irrelevant when regarded as isolated facts, may, when taken together, throw light upon the question at issue. Social evidence, then, has an advantage over legal evidence in that it can include facts of slight probative value. The author treats evidence under three heads-real, testimonial, and circumstantial evidence. Real evidence-that is, facts which the investigator sees for himself-needs little discussion. Testimonial evidence the assertions of human beings whether oral or documentary-is discussed at some length. Stress is laid on the necessity of knowing whether the witness is a first-hand observer or is quoting what he has heard at second or third hand, how reliable he is in general, and whether he has any bias in the present matter.

The two factors which condition the value of a witness' testimony are his competence and his bias. Competence includes both the witness' opportunity to know the facts and the way in which he has used this opportunity. Bias includes those ideas and emotions of the witness which may prejudice his judgment.

Bias is more

Of the two factors mentioned here, probably competence is the one which investigators are most likely to overlook. If a witness has had an opportunitly to know the facts, too little inquiry is made as to how he used that opportunity and how competent he is to give a reliable account of what he saw or heard. easily allowed for and even has a value of its own. gator who has tried to get at the facts underlying a out knows how the prejudiced testimony of one side throws light on the prejudiced evidence of the other, and of the caution with which any statement must be handled which has behind it strong partisan feeling.

Every investistrike or lock

Circumstantial evidence is evidence to a fact which, if proved, tends to establish the point at issue. As an illustration, the author supposes that the point to be determined is whether or not a man feels affection for his family-"a question which is of practical importance sometimes in case decisions." The wife perhaps testifies that he does not care for her or their children. This is testimonial evidence, and needs testing only as to her competence and possible bias.

When, however, she names certain cumulative circumstances from which might be inferred the state of the man's affections; when she says that he gives her $6 a week out of $22, that he spends over half his leisure time away from home, that he is irritable when he does appear there, etc., these statements, which would be direct testimony as to his habits are only indirect testimony as to his indifference. Note in the foregoing the cumulative effect of adding item to item of indirect evidence, each a comparatively weak basis of inference in itself, but gaining in cogency with every circumstance added.

*

*

It is in the use of indirect or circumstantial evidence that investigations may differ most widely from court procedure.

Social

workers may fairly and reasonably take cognizance of the cumulative effect of many small items of evidence, each one of which by itself might be inadmissible under the strict rules of court testimony. This in turn throws on the worker the responsibility of examining each item carefully for evidence of carelessness, incompetence, or bias on the part of the witness.

The comparative value of each of these three forms of evidence, the risks involved in handling them, and the mental predispositions against which the worker must be on his guard when dealing with them are discussed somewhat fully, this discussion occupying about one-fourth of the book. The remainder deals with the application of these principles to the peculiar problems of case workers. There is much in this part of discussion of interest to the more general worker, but little of direct value.

THIRD ANNUAL MEETING OF THE INTERNATIONAL ASSOCIATION OF INDUSTRIAL ACCIDENT BOARDS AND COMMISSIONS.

The beneficial effects upon both employers and employees resulting from the practical operation of workmen's compensation laws in the various States have become so generally recognized that there is little likelihood of a return to the old method of employers' liability, with the attendant costly litigation, its delays and disappointments, and the bad feeling thereby engendered. Thirty-seven States, the Territories of Alaska and Hawaii, and the Federal Government now have workmen's compensation laws. As might be expected, there is a lack of uniformity in the legislation, and it was thought desirable by some of those charged with the duty of administering these laws that an organization should be formed for the purpose of bringing together administrative officials to discuss the interpretation of the laws and the puzzling problems of administration and to adopt, so far as possible, uniform practices and statistics. Accordingly, on April 14 and 15, 1914, representatives from several newly created industrial accident boards and commissions met at Lansing, Mich., and formed the National Association of Industrial Accident Boards and Commissions.

As a means of carrying out the purposes of this association, three conferences have since been held. The first, a special meeting, was held in Chicago January 12 and 13, 1915. This meeting was devoted largely to a discussion of standardization of accident statistics.

At the second annual meeting, held in Seattle, Wash., September 30 to October 2, 1915, the Province of Ontario, Canada, was received

as a member, and the title of the association was changed to the International Association of Industrial Accident Boards and Commissions. The most important action taken at this second annual meeting was the adoption of the tentative report of the committee on statistics and compensation insurance cost and the continuance of this committee.

The third annual meeting was held at Columbus, Ohio, April 25 to 28, 1916. The proceedings of this meeting are published in full in Bulletin 210, just issued by the Bureau of Labor Statistics of the United States Department of Labor. Among the resolutions adopted by the association at this meeting was one requesting the Bureau of Labor Statistics to publish from time to time the proceedings of the conventions and conferences of the association. Another most important action at the Columbus meeting was the election of the United States Commissioner of Labor Statistics as secretary-treasurer of the association.

