TABLE 2.-ORDER OF FREQUENCY OF CONSULTATION IN THE SEPARATE CITIES OF THE 20 SOURCESa MOST OFTEN USED IN THE THREE CITIES TAKEN TOGETHERb

a Exclusive of private and public charitable and social agencies. b The 20 sources most used in the three cities taken together were selected as follows: For each city all the sources were numbered in order of frequency of consultation, beginning with the source most frequently consulted. The numbers indicating the order of frequency of each source in the three cities were then added together. The 20 sources showing the smallest resulting totals are included in the table.

c Same number of consultations with lawyers and with fellow church members, in records of first city.

d Same number of consultations with former neighbors and with courts, in records of second city.

e Truant officers were consulted same number of times as nurses, in records of second city.

f Lodgers were consulted same number of times as lawyers, in records of second city.

8 Same number of consultations with teachers, with police, and with nurses in records of third city.

h Former tradesmen and foreign consuls were consulted same number of times as health departments, in records of third city.

i District or county attorneys were consulted same number of times as present tradesmen, in records of third city.

neighbors" actually heads the list of sources, and it stands third in the second city; "present" and "former landlords" in the second city and "present landlords" in the third are also used very frequently. The significance of these differences in practice as between the first and the second and third cities, and the serious dangers involved in the use of neighborhood sources at all are brought out in the chapter on Neighborhood Sources. Another difference is in the use of "policemen," which may be accounted for by the fact that no study of probation work, either adult or juvenile, was possible in the second and third cities, while in the first city both of these and a number of other forms of work in which the evidence of the police would be important were included. The marked differences in the use of "teachers" may be accounted for by the fact that the proportion of agencies studied which dealt exclusively with children was smallest in the third city and largest in the second.

But when we consider that social case work in these many forms and these different places has not been developed out of any formulated and systematically transmitted experience, for the most part, but has felt its way quite independently, often, to processes that grew out of the necessities of the day's work, we shall find the resemblances in practice as shown in these statistics quite as striking as the differences, especially if we compare the larger groupings of sources shown in the large general Table A in Appendix II. Here the fact stands out that, with many variations within each group and variations as between groups allowed for, there still has been frequent use in all three cities of the sources classified under social agencies and churches, doctors and health agencies, former and present neighborhoods, relatives, former and present employers, schools, friends, and public records.

As to the shifting of emphasis upon one or another of these, as we change from family work to children's work, from one form of children's work to another, or from all to medical-social service, the figures show some interesting variations. Table B in Appendix II tells the story in detail. Its most important features are brought out in Tables 3 and 4 on the pages immediately following.

The first city was chosen for making this particular comparison because it was the only one of the three (at the time the study was made at least) in which the standard of treatment was quite as

TABLE 3.-TWENTY SOURCES MOST USED BY AGENCIES DOING WORK WITH CHILDREN, FAMILY WORK, AND MEDICAL-SOCIAL WORK IN FIRST CITY, IN ORDER OF FREQUENCY OF USE

a Exclusive of public and private charitable and social agencies.

b Same number of consultations with courts as with present neighbors, in records of work with children.

c Same number of consultations with nurses as with present landlords, in records of work with children.

d Same number of consultations with marriage records as with teachers, in records of family work.

e Same number of consultations with prison and reformatory officials as with truant officers, in records of family work.

f Same number of consultations with health departments as with present employers, in records of medical-social work.

8 Same number of consultations with marriage records and lawyers as with former neighbors, in records of medical-social work.

h Same number of consultations with police as with medical-social service departments, in records of medical-social work.

Same number of consultations with a number of other sources as with birth records, in records of medical-social work.

high in children's work as in family work, and in which medicalsocial work could safely be compared with both. The figures for

TABLE 4.-USE OF THE 20 SOURCESa MOST USED IN THE FIRST CITY, BY AGENCIES DOING WORK WITH CHILDREN, FAMILY WORK, AND MEDICAL-SOCIAL WORK IN THAT CITY

a Exclusive of social and charitable agencies, public and private.

b Only the first consultation with each source used was counted, in any one case.

each separate agency in the first city are given in Table B in Appendix II, thus making it possible for a student interested in forms of organization to compare the classifications in greater detail.

II. PRINCIPLES GOVERNING TIIE Choice of SOURCES After a social organization has abandoned the old idea that a more or less searching cross-examination of one witness (of the client, that is) is an adequate basis for action-even after it has advanced beyond this stage, its earlier adventures in social diagnosis are likely to bring endless variations on the theme of the client's present situation. It will hear this described again and again by charitable individuals and social workers, and will assume

that, in listening to these detailed statements, it is gathering valuable diagnostic material. "All these people," said one critic after examining a group of case records, “seem to have been born just about two years ago. There is a great deal set down about what they and the charities have done since, but they have no past.”

1. First Principle of Choice. When, the first interview over and a home visit paid, we find ourselves with a group of clues, perhaps the most important principle of action is to strike out boldly for bistory and avoid, for the time being at least, those references whose point of view is most like our own. Where there has been an investigation already, we should profit by it, of course, but where there has not, we cannot afford to lose a moment before consulting the people who knew our client at his best, if he ever had a best; who knew him at least at other times and in other relations. These can give, one from this angle, one from that, the aspects that will help us to see our subject in the round. Any notion of social efficiency that excludes the humblest sources of knowledge and help makes for inefficiency in the long run. Too often we confer with the sources that are nearest to us, and neglect those that are nearest to our client. Other things being equal, the evidence of those who knew him before the present difficulties developed is worth ten times as much as the evidence of those who have known him since. But this must not be construed as referring to length of acquaintance, merely. We can know people a long time and really know them very little. The nature of the relation and the opportunities that it has given for self-expression are the important things.

2. Second Principle of Choice. In order that we may never have to pay two visits where one can be made to serve, a second principle will be found useful in making our choice of diagnostic sources, and in preparing, at the same time, for the co-operation which will probably be needed later in treatment. This principle is to seek first those sources that are likely to be rich in history only and seek later those most likely to be rich also in co-operation. It is wise to build in our background and get as clear a picture of

1 Dr. S. Weir Mitchell says of the family doctor, who belongs to this longacquaintance group, "He is supposed to have some mysterious knowledge of your constitution, and yet may not have asked you a medical question in months and years."-Doctor and Patient, p. 28.