method of producing correspondence was changed to the phonographic, and a like record was kept. This record showed that the output of 23 operators, amounting to 730,504 words of letters, etc., produced in 693 working hours, was at the rate of 17.56 words a minute, or an increase in rate of approximately 20 per cent.

The above figures, however, do not set forth with complete justice. to the dictation machines the comparative efficiency of the two methods, since in the case of the stenographers they had had years of experience, whereas none of them had ever before used the dictation machine. For this reason, at the outset, the rate of output by means of the dictation machine was yery low; in fact, it was lower than the rate of output by the stenographic method. As the test progressed, however, the increase of rate of output was pronounced and steady, so that, while the rate was only 14 words a minute during the first week, during the seventh or last week of the test the rate was over 32 words a minute, this last rate showing an increase over the stenographic method of more than 100 per cent. It is confidently believed that the rate of output by the dictation-machine method would have increased still more, but as the test had already lasted seven weeks and the results seemed to be conclusive, it was not regarded necessary to continue the trial longer.

The statistics kept showed only the saving in time or the increase of output by the stenographers. It took no account of the saving in time which accrued to the dictators, whose time is more valuable, on account of their higher compensation, but impossible to measure in a true comparison with the stenographic method on account of the following circumstances: The difference of method by the dictator in using the stenographer and the dictation machine is radical. The usual practice by the former method is for the dictator to take a succession of cases, go over them, prepare in each case the reply he intends to make, after which he calls the stenographer and proceeds with his dictation. By this time the dictator has in his mind some 20 to 30 replies, but he has read so many letters and consulted so