At this conference addresses of timely interest were made covering such subjects as the conflicts between Federal and State jurisdictions in commerce cases, merit rating in workmen's compensation insurance, the why and how of uniform industrial accident statistics for the United States, the use of accident statistics for accident prevention, educational work in accident prevention, the theory and practice of compensation, a comparison of the methods of dealing with permanent partial disability cases, Ohio's experience with State insurance, the relation of workmen's compensation to old age, health, and unemployment insurance. As a part of the meeting the first sessions of the newly organized medical section were held, and a safety committee was organized to form the nucleus of a proposed safety section, later indorsed by the association, to advance the promotion of accident prevention.

Boston was selected as the place for the next convention, to be held during the third week in August, 1917.

REPORT OF THE ILLINOIS PENSION LAWS COMMISSION.

The Forty-ninth General Assembly of the State of Illinois, in January, 1916, created the Illinois Pension Laws Commission and placed upon it the responsibility of investigating the operation of all the State pension laws, to ascertain the present and probable future cost of maintaining the funds created by those laws, and to collect all available information in regard to the operation of similar laws in other States and countries. The report of the commission has been

issued in two parts embraced in one volume.1 Part I is entitled Investigations with certain comparative studies, and presents the history of pensions for public employees in other countries, reviews in detail the pension laws of Illinois, recites the terms of the leading pension laws of other States in this country, sets forth the results of a comprehensive actuarial investigation of the five chief public pension funds in Illinois, and gives important statistics concerning other pension funds and the extent of present and possible future pension legislation for public-service employees in the State. Part II outlines briefly the scope and main features of the pension problem, states the essential principles and provisions which, in the opinion of the commission, should constitute a normal or standard pension plan, and makes specific recommendations regarding changes that should be made in some of the existing pension acts pending further investigation and study of the pension problem.

Information as to the operation of the pension laws was obtained in Chicago personally by the commission and in all the other cities, towns, and villages having a population of 5,000 or more from the mayors thereof, who, however, in some instances did not furnish detailed reports as to the present cost, so that the investigation in this particular, outside of Chicago, does not appear to have been complete. Data as to future costs, it is stated, were somewhat more difficult to obtain, since many considerations which can not be positively known enter into the calculation. The commission found 15 pension acts on the statute books on January 1, 1916. One of these, however, does not involve the expenditure of public funds, one was not put into effect, and a third was declared invalid. No summary statement of the operation of the remaining 12 is given in the report, but the following table has been compiled to show the number of employees affected, the number of pensioners, the amount paid to pensioners, and the estimated future cost of maintaining the five chief public pension funds existing on January 1, 1916:

1 Report of the Illinois Pension Laws Commission, 1916. Springfield, 1917. 310 pp.

STATEMENT OF THE OPERATION OF FIVE PENSION FUNDS IN ILLINOIS, SHOWING THE NUMBER OF EMPLOYEES AFFECTED, THE NUMBER OF PENSIONERS ON JAN. 1, 1916, THE AMOUNTS RECEIVED AND DISBURSED DURING 1915, AND ESTIMATED FUTURE PAYMENTS.

1"After a system becomes such that the number of employees remains constant, new entrants coming in only to fill vacancies caused by withdrawals and deaths, then there will come a time when pension payments will be the same from year to year. When this time comes, the system is said to be carrying its ultimate normal load."

Of this amount, $117,209.35 was contributed by employees.

Of this amount, $129,875.55 was deducted from teachers' salaries.

Of this amount, $134,127.22 was received from employees.

No pension payable until the expiration of 5 years from July 1, 1911.

Estimated; outside of Chicago and Peoria.

During the period Dec. 30, 1915, to Sept. 30, 1916. This amount includes $198,014.20 received from teachers, which, in turn, includes $33,175.13 in salary deductions.

On the roll Sept. 30, 1916.

• Disbursements from Jan. 1, 1916, to Sept. 30, 1916.

In Part II the commission announces its opinion that the various pension schemes now existing under the laws of Illinois for public employees not only "are inharmonious and often contradictory with reference to each other, but that with perhaps a single minor exception they are financially unsound and moving toward a crisis. It shows the urgent need for the adoption by law of a revised pension plan based on true principles and for the prompt merging of existing pension systems into that revised plan on terms not too burdensome to the contributing public authorities or to the contributing employees. The general condition of pension systems operating under the laws of Illinois may be correctly described as one of insolvency."

To reach a solution of the pension problem our most promising course would seem to be to proceed to develop a pension system in line with definite principles which we conceive to be fundamental to efficiency in the public service and to the welfare of employees.

The report then discusses in detail several principles which should control in working out such a revised pension system, as follows:

1. The theory of public employee pensions.

2. The question of who should be beneficiaries-whether employees only, or their widows and children also